APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Aug. 1978, p. 213-216 0099-2240/78/0036-0213$02.00/0 Copyright i 1978 American Society for Microbiology
Vol. 36, No. 2
Printed in U.S.A.
Progesterone Biotransformation by Plant Cell Suspension Cultures B. YAGEN,1 G. E. GALLILI,2 AND R. I. MATELES2* School of Pharmacy' and Laboratory of Applied Microbiology,2 Hebrew University-Hadassah Medical School, Jerusalem, Israel
Received for publication 7 March 1978
Progesterone was converted to 5a-pregnane-3a-ol-20-one, A4-pregnene-20a-ol3-one, M&-pregnene-14a-ol-3,20-dione, A4-pregnene-7fl,14a-diol-3,20-dione, and A4pregnene-6,8,lla-diol-3,20-dione by cell cultures of Lycopersicon esculentum. Cell cultures of Capsicum frutescens (green) metabolized progesterone to A4-pregnene20a-ol-3-one in very high yield, and Vinca rosea yielded A4-pregnene-20,f-ol-3one and A4-pregnene-14a-ol-3,20-dione. A stereospecific reduction of the keto groups and a double bond and stereospecific introduction of hydroxyl groups at the 6, 11, and 14 positions have been observed. The mono- and dihydroxylated progesterones have not previously been reported as metabolic products of progesterone by plant cell systems and represent de novo hydroxylation of a nonglycosylated steroid. We have been studying biotransformations of progesterone and other steroids by plant cell suspension cultures, with the aim of determining the structure of the metabolites and the yields obtained with the various cultures. Microbial transformations of steroids have been extensively investigated (5, 15). There are relatively few reports, however, on the transformation of steroids by plant cell cultures (19). These studies have been conducted mainly with various pregnane and androstane derivatives, and the most common types of biotransformations have involved stereospecific reductions of the 3 and 20 keto groups as well as the 4,5 double bond (10, 12-14, 23). In addition, formation of conjugates as the glucosides and pahnitate is commonly found (9, 10, 13, 14). Oxidation of a hydroxyl group to a keto group has been noted for testosterone (13), and hydroxylation de novo of the glycoside digitoxin has been found in the 16,/ (1, 19) and 12/3 (1, 24) positions. Recently we reported the de novo hydroxylation of a non-glycosylated steroid, progesterone (VII), by suspension cultures of Vinca rosea (11) and other plant cell suspension cultures (27). This paper reports the results of biotransformation of progesterone to 5a-pregnane-3,8-ol-20one (I), A4-pregnene-20a-ol-3-one (II), A4-pregnene-14a-ol-3,20-dione (III), A4-pregnene-6f?,14a-diol-3,20-dione (IV), A4-pregnene-6,1,11adiol-3,20-dione (V), and A4-pregnene-20/8-ol-3one (VI) by Lycopersicon esculentum, Capsicum frutescens, and Vinca rosea suspension cultures.
MATERIALS AND METHODS Reagents. All chemicals and solvents were of analytical grade. The solvents were distilled before use. Progesterone (VII) (Makor Chemicals, Ltd., Jerusalem, Israel) was purified by recrystaization from acetone-hexane until white crystals were formed which melted at 130 to 1310C and were pure as demonstrated by thin-layer chromatography (TLC) using system I (see below) and by gas chromatography (GC). These crystals were free of any hydroxylated progesterone. The chromatographic standards (Fig. 1) 5a-pregnane3,B-ol-20-one (I), A4-pregnene-20a-ol-3-one (II), and A4-pregnene-20fB-ol-3-one (VI) were purchased from Makor Chemicals, Ltd. The standards M-pregnene14a-ol-3,20-dione (III) and A&4-pregnene-6.8,14a-diol3,20-dione (IV) were a gift of The Upjohn Co., and A4pregnene-6#,Bla-diol-3,20-dione (V) was prepared in this laboratory (7). Growth conditions. Cultures, well adapted to suspension culture, were inoculated into 250-ml Erlenmeyer flasks containing 100 ml of Murashige-Skoog medium (18) with 1 jug of 2,4-dichlorophenoxyacetic
acid per ml. Incubation was carried out at 280C on a gyratory shaker at 150 rpm. After 1 week of growth, progesterone (VII) dissolved in 95% ethanol (30 mg/ml) was added to a final concentration of 300 mg/liter of broth, and incubation was continued for a further 10 to 14 days. For isolation and identification of metabolites, several flasks were pooled. Plant cell cultures. A large number of plant cell cultures were examined for their ability to produce biotransformation products from progesterone (VII). The selected cultures were L. esculentum, V. rosea, C. frutescens (green), C. frutescens (red), Phaseolus sp., Nicotiana tabacum, Daucus carota, Phaseolus vulgaris, Cucumis melo, and Ammi visnaga, all of which 213
YAGEN, GALLILI, AND MATELES
214
0'
H I
APPL. ENVIRON. MICROBIOL.
5ai-pregnane-3BO-l-20-one
II
A4_pregnene-20a-ol-3-one
VI A4_pregnene-20B-ol-3-one
R1 R1
R2
R3
14a-hydroxyprogesterone (a4-pregnene-l14a-ol-3,20-dione) 60,14a-dihydroxyprogesterone (& 4-pregnene-6B,14a-diol-3,20-dione) H V 6B,11-dlhydroxyprogesterone (C1 -pregnene-60,lla-diol-3,20-dione) H VI1 progesterone (a 4-pregnene-3,20-dione) FIG. 1. Steroids produced by biotransfonnation ofprogesterone by plant cell suspension cultures. III IV
OH OH
H
OH OH H
H H OH H
showed indication of biotransformation products. Extraction and cleanup. At the end of the incubation period, the plant cells together with medium were extracted with 2 volumes of chloroform, and the chloroform layers were pooled. After the extracts had been dried over MgSO4, the solvent was removed by evaporation on a rotary evaporator at 370C. Detection of steroids was carried out by TLC using appropriate systems for comparison with known standards run in parallel. Proof that TLC spots originated from the added progesterone (VII) rather than from endogenous metabolites was obtained by using ['4C]progesterone and scanning the plates. Only spots containing radioactive label were dealt with further. Yields were estimated by quantitative GC. The oil obtained after evaporation of the chloroform was treated with an etheric solution of diazomethane to convert fatty acids to their methyl esters. The reaction mixture was left for 0.5 h at room temperature, and the ether was evaporated, yielding an oil. This oil was chromatographed on a column filled with Sephadex LH 20, using 40% chloroform in petroleum ether (40 to 600C) as eluant. Partly purified steroid fractions were rechromatographed on Florisil (100 to 200 mesh; Merck, Darmstadt, W. Germany) and eluted from the column by means of a stepwise gradient of ethyl acetate in petroleum ether. This process removed from the Florisil the
unreacted progesterone and additional fractions, which were analyzed and identified by using nuclear magnetic resonance and mass spectrometry, infrared, ultraviolet, and optical rotation spectrophotometry, and GC and TLC. The separation of the various fractions was followed by TLC and by GC, which was also used for quantitative analysis. TLC was carried out on precoated Kieselgel GF 254 plates with a layer thickness of 0.25 mm (Merck). The progesterone (VII), 5a-pregnane3,f-ol-20-one (I), A4-pregnene-20a-ol-3-one (II), and '&-pregnene-20fi-ol-3-one (VI) were separated on these plates and developed by 40% acetone in hexane (system I). The A4-pregnene-14a-ol-3,20-dione (III),
A&4-pregnene-6,,14a-diol-3,20-dione
(IV), and A4-preg-
nene-6.i,11a-diol-3,20-dione (V) were separated by de-
velopment by 80% ethyl acetate in hexane (system II). GC for measurement primarily of non- or monohydroxylated compounds was carried out on a column (6 feet by 0.25 inch [ca. 1.83 m by 6.35 mm]) of 3% OV-17 on Chromosorb W, 80 mesh, operated at 2500C with N2 as the carrier gas, using a Varian 1200 GC with a flame ionization detector. RESULTS Of the cultures that gave some indication of producing biotransformations of progesterone
VOL. 36, 1978
PROGESTERONE BIOTRANSFORMATION
(VII), several which gave the strongest response studied in more detail. These cultures were L. esculentum, V. rosea, and red and green varieties of C. frutescens, and the metabolites found are reported below. The physical characteristics of the compounds isolated were identical to those of the standards. In every case unreacted progesterone (VII) could be recovered from the incubation mixture also, and all of the progesterone added initially was accounted for. From cultures of L. esculentum it was possible to isolate five biotransformation products: 5apregnane-3,8-ol-20-one (I)-molecular ion at mWe 318; NMR(CDC13): 80.6, 0.83, and 2.1 (3 CH3, s), 3.6 (H, s, br). A4-Pregnene-20a-ol-3-one (11)-molecular ion at m/e 316; NMR(CDCI3): 80.73, 1.2 (2 CH3, s), 1.23 (CH3, d, j = 6 Hz), 3.65 (H, s, br), 5.7 (H, s). A4-Pregnene-14a-ol-3,20dione (III)-molecular ion at m/e 330; NMR(CDC13): 80.075, 1.13, 2.1 (3 CH3, s), 3.16 (H, d, j = 8 Hz), 5.66 (H, s). A4-Pregnene-6f?,14adiol-3,20-dione (IV)-molecular ion at m/e 346; NMR(CDC13): 80.8, 1.36, 2.1 (3 CH3, s), 3.1 (H, d, j = 8 Hz), 4.36 (H, s, br), 5.75 (H, s). A4were
Pregnene-6,8,lla-diol-3,20-dione
(V)-molecular ion at m/e 346; NMR(CDC13): 80.68, 1.45, 2.07 (3 CH3, s), 4.1 (H, s, br), 4.3 (H, s), 5.75 (H, s). The yield of each product was 3 to 5% of the initial quantity of progesterone added. The cultures of V. rosea yielded A4-pregnene20/i-ol-3-one (VI) and A4-pregnene-14a-ol-3,20dione (III), each in yields of 2 to 3%. For VI the molecular ion was at m/e 316; NMR(CDC13): 80.76 (CH3, s), 1.12 (CH3, d, j = 6 Hz), 1.2 (CH3, s), 3.66 (H, s, br), 5.7 (H, s). The culture of C. frutescens (green) converted progesterone, in yields of 60 to 90%, to only one product, readily identified as A4-pregnene-20aol-3-one (11). DISCUSSION Progesterone (VII) is readily metabolized by Digitalis purpurea tissue cultures to 5a-pregnane-3,20-dione and 5a-pregnane-3,8-ol-20-one (I) (3, 5, 20). Furthermore, Rosa sp. tissue cultures were reported to form A4-pregnene-20,8-ol3-one (VI) and A4-pregnene-20a-ol-3-one (II) (12). In this investigation, cultures of L. esculentum and C. frutescens (green) were able to reduce stereospecifically the carbonyl at C20 to yield A4-pregnene-20a-ol-3-one (II), in contrast to V. rosea cultures, which were capable of converting it to the 20,8-hydroxy isomer (VI). The 20,B isomer could not be detected in the cultures of L. esculentum, nor the 20a isomer in V. rosea cultures. The culture of C. frutescens (green) was relatively resistant to inhibition of growth caused by addition of progesterone and
215
was capable of converting it to A4-pregnene-20aol-3-one (II) in yields of 60 to 90%. Such a high yield with a relative absence of by-products is reminiscent of the conversions obtained with the yeast Rhodotorula longissima (4). The 20a reduction has been reported to be carried out by yeasts (3), algae (17), plants (16), and mice (25), but not by bacteria or molds. The cultures of L. esculentum also yielded 5a-pregnane-3,8-ol-20one (I), as has been reported for suspension cultures of D. purpurea (12), D. deltoidea (23), N. tabacum (8), and other plants. This product forms as a result, of stereospecific reduction of the A4-double bond, yielding the trans configuration of the A, B steroid rings, and stereospecific reduction of 3-carbonyl, forming only the 3f,-alcohol. In addition, A4-pregnene-14a-ol-3,20dione (III) was isolated from cultures of L. esculentum. This stereospecific de novo hydroxylation in the 14a position is a common hydroxylation carried out by many strains of Curvularia lunata (28) and Naematoloma sublateritium (21), but has not been previously reported as occurring in plant cell cultures. This metabolite was also isolated from cultures of L. esculentum and V. rosea. Additional new metabolites from the L. esculentum incubation were identified as A4-pregnene-6f8,14a-diol-3,20-dione (IV) and A4-pregnene-6f8,lla-diol-3,20-dione (V). These products resulting from hydroxylation in the 14a and 6,f positions indicate that certain plant cell cultures are capable of introducing a hydroxyl group apparently without the need for the steroid to be conjugated to a sugar moiety. The C6 hydroxylation performed by L. esculentum on an activated allylic position is analogous to non-stereospecific hydroxylation obtained by the use of mercuric acetate (26). The stereospecific hydroxylations of nonactivated positions, as in the case of 14a and hla, performed by the plant cell cultures (as well as by microorganisms) cannot, however, be achieved easily by chemical synthesis. The A4-pregnene-6,f,14a-diol-3,20dione (IV) metabolite is also readily obtained with C. lunata (28), Circinella sp. (21), and Trichothecium roseum (20), while the 6fl,11adiol (V) could be obtained from Aspergillus and Penicillium (6). Although until recently only reductions of unglycosidated steroids have been reported, apparently some cell cultures can carry out oxidative reactions, namely hydroxylations, on unglycosidated steroids. In addition, reductive and oxidative reactions can be performed by the same plant cell culture. The introduction of a hydroxyl group into one or more positions of progesterone (VII) by means of the cultures of V. rosea and L. esculentum was stereospecific. Cultures of L. esculentum yielded hydroxylations in the 6/?, 11a, and 14a positions, whereas V. rosea cultures
216 YAGEN, GALLILI, AND MATELES carried out hydroxylation in the 14a position. Reduction of progesterone by cultures of L. esculentum and C. frutescens led to the 20fi-hydroxy derivative (VI), while cultures of V. rosea produced the 20a-hydroxy isomer (II). Thus, if a new center of asymmetry was formed by hydroxylation or reduction, only one of the possible epimers arises in each case. This is similar to the situation found with microorganisms. Although the yields are generally low as compared to those obtained with microorganisms, the occasional very high yield suggests the possibility of finding additional examples of culturesubstrate combinations with high yields and suggests the desirability of determining whether, in cultures such as Capsicum, there is a connection between the high yield and the apparent insensitivity to steroid inhibition. Although it does not appear likely that plant cell cultures can compete with microorganisms for use in the manufacture of steroid drugs, there is a possibility that plant cell cultures might be used for enzymic conversion to optically active forms, or for specific biotransformation of natural compounds for producing rare or unusual steroids which are difficult to obtain from other natural or synthetic sources. ACKNOWLEDGMENITS This research was supported by grants from the National Council for Research and Development, Jerusalem, Israel, and the Lewis and Rosa Strauss Memorial Fund, Washington, D.C. We thank P. W. O'Connell, The Upjohn Co., Kalamazoo, Mich., for supplying steroid standards. LITERATURE CITED 1. Aifermann, A. W., H. M. Boy, P. C. Doller, W. Hagedorn, M. Heins, J. Wahl, and E. Reinhard. 1977. Biotransformation of cardiac glycosides by plant cell cultures, p. 125-141. In W. Barz, E. Reinhard, and M. H. Zenk (ed.), Plant tissue culture and its bio-technological application. Springer-Verlag, Berlin. 2. Behrend, J., and R. L. Mateles. 1975. Nitrogen metabolism in plant cell suspension cultures. I. Effect ofamino acids on growth. Plant Physiol. 56:584-589. 3. Carvajal, F., 0. F. Vitale, M. J. Gentles, H. L. Herzog, and E. B. Hershberg. 1959. Microbial transformation of steroids. VI. Stereospecific reductions of the 20-carbonyl group. J. Org. Chem. 24:695-698. 4. Chang, V. M., and D. R. Idler. 1961. a-Reduction of the 20-carbonyl group in C-21 steroids by Rhodotorula longissima. Can. J. Biochem. Physiol. 39:1277-1285. 5. Charney, W., and H. L. Herzog. 1967. Microbial transformation of steroids. Academic Press Inc., New York. 6. Dulaney, E. L, W. J. McAleer, M. Koslowski, E. 0. Stapley, and J. Jaglom. 1955. Hydroxylation of progesterone and 11-deoxy-17-hydroxycortecosterone by Aspergillus and Penicillium. Appl. Microbiol.
3:336-340.
7. Dulaney, E. L., E. 0. Stapley, and C. Hlavac. 1955. Hydroxylation of steroids, principally progesterone, by a strain of AspergiUus ochraceus. Mycologia 47:464 474. 8. Furuya, T., M. Hirotani, and K. Kawaguchi. 1971. Biotransformation of progesterone and pregnenolone by plant suspension cultures. Phytochemistry
APPL. ENVIRON. MICROBIOL. 10:1013-1017. 9. Furuya, T., M. Hirotani, and T. Shinohara. 1970. Biotransformation of digitoxin by suspension callus culture of Digitalis purpurea. Chem. Pharm. Bull. 18:1080-1082. 10. Furuya, T., K. Kawaguchi, and M. Hirotani. 1973. Biotransformation of progesterone by suspension cultures of Digitalis purpurea cultured cells. Phytochemistry 12:1621-1626. 11. Gallili, G. E., B. Yagen, and R. I. Mateles. 1978. Hydroxylation of progesterone by plant cell suspension cultures of Vinca rosea. Phytochemistry 17:578. 12. Graves, J. M. H., and W. K. Smith. 1967. Transformation of pregnenolone and progesterone by cultured plant cells. Nature (London) 214:1248-1249. 13. Hirotani, M., and T. Furuya. 1974. Biotransformation of testosterone and other androgens by suspension cultures of Nicotiana tobacum (bright yellow). Phytochemistry 13:2135-2142. 14. Hirotani, M., and T. Furuya. 1975. Metabolism of 5,Bpregnane-3,20-dione and 3f,-hydroxy-5,B-pregnan-20one by Digitalis suspension cultures. Phytochemistry 14:2601-2606. 15. Iizuka, H., and A. Naito. 1967. Microbial transformation of steroids and alkaloids. University of Tokyo Press, Tokyo. 16. Luedemann, G. W., W. Charney, A. Mitchell, and H. L Herzog. 1959. Stereospecific reduction of prednisone by alfalfa seedlings. J. Org. Chem. 24:1385-1386. 17. Luedemann, G., W. Charney, A. Woyciesjes, E. Pettersen, W. D. Peckham, M. J. Gentles, H. Marshall, and H. L Herzog. 1961. Microbial transformation of steroids. IX. The transformation of Reichstein's compound S by Scenedesmus sp. Diketopiperazine metabolites from Scenedesmus sp. J. Org. Chem.
26:4128-4130. 18. Murashige, T., and F. Skoog. 1962. A revised medium for rapid growth and bioassays with tobacco tissue
cultures. Physiol. Plant. 15:473497. 19. Reinhard, E. 1974. Biotransformations by plant tissue cultures, p. 433459. In H. E. Street (ed.), Tissue culture and plant science. Academic Press, London. 20. Schubert, A., K. Heller, D. Onken, K. Zetache, and B. Kluger. 1960. Microbielle Hydroxylierung von Progesterone in 6- und 14-Stellung. Z. Naturforsch. Teil B
15:269. 21. Schubert, A., D. Onken, R. Sibert, and K. Heller. 1958. Microbielle Hydroxylierung von Steroiden in 9und 14-Stellung. Chem. Ber. 91:2549-2552. 22. Stohs, S. J. 1977. Metabolism of steroids in plant tissue cultures, p. 142-150. In W. Barz, E. Reinhard, and M. H. Zenk (ed.), Plant tissue culture and its bio-technological application. Springer-Verlag, Berlin. 23. Stohs, S. J., and M. M. El-Olemy. 1972. Metabolism of progesterone by Dioscorea deltoidea suspension cultures. Phytochemistry 11:1397-1400. 24. Stohs, S. J., and H. Rosenberg. 1974. Steroids and steroid metabolism in plant tissue cultures. Lloydia 38:181-194. 25. Weinstein, W., H. R. Lindner, and B. Eckstein. 1977. Thymus metabolism progesterone-possible enzymatic marker for T lymphocytes. Nature (London) 266:632-633. 26. Wiberg, K. B., and S. D. Nielsen. 1964. Some observations on allylic oxidation. J. Org. Chem. 29:3353-3361. 27. Yagen, B., G. E. Gallili, and R. I. Mateles. 1978. Progesterone biotransformation by plant cell cultures, p. 68-73. In A. W. Alfermann and E. Reinhard (ed.), Production of natural compounds by cell culture methods. Gesellachaft fur Strahlen-und Umweltforschung mbH, Munich. 28. Zetsche, K. 1961. Chemical-physiological investigations concerning the hydroxylation of steroids by fungi of the genus Curvularia. Arch. Mikrobiol. 38:237-271.
Vol. 36, No. 2
Printed in U.S.A.
Progesterone Biotransformation by Plant Cell Suspension Cultures B. YAGEN,1 G. E. GALLILI,2 AND R. I. MATELES2* School of Pharmacy' and Laboratory of Applied Microbiology,2 Hebrew University-Hadassah Medical School, Jerusalem, Israel
Received for publication 7 March 1978
Progesterone was converted to 5a-pregnane-3a-ol-20-one, A4-pregnene-20a-ol3-one, M&-pregnene-14a-ol-3,20-dione, A4-pregnene-7fl,14a-diol-3,20-dione, and A4pregnene-6,8,lla-diol-3,20-dione by cell cultures of Lycopersicon esculentum. Cell cultures of Capsicum frutescens (green) metabolized progesterone to A4-pregnene20a-ol-3-one in very high yield, and Vinca rosea yielded A4-pregnene-20,f-ol-3one and A4-pregnene-14a-ol-3,20-dione. A stereospecific reduction of the keto groups and a double bond and stereospecific introduction of hydroxyl groups at the 6, 11, and 14 positions have been observed. The mono- and dihydroxylated progesterones have not previously been reported as metabolic products of progesterone by plant cell systems and represent de novo hydroxylation of a nonglycosylated steroid. We have been studying biotransformations of progesterone and other steroids by plant cell suspension cultures, with the aim of determining the structure of the metabolites and the yields obtained with the various cultures. Microbial transformations of steroids have been extensively investigated (5, 15). There are relatively few reports, however, on the transformation of steroids by plant cell cultures (19). These studies have been conducted mainly with various pregnane and androstane derivatives, and the most common types of biotransformations have involved stereospecific reductions of the 3 and 20 keto groups as well as the 4,5 double bond (10, 12-14, 23). In addition, formation of conjugates as the glucosides and pahnitate is commonly found (9, 10, 13, 14). Oxidation of a hydroxyl group to a keto group has been noted for testosterone (13), and hydroxylation de novo of the glycoside digitoxin has been found in the 16,/ (1, 19) and 12/3 (1, 24) positions. Recently we reported the de novo hydroxylation of a non-glycosylated steroid, progesterone (VII), by suspension cultures of Vinca rosea (11) and other plant cell suspension cultures (27). This paper reports the results of biotransformation of progesterone to 5a-pregnane-3,8-ol-20one (I), A4-pregnene-20a-ol-3-one (II), A4-pregnene-14a-ol-3,20-dione (III), A4-pregnene-6f?,14a-diol-3,20-dione (IV), A4-pregnene-6,1,11adiol-3,20-dione (V), and A4-pregnene-20/8-ol-3one (VI) by Lycopersicon esculentum, Capsicum frutescens, and Vinca rosea suspension cultures.
MATERIALS AND METHODS Reagents. All chemicals and solvents were of analytical grade. The solvents were distilled before use. Progesterone (VII) (Makor Chemicals, Ltd., Jerusalem, Israel) was purified by recrystaization from acetone-hexane until white crystals were formed which melted at 130 to 1310C and were pure as demonstrated by thin-layer chromatography (TLC) using system I (see below) and by gas chromatography (GC). These crystals were free of any hydroxylated progesterone. The chromatographic standards (Fig. 1) 5a-pregnane3,B-ol-20-one (I), A4-pregnene-20a-ol-3-one (II), and A4-pregnene-20fB-ol-3-one (VI) were purchased from Makor Chemicals, Ltd. The standards M-pregnene14a-ol-3,20-dione (III) and A&4-pregnene-6.8,14a-diol3,20-dione (IV) were a gift of The Upjohn Co., and A4pregnene-6#,Bla-diol-3,20-dione (V) was prepared in this laboratory (7). Growth conditions. Cultures, well adapted to suspension culture, were inoculated into 250-ml Erlenmeyer flasks containing 100 ml of Murashige-Skoog medium (18) with 1 jug of 2,4-dichlorophenoxyacetic
acid per ml. Incubation was carried out at 280C on a gyratory shaker at 150 rpm. After 1 week of growth, progesterone (VII) dissolved in 95% ethanol (30 mg/ml) was added to a final concentration of 300 mg/liter of broth, and incubation was continued for a further 10 to 14 days. For isolation and identification of metabolites, several flasks were pooled. Plant cell cultures. A large number of plant cell cultures were examined for their ability to produce biotransformation products from progesterone (VII). The selected cultures were L. esculentum, V. rosea, C. frutescens (green), C. frutescens (red), Phaseolus sp., Nicotiana tabacum, Daucus carota, Phaseolus vulgaris, Cucumis melo, and Ammi visnaga, all of which 213
YAGEN, GALLILI, AND MATELES
214
0'
H I
APPL. ENVIRON. MICROBIOL.
5ai-pregnane-3BO-l-20-one
II
A4_pregnene-20a-ol-3-one
VI A4_pregnene-20B-ol-3-one
R1 R1
R2
R3
14a-hydroxyprogesterone (a4-pregnene-l14a-ol-3,20-dione) 60,14a-dihydroxyprogesterone (& 4-pregnene-6B,14a-diol-3,20-dione) H V 6B,11-dlhydroxyprogesterone (C1 -pregnene-60,lla-diol-3,20-dione) H VI1 progesterone (a 4-pregnene-3,20-dione) FIG. 1. Steroids produced by biotransfonnation ofprogesterone by plant cell suspension cultures. III IV
OH OH
H
OH OH H
H H OH H
showed indication of biotransformation products. Extraction and cleanup. At the end of the incubation period, the plant cells together with medium were extracted with 2 volumes of chloroform, and the chloroform layers were pooled. After the extracts had been dried over MgSO4, the solvent was removed by evaporation on a rotary evaporator at 370C. Detection of steroids was carried out by TLC using appropriate systems for comparison with known standards run in parallel. Proof that TLC spots originated from the added progesterone (VII) rather than from endogenous metabolites was obtained by using ['4C]progesterone and scanning the plates. Only spots containing radioactive label were dealt with further. Yields were estimated by quantitative GC. The oil obtained after evaporation of the chloroform was treated with an etheric solution of diazomethane to convert fatty acids to their methyl esters. The reaction mixture was left for 0.5 h at room temperature, and the ether was evaporated, yielding an oil. This oil was chromatographed on a column filled with Sephadex LH 20, using 40% chloroform in petroleum ether (40 to 600C) as eluant. Partly purified steroid fractions were rechromatographed on Florisil (100 to 200 mesh; Merck, Darmstadt, W. Germany) and eluted from the column by means of a stepwise gradient of ethyl acetate in petroleum ether. This process removed from the Florisil the
unreacted progesterone and additional fractions, which were analyzed and identified by using nuclear magnetic resonance and mass spectrometry, infrared, ultraviolet, and optical rotation spectrophotometry, and GC and TLC. The separation of the various fractions was followed by TLC and by GC, which was also used for quantitative analysis. TLC was carried out on precoated Kieselgel GF 254 plates with a layer thickness of 0.25 mm (Merck). The progesterone (VII), 5a-pregnane3,f-ol-20-one (I), A4-pregnene-20a-ol-3-one (II), and '&-pregnene-20fi-ol-3-one (VI) were separated on these plates and developed by 40% acetone in hexane (system I). The A4-pregnene-14a-ol-3,20-dione (III),
A&4-pregnene-6,,14a-diol-3,20-dione
(IV), and A4-preg-
nene-6.i,11a-diol-3,20-dione (V) were separated by de-
velopment by 80% ethyl acetate in hexane (system II). GC for measurement primarily of non- or monohydroxylated compounds was carried out on a column (6 feet by 0.25 inch [ca. 1.83 m by 6.35 mm]) of 3% OV-17 on Chromosorb W, 80 mesh, operated at 2500C with N2 as the carrier gas, using a Varian 1200 GC with a flame ionization detector. RESULTS Of the cultures that gave some indication of producing biotransformations of progesterone
VOL. 36, 1978
PROGESTERONE BIOTRANSFORMATION
(VII), several which gave the strongest response studied in more detail. These cultures were L. esculentum, V. rosea, and red and green varieties of C. frutescens, and the metabolites found are reported below. The physical characteristics of the compounds isolated were identical to those of the standards. In every case unreacted progesterone (VII) could be recovered from the incubation mixture also, and all of the progesterone added initially was accounted for. From cultures of L. esculentum it was possible to isolate five biotransformation products: 5apregnane-3,8-ol-20-one (I)-molecular ion at mWe 318; NMR(CDC13): 80.6, 0.83, and 2.1 (3 CH3, s), 3.6 (H, s, br). A4-Pregnene-20a-ol-3-one (11)-molecular ion at m/e 316; NMR(CDCI3): 80.73, 1.2 (2 CH3, s), 1.23 (CH3, d, j = 6 Hz), 3.65 (H, s, br), 5.7 (H, s). A4-Pregnene-14a-ol-3,20dione (III)-molecular ion at m/e 330; NMR(CDC13): 80.075, 1.13, 2.1 (3 CH3, s), 3.16 (H, d, j = 8 Hz), 5.66 (H, s). A4-Pregnene-6f?,14adiol-3,20-dione (IV)-molecular ion at m/e 346; NMR(CDC13): 80.8, 1.36, 2.1 (3 CH3, s), 3.1 (H, d, j = 8 Hz), 4.36 (H, s, br), 5.75 (H, s). A4were
Pregnene-6,8,lla-diol-3,20-dione
(V)-molecular ion at m/e 346; NMR(CDC13): 80.68, 1.45, 2.07 (3 CH3, s), 4.1 (H, s, br), 4.3 (H, s), 5.75 (H, s). The yield of each product was 3 to 5% of the initial quantity of progesterone added. The cultures of V. rosea yielded A4-pregnene20/i-ol-3-one (VI) and A4-pregnene-14a-ol-3,20dione (III), each in yields of 2 to 3%. For VI the molecular ion was at m/e 316; NMR(CDC13): 80.76 (CH3, s), 1.12 (CH3, d, j = 6 Hz), 1.2 (CH3, s), 3.66 (H, s, br), 5.7 (H, s). The culture of C. frutescens (green) converted progesterone, in yields of 60 to 90%, to only one product, readily identified as A4-pregnene-20aol-3-one (11). DISCUSSION Progesterone (VII) is readily metabolized by Digitalis purpurea tissue cultures to 5a-pregnane-3,20-dione and 5a-pregnane-3,8-ol-20-one (I) (3, 5, 20). Furthermore, Rosa sp. tissue cultures were reported to form A4-pregnene-20,8-ol3-one (VI) and A4-pregnene-20a-ol-3-one (II) (12). In this investigation, cultures of L. esculentum and C. frutescens (green) were able to reduce stereospecifically the carbonyl at C20 to yield A4-pregnene-20a-ol-3-one (II), in contrast to V. rosea cultures, which were capable of converting it to the 20,8-hydroxy isomer (VI). The 20,B isomer could not be detected in the cultures of L. esculentum, nor the 20a isomer in V. rosea cultures. The culture of C. frutescens (green) was relatively resistant to inhibition of growth caused by addition of progesterone and
215
was capable of converting it to A4-pregnene-20aol-3-one (II) in yields of 60 to 90%. Such a high yield with a relative absence of by-products is reminiscent of the conversions obtained with the yeast Rhodotorula longissima (4). The 20a reduction has been reported to be carried out by yeasts (3), algae (17), plants (16), and mice (25), but not by bacteria or molds. The cultures of L. esculentum also yielded 5a-pregnane-3,8-ol-20one (I), as has been reported for suspension cultures of D. purpurea (12), D. deltoidea (23), N. tabacum (8), and other plants. This product forms as a result, of stereospecific reduction of the A4-double bond, yielding the trans configuration of the A, B steroid rings, and stereospecific reduction of 3-carbonyl, forming only the 3f,-alcohol. In addition, A4-pregnene-14a-ol-3,20dione (III) was isolated from cultures of L. esculentum. This stereospecific de novo hydroxylation in the 14a position is a common hydroxylation carried out by many strains of Curvularia lunata (28) and Naematoloma sublateritium (21), but has not been previously reported as occurring in plant cell cultures. This metabolite was also isolated from cultures of L. esculentum and V. rosea. Additional new metabolites from the L. esculentum incubation were identified as A4-pregnene-6f8,14a-diol-3,20-dione (IV) and A4-pregnene-6f8,lla-diol-3,20-dione (V). These products resulting from hydroxylation in the 14a and 6,f positions indicate that certain plant cell cultures are capable of introducing a hydroxyl group apparently without the need for the steroid to be conjugated to a sugar moiety. The C6 hydroxylation performed by L. esculentum on an activated allylic position is analogous to non-stereospecific hydroxylation obtained by the use of mercuric acetate (26). The stereospecific hydroxylations of nonactivated positions, as in the case of 14a and hla, performed by the plant cell cultures (as well as by microorganisms) cannot, however, be achieved easily by chemical synthesis. The A4-pregnene-6,f,14a-diol-3,20dione (IV) metabolite is also readily obtained with C. lunata (28), Circinella sp. (21), and Trichothecium roseum (20), while the 6fl,11adiol (V) could be obtained from Aspergillus and Penicillium (6). Although until recently only reductions of unglycosidated steroids have been reported, apparently some cell cultures can carry out oxidative reactions, namely hydroxylations, on unglycosidated steroids. In addition, reductive and oxidative reactions can be performed by the same plant cell culture. The introduction of a hydroxyl group into one or more positions of progesterone (VII) by means of the cultures of V. rosea and L. esculentum was stereospecific. Cultures of L. esculentum yielded hydroxylations in the 6/?, 11a, and 14a positions, whereas V. rosea cultures
216 YAGEN, GALLILI, AND MATELES carried out hydroxylation in the 14a position. Reduction of progesterone by cultures of L. esculentum and C. frutescens led to the 20fi-hydroxy derivative (VI), while cultures of V. rosea produced the 20a-hydroxy isomer (II). Thus, if a new center of asymmetry was formed by hydroxylation or reduction, only one of the possible epimers arises in each case. This is similar to the situation found with microorganisms. Although the yields are generally low as compared to those obtained with microorganisms, the occasional very high yield suggests the possibility of finding additional examples of culturesubstrate combinations with high yields and suggests the desirability of determining whether, in cultures such as Capsicum, there is a connection between the high yield and the apparent insensitivity to steroid inhibition. Although it does not appear likely that plant cell cultures can compete with microorganisms for use in the manufacture of steroid drugs, there is a possibility that plant cell cultures might be used for enzymic conversion to optically active forms, or for specific biotransformation of natural compounds for producing rare or unusual steroids which are difficult to obtain from other natural or synthetic sources. ACKNOWLEDGMENITS This research was supported by grants from the National Council for Research and Development, Jerusalem, Israel, and the Lewis and Rosa Strauss Memorial Fund, Washington, D.C. We thank P. W. O'Connell, The Upjohn Co., Kalamazoo, Mich., for supplying steroid standards. LITERATURE CITED 1. Aifermann, A. W., H. M. Boy, P. C. Doller, W. Hagedorn, M. Heins, J. Wahl, and E. Reinhard. 1977. Biotransformation of cardiac glycosides by plant cell cultures, p. 125-141. In W. Barz, E. Reinhard, and M. H. Zenk (ed.), Plant tissue culture and its bio-technological application. Springer-Verlag, Berlin. 2. Behrend, J., and R. L. Mateles. 1975. Nitrogen metabolism in plant cell suspension cultures. I. Effect ofamino acids on growth. Plant Physiol. 56:584-589. 3. Carvajal, F., 0. F. Vitale, M. J. Gentles, H. L. Herzog, and E. B. Hershberg. 1959. Microbial transformation of steroids. VI. Stereospecific reductions of the 20-carbonyl group. J. Org. Chem. 24:695-698. 4. Chang, V. M., and D. R. Idler. 1961. a-Reduction of the 20-carbonyl group in C-21 steroids by Rhodotorula longissima. Can. J. Biochem. Physiol. 39:1277-1285. 5. Charney, W., and H. L. Herzog. 1967. Microbial transformation of steroids. Academic Press Inc., New York. 6. Dulaney, E. L, W. J. McAleer, M. Koslowski, E. 0. Stapley, and J. Jaglom. 1955. Hydroxylation of progesterone and 11-deoxy-17-hydroxycortecosterone by Aspergillus and Penicillium. Appl. Microbiol.
3:336-340.
7. Dulaney, E. L., E. 0. Stapley, and C. Hlavac. 1955. Hydroxylation of steroids, principally progesterone, by a strain of AspergiUus ochraceus. Mycologia 47:464 474. 8. Furuya, T., M. Hirotani, and K. Kawaguchi. 1971. Biotransformation of progesterone and pregnenolone by plant suspension cultures. Phytochemistry
APPL. ENVIRON. MICROBIOL. 10:1013-1017. 9. Furuya, T., M. Hirotani, and T. Shinohara. 1970. Biotransformation of digitoxin by suspension callus culture of Digitalis purpurea. Chem. Pharm. Bull. 18:1080-1082. 10. Furuya, T., K. Kawaguchi, and M. Hirotani. 1973. Biotransformation of progesterone by suspension cultures of Digitalis purpurea cultured cells. Phytochemistry 12:1621-1626. 11. Gallili, G. E., B. Yagen, and R. I. Mateles. 1978. Hydroxylation of progesterone by plant cell suspension cultures of Vinca rosea. Phytochemistry 17:578. 12. Graves, J. M. H., and W. K. Smith. 1967. Transformation of pregnenolone and progesterone by cultured plant cells. Nature (London) 214:1248-1249. 13. Hirotani, M., and T. Furuya. 1974. Biotransformation of testosterone and other androgens by suspension cultures of Nicotiana tobacum (bright yellow). Phytochemistry 13:2135-2142. 14. Hirotani, M., and T. Furuya. 1975. Metabolism of 5,Bpregnane-3,20-dione and 3f,-hydroxy-5,B-pregnan-20one by Digitalis suspension cultures. Phytochemistry 14:2601-2606. 15. Iizuka, H., and A. Naito. 1967. Microbial transformation of steroids and alkaloids. University of Tokyo Press, Tokyo. 16. Luedemann, G. W., W. Charney, A. Mitchell, and H. L Herzog. 1959. Stereospecific reduction of prednisone by alfalfa seedlings. J. Org. Chem. 24:1385-1386. 17. Luedemann, G., W. Charney, A. Woyciesjes, E. Pettersen, W. D. Peckham, M. J. Gentles, H. Marshall, and H. L Herzog. 1961. Microbial transformation of steroids. IX. The transformation of Reichstein's compound S by Scenedesmus sp. Diketopiperazine metabolites from Scenedesmus sp. J. Org. Chem.
26:4128-4130. 18. Murashige, T., and F. Skoog. 1962. A revised medium for rapid growth and bioassays with tobacco tissue
cultures. Physiol. Plant. 15:473497. 19. Reinhard, E. 1974. Biotransformations by plant tissue cultures, p. 433459. In H. E. Street (ed.), Tissue culture and plant science. Academic Press, London. 20. Schubert, A., K. Heller, D. Onken, K. Zetache, and B. Kluger. 1960. Microbielle Hydroxylierung von Progesterone in 6- und 14-Stellung. Z. Naturforsch. Teil B
15:269. 21. Schubert, A., D. Onken, R. Sibert, and K. Heller. 1958. Microbielle Hydroxylierung von Steroiden in 9und 14-Stellung. Chem. Ber. 91:2549-2552. 22. Stohs, S. J. 1977. Metabolism of steroids in plant tissue cultures, p. 142-150. In W. Barz, E. Reinhard, and M. H. Zenk (ed.), Plant tissue culture and its bio-technological application. Springer-Verlag, Berlin. 23. Stohs, S. J., and M. M. El-Olemy. 1972. Metabolism of progesterone by Dioscorea deltoidea suspension cultures. Phytochemistry 11:1397-1400. 24. Stohs, S. J., and H. Rosenberg. 1974. Steroids and steroid metabolism in plant tissue cultures. Lloydia 38:181-194. 25. Weinstein, W., H. R. Lindner, and B. Eckstein. 1977. Thymus metabolism progesterone-possible enzymatic marker for T lymphocytes. Nature (London) 266:632-633. 26. Wiberg, K. B., and S. D. Nielsen. 1964. Some observations on allylic oxidation. J. Org. Chem. 29:3353-3361. 27. Yagen, B., G. E. Gallili, and R. I. Mateles. 1978. Progesterone biotransformation by plant cell cultures, p. 68-73. In A. W. Alfermann and E. Reinhard (ed.), Production of natural compounds by cell culture methods. Gesellachaft fur Strahlen-und Umweltforschung mbH, Munich. 28. Zetsche, K. 1961. Chemical-physiological investigations concerning the hydroxylation of steroids by fungi of the genus Curvularia. Arch. Mikrobiol. 38:237-271.
