COLICINS AND OTHER BACTERIOCINS 387 



the final number of bacteria killed, which proves to be propor- 

 tional to the amount of colicin added. In this way one measures 

 the number of "lethal particles" in the sample of colicin. Hence 

 the most informative method of colicin titration consists in deter- 

 mining the total number of cells killed by a suitable quantity of 

 colicin. The results may then be expressed in lethal particles 

 per unit volume of the colicin preparation (Jacob, Siminovitch, 

 and Wollman, 1952). 



The action of colicin on sensitive bacteria therefore appears to 

 be very similar to the bactericidal action of phage : either may be 

 described as irreversible fixation of lethal particles on specific 

 receptors. Once fixed, colicin interferes strongly with bacterial 

 metabolism (Jacob, Siminovitch, and Wollman, 1952). When 

 colicin ML, for instance, is added in excess to sensitive bacteria, 

 growth is almost immediately arrested, but no lysis occurs. 

 Respiration remains constant during 30 to 60 minutes. Syn- 

 theses of ribo- and deoxyribonucleic acids as well as induced en- 

 zyme synthesis are promptly inhibited. Furthermore, addition 

 of colicin to bacteria previously infected with phage blocks the 

 development of the phage. Thus in their action on sensitive 

 bacteria, the parallelism between colicin and virulent phages 

 such as T2 appears once more very striking. More especially, 

 since colicins do not induce bacterial lysis nor reproduce in sensi- 

 tive cells, their behavior is very reminiscent of that of ultraviolet 

 inactivated phages or phage ghosts. 



3. Colicinogenic Bacteria 



Problems met with in the study of colicinogenic bacteria are in 

 many respects analogous to those of lysogeny. They deal mostly 

 with the mode and genetic determination of colicin production. 

 For technical reasons, answers to these problems are more difl^- 

 cult to obtain with colicinogenic than with lysogenic bacteria. 

 On the one hand, colicins not being reproduced by the bacteria 

 they kill, titration is relatively laborious and insensitive. On the 

 other hand, the colicinogenic character cannot be acquired by 

 infection and therefore many types of experiment which have 



