102 



NATURE, FORMATION, AND ACTIVITIES 



other polyenes was studied b}- Henis and 

 Grossowicz (19G0). 



Other Antibiotics 



According to Aizawa (lOoo), both the 

 respiration and the adaptive oxidation of 

 mannose and galactose by Candida albicans 

 are inhibited by aureothricin, candimycin, 

 eurocidin, and trichomycin; they are not af- 

 fected, however, by penicillin G, chlortetra- 

 cycline, oxytetracycline, chloramphenicol, or 

 dihydrostreptomycin. 



Abraham (1959) defined some of the main 

 features of bacterial metabolism that are 

 open to selective attack by antibiotics. The 

 formation of the cell walls in liacteria and 

 actinomycetes differs from that of other 

 kinds of cells, although the precise nature 

 of the reactions that are inhibited has yet 

 to be elucidated. The same is true of certain 

 highly specific properties of the bacterial cell 

 which make them sensitive or resistant to a 

 given antibiotic. 



Dependence of JVIicroorganisnis upon 

 Specific Antibiotics 



Among the phenomena related to the 

 antibacterial properties of streptf)mycin, 

 the development of resistance and of depend- 

 ence is of particular interest. The problems 

 related to resistance will be considered in 

 detail in Chapter 10. The problem of de- 

 pendence was first observed by Miller and 

 Bohnhoff (1950). Each of 18 strains of menin- 

 gococci yielded two variants. One variant, 

 designated as A, grew in large yellowish 

 colonies on streptomycin-free and strepto- 

 mycin-containing media; it retained the 

 original virulence for mice. The other, E, 

 appeared in greatest numbers in concentra- 

 tions of 100 and 400 mg per ml of strepto- 

 mycin. Its colonies varied in size and color, 

 depending upon the concentration of the 

 antibiotic in which they were developed, 

 and were dependent on the pi'esence of 



streptom3^cin for multiplication. This de- 

 pendence was demonstrable not only in 

 vitro but also in vivo, since the organism 

 exhibited no \'irulence for mice unless strep- 

 tomycin was administered to the animals 

 after infection. Both variants retained the 

 characteristic sugar fermentations and type 

 specificity of the parent strain. 



The production of streptomjTun-depend- 

 ent strains has also been reported for a 

 number of other bacteria, including E. coli, 

 Ps. aeruginosa, B. subtihs, Staph, aureus, 

 and xl/. tuberculosis. 



In a study on the distribution of depend- 

 ent cells of E. coli in a broth culture of this 

 organism, Iverson and Waksman (1948) 

 found that one dependent cell was present 

 among each 1.5 billion normal sensitive cells. 

 Streptomycin, and not any accompanying 

 impurity, was reciuircd for growth of the 

 dependent organisms. IMannosidostrepto- 

 mycin and dihydrostreptomycin were also 

 effective in favoring growth of dependent 

 strains; but streptomycin that had been 

 inacti\'ated by cysteine and hydroxylamine 

 was ineffective, as were streptidine and 

 streptamine. 



Xewcombe and Xyholm (1950) have 

 shown that streptom3Tin-dependent fcms 

 of E. coli differ among themselves not only 

 in the degree of their dependence but also 

 with regard to other compounds that have 

 the capacity to replace streptomycin. De- 

 pendent strains give double mutants, arising 

 from a second mutation at the original locus, 

 thus forming a continuous series with respect 

 to degree of resistance. 



Szybalski and Cocito-\andermeulen 

 (1958) identified among the streptomycin- 

 dependent mutants of E. coli four nutritional 

 groups: (1) Growth is supported solely by 

 streptomycin, or by its dihydro, hj^droxy, 

 and desoxydihydro derivatives. (2) Catenu- 

 lin and neamine, the neomycin-related 

 antibiotics, can substitute for streptomycin. 

 (3) Streptobiosamine, a streptomycin deg- 



