ANTIBACTERIAL SUBSTANCES FROM BACTERIA 173 



scope of this book. We shall confine the discussion to a few illustrative examples, 

 and note some of the more recently described substances, with especial emphasis 

 on those that have proved their therapeutic value. 



Antibacterial Substances from Bacteria. 



Ps. pyocyanea was among the first bacteria to be studied from the point of 

 view of its elaboration of antibacterial substances. Cultures of the organism 

 yield a number of them : — pyocyanase, first prepared by Emmerich and Low 

 (1899), and Emmerich, Low and Korschun (1902), depends upon its content of 

 fatty acids of high molecular weight for its activity (Birch-Hirschfeld 1934, Hettche 

 1934) and affects a large variety of Gram-positive and Gram-negative bacilli 

 including pathogenic cocci. Salmonella bacilli, the diphtheria bacillus (Wagner 

 1929, Kramer 1935) and Br. abortus (Kocholaty 1942) ; pyocyanin, a respiratory 

 pigment (see Ehrismann 1934, and Chapter 3) ; a-hydroxyphenazine and a colour- 

 less compound having a strongly lytic action on F. cJiolerce (Schoental 1941). 

 lodinin, an antibacterial substance from Chroma, iodinum is a N-oxide of a hydro- 

 phenaziiie, and is apparently specifically antagonized by naphthoquinone (Mcllwain 

 19436). 



Bacteriolytic enzymes of bacterial origin were described by Malfitano (1900, 

 1903). Nicolle (1907) isolated from B. suhtilis a lytic agent active against saline 

 suspensions of staphylococci, pneumococci, the anthrax, plague and glanders 

 bacilli, and salmonellse (see also Jobling and Petersen 1914, Jobling, Petersen and 

 Eggstein 1915, Sartorius 1924, Much 1925). 



The inhibitory actions of certain bacteria were not obviously due to separable 

 enzymes. Thus, certain strains of Bad. colt, which inhibited the in vitro and in 

 vivo growth of other bacteria, would act only if whole, living cells were present. In 

 culture, they overgrew and eliminated SalmoneUa and dysentery bacilli, streptococci, 

 staphylococci, B. anthracis and C. diphtherice (Nissle 1930, 1932, 1933, Gundel and 

 Kliewe 1932, Koch and Kramer 1932, Besta and Kuhn 1934, Makowsky 1936, Eeploh 

 1937, Maner 1939). The i7i vivo inhibition is not marked. For example, Gundel 

 and Kliewe infected mice subcutaneously with mixtures of Bact. coli and a lethal 

 dose of B. anthracis. A immber of the mice survived, and the survival time of 

 those that died was increased, as compared with mice receiving the lethal dose 

 alone. Other observers have reported the inhibition of C. diphtherice by various 

 streptococci. Inhibition by the live cultures was observed in the subcutaneous 

 tissue of the guinea-pig (Besta and Kuhn 1934, Weigmann and Holzl 1940, Holzl 

 1941), and Duliscouet (1939) reported on the successful treatment of throat carriers 

 of C. diphtherice by killed cultures of an inhibitory organism. 



Two of the most interesting developments in the isolation of antibacterial 

 substances we owe to the work of Dubos. The first was directed to a specific 

 end, the isolation of a bacterium whose enzymes would destroy the polysaccharide 

 substance in the capsules of Type III pneumococcus. This was accomplished 

 by testing large numbers of bacterial strains, mostly spore-bearing aerobic bacilli, 

 in a medium containing the polysaccharide as the sole source of carbon. A bacillus 

 which grew in this medium was trained to produce large amounts of a polysaccharide- 

 splitting enzyme, which in a purified form was subsequently used with success 

 in the treatment of experimental Type III pneumococcus infection of the rabbit 

 (see Chapter 74). The enzyme was specific, and affected only Type III pneumo- 

 cocci. Dubos extended this search to bacilli with a wide range of antibacterial 

 action by serial subculture of samples of soil in a medium containing large quantities 



