BACTERIAL NUTRITION: A BRIEF NOTE ON THE 

 PRODUCTION OF EREPSIN (PEPTOLYTIC 

 ENZYME) BY BACTERIA 



NATHAN BERMAN and LEO F. RETTGER 

 From the Sheffield Laboratory of Bacteriology and Hygiene, Yale University 



In a recent publication (1) the authors stated that gelatin- 

 non-liquefying bacteria of the Bacillus coli type are able to exert 

 but little, if any, proteolytic action on Witte's peptone and on 

 partially purified proteoses. The biuret test was employed in 

 the earlier experiments, and the results were not expressed in 

 mathematical terms. In later studies the changes in the pro- 

 tein content of the media have been recorded in definite figures. 

 Furthermore, the Sorensen method for the determination of 

 amino acids has been used along with the biuret test. 



The published results in a few instances gave unmistakable evi- 

 dence of a reduction in the amount of biuret-giving substances 

 in the media containing Witte's peptone, when B. coli, B. typhi 

 and B. pullorum were employed. These reductions were only 

 slight, however, and were noticeable only after two weeks' in- 

 cubation. While the changes in the protein content of the media 

 suggested the possibility of ereptic enzyme action, no definite 

 conclusion was warranted. Subsequent investigation has, in a 

 measure at least, clarified the situation. 



In a medium containing 1 per cent peptone, 0.25 per cent 

 Liebig's meat extract and 0.5 per cent of sodium chloride B. coli, 

 B. typhi, and B. para^?/7?/l^ A, under optimum cultural conditions 

 have slowly acted upon the biuret-positive substances in the 

 peptone. While at least two weeks were usually required to 

 produce any appreciable decrease in the ''protein," the loss 

 amounted to from 10 per cent to 40 per cent of the original when 

 the digestion was allowed to continue for at least three weeks. 



It was impossible to bring about a decomposition of more than 



537 



