only deviated from that of the Lowry assay by 14 -18%, offering further alternatives. In the Lowry system, viscumin again gave similar color yields compared to ConA, while the reactivity of ricin and RCA was slightly smaller and similar to that of BSA (Fig. ). The accuracy of protein quantitation (maximum standard deviations of 10%) was significantly increased relative to the Bradford assay, due to the higher slope of the response curves. With respect to the isolated subunits, the color development for viscumin B chain exceeded only slightly that of viscumin, while the A-chain response to the Lowry reaction increased drastically as compared to the relative Coomassie blue dye response, approaching that of the whole AB toxin (Fig. ). Thus, even the quantitation of the toxic A chain could be satisfactorily achieved by application of the Lowry method using BSA as standard.
The reported results emphasize the critical importance of the selection of an appropriate colorimetric method and protein standard for the determination of protein concentration. This selection becomes particularly relevant in the standardization of proteins of biopharmaceutical interest, such as viscumin and related toxins. We show that the Lowry assay is the method of choice for quantitation of viscumin and its toxic A subunit and related toxic agglutinins, using ConA and BSA, respectively, as standards.
Acknowledgments. This work was supported by Grant PB98-0650 from the Direccio ´n General de Investigacio ´n Cientı ´fica y Te ´cnica (Madrid, Spain) and the Volkswagen Foundation (Hannover, Germany). We thank Acciones Integradas Hispano Alemanas HA1997-0064 for coverage of traveling expenses.