Summary
Previous work has shown that at least certain antibodies may be brought down in the course of a specific precipitin reaction. Antitoxins (Dehne & Hamburger, Weill-Hallé and Lemaire); fixation antibodies (Gay); and the protective bodies of antipneumococcus serum (Gay and Chickering) are instances of this phenomenon. It seemed advisable to undertake a more systematic study of the several types of antibodies to learn how frequent this occurrence may be. It would be of great service to separate out antibodies in this manner as they could then be studied in a condition relatively free from other proteins and certain very practical results might be obtained with curative sera.
Attention is drawn to the fact that in the combinations noted above in which successful results have been obtained, the antibody adherent to the precipitate may have been originally associated with either the precipitinogen or with the precipitin. This is a matter not only of theoretic interest, but of practical importance in concentrating antisera. For this reason we have, so far as possible, endeavored to trace the antibody when it is associated with both factors in the precipitin reaction.
Our results have, for various reasons, been entirely negative, at least in so far as separating out antibodies is concerned. They seem, however, of value as being the first attempt at a systematic study of this kind and as saving others from fruitless endeavor.
The previous method of bringing down antitoxins with the precipitin has been shown in those instances in which the antitoxic serum served as the precipitinogen in the reaction. We have attempted to reverse the phenomenon by producing both antitoxins and precipitins to the diphtheria bacillus in rabbits and horses. Our results hitherto have been without result, owing to the fact that we have not succeeded in producing precipitins in these animals.
Experiments with various combinations have shown that both bacteriolysins and hemolysins when associated either with the precipitinogen serum or with the precipitin serum are not brought down in the formation of the precipitate. Similar negative results were obtained with artificial bacterial agglutinins and hemagglutinins.
Experiments on the fate of the fixation antibodies in the precipitin reaction shown, in general, that when the precipitate is produced by adding serum to its antiserum the fixation complex is present in the precipitate. This is by no means universal, as others have pointed out, and the fixation complex may lie in the supernatant fluid, or in both the supernatant fluid and the precipitate. The difference in results would seem to depend somewhat upon the rapidity of the precipitate formation without relation to its actual intensity. When bacterial extract is added to an immune serum the fixation complex seems usually present in the supernatant fluid.
In certain combinations it seems definitely shown that the fixation complex is present in that fraction (supernatant or precipitate) in which the protective bodies are absent. Thus in the case of pneumococcus precipitate produced by adding the extract of pneumococous to antiserum from the horse, the protective bodies are present in the precipitate and the fixation complex is present in the supernatant fluid. The exact reverse is true in a combination of rabbit antihorse serum and horse antipneumococcus serum.
We have not been able to find precipitins in the antisera to meningococcus and streptococcus from the horse which are used for curative purposes, and which were produced in a manner similar though perhaps not identical to the pneumococcus serum which we previously investigated.
