CHANGES INDUCED IN THE O ANTIGENS OF SALMONELLA 1 



D. W. BRUNER and P. R. EDWARDS 



Department of Animal Pathology, Kentucky Agricultural Experiment Station, 

 Lexington, Kentucky 



Received for publication December 11, 1947 



It is possible to induce profound changes in the flagellar antigens of Salmonella 

 through growth in media containing H antisera. These changes were recently 

 summarized by Edwards and Moran (Proc. Soc. Exptl. Biol. Med., 61, 242). 

 Hitherto, the only changes reported in the O antigens were those associated with 

 variation between smooth and rough forms and those occurring in natural form 

 variation. In this laboratory attempts to transform O antigens by the method 

 of Boivin et al. (Experientia, 2, 139) and by growth in various combinations of 

 S and R antisera and S vaccines have been unsuccessful. However, by a modi- 

 fication of the method of Gard (Z. Hyg., 120, 615) in which absorbed O anti- 

 serums were used in high concentration, it was possible to bring about certain 

 changes. 



When S. anatum (III,X,XXVI: e,h-l,6) was cultivated in semisolid medium 

 containing III,X,XXVI serum that had been absorbed with a type having O 

 antigens III,XV, the organisms gradually spread through the medium. From 

 the spreading growth was isolated a form that was indistinguishable from S. 

 newinglon (III,XV:e,h-l,6) by agglutination and absorption tests. Absorption 

 of the III,X,XXVI serum by S. newington, S. Cambridge (III,XV: e,h-l,w), or S. 

 new-brunswick (III,XV: l,v-l,7) gave the same results. The induced III,XV: 

 e,h-l,6 form was then reverted to a typical S. anatum strain by cultivation in 

 absorbed III,XY serum. Similarly, S. meleagridis (III,X,XXVI: e,h-l,w) was 

 changed to a form indistinguishable from S. Cambridge (III,XV: e,h-l,w). 



Although filtrates similar to those employed by Boivin were not used in the 

 experiments, it must be remembered that the serums were absorbed with very 

 large doses of bacteria and probably contained dissolved antigens as well as 

 metabolic products. Thus the principle that induced the changes may be the 

 same as that involved in Boivin 's work with Escherichia coli. Further, attention 

 should be called to the fact that the changes described are only transformations 

 between subgroups of the same O group. Whether similar experiments will 

 permit transformation between distinct O groups or whether they will lead only 

 to rough variation remains to be determined. 



1 The investigation reported in this paper is connected with a project of the Kentucky- 

 Agricultural Experiment Station and is published by permission of the Director. It was 

 supported in part by a research grant from the U.S. Public Health Service. 



449 



[Reprinted by permission of The Williams & Wilkins Company from Journal of 

 Bacteriology 55 : 449, 1948] 



Il8 



