Lysis of tumor cells by human blood monocytes by a mechanism independent of activation of the oxidative burst.

Tumorilytic human blood monocytes recognize and destroy neoplastic cells by a mechanism that is nonphagocytic and requires cell-to-cell contact. The mechanism of cytolysis subsequent to binding is controversial. Release of reactive oxygen intermediates by activated rodent macrophages has been suggested as an important mechanism for tumor cell lysis in some short-term cytotoxicity assays. We examined whether oxygen intermediates are also responsible for mediating the lysis of adherent human tumor cells in a long-term (72-h) tumoricidal assay. Human blood monocytes were incubated with medium, concanavalin A-stimulated lymphokine [macrophage-activating factor (MAF)], lipopolysaccharide endotoxin, or human recombinant gamma interferon for 24 h prior to the addition of [125I] iododeoxyuridine-labeled A375 melanoma cells. The following evidence indicated that monocyte-mediated tumor cell lysis was independent of superoxide anion (O2-) and H2O2 production: (a) although human blood monocytes incubated for 24 h with gamma interferon produced twice as much O2- as control or MAF-treated monocytes, gamma interferon did not activate monocyte tumoricidal activity unless combined with lipopolysaccharide endotoxin, 0.2 ng/ml or more; (b) incubating the monocytes with 10 nM phorbol myristate acetate for 0.5 h stimulated O2- production but no cytotoxicity; (c) the cytolytic activity of MAF-treated monocytes was not decreased in the presence of catalase or superoxide dismutase; and (d) finally, peripheral blood monocytes were isolated from six patients with chronic granulomatous disease, activated by MAF or lipopolysaccharide endotoxin, and then assayed for tumoricidal activity. While these activated chronic granulomatous disease monocytes did not produce O2- or H2O2, tumor cell lysis was normal in all six patients. Hence, lysis of tumor cells in a 72-h assay is not dependent upon the generation of O2- and/or H2O2 and is intact in chronic granulomatous disease monocytes.

Effect of monoclonal antibodies (MoAb) to class I and class II HLA antigens on lectin- and MoAb OKT3-induced lymphocyte proliferation.

We have examined the effect of several monoclonal antibodies (MoAb) to monomorphic determinants of class II HLA antigens, and MoAb to monomorphic determinants of class I HLA antigens and to beta-2-microglobulin (beta 2-mu) on lectin- and MoAb OKT3-induced proliferation of human peripheral blood mononuclear cells (PBMNC) and cultured T cells (CTC). Some, but not all, anti-class II HLA MoAb inhibited the proliferative response of PBMNC to MoAb OKT3 and pokeweed mitogen (PWM). The degree of inhibitory effect varied considerably. This effect was not limited to anti-class II HLA MoAb since anti-class I HLA MoAb and anti-beta 2-mu MoAb also inhibited MoAb OKT3- or PWM-induced proliferative responses. In contrast, the response of PBMNC to phytohemagglutinin (PHA) and concanavalin A (Con A) was not blocked by any anti-class II HLA MoAb. However, some anti-class II HLA MoAb also inhibited the proliferative response of CTC plus allogeneic peripheral blood adherent accessory cells (AC) to PHA or Con A as well as to MoAb OKT3 or PWM. This may be attributable to the substantially greater class II HLA antigen expression by CTC than by fresh lymphocytes. Pretreatment of either CTC or AC with anti-class II HLA MoAb inhibited OKT3-induced proliferation. In contrast, pretreatment of CTC, but not AC, with anti-class I HLA MoAb inhibited the proliferative response of CTC to OKT3. Pretreatment of CTC with anti-class I HLA MoAb inhibited PHA-, Con A and PWM-induced proliferation, to a greater degree than the anti-class II HLA MoAb. It appears as if lymphocyte activation by different mitogens exhibits variable requirements for the presence of cells expressing major histocompatibility determinants. Binding of Ab to membrane markers may interfere with lymphocyte-AC cooperation, perhaps by inhibiting binding of mitogens to their receptors or by interfering with lymphocyte and AC function. We also have examined the role of class II HLA antigens on CTC by depleting class II HLA-positive cells. As expected, elimination of class II HLA-positive AC with anti-class II HLA MoAb plus complement caused a decrease in proliferation of CTC in response to all the mitogens tested. In contrast, elimination of class II HLA-positive CTC was shown to clearly increase proliferation of CTC, perhaps because this may deplete class II HLA-positive suppressor cells.

Human lymphokine preparations which generate tumoricidal properties of human monocytes in vitro may be distinct from gamma interferon.

The purpose of these studies was to determine whether stimulated human lymphocytes produce lymphokines distinct from IFN gamma, that can activate human blood monocytes to lyse tumor cells. We undertook this investigation because of the controversy concerning whether MAF and IFN gamma are the same molecule. Crude lymphokine preparations prepared from normal human mononuclear cells incubated with Con A and rich in MAF activity also contained 1000 U/ml IFN gamma as measured by the virus neutralization assay. However, the induction of tumoricidal activity in monocytes by the lymphokine preparation could be dissociated from the IFN gamma activity, based on the following data: (1) Heat treatment (100 degrees C for 2 min) removed the antiviral activity of the lymphokine yet did not diminish its MAF-like activity when measured in a 72 h cytotoxicity assay against 125I IUdR-labeled human A375 melanoma cells. (2) Likewise, treatment of this lymphokine preparation with a twofold excess of anti-IFN gamma antibody neutralized antiviral activity but once again had no effect on its ability to activate monocyte tumoricidal function. In contrast, both heat treatment and anti-IFN gamma antibody abolished monocyte activation by equivalent units of human recombinant IFN gamma. Taken together, these data suggest that there is a molecule(s) distinct from IFN gamma which can activate human monocytes for tumoricidal function. Furthermore, this dissociation of MAF and IFN gamma activity was dependent on the use of a long-term (72 h) assay, since activation of tumoricidal activity in an 18-24 h assay appeared to be attributable solely to IFN gamma.

HTLV-I infection of T and B cells of a patient with adult T-cell leukemia-lymphoma (ATLL) and transmission of HTLV-I from B cells to normal T cells.

We analysed the DNA of different tissues of a patient (HS) with adult T-cell leukemia/lymphoma virus (HTLV-I). We detected viral sequences in fresh specimens from spleen, thymus, liver, skin and peripheral blood neoplastic lymphocytes. The pattern of HTLV-I integration is identical in the leukemic cells and in all other tissues analysed, but the signal intensity is strongest in the leukemic cells, indicating the source of HTLV-I proviral sequences was the leukemic T-cells which had infiltrated these tissues. In fact, the cultured skin fibroblasts of the patient did not contain HTLV-I sequence. However, cultured lymphocytes of this patient was consistently an immortalized B-cell line containing HTLV-I sequences in a manner indicative of a polyclonal infection. This cell line was also infected with the Epstein-Barr virus (EBV). In order to determine whether HTLV-I alone was sufficient for B-cell immortalization, we obtained single cell clones by limiting dilution. The DNA of all the cell clones that we analysed contained both the HTLV-I and EBV genomes, suggesting that immortalization of the B-cell was more likely due to the EBV rather than HTLV-I. Infectious HTLV-I viruses produced by the B-cell line still had the propensity to infect and transform T-lymphocytes in normal human umbilical cord blood. Unlike the parental B cells, the transformed T lymphocytes were clonally selected. Our results indicate that although the predominant infected cell population of the patient was his leukemic T lymphocytes, some of his EBV-positive B-lymphocytes were also polyclonally infected. The latter had a growth advantage in culture over the T lymphocytes but the virus produced by these immortalized B cells has not been adapted and has maintained its tropism for T cells.

Constitutive production and release of a lymphokine with macrophage-activating factor activity distinct from gamma-interferon by a human T-cell leukemia virus-positive cell line.

Culture supernatant fluids from the human T-cell leukemia virus-positive cell line C10/MJ-2 were found to contain a soluble factor with macrophage-activating factor (MAF) activity. The MAF activity of this culture supernatant fluid was stable at 100 degrees for 2 min and was unaffected by treatment with human anti-gamma-interferon (IFN-gamma) monoclonal antibody. Both treatments neutralized greater than 97% IFN-gamma activity from the supernatant fluid as measured by virus neutralization. Furthermore, this MAF activity was not due to contamination with endotoxins since the Limulus amebocyte lysate assay was negative (less than 0.125 ng/ml) and preincubation of the C10/MJ-2 supernatant fluid with polymyxin B did not diminish its activating potential. By contrast, human IFN-gamma rendered human monocytes tumoricidal only when combined with Salmonella typhosa lipopolysaccharide (LPS) at a minimum dose of 0.2 ng/ml. The concentrations of both LPS and IFN-gamma were crucial to achieve activation since IFN-gamma at doses less than 10 units/ml did not activate human monocytes even when combined with maximal doses of LPS (0.5 ng/ml). Finally, when human IFN-gamma admixed with LPS was preincubated with polymyxin B, its activating potential was completely abrogated. Collectively, these data suggest that the human T-cell line C10/MJ-2 constitutively produces a diffusable product distinct from IFN-gamma which activates human monocytes to lyse tumor cells. Thus, this cell line could provide a good source of a unique human MAF for future purification procedures.