TOXIN PRODUCTION AND PATHOGENICITY 465 



antigenic relationship between these species, or between either of them and any sero- 

 logical type of C. diphtherice. Similarly, Bailey (1925) studied two strains of toxigenic 

 C. diphtherice, two of G. hofmanni and three of C. xerosis, together with seventeen non- 

 toxigenic strains of C. diphtherice. He was unable to detect any significant antigenic 

 relationship between these two species of diphtheroids and the diphtheria bacillus. 



The fact that agglutination tests fail to reveal any antigenic relationship between 

 G. diphtherice and the various diphtheroid organisms, does not, of course, mean that there 

 is no sharing of antigenic components between them. It means only that, if such sharing 

 exists, the shared antigens, either because of their position in the bacterium or for some 

 other reason, are not concerned in the agglutination of the intact bacterial cells. It is not, 

 therefore, surprising to find that certain antigenic components are distributed widely among 

 the corynebacteria, and even among related genera. Thus Krah and Witebsky (1930) 

 record that alcoholic extracts of the diphtheria bacillus, of certain diphtheroids, and of 

 the tubercle bacillus, all fix comjilement in the presence of an antiserum prepared against 

 any one of these organisms. Again, a strain of G. diphtherice that has undergone antigenic 

 variation, and lost its type-specific surface antigen, may agglutinate with antisera prepared 

 against diphtheria bacilli of other serological types, or against various diphtheroids. Neill 

 and his colleagues (1931) have described such a strain. 



Toxin Production and Pathogenicity. 



The type species, C. diphtherice, is an important human pathogen, giving rise 

 to a characteristic and often fatal disease, the lesions of which are, in the main, 

 produced by the action of a powerful exotoxin. This diffuses throughout the body 

 from the primary focus of infection, which is most frequently situated in the tonsillar 

 region. The pathogenesis of diphtheria in man, and its diagnosis, prevention and 

 treatment so far as these depend on bacteriological methods, are dealt with in 

 Chapter 61. The characters of the specific toxin, its effects on certain labora- 

 tory animals, and its immunological relationships with toxins produced by other 

 species and types within this genus, are, however, so important as differential 

 criteria, that it is necessary to discuss them here. 



The production of diphtheria toxin in artificial culture has been the subject 

 of a large series of empirical observations. From these it is clear that optimal 

 toxin production demands conditions which are not necessary for optimal growth. 



Davis and Ferry (1919) tested the value of media prepared from beef infusion, peptone, 

 and meat extract in various combinations with each other. The presence of beef infusion 

 was found to be essential. This was later disproved by Wadsworth and Wheeler (1934), 

 who devised a synthetic medium containing 2 per cent, of peptone, on which a toxin 

 of liigh potency can be prepared (Eaton 1936, Pappenheimer and Robinson 1937). Hartley 

 and Hartley (1922) tested various specimens of peptone, and found that, while each 

 brand of peptone gave a characteristic curve of toxin production, it was impossible to 

 predict the value of any one brand by a prehminary chemical analysis. Hartley (1922) 

 pointed out the superiority of a tryptic digest of horse muscle as a medium for the pro- 

 duction of toxin, and his findings have been confii-med and extended by Watson and Lang- 

 staff" (1927), who also confirm his observation that the value of such a medium is markedly 

 infiuenced by the method of sterihzation. In addition to growth-promoting substances, 

 there are apparently toxin-inducing substances present, which are very labile to heat 

 when the pH is at neutrahty, or on its alkaline side. Autoclaving at pH 8-0 or over may 

 completely destroy the value of a medium for toxin production. This may be avoided 

 by filtration through a Seitz press, followed by a short steaming. The addition to the 

 medium of maltose, or certain other energy sources, consideraljly increases the toxin yield, 

 particularly if sodium acetate or lactate is also added (Pope 1932, Ramon and Berthelot 

 1932, Pope and Smith 1932, Pope and Healey 1933a, Pope and Linggood 1939). The 



