ISOLATION AND IDENTIFICATION OF ANTIBIOTICS 



41 



clium may also affect the ease of puritication. 

 Thus, it is wise to have close cooperation 

 between workers invoh'ed in media cle\'elop- 

 ment and those who carry out the extraction 

 of an antibiotic. 



Methods of Characterization 



The characterization of an antibiotic in- 

 volves the determination of its properties: 

 physical, chemical, and biological. Plants, 

 animals, and other li\'ing systems camiot be 

 easily classified on the basis of a single char- 

 acteristic which would be different for each 

 ]i\-ing form. Similarly, ouv cannot usuall}' 

 differentiate between the products of li\'ing 

 systems on the basis of one single criterion; 

 this is true particularly of the \-arious anti- 

 biotics. A number of properties should be 

 studied, and an^^ comparison between anti- 

 biotics must be based on a number of char- 

 acteristics. 



It is impossible to make a list of "classical" 

 criteria which would always permit one to 

 differentiate between two antiliiotics and 

 which would always be adeciuate for the full 

 characterization of an antibiotic. A list 

 satisfactory at one time will be inade(iuate a 

 few years or even a few months later. Scien- 

 tific methods are forever changing. The 

 investigator must be adaptable and refrain 

 from taking a dogmatic stand. 



We will discuss various criteria which ha^'e 

 been used in the characterization of anti- 

 biotics. In the keys found in Part H of this 

 book we will try to use these characteristics 

 to single out each antibiotic which had been 

 described by the time that the compilation 

 was completed. This will not always be 

 possible, since some substances have not 

 been described in sufhcient detail. In such 

 cases we simply group similai' substances 

 together. 



Nature of Antibiotic-producing Organisms 



It has been observed that certain types of 

 cultures tend to form certain types of anti- 



biotics. For example, members of the Strcpto- 

 myces griseus group ha\'e been found to 

 produce streptomycin, cycloheximide, gri- 

 sein, and candicidin, to name only a few. 

 Similarly, strains of S. fradiae tend to pro- 

 duce antibiotics of the neomycin tN'pe and 

 fradicin. However, the ability of a cultiu'e to 

 produce a particulai' antibiotic is hardly a 

 criterion for the identification of the organ- 

 ism or of the antibiotic. This is due partly to 

 the confused state of the systematics of this 

 group of organisms. Further, it has been 

 shown that the same antibiotic can be pro- 

 duced by a \-ariety of actinom^'cete species 

 or e\'en by \'arious genera. In this respect 

 the paper of Fishei- and coworkers (19ol) is 

 of interest. These authors isolated and 

 chemically characterized actinomycin from 

 a strain of Micromonospora. This antil)iotic 

 is usually j^roduced l)y streptomycetes and 

 is known to be a mixture of closely related 

 substances. It is then conceivable that the 

 actinomycin elaborated by the Micro- 

 monospora might ha\'e been different from 

 the actinomycins of the streptomycetes. But 

 any difference must ha\'e been minor indeed, 

 since the amino acid content of the Micro- 

 monospora actinomycin was not unusual. It 

 is also unlikely that a Micromonospora would 

 be confused with a Strcptomyces. In this light, 

 one should consider with reserve the state- 

 ment of Krassilnikov (1960) that authors 

 who claim that the same antibiotic can be 

 produced by different species "are mistaken 

 either in speci(>s identification or in the 

 accuracy of detei'mining the nature of the 

 antibiotic." 



Teillon (1*.)."):)) suggested a scheme for the 

 separation of antibiotic-producing organ- 

 isms on the basis of their sensiti\'ity to 

 known antibiotics (Table l(i). 



A ntim icrobial Spectra 



The range of antimicrobial activity of any 

 antibiotic should be determined against 

 aerobic l)acteria, both gram-positive and 



