PRODUCTION OF ANTIBIOTIC SUBSTANCES BY ACTINOMYCETES* ^†

Abstract

It has now been definitely established that a considerable proportion of all actinomycetes that can be isolated from soils or other natural substrates have the capacity of inhibiting the growth of, and even of destroying, bacteria and other microorganism.5 This was brought out emphatically in several of the surveys that have been made on the distribution of antagonistic properties among actinomycetes.1, 2, 4, 6, 10 Twenty to 50 per cent of all the cultures tested, whether freshly isolated from natural substrates or taken from culture collections, were found to possess antagonistic properties. The selective antimicrobial activities of actinomycetes differ greatly, both quantitatively and qualitatively, as could easily be demonstrated by their respective antibiotic spectra. The nature of the active agents or the antibiotics produced by these organisms depends upon the species; frequently upon the strain; the composition of the medium in which it is grown, and the conditions of cultivation.7, 8 Several distinct antibiotics have now been isolated from cultures of actinomycetes. Some, namely, actinomycetin, micromonosporin, myce-tin, and actinomyces lysozyme, have been only partly purified, whereas others, including actinomycin, proactinomycin, streptothricin, and streptomycin, have been isolated and crystallized. These substances differ greatly in their chemical structure, antimicrobial properties, toxicity to animals, and in vivo activity. Some of the antibiotics are produced in simple synthetic media; others are formed in complex organic substrates; still others, like streptomycin, require the presence in the medium of a specific nutritive substance, an “activity factor,” which is either a precursor or a prosthetic group of an enzyme system essential for the production of the antibiotic agent. Although this “activity factor” can be synthesized by Streptomyces griseus, its addition to the medium favors the rapid production of streptomycin. Streptomyces griseus can, therefore, grow well in substrates lacking the above factor, but the production amount of streptomycin is slower. However, when the mycelium thus formed is used as a constituent of fresh medium, streptomycin will be produced satisfactorily. The following investigations were undertaken for the purpose of elucidating several important problems pertaining to the production of antibiotic substances by actinomycetes, namely: (1) the use of selective antibacterial spectra, for the purpose of establishing the formation of antibiotic type compounds by various species or strains; (2) the ability of different organisms to produce the same antibiotic substance; (3) the variation in chemical nature and in biological properties of a given antibiotic substance, formed by different organisms or under different conditions of culture. A large number of cultures of actinomycetes were isolated from differently treated soils, composts, and other substrates, and were tested for their antibiotic potcncy against several test bacteria. For isolation purposes, either ordinary synthetic media (glucose-asparagine agar) or organic media (nutrient agar, peptone-glucose agar) were used. In some cases, the following special selective media were also employed: TB agar, consisting of 1.5 per cent washed agar, 0.05 per cent KH2PO4, 1 per cent glucose, and washed viable cells of Mycobacterium tubercdosis H37, as the sole source of nitrogen. (Because of the difficulty of suspending the tubercle bacilli homogeneously throughout the medium, growth of the antagonist upon the small cell aggregates, rather than the production of a clear zone of lysis, was taken as the criterion of attack upon the tubercle organism.) Bacterial agar, namely, washed mineral-glucose agar, as prepared above, to which centrifuged and washed cells of various gram-positive and gram-negative bacteria had been added. Streptomycin-enriched agar, comprising nutrient agar to which had been added 5 to 25 mg. of purified streptomycin per ml. of medium. The cultures of actinomycetes that were isolated from the TB agar were tested by the streak method. Nutrient agar and glucose asparagine agar were used, with Mycobacterium phlei as test organism. This saprophyte grows rapidly and can be handled more easily than the pathogenic mycobacteria. The reason for adding the H37 tubercle bacillus to the washed agar medium was that this organism allows better growth of antagonistic microorganisms (bacteria, fungi, and actinomycetes) than Mycobacterium phlei. Those actinomycetes that were found by the streak method to be most active against Mycobacterium phlei were then grown in nutrient broth and in glucose-asparagine media containing 0.25 per cent agar, in order to obtain better surface pellicle formation. After incubation for 8 days at 28° C, the filtrates were sterilized, by heating at 70° C for 10 minutes, and tested. The results presented in table 1 show that more actinomycetes were antagonistic to Mycobacterium phlei when tested on glucose-asparagine agar, than on nutrient agar. With the synthetic medium, twelve per cent of the cultures exhibited no antagonistic activity by the streak test, as compared with 35 per cent inactive cultures on the organic medium. The highest activity, namely 9 to > 9 mm. zone of inhibition, was ahown by 59 per cent of the cultures on the synthetic, and by only 24 per cent on the organic, medium. When these cultures were grown in liquid or in semi-solid agar media, and the filtrates tested, the reverse was found to be the case, the organic broth proving to be a better medium for the production of antibiotic substances than the synthetic medium. This may be due to the larger amount of growth on the organic medium. The percentage of actinomycetes active against Mycobacterium phlei was fairly high and was well correlated with the distribution of antagonistic actinomycetes previously reportcd for the gram-positive and gram-negative bacteria.4, 6 The enrichment of the soil with living or dead TB cells appeared to exert no appreciable stimulation upon the developnient of antagonists, as compared with the control soil. Although the enriched soils showed a marked stimulation of the total number of actinomycetes, these organisms did not necessarily posscss antagonistic properties.9 Many of the strains developing on the TB-washed agar plate, to the extent of about 20 per cent of the total number, were found to belong to the Streptomyces coelicolor group, which comprises organisms not antibacterial in nature. A number of actinomycetes were picked at random from plates prepared from various natural materials. These cultures were tested, by the streak method, against Eschaichia coli and Bacillus subtilis. The results presented in table 2 show that, as a rule, a much larger proportion of the cultures were active against Bacillus subtilis than against Escherichia coli. Some of the cultures had about cqual activity against both organisms, and one had greater activity against the gram-negative, than against the gram-positive, test-organism. Just as in the casc of the actinomycetcs active against Mycobacterium phlei, more organisms displayed antagonistic action, by the streak method, on glucosc-asparagine agar than on the nutrient agar. As compared to the 85 per cent of the cultures that showed lack of activity against Escherichia coli, when tested on the asparagine agar, 93 per cent of the cultures were inactive, when grown on nutrient agar; the corresponding percentages showing no activity against. Bacillus subtilis were 41 and 64 per cent. The cultures giving the highest activity against E. coli were 6 and 3 per cent, and against B. subtilis, 44 and 16 per cent, when tested on glucose-asparagine and nutrient agar, respectively. The reverse was true, when the culture filtrates were tested, nutrient broth proving to be a better medium than glucose-asparagine broth for the production of the antibiotic agents. The results obtained for the soil enriched with washed viable cells of gram-negative bacteria show that such treatment did not bring about any significant stimulation of actinoinycetes antagonistic to either Escherichia coli or to Bacillus subtilis, as compared with the unenriched control soil. This confirms the results reported elsewhere,9 concerning the doubtful significance of the soil enrichment procedure for the isolation of specific antagonistic organisms. Ninety-one cultures were selected at random and tested for their antagonistic properties against various bacteria, by the agar streak method. One or more of the following three media were used: nutrient agar, glucose-peptone agar, and glucose-asparagine agar. Of these cultures, 47 showed some activity against either, or both, of the gram-negative bacteria, on one or more of the above media. Tile distribution of the antibiotic activity of the cultures against the two bacteria is brought out in table 3. In general, there were more actinomycetes active, in the case of all three media, against Pseudomonas aeruginosa alone, than against Escherichia coli alone or against both bacteria. Glucose-aspkagine agar gave zones of bacterial inhibition which were wider than those observed on the organic media. The interesting phenomenon, in this connection, is that Pseudomonus aeruginosa, an organism which is usually recognized as highly resistant to most of the known antibiotic agents, was found to be more sensitive, at least as measured by the streak method, to the action of antagonistic actino-mycetes than the ordinarily more sensitive Escherichia coli. Whether this is due to the formation by these actinomycetes of as yet unknown substances active against Pseudomoms aeruginosa, or whether the presence of the cells of the antagonist influences, in some manner, the sensitivity of Pseudomoms aeruginosa, still remains to be determined. The fact that Streptomyces lavendulae showed, by the streak method, some activity against Bacillus mycoides, whereas the antibiotic streptothricin produced by S. lavendulae is inactive against this bacterium, may also be due to factors still unrecognized, at present. A detailed study was next made of the antagonistic properties of the actinomycetes that had been isolated from the washed agar media enriched with Escherichia coli and Bacillus subtilis cel!s. In all, 44 cultures were tested by the streak method, using three different media and four test bacteria. The results (table 4) show that greater inhibition was exhibited by the actinomycetes on all the media against gram-positive, than against gram-negative, bacteria. There were marked differences, however, in the sensitivity of the different bacteria within each group. When the two aerobic spore formers were compared, Bacillus mycoides was found to be more resistant than Bacillus subtilis to most of the actinomycetes, both on nutrient agar and on glucose-tryptone agar. On the synthetic medium, however, the two spore formers exhibited very little difference in sensitivity to the various actinomycetes. A more detailed examination was made of the behavior of these two test organisms toward a number of actinomycetes. These results can be only briefly summarized here. In the case of nutrient agar, 20 cultures were more active against Bacillus subtilis than Bacillus mycoides; 7 were equally active against botth bacteria; and 3 were more active against B. mycoides than against B. subtilis. The corresponding ratios of sensitivity of the two bacteria, to various actinomycetes grown on glucose-asparagine agar, were 9:20:9 (greater: equal: lower activity against Bacillus subtilis), and on glucose-tryptone agar, 22:6:10. A comparison of the sensitivity of the two gram-negative bacteria brought out the fact that, on all the media used, Aerobacter aerogenes was more resistant to the action of actinomycetes than was Escherichia coli. But here, as well, a more detailed examination of the plates revealed certain striking differences: on nutrient agar, 20 cultures were more active against Escherichia coli than against Aerobacter aerogenes, 2 were active alike against both bacteria, and. 3 were more active against A. aerogenes than against E. coli. The corresponding ratios for the sensitivity of the two bacteria on the other two media were 35:1:2 and 14:0:2. Streptomycin is the most important antibiotic so far obtained from actinomycetes. It was found to be produced by two strains of Streptomyces griseus isolated, in 1943, from two distinct sources. The original culture of Streptomyces griseus, which has been kept in the culture collection since 1915, does not possess now the property of producing streptomycin, and, when recently tested, was found to exert only a very weak antibiotic effect against Bacillus subtilis and none against Escherichia coli. In order to determine whether the capacity to form streptomycin is characteristic only of freshly isolated strains of Streptomyces griseus, and how widely distributed this capacity is among members of the S. griseus group of actinomycetes, a number of cultures, recognized by macroscopic appearance to be typical S. griseus, or to be forms closely resembling it, were isolated from different substrates. These cultures were first tested for antibiotic activity by the agar streak method, using nutrient agar and glucose-tryptone agar. Streptomycin is produced largely on the first medium and only to a limited extent on the second. The results presented in table 5 show that the various Streptomyces griseus-like cultures thus isolated were far more active against gram-positive, than gram-negative, bacteria. They showed greater activity against gram-negative bacteria on nutrient agar, than on glucose-tryptone agar. Even that activity, however, was limited in nature, as compared with the streptomycin-producing strains of Streptomyces griseus. When grown in liquid media, none of the freshly isolated Streptomyces griseus strains produced the typical streptomycin. Only one culture, namely, 25-G, was capable of yielding an active filtrate, as shown in table 6. The antibiotic nature of this filtrate appeared to be different from both strep-tothricin and streptomycin, as shown by its antimicrobial spectrum. These results prove the principle that the production of a given antibiotic by actinomycetes is characteristic of the strain of organism, rather than of the species. As a result of a comprehensive study of the antimicrobial properties of a large number of cultures of actinomycetes isolated from different sources, and tested on the various media listed above, the general types of activity can be summarized as follows: Certain cultures of actinomycetes are largely active against fungi, but not at all, or only to a limited extent, against bacteria. Some cultures are active, primarily, against other actinomycetes, and only to a limited extent against bacteria. A large number of actinomycetes affect mycobacteria, notably Mycobacterium phlei, but not at all, or only to a limited extent, various other bacteria. Many actinomycetes are active against gram-positive bacteria and mycobacteria, but not against gram-negative bacteria. Various actinomycetes are active against both gram-positive and gram-negative bacteria, with certain marked variations, namely: some are more active against gram-positive bacteria; some are more active against gram-negative bacteria; some are equally active against both groups of bacteria.The antimicrobial properties of a given actinomyces culture also depend upon the composition of the medium in which it is grown. One organism is equally active against Escherichia coli and Aerobacter aerogenes on nutrient agar, whereas it inhibits only E. coli on glucose-asparagine or glucose-tryptone media. Another culture is equally active against Bacillus mycoides and Bacillus subtilis, on nutrient agar only. A third culture may produce equal inhibition on all three media, against the two gram-positive bacteria. A fourth culture, which is more active against Bacillus subtilis on glucose-asparagine agar, may be more active against Bacillus mycoides on nutrient agar. Finally, there is a group of properties characteristic of certain actinomycetes, namely, activity against bacteriophages; this activity may or may not extend to the host bacteria. Some of the specific antimicrobial agents or antibiotics produced by actinomycetes have been isolated from suitable cultures of the organisms, either in a crude or in a pure state. A much larger number of substances still remains to be isolated, however, since it can be established beyond any doubt, by testing of the crude filtrates, that such substances exist. Among those agents that have not as yet been isolated, but the formation of which is unquestioned, the following may be given here only brief consideration: Substances active against Mycobacterium phlei, but not aguinst gram-positive or gram-negative eubacteria. A certain strain of Streptomyces lavendulae (No. 3465), grown in glucose tryptone medium, was found to produce in stationary culture, after 6 and 11 days’ incubation, a factor which was heat stable and which inhibited Mycobacterium phlei in a dilution of 1:300. This filtrate had no activity against Bacillus subtilis, Bacillus mycoides, Escherichia coli, and Pseudomonas aeruginosa. The same observation was made in the case of another culture, which was also found to belong to the Streptomyces group. Substances active against bacteriophages. Certain cultures of actinomycetes have been found to produce factors present in the culture filtrates, which were devoid or nearly devoid of antibacterial properties, at least in the concentration tested, yet which inhibited the lysis of bacteria by phages. Whether the action of these factors is directed against the phage particles or against the host cells, resulting in the alternation of their metabolism so as to prevent phage infection or multiplication, still remains to be determined. The production of these antiphage filtrates is dependent upon the composition of the medium, in a manner similar to the production of true antibiotics. The antiphage action of the filtrates is also specific, some phages being affected, and not others. This selectivity is comparable to that of the selective antibacterial action of known antibiotics. Agents active against animal viruses. Certain culture filtrates of various actinomycetes appear capable of exerting a slight, but definite, inactivating effect upon animal viruses. In one case, the filtrate was not antibacterial, although test bacteria were found to be inhibited on certain media, as measured by the streak method. From the active filtrate of another culture, actinomycin was isolated, as the virus-inactivating agent. Antibiotics active against bacteria and fungi. Although several antibiotics have already been isolated from actinomycetes, still others remain to be obtained. Those isolated, so far, vary greatly in their chemical nature and in their selective antimicrobial action. Among the problems bearing upon the isolation of antibiotic substances, two deserve special consideration: It is now well established that certain actinomycetes produce more than one antibiotic substance. Streptomyces griseus produces streptomycin and another, still undescribed factor, present in the mycelium of the organism. This factor is soluble in organic solvents and differs markedly, in its antibacterial spectrum, from streptomycin. The same antibiotic may be produced by different species of actinomycetes. Certain unpublished data from our laboratory have established, beyond doubt, that actinomycin is produced by several species of Streptomyces. The same is true of streptothricin. A given antibiotic may, therefore, be identical, even when produced by different actinomycetes, as shown by its chemical composition and antibiotic spectrum. In the case of other antibiotics, however, certain differences in the specific antibacterial spcctra of the same type of substance produced by different organisms appear to point to differences in the chemical structure of the antibiotic. The present information is too limited, however, to justify broad generalizations. However, it opens some interesting possibilities. The results of a survey of a large number of actinomycetes, isolated from different substrates and cultured on different media, permit certain broad generalizations: A large proportion of the actinomycetes inhabiting natural substrates have the capacity to inhibit the growth of bacteria and other microorganisms. The ability of these actinomycetes to exert an inhibiting effect upon microorganisms is highly specific. This selectivity depends, not only on the strain of organism, but also on the medium in which it is grown and conditions of growth. Antagonistic actinornycetes produce a variety of antibiotics that vary in chemical nature, in antimicrobial action, in toxicity to animals, and in their chemotherapeutic potentialities. The antibiotics that have, so far, been isolated from actinomycetes vary in the degree of purification. Some are crude preparations, whereas others have been crystallized, and considerable information has been gained concerning their chemical nature. They include: actinomycctin, actinomyces lysozyme, actinomycin, micromonosporin, streptothricin, streptomycin, and mycetin. Some actinomycetes produce more than one antibiotic substance. Some antibiotics are produced by several different organisms. There is some evidence, however, that, although the same general type of substance may be formed by t8he various cultures, it may not possess exactly the same antibiotic spectrum; this suggests the possibility of variations in the chemical structure of the agent.