The method of partial synthesis of antigens as employed in the foregoing experiments obviously cannot be substituted for the chemical study of natural antigens. But some questions of a rather general nature not easily accessible to investigations of the latter sort, may be approached by the use of artificial protein compounds. Thus the results reported indicate a peculiarity of certain chemical structures such as acid radicals.
The group of immune sera obtained by injecting azoproteins made from non-acid azo components had a wide range of activity. Substituents like CH3, OCH3, NO2, Cl, Br, I, in the aromatic nucleus altered the reactions to a moderate degree only, in most cases. The effects were dependent more on the position than on the nature of the substituents. Two substances were found, however, which had a pronounced effect on the specificity of the compound protein, namely acetyl-para-phenylenediamine and para-aminoacetophenone. In consideration of the above facts it is uncertain whether the antigenic changes noticed by Obermayer and Pick (7) after treating proteins with nitric acid, nitrous acid or iodine are mainly due to the substitution of hydrogen in the benzene ring by NO2 and I, as is the general belief, or to other changes of the protein. This question had been raised already by the observation that the proteins treated with HNO3 or HNO2 containing respectively the nitro or the diazo group, did not differ substantially in their serological properties (8).
The antigens made from acid compounds form a group with distinctive features. In the first place the presence of acid radicals destroys the reactivity with the immune sera for the non-acid substances. This influence is so marked that even the reaction with the species specific part of the protein, if such is present, appears to be diminished. Also the sera produced with the acid antigens react but feebly with the non-acid azoproteins. Accordingly it was possible to show that by hydrolysis of the ester of an aromatic acid contained in an azoprotein the serological reactions of the antigen underwent a radical change.
The presence of a free carboxyl group in the antigens not only determines the characteristics mentioned but there is reason to believe that it increases markedly the degree of specificity exhibited by the antigens and the corresponding immune sera, when cross-tests are made with a number of acid azoproteins and their antisera. This is brought out by a comparison of the results of the present investigation with those described previously (3). It is of interest in this respect that the specific carbohydrates found by Avery and Heidelberger in pneumococci and pneumobacilli are mostly, if not in all cases, compounds of distinctly acid character.