Production of an anti-tumour cytotoxin by human monocytes: comparison of endotoxin, interferons and other agents as inducers.

Production of an anti-tumour cytotoxin by human monocytes: Comparison of endotoxin, interferons and other agents as inducers

The mechanism of macrophage-induced tumour-cell stasis and cytotoxicity are unclear. In some assays direct macrophage tumour-cell contact appears necessary (Stewart et al., 1976); in others the antitumour effects are mediated by factors released from the macrophages. Such factors include arginase (Currie, 1978), possibly the complement breakdown product, C3a (Ferluga et al., 1978; but see Goodman et al., 1980), proteolytic enzymes (Adams, 1980) and the tumour-necrosis factors (TNFs) of mice and rabbits (Mannel et al., 1980;Matthews, 1978Matthews, , 1981a). Recently we have described an anti-tumour cytotoxin which is synthesized by human monocytes on endotoxin challenge in vitro (Mathews, 1981b). This cytotoxin is not an arginase, proteolytic enzyme or C3a, but does resemble rabbit TNF.
The aim of this study is to determine whether the human monocyte cytotoxin can be induced by interferons or other macrophage-activating agents.
Monocytes.-The method of isolation was essentially that described for rabbit monocytes (Matthews, 1978). Mononuclear cells, isolated from the heparinized blood of healthy donors using Hypaque-Ficoll, were washed x 2 and suspended at 5 x 106/ml in Eagle's minimum essential medium supplemented with 10% foetal calf serum (MEM/FCS). Volumes of 0 4 ml were incubated for 11 h at 37°C in 5% C02, 95% air in the wells of plastic Linbro plates (24 wells, 2 cm2 in area). Non-adherent cells were removed by 2 washes with warm medium leaving preparations of > 80% monocytes. Dilutions of test substances (0.4 ml in MEM/FCS) were added and the cells were incubated at 37 TC for 20 h. Supernatants were collected, centrifuged and stored at -70°C until assay.
None of the materials was itself toxic to the target cells at the concentrations used.
Cytotoxin assay.-Mouse L929 target cells in micotitre trays were used as described in detail in Matthews (1981b). Briefly, monocyte supernatant dilutions (3 dilutions in triplicate) were incubated with a monolayer of L929 cells for 20 h in the presence of 1 ,ug/ml actinomycin D. Dead cells detached from the plastic and the adherent viable cells were stained with crystal violet. The amount of dye bound is proportional to the number of viable cells in the well, and was quantitated photometrically using a Titertek Multi-skan photometer. The percentage cytotoxicity was calculated for each supernatant dilution from the formula 100 (a-b)/ (ac) where a, b and c are the absorbances of wells with respectively L929 cells and medium, L929 cells and monocyte supernatant and no cells. The titre (defined as dilution causing 50% cytotoxicity) was then calculated from the graph of cytotoxicity vs logio dilution, using the least-squares method with the aid of a programmable calculator.
Comparison of macrophage-activating agents as cytotoxin inducers.-The results are shown in the Table. In all tests endotoxin was included as the positive control.
t Supernatant from monocyte-depleted lymphocytes (5 x 106/ml) incubated for 3 days in MEM/FCS with BCG (107 viable organisms/ml). inducer, polyinosinic-polycytidylic acid (poly 1,C) (test 3). A number oflymphokine preparations (either antigenor mitogenstimulated) were tested and some of these did induce the cytotoxin, but never as effectively as endotoxin (test 4). The lymphokine preparation presumably contained y-IFN as well as other factors. Although a-and 3-IFNs were not cytotoxin inducers, we cannot exclude the possibility that y-IFN is effective. Some micro-organisms or their products gave titres as high as endotoxin, including BCG, C. Parvum and the yeast cell-wall preparation zymosan (tests 5, 7, 8). Although these inducers can all activate the alternative pathway of complement, this appears not to be the critical factor, as inulin and rabbit erythrocytes, two other activators of the alternative pathway of human complement, did not induce cytotoxin (data not shown). Muramyl dipeptide, a purified mycobacterial product with macrophage-activating properties, did not induce cytotoxin (test 6).
Phorbol myristate acetate (PMA) acts as an inducer, but was less potent than endotoxin (test 12). PMA can cause differentiation along the monocyte/macrophage pathway, and it may act in part by increasing the pool of cytotoxin-producing cells.
To check whether the cytotoxin inducers other than endotoxin were effective because of endotoxin contamination, each was tested in the presence of polymyxin B, an antibiotic which neutralizes endotoxin. None of the inducers, when tested at the lowest dose capable of inducing maximum amounts of cytotoxin, was neutralized by 5 Hug/ml polymyxin B.
It is probable that the same cytotoxin is induced by the various agents described above, for the following reasons. Although physico-chemical analysis of the cytotoxins induced by the various agents is incomplete, no differences have emerged as yet. Secondly, the dose-response curves have similar slopes. Thirdly, two inducers used together are no more effective than either alone. Lastly, after pulsing with sub-toxic doses of one inducer for 0-20 h, the same or other inducer is never effective at producing significant amounts of cytotoxin over a second 20 h period.
In conclusion, the human monocyte cytotoxin can be induced by certain macrophage stimulants but not by others. BCG, C. parvum, zymosan and PWM are potent inducers, lymphokines and PMA are less effective and o-and ,8interferons, poly I,C, MDP, Con A and PHA are ineffective.