272''
.A. L
?;USSI3AUhl,
G. RR.\R.\ZOK, T. L. P O P P E R
Michaelis-Menten constant having the value of 1.7 x 10-6M. The requirement of sulfhydryl groups for citrovorum factor cyclodehydrase activity is indicated by the inhibitory effect of p-chloromercuribenzoate and o-iodosobenzoate on the rate of formation of the intermediate compound from citrovorum factor and the reversal of inhibition in both cases by reduced glutathione (Fig. 6). Rabinowitz and Price? have reported the occurrence in Clostridium cylindrosporum of an enzyme, formimino-THF cyclodeaminase, which catalyzes a reaction which is analogous to that proposed for citrovorum factor cyclodehydrase. ATP is not required in the reaction they describe.
Experimental Apparatus and Materials.--A Beckman model DU spectrophotometer and 1 cm. Corex cuvettes were used for the enzyme assays. 1 cm. quartz cuvettes were used for the determinations of protein concentration from the ultraviolet (8) J. C. Rabinowitz and W. E. Pricer, Jr., TIUSJOURNAL,78, 5702 (1956).
[CONTRIBUTION FROM
THE
AND
E.
P.OLIVETO
1-01. so
absorption by the method of Warburg and Christiang and a Cary recording spectrophotometer was used for the comparison of the spectra of anhydroleucovoriri and the enzymatically-formed intermediate in the region 320-360 nip. A Beckman model G pH meter was used for PH determinations. Calcium leucovorin was generously supplied by the Lederle Laboratories Division of the American Cyanamid Company, Pearl River, N. Y . , and Solka-Floc BW 200 used for the preparation of carboxymethyl-celldose3 (sodium form), was a gift of the Brown Company, Boston, Mass. ATP was obtained from the Sigma Chemical Company, St. Louis, Mo. Sheep liver acetone powder was prepared as previously described.2 Hydroxylapatite was prepared by the method of Tiselius, et aZ.,'O and calcium phosphate gel was prepared by the method of Keilin and Hartree." Dowex-1 (200-400 mesh) (Microchemical Specialties Company, Berkeley, Calif.) was converted t o the chloride form and columns were prepared by the method of Cohn.Bb (9) 0. Warburg and W. Christian, Biochem. Z., 310, 384 (19411942). (10) A. Tiselius, S. Hjert6n and b. Levin, Arch. Biochem. Biophys., 66, 132 (1956). (11) D.Keilin and E. F. Hartree, Fvoc. Roy. SOL.Lo?zdoiz, Series B. 124, 397 (1938).
BERKELEY. CALIFORNIA
CHEMICAL RESEARCH DEPARTMENT, S C H E R I N C CORP.]
Introduction of Oxygen into Ring B of Corticoids BY A. L. NUSSBAUM, G. BRABAZON, T. L. POPPER AND E. P. OLIVETO RECEIVED JASUARY 2, 1958 Epoxidation of 6-dehydroprednisone acetate gave the 6a,7a-epoxide as well as some 17-ketone V. T h e former was converted to the 7a-hydroxy derivative either directly with chromous acetate or via a bromohydrin. Corresponding reactions with 6-dehydrocortisone acetate led t o 7a-hydroxycortisone acetate. T h e latter was converted microbiologically t o the new 7a-hydroxyprednisone. T h e A4-compound dehydrates normally, but the A1s4-dieneis resistant to dehydration. Arguments in support of the indicated structures are presented.
The introduction of an additional double bond a t 11.88 p, assigned to the epoxide f ~ n c t i o n . ~The into the prednisone molecule by Gould, et aL,l and nature of the newly-introduced oxygen was further by Agnello and Laubach2 furnished a means for indicated by reversion of I I a to starting material further modification of the steroidal B ring in struc- upon treatment with hydriodic acid.5 Stereotures possessing physiological activity. Prefer- chemical assignment of the epoxide function to the ential attack of the 6,7-double bond by peracids a-side of the molecule rests on steric considerations was observed when 17a ,2 1-dihydroxy-A1r4s6-pregna- (rule of the rear6) as well as the nature of certain trien-3,11,2O-trione 21-acetate1,2 (I) was exposed transformation products now to be described. Halohydrins I11 were prepared by standard to such reagents. Two products were isolated: the 6o,7o-epoxide I I a and its corresponding 17- procedures. Thus, a bromohydrin IIIa, a chlorohydrin IIIc and a fluorohydrin IIIe, as well as the ketone V. The assignment of structure of substance I I a fol- corresponding diacetates IIIb, d and f were oblows from the following considerations : Elemen- tained. Normal oxide opening would lead to the tary analysis of both 21-acetate and the corre- halohydrins formulated as in 111, Chart I (6@-halo, sponding 21-01 I I b showed the presence of an addi- 7a-hydroxy) and a consideration of their ultrational oxygen atom in the molecule. The ultra- violet absorption spectra lends support to this forviolet spectrum of the cross-conjugated trienone mulation. There is a progressive drop in the o,@starting material had disappeared, and the new unsaturated ketone absorption ; thus, bromohydrin substance had instead a single absorption maxi- IIIa has its maximum a t 244 m p , chlorohydrin mum a t 240 m p . This excluded attack a t either of I I I c a t 240 mp and fluorohydrin IIIe a t 236 mp. the two ring A double bonds, a consideration that A similar picture has been observed by Bird and cowas confirmed by the infrared spectrum. The lat(4) W, A. Patterson, Anal. Chem., 26, 823 (1954),finds an absorpter showed absorption peaks a t 6.18 and 6.24 p, tion band a t ll.B 1.1 for a series of simple aliphatic epoxides. ) E. P. Oliveto, C. Gerald and E. B. Hcrshberg, 'THIS JOURNAL, characteristic of cross-conjugated dienones, and 7 9(, 53396 (1957). (1) D. Gould, E. I.. Shapiro, H. L. Herzog. M. J. Gentles, E. B. Hershberg, W. Charney, M. Gilmore, S. Tolksdorf, M. Eisler, P. J,. Perlman and M . Pechet, THISJOURNAL, 79, 502 (1957). (2) E. J. Agnello and G. D. Laubach, ibid., 79, 1257 (1957). (3) J. Fried, R. W. Thomu and A. Klingsberg, ibid., 7 5 , 57G4 (1953).
(6) L. Fieser, Erperiewtia, 6,312 (1950). (7) Oxides generally open to t h e trans-diaxial halohydrins; see, f o r instance, A. Fiirst and P. A. Plattner, Abstracts, 12th International Congress of Pure and Applied Chemistry, hrew York. 19,51, 11. 409. Abnormal ring openings, however, have been reported: cf. i i i l r r al.. W. S. Knowles and Q . E. Thompson, THISJOURNAL, 79, 3212 (1037).
OXYGENINTO RINGB
June 5, 1958
OF
CORTICOIDS
2723
CHARTI CHzOAc
CHzOR
c=o
c=o
I
I
0
.. 0
. .
0
CHzOAc
CHzOAc
c=o
c=o
II
b; R = Ac
I
x
workerss in a structurally related A4-cholesten-3one series substituted in the 6i3-position. This effect was not observed in the 6a-position, and certainly could not be expected if the halogens were located a t C-7, where intimate interaction of the aelectron system is unlikely. Thus, formula I11 (Chart I) for the halohydrins is supported and consequent assignment of the oxide I1 to the a-side follows. Attempts to oxidize the bromohydrin IIIa, either with chromium trioxide-pyridineg or chromic acid in acetonelo were unsuccessful, as were several dehydration attempts, either on I I I a itself, or v i a a 70-mesylate. l1 Treatment with zinc in ethanol furnished a new hydroxyprednisone acetate, formulated as IVa. This compound had the correct elementary analysis, as did a diacetate IVb, and was considerably more polar in a paper chromatographic system12 than the starting material. Here again several attempts a t oxidation and dehydration were unsuccessful, but conversion to a mesylate followed by treatment with refluxing 2,6-lutidine gave a mixture of products, the ultraviolet spectrum of which showed some absorption a t 297 mp, indicating that partial dehydration to the triene I had occurred. I n order to bolster the structure assigned to the presumed 7a-hydroxyprednisone acetate IVa, we subjected substance IIa, the 6a,7a-epoxide, to the action of chromous acetate. This reagent has been used successfully to convert 16a,17a-epoxy(8) C. W. Bird, R. C . Cookson and S. H. Dandegaonker, J. Chem. 3675 (1950). (9) G. I. Poos, G. E. Arth, R. E. Beyler and L. H. Sarett, THIS J O U R N A L , 76, 422 (1953). (10) R. G. Curtis, I. A. h'. Heilbron, E. R. H. Jones and G. F. Woods, J. Chem. SOC.,461 (1953). (11) N. L. Wendler, D . T a u h , S. Dohriner and D. K . Fukushima. THIS J O U R N A L , 7 8 , 5027 (1956). (12) G. M. Shull, D. A. Kita and J. W. Davisson, U.S. Patent 2,658,023. SOC.,
0
f
F
Ac
2 1-acetoxy-A4-pregnene-3,11,20-trione into 16a,2 1dihydroxy - A4 - pregnene - 3,11,20 - trione 21acetate,l 3 an analogous case. Prednisone acetate itself proved to be stable toward this reagent. When 6a,7a-epoxy- 17a,2 1-dihydrox y-A1,4-pregnadiene-3,11,20-trione 21-acetate (IIa) was treated with chromous acetate,14 there was obtained, in low yield, the same 7a-hydroxyprednisone acetate IVa, that had been previously formed by debromination of bromohydrin I I I a with zinc. The resistance of IVa toward dehydration must, however, be considered anomalous in view of the fact that 7a-hydroxydesoxycorticosterone acetate can be converted easily into the corresponding dienWe were able to confirm this observation in the case of a compound possessing the full cortical side chain. 6-Dehydrocortisone acetate16 was converted into its 6a,7a-oxide VIII, and a corresponding bromohydrin I X 1 7 was prepared. Debromination with zinc in ethanol gave a highly polar hydroxylated derivative of cortisone acetate, undoubtedly 7a,l7a,21-Trihydroxy-A4-pregnene-3,11,20trione 21-acetate (X). The latter compound, unlike its 1-dehydro analog, was dehydrated with expected ease. When i t was subjected to ultraviolet measurement under alkaline conditions,ls the original maximum a t 238 mp shifted to that of the dienone at 282 mp. The fact that the two 7ahydroxycorticoids IV and X differed only in the A1(13) P. L. Julian, W. Cole, E. W. Meyer and B. M. Regan, THIS 7 7 , 4601 (1955). W. Cole and P: L. Julian, J . Org. Chem., 19, 131 (19.54). Ch. Meystre, E. Vischer and A. Wettstein, Helv. Chim. Acta, (1955). V. R. Mattox, E. L. Woroch, S. A. Fleisher and E. C. Kendall, J. Biol. Chem., 197, 261 (1952). (17) F. B. Colton, in U.S. Patent 2,738,348 (March 13, 1957), reported an a-oxide and the corresponding bromohydrin from 6dehydrocortisone acetate. However, preparation of our compounds are given in the Experimental portion, since physical constants of the two series are quite different. (18) A. S. Meyer, J. Bid. C h e w . , 203, 4ti9 (1953). JOURNAL,
(14) (15) 38, 38 (16)
A. L. NUSSBAUM, G. BRABAZON, T. L. POPPERAND E. P. OLIVETO
2724
CHARTI1 CHzOAc
CHzOAc
I c-0
c=o
Vol. 80
E. B. Hershberg for his kind help with the preparation of the manuscript. Experimental'3
6a ,7a-Epoxy- 1 7,2~1-dihydroxy-A1 4-pregnadiene-3, 11JOtrione 21-Acetate (IIa).-17a,2l-Dihydr0xy-A~~~~~-prcg~iatriene-3,11,20-trione 21-acetate (I),l 24.66 g., was dissolved in 2.7 1. of methylene chloride, and 1.4 1. of ethereal monoperphthalic acid (47.3 mg. of peracid/ml.) was added. After 65 hours at room temperature, 3.1 1. of methylene cl~loridc was added, and the solution washed twice with 2.5 1. of saturated sodium carbonate solution. The latter was backextracted with methylene chloride, and the combined organic layers dried and concentrated. By differential crystallization from isopropl-1 ether, a total of 19.1 g. of IIa as well as 266 mg. of the degradcd 6a,7a-epo~y-A~~~-androstadiene-3,11,17-trione (V) could be isolated (see later). (In anothcr run, 17.5 g. of I gave 6.89 g. of I I a and 4.70 g. of V.) Substance IIa had a melting point of %57-264", e;:;"" 15,130, Recrystallizations from isopropyl ether gave an analytical 15,600, sample, m.p. 262-276' dec., [ c Y ] * ~ Df l 5 6 . 3 , E$";:' Rf0.35 to 0.4312; Xp:l a t 2.96, 5.72, 5.84, 6.02, 6.18, 6.26, 7.87, 8.11 and 11.88 p . Anal. Calcd. for C23H260,: C, 66.65; II, 6.32. Found: C. 66.95:, H., 6.42. ' I 6a,7a-Epoxy-17a,21-dihydro~y-~~~~-pregnadiene-3,11,20VI1 trione (IIb).-Acetate I I a , 900 mg., was dissolved in 200 nil. of methanol and 2.33 nil. of a 4si2 aqueous sodium hyZn/ElOH I t droxide solution was added. After 25 minutes a t room temperature, the solution was neutralized with acetic acid and O / v v''.OH C ) / v l' concentrated to a small volume. Water was added, and I s Br I S resulting crystals filtered; yield 402 mg., 1n.p. 23j-240 . Recrystallizations from acetone gave an analytical sample, 14I 800.> double bond present in I v but absent in X was m.P. 250-265", [ a ] " D 3-86.5"(pyridine),fI'& '":A:: a t 3.00, 5.80(sh), 5.85, 6.01 and 6.15 p . proved by the microbiological introduction Anal. Calcd. for CZ1H2106:C, 67.73; H, 6.50. Found: of this double bond into When the latter com- C, 68.02; H, 6.42. pound was subjected to the dehydrogenating action Reaction of IIa with Hydriodic Acid.-6a,'ioc-Epoxy-iia,of B. sphaericus, the crude extract gave IVb after 21-dihydro~y-A.~~~-pregnadiene-3,11,20-trione21-acetate (IIa) (200 mg.) was dissolved in 25 ml. of methylene chloacetylation and chromatography. Aqueous 48% hydriodic acid (2 ml.) was added, and It was mentioned earlier that when 17a,21-di- rlde. the solution was shaken in the dark for 15 minutes at room hydro~y-A~$~,~-pregnatriene-3,11,20-~iOne "-acetemperature. After dilution with 2G mi. of water, the ortate (I) was subjected to peracids, a second prod- ganic layer was separated and the aqueous phase was further uct was isolated in addition to the (ja,ia-epoxide extracted twice with methylene chloride. Combined extracts were washed with 20-ml. portions each of 5% sodium 1 1 ~ .This was not the corresponding bicarbonate and 5%, sodium carbonate solutions, and water. since i t did not give a purple color with 2,395-tri- -iJter drying and vacuum concentration, the residue was phenyltetrazolium chloride, l9 indicating the ab- crystallized from methylene chloride-isopropyl ether to give Sence of the dihydroxyacetone side chain. The in- 175 mg. of 17~,2l-dihydroxy-A~,~,~-pregnatrierie-3,1 1,20frared spectrumshowed absence of h y ~ o x y lab- trione 21-acetate (I), identical with genuine material in spectral and melting point comparison. sorption and a band that could be attributed to a 6p-Bromo-7a, 17a,2l-trihydro~y-A',~-pregnadiene-3,11,five-membered ring carbonyl. The compound gave 20-trione 21-Acetate III(a).-The epoxide 11%(2.5 g.), (lisa positive Zimmermann20 test and, accordingly, solved in 150 ml. of chloroform, was shaken with 63 1111. of ~~,701-epoXy-~i,4-androstadiene-3, 11,- 48% aqueous hydrobromic acid. The mixture was extracted with ethyl acetate, and the extract washed with li-trionei was assigned to the substance in ques- water, 574 aqueous sodium sulfate and again with lvater. tion. This structure was proved by the following Drying and concentration followed by crystallization from transformations: (1) reaction of V with hydriodic acetone-isopropyl ether gave a product which u n s seen to heterogeneous by paper c11romatw3raphY1* and could acid gave Al,4,6-androstatriene-3,11, li-trione (VI) be not be purified by recrystallization. The dificulty appeared a known the latter, when treated to be partial hydrolysis of the acetate a t C-21, so the ciitirc with perphthalic acid, reverted to V; (3) degrada- crystalline residue was dissolved in i n 1111. of pyridine and tion of 6a,7a-epoxy-l'ia,2l-dihydroxy-Al,~-pregna- a n equal volume of ncctic anhydride was added. After f(mr minutes, the solution \vas diluted iTith 75 ml. of water anti diene-3,11,2()trione (11b) by sodium bismUthate?l the resulting solid filtered, washed with water a i d recrvsDegradation Of the side furnished tallized from acetone-isopropyl ether, yielding 1.34 g., myp. by peracids had been observed earlier.22 190-194'. An analytical sample was prepared by scvcral from .the same solvent system: 111.p. 195~ ~ ~ ~ ~ are lindebted ~ to d D ~~ D.~ ~ .recrystallizations ~ ~ -Rt 0.15, ~ ~ 16,770, 198' dec., [ o ~ ] t~ +153.1", ~D Xi;:"' at H . R. Barton, J. Meinwald and R. K. Hill for help- 2.99, 3.10, 5.70, 5.76, Ij.82, 6.01, 6.17 and 8.09 p . fd advice, to Dr. UT. Charney for the microA n d . Calcti. for CsaH2;0;Br: Br, 16.11. Found: Ijr, biological transformation, and especially to Dr. 15.47. A diacetate I I I b was prepared with :icetic :i1111ytlritlc (19) R. B. Burton, A . Zaffaron; and E . H . Keutmann, J . B i d . Cherrt., pyridine in the usual manner, and crystallized from Illetllyl3
+
Pi')'--
v'
tb
x.
v,
~
v.
188, 763 (1951). (20) W. Zimmermann, Z . ghysiol. Chent., 246, 47 (1936). (21) C . J . W. Brookes and J. K . Norymberski, Bioche??~. J . , 56, 371 (1953). ( 2 2 ) N . S. Leeds, I). K . Fukushima and T. 1;. Gallagher, T H I S J O U R N A L , 76, 3265 (1954).
(23) All melting points mere taken o n a Kofler bluck. I
