IL-13 production by NK cells: IL-13-producing NK and T cells are present in vivo in the absence of IFN-gamma

In this study, we demonstrate that human NK cells, human NK clones, the human NK cell line (NK3.3), and a population of murine NK cells can produce the type 2 cytokine IL-13 in response to IL-2 or phorbol myristate acetate plus ionomycin. IL-2 rapidly induced new IL-13 mRNA and protein synthesis in the NK3.3 cell line. Six of 12 human NK clones tested produced IL-13 protein in response to IL-2 or phorbol myristate acetate and ionomycin. Intracellular analysis revealed that approximately 2% of human peripheral NK cells produced IL-13 protein in response to IL-2. Isolated NK cells from SCID and RAG-2 knockout (-/-) mice that lack T and B cells as well as normal mice also can produce IL-13 mRNA and protein in response to IL-2. We hypothesized that in the absence of IFN-gamma, IL-13-producing NK cells may predominate in vivo. Utilizing IFN-gamma knockout (-/-) mice as a model system, IL-2-activated liver NK and T cells expressed 10-fold more IL-13 and IL-5 mRNA and protein than normal controls following IL-2 treatment in vitro. These results suggest that in the absence of IFN-gamma, an IL-13- and IL-5-producing NK and T cells predominate in vivo. The existence of this cell type has important implications in innate immunity given that the balance between IFN-gamma and IL-13/IL-5-producing NK cells may influence the early development of a cell-mediated or humoral immune response.

The Ly-49 family: regulation of cytokine production in murine NK cells

Most proteins encoded by members of the Ly-49 gene family are class I-recognizing receptors on murine natural killer (NK) cells. Class I recognition by Ly-49 receptors usually results in inhibition of NK cell lysis of target cells. However, NK cells function not only in a lytic capacity, but also can mediate cytokine production. In this report we have demonstrated the ability of Ly-49A and Ly-49G2 to inhibit production of cytokines by NK cells by showing that specific antibodies against these gene products stimulate cytokine production. Murine NK cells were cultured in the presence of P815 (H2-Dd), and supernatants were analyzed for the production of interferon-gamma (IFN-gamma), tumor necrosis factor alpha (TNF-alpha), and granulocytemacrophage colony-stimulating factor (GM-CSF). NK cell populations were sorted for Ly-49A+ or Ly-49G+ subsets, and these subsets were analyzed for their ability to alter cytokine induction by target cell interaction. In the presence of target cells expressing the appropriate class I molecules, Ly-49A and G2 were found to inhibit cytokine induction by NK cells. Examination of mRNA for IFN-gamma and GM-CSF indicated that Ly-49 receptors increased mRNA levels of NK cells. These results demonstrate that class I binding of these NK receptors can inhibit production of important physiological cytokines, in addition to the regulation of cytotoxic activity.

Fas involvement in human NK cell apoptosis: lack of a requirement for CD16-mediated events

Propriocidal regulation of T cells refers to apoptosis induced by interleukin-2 (IL-2) activation with subsequent antigen receptor stimulation. We previously reported that natural killer (NK) cells also exhibit propriocidal death. Cell death can be induced following occupancy of the Fc gamma RIII (CD16) receptor when NK cells were pretreated with IL-2, IL-12, or IL-15. Here we show other triggering receptors on NK cells such as CD44, anti-NK-receptor antibodies, and pharmacological activation can result in the cell death signal. Requirement for cell interactions indicated that cell contact was required; however, unlike cell-mediated lysis, extracellular calcium was not required. Like T cells, the process of cell death for NK cells was receptor-induced apoptosis. Activation-induced apoptosis of T cells is mediated by members of the tumor necrosis factor (TNF) cytokine superfamily. We examined the involvement of TNF receptor family members or Fas in this rapid cell death. Antibody directed against Fas, TNFR60, TNFR80, LTBR, and LT alpha failed to inhibit receptor-induced death. Therefore, NK cells appear to demonstrate a rapid apoptotic episode when CD16 is cross-linked, but the mechanism of this apoptosis is quite different than was observed in T cells with CD3. The direct examination of the Fas pathway on activated NK cells revealed that susceptibility required longer treatment times and IL-2 activation. This susceptibility was paralleled by increased Fas-ligand expression. Therefore, NK cells can demonstrate an apoptotic response to CD16, CD44, NK receptors, and Fas. The enumeration of ligands capable of eliciting NK cell death and the in vivo relevance of this observation require further study.

Selective inhibition of human and mouse natural killer tumor recognition using retroviral antisense in primary natural killer cells: involvement with MHC class I killer cell inhibitory receptors

The natural killer tumor recognition (NK-TR) protein has been shown to be a necessary component for the killing of NK-sensitive and virus-infected targets by the rat RNK-16 cell line. Class I-recognizing killer cell inhibitory receptors (KIR) have been found in the human (p58; NKAT family) and mouse (Ly-49 family). The principal functional characteristic of these receptors is their ability to block NK cell lysis by recognition of selected class I molecules on target cells. In the present study, we examined whether abrogation of NK-TR expression by retroviral infection of primary human or mouse NK cells with virus-producing antisense NK-TR also would demonstrate loss of non-MHC-restricted killing and whether the NK-TR was associated with KIR function in humans or with Ly-49 in the mouse. Using short term culture of fresh human or mouse NK cells, antisense NK-TR-treated NK cells demonstrated strong selective reduction of NK cytotoxicity. NK-TR was necessary for lytic activity even when KIR function was blocked by Ab in experiments involving NK3.3 lysis of HLA.cw3-expressing targets or killing of Dd targets by Ly-49A+ or Ly-49G2+ mouse NK cells. These studies extend our previous studies in rat NK cell lines to demonstrate that primary mouse and human NK cells require NK-TR for non-MHC-restricted lysis of tumor and virus-infected targets. In addition, the reversal of KIR or Ly-49 inhibition of NK cell lysis requires NK-TR expression for cellular killing in both human and mouse.

Selective inhibition of human and mouse natural killer tumor recognition using retroviral antisense in primary natural killer cells: involvement with MHC class I killer cell inhibitory receptors

The natural killer tumor recognition (NK-TR) protein has been shown to be a necessary component for the killing of NK-sensitive and virus-infected targets by the rat RNK-16 cell line. Class I-recognizing killer cell inhibitory receptors (KIR) have been found in the human (p58; NKAT family) and mouse (Ly-49 family). The principal functional characteristic of these receptors is their ability to block NK cell lysis by recognition of selected class I molecules on target cells. In the present study, we examined whether abrogation of NK-TR expression by retroviral infection of primary human or mouse NK cells with virus-producing antisense NK-TR also would demonstrate loss of non-MHC-restricted killing and whether the NK-TR was associated with KIR function in humans or with Ly-49 in the mouse. Using short term culture of fresh human or mouse NK cells, antisense NK-TR-treated NK cells demonstrated strong selective reduction of NK cytotoxicity. NK-TR was necessary for lytic activity even when KIR function was blocked by Ab in experiments involving NK3.3 lysis of HLA.cw3-expressing targets or killing of Dd targets by Ly-49A+ or Ly-49G2+ mouse NK cells. These studies extend our previous studies in rat NK cell lines to demonstrate that primary mouse and human NK cells require NK-TR for non-MHC-restricted lysis of tumor and virus-infected targets. In addition, the reversal of KIR or Ly-49 inhibition of NK cell lysis requires NK-TR expression for cellular killing in both human and mouse.

Bone marrow and thymus expression of interferon-gamma results in severe B-cell lineage reduction, T-cell lineage alterations, and hematopoietic progenitor deficiencies

Interferon-gamma (IFN-gamma) is an immunoregulatory lymphokine that is primarily produced by T cells and natural killer cells. It has effects on T-cell, B-cell, and macrophage differentiation and maturation. We have developed transgenic mice that express elevated levels of IFN-gamma mRNA and protein by inserting multiple copies of murine IFN-gamma genomic DNA containing an Ig lambda-chain enhancer in the first intron. The founder line carrying eight copies of this transgene has eightfold to 15-fold more IFN-gamma-producing cells in the bone marrow and spleen than do nontransgenic littermates. Transgenic mice show a pronounced reduction in B-lineage cells in the bone marrow, spleen, and lymph nodes. In addition, single positive (CD4+,CD8- and CD4-,CD8+) thymocyte numbers are increased twofold, yet the number of splenic T cells is reduced by 50%. There is also a twofold to threefold decrease in the frequency and total number of myeloid progenitors in the bone marrow. Granulomatous lesions and residual degenerating cartilaginous masses are also present in the bones of these mice. Overall, our data show that the abnormal expression of IFN-gamma in these transgenic mice results in multiple alterations in the immune system. These animals provide an important model to examine the role of IFN-gamma expression on lymphoid and myeloid differentiation and function.

Characterization of a non-granule associated pore-forming protein in agranular lymphocytes

Recently, two populations of small lymphocytes (SL) have exhibited non-major histocompatibility complex (MHC) restricted lysis. Recent studies by numerous laboratories have demonstrated that resting T cells triggered through CD3 and CD28 costimulations can result in immediate, non-MHC restricted killing. Our recent studies with CD3-, CD56+ SL demonstrated that although these cells contained no cytoplasmic granules detected with electron microscopy, they mediated potent NK and ADCC activity. In the present study, we have used a Western blotting technique that allows for the detection and quantitation of total cellular levels of pore-forming protein (PFP). We have found that freshly isolated peripheral non-granulated lymphocytes (both CD3+ and CD3-) contain PFP. In addition, CD3-, CD56+ SL contain levels of PFP similar to those of the highly granular CD3- LGL. In search of non-granule PFP, we exploited the rat NK (RNK) cell lines as a source of other potential cytotoxic factors. A membrane associated PFP, based on Western blotting, was isolated from rat RNK cells. Unlike PFP isolated from granules, this PFP was active after culture in Ca(2+)-containing medium. However, the lytic activity isolated from the non-granule PFP of RNK cells was inhibited by monoclonal antibodies to PFP. Collectively, these studies indicate that PFP is present in small agranular lymphocytes (both NK and T cells) and that it is not stored in large cytoplasmic granules. The implication of our results for the acquisition and activation of lytic ability in NK and T cells will be discussed.

CD3+, CD56+ non-MHC restricted cytotoxic T lymphocytes in two fraternal AIDS patients: a case report

We describe two brothers with hemophilia and AIDS, with an unusually large percentage of CD3+, CD56+ lymphocytes. They experienced no major complications with opportunistic infections (OI) and infrequent secondary infections, even though they had nearly 0% CD4 lymphocytes for 3 years. Both patients died in 1991 of progressive cardiomyopathy. The patients' lymphocytes were immunophenotyped by flow cytometry and analyzed for functional cytolytic activity against K562 and HIV infected HUT 78 cell lines. Single color CD4 counts were 2-9% for 4 years. Dual color CD4 counts at our laboratory demonstrated 0-1% CD4 for the last 6 months. When CD3+ lymphocytes were examined, both patient 1 and patient 2 demonstrated a significantly higher proportion and absolute number of CD3"bright"+, CD56+ double-positive cells, 47% and 22%, respectively, compared to other HIV-positive children with hemophilia (< or = 2%). Functional studies with the K562 target cell line demonstrated the highest natural killer (NK) lymphocyte activity in patient 1 that could not be augmented by in vitro addition of IL-2, whereas patient 2 had no NK activity unless IL-2 was added. Functional studies with HIV-infected HUT-78 cells demonstrated patient 2 had cytolytic activity against HIV-infected cells and patient 1 had high nonspecific cytolytic activity even against uninfected HUT-78 cells, whereas controls had minimal activity to HUT-78 cells or HIV-infected HUT-78 cells. The case report raises a speculative question requiring a larger database, whether the anti-HIV activity and/or unusual clinical course without typical O.I. of some AIDS patients may be related to the presence of CD3"bright"+, CD56+ lymphocytes of their immune system.

Direct evidence for release of pore-forming protein during NK cellular lysis

In the present study, we have examined the exocytosis model for cellular lysis. Using monoclonal antibodies reactive with human pore-forming protein (PFP), we examined the localization of PFP at the interaction site of natural killer (NK) cells and the NK tumor targets K562 and Molt-4 as well as during antibody-dependent cellular cytotoxicity. Following the interaction of effector-target cell contact, an increased frequency of PFP was detected on the effector surface, in the micro-environment, and on the target surface of the interaction site. This temporal deposition of PFP was paralleled by loss of target cell integrity and release of chromium. In addition, selective deposition of PFP was seen at the interaction site of the target cell compared to other target cell regions. Collectively, these results are consistent with the exocytosis model and further support the hypothesis that PFP is one of the secreted moieties involved in NK cellular cytotoxicity.

Comparative studies of CD3- and CD3+ CD56+ cells: examination of morphology, functions, T cell receptor rearrangement, and pore-forming protein expression

Both CD3- and CD3+ CD56+ effector cells can mediate non-MHC-restricted lysis in the absence of activation. Previous studies have shown that both of these subsets can be augmented with IL-2. In the present study, we have examined further the phenotypic markers expressed on these cells as well as the functional capacities of these subsets, including LAK activity, cytokine expression, and pore-forming protein (PFP) production. In addition, these populations were analyzed for clonality by Southern blot analysis of the T cell receptor beta chain gene constant region. The CD3-, CD56+ and CD3+, CD56+ lymphocytes were quite similar in their phenotypic markers, although the CD3+, CD56+ lymphocytes lacked high levels of IL-2 receptor beta chain and did not express CD16. The CD3+, CD56+ lymphocytes mediated non-MHC-restricted lysis, but failed to express LAK activity or be induced by IL-2 to secrete IFN gamma, a characteristic of the CD3-, CD56+ lymphocytes. The T cell receptor beta chain gene pattern of the CD3+, CD56+ lymphocytes was characteristic of a polyclonal cell population. Of interest, both populations of cells appeared morphologically to be large granular lymphocytes that contain PFP in their cytoplasmic granules. Therefore these CD56+ subsets provide a new model to study several questions related to non-MHC-restricted target cell lysis, including the identification of novel receptors involved in target cell recognition and/or triggering as well as the biochemical pathways implicated in cellular lysis.

CD3 negative "small agranular lymphocytes" are natural killer cells

We describe here that CD3-, CD16+ and/or CD56+ small lymphocytes, in a highly reproducible fashion, mediate a significant level of K562 killing that is, on a "per cell" basis, comparable to the cytolytic activity of CD3- LGL. The CD3- small lymphocytes appeared to have no granules based on light and electron microscopy and lack of right-angle scatter on the FACS; we thus refer to them as small "agranular" lymphocytes (SAL). The lytic activity against K562 is inhibited by treatment with either L-leucine methyl ester or EGTA, which are reported to effect granule-dependent killing. We suggest that the SAL have lytic molecules in their cytoplasm (which are sensitive to these treatments) but that these molecules are not organized into discrete granules as found in LGL. The CD3- SAL are phenotypically very similar to LGL and both SAL and LGL mediated equal and reproducible antibody-dependent cell-mediated cytotoxicity. These observations force redefinition of the concept of NK cells to include both CD3- LGL and CD3- SAL.

Analysis of rat natural killer cytotoxic factor (NKCF): mechanism of action and relationship to other cytotoxic/cytostatic factors

Natural killer cytotoxic factor (NKCF) has been proposed as one of the factors that mediates lysis induced by natural killer (NK) cells. Recently, an excellent source of NKCF has been found to be the rat large granular lymphocyte (LGL) tumor (RNK) cell line. In this study, the kinetics of lysis of the NK-sensitive, tumor target YAC-1 by the RNK-NKCF was analyzed and found to parallel that seen with NK cell-mediated killing. RNK-NKCF was also capable of killing the NK-resistant target cell, MBL-2, over a longer time period. This study utilized monoclonal antibodies (mAbs) prepared against granule protein, previously termed "anti-NKCF mAbs." These mAbs established the nature of RNK-NKCF as compared to other known cytotoxic factors in combination with studies that show that RNK-NKCF causes both 51Cr release and nuclear degradation. Antibody inhibition experiments have verified that RNK-NKCF is unique from tumor necrosis factor (TNF), leukoregulin, or complement. Anti-NKCF mAbs were capable, however, of neutralizing the RNK cell granule activity against YAC-1 tumor target cells. Based on these results, the ability of anti-NKCF mAbs to neutralize the cytolytic function of pore-forming protein (PFP), a component of these granules, was analyzed. In these experiments, the antibodies were found to inhibit the hemolytic activity of granules. Interestingly, the antibodies were effective in inhibiting the activity of unbound granule proteins as well as those bound to sheep red blood cell (SRBC) targets. Further studies to examine the target lysis requirements demonstrated that in contrast to PFP, the RNK-NKCF was able to lyse the tumor target in the absence of calcium. In addition, treatment of targets with RNA and protein synthesis inhibitors indicated that the mechanism of lysis of NKCF is quite unique from other defined cytotoxic moieties.

CD3 negative "small agranular lymphocytes" are natural killer cells

We describe here that CD3-, CD16+ and/or CD56+ small lymphocytes, in a highly reproducible fashion, mediate a significant level of K562 killing that is, on a "per cell" basis, comparable to the cytolytic activity of CD3- LGL. The CD3- small lymphocytes appeared to have no granules based on light and electron microscopy and lack of right-angle scatter on the FACS; we thus refer to them as small "agranular" lymphocytes (SAL). The lytic activity against K562 is inhibited by treatment with either L-leucine methyl ester or EGTA, which are reported to effect granule-dependent killing. We suggest that the SAL have lytic molecules in their cytoplasm (which are sensitive to these treatments) but that these molecules are not organized into discrete granules as found in LGL. The CD3- SAL are phenotypically very similar to LGL and both SAL and LGL mediated equal and reproducible antibody-dependent cell-mediated cytotoxicity. These observations force redefinition of the concept of NK cells to include both CD3- LGL and CD3- SAL.