MICROBIOLOGICAL TRANSFORMATIONS OF STEROIDS. IX

COMMUNICATIONS. TO THE EDITOR. Vol. 75. CoA, also occurs ... of oxidative decarboxylation of pyruvate. These data confirm the previous suggestion^^*^...
0 downloads 0 Views 264KB Size
Vol. 75

COMMUNICATIONS TO THE EDITOR

5768

chain of 4-cholestenone to a (2%compound, 3-keto4-etiocholenic acid, is accomplished by Proactinomyces. In these microbiological transformations, we have found that it is possible to degrade the side chain of C21 steroids to Cl9 compounds by a simple and novel one step process without the necessity of protecting the sensitive 3-keto-A4system. The methylene chloride extractives of the beer LABORATORY OF BACTERIOLOGY L. P. HAGER from the Gliocladium were chromatographed over ~ OF ILLINOIS I. C. GUNSALUS alumina to yield 11, m.p. 174-176', [ a I z 3+194O UNIVERSITY URBANA ( c 0.8695 in CHCIa), (Anal. Calcd. for C19Hzs02: RECEIVED OCTOBER 2, 1953 C, 79.68; H, 9.15. Found: C, 79.53; H, 8.84 and 111, m.p. 190-192", [ a ] 2 3f107' ~ (c 0.6685 in CHCla) (Anal. Calcd. for C19H2603: C, 75.46; MICROBIOLOGICAL TRANSFORMATIONS OF H, 8.67. Found: C, 75.39; H, 8.47). ComSTEROIDS. IX DEGRADATION OF Ca1 pound I11 yielded a monoacetate (V), m.p. 202STEROIDS TO Cip KETONES AND 205O, [ a ] 2 3+114O ~ (c 0.9353 in CHC13) (Anal. TO TESTOLOLACTONE Calcd. for C21H2804: C, 73.22; H, 8.19. Found: Sir: In continuing our studies on microbiological C, 72.97; H, 8.01). By similar techniques the transformations of steroid^,'^^*^ we wish to report beer from the Penicillium yielded I1 from I. The physical constants of I11 and V are identical the degradation of CZIcompounds, particularly pro- with those reported by Ehrensteh8 Oxidation of gesterone (I),to CIScompounds: e.g., 4-androstene3,17-dione (11), 6/3-hydroxy-4-androstene-3,17-di-I11 produced 4-androstene-3,6,17-trione (VII), m.p. one (111) and testololactone (IV).4 These results 220-224', whose infrared spectrum was identical with an authentic sample. are given in the table. Microbiological transformation of progesterone (I) and 17a-hydroxyprogesterone (VI) with AsperBIOCONVERSION OF CI1 TO CI9STEROIDS giZlus j a w s yielded testololactone (IV), m.p. 210Substrate Product Microorganism Progesterone Androstenedione (11) Gliocladium catenu- 212", [ Q ] * ~ D +43" (c 1.00 in CHC13), (Anal. for C19H2603: C, 75.46; H, 8.67. Found: (1) + other oxygenated laturn, Penicil- Calcd. steroids lium lilacinurn C, 75.61; H, 8.51); androstenedione (11)and other Thorn, Aspergil- oxygenated steroids. Similarly I was transformed to I V by Penicillia. The product obtained by lus &VUS these bioconversions was identical in all physical 6j3-Hydroxyandrostene- Gliocladium catenuproperties with IV obtained by chemical means,* dione (111) lalum Testololactone (IV) Aspergillus $flaws, the structure of which has been definitely establi~hed.~ other oxygenated Penicillium When the fermentation is interrupted before the steroids adametzi substrate is completely transformed, I1 as well as I V l7a-Hydroxy- Testololactone (IV) Aspergillus $avus can be isolated; however, when the substrate is progesother oxygenated CoA, also occurs, ;.e., reaction 1. Thus over a 200-fold range of purity, the lipoic dehydrogenase activity of E. coli fraction B, a t least in high substrate concentration, is sufficient to account for the total rate of oxidative decarboxylation of pyruvate. These data confirm the previous suggestion^^*^ that the B fraction of E. coli is a dehydrogenase enzyme common t o the keto acid systems and that it links lipoic acid to DPN.

+

-+

terone (VI) Other substrates

steroids (5)

PIT3 I c -0

0

Gliockzdium catenzllatum (A.T.C.C. 10523) converts I to I1 and 111, while a strain of Penicillium lilacinurn Thorn*converts I to I1 plus other oxygenated steroids. The fermentation and extraction techniques used have been previously described.1, Turfitt' has shown that degradation of the side (1) D. H. Peterson, H. C. Murray, S. H. Eppstein. P. D Meister, L. M. Reineke, A. Weintraub and H. M. Leigh, TRISJOURNAL, 7 6 , 421 (1953), and earlier references (2) H. C. Murray and D. H. Peterson, U. S. Patent 2,602,769, C A . 46, 8331 (1952). (3) Microbiological transformations of steroids have also been reported by D. Perlman, e l al., TXIS JOWRNAL. 74, 2128 (1952); J. Fried, el al, ibid., 74, 3962 (1952); 0. Mancera, e; al., ibid.. 74, 3711 (1952): and F. W. Kahnt, et al, Ezperientia, 8, 422 (1952) ( 4 ) H. Levy and R. P. Jambsen, J . Bid. Ckem., 171, 61 (1947). (5) Studies have indicated that desoxycortfcwterone and Relchstein's Compounds S can be converted to 4-androotene-3,17-dione (11) or testololactone (IV) and other oxygenated steroids by these cultures as well as by various Aspergilli and Penicillia. (8) Oxidation of 140-hydroxyprogesterone to 1401-hydroxy-4-nndrostened,l74ione by this micro6rganism was reported by our group at 123rd Meeting Am. Chem. SOC., Los Angelea, California, March 1519, 1953 (Division of Biological Chemistry, Abstract 5C), (7) G . 12. Turfitt, Biochcm. J., 41, 376 (1948).

R

II1,R -OH 11, R = -0Ac 1-11, R = -0 ( 8 ) C. P. Balant and M. Ehrenstein, J . Org. Chsm., 17, 1587 (1952). (9) M. F. Murray, K. L. Pederson, B. A. Johnson and A. C Ott, =IS

JOURNAL, to be published.

Nov. 20, 1953

COMMUNICATIONS TO THE EDITOR

5769

probable and since measurement of the acidity of ZrCL solutionss made it unlikely that an infinite series of polymers exists a t high acidities, equilibrium ultracentrifugations of Zr(1V) were carried out in chloride and perchlorate s o l ~ t i o n s . ~ The data were recently augmented and reanalyzed by a modification of the method of Lammlo which was suggested to us by Professor George Scatchard, details of which will be published separately. In this computation the charge of the polymer units was considered, and estimates of this charge were obtained by ultracentrifugations under a variety of conditions. Centrifugation of 0.05 M Zr(1V) solutions in 1 M HC1-1 M MC1 (where M was Li, Na and Cs) revealed the existence of only one principal species of Zr(1V) with an apparent degree of polymerization of 3.0 and charge Z’ < 1 per monomer unit. At considerably higher and lower acidities (3 M HC1 and 0.1 M HCl-1.9 M NaC1) mixtures (10) D. R. Colingsworth, M. P. Brunner and W. J . Haines, THIS were found with apparent degree of polymerization JOURNAL, 74, 2381 (1952). varying between cu. 2 to 2.6 (3 M HCI) and 4 to 5.4 (11) W. J. Haines, “Recent Progress in Hormone Research,” Vol VII, Academic Press, Inc., New York, N. Y., 1952, p. 255. (0.1 MHCl). (12) J. von Euw and T. Reichstein, Helo. Chim. A d a , 26, 988 Similar low degrees of polymerization were (1942); 24, 879 (1941). found in ultracentrifugations in perchlorate solu(13) H. L. Mason and W. W. Engstrom, Physiol Rev., SO, 321 tions (1 M HC104-l M NaC104). In this medium (1950). D. H. PETERSON the zirconium particles appeared to carry a considS. H. EPPSTEIN erable charge and hence preliminary estimation of RESEARCH LABORATORIES P. D. MEISTER the degree of polymerization is somewhat more unH. C. MURRAY THE UPJOHNCOMPANY KALAMAZOO, MICHIGAN H. M. LEIGH certain than for the chloride solutions. The most A. WEINTRAUB probable degree of polymerization for 0.05 and 0.12 L. M. REINEKE M Zr(1V) solutions in this medium was 3, with an RECEIVED OCTOBER19, 1953 outside possibility that i t may be as high as 4.5. There was no indication of an increase in degree of polymerization with concentration. These results HYDROLYTIC POLYMERIZATION OF may be compared with the (weight average) degrees ZIRCONIUM( IV)! of polymerization (Nw)estimated by Connick and Sir: Reas.* These authors report N , = 18 for a conGraner and Sillen2suggested that in the hydroly- siderably more dilute solution (0.03 M Zr(1V)-1 M sis of Bi(II1) a continuous series of particles is HClOI-1 M LiC104) and a value of Nwbetween 10 formed, all in equilibrium with each other and rang- and 300 a t a higher acidity (0.17 Zr(IV)-2MHC104). The results of the ultracentrifugation experiments ing in size from monomers to “infinitely” large polymers, the exact distribution depending on acidity described here thus indicate that Zr(1V) in strongly and concentration. Recently Connick and Reasa, acidic solutions ( M H + > 0.1) does not show conin an attempt to interpret solvent extraction data tinuous polymerization with high molecular weight on Zr(IV), advanced the same hypothesis of con- products but rather forms only low molecular tinuous polymerization and equilibrium between weight polymers with trimers apparently predomithe species and postulated the existence of high mo- nating a t acidities near 1 M . lecular weight particles in acidic solutions of Zr(1V). (8) K. A. Kraus and S. Y. Tyree, Jr., Report ORNL-499 (Sept. Other ~ o r k e r s ~drew * ~ , the ~ conclusion that only 1949). (9) J. S. Johnson and K. A. Kraus, Reports ORNL-607 (1949), low molecular weight polymers are formed in (1951). strongly acidic media. High molecular weight ORNL-1053 (10) 0. Lamm, Arkio Kcmi, Mincrol. Ccol., 17A,No. 25 (1944). polymers, not in equilibrium with the more “norOAKRIDGENATIONAL LABORATORY mal” species, are apparently formed under consid- CHEMISTRY DIVISION KURTA. KRAUS erably drastic conditions (e.g., lower acidity or after OAKRIDGE,TENNESSEE JAMESS. JOHNSON boiling) RECEIVED SEPTEMBER 22, 1953 Since the assumption that high molecular weight polymers are in equilibrium with low molecular REDUCTASE, A FERRO-FLAVOweight polymers and monomers appears rather im- DPNH-CYTOCHROMEPROTEIN’ (1) This document is baaed on work performed for the Atomic Sir : Energy Commission at the Oak Ridge National Laboratory: “HydrolWithin the past year several flavoprotein enytic Behavior of Metal Ions. 11,” previous paper, THISJOURNAL, 72, zymes have been shown to contain heavy metals as 3901 (1950). (2) F. Grant%and L G. Sillen Acfo Chcm. Scand., 1, 631 (1947). part of their prosthetic groups. Copper was iden(3) R. E.Connick and W. H. Reas, T H IJOURNAL, ~ 73, 1171 (1951). tified as a constituent of butyryl CoA dehydrogen(4) M. Adolf and W. Pauli, Kolloid Z., 29, 173 (1921).

completely utilized, 11 could not be found. This evidence suggests that the reaction to form IV proceeds through 11. I n previous papers from these laboratories, 6 and 11 hydroxylation and reduction of the double bond in the A ring by microorganisms have been described. The present communication describes the formation of 4-androstene-3,17-dione from Czl steroids. It is interesting to note that these metabolic end products from steroidal substrates are similar to those produced by higher vertebrates.11J2J3 Acknowledgment.-Authentic testo olactone was kindly furnished us by M. F. Murray and B. A. Johnson of these laboratories. For technical assistance we are grateful to J. R. Heald, J. I. Mejeur, I. N. Pratt, H. Triemstra, H. AT. Woltersom and G. Staf‘ien. 1.2,3910

.’

(5) G. Jander and K. F. Jahr. Kolloid-Bcih., 43, 295 (1936). (6) B. A. Lister and L. A. McDonald, J . Chcm. SOC.,4315 (1952). (7) See, e a . , R. Ruer, 2. anorg. u. ollgcm. Chcm., I S , 282 (1905).

(1) Paper IV in a series entitled Studies on Diphosphopyridine Nucleotide-Cytochrome c Reductase. For paper 111 see L. P. Vernon, H. R. Mahler and N. K. Sarkar, 1. B i d . Chcm., 199, 598 (1952).