616 PRELIMINARY NOTES

in dry I N HCl-methanol . The hydrazones were isolated from the bulk of the methyl esters by chromatography on magnesia s. The hydrazones of the keto acid methyl esters were then separated from other classes of hydrazones by chromatography on grade-III (ref. 7) alumina. These keto acid derivatives were then fractionated by hexane-acetonitr i le partition chromatography s. The methyl keto esters were re- generated from the hydrazones by reaction with an excess of acidic acetone. The isolated ketostearate esters from milk fat were identified by gas chromatography on Apiezon L, infrared examination, saponification equivalent, carbonyl-oxygen content, and products of the Beckmann transformation s. The dicarboxylic acids from hydrolysis of the Beckmann amides were determined by gas chromatography of their dimethyl esters on Apiezon L. The amines were determined by spectrophotometric measurement of their 2,4-dinitrophenyl derivatives after separation of the derivatives on a hexane- acetonitrile column s. Keto acids were also concentrated from fat without hydrazone formation by chromatography of both native fat and methyl ester preparations on grade-II I alumina. The ketoglycerides and methyl keto esters were adsorbed from hexane solutions and eluted in a high degree of purity with methylene chloride. Detailed reports of this work will be presented.

We are indebted to Dr. W. I. PATTERSON for his interest and helpful advice.

Department of Dairy Science, University of Maryland, College Park, Md. and Dairy Products Laboratory,

Eastern Utilization Research and Development Division, Agricultural Research Service,

U.S. Department of Agriculture, Washington, D.C. (U.S.A.)

MARK KEI~NEY

IRA KATZ DANIEL P . SCHWARTZ

1 A. T. JAMES AND J. WEBB, Biochem. J., 66 (1957) 515 • 2 K. BLOCH, P. BARONOWSKY, H. GOLDFINE, W. J. LENNARZ, I'l. LIGHT, :\. T. NORRIS AND

G. SCHEUERBRANDT, Federation Proc., 2o (I96I) 92I. 3 A. T. JAMES AND J. B. MARSCH, Biochim. Biophys. Acts, 57 (1962) 17o. 4 D. I~. BLOOMFIELD AND K. BLOCH, J. Biol. Chem., 235 (196o) 337. 5 M. FLAVIN AND C. SLAUGHTER, J . Biol. Chem., 235 (196o) 1112. 6 D. P. SCHWARTZ, O. W. PARKS AND M. KEENLY, Anal. Chem., 34 (1962) 669. 7 I t . BROCKMANN AND H. SCHODDER, Chem. Ber., 74 (1941) 73- s E. A. CORBIN, D. P. SCHWARTZ AND M. KEENLY, J. Chromatog., 3 (196o) 322. 9 j . Ross , A. I. GEBHART AND J. F. GERECHT, J. Am. Chem. Soc., 71 (I949) 282.

Received June I8th, 1962 Biochim. Biophys. Acts, 62 (1962) 615-6 t6

PN 1121

Transethylation in antibiotic biosynthesis

II. Production of the 2'-ethoxy analogue of griseofulvin by biosynthesis

The methyl groups of several antibiotics are derived from transmethylation reactions. Ethionine has been shown to serve as an inhibitor of transmethylation with Strepto- myces viridifaciens at concentrations producing partial inhibition of growth of the streptomycete 1. 6-Demethylchlortetracycline is produced in place of the normal metabolite, chlortetracycline. In other instances, ethionine substitutes for methionine,

Biochim. Biophys. Acla, 62 (1962) 610-619

PRELIMINARY NOTES 617

resulting in transethylation in place of the normal transmethylation. In an other publication 2, the first recorded instance of transethylation in antibiotic biosynthesis was reported, with the formation of an oxytetracycline analogue, the N-methylethyl derivative of oxytetracycline. In this paper, it is demonstrated that an O-substituted antibiotic may also result from ethyl-transfer reactions.

Griseofulvin was employed for study of the effect of ethionine on antibiotic biosynthesis because it is known to contain three methyl groups derived from cholinO. The remainder of the carbons are derived from head-to-tail condensation of acetatO. Griseofulvin, therefore, is analogous in its derivation to the tetracyclines, in which methyl transfer and head-to-tail acetate condensation are important in the synthetic scheme 5. Based on the previously reported work with the tetracyclines, it was anti- cipated that either demethyl analogues or ethoxy analogues of griseofulvin could be produced under appropriate fermentation conditions.

Of the griseofulvin-producing cultures available, Penicillium griseofulvum pro- duced the highest concentrations of the antibiotic and was selected for the investi- gation. Methionine auxotrophs were desired to intensify the stress applied by ethionine during the fermentation. The following procedure proved satisfactory for isolation of the auxotrophs. A modified Czapek-Dox medium containing 5 % lactose as the carbohydrate source was employed to obtain conidia for purposes of irradiation. The medium also served as the minimal culture medium. A maximal medium was obtained by addition of ~5 vitamins, 21 amino acids, adenine, guanine, and uracil. Methionine was present at ioo/~g/ml. Mutation of the parent culture for auxotroph production was applied by suspending spores from agar slants in water containing 0.02 % Tween 80, followed by filtration through a loose absorbent-cotton filter, and washing with water. The spores were treated by exposure to ultraviolet radiation for 4 min at a distance of IO in from the light source (Hanovia Lamp Bulb No. 83A-I) to obtain a 99.3 % kill. Treated spores were plated on maximal medium. After sporu- lation, colonies were replicated to minimal and maximal medium by the velveteen technique, and auxotrophs were selected. Of the i i auxotrophs obtained, 6 were found to grow with methionine as the sole supplement to the minimal medium. The 6 anxotrophs also grew to a varied extent with cystine and 3 were capable of utilizing sodium thiosulfate, indicating that the auxotrophs were genetically blocked at different points in the biosynthetic pathway.

No chemically defined medium is available for griseofulvin production. A medium of low methionine content, which proved suitable for the experiments, had the following composition: corn steep liquor, 2 %; lactose, 7 %; KH2PO4, o.4 %; KC1, 1%; mineral oil, 0.5 %. Based on analysis of the corn steep liquor, this medium contained 15o t~g/ml methionine and was capable of supporting production of about IOOO/~g/ml griseofulvin by the submerged cultivation technique. DL-Ethionine was added on the third day of incubation, after significant growth of Penicillium griseo- fulvum had taken place.

Griseofulvin was assayed by a plate-disc procedure, employing peptone-glucose- agar medium with a heavy inoculum of conidia of Botrytis allii. Zones of inhibition were read after 3 days of incubation at 28 °. Growth inhibition was not complete but areas in which the hyphae were stunted and curled could be distinguished readily from areas containing normal mycelial development.

Paper chromatography was employed for detection of new griseofulvin analogues.

Biochim. Biophys. dcta, 62 (I962) 616-619

618 P R E L I M I N A R Y N O T E S

A two-phase system employing formamide as a stationary phase and benzene- Skellysolve B (I : I ) as the mobile phase proved suitable. The RF for griseofulvin was approx, o.5. A minimum of 5o/xg of griseofulvin was necessary to produce a recogniz- able zone either by ultraviolet scanning or by detection of fluorescence with a phosphor screen. To attain the necessary concentration for paper chromatography, the anti- biotic was removed from broth by two extractions with equal volumes of ethyl acetate, evaporated to dryness, and dissolved in a small volume of ethyl acetate.

Ethionine proved highly toxic to growth of P. griseofulvum; growth and anti- biotic formation ceased after addition of 2oo/,g/ml to the parent culture on the third day of incubation. One of the auxotrophs also proved sensitive to ethionine, four showed reduced growth and antibiotic production in the presence of ethionine, and one was resistant. In the paper-chromatographic analysis, a minor amount of a new ultraviolet-absorbing fluorescing component at RF o.6 7 was observed in the ethionine- supplemented fermentation broths of three auxotrophs (Table I).

TABLE i E F F E C T O F E T H I O N I N E O N G R O V , ' T H A N D A N T I B I O T I C P R O D U C T I O N

BY Penicillium griseofulvum AUXOTROPHS

A uxotroph Ethionine addition Relative growth Griseoful7 iJ* R F o.6 7 ultraviolet No. Type (2oc~ !*g/ml) o] culture activity (Ftg/ml) absorption

P a r e n t - - 4 -b i o o o - -

+ I + N A * N R * *

i M e t h i o n i n e - - 4 + 8 0 0 - -

+ 4 + 9 2 o +

2 M e t h i o n i n e - - 4 q- i i o o - -

+ 2 + 2 5 0 +

3 L e a k y m e t h i o n i n e - - 4 + 6 0 0 -

+ i + N A N R

4 M e t h i o n i n e o r N a ~ S ~ O a - - 4 + 6 8 5 - -

+ 2 + 165 - - -

5 M e t h i o n i n e o r N a e S 2 0 a - - 4 - - lO6O - -

+ z + 2 6 0 T r * * *

6 M e t h i o n i n e o r N a 2 S 2 0 a - 4 + 1 1 2 5 -

+ 2 + 18o - -

* Na -- inactive. ** NR = not tested. *** Tr -- light zone.

Studies with DL-ethionine labeled with 14C in the terminal carbon of the ethyl group suggested that the new product resulted from ethyl transfer. Fermentations conducted with the labeled ethionine produced two fluorescent spots on paper chromatograms. It was estimated by densitometer readings in the ultraviolet region that the new RF 0.67 component represented about o/ and the griseofuhdn corn- / o

ponent at 0.5 RF represented 95 % of the total ultraviolet-absorbing products pro- duced. The major portion of radioactivity in the paper chromatogram appeared at the origin and at the RE 0.67 region, with only a trace at the RF o.5 region.

Isolation and crystallization of the new ultraviolet-absorbing component was accomplished by ethyl acetate extraction of broth, chromatography of a concentrate on sheets of Whatman 3 MM paper, elution from the paper with benzene containing

ldiochim. Biophys. Acta, 6-, (1962) 616 619

PRELIMINARY NOTES 619

IO % methanol, concentration to dryness, and chromatography of an ethyl acetate solution on a column of acid-washed alumina. Crystallization was accomplished by dilution of the effluent with petroleum ether.

The crystalline product was shown to be characteristic of a 2'-ethoxy analogue of griseofluvin (7-chloro-4,6-dimethoxy-2'-ethoxy-6'-methylgris-2'-en-3,4-dione) by nuclear magnetic resonance (NMR) spectrum. The m.p. (202-208 °) was in agreement with the literature value of 205-206 °. The 2'-ethoxy analogue of griseofulvin had been produced by chemical modification of griseofulvin during the period of great interest in possible agricultural application of this antibiotic 6. It was dropped from active consideration because of poor translocation in plants.

The 2'-ethoxy derivative was synthesized in our laboratories following literature methodsL The m.p. agreed with the literature value and the NMR of the chemically synthesized 2'-ethoxy analogue was identical with the product obtained by fermen- tation. The crystalline 2'-ethoxy analogue of griseofulvin was assayed by the plate-disc procedure for activity against Botrytis allii and was found to be approximately twice as active as griseofulvin.

The isolation of a 2'-ethoxy analogue of griseofulvin from an ethionine-supple- mented fermentation demonstrates that the transethylation reaction may be of greater significance in microbial biosynthesis than previously realized. S-Adenosyl- ethionine has been isolated from yeast cells grown in the presence of ethionine s. It will participate in transethylation reactions with L-homocysteine to form ethionine with an in vitro enzyme system derived from Torulopsis ultilis 9. Furthermore, ethionine has been reported to be a common metabolite in microorganisms 10. Ethyl transfer must be considered as the source of ethyl groups found to be present in normal antibiotics. It also provides the mechanism for controlled biosynthesis of new antibiotics.

We are indebted to Dr. N. R. TRENNER of the Merck Sharp and Dohme Research Laboratories for performing and interpreting the NMR spectrum.

Merck Sharp and Dohme Research Laboratories, Division of Merck ~ Co., Inc., Rahway, N.J. (U.S.A.)

MARION JACKSON

EUGENE L. DULANEY

IRVING PUTTER

HENRY M. SHAFER

FRANK J . WOLF

H. BOYD WOODRUFF

1 D. HENDLIN, E. L. DULANEY, D. DRESCHER, T. COOK AND L. CHAIET, Biochim. Biophys. Acta, 58 (1962) 635.

2 ]~. L. DULANEY, I. PUTTER, D. DRESCHER, L. CHAIET, \V. J. MILLER, F. J. WOLF AND D. ]7IENDLIN, Biochim. Biophys. Acta, in the press.

a D. J. D. HOCKENHULL AND W. F. FAULDS, Chem. and Incl., (1955) 139o. 4 A. J. BIRCH, 1:{. A. MASSY-WESTROPP, R. W. RICKARDS AND H. SMITH, J. Chem. Soc., (1958) 360. 5 p. L. THOMSON, J. Chem. Soc., (1962) 425 . 8 S. H. CROWDY, J. F. GROVE AND P. MCCLOSKY, Biochem. J., 72 (1959) 241.

J. F. GROVE, J. MACMILLAN, T. P. C. MULHOLLAND ANn M. A. THOROLD ROGERS, J. Chem. Soc., (1952 ) 3977.

s F. SCHLENK AND J. A. TILLOTSON, J. Biol. Chem., 206 (1954) 687. 9 W. L. PARKS, J. Biol. Chem., 232 (1958) 169.

10 j . F. FISHER AND M. F. MALLETTE, J. Gen. Physiol., 45 (1961) I.

Received June i6th, 1962

Biochim. Biophys. Acta, 62 (1962) 616-619