Pertussis toxin (PT), preactivated with 20 mm dithiothreitol (DTT), was incubated with different serum dilutions (1/10-1/200) before addition to the reaction mixture. Final concentrations of the reagents were: PT (50 ng/ml), dithiothreitol (DTT)(less than or equal to 0.3 mM), bovine transducin (2 micrograms/ml), ATP (1 mM). GTP (1 mM), lysophosphatidylcholine (LTC) (0.1 mg/ml), sodium acetate (NaAc) (0.1 M), Tris-HCl, pH 7.1 (0.06 M) and 32P-NAD+ (10 microCi 28 Ci/mM). The reaction was stopped by precipitation with 10% TCA (w/v), the pellet was collected and the samples were submitted to SDS-PAGE followed by autoradiography. ADP-ribosylation was detected as the radiolabelling of a protein band (m.w. approximately 40 kD) corresponding to the alpha-subunit of transducin. 32P-ADP incorporation was a linear function of PT concentration. By this assay quantitative differences in PT neutralizing antibodies were found between sera which were not revealed by measuring PT neutralizing antibodies by a Chinese hamster ovary (CHO) cell culture test or by an enzyme-linked immunosorbent assay (ELISA). The conditions of the ADP-ribosylation of bovine transducin have been optimized to permit detection of the enzymic activity of as low amounts of PT as 0.5 ng. This opens the possibility of the study of the presence and avidity of neutralizing antibodies to PT in post-vaccination sera without preceding purification and concentration of the antibodies and thus may provide a useful tool for evaluation of whooping cough vaccines.