724 SALMONELLA 



Conradi (1906) found that a filtrate from an autolysed saline suspension of /SoZm. typhi, 

 after evaporation at 35° C. to one-tenth of its original volume, was toxic for guinea-pigs 

 in a dose of 0-2 ml. Yamanouchi (1909) reported that filtrates of cultures of Salm. typhi 

 in 5 per cent, peptone water were toxic for rabbits in a dose of 0-5-1 ml. per kilo body- 

 weight. Arima (1912) grew Salm. typhi on weakly alkaline agar, suspended the growth 

 in saline and later separated the bacillary bodies by centrifugation. The supernatant 

 fluid was pipetted off and examined separately. The bacillary bodies were washed three 

 times, ground up, and extracted with saline. Both solutions were toxic for rabbits, less 

 so for guinea-pigs and mice. Douglas (1921) showed that tryptic digests of acetone- 

 extracted t5rphoid bacilli were highly toxic. The statements with regard to the thermo- 

 stability of these toxic extracts or filtrates are somewhat conflicting. But most workers 

 report them as heat-resistant. Taken as a whole these observations indicate that tjrphoid 

 bacilli elaborate toxic substances which are, in the main, retained within the bacterial cell, 

 though they are liberated into the surrounding medium when autolysis occurs. There 

 is, at the moment, no evidence that the typhoid bacillus produces an exotoxin in the 

 ordinary sense of that term. 



Most other salmonellse that have been examined behave, in this respect, in the same 

 way as Salm. typhi, though the toxicity of kiUed cultures or extracts of the different 

 species may show significant differences in toxicity for laboratory animals, particularly 

 the rabbit (White 1926, 19296). 



The two natural pathogens of rodents, Salm. typhi-miirium and Salm. enteritidis, may 

 be considered in rather more detail from this point of view, since it happens that we have 

 recently obtained some knowledge of the constitution of one, at least, of their toxic products. 



Ecker (1917), Ecker and Richardson (1925), Branham (1925), Ecker and Rimington 

 (1928) and Menten and King (1930) noted the toxicity for rabbits or mice of broth filtrates 

 of Salm. typhi-murium and of certain other salmonellse ; and Casper (1928-29) 

 prepared a toxic extract of the bacterial bodies by extraction with antiformin. Ecker 

 and Rimington (1928), Menten and King (1930) and Casper (1928-29) were able, by various 

 methods of adsorption or precipitation, to obtain a partially purified toxic fraction from 

 these crude filtrates or extracts. The fractions so obtained were thermostable. The 

 purification was very incomplete, and though qualitative tests indicated the presence 

 of polysaccharide constituents, no serious attempt was made to determine the chemical 

 constitution of the toxic components. Martin ( 1934) showed that the greater part of the 

 toxic material present in a broth culture of Salm. typhi-murium was contained in the 

 bacterial cells ; and that these constituted the most favourable starting material for 

 obtaining a purified toxin. His method of purification is considered below. 



Independent studies on the immunological properties of chemical fractions isolated 

 from Salm. typhi-murium and Salm. enteritidis byBoivin and his colleagues (1933, 1934, 

 1935) and by Raistrick and Topley (1934), Delafield (1934) and Martin (1934) have thrown 

 further Ught on this problem. These studies were undertaken, in the main, with the object 

 of obtaining a chemically pure immunizing substance from these organisms. The point 

 that concerns us here is that fractions obtained by both groups of workers were found 

 to be highly toxic. 



Boivin and his colleagues obtained their fractions by extraction with trichloracetic 

 acid, followed by alcohol precipitation. Raistrick and Topley extracted the bacilli with 

 acetone, digested them with trypsin, and submitted the tryptic digest to fractional pre- 

 cipitation with alcohol. 



The final product obtained by both groups of workers was free from protein, as judged 

 by the ordinary chemical tests, but contained a polysaccharide component that gave 

 rise to reducing sugars on hydrolysis. Nitrogen and phosphorus were also present, and 

 the results of micro-combustion analysis, taken with various qualitative tests, suggested 

 that the active fraction consisted of a complex polysaccharide combined with a phospha- 

 tide. The polysaccharide can be separated from the phosphatide residue by heating with 

 weak acetic acid. When this is done, the polysaccharide is entirely, and the phosphatide 



