The plant toxin saporin is a ribosome-inactivating protein which inhibits protein synthesis and growth of both normal and tumour cells. Its cytotoxic activity can be increased by coupling with antibodies recognizing cell surface antigens. In this work we performed experiments to test the hypothesis that saporin induces cell death via apoptosis. Exposure to saporin induced apoptosis in different cellular models, such as human peripheral blood B lymphocytes and neutrophils, in the Daudi B-cell line, and in the haemopoietic cell lines HL-60 and TF-1. This was indicated by: (a) the appearance of typical morphological features such as chromatin condensation, nuclear fragmentation and blebbing of plasma membranes: (b) DNA degradation into oligonucleosomal fragments: (c) the appearance of apoptotic cells on DNA flow cytometry as a cell population with reduced DNA content (A0 region). The fraction of cells showing features of apoptosis ranged from 19 +/- 5% for TF-1 cells to 35 +/- 8% for neutrophils. In experiments with normal peripheral blood B lymphocytes or with Daudi cells, we compared the activity of native saporin with that of an immunotoxin hybrid molecule obtained by binding the toxin to two bispecific antibodies recognizing both saporin and the B lymphocyte-specific antigen CD22 (Sap/BsAb complexes). Saporin bound to the antibodies was 2-3 logs more effective than native saporin in inducing apoptosis, with maximal inhibitions being observed at concentrations of 10(-6) M for native saporin and 10(-9)-10(-8) M for the hybrid molecules. These findings indicate that treatment with saporin results in apoptosis of target cells and suggest that this may be relevant to the therapeutic use of saporin-containing immunotoxins. In fact, if used in vivo as an immunotoxin, its cytotoxic activity could be devoid of more extensive and non-specific tissue damaging effects as would be the case if saporin induced necrosis of target cells.