Studies on the activity of barnase toxins in vitro and in vivo.

Pseudomonas exotoxin A (PE) is a protein toxin composed of three structural domains which are responsible for cell binding (domain Ia, amino acids 1-252), translocation into the cytosol (domain II, amino acids 253-364) and ADP-ribosylation activity (domain III, amino acids 405-613). We have previously described (Prior, T. I., FitzGerald, D. J., and Pastan, I. (1992) Biochem. 31, 3555-3559) a molecule composed of amino acids 1-412 of PE and the extracellular ribonuclease of Bacillus amyloliquefaciens, barnase (Bar), and shown that this protein (PE1-412-Bar) is toxic to cells due to its ribonuclease activity, which had been delivered to the cytosol. We have now used this model to evaluate the role played by the carboxyl end of domain II (amino acids 347-364), domain Ib, and the amino end of domain III (amino acids 405-412) in the translocation event. Toxins completely lacking domain III, termed PE1-380-Bar, or both domains Ib and III, termed PE1-364-Bar, were equally cytotoxic to a murine fibroblast cell line (L929) as was PE1-412-Bar. Extending the deletion to include part of the E-helix and all of the F-helix of domain II (amino acids 347-364) resulted in a toxin (PE1-346-Bar) that was 10-fold less toxic. Previously tested on only murine cell lines, we demonstrate that barnase toxins are cytotoxic also to a variety of human cell lines. Cytotoxicity was assessed by measuring inhibition of DNA synthesis. Surprisingly, PE1-380-Bar is not lethal when injected into mice, either intraperitoneally or intravenously, at 9 nmol, which is 2200-fold more than the amount required for killing by PE (4 pmol). In cell culture these barnase-containing toxins are 100-fold less toxic to murine fibroblast cells than PE. Barnase toxin has a greater survival time in the blood of mice than PE, with a half-life of 102 min. We conclude that domain II is sufficient to transport proteins into the cytosol. Further, since domain Ia can be replaced with other cell targeting moieties, we propose that barnase-toxins should be evaluated for utility in targeted cancer therapy.