Agonist and antagonist effects of interferon alpha and beta on activation of human macrophages. Two classes of interferon gamma receptors and blockade of the high-affinity sites by interferon alpha or beta

H2O2-releasing capacity and limited antitoxoplasma activity could be induced in human macrophages (derived from monocytes cultured greater than or equal to 5 d) but not in monocytes themselves (cells cultured less than or equal to 4 d) by a further 3-d incubation with pure natural or rIFN-alpha or -beta. More than 3 pM (10 U/ml) of these IFNs was required, with greatest effects at approximately 300 pM (10(3) U/ml). At 300 pM, H2O2-releasing capacity was enhanced 4.4 +/- 1.6-fold over medium control (mean +/- SD for natural INF-alpha, rIFN-alpha A, rIFN-alpha D, and rIFN-beta) compared to an 8.4 +/- 4.8-fold increase with rIFN-gamma (100 pM, 100 U/ml) in the same experiments. Unexpectedly, low concentrations of IFN-alpha or -beta (3 fM-300 pM) blocked induction of H2O2-releasing capacity by rIFN-gamma (10 pM), with a 50% inhibitory dose of approximately 80 fM. However, IFN-alpha or -beta (3 fM-300 pM) could not inhibit the effect of higher concentrations of rIFN-gamma (1 nM). In contrast to results with monocytes or young macrophages, Scatchard plots of binding of 125I-rIFN- gamma to mature macrophages (day 8 of culture) indicated two classes of binding sites: approximately 2,000 high-affinity sites (Kd approximately 0.43 nM) and approximately 23,000 low-affinity sites (Kd approximately 6.4 nM) per cell. Binding of 125I-rIFN-gamma to the high- but not the low-affinity sites was blocked by simultaneously added IFN- alpha or -beta, with a 50% inhibitory dose of approximately 2 U/0.25 ml (approximately 2 pM), or reversed by subsequently added IFN-alpha or - beta. Thus, differentiation of human mononuclear phagocytes in vitro is accompanied by the emergence of (a) an agonist response to submicromolar concentrations of IFN-alpha or -beta, (b) antagonism of the effect of picomolar IFN-gamma by femtomolar IFN-alpha or -beta, (c) two classes of IFN-gamma-Rs, and (d) nonstimulatory binding of IFN- alpha or -beta to the high- but not the low-affinity IFN-gamma-Rs, with higher affinity than rIFN-gamma itself. We speculate that traces of IFN- alpha or -beta derived from stromal cells, parenchymal cells, or resident macrophages may dampen the activation of mature tissue macrophages by the small amounts of IFN-gamma that diffuse from inflammatory sites into normal tissues. Such a mechanism could constrain the potentially destructive phenomenon of macrophage activation to areas where monocytes have recently immigrated and/or the concentration of IFNs is high.