Experimental studies on the significance of endotoxin in the pathogenesis of brucellosis.

The participation of an endotoxin in the pathogenesis of brucellosis has received scant recognition although clinical observations have suggested that a Brucella toxin might be responsible for some of the manifestations in the acute phases of the disease. This has been especially obvious in patients whose illness has been due to Brucella melitensis. In general, brucellosis caused by Br. melitensis is a more severe disease than that resulting from Brucella abortus (1). It has also been observed in experimentally infected guinea pigs that Br. melitensis caused a more severe type of illness than did Br. abortus, although the abortus strains produced a larger number of demonstrable lesions in the tissues (2). Experience with aureomycin in the treatment of seriously ill patients with brucellosis has also suggested that an endotoxin might be responsible for some of the reactions observed after an initial dose of the antibiotic (3). Within eight to twelve hours after the first oral dose of aureomycin, an abrupt rise in temperature occurred, which was associated with a shock-like reaction, hypotension and tachycardia. Smaller initial doses of aureomycin eliminated the more severe reactions. A possible explanation for these clinical phenomena was that the antibrucella activity of the aureomycin resulted in the liberation of an agent or agents from the Brucella cells that provoked the undesired reactions. Further clinical observations that have promoted the present study were related to the dramatic improvement that occurred following the administration of adrenocorticotropin (4). No therapeutic agent available thus far for human brucellosis has resulted in a more rapid improvement in the condition of the acutely ill patient than that which has accompanied the use of either cortisone or adrenocorticotrophic hormone, even though the Brucella organisms have not been simultaneously eliminated from the tissues or blood stream. In view of the questions raised by these clinical observations, experiments were carried out to determine if the endotoxins of representative strains of the three species of Brucella differed in their toxicity for animals. The investigations were then extended to see if antibiotics or cortisone offered protection against the endotoxins. And finally, the acute lethal effect of the endotoxins was compared with that of living Brucella organisms. The concept that the endotoxin of Brucella organisms is related to the manifestations of the disease is in accord with many studies that have been carried out with the toxins of other gram-negative bacteria (5). It is now generally agreed that the endotoxins of gram-negative organisms are associated with somatic antigens located on or near the surface of smooth cultures of the cells (6). Little or none of the endotoxic substance is associated with the surface of rough colonies. These surface factors include the 0 and Vi antigens, and are composed of carbohydrate-lipoid-protein complexes, in which the carbohydrate is probably the major component (7). Miles and Pirie (8) in their basic studies on a smooth culture of Br. melitensis isolated a complete somatic antigen, which was toxic and which had as its "core" an amino polyhydroxy antigen, designated as AP. They stated that AP was analogous to the SSS isolated by Raistrick and Topley (9) from Bacterium aertrycke and (10) from Bacterium typhosum, and by Morgan (11) from Bacterium shigae. Furthermore, AP was similar to the polysaccharide complex isolated by Boivin and Mesrobeanu (12) from different species of gramnegative organisms. AP also contained the M and A Brucella antigens of Wilson and Miles (13). Paterson, Pirie, and Stableforth (14) have also isolated a complete antigen from a smooth culture of Br. abortus, similar to, but not identical with that obtained from Br. melitensis. The available information suggested that endotoxin from smooth cultures of Brucella could be obtained by several