1942b), activity in vivo (Metzger, Waksman, Pugh, 1943; Robinson, 1943),

Transcription

1 PRODUCTION AND ACTIVITY OF STREPTOTHRICIN1.2 SELMAN A. WAKSMAN New Jersey Agricultural Experiment Station, Rutgers University Received for publication April 24, 1943 INTRODUCTION The actinomycetes represent a widely distributed group of organisms which comprise many forms possessing marked antagonistic properties against bacteria, fungi and other microorganisms (Waksman, 1941; Waksman, Horning, Welsch and Woodruff, 1942; Welsch, 1942). The antibiotic substances produced by actinomycetes vary greatly in chemical nature, in specific antibacterial action, and in toxicity to animals. Actinomycin is by far the most toxic of these substances and streptothricin apparently the least. Because of this low toxicity, as well as its solubility in water, selective action against gram-negative and gram-positive bacteria (Waksman and Woodruff, 1942a; Waksman and Woodruff, 1942b), activity in vivo (Metzger, Waksman, Pugh, 1943; Robinson, 1943), relative stability and heat resistance, all of which make the antibacterial properties of streptothricin of particular importance, it was selected for further study. Streptothricin is produced by an aerobic, conidia-forming species of actinomyces which was identified, on the basis of its pigmentation and certain cultural characteristics, with an organism, isolated from the soil many years prenvously, as Actinomyces lavendulae (Waksman, Horning, Welsch and Woodruff, 1942; Waksman and Woodruff, 1942a). This organism is grown in simple media contamning glucose or starch as a source of carbon, a protein digest (known under the trade name of tryptone) or different amino acids (glycine, glutamic acid) as a source of nitrogen, with limited concentrations of inorganic salts and tap water. The media are distributed in shallow layers in flasks, sterilized as usual, inoculated, and incubated at 28 C. for 7-12 days. The antibiotic substance is readily secreted into the metabolic solution. It is isolated from the culture 'filtrate by adsorption on active charcoal (norit-a), from which it is removed by treatment with acidified alcohol. The alcoholic extract is neutralized with NaOH solution and treated with 10 volumes of ethyl ether. This results in a highly concentrated 1 Journal Series Paper, New Jersey Agricultural Experiment Station, Rutgers University, Department of Soil Microbiology. 2 The author gratefully acknowledges the assistance obtained in this work from several collaborators. Dr. W. Kocholaty, of the University of Pennsylvania, first isolated in our laboratory the strain of the organism producing streptothricin and first tested it for its antagonistic action; Dr. E. Horning isolated in our laboratory several other species which produced streptothricin-like substances; Dr. H. B. Woodruff, now with Merck & Co., collaborated in the earlier work on the development of suitable media for the growth of the organism, as well as in the isolation and concentration of the streptothricin; Mr. A. Schatz, now with the U. S. Army, assisted in some of the experimental work reported in this paper. The author is also indebted to Merck & Co. for the concentrated and purified streptothricin used in some of these experiments. 299

2 300 SELMAN A. WAKSMAN aqueous solution of streptothricin. The active solution may be further concentrated by evaporation at reduced pressure. Although the original culture of A. kavendulae isolated from the soil in 1915 is still available in the culture collection, some of its cultural characteristics have become changed to a considerable extent, so that the culture now differs from its original description. On careful testing, this culture was found to possess antagonistic properties, but only to a very limited extent; it was largely active against various gram-positive spore-forming bacteria. Considerable difficulty in obtaining the active substance has been experienced by various investigators to whom cultures of streptothricin-producing strains of A. lavendulae have been submitted. Because of this and also because of the paucity of information concerning the best methods for culturing actinomycetes as well as of their physiological properties, the following investigations were undertaken. The primary purpose was to establish the optimum conditions for the production of streptothricin, the relation of this substance to the growth and physiology of the organism, strain variation and its formation by related organisms, and the mechanism of its production.3 EXPERIMENTAL. Methods Actitnomyce8 lavendulae is capable of producing streptothricin only when grown in very shallow layers in stationary cultures,,the presence of abundant oxygen being essential to this process. A culture medium consisting of 10 grams anhydrous glucose (or starch), 5 gm. protein digest, 0.5 gm. K2HPO4, 0.5 gm. NaCl, and 0.1 gm. FeSO4, and 1000 ml. tap water, is inoculated with the lavender to rose-lavender colored spores of the organism grown on synthetic (glucoseasparagine) agar for 5-10 days, at 280. When the inoculated spores drop to the bottom of this medium, they germinate and grow there into colonies, but do not give rise to a surface pellicle or produce only a ring on the surface along the glass; under these conditions only a little streptothricin is produced. The addition of 0.25 per cent agar to the medium was found to overcome this diffi culty; however, the presence of agar makes the process of filtration of the cultures and the extraction of the active material rather cumbersome. In most of this work, 250 ml. portions of medium were placed in 1 liter Erlenmeyer flasks and were sterilized either in flowing steam, when glucose was used, or under pressure with starch as the carbohydrate source. The ability of actinomycetes to grow in a submerged condition and to form streptothricin, provided the cultures are aerated by agitation, has recently been demonstrated by Woodruff and Foster (1943). A comparison has, therefore, been made of the production of streptothricin in stationary and in shaken cultures. The antibiotic activity of streptothricin was measured by the agar plate- ' The results presented in this paper and those of Woodruff and Foster presented elsewhere (1943) may be considered as supplementing one another, since they were carried out with the same organism, but in different laboratories.

3 PRODUCTION AND ACTIVITY OF STREPTOTHRICIN dilution method, using several test organisms (Waksman and Woodruff, 1942b). The results are reported in terms of dilution units, as expressed by the ratio of the volume of test medium used (10 ml. agar) to the amount of culture required to give complete inhibition of the test organism. If partial inhibition was obtained in one dilution and complete inhibition in the next lower dilution, the end point was interpolated between these two dilutions. It has been pointed out elsewhere (Waksman and Homing, 1943) that this method has certain advantages and disadvantages, when compared with the dilution culture and agar-cup methods. When it is desired to employ several test organisms, especially when the active substance has not yet been isolated in a pure state, and when accurate quantitative yields are of only secondary consideration, this method is far superior to any other now in use for the study of antibiotic substances. The four test organisms commonly employed in this laboratory for the study of antibiotic substances are Escherichia coli, Bacillus subtilis, Bacillus mycoides and Sarcina lutea. Streptothricin has very little activity against the third and itq action against the fourth is similar to that upon E. coli. Hence, only the results obtained against the first two organisms are reported here. It is of particular interest to note that the ratio of activity of streptothricin against E. coli and B. subtilis is about 1: 5 to 1: 10. Although in most cases the results are presented in terms of both units, only one of these can readily be used as the test organism for measuring the activity of streptothricin. Since different strains obtained from different laboratories may vary in their sensitiveness, only well-defined test strains are to be employed. Strain specificty Several cultures of actinomycetes isolated from soil and from dust were found capable of producing streptothricin or a streptothricin-like substance (Waksman, Homing, Welsch and Woodruff, 1942). The culture of A. lavendulae largely used in this work was found to be made up of different strains which varied in their capacity to produce the antibiotic agent, as shown in table 1. These strains were isolated from the mother culture by plating suspensions of spores from agar slants and picking individual colonies. Some of the strains thus isolated, notably Nos. 3, 8, 12 and 14, gave fairly high antibacterial activity, whereas other strains, such as Nos. 4, 5,.6 and 10, produced no activity at all or only very little when grown in the same medium and under the same conditions of culture. Because of the great variation in the growth of the organism in stationary cultures and the resulting yields of streptothricin, it is not possible to state definitely that these variations are of a permanent nature. Considerably less variation was obtained in shaken cultures. Two of the above strains, Nos. 8 and 14, were used chiefly in the experiments reported here. These two strains also varied from one another, as brought out in table 2. In shaken cultures, No. 8 gave greater activity, both in the tryptone and in the glycine media, than did No. 14; in stationary cultures the reverse was true. 301

4 302 SELMAN A. WAKSMAN TABLE 1 The production of streptothricin by different strains of A. Iavendulae Glucose-tryptone-soft agar medium STRAIN NUMBICR E. COUI D. SUTLS units ##its > > TABLE 2 Comparative activity of two strains of A. lavendulae 1 per cent starch media STRAN SOURCE OA O IN ION GRO NUMME NITROGEN CULTURES wu'o RW E. coli B. subliis u g. S 100mṚditS 8 Tryptone Shaken , Tryptone Shaken Tryptone Shaken , Tryptone Shaken Glycine Shaken Glycine Shaken Glycine Shaken Glycine Shaken Tryptone Stationary Tryptone Stationary Glycine Stationary Glycine Stationary Influence of nutrition of the organism upon the production of streptothricin In view of the importance of the carbon and nitrogen sources in the growth of A. kavendulae and in the production of the streptothricin, the results of several typical experiments are reported here. Four different forms of nitrogen, in concentrations of per cent, with

5 PRODUCTION AND ACTIVITY OF STREPTOTHRICIN 303 starch (1 per cent) as the added source of carbon were compared, using stationary cultures. The maximum production of streptothricin was obtained in seven 10 z a 0 TRA YAP?QNEV-A -- ASPAPQA 6/*VN LvtrAM/C AC/P ---' LYC/NE h.5 I I-- 4 l wwr* w~~--w-71 %s-- / -- 0v DAYS OF INCUBATION PIG. 1. INFLuENCE of DIFFERENT Fo0ma of NITROGEN UPON THE PRODUCTION OF STEPToTERC1N-E. COu UNITS 30 C~LUCO.$E _. /ALTO3E S ratch 2 z 20 -J hii I- U 1 FIG shaa/rnevniruars DAYS OF s5 TArAOAmRY CLTraqES INCUBATION INFLUENCE OF DIFFERENT CABEON SOURCES UPON THE PRODUCTION OF STREPTOTHRICIN days, followed by a rapid decrease. The protein digest tryptone and glutamic acid proved to be the best sources of nitrogen; however, growth was not uniform in all the flasks, especially in the glutamic acid cultures, some showing little if any

6 304 SELMAN A. WAKSMAN activity. The activity with glycine was not so high but tended to remain constant (fig. 1). Shaken cultures were found to give similar results, with a slight difference, tryptone giving the highest activity in 3 days (200 E. coli units) and glycine in five days (150 units); asparagine gave the lowest activity in shaken cultures as well. 4o - GLYCINE -STARCH MEDIUM ~)30 TIRYPTONE-STARCH MEDIUM lll. DAYS OF INCUBATION FIG. 3. INFLUENCE OF NITROGEN SOURCES UPON THE GROWTH OF A. LAVENDUIAE AND FORMATION OF STREPTOTHRICIN A comparative study of the effect of different carbon sources upon the produc tion of streptothricin gave the highest activity for starch, followed by maltose or by glucose. Starch had the added advantage that one could do away with the use of agar in stationary cultures, since the spores tended to remain on the surface of the medium and form a pellicle more readily; the starch was gradually decomposed by the growing organism. No reducing sugar was ever demonstrated in the starch media; either the starch was hydrolyzed to the dextrin stage

7 PRODUCTION AND ACTIVITY OF STREPTOTHRICIN and directly assimilated by the organism, or it was broken down to compounds not readily demonstrated, or the sugar was consumed as soon as formed. Only limited acid production by the organism was demonstrated in the above media. A comparison of the three best carbon sources, with tryptone (0.5 per cent) as the source of nitrogen, upon the production of streptothricin in shaken and in stationary cultures, is brought out in figure 2. The shaken cultures gave much greater activity in a shorter period of time. A detailed study of the growth of the organism, as measured by dry weight of mycelium and production of streptothricin, using shaken cultures, pointed to a correlation between the two, as illustrated in figure 3. This parallelism tends to indicate that the formation of the antibiotic substance is a function of the growing organism, under favorable conditions, and not of the autolyzing cells. TABLE 3 Influence of depth of medium upon the production of streptothricin Starch-glycine medium, stationary cultures VOLUME 0o 3mE PEE NCUAION ACTIITY E. coli B. sublilis i. days units units > > Influence of aeration upon the production of streptothricin Among the various factors influencing the formation of streptothricin, none is more important, in addition to the essential nutrients, than the air supply. This is illustrated by the following experiment. Different volu-mes of the starch-glycine agar-free medium were placed in 1 liter Erlenmeyer flasks, so as to obtain varying depths of medium. The fasks were inoculated with a spore suspension of an active strain of the organism and incubated under identical conditions, in a stationary state. The antibiotic activity of the cultures was measured after varying periods of incubation (table 3). An inverse proportion was obtained between the volume of the medium in the flask and the production of streptothricin. The shallowest layer, 100 ml. per liter flask, gave as high activity as that obtained in shaken cultures using the same medium. Various other surfaces, such as cotton, paper, peat and straw, were compared for the growth of A. tavendulae and for the production of the streptothricin. These substrates were saturated with the glucose-tryptone medium. The best activity was obtained on the cotton medium. However, none proved superior

8 306 SELMAN A. WAKBMAN to, if as good as, the ordinary starch-tryptone or starch-glycine medium, either in very shallow stationary layers or in shaken cultures. Influence of temperature and other factors The temperature of incubation of the cultures is known to have an important effect upon the production of antibiotic substances by microorganisms. This was also found to hold true of streptothricin, as brought out in table 4. The highest activity was obtained at the lowest temperature of incubation, namely at room temperature. A more detailed study was made of the production of streptothricin at 23 and 2800., using varying incubation periods and sources of nitrogen; the lower temperature invariably gave the greater antibiotic activity. TABLE 4 Influence of temperature upon the production of streptothricin Station,ary cultures ACTIVITY TEMPERATURE 01 INCUBATION PERIOD O INCUBATION.. E. coli B. snlilis C. days units units < <10 <10 A detailed study of the effect of reaction upon the growth of A. lavendulae and the production of streptothricin in tryptone-starch medium brought out the surprising fact that variations in ph of the original culture between 5.0 and 7.5 had very little effect. In the more acid cultures, there was a gradual increase in the ph value with the growth of the organism, due to the production of ammonia from the tryptone; in the media with the less acid reactions, there was at first a decrease in ph value, soon followed by an increase. The addition of yeast extract to the standard tryptone-starch medium either had no effect at all or suppressed streptothricin production. Correlation between growth of A. lavendulae and the production of 8treptothricin Attention has been directed above to the close correlation between the growth of the organism and the production of streptothricin. The results of another experiment, reported in table 5, show that in stationary as well as in shaken

9 PRODUCTION AND ACTIVITY OF STREPTOTHRICIN cultures, with tryptone or with glycine as sources of nitrogen, growth and activity reached a maximum and then declined, the maximum for the first preceding somewhat that of the second. Since the nitrogen in the dry mycelium varied between 7 and 9 per cent, growth may be expressed by the dry weight of the mycelium or by its nitrogen content. One must, therefore, conclude that the production of streptothricin is not a result of autolysis of the mycelium but is due to cell nutrition or to cell synthesis. This renders the mechanism of the production of this substance distinct from that of tyrothricin, for example, which is a result of autolysis of the bacterial cells, or of penicillin, which is produced at a much later stage of growth of the organism, namely, when it reaches an alkaline reaction. TABLE 5 GrOwth of A. lavendulae and production of streptothricin Tryptone-starch medium ACTIVITY AERATION INCUIBATION STARCH DRY WEIGHT NTROGEN IN LEFT OF MYCELIUM HYCELIUM- E. coli B. subtilis days ing. mg. nuiss units Shaken Shaken Shaken Shaken Shaken Shaken Stationary Stationary.10 0* Stationary.14 0* * Small amount of starch left at bottom of medium. 307 The efficiency of utilization of the carbon and the nitrogen by A. lavendulae is very high. At the maximum growth stage, 65 per cent of the nitrogen in the glycine added to the medium became converted into actinomyces cell substance. Since as much as mg. of mycelial growth was obtained from 1 gm. of raw starch, the efficiency of utilization of the carbon, considering the carbon content of the starch as well as of the glycine, is about 40 per cent. Bacteriostatic spectrum of streptothricin It has already been demonstrated (Waksman and Woodruff, 1942), by the use of crude culture filtrate as well as concentrated crude preparations, that streptothricin has a selective bacteriostatic effect against both gram-negative and grampositive bacteria. Additional results are reported here, using a salt-free, purified streptothricin preparation (table 6); these results are definitely similar to those presented previously. The selective action is very striking. B. subtilis is by far the most sensitive organism to the action of streptothricin, Bacillus mycoides

10 308 SELMAN A. WAKSMAN and BaciUus cereus are less sensitive than most gram-negative bacteria. The same is true of the variation among the gram-negative bacteria, as illustrated by the two extremes, namely Shigella gallinarum and Pseudomornas fluorescens. Whereas Sarcina lutea is about as sensitive to streptothricin as E. coli, the ratio in sensitivity between the two against actinomycin is about 10,000:1. TABLE 6 Antibacterial spectrum of purified streptothricin ~~UNITS Of ORlGANIZSMX ORGAMSM UNITS OF ACTIVITY OGNS AORGANISMATITNSY Escherichia coli ,000 Bacillus subtilis.750,000 Aerobacter aerogens ,000 Bacillus mycoides. <10,000 Pseudomonas fluorescens... <1,000 Bacillus cereus... <1,000 Shigella gallinarum ,000 Staphylococcus aureus.. 200,000 Shigella dysenteriae, ,000 Sarcina lutea.100,000 Shigella dysenteriae, ,000 Shigella dysenteriae, ,000 Shigella paradysenteriae... 5,000a DISCUSSION Among the thirty or more antibiotic substances which have already been isolated from different groups of microorganisms, only four so far either have found practical application or show promise of such application, because of their low toxicity to animals and their in vivo activity against pathogenic bacteria. These are, in order of their discovery, 1. pyocyanase, a substance which has had a varied history and for which, at various times, many clims have been made but not always substantiated; 2. penicillin, which recently has found quite extensive practical application; 3. tyrothricin (gramicidin and tyrocidine), which has become recognized as an agent with certain definite potentialities; 4. streptothricin, a substance active not only against certain gram-positive but also against gram-negative bacteria. These four substances have been isolated from cultures of different microorganisms, representing gram-negative non-spore-forming bacteria, fungi, grampositive spore-forming bacteria, and actinomycetes, in the order listed above. These substances vary not only in origin, but also in chemical nature, solubility, and specific antibacterial activity, i.e., their respective bacteriostatic spectra. The limited information concerning their mode of action upon bacteria definitely suggests great variation in this respect as well. Because of these variations, one can easily understand why the substances should vary also in their activity in the animal body. The latest addition to the list of active compounds, streptothricin, is one of a series of five substances which have so far been obtained from actinomycetes, a group of organisms containing a large percentage of forms with antagonistic properties. Streptothricin is a nitrogen-containing base. It is insoluble in ether and soluble in water. It has a low toxicity to animals. It is active both against

11 PRODUCTION AND AClIVITY OF STREPTOTHRICIN 309 certain gram-positive and gram-negative bacteria. It has been isolated in a concentrated form, although it has not been crystallized as yet. The presence of certain amino acids or a mixture of amino acids (polypeptides) in the medium is favorable to the formation of streptothricin. Although the organism grows well with NaNO3 as a source of nitrogen, it produces little of the antibiotic substance. One may, therefore, conclude that streptothricin is largely formed from one or more of the amino acids. The fact that larger amounts of the active substance are produced in the presence of an additional carbohydrate, such as starch, maltose or glucose, does not detract from this conclusion. The carbohydrate may serve only as a nutrient for cell synthesis, whereas the nitrogen source contributes to the formation of streptothricin. The direct parallelism between the growth of the organism and the production of the substance serves to substantiate this theory. Aerobic conditions are absolutely essential to the production of streptothricin. The results presented in this paper amply justify this conclusion. In spite of the fact that the organism may produce in stationary cultures a mass of growth on the bottom of the flask, comparatively little streptothricin is produced if no pellicle is formed. Acid conditions of the medium do not interfere with the production of streptothricin; this may possibly be due to its basic character, since this substance forms salts readily and is isolated by removal in an acid solution. Several actinomycetes were found to produce an antibiotic substance which was, on the basis of its selective antibacterial action, similar to streptothricin. It is not known as yet whether the differences observed in the production of this agent by the different organisms are rather quantitative in nature or whether there is actually a chemical difference in the nature of the substance produced by the various organisms. SUMMARY Different strains of Actinomyces klvendulae isolated from soil, dust or an active mother culture were found to vary in their ability to produce the antibiotic substance streptothricin. The formation and accumulation of streptothricin is largely controlled by the conditions of nutrition of the organism and the aeration of the culture. Streptothricin is formed largely when a protein digest or certain amino acids, namely, glutamic acid, glycine or asparagine, are used as sources of nitrogen. When glucose is used as the additional carbon source, a small amount of agar (0.25 per cent) was found to favor the growth of the organism and results in the maximum production of streptothricin. When starch is used in place of glucose, no agar is required. The starch is consumed rapidly, but no reducing sugar could be demonstrated. Supplementary additions of yeast extract and similar materials had an injurious effect upon the production of streptothricin in stationary cultures. Aeration was found to be one of the most essential factors in the production of streptothricin. Either very shallow layers must be used or the culture must be

12 310 SELMAN A. WAKSMN kept in a well-aerated submerged state. In the case of stationary cultures, the formation of a surface pellicle is essential for the optimum production of the antibiotic substance. An incubation temperature of C. for 7-10 days proved to be optimum for the production of streptothricin in stationary cultures; for shaken cultures growing at 2800., 2-4 days incubation was sufficient. The reaction of the medium, ranging between ph 5.0 and 7.5, had little effect upon growth and streptothricin formation. With the less acid reactions, the culture medium first became acid, as a result of the growth of the organism, and then changed to alkaline. A certain parallelism was found to eist between the growth of the organism and the formation of streptothricin. This antibacterial substance is a metabolic waste product, which is apparently produced from certain amino acids or polypeptides. REFERENCES METZGER, H. J., WAKSMAN, S. A., AND PUGH, L. H In vivo activity of streptothricin against Brucella abortu. Proc. Soc. Exptl. Biol. Med., 51, ROBINSON, H Studies on the toxicity of streptothricin. In press; Thesis. Rutgers University. WAxKSmAN, S. A Antagonistic relations of microorganisms. Bact. Revs., 5, WAKSMAN, S. A., AND HORNING, E. S Distribution of antagonistic fungi in nature and their antibiotic action. Mycologia, 35, WAKsMAN, S. A., HORNING, E. S., WELSCH, M., AND WOODRUFF, H. B Distribution of antagonistic actinomycetes in nature. Soil Sci., 54, WAKSMAN, S. A., AND WOODRUFF, H. B. 1942a Streptothricin, a new selective bacteriostatic and bactericidal agent, particularly active against gram-negative bacteria. Proc. Soc. Exptl. Biol. Med., 49, WAKSMAN, S. A., AND WOODRUFF, H. B. 1942b Selective antibiotic action of various substances of microbial origin. J. Bact., 44, WELSCH, M Bacteriostatic and bacteriolytic properties of actinomycetes. J. Bact., 44, WOODRUFF, H. B., AND FOSTER, J. W Cultivation of actinomycetes under submerged conditions, with special reference to the formation of streptothricin. J. Bact., 45, 30; Arch. Biochem. In press.