Faculty Bibliography

Incubation of HeLa cell cultures with human interferon-gamma (IFN-gamma) increased the binding of radioiodinated human tumor necrosis factor (TNF) to specific cell surface receptors (TNF-R). IFN-gamma also produced a proportionate increase in receptor-mediated endocytosis of TNF. TNF-R expression was significantly increased after 6 h of exposure to IFN-gamma (100 units/ml), and it remained elevated in the continuous presence of IFN-gamma for at least 20 h. Incubation of cells with IFN-gamma in the presence of cycloheximide, followed by treatment with actinomycin D and reversal of the inhibition of protein synthesis, also resulted in increased TNF-R expression as compared to cultures subjected to the same treatments in the absence of IFN-gamma. These results suggest that IFN-gamma can directly stimulate accumulation of the mRNA for TNF-R and that TNF-R is among the cellular proteins whose synthesis is increased by IFN-gamma

The mechanism of human peripheral blood monocyte-mediated cytotoxicity was investigated using the HT-29 human colon adenocarcinoma line, A673 human rhabdomyosarcoma line, and A375 human melanoma line as target cells. Pretreatment of these target cells with 100 U/ml of recombinant human interferon (IFN)-gamma for 48 hr increased their susceptibility to monocyte killing. Increased susceptibility to the lytic action was particularly pronounced at low effector/target cell ratios. Unlike IFN-gamma human IFN-alpha did not potentiate monocyte cytotoxicity, and pretreatment of HT-29 with IFN-alpha also had virtually no effect on their susceptibility to monocyte killing. However, IFN-gamma appeared to prime either monocytes or target cells to become responsive to IFN-alpha. Our data suggest that IFN-gamma can promote the killing of tumor cells by monocytes through two separate actions, one on the monocyte and one on the target cell

Incubation of several human tumor cell lines with human interferon-gamma (IFN-gamma) increased the specific binding of subsequently added 125I-labeled recombinant human tumor necrosis factor (TNF). A similar increase in TNF binding was seen in murine L929 cells after incubation with murine IFN-gamma, but not after incubation with human IFN-gamma. Increased TNF binding to cells incubated with IFN-gamma was due to an increase in the number of TNF receptors, with no demonstrable change in binding affinity. In one out of two human cell lines tested, IFN-alpha and IFN-beta also produced increased TNF binding, albeit with a lower efficacy than IFN-gamma. A maximal increase in TNF binding was seen after about 6 to 12 hr of incubation with IFN. Increased TNF binding due to enhanced TNF receptor expression may contribute to the enhancement of TNF cytotoxicity seen in some tumor cell lines after INF treatment. Modulation of TNF receptor expression by IFN may also influence other biological activities of TNF

Tumor necrosis factor (TNF) is a monocyte-derived protein cytotoxic or cytostatic for some tumor cell lines. Here we show that highly purified E. coli-derived recombinant human TNF stimulated the growth of human FS-4 diploid fibroblasts. Stimulation of cell growth was demonstrable at a TNF concentration of 10 pg/ml (3 X 10(-13) M). Maximal stimulation was attained at TNF concentrations of 10 ng/ml (3 X 10(-10) M) or higher. Growth-stimulatory activity of TNF was inhibited by an mAb neutralizing the cytotoxic activity of TNF. Growth stimulation was not inhibited by another mAb specific for TNF, lacking neutralizing activity for the cytotoxic activity of TNF. Growth stimulation by TNF was more marked and more sustained in the presence of greater than or equal to 10% FCS than in medium with less than or equal to 5% FCS. Addition of TNF to confluent FS-4 cultures also produced a marked stimulation of cell growth in the presence of fresh FCS, while a much less marked stimulation was seen in the absence of FCS. Stimulation of confluent cultures by TNF in serum-free medium was enhanced by insulin, suggesting that insulin or insulin-like growth factor(s) in the serum can act synergistically with TNF in producing growth stimulation. While the growth-stimulatory effects of TNF and insulin were synergistic, the actions of TNF and epidermal growth factor (EGF) were less than additive, suggesting that TNF and EGF may activate identical or similar pathways. We conclude that stimulation of cell growth is probably a physiological function of TNF, and that the cytotoxic and cytostatic actions of TNF may be the result of an anomalous growth signal transduction in neoplastic cells lacking the constraints of normal growth control mechanisms

Production of interferon (IFN)-gamma by peripheral blood leukocytes (PBL) was examined in cultures of unseparated fresh whole blood exposed to phytohemagglutinin (PHA), concanavalin A (Con A), or pokeweed mitogen (PWM). The yield of IFN-gamma was measured by a newly developed immunoradiometric assay. Nine of 14 patients with acute pulmonary tuberculosis (TB) showed a depressed IFN-gamma response to Con A and/or PWM. Only four of these TB patients also showed a depressed IFN-gamma response to PHA. Stimulation of the patients' PBL cultures with PHA in the presence of exogenous interleukin 2 (IL 2) produced normal IFN-gamma yields in all but the most severely depressed patients. PBL cultures of TB patients with defective IFN-gamma production in response to mitogenic lectins also produced less IFN-gamma after stimulation with tuberculin PPD. Although some patients showed a moderate degree of lymphopenia, their OKT4/T8 lymphocyte ratios were mostly normal or close to normal, with the notable exception of one TB patient who has been diagnosed to have the acquired immune deficiency syndrome (AIDS)

Ion-exchange and gel filtration chromatography were used to purify partially a 33,000-dalton (33 kD) 2',5'-oligoadenylate synthetase and a 110,000-dalton (110 kD) 2',5'-oligoadenylate synthetase from HeLa cells treated with alpha-interferon (IFN-alpha). The 33-kD enzyme was optimally activated only when double-stranded RNA was added in 100 to 1000-fold excess of the concentration activating the 110-kD enzyme. Certain double- or triple-stranded RNAs, which were observed to activate the 110-kD enzyme, failed to activate the 33-kD enzyme, even when added at high concentration. The 110-kD enzyme manifested an alkaline pH optimum of 7.5 or more and the 33-kD enzyme an acidic pH optimum of 6.0 or less. The results suggest that intracellular activation of the two forms of 2',5'-oligoadenylate synthetase may occur under markedly different conditions