826 BRUCELLA 



from the three species, and considerable cross-precipitin reactions between the three 

 endo-antigens and their respective antisera. Br. abortus and Br. suis endo-antigens 

 reacted similarly, though not identically, and Br. melitensis endo-antigen was more sharply 

 distinguished from the other two (PenneU and Huddleson 1938). Huddleson (1943) has 

 more recently reported that neither the " complete " antigen nor the " endo-antigen," which 

 are both antigenic, wlU mduce a significant degree of active immunity to experimental 

 infection in the guinea-pig, but that a watery extract of crushed Uving Br. abortus or 

 Br. suis is highly effective against infection with Br. abortus. 



Miles and Pirie (1939a, b) attempted to separate from Br. melitensis the A and M antigens 

 postulated by Wilson and Miles (1932). By progressive degradation of antigenic material 

 obtained by the gentlest possible treatment of the bacterium, they prepared a series of 

 substances. The most degraded product was a formyl amino-polyhydroxy compound 

 containing carbohydrates with a probable molecular weight of about 3,300. This su))- 

 stance inhibited agglutination of Br. melitensis by homologous antibody, but did not 

 itself react with antisera, i.e. it was an inhibiting hapten. Next in complexity was a 

 substance analogous with the " endo-antigens " described above. It was antigenic, toxic, 

 and precipitated with antisera in a dilution of 1 : 5 mUhon. Its molecular weight was 

 about 1,000,000. On hydrolysis it yielded a phosphohpin, phosphate, and the formyl 

 ammo compound. This antigen was in turn derived from a more polymerized material 

 combined with a protein-like material ; and finally, this last sjiibstance, combined with 

 one-third to one-quarter its weight of a mixture of hpins and phosphoUpins, constituted 

 the relatively unstable native antigen. The native antigen, which comprised about 

 10 per cent, of the dry weight of the cell, was in a state of greater aggregation than the 

 simple antigen, and weight for weight was more antigenic and less toxic. The native 

 antigen consequently appears to be a large complex of a protein-hke substance, with 

 two sets of phosphohpins differing in their readiness of separation from the complex, 

 the second of which, together with a polymerized formyl amino-polyhydroxy compound, 

 determines the specificity and antigenicity of the main antigen of Br. melitensis. In the 

 Ught of these findings it seems probable that the antigenic fractions of bruceUae prepared 

 by trichloracetic acid extraction or tryptic digestion of the baciUi consisted of mixtures 

 in varying proportions of native antigen and its various degradation products. Miles 

 and Pirie failed to separate the A and M antigens, but, as in previous work, they obtained 

 evidence that the antigens of the S-forms of Br. abortus and Br. melitensis showed a quali- 

 tative similarity, but a quantitative difference in distribution. 



The apparent efl&cacy of hve as compared with dead vaccines, led Priestley (1938) 

 to search — without success — for a labUe antigen in Br. abortus, similar, perhaps, to the 

 Vi antigen of Salm. typhi. Topping (1934) isolated substances of a nucleo-protein nature 

 from Br. abortus and Br. melitensis, which were feebly reactive with heterologous and 

 homologous sera. The reactivity may have been due to contamination with the main 

 soluble antigen. Stahl, PenneU and Huddleson (1939) extracted non-toxic " protein- 

 nucleates " from aU three species ; those from S strains were serologically reactive, but 

 only group-specific, and constituted about 14 per cent, of the dry weight of the cell ; 

 those from R strains constituted about 18 per cent, of the weight of the cell, and were 

 serologically mactive. The serological reactivity of the S protein-nucleates was associated 

 with the protein, and not with the nucleic acid part of the fractions. Stahl (1941) found 

 5-6 per cent, of Hpins in the dry cells. They were non-toxic and serologically inactive. 



Pathogenicity.— 5r. melitensis, Br. abortus, and Br. suis are all infective for 

 man and animals. Though undulant fever is the most characteristic result of 

 infection in man, numerous other clinical manifestations occur. Often the infection 

 remains latent, causing no recognizable symptoms of disease. Since we cannot 

 carry out large-scale experiments on man under similar conditions, it is impossible 

 to make any definite statement on the comparative virulence of these three organ- 

 isms, but the limited observations on human volunteers of Morales-Otero (1929, 



