The essential points presented in this paper are:
Certain antipneumococcus horse sera, though showing powerful precipitation reactions with pneumococcus antigen, do not sensitize similar antigens to alexin fixation.
Such sera not only fail to fix alexin in the presence of antigen as described above, but, in addition to this, they will prevent alexin fixation when added, in small amounts, to tubes of pneumococcus antigen sensitized with antipneumococcus rabbit serum. They, thus, prevent the fixation of alexin by a sensitized antigen which would ordinarily have given a positive fixation. In order to obtain this interfering action, it is necessary to add the horse serum to the sensitized antigen before the alexin is added. If the alexin is added first and given thirty minutes at 37.5°C., subsequent addition of the horse serum does not interfere with the fixation.
This interfering effect of the horse serum can be removed by absorbing with sensitized pneumococci.
While such absorption completely removes the interfering effect from the horse serum, our experiments attempting to show that a fixing sensitizer is then left unhampered in the horse serum have not been conclusive up to the present time, but experiments like Experiment IV of the reported series suggest such a condition. This is being subjected to further study.
We cannot positively determine, as yet, whether interference with the attachment of alexin is due to a hitherto undescribed substance in horse serum, separate from the sensitizer, or whether the phenomenon depends upon adventitious conditions, such as the association of the sensitizer with serum constituents which alter its colloidal state or in some other way indirectly modify the physical conditions usually existing in such immune sera.
However this may be, our results appear to indicate definitely that the failure of a specific immune serum to sensitize antigen for alexin fixation, while powerful in its agglutinating and precipitating effects, does not prove the absence of a “fixing” antibody. This confirms our view that the failure of any individual antibody effect in an immune serum may be due to secondary and environmental causes rather than to absence of a given “antibody.”
Incidentally, the experiments show that the titration of the antibody potency of horse serum for therapeutic use is not a reliable method, since such sera, extremely powerful in their agglutinating and precipitating effects and containing a high concentration of antibodies, may be entirely incapable of giving rise to alexin fixation reactions for the reasons demonstrated.
Whether or not the property observed by us is universal for horse sera or is at all common for immune sera in general has not been determined as yet. It may well be that the body we have described is a constituent of exceptional specimens only. Because of its possible practical importance, however, this matter will be extensively investigated both in connection with pneumococcus and other sera.
In closing, we would again call attention to the fact that the residue antigens produced as described, both with pneumococci and a great many other bacteria, constitute the most easily managed and convenient antigens for specific precipitations and complement fixations. With the pneumococcus, it is merely necessary to grow on agar surfaces properly enriched, wash off with salt solution, shake for an hour in slightly alkalin salt solution, filter, boil with acid, clear by filtration and neutralize. The resultant water-clear fluid precipitates powerfully and possesses none of the disturbing features for complement fixations possessed by bacterial suspensions or extracts.