INVESTIGATIONS ON AN IMMUNOASSAY OF ERYTHROPOIETIN

It is presently possible to perform determinations of plasma erythropoietin, but such procedures are fraught with difficulties. The current methods bear the handicap inherent in bioassay: they are insensitive, time-consuming and expensive in that they require animals and many materials. Since the antigenicity of human urinary erythropoietin has been established by Schooley and Garcia,l one may hypothesize that an immunoassay is theoretically possible. This would be a vast improvement when compared to current methods, but such an immunoassay is not possible without first obtaining a specific antibody. We were able to produce a precipitating neutralizing antibody against human erythropoietin by injecting rabbits with a human plasma-erythropoietin preparation. As the antigen used for immunization contained several other human plasma proteins besides erythropoietin, precipitating antibodies directed against these proteins were produced at the same time and several precipitation lines were obtained with the agar diffusion technique (FIGURE 1) . The antiserum showed a continuous precipitation line with human, rat and rabbit erythropoietin using the same technique (FIGURE 1 ) . By using this cross reactivity, we were able to produce an antiserum specifically directed against human erythropoietin and not against other human plasma proteins. We immunized rabbits with a rat erythropoietin preparation. Although the produced antiserum showed several precipitation lines with rat erythropoietin preparations, only one was seen with a human erythropoietin preparation, as well as with a rabbit erythropoietin preparation (FIGURE 2) . The antiserum showed only one precipitation line in the al globulin range with a human and a rabbit erythropoietin preparation in the immunoelectrophoresis (FIGURE 3 ) . The antiserum depressed the endogenous erythropoietin activity in the normal rat (TABLE 1 ) and neutralized human erythropoietin in the polycythemic hypoxic mouse assay (TABLE 2) . We were able to dissociate the erythropoietin activity from an immune precipitate of a human erythropoietin preparation and this antiserum at pH 2.5 followed by Sephadex G 200 gelfiltration (TABLE 3 ) . The anti-rat erythropoietin antiserum again showed only one precipitation line in the a1 globulin range with the aforementioned dissociation product in the immunoelectrophoresis as did an anti human erythropoietin antiserum. To investigate the possibility of an immunoassay of erythropoietin, we carried out some preliminary experiments with different plasma samples using the agar diffusion technique. With normal plasma a precipitation line was seen up to a dilution of 1 :4. With plasma of patients with aplastic anemia a precipitation line up to a dilution of 1 :256 was seen. Plasma of two patients with chronic nephritis, uremia and anemia did not give any visible reaction.