A comparison of antibody-dependent cellular cytotoxicity (ADCC) mediated by murine and human lymphoid cell populations

Characterization of IL-2-induced murine cells which exhibit ADCC activity

The incubation of murine splenocytes in recombinant interleukin 2 (IL-2) gives rise to both lymphokine-activated killer (LAK) cells capable of lysing fresh tumor cells and cells capable of mediating antibody-dependent cellular cytotoxicity (ADCC) in the presence of anti-H2 allosera. A similarity between these two IL-2-induced cell populations was found. The precursors of the cells mediating these activities were shown to be ASGM1 positive, Thy 1 negative, and radiosensitive. Cells taken from the spleen, thymus, and bone marrow were able to mediate ADCC after culture in IL-2. The effector cell was either Thy 1 positive or negative and was less affected by anti-Thy 1 plus C' treatment than cells which mediated LAK activity. In addition ADCC was exhibited in IL-2-cultured splenocytes from various murine strains and correlated with their LAK activity with one exception. While IL-2-cultured C57BL/6 splenocytes exhibited LAK activity similar to that of C3H LAK cells, no ADCC activity could be demonstrated in C57BL/6 cells. Study of the difference in the ability of these two strains to mediate ADCC revealed that IL-2-induced FcR+ cells in C3H thymocytes, but not in C57BL/6 thymocytes. Based on FACS analysis and on the radiosensitivity of the induction of both FcR+ cells and ADCC, it was suggested that IL-2 was expanding a small FcR+ cell population rather than inducing an increase in FcR density on the cell surface. The relationship between the IL-2-induced ADCC mediator and other IL-2-induced cells, as well as ADCC effector cells, and the possible implications of the results for the in vivo therapy of cancer based on ADCC are discussed.

Higher ADCC of murine peritoneal cells after immunization with allogenic tumor cells as compared with stimulation by adriamycin, BCG, and thioglycolate

Normal peritoneal cells or spleen cells from C57BL mice could not lyse SRBC in an ADCC assay. After intraperitoneal injection of Adriamycin, BCG or thioglycolate the ADCC of peritoneal cells toward antibody-coated SRBC was elevated to 30% in contrast to the ADCC of spleen cells. However, peritoneal cells but not spleen cells of mice immunized with allogenic tumor cells (DBA SL2) showed ADCC levels at least two times higher than the levels observed after stimulation by other agents. Maximal ADCC levels (55.8%) were observed 10 to 15 days after immunization. Direct cytotoxicity towards SRBC increased to a maximum of 17.7% at 9 days after immunization. The effector cells in this system are thought to be macrophages, for ADCC activity was only present in the plastic-adherent cell fraction. Cell to cell contact was necessary for ADCC to occur; nonsensitized erythrocytes were not lysed when added to a mixture of effector cells and sensitized erythrocytes. Concentrations of antibody of 1 pg/ml were sufficient to induce ADCC, and effector cell to target cell ratios could be as low as 0.05. The finding that macrophages of mice immunized with allogenic tumor cells exhibit higher ADCC levels than macrophages elicited in other ways can contribute to the investigation of combined cancer therapy with antibodies and biological response modifiers.

Modulation of nonspecific cell-mediated growth inhibition by estrogen metabolites

Chronic exposure of mice to estrogens such as 17-beta estradiol and diethylstilbestrol inhibits natural killer cell-mediated cytotoxicity in vivo. In this report, we investigated the direct in vitro effects of 17-beta estradiol and its major metabolites on nonspecific effector cell function measured as the ability of naive lymphocytes to inhibit the growth of the YAC-1 lymphoma, a classical natural killer-sensitive target cell. Without exception, the effects of individual estrogen metabolites on the growth inhibitory properties of these cells were accompanied, at every concentration of compound, by identical effects on the blastogenic response of lymphocytes to the T cell lectin phytohemagglutinin. These observations suggested membrane-mediated immunomodulation of lymphocyte function by estrogen metabolites. As suggested by previous studies with quinone metabolites of benzene, the catechol estrogen metabolite 2-OH estrone was significantly more potent than the parent compound at suppressing lymphocyte functions in vitro; however, dosing regimens of 2-OH estrone that suppressed blastogenic response in vivo failed to inhibit nonspecific cell-mediated growth inhibition.

Inability to detect a K cell in bovine peripheral blood leukocytes

Antibody-dependent cellular cytotoxicity to viral-infected cells, chicken red blood cells, and tumor cells was tested using different effector cell populations: polymorphonuclear cells, mononuclear cells, and mononuclear cells separated into adherent and nonadherent populations by Sephadex G-10. Polymorphonuclear cells were the most efficient mediators of antibody-dependent cellular cytotoxicity against most targets, although a combination of G-10 adherent and polymorphonuclear cells was more efficient in killing infectious bovine rhinotracheitis virus-infected cells than either single cell population. Removal of G-10 adherent cells from the mononuclear cell population removed all antibody-dependent cellular cytotoxicity from that population, indicating the lack of a typical K cell in bovine peripheral blood.

Production of target-specific effector cells using hetero-cross-linked aggregates containing anti-target cell and anti-Fc gamma receptor antibodies

Rabbit anti-2,4-dintrophenyl (DNP) antibodies or their F(ab')2 fragments were chemically cross-linked to the anti-mouse Fc gamma R monoclonal antibody 2.4G2 or to its Fab fragment. P388D1 cells were incubated with heteroaggregates between 2.4G2 and anti-DNP (anti-Fc gamma R X anti-DNP) and washed. The resulting cells lysed 2,4,6-trinitrophenyl chicken erythrocytes (TNP CRBC) in a hapten-specific manner. The lysis was inhibited by free hapten but was resistant to inhibition by immune complexes. Other cells coated with antibody heteroaggregates also mediated lysis of TNP-modified target cells. For example, mouse resident peritoneal exudate cells (PEC) lysed TNP CRBC and bacillus Calmette-Guérin-activated PEC lysed both TNP CRBC and TNP tumor targets. Human neutrophils, when incubated with heteroaggregates containing the anti-human neutrophil Fc gamma R antibody 3G8 and anti-DNP also lysed TNP CRBC and TNP-modified tumor cells. To test whether linkage to Fc gamma R was required for lysis, F(ab')2 fragments from the anti-KdDd monoclonal antibody 34-1-2 were cross-linked to anti-DNP F(ab')2 fragments. P388D1 cells (which express Kd and Dd) were then incubated with these heteroaggregates and washed, and their abilities to form conjugates and lyse TNP CRBC were compared with those of P388D1 cells treated with anti-Fc gamma R X anti-DNP. In both cases, P388D1 cells formed conjugates. However, only the cells treated with anti-Fc gamma R X anti-DNP mediated lysis to a significant extent. We conclude that heteroaggregates containing anti-Fc gamma R and anti-target cell antibodies can be used to create potent effector cells against red cell and tumor targets and that bridging of effectors with target cells directly to Fc gamma R on effector cells is required for lysis.

Modulation of antitumoral antibody-dependent cellular cytotoxicity and natural killer activity by Adriamycin and daunorubicin

Mouse effector cells isolated from various anatomical sources failed to mediate antibody-dependent cellular cytotoxicity (ADCC) against alloantiserum-coated L1210 murine leukemia cell targets, whereas rat spleen cells appeared to be potent mediators of this activity. Following suppression of effector cell function 3 days after a single drug injection, the nylon wool non-adherent population of rat spleen cells from daunorubicin (DM)-treated rats demonstrated an increased ability to mediate ADCC compared to controls. Alternatively, although suppression occurred at day 7, no rebound enhancement was demonstrated by the same cell population isolated from Adriamycin (AM)-treated animals for as long as 12 days post-injection. Natural killer (NK) activity, measured as the ability of the nylon wool non-adherent rat spleen cell population to lyse uncoated L1210 cells, was modulated by drug treatment in a similar manner at each time point although the changes were not significant. In contrast to NK cells for which a substantial amount of activity remained adherent to nylon wool, all K cell activity was found in the non-adherent spleen cell population. The effector cell, in both cases, was not susceptible to antithymocyte serum plus complement treatment; however, NK activity appeared trypsin-sensitive whereas K cell activity did not. Therefore, AM and DM demonstrated different activities with regard to in vivo modulation of antitumoral ADCC.

Immunotherapy of murine leukemia. VII. Prevention of Friend leukemia virus-induced immunosuppression by passive serum therapy

Previous studies have suggested that the passive therapy of Friend leukemia virus (FLV)-induced disease with chimpanzee anti-FLV serum operates by reducing the level of infectious virus in the treated animal below the immunosuppressive threshold, thereby allowing the host to mount anti-viral immune responses which are responsible for long-term protection. The present study was undertaken to examine directly the effect of passive serum therapy on the marked immunosuppression induced by FLV in progressively infected mice, as well as to determine whether virus-specific host cellular immune effector functions are augmented in serum-protected animals. Using a variety of assays of host immunocompetence, including natural killing (NK), antibody-dependent cellular cytotoxicity (ADCC) in vivo and in vitro induction of allogeneic killers, and mitogen blastogenesis, a marked compartmentalization of FLV immunodepression was observed in progressively infected DBA/2 mice, possibly reflecting the distribution of FLV target cells in various host lymphoid populations. Thus, spleen-cell functions were suppressed most rapidly and to the greatest degree, followed by peritoneal cells and peripheral blood lymphocytes, while lymph node cells and thymocytes maintained normal levels of activity. In contrast, serum-protected mice demonstrated no sign of FLV-induced immunosuppression regardless of the host effector-cell population or immune function examined. However, we were not able to identify host anti-viral cellular immune functions which are significantly enhanced in serum-protected animals; thus the specific role of the host immune system in the passive serum therapy of FLV-induced disease remains undefined at the present time.

Failure to detect killer cell activity in rabbits

Rabbit lymphoid cells from spleen, peripheral blood, and peritoneal cavity lacked killer (K)-cell activity against cell lines of rabbit and human origin, including virus-infected human tumor cells. This lack of activity was not affected by antibody concentration, source of antibodies, effector/target cell ratio, or length of assay. Rabbit leukocytes, however, were capable of lysing antibody-coated chicken erythrocytes. Hamster leukocytes, serving as a known source of K cells, mediated antibody-dependent cellular cytotoxicity against all targets. EA-rosette assays and mixed effector cell competition tests suggested a deficiency in rabbit K-cell activity which is not a result of an inherent lack of Fc receptor-positive cells or of some suppressor mechanism operating in the rabbit cell populations. Our data support the concept that antibody-dependent cellular cytotoxicity may not be a significant in vivo immune mechanism in certain species.

Modulation of antibody-dependent cellular cytotoxicity against chicken erythrocyte targets by adriamycin and daunorubicin

Peritoneal exudate cells isolated from Adriamycin (AM) or daunorubicin (DM)-treated mice demonstrated an increased ability to mediate antibody-dependent cellular cytotoxicity (ADCC) against antibody-coated chicken red blood cell targets. Following initial suppression of this cell-mediated function 3 days after a single injection, increased effector cell efficiency occurred to an equal degree in both groups of drug treated mice by day 10 compared to controls. This increase in ADCC activity occurred in parallel with a decrease in the total number of peritoneal cells recovered. It is hypothesized that the drugs acted to modulate ADCC in two ways: 1) suppression by induction of suppressor cell activity, and 2) enhancement by elimination of suppressor cells which resulted in increased effector activity of the remaining cells.

Monoclonal antibodies in cell-mediated cytotoxicity against human melanoma and colorectal carcinoma

Hybridoma-derived monoclonal anti-melanoma antibodies and anti-colorectal carcinoma antibodies were found to mediate in vitro antibody dependent cell-mediated cytotoxicity (ADCC) reactions against melanoma and colorectal carcinoma cells, respectively. The antigen(s) detected in ADCC on melanoma cells maintained for more than one hundred passages in tissue culture were also found on two recently established melanoma cell lines. These antigens were not detected on skin fibroblasts of the same patients from whom the melanomas were obtained. The ADCC reactivities of anti-melanoma and anti-colorectal carcinoma antibodies were found to be specific for melanoma cells and colorectal carcinoma cells, respectively.