ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Aug. 1974, p. 144-149Copyright 0 1974 American Society for Microbiology

Vol. 6, No. 2Printed in U.S.A.

Antibiotic G-418, a New Micromonospora-ProducedAminoglycoside with Activity Against Protozoa and Helminths:Fermentation, Isolation, and Preliminary Characterization

G. H. WAGMAN, R. T. TESTA, J. A. MARQUEZ, AND M. J. WEINSTEINMicrobiology Division, Schering Corporation Bloomfield, New Jersey 07003

Received for puWlication 25 March 1974

Antibiotic G-418 is a new aminoglycoside produced as the major component bya new species of Micromonospora, M. rhodorangea NRRL 5326. The antibiotic isprepared by submerged fermentation in a soybean-dextrin medium. AntibioticG-418 is adsorbed on a cationic-exchange resin and separated from otherimpurities by passing it down a Dowex (1 x 2) resin column. The antibiotic,which contains 2-deoxystreptamine, has broad-spectrum antibacterial activityand is highly active against protozoa, amoebae, tapeworm, and pinworminfections in mice. This report describes the taxonomy of the organism, andfermentation, isolation, and preliminary characterization of antibiotic G-418.

A new antibiotic, G-418, has been foundwhich has broad-spectrum antibacterial activityin vitro and in vivo but is of special interest dueto its high level of activity against protozoa,cecal amoebiasis, and helminths (G. H. Wag-man, R. T. Testa, J. A. Marquez, J. A. Waitz,and M. J. Weinstein, Prog. Abstr. Intersci.Conf. Antimicrob. Ag. Chemother., 13th, Wash-ington D.C., Abstr. 138, 1973). AntibioticG-418, which is essentially one major compo-nent, is recovered by methods similar to thoseused for other aminoglycosides. It is producedby a new species of Micromonospora namedM. rhodorangea.

MATERIALS AND METHODSTaxonomy. In the tabular descriptions of the

growth of M. rhodorangea on various media, two colordesignates are used. The first was taken from theColor Harmony Manual (1), and the description forthis designate was taken from the Descriptive ColorName Dictionary (4). The second color designate is asynonym or near synonym of the first and was takenfrom the national Bureau of Standards circular no.553 (5).

Fermentation conditions. The inoculum for M.rhodorangea was prepared with a medium containing(grams per liter): beef extract, 3; tryptose, 5; yeastextract, 5; dextrose, 1; starch, 24; and calcium car-bonate, 2. The organism was incubated at 35 C on arotary shaker for 2 to 5 days. An inoculum of 5% byvolume was used for all fermentations.

All fermentation studies were carried out in dupli-cate or triplicate in 500-ml Erlenmeyer flasks con-taining 100 ml of medium or in 14-liter fermentorscontaining 10 liters of medium. The fermentationmedium containing (grams per liter): soybean meal,30; dextrin, 50; dextrose, 5; calcium carbonate, 7; and

cobalt chloride, 2.4 x 10-'; was chosen for furtherexperimentation.

Antibiotic isolation. A schematic representation ofthe isolation process is shown in Fig. 1. The antibioticwas adsorbed from the fermentation broth by anion-exchange procedure. Oxalic acid was added to thewhole broth to precipitate calcium ions, and the pH ofthe fermentation was further adjusted to 2 withsulfuric acid to release the major part of the antibioticfrom the mycelium. After filtration, the clarifiedbroth was neutralized with ammonium hydroxide, theantibiotic was adsorbed on Amberlite IRC-50 (20 to 50mesh) ion-exchange resin in the NH4+ cycle, and thespent broth was discarded.The antibiotic was eluted from the resin with 2 N

ammonium hydroxide, and the eluate was evaporatedto dryness. The material produced in this way con-sisted of one major active component and at least twominor components detectable by bioautography.

BROTH

TREAT WITH OXALIC ACIDIAJTTOpH 2.0 AND FILTER

MYCELIAL CAKEFILTRATE CALCIUM OXALATE

pH 7.0, CHARGE ON IRC-SO NH4'I SPENT'BROTH

CHARGED IRC-50 RESIN

ELUTE WITH 2N NH4OH

I SPENT RESINIRC-50 ELUATE

CONCENTRATE, TREAT WITH H2S04; pH 4.5

ACID SOLUTIONPASS THROUGH IRA 401S (OH-) RESIN COLUMN

SPENT RESIN401S ELUATE

_ PASS 401S ELUATE THROUGH DOWEX 1X2 RESIN COLUMN

SPENT RESINDOWEX 1X2 ELUATE

CONCENTRATE AND LYOPHILIZE

ANTIBIOTIC G-418

FIG. 1. Flow chart detailing isolation of antibioticG-418.

on July 7, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

NEW MICROMONOSPORA-PRODUCED AMINOGLYCOSIDE

A typical paper chromatographic pattem of themixture is shown in Fig. 2 (left).

Separation and purification of antibiotic G-418were achieved by first converting the lyophilizedIRC-50 resin eluate to its sulfate salt and then to thefree base via an IRA 401S resin column. It waspurified further by passing the resin effluent througha Dowex (1 x 2) resin column (OH- cycle). Lyophili-zation of this base material yielded an off-white pow-der.Chromatographic methods. Chromatography of

the antibiotic or its hydrolysis products was carriedout by development on Whatman no. 1 paper inseveral solvent systems. These were: (i) 2-butanone-tertiary butanol-methanol-6.5 N ammonium hydrox-ide (16:3:1:6); (ii) n-butanol-pyridine-acetic acid-water (6:4: 1:3); and (iii) propanol-pyridine-aceticacid-water (15:10:3:12). Development times forthese chromatograms were 24 h for 2-butanone-terti-ary butanol-methanol-6.5 N ammonium hydroxideand 16 h for n-butanol-pyridine-acetic acid-water andpropanol-pyridine-acetic acid-water. The latter twosystems were developed ascending, and the formerwas developed in a descending mode. Other solventsused for comparative purposes are detailed in Table 9.

Detection of antibiotic activity is by the usualbioautographic technique of overlaying the paper onagar plates seeded with Staphylococcus aureus ATCC6538P and incubating the plates overnight at 37 C.

Detection of ninhydrin-positive spots was carriedout by spraying the papergrams with 0.25% ninhydrinin a mixture of pyridine-acetone (1:1) and heating at105 C for several minutes to develop the coloredzones.

Hydrolysis of antibiotics. All antibiotics werehydrolyzed for comparative purposes as the free bases(10 mg/ml) in 6 N hydrochloric acid in sealed tubes at100 C for 2 h.

RESULTS AND DISCUSSIONTaxonomic studies. Micromonospora rhodo-

rangea has been classified as a new species ofMicromonospora based upon its taxonomic andgrowth properties on a number of standard agarmedia. On some of these media, the colonieshave a red-orange appearance. Therefore, themicroorganism has been designated Mi-cromonospora rhodorangea, with the productionof antibiotic G-418 being a characteristic of thisspecies. M. rhodorangea has been deposited atthe Northern Utilization Research and Develop-ment Division, U.S. Department of Agriculture,Peoria, Ill., and added to its collection of micro-organisms as NRRL 5326.The microorganism has the following micro-

scopic, macroscopic, and biochemical proper-ties. Macroscopic observations of a 30-day-oldculture incubated at 24 to 26 C in a 3% N-Z-amine type A-1% dextrose-1.5% agar mediumshow poor growth with no visible distinguishingcharacteristics. M. rhodorangea is a non-sporulating species under this particular condi-

tion of growth; occasionally, chalmydosporesare seen. Microscopic observations of the sameculture indicates that the mycelia range from0.4 to 0.8 ,um in diameter and are sparselybranched.

Observations of colonial growth on variousmedia are shown in Table 1; the carbohydrateutilization pattern is illustrated in Table 2; andutilization of nitrate sources is shown in Table3.M. rhodorangea is most closely related to, but

different from, M. purpurea. Some differencesbetween these two organisms are shown inTable 4. Note particularly the growth responseon glucose asparagine agar, Emerson agar, to-mato paste-oatmeal agar, and Czapek agar.Nitrate is reduced to nitrite by M. rhodorangeabut is variable for M. purpurea. Several car-bohydrates are utilized to a different extent bythe two organisms, and the salt tolerance,critical for identification of micromonosporae, islower for M. rhodorangea than for M. purpurea.M. rhodorangea has been classified as a new

species based upon its taxonomic and growthproperties on a variety of standard agar media.It is most closely related to M. purpurea, but itis readily differentiated from the latter orga-nism.Fermentation studies. In the submerged

fermentation studies carried out with M. rhodo-

- G-418 COMPLEXGG-418

1! ~~~*

- ORIGIN

FIG. 2. Bioautographic comparison of the crudeand purified antibiotic G-418 after chromatography onWhatman no. 1 paper with a solvent composed of 2-butanone-tertiary butanol-methanol-6.5Nammoniumhydroxide (16:3:1:6); plated against S. aureus.

VOL. 6, 19'74 145

on July 7, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

ANTIMICROB. AG. CHEMOTHER.

TABLE 1. Growth of Micromonospora rhodorangea onvarious media

Media or observations Results

Glucose asparagineagar ............. No growth

Gelatin ............. Weakly liquefied

Milk ...............

Sucrose ............

Starch .............

Growth good, hydrolysis com-plete, colony plicate, g6pabrite coral-red, vivid red-dish orange-34

Utilized

Growth good, hydrolysis com-plete

Cellulose ... . Decomposed

Nitrate reduction ... Positive

Temperature ....... Grows well at 28 C, growthfair to poor at 37 C, nogrowth at 45 C or above

Aerobic or anaerobic Aerobic

Czapek sucrose agar(Difco) ...........

Bennett agar .......

Tomato paste-oat-meal agar ........

Glucose yeast extractagar .............

Sucrose nitrate agar(Czapek agar)

Tyrosine agar withyeast ............

Growth poor, flat, color: g7plburgundy, dark grayish,reddish brown 47

Growth good, plicate, mem-branous, no aerial myce-lium, no diffusible pigment,color g6pg chinese red, deepreddish orange-36

Growth poor, not recordable

Growth moderate, membra-nous, no aerial mycelium,no diffusible pigment, colorg6pg bam red, mode red-dish brown-43

Growth poor, flat, g7 ½/2 pidark wine, dark reddishbrown-44

Growth good, plicate, crystalsdissolved, dark amber dif-fusible pigment

TABLE 1. Continued

Media or observations Results

Tyrosine agar withbeef .............. Observations at 2, 7, and 14

days (2): growth poor, flat,crystals, not dissolved,slight amber diffusible pig-ment

Peptone iron agar ... Observations at 2, 7, and 14days: no growth

Bromo cresol purplemilk ............ Completely peptonized, dark

maroon

TABLE 2. Carbohydrate utilization byMicromonospora rhodorangea

Carbohydrate Growth

Arabinose . .................. ModerateCellulose . .................. FairGlucose . .................. GoodGalactose ........ ........... FairLactose ......... .......... FairLevulose . .................. PoorMannose . .................. GoodRaffinose ........ ........... PoorRhamnose ........ ........... PoorStarch ......... .......... GoodSucrose . .................. GoodXylose . .................. GoodInositol ......... .......... PoorMannitol ........ ........... PoorSorbitol . .................. PoorControl (0.5% yeast extract) ............. Poor

rangea, decreasing the levels of the soybeanmeal or dextrin did not significantly affect theamount of antibiotic produced. Other starchsources could be substituted for the dextrinwithout significantly affecting antibiotic pro-duction. The addition of various complex ni-trogenous sources to this medium did not resultin increased yields. The substitution of ex-tracted soy flour for the soybean meal at a levelof 20 g/liter resulted in slightly higher antibiotictiters in an assay similar to that described byOden et al. (3) for gentamicin. This observationwas more evident in fermentations carried out in14-liter fermentors. The medium containing soyflour was used for additional experimentation.

Similar antibiotic titers were obtained withinocula ranging from 1 to 10% after 7 days ofincubation at 28 C (Table 5). A faster initiationof antibiotic production occurs with the higherpercentage of inoculum.

Potato slice ...... No growth

146 WAGMAN ET AL.

on July 7, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

NEW MICROMONOSPORA-PRODUCED AMINOGLYCOSIDE 147

TABLE 3. Utilization of nitrogen sources byMicromonospora rhodorangea

Nitrogen sources Growth+ 1% glucose

0.5% Yeast extract(Difco) ........ Growth moderate, membranous,

no aerial mycelium, no diffusi-ble pigment, color g6 1/2 pgbarn red, deep reddishbrown-41

1.0% N-Z-aminetype A (Shef-field) .Growth fair, flat, furrowed, no

aerial mycelium, no diffusiblepigment, g6pe chinese red,deep reddish orange-36

1% Asparagine ... Growth poor, flat, no aerial my-celium, no diffusible pigment,color g7pi dark wine, dark red-dish brown-44

1% Sodium nitrate No growth

1% Ammonium ni-trate .......... No growth

TABLE 4. Growth differences between M.rhodorangea and M. purpurea

Growth and/or colorMedium

M. rhodorangea M. purpurea

Glucose asparagine agar No growth Growth moder-ate

N-Z-amine-dextrose agar Growth poor Growth goodGelatin liquefaction Weakly lique- Liquefied

fledMilk Vivid reddish Strong yellow-

orange brownNitrate reduction Positive VariableEmerson agar Growth fair to Growth good

poorTomato paste-oatmeal Growth poor Growth good

agarCzapek agar Poor Good

Dark red-brown Very dark red-purple

Peptone iron agar No growth Growth fairCarbohydrate utilization

Levulose Poor ModerateD-Arabinose Moderate PoorRibose Poor Moderate

NaCl tolerance Up to 1.5% Up to 3.0%

Fermentations incubated at 35 C resulted ina more rapid, as well as a higher total, antibioticyield than fermentations run at 28 C (Table 6).Peak antibiotic production occurs between thedays 5 and 7 of fermentation.One major and two minor zones are present;

the major component is identified as antibioticG-418.Characterization of the antibiotic. The

antibiotic complex was compared with a varietyof antibiotics by paper chromatography andsubsequent bioautography. In a series of solventsystems (Table 7) the R, values of antibioticG-418 were distinguishable from all other gen-tamicin-related antibiotics. This table showsa comparison of antibiotic G-418 with thegentamicin complex, gentamicins Cl, C2, C1a,and neomycin, kanamycin, and paromomycin.

Figure 2 (right) shows a bioautogram of thepurified antibiotic G-418 compared with thecomplex in the 2-butanone-tertiary butanol-methanol-6.5 N ammonium hydroxide system.Material of this quality was assigned an anhy-drous potency of 1,000 Ag/mg as antibioticG-418 base.

Antibiotic G-418 was also differentiated fromother related antibiotics by a paper chromato-graphic comparison of their acid hydrolysisproducts.

Figure 3 illustrates the hydrolytic pattern ofthe base form of the antibiotics G-418, ver-damicin, gentamicin, kanamycin, sisomicin,and tobramycin in the n-butanol-pyridine-acetic acid-water chromatographic system; theorigin is to the left. Visualization was done byuse of ninhydrin.The hydrolytic patterns of the listed antibiot-

ics resulting after chromatography in the propa-nol-pyridine-acetic acid-water solvent system

TABLE 5. Effect of inoculum size on antibioticproduction

Antibiotic G-418 production (gg/ml)Inoculum (%)

3 days 7 days

1.0 <3 1792.0 <3 1803.0 4 1795.0 5 1867.5 7 21710.0 7 231

TABLE 6. Effect of temperature on antibioticproduction

Antibiotic G-418 production (gg/ml)Temp (C)

3 daysa 5 days 7 days

28 41 172 23635 165 314 388

a Fermentation time.

VOL. 6, 1974

on July 7, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

TABLE 7. Comparative R, values of antibiotic G-418 and other aminoglycoside antibiotics

R, values of antibiotics

Paper chromatographic system Anti- Gentamicin Paromo-biotic components Neomycin Kanamycin mycinG-418 mycinc,

80% Methanol containing 3% NaCl,ascending ............................... 0.57 0.57 0.56 0.48 0.0, 0.17 0.0,0.28 0.0, 0.28

Propanol-pyridine-acetic acid-water(6:4:1:3), ascending.0.35 0.34 0.30 0.22 0.05 0.08 0.07

80% Phenol, ascending.0.45 0.45 0.45 0.45 0.0-0.12 0.0-0.17 0.0-0.2Chloroform-methanol-concn ammonia (1:1:1)

on Chromar sheet 500, ascending.0.41 0.74 0.65 0.57 0.03 0.07 0.04Chloroform-methanol-17% ammonia (2:1:1),

descendingb.0.05 0.67 0.40 0.21 0.0 0.0 0.0

aPaper buffered with 0.95 M Na2SO4 + 0.05 M NaHSO4.° R, = distance of zone from origin/distance from origin to end of paper; where t = 6 h.

G-418VERDAMICIN

CGENTAM IC IN

KANAMYCIN

SISOMI CINTOBRAMYCIN

FIG. 3. Comparison of hydrolytic pattern of antibiotic G-418 with other aminoglycoside hydrolysates.Chromatography on Whatman no. 1 paper with a solvent composed of n-butanol-pyridine-acetic acid-water(6:4:1:3). Zones visualized by use of ninhydrin. Origin is at left.

are shown in Fig. 4. In both systems thehydrolysates of antibiotic G-418 indicate thepresence of one product positive to the ninhy-drin reagent which has the same migration as2-deoxystreptamine and which is also commonto the other compounds to which it was com-pared. The rate of hydrolysis for tobramycin isslower than the other antibiotics under theseconditions, and the zone for 2-deoxystreptamineis, therefore, not evident in these chromato-grams.

Some of the chemical and physical dataobtained for antibiotic G-418 are shown inTable 8.

Antibiotic G-418 is isolated as an amorphouswhite powder with a melting point of 138 to144 C. It does not have any absorption in theultraviolet range (220 to 400 nm). It is stable toboiling for at least 30 min in 0.1 M buffers in thepH range of 2 to 10. The base is soluble in wateror methanol and the sulfate is soluble in water.

In summary, taxonomy, fermentation, and

148 WAGMAN ET AL. ANTIMICROB. AG. CHEMOTHER.

:...

on July 7, 2018 by guesthttp://aac.asm

.org/D

ownloaded from

NEW MICROMONOSPORA-PRODUCED AMINOGLYCOSIDE 149

G -418

VERDAMICIN

GENTAMICIN

KANAMYCIN

SISOMICIN

TOBRA.MYCIN

.............. .... ...I': :. !:

:,. .: e2: .::. :: ::!:. .:: i.: ::..:* : :. : :.:.: .::: :: ::: .i::.: ..

::: .: :' 'ffix: : :: ::: :.::: ...; :..:. .::: ::. :. :.:....... :

:; ..

|''. :.... ,. ::S.:. ::

.: .:: :.:::: :: ::. :.:. .: :.

.: :.:

:::.....

.:....X....:

.. b.'}.:: '!.. ::'..:::::

::e ..:: :::::..:::::: :.

.:..: :..: .::W :..1x

.. :':: '':.* .. 3;;g:

'::E ..:: '!::: ....i.:; : .. .: .;:: :: .:.. .:. :: .. .: :.: ::i:.: ::.: :, .... > ffloa ,,: ...... }}: .:}.

Ww eS .::.wor' .'., .:::}:}N ..:::}. ::

;:::.....

X dr,:; :: : : ..FIG. 4. Comparison of hydrolytic pattem of antibiotic G-418 with other aminoglycoside hydrolysates.

Chromatography on Whatman no. 1 paper with a solvent composed of propanol-pyridine-acetic acid-water(15:10:3:12). Zones visualized by use of ninhydrin. Origin is at left.

TABLE 8. Chemical and physical properties ofantibiotic G-418 sulfate and base'

Determinations Base Sulfate

Elemental analysisC 49.05 31.79H 8.49 6.65N 11.05 7.630 (by difference) 31.41S04 24.87

Rotation: [a 126 (water, C = +148.80 + 104.400.3%)

Equivalent wt 142.3PKa 8.0

1Samples dried at 60 C; p < 5 mm.

isolation of antibiotic G-418 have been reported.By its chromatographic mobility, hydrolysisproducts, and chemical and physical data it hasbeen demonstrated that antibiotic G-418 is a

new substance.

Details of the biological activity of antibioticG-418 will be reported in a separate publication.

ACKNOWLEDGMENTSWe wish to acknowledge the technical assistance of Bruce

Tilley, Janet V. Bailey, and Alphonse Murawski in thesestudies. We also thank George M. Luedemann and Ann C.Horan for the taxonomic data, and Edwin M. Oden forperforming the microbiological assays.

LITERATURE CITED1. Container Corporation of America. 1958. Color harmony

manuel, 4th ed. Container Corporation of America, NewYork, New York.

2. Gordon, R. E., and M. M. Smith. 1955. Proposed group ofcharacters for the separation of streptomyces and no-cardia. J. Bacteriol. 69:147-150.

3. Oden, E. M., H. Stander, and M. J. Weinstein. 1964.Microbiological assay of gentamicin, p. 8-13. Antimi-crob. Ag. Chemother. 1963.

4. Taylor, H. D., L. Knoche, and W. C. Granville. 1950.Descriptive color name dictionary. Container Corpora-tion of America, New York.

5. U.S. Government Printing Office. 1955. National Bureauof Standards circular no. 553. U.S. Government PrintingOffice, Washington, D.C.

VOL. 6, 1974

N...:

on July 7, 2018 by guesthttp://aac.asm

.org/D

ownloaded from