The in vitro effect of C-AGP (pure α1-acid glycoprotein from the ascitic fluid of cancer patients) on NK cell cytotoxicity was tested using normal healthy human PBMC. C-AGP had no inhibitory effect on basal NK cell activity. C-AGP selectively suppressed the augmentation of NR cell activity by rIFNαA and rIFNγ, but C-AGP did not prevent the NK activation by rIL-2. NK cells in PBMC treated with C-AGP for 12 h and then washed just once, to remove the C-AGP, fully recovered the ability to respond to rIFNαA. However, after the treatment of PBMC with C-AGP for 5 or 6 days, NK cells failed to respond to rIFNαA, in spite of washing to remove C-AGP from the cultures. Monocytes were necessary for the suppressive effect of C-AGP on rIFNαA activation of NK cells. Indomethacin restored the ability of NK cells to respond to rIFN(yA in C-AGP treated PBMC. These results suggest that monocytes are able to selectively suppress the response of NK cells to IFNs in the presence of, or following treatment with C-AGP.