Steroids. CXXIV.1 Studies in Cyano Steroids. Part I. The Synthesis of a

Steroids. CXXIV.1 Studies in Cyano Steroids. Part I. The Synthesis of a Series of C-6-Cyano Steroid Hormones. A. Bowers, E. Denot, María Blanca Sánc...
0 downloads 0 Views 1MB Size
Oct. 5, 1959

C-6-CYANO STEROID

16,3p-Dihydroxy-l6~u,17a-oxido-~~-pregnen-20-one (VII). -1@,3P-Dihydro~y-A~~~~-pregnadien-2O-one (IV) (895" mg.) was dissolved in 68 ml. of methanol and cooled to 10 At this temperature, 1.8 ml. of 4 N sodium hydroxide and 3.6 ml. of 30y0 hydrogen peroxide were added, and the resulting solution was allowed to stand a t 5' overnight. I t then was neutralized with acetic acid, 100 ml. of water was added, and the volume of the resulting suspension was reduced in vacuo to ca. 120 ml. Cooling and filtration gave 693 mg., m.p. 190-197 . Recrystallization from benzene gave the analytical sample, m.p. 200-202" (after drying for three kh'~j0'2.88, 5.90 p . hours a t 100"in uacuo); [ C X ] * ~ D-8.9'; A n a l . Calcd. for C2,Ha004:C, 72.80; H, 8.73. Found: C, 72.80; H, 8.67. 16P-Bromo-lfi,3p, 17~~trihydroxy-Y-pregnen-20-one 1,3Diacetate (VIII).-The foregoing epoxide (VII, 1.25 g.) was acetylated with pyridine-acetic anhydride and, upon the usual workup (recrystallization from isopropyl ether), gave 1.4 g. of a diacetate, m.p. 158-162. This was dissolved in 50 ml. of glacial acetic acid, 5 ml. of a solution of acetic acid saturated with hydrogen bromide was added, and the reaction was allowed to proceed for 1 hour a t room temperature. The solution then was quenched in a large amount of water, the resulting solid was filtered off (1.53 g., dec. 190-195") and used as such in the next step. 1@,3P,17~u-Trihydroxy-A~-pregnen-2O-one1,3-Diacetate (IX).-Bromohydrin VI11 (980 mg.) was dissolved in 25 ml. of tetrahydrofuran, 0.24 ml. of triethylamine was added, and the solution was hydrogenated over 400 mg. of 570 palladium-on-charcoal. After one mole of hydrogen had been absorbed, the catalyst was filtered off, amine hydrobromide mas washed out in a separatory funnel, and the organic layer was dried and concentrated in vacuo. The residual oil was chromatographed over Florisil, and from the benzene-ether eluates 578 mg. of I X , m.p. 174-179", was obtained. Recrystallization from isopropyl ether gave an

.

[COXTRIBUTION FROM

Steroids. BY

CXX1V.l

THE

HORMONES

5233

analytical sample, m.p. 1T9-180°, [ c Y ] ~ ~ D-34.3; XNujol at 2.90,2.95, 5.77,5.86, 8 . 1 0 ~ . A n a l . Calcd. for C2jHa606: C, 69.42; H, 8.39. Found: C, 69.43; H, 8.28. 1@,3~,17~,21-Tetrahydroxy-P6-pregnen-2O-one 1,3,2l-Triacetate(X).-The foregoing diacetate I X (428 mg.) was dissolved in 7 ml. of chloroform, and 1.8 ml. of a solution of bromine in chloroform (3.68 g. in 40 ml.) was added. Upon the addition of a small amount of hydrogen bromide in chloroform, the reaction mixture decolorized. After standing for 25 minutes a t room temperature, 2 g. of potassium acetate, 7.2 ml. of methanol and 850 mg. of sodium iodide were added, and the mixture stirred a t 45" for 45 minutes. Icewater (20 ml.) was added, and the color of the liberated iodine was discharged with a dilute solution of hydrazine. The aqueous phase was extracted with chloroform, the latter extract was washed with water, dried over potassium acetate, and concentrated to dryness in z~aczto. Potassium acet a t e (2 g.), 2 ml. of water and 20 mt. of acetone were added, and the mixture was refluxed for 17 hours. It then was concentrated t o a low volume, extracted with chloroform, the extract washed with water and dried over magnesium sulfate. Filtration and concentration in vacuo gave an oil which was chromatographed over Florisil (30 g.). Eluates with benzene-ether (1: 1) gave 43.5 mg. of crystalline X, m.p. 169-177', positive color reaction with red tetrazolium; X";jo1 2.86, 5.78ibroad) p . 1a. 17a .2 l-Trihvdroxv-A4-~reanene-3,20-dione (XI).-The foregoing triacetate (X,3 8 m i . ) mas subjected'to the action of Flavobacterium dehydrogenans, as described for the preparation of V. The crude extract was chromatographed over silicic acid," and with 1% methanol in chloroform, 19 mg. of the a,@-unsaturated ketone X I was eluted, m.p. 203-207" (soft. a t 180). I t s infrared spectrum was identical with XI obtained by a different route.24 BLOOMFIELD, N. J.

RESEARCH LABORATORIES OF SYSTHEX, S. A,]

Studies in Cyano Steroids. Part I. The Synthesis of a Series of C-6-Cyano Steroid Hormones

A. BOWERS,E.DENOT,h f A R f A BLANCASANCHEZ, L. &I.

SANCHEZ-HIDALGO AND

H. J. RINGOLD

RECEIVED APRIL 13, 1959 The fission of steroid 3@-acetoxy-5a,6cu-epoxides and 3-cycloethylene-ketal-5cu,6a-epoxideswith potassium cyanide has been studied. Dependent upon the conditions the reaction was attended with or without concomitant elimination of the initially formed 5 ~ h y d r o x y group l t o afford either the 5a-hydroxy-6-cyanide or the A5-6-cyanide. Under more vigorous conditions a double elimination a t C-3 and C-5 occurred t o form A3a5-6-cyanides. Manipulation of appropriate intermediates gave the 6-cyano-A4-3-ketones, which exist mainly in the enol form. Application of this general method led t o t h e synthesis of the 6a-cyano derivatives of progesterone, 17a-acetoxyprogesterone, testosterone and cortisone.

In 1953, Fried and Sabo made a significant contribution to steroid hormone studies by disproving the then widely held view that structural modification of the adrenal hormones cortisone and hydrocortisone always led to a decrease in biological activity. They demonstrated that introduction of a chlorine2 or a fluorine3 atom a t C-~(CY) significantly enhanced the anti-inflammatory activity of the parent hormone. This finding stimulated considerable interest in modified steroid hormones and prodigious efforts have been made to introduce atoms or groups or further unsaturation a t key positions throughout the steroid nucleus. For both chemical and empirical reasons the position that has attracted the most attention has been C-6 and the earliest modification was the (1) Steroids CXXIII, R. Villotti, C Djerassi and H. J. Ringold

THISJOURNAL, 81, 4566 (1959). ( 2 ) J. Fried and E. F. Sabo, t b z d , 76, 2273 (1963). (3) I. Fried and E. F. Sabo, ibid., 76, 1455 (1954).

preparation of a series of 6-hydroxy (or acetoxy) and 6-keto steroid hormones. Amongst the more important may be mentioned the 60-hydroxy ~~~ 4,6*7 desoxyanalogs of t e s t o s t e r ~ n e ,progesterone, c o r t i c o ~ t e r o n eReichstein's ~ ~ ~ ~ ~ ~ ~ Compound S7*10J1 and cortisone.I1 The corresponding C-6 ketones of all the above compounds have also been pre(4) C. Amendolla, G. Rosenkranz and F. Sondheimer, J . Chem. Soc., 1226 (1954). ( 5 ) J. Romo, G. Rosenkranz, C. Djerassi and F. Sondheimer, J . Org. Chcm., 19, 1509 (1954). (0) M. Ehrenstein and T. 0. Stevens, ibid., 8, 318 (1940). (7) C. P. Balant and M. Ehrenstein, ibid., 17, 1587 (1952). (8) P. T. Herzig and M . Ehrenstein, ibid., 16, 1050 (1951). (9) S. H. Eppstein, P. D. Meister, D. H . Peterson, H . C. M u r r a y , H . M . Leigh, D. A. Lyttle, L. M . Reineke a n d A. Weintraub, Tms JOURNAL, 76, 408 (1953). (10) D. H . Peterson, S.H . Eppstein, P. D . Meister, B. J. Magerlein, H. C. M u r r a y , H. M. Leigh, A. Weintraub and L. M. Reineke, ibid., 76, 412 (1953). (11) F. Sondheimer, 0. Maneera and G. Rosenkranz, ibid., 76, 5020 (1954).

5234

A. BOWERS,E. DENOT,21. B. SANCHEZ, L. M. SANCHEZ-HIDALGO AND H. J. RINCOLD 1701. 81

pared4>"-l4 and the 6a-hydroxy analogs of progesteroiie,6 desoxycorticosterone7~scortisonelj and P:ii ti31 steroid S= formulas OH+, Cl+ predni~one'~ are known. However in every case the introduction of an x dr S0.f oxygen function a t C-6 led to a decrease in biI I1 ological activity. Improved biological activity Electrophilic cyanide (CN+) is not readily was first realized in a C-G modified steroid hormone but such compounds as cyanogen and the ~~ with the introduction of the C-6-methyl g r o ~ p ' ~ -available] cyanogen halides can be considered as pseudoand very high biological activities have been reported for many G ~ r - f l u o r o ~ and ~ - ~ ~6a-chlor0~~electrophiles. Attempts were made to introduce a cyano group into C-6 by treatment of either A3,5steroid hormones. or the 3-ethyl Among the C-6-nitro steroid hormones reported androstadien-3,17P-diol "S," both enol ether39 of Reichstein's compound r e ~ e n t l y ~the ~ . ~high ' oral progestational activity of 6~r-nitro-17a-acetoxyprogesterone~~ is worthy of with cyanogen and cyanogen iodide under many varied conditions. However, under no circumnote. stances could a 6-cyano compound be isolated. As a continuation of our studies relating to the Either starting material was recovered or the corsynthesis of steroid hormones with electronegative responding A4-3-ketone obtained. This approach substituents a t C-6,28~29,31--27 we now report the was abandoned in favor of a route via a 3a16asynthesis of a series of C-6 cyano steroid hor- epoxide. mones. The fission of unsymmetrical epoxides is compliPrevious publications from these laboratories cated by many factors,40but it has been observed have demonstrated that Bo-hydroxy,j 6P-nitro3' without exception that steroid 5a,6a-epoxides and BP-~hloro~~-A~-3-ketones are synthesized readily undergo nucleophilic attack a t C-6 ( P ) to afford by electrophilic attack a t C-6 of a A3,j-dienyl ace- the diaxial 6P-substituted sa-alcohol; examples tate or enol ether system (I+II). include fission of such an epoxide with hydrochloric acid,41 hydrofluoric acid, 30 boron trifluo(12) A. Butenandt and B. Riegel, Ber., 69, 1163 (1936). (13) M.Ehrenstein, J . Org. Chem., 4, 506 (1939). nitric acid,37 methylmagnesium (14) R. B. Moffett, J. E. Stafford, J. Linsk and \V. lf.Hoehn, THIS halide,16-z7343 lithium aluminum hydride44 and JOURNAL, 68, 1857 (1946). acetic acid.45 To the authors' knowledge the fission (15) J. A. Edwards, J. Iriarte, C. Djerassi and H. J. Ringold, forthof epoxides with nucleophilic cyanide ion has not coming publication from these laboratories. (16) G. B. Spero, J. L. Thompson, B. J. Magerlein, A. R. Hanze. been recorded, but it appeared, nevertheless, H.C. Murray. 0. K. Sebek and J. A . Hogg, THISJ O U R N A L , 78, G213 to represent a plausible approach to ti-cyano ster(1956). oids. (17) H . J. Ringold. E. Batres and G. Roseukranz, J . Org. Cheln., 22, ri suitable starting material for preliminary 99 (1957). (18) G.B. Spero, J. L. Thompson, F. H . Lincoln, W. P. Schneider studies was pregnenolone acetate-oc-epoxidez8(IIIa), and J. A. Hogg, THISJOUKNAI., 79, 1515 (1957). but prolonged heating of this epoxide in ethanol or (19) D. Burn, B . Ellis, V. Petrow, I . A. Stuart-\Vebb and D. h l . I-propanol solution with an excess of potassium Williamson, J . C h e m . Soc , 4092 (1957). cyanide led only to the hydrolysis of the acetate (20) If. Ackroyd, W. J . Adams, B. Ellis, V. Petrow and I. A. Stuart-Webb, i b i d . , 4099 (1957). group to afford pregnenolone-a-epoxide (IIIb). 4 6 , 4 7 (21) V. Grenville, D. K. Patel, Y , Petrow, I . A. Stuart-Webb and However, when the reaction was carried out in a D . M. Williamson, ibid.,4105 (1957). sealed tube a t 150' for 16 hours a product was ob(22) G. Cooley, B. Ellis, D. N. Kirk and V. Petron-, ibid.,4112 260 mg, E 20,890, the tained in 24y0 yield, "::A: (1957). (23) A. Bowers and H . J . Riugold, T m s JOURNAL, SO, 3091 (1958). infrared curve of which had strong absorption (24) J. C. Babcock, E. S. Gutsell, >I, E. Herr, J. A . Hogg, J. C. bands a t 2190 (>C=C-)48 and 1695 ern.-' (2029,31--34s42

Stucki, L. E . Barnes and TV. E . Dulin, ibid., 80,2904 (1958). (25) J. A. Campbell, J . C. Babcock and J. A . Hogg, ibid., 80, 4717 (1958). (26) R . Villotti, C . Dierasai and II J . Kingold, i b i d . , 81, LdWi

(1959). (27) H. J. Kingold, J . Perez Kuelas, JZ Uatres and C . I),iera.;i, ibid., 81, 3712 (1959). (28) A. Bowers and H. J. Ringold, Tetrahedrori, 3, 14 (19.36). (29) A. Bowers and H . J. Ringold, THISJ O U R N A L , 80, 4423 (1958). (30) J. A. Hogg, G . B . Spero, J. I,. Thompson, B. J. Magerlein, TT'. P. Schneider, D . H . Peterson, 0. K . Srbek, H . C . Murray, J. C . Babcock, K. L. Pederson and J. A. Campbell, Chenzisl;,y & I n d u s l r y , 1002 (1956). (31) A. Bowers, I, C. Ihdfier and H. J Ringold, Trlviiiicdvoii, i n Iiress

(1959). (32) A. Bowers, I: Deriot, I f . 1s. S d n c h e z and €1. J. Ringold i b i L , i n press (1959). (33) J. S . Mills, A. Bowers, C . C . Campillo, H. J. Ringold and C .

Djerassi, THISJOURNAL,81, 12ti4 (19.59). (34) J. A. Edwards, A. Zaffaroni, H. J. Ringold and C. Djeraiii, Proc. C h e m . Sac., Feb., 1959. (35) H . J. Ringold, 0. hlancera, C . Djerassi, A, Bowers, E. Batres. H. Martinez, E . Necoechea, J. A . Edwards, M. Velasco, C . C. Campillo and R. I. Dorfman, THISJ O U R N A L , 80, 6464 (1958). (36) A. Bowers, h l . €3. Sdnchez and H . J. Ringold, ibid., 81, 3702

(1959). (37) A. Bowers, L. C . Ibanez and H. J , Ringold, ibid., 81, 8707 11959).

CIK ( 3 8 ) I:. Westphal, S a f u r u ' i s s . , 24, 6Yti (19:trj). (89) P.I*. Julian, E. \V. Meyer, W. J , Karpel and K , C d e , r H l s JOI-RXAI., 73, 1982 (1951). (40) For a recent and critical review see E . L. Eliel ill "Steric Effects in Organic Chemistry," edited by hf. S. Newman, John Wiley and Sons, Inc., Pl'ew York, X. Y.,1956,pp. 106-114. anti references cited therein. (41) Z . Hattori, J . I'hiivnt. SOL.J a p a z . 60, 125 (1440); i'. .4., 34,

7294 (1940). (42) H . B. Henbest and T . I . Wrigley, J . Cheru. S a c . , $765 (1957). (43) M.I. Ushakov and 0. S. hladaeva, J Gel:. Chein. ( U . S S . R . ) , 9,436 (1939). (14) S. A. Julia. P. A . Plattner and H. Heusser, B e l v . Chirrz. . l c l u , 3 5 , 663 (1952). (45) 2. Hattori J . Pharnz. .Sor. J a p a ~ i ,69, 411 (1R31l), (-. R . , 33, 6622 (19391. (46) T h i s compound h a s not been adequately described in t h c literature, b u t i t s structure was proved readily when i t smoothly a f forded 5ar,6a-oxidopregnan-3~-ol-2O-one-3-acetate~~ upon acetylation. (47) T h e use of potassium cyanide f o r t h e hydrolysis of steroid esters has been reported previously; cf. Q. R . Peterson, THISJ O U R N A I . , 77, 1743 (1955). (48) L. J. Bellamy, "The Infrared Spectra of Complex .\Iolecules," John Wiley and Sons, Inc., New Pork, h'. Y . , 195t3, p p , 223-

226.

Oct. 5, 1959

C-6-CYANO STEROID

CHs I

P

IIIa. R = R ' = COCH, b. R = H ; R'=COCH, C, R-COCH,; R'=OCOCH, d. R = H ; R ' = O H OH

+ VIIC

5235

HORMONES

Iv

P'

IVa, R = H; R' = COCH3 b. R = R ' = C O C H j c R = H ; R'=OH d R = COCH3; R ' = OCOCH,

d3 NdBH,

R

COCH, H b R==O

X

IXa R=
Cproduct was a mixture of Ga-cyanoprogesterone and OH), 2200 (>C=C-)48 and 1695 cni.-' (20- its enol form XIIa, the latter compound predominatI ing. I n the presence of a drop of 5% potassium CN hydroxide solution there was a shift of maximum ketone) On the basis of these spectral data and absorption to 338-340 mp, t 18,700, consistent the elemental analytical results i t was formulated as 6-cyano-As pregnen-3P-ol-20-one (V) .49 The second product contained one more additional oxygen atom (elemental analysis) and did not -0 CN exhibit selective absorption in the ultraviolet but

@

had a band a t 2220 cm.-' (>&-CNT)i8 in the infrared. It formed a monoacetate after treatment with acetic anhydride and pyridine which still showed strong hydroxyl band absorption in the infrared. These data were clearly consistent with this product being either 6a- or Gp-cyanopregnan3P,5a-diol-20-oneso (VIa) or (VIIa). A further modification was to carry out the fission of the epoxide I I I a by heating in ethylene glycol with potassium cyanide for 3.5 hours a t 90-95' when three products were obtained. The least polar toward alumina was 5a,6a-oxidopregnan-3/3-01-2O-one (IIIb) and the second product was the same 6-cyano-3/3,5a-diol (VIa) or (VIIa) which was isolated from the 140" experiment. The third product was a new compound which did not exhibit selective absorption in the ultraviolet, but showed a band in the infrared a t 2225 cm.-l (-CSN)~* and formed a monoacetate under mild acetylation conditions which still displayed a strong hydroxyl band in the infrared. I t was concluded therefore that these two compounds differed only in the stereochemistry of the nitrile group a t C-6 and they were formulated as VIa and VIIa, respectively. Chemical evidence supporting this conclusion was forthcoming. Oxidation of both VIa and VIIa with 8 N chromic acid in acetone solution51 led to the corresponding 6a- and 6Pcyano-5a-hydroxy-3,20-diones IXa and VIIIa. It is well known that C6-substituted 5a-hydroxy-3ketones smoothly dehydrate under very mild alkaline23-2i or acid c o n d i t i o n ~ ~ ato -~~ afford the corresponding 6a-substituted-A4-3-ketones. However treatment of both VIIIa and IXa with 0.270 methanolic potassium hydroxide for 18 hours at CHZOH I

(49) R. E. Jones and F. W.Kocker, T H I S JOURNAL, 76,

Xmax 223 mM, nitrile.

c

3682 (1954), report 13,680, for t h e vinyl

C-CN

@

0

(50) T h e stereochemistry of t h e nitrile group is discussed in t h e sequel and shown t o be Ba. (51) Cf. (a) K. Bowden, I. M. Heilbron, E. R . H. Jones and B C 1,. Weedon, J . Chem. SOC.,39 (1946); (h) A. Bowers, T. G . Halsall. E . R. H. Jones and A. J. Lemin, ibid., 2548 (1953).

A

Similarly treatiilent of with the enolate ion VIIIa and IXa with anhydrous hydrogen chloride in acetic acid followed b y precipitation with water afforded the same ketoenol mixture X I I a which was obtained by the alkaline dehydration; Amax 230 and 288-290 mp, E 5130 and 10,250, respectively, changed by a trace of alkali to 338-340 mp, e 18,200. A satisfactory elemental analysis could not be obtained for this product and attempts to obtain a crystalline solid were abortive, but acetylation of both of the keto-enol mixtures derived from the ketones VIIIa and I X a with acetic anhydride and acetyl chloride led in each case to the same enol264 mp, E 18,200. h4ild alkaline acetate XIa, Lax hydrolysis of this enol-acetate X I a afforded the solid, non-crystalline keto-enol mixture of 6-cyanoprogesterone (XII) identical to that obtained previously, as evidenced by ultraviolet and infrared spectroscopy. Thus, the two isomeric ketones VIIIa and IXa clearly differed only in the stereochemistry of the nitrile group a t C-G. The initial product from the diaxial epoxide cleavage would be the trans-5a-hydroxy-GP-nitrile VIIa which under the basic conditions of the reaction partly epimerized a t C-G to afford some of the Ga (equatorial) nitrile VIa. It was possible to deduce the stereochemistry a t C-6 of the two products from a consideration of their molecular rotations. I t is well known that 6a-substituted A4-3-ketones are more dextrorotatory than their GP-epi~ners:~and i t can be seen that this effect a t C-6 is of a geueral nature and is not restricted only to C-6 substituted (-vD A4-3-ketones since ~holestan-3P,3a,6a-triol~~ +88") is more dext.rorotatory than cholestan-3P,( M D +13"). 5a,Gp-tri01~~ (52) Cf. R. H. Lenhard and S. Bernstein, THISJ O U R N A L , 78, 989 (19561, report Aman

320-321 and 391-302 m p , e 23,600 and 3,800, respectively, for ( h e system B which changes t o Ho Xmax 390-391 mp, E 74,400, in ethanol conB taining 1% of potassium hydroxide. (53) For a summary of t h e molecular rotation d a t a from seven different pairs of epimeric (2.6 substituted A4-3-ketones cf. ref. 33. (54) V. Prelog and E. Tagmann, Helu. Chim. A d a , 27, 1867 (1944), report t h e preparation of this compound b u t its rotation has not been recorded. Consequmtly i t was prepared by t h e osmium tetroxide oxidation of cholesterol and its constants are reported in t h e Experimental section. (55) L. Ruzicka and V. Prelog, i b i d . , 26, 975 (1943).

Oct. 5, 1959

C-6-CYANO STEROIDHORMONES

5237

The two epimeric 6-cyanopregnan-3/3,5a-diol-hydrolysis of either XVa or XVIa led to the same 20-ones VIa and VIIa had molecular rotations of non-crystalline mixture of 6a-cyanoprogesterone +258 and +lo% Accordingly the more dextro- in equilibrium with its enol form X I I a as was obrotatory compound was assigned the 6a-cyano tained from VIIIa and I X a as described above. Thus, two different approaches to 6-cyano-A4-3configuration (VIa) and its epimer VIIa the 6Bketones had been established and application of cyano structure. Convincing chemical proof that these stereo- these methods led to the synthesis of the 6-cyano testosterone, chemical assignments were correct was forthcoming analogs of A4-androstene-3,17-dione, when treatment of GP-cyanopregnan-3P,5a-diol- 17a-acetoxyprogesterone and cortisone. Potassium cyanide cleavage of 5a,6a-oxidoan20-one (VIIa) (axial cyano group) with potassium cyanide in ethylene glycol a t 95" for 3 hours led drostan-3P,17@-dioldiacetate60 (IIIc) for 3.5 hours in to the formation of Gc~-cyanopregnan-3/3,5a-diol- ethylene glycol a t 90-95' led to the isolation of 20-one (VIa) (equatorial cyano group) in 25% three products which were separated by crystalliyield. Also, in one instance the oxidation of VIIa zation and chromatography over alumina. They 17P-diol (IIId), 6aled not to 6p - cyanopregnan - 501 - 01 - 3,20 - dione were 5a,6a-oxidoandrostan-3P, cyano-androstan-3/?,5ar,17B-triol (VIc) and its 6P(VIIIa) but to its 6a-epimer IXa. An alternate approach to 6-cyano-A4-3-ketones epimer VIIc. By analogy with the structures then was developed from 3-cycloethylene-ketal- assigned to the corresponding pregnane derivatives 5a,Ba-epoxides. Progesterone-3,20-biscycloethyl- VIa and VIIa the most dextrorotatory of the two ene-keta1-5a,6a-epo~ide~~ (XIIIa), for example, compounds ( M D f 0 compared to MD - 99) was was found to be superior to pregnenolone acetate- assigned the 6a-configuration. Attempts to chemia-epoxide (IIIa) as a substrate for conversion into cally interrelate these two epimeric nitriles by acid 6-cyanoprogesterone (XIIa) since even when a treatment were unsuccessful. For example, treatdouble elimination occurred a t C-3 and C-5 the ment of both VIc and VIIc with anhydrous product was 6-cyano-3- (2 '-hydroxyethyl) - A r5-preg- hydrogen chloride in acetic acid led smoothly to the nadien-20-cycloethylene-ketal (XVIa), clearly a corresponding diacetates VId and VIId. No insuitable compound for conversion into 6-cyano- version of the nitrile group in VIId could be deprogesterone (XIIa). In a manner similar to the tected. Upon oxidation with 8 N chromic acid in acetone potassium cyanide fission of IIIa, the products obtained from X I I I a varied with the length of the solution, VIc and VIIcS1 afforded the correspondreaction time. When X I I I a was heated under ing 3J7-diketones I X b and VIIIb, respectively. reflux with potassium cyanide in ethylene glycol Sodium borohydride reduction of VIIIb led to a mixture of VIIc (52%) and the 3a-epimer X (28%). for 35 minutes the product was 6,-cyanopregnan5a-ol-3,20-biscycloethylene-ketal(XIV) , It did The unusually high yield of a 3a-alcohol from the not exhibit selective absorption in the ultraviolet borohydride reduction of a C-3-ketone was attriband displayed bands a t 3450 (-OH) and 2230 uted to the increase in steric hindrance caused by cm.-' (-CSN).~~ The stereochemistry of the the 5a-hydroxyl group to the approach of the boronitrile group was not proved but the 6a-(equatorial) - hydride ion from the a-face with a consequent deconfiguration was assigned on the basis of molecular crease in the amount of the 3/3-alcohol formed. rotation data.57 When the reaction time was Both of the ketones VIIIb and I X b afforded a extended to 3.5 hours two different products were non-crystalline mixture of 6-cyano-A4-androsteneobtained. The less polar of the two toward 3,17-dione and its enol form X I I b on acid treatment. alumina exhibited spectral data appropriate for An extension to the androstane series of the a vinyl nitrile,48r49 namely, maximum absorption method used to synthesize 6a-cyanoprogesterone in the ultraviolet a t 222 mp, e 10,000, and a strong from progesterone-bisketal-a-epoxide (XIIIa) led band in the infrared a t 2200 cm.-l. Analytical to the synthesis of 6a-cyanotestosterone (XIIc). data substantiated the assigned structure as be- Testosterone acetate 3-cycloethylene-ketal-5a,Gcring G-cyano-A5-pregnen-3,20-biscycloethylene-epoxide (IIIb) after treatment with potassium ketal (XVa). The second product was isomeric cyanide in ethylene glycol under reflux for 4 hours with XVa and had maximum absorption in the afforded a mixture of 6-cyano-A5-androsten-17Bultraviolet a t 282 mp, E 20,000. In the infrared ol-3-one-3-cycloethylene-ketal (XVb), Xmax 224 it had bands at 3350 (-OH), 2200 (nitrile group rnp, E 13,000, and the 6-cyano-enol ether (XVIb), attached to an unsaturated carbon atom) ,48 1620 Amax 284 mp, E 18,750. Acid hydrolysis of XVIb and 1590 cm.-l (A3,j-diene-enol ether system).j5 with hydrochloric acid in acetic acid led to a nonThese data could only be accommodated by crystalline mixture of 6a-cyanotestosterone acetate structure XVIa. A double elimination took place and its enol form XIIc, acetylation of which with a t C-3 and C-5 analogous to the formation of IV acetic anhydride and acetyl chloride afforded a from IIIa.b9 Perchloric acid in tetrahydrofuran homogeneous enol acetate (XIc) . I n a like manner 6a-cyano-170-acetoxyprogester(56) G.Cooley, B. E. Ellis, D . N. Kirk and V. Petrow, J . Chem. SOC., 4112 (1957). one (XIIe) was prepared. Treatment of 17a(57) M D 5a,6a-epoxide IIIb - MD 5a-ol-6&nitrile VIIa = -78; hydroxyprogesterone- 3,20 - biscycloethylene - ketalM D 5a,6a-epoxide IIIb - MD 5a-ol-6a-nitrile VIa = -228; M D ja,Ba-epoxide (XIIIc) with potassium cyanide in XI11 - M D X I V = -209; more compatible with a 6a-configuration the usual way led to a mixture of 6-cyano-d5of the nitrile group than 60 for compound X I V . (58) Strong doublets in this region have been observed in these pregnen-17a-ol-3,20-biscycloethylene-ketal(XVc) laboratories for many AW-diene-3-ethyl-enol ethers. and the A3r5-3-hydroxyethyl enol ether XVIc. (59) The fission of a AL3-cycloethylene ketal t o afford a 3-(2',Acid hydrolysis of either XVc or XVIc followed by hydroxy-ethyl)-A'ts-dienyl ether has been reported previously; cf. ref. 52.

(60) L. Ruzicka and A. C. Muhr, Helu. Chim. A d a , 27, 503 (1944).

5238

A. BOWERS,E. DENOT,M. B. SANCHEZ, L. 1cI. SANCHEZ-HIDALGO AND H. J. RINGOLD Vol. 81



b, R = C ( CHZOH

CH,OH~

0-

OJ [--OH

I

iI

acetylation with acetic anhydride and acetyl cortisone-bisketa1-5a,Ga-epo~ide~~ (XIXb) . Cleavchloride afforded 6-~yano-A~>~-pregnadien-3,17aage of the latter epoxide with potassium cyanide diol-20-one diacetate (XId), whence mild alkaline led to both the vinyl cyanide XXb and the Ghydrolysis afforded a mixture of Ga-cyano-lia- cyano-enol ether XXIb. Hydrolysis of XXIb with acetoxyprogesterone and its enol form XIIe. perchloric acid in tetrahydrofuran gave a mixture Thus the synthesis of a 6-cyano analog of an of Ga-cyanocortisone and its enol from X X I I androgenic, a progestational and an oral progesta- characterized by the double maximum in the tional hormone had been accomplished. However, ultraviolet, Xmax 220-222 and 286-288 mp, E to complete our preliminary studies of the biological 42iO and 10,750. effect of a C-6 cyano group it was deemed necessary Experimental6 to synthesize a 6-cyano cortical hormone. Treatment of A3~5-Androstadiene-3,I’IP-diol Diacetate I n view of the instability of the cortical side with Cyanogen and Cyanogen Iodide.-A solution of chain to potassium cyanide6I i t was necessary to A3~5-androstadiene-3,1i~-diol diacetate38 (500 mg.) in dry suitably protect this grouping and a convenient ether a t 0 ” was saturated with cyanogen and then kept at 0” starting material appeared to be the bismethylene for 24 hours. Addition of water and isolation with ether dioxy derivative62 of cortisone (XVII; Fig. 2 ) . afforded unchanged enol acetate. This experiment was repeated using the following solvents instead of ether: diThis compound readily formed the corresponding methylformamide, nitromethane and chloroform containing As-3-cycloethylene-ketalXVIII, whence epoxida- 2 7 , pyridine; in every case only starting material could be tion with permonophthalic acid afforded the cor- recovered. Similarlb-, abortive results were obtained when iodide was substituted for cyanogen. responding 5a,6a-epoxide XIX. Following the cyanogen Treatment of the 3-Ethyl Enol Ether of Reichstein’s Comusual procedure, cleavage of X I X with potassium pound “S” Acetate with Cyanogen or Cyanogen Iodide.cyanide in ethylene glycol led to a mixture of the ?he experiments described above with cy&iog& and cyanovinyl nitrile XXa and the B-~yano-A~~~-enol ether gen iodide were repeated using t h e ?,-ethyl enol ether of XXIa. However, all attempts to hydrolyze either Reichstein’s compound “S” a ~ e t s t e . 3 I~n every case only starting material or Rcictistein’s compound “S” acetntr X X or X X I with aqueous formic acid, acetic acid, could be isolated. hydrogen chloride in acetic acid (with or without Treatment of 5a,6a-Oxidopregnane-3~-ol-ZO-one Acetate water being present) failed to afford 6-cyanocorti- (IIIa) with Potassium Cyanide .-(a) Potassium cyanidc (2.0 g.) was added t o a solution of the epoxide IIIaZS(1.0 sone (XXII). It appeared that the 6-cyano-A4- g.) in ethanol (50 cc.) a n d heated i i r i d ~ reflux r for 2-1 houri. 3-ketone system was destroyed by the rather (63) J. A . Edwards and H. .I.Kingold, forthcoming Iniblicatioti vigorous acid conditions required to regenerate the cortical sidechain from the bisrnethylene dioxy from these laboratories. (04) Melting points are uncorrected. Rotations nere measured i l l protective grouping. ,4n alternate, successful syn- chloroform solution unless stated otherwise and the ultraviolet a b thesis of 6-cyanocortisone then was carried out from sorption spectra in 9.5% ethanol solution. \Te are grateful t o Dr. I,. J (61) ~‘-Pregnen-17cr.21-diol-20-one-3-cycloethylene-ketal (Com-

pound “S” monoketal) could not be recovered after treatment with potassium cyanide in ethylene glycol under reflux for 1 hour. (62) R. E. Beyler, R. M . Moriarty, F. Hoffman and 1, H. Sarett, T H I S J O U R N A L , 80, l j l i (1958).

Throop and his staff for these measurements and for t h e infrared spect r a which were obtained with a Perkin-Elmer model-21 spectrophotometer with a sodium chloride ptism. T h e alumina used in this work had previously been suspended in boiling ethyl acetate for 6 hours, filtered and dried at 100’. T h e elemental analysis irere carried out by Dr. .2.Rernhardt, Mulheim, Ruhr, Germany.

Oct. 5, 1959

C-6-CYANO STEROID

HORMONES

5239

on admixture with the same Zroduct obtained from the Addition of water and extraction with ethyl acetate afforded previous experiments, ID +74 . 5a,6a-oxidopregnane-3fi-ol-20-one ( I I I b ) (910 mg.), m.p. Further elution with ether-acetone (90: 10, 1,500 cc.) af177-184', raised by one crystallization from acetone-hexane ~ raised by one crystallization from forded 6P-cyanopregnane-3P,5a-diol-20-one(VIIa)66 (1.19 t o 188-190", [ a ] +17', g . ) , m.p. 206-212" raised by crystallizations from ethyl aceacetone-hexane to 188-290", [a] +17", lit.65 reports m.p. ~ The m.p. was strongly detate t o 223-225', [ a ]+30°. 185-187 O . 284-286 mp, e Anal. Calcd. for C21H3203: C, 75.86; H , 9.70. Found: pressed on admixture with VIa, : : :A 3500, 2240 and 1690 cm.-l. C, 75.79; H , 9.82. Anal. Calcd. for Cz2HssOaN: C, 73.50; H , 9.25; 0, Acetylation of I I I b (acetic anhydride-pyridine) readily 13.35; N, 3.90. Found: C, 73.38; H , 9.09; 0, 13.44; afforded I I I a identical with an authentic sample.28 Similar treatment of I I I a with potassium cyanide in 1- N, 4.02. propanol led only to I I I b . Acetylation of VIIa with a n excess of acetic anhydride in Dvridine solution at room temmrature for 16 hours afforded (b) Potassium cyanide (2.0 g.) was added t o a solution of the epoxide I I I a (1.0 9.) in ethanol (50 cc.) and heated in a 6&cyanopregnane-3~,5a-diol-iO-one3-acetate (VIIb), n1.p. ~ ; : :A 3400, 2200, 1720, 1685 and 1245 sealed tube a t 150 G t IO' for 16 hours. Addition of water 225-227', [ a ]+'io; and extraction with ethyl acetate afforded a product which cm.-l. was adsorbed from benzene-hexane (80:20, 100 cc.) onto Anal. Calcd. for C2&504NL C, 71.79; H , 8.79; 0, alumina (40 g.). Elution with benzene-hexane (80:20, 15.94. Found: C , 71.69; H , 8 . t 3 ; 0, 16.26. (I\') (240 250 cc.) afforded 6-~yano--A~,~pregnadiene-2O-one Cholestane-3~,5a,6a-triol.-Osmiumtetroxide (280 mg.) mg.), m.p. 119-125", raised by several crystallizations from was added to a solution of cholesterol (350 mg.) in pyridineacetone-hexane to 138-142", [ a ] n - l o l a , AgzH260 mp, E chloroform ( l : l , 20 cc.). After 5 days a t room tempera20,890; : :A: 2190, 1695, 1620 and 1585 cm.-l. ture the solvent was removed zn z'acuo and the residue was Anal. Calcd. for C22HpgOK: C, 81.69; II, 9.04; 0, treated with a suspension of lithium aluminum hydride 4.93; N, 4.33. Found: C , 81.33; H , 9.16; 0, 5.42; N, (1.0 9.) in tetrahydrofuran (50 cc.) under reflux for 18 hours. 4.38. After decomposition of the excess of reagent with ethyl ace(c) Potassium cyanide (4.0 9.) was added t o solution of tate, dilute hydrochloric acid was added. Isolation with the epoxide 111%(2.0 9.) in ethylene glycol (120 cc.) and ethyl acetate then afforded a product which was adsorbed heated under reflux for 1 hour. Addition of water and iso- fiom benzene onto alumina (20 g , ) . Elution with benzenelation with ethyl acetate gave a product which was adsorbed ether (70:30.500cc.) afforded cholestane-30.5a.6a-triolf210 from benzene (50 cc.) onto alumina (80 g.). Elution with mg.), m.p. 227-232 raised by crystallizations from aceton'e to 232-234", [ a ] D $21'; lit.64reports m.p. 236-238'. benzene (900 cc.) and one crystallization of the product from acetone-hexane afforded 6-cyano-A3z5-pregnadietie-20Epimerization of 6p-Cyanopregnane-3P,50-diol-ZO-one one ( I V ) (970 me.), m.p. 133-136", 260--262 mfi, E (VIIa).-Potassium cyanide (400 mg.) was added to a solu19,050, undepressed on admixture with the product isolated (VIIa) in ethyltion of 6P-cyanopregnane-3fi,5a-diol-2O-one in the preceding experiment (b). ene glycol (10 cc.) and heated a t apmosimately 90" (steam(d) Potassium cyanide (2 g.) was added to a solution of bath) for 4 hours. Addition of water and isolation with the epoxide (1.0 g.) in ethylene glycol (60 cc.) and heated a t ethyl acetate afforded a product which was adsorbed from 140" for 1 hour. -4fter cooling, addition of water and isola- benzene onto alumina 20 g. Elution with benzene-ether tion with ethyl acetate gave a product which was adsorbed (70:30, 250 cc.) afforded 6a-cyanopregnane-3fi,5a-diol-20from benzene (100 cc.) onto alumina (40 g.). Elution with one (VIa), 51 mg., m.p. 251-256', raised by one crystallizabenzene-ether (90: 10, 400 cc.) afforded a semi-crystalline tion t o 256-258', undepressed on admixture with an au" : :A 222 m r , E 10,230, which after thentic sample; l a ] +73". fraction (220 mg.), ~ further chromatography over alumina afforded 6-cyano-As(VIa). Oxidation of 6a-Cyanopregnane-3fi,5~-d~ol-ZO-one pregnene-3fi-ol-20-one (I-)(85 mg.), m.p. 132-137", raised -The diol VIa (500 mg.), m.D. 256-269 , in acetone (15 by several crystallizations from aqueous methanol t o 149cc.) was treated a t 0' t&h an excess of 8 A' chromic acid in ~ A%:H 220-222 miLC, E 10,000;; :A: 2200,1695 the usual way.6l Addition of water and filtration afforded l52O, [ a l -23", and 1625 cm. -1. 6a-cyanopregnane-5a-ol-3,20-dione ( I X a ) (420 mg.), m.p. Anal. Calcd. for C22H3102N: C, 77.37; H , 9.15; X, 238-245', raised by crystallizations from acetone-hexane to 242-244', [a]D +51'; 278-288 I n U , E 53; A":, 4.10. Found: C, 77.62; H , 9.23; S,3.98. 3320,2230, 1720 and 1685 cm.-l. Further elution with ether (,500 cc.) gave 6a-cyanopreg Anal. Calcd. for C Z ~ H ~ ~ OC,~ N 73.91; : H, 8.74; 0, nane-3/3,5a-diol-20-one (VIa) (160 rng.), m.p. 220-230 , 13.43; N,3.92. Found: C, 73.84; H,8.54; 0, 13.58; N, raised by cry;tallizations from ethyl acetate t o 258-259', [ a ]+72'; ~ " : :A 240-246 and 280-288 m p , E 21 and 34, 3.82. 3400, 2220 and 1690 crn.-l. respectively; ; :A: Oxidation of 6P-Cyanopregnane-3fi,5a-dioi-20-one (VIIa). Anal. Calcd. for Cz~Ha303X: C, 73.50; H , 9.25; 0, -The diol VIIa (200 mg.), n1.p. 222-224', in acetone (20 13.35; 5 , 3 . 9 0 . Found: C,73.46; H,9.01; 0, 13.51; X, cc.) was treated a t 0" with an excess of 8 iV chromic acid in the usual way.61 Addition of water and filtration afforded 4.12. Acetylation of VIa with a n excess of acetic anhydride in 6fi-cyanopregnane-5a-ol-3,2O-dione(VIIIa) (165 mg.), pyridine solution a t room temperature for 18 hours smoothly m.p. 243-246", raised by one crystallization from acetone6a-cyanopregnane-3~,5a-diol-20-one3-acetate hexane t o 246-248", depressed (220-242') on admixture afforded ~ ;:A: %OO, 1730 and 1690 an.-'. ~ ::A: 3300, 2205, with I l i a ; [ a ]4-35'; (VIb), 1n.p. 245-247', [ a ] +47'; Anal. Calcd. for C22H31O3X: C, 73.91; H, 8.74; 0, 1720(sh), 1700 and 1260cm.-l. Anal. Calcd. for C, 71.79; H , 8.79; S, 13.43; r\',3.92. Found: C , 73.75; H, 8.71; S , 3 . 9 9 . I n another experiment apparently identical t o the one 3.49; 0, 15.93. Found: C, 71.67; H , 8.65; S,3.19; described above the product was 6a-cyanopregnane-5a-010, 16.43. 3,20-dione ( I X a ) (86% yield), m.p. 239-243", raised by one (e) Potassium cyanide (20 9.) was added to a solution of crystallization from methanol-chloroform to 243-245', pregnenolone acetate epoxide ( I I I a ) in ethylene glycol (600 [ a ]+49', ~ undepressed on admixture with authentic IXa. cc.) and heated a t 90" (steam-bnth) for 3.5 hours. AddiThe infrared spectra of the two compounds were identical. tion of water and extraction with ethyl acetate afforded a 6-Cyanoprogesterone (XIIa).-(a) Potassium hydroxide product which was adsorbed from benzeie (300 cc.) onto (100 mg.) in methanol (10 cc.) was added to a solution of 6aalumina (500 g.). Elution with benzene-ether (50: 50, cyanopregnan-5a-ol-3,20-dione ( I X a ) (400 mg.) in methanol 1,500 cc.) gave a fraction which after one crystallization from (66) The 60 (axial) nitrile VIIa might reasonably have been exacetone-hexane afforded pregnenolone-5~u,6a-epoxide ( I I I b ) , 1.3 g., n1.p. 186-188" undepressed on admixture with the pected to have been eluted before the 6a (equatorial) nitrile VIa product obtained in experiment (a) above. Further Possibly hydrogen bonding between the 6u-cyano group and the 5aelution with ether (1,300 cc.) gave 6a-cyanopregnane-3@,5a- hydroxyl group in V l a reduces the polarity of this compound toward diol-20-one (VIa) (1.95 g.), m.p. 252-255' raised by one alumina. However two recent publications discuss examples at C-6 and C-7 where the axially substituted compounds are more polar than crystallization from ethyl acetate t o 257-259" undepressed (65)

~

Y. Urushibara, M. Chuman and S. Wada, Bull. Chem. SOC. ~ a4, 83p (1951); ~

c. A~ . , 47,59561 ,

(1953).

the equatorial isomers; cf. 0. Wintersteiner and M . Moore, THIS (1958),and W. J. McAleer, M. A. Kozlowski, T. H. Stoudt and J. M. Chemerda, J . Org. Chem., as, 958 (1958). JOURNAL, 81, 422

5240

A. BOWERS,E. DENOT,ILL B. SANCHEZ, L. hl. SANCHEZ-HIDALGO AND H. J. RINGOLD Vol. 81

(40 cc.). After 18 hours a t room temperature the solution cc.) was added and the product was extracted with ethyl was diluted with water, neutralized with acetic acid and acetate. The combined extracts were washed with sodium then extracted with ethyl acetate. After washing with wa- bicarbonate solution (570), water and then dried over soter, the dried solution (NazSOa) was evaporated t o afford a dium sulfate. Removal of the solvent afforded 6-cyanopronon-crystalline mixture of 6a-cyanoprogesterone and its enol gesterone (XIIa), Amax 228-230 and 288-299 mp, E 4,250 and XEtOH 228-230 and form X I I a , m.p. 91-98', [ a ] -88"; ~ 10,075, respectively. This product was dissolved in acetic 288-290 mu, E 4,300 and 11,300, respectively; XE'oH'KoH' anhydride (10 cc.) and acetyl chloride (10 cc.) and heated 338-340 mp, e 18,700; 3500, 2220, 1700, 16%; 1640 under reflux in a nitrogen atmosphere for 2 hours. Removal and 1580 ern.-'. Similar alkaline treatment of 6P-cyano- of the solvent a t 95" in in vacuo and crystallization of the pregnan-5a-ol-3,20-dione(VIIIa) led to the same product; residue from methanol afforded 6-cyano-A3P-~regnadiene-3X% iH : 228-230 and 288-290 mu, E 5,050 and 10,300. The 01-20-one acetate (XIa) (285 mg.), m.p. 14:-151°, raised by infrared spectra were essentially identical. crystallizations from methanol to 168-1'70 , undepressed on ( b ) Dry hydrogen chloride x a s bubbled through a solu- admixture with a n authentic sample;" : :A 264 m,u, e, ( I X a ) (50 mg.) in tion of 6cu-cyanopregnan-5a-ol-3,20-dione 18,500. Potassium Cyanide Cleavage of 5~~,6a-Oxidoandrostaneacetic acid (5.0 cc.) for 10 minutes a t 15" and then the flask was tightly stoppered. After a further 2.5 hours a t 15' 3/3,17p-diol Diacetate (IIIc).-Potassium cyanide (14 g.) was addition of water and filtration afforded X I I a (41 mg.), m.p. added to a solution of 5a,6a-oxidoandrostane-3~,li~-diol 230 and 288-290 mp, E 5130 and 10,250, re- diacetate (TIIc) (7.2 g.) in ethylene glycol (420 cc.) and 97-112'; "::A: heated on the steam-bath for 3.5 hours. Addition of water 338-340 mu, E 18,200. The infrared spectively; XE::E(KoH' spectrum was essentially identical with the product obtained and isolation with ethyl acetate afforded a product which was heated, under reflux for 10 minutes with benzene (400 from the two previous experiments. Hydrogen chloride in acetic acid treatment of VI11 also cc.) (a portion did not dissolve). After cooling t o room temled t o X I I a , m.p. 103-120", 230 and 288-290 mu, e perature filtration afforded 6~-cyanoandrostane-3p,ja,l7!3triol (VIIc) (2.15 g.), m.p. 299-303', raised by crystalliza4,850 and 9,970, respectively. ~ (ditioxis from aqueous methanol t o 309-311', [ a ] -31" 6-Cyan0-A~,~-pregnadiene-3-01-20-0ne Acetate (XIa).--A 3350, 3450 and 2220 cm.-'; VIIc did not solution of 6-cyanoprogesterone ( X I I a ) (400 nig.) (prepared oxane); as in method a above) in acetic anhydride-acetyl chloride exhibit selective absorption in the ultraviolet. (1:1, 15 cc.) was heated under reflux for 2 hours in a nitroAnal. Calcd. for Cz&~Oax: C, 72.03; H, 9.37; N, 4.20. gen atmosphere. The solvent then was removed a t 95" in Found: C, 71.65; H , 9.22; N,3.92. vacuo and the residue crystallized from methanol t o afford 6The filtrate from the benzene treatment of the product ~yano-A~~~-pre~nadiene-3-01-20-one acetate ( X I a ) (225 mg.), then was adsorbed onto alumina (200 g.). Elution with m.p. 147-160 , raised by crystallizations from methanol t o benzene-ether (50: 50, 2 1.) afforded 5a,6a-oxidoandrostane168-170", [ a ] D -35", 264 mu, e 18,200; 2200, 3P,17P-diol ( I I I d ) (1.31 g.), m.p. 190-195", raised by c;zx1770, 1700 and 1670 cm.-l. tallizations from ethyl acetate to 203-204", [ a ] D - i o Anal. Calcd. for Cz4Ha10&: C , '75.56; H, 8.19; 0, This product was identical in every respect to the product ob12.58. N, 3.67. Found: C, 75.27; H, 8.36; 0, 12.54; tained by an alkaline hydrolysis (3% KOH-MeOH, 45 min. s,3.75. reflux) of IIIc. Similarly enol acetylation of 6-cyanoprogesterone ( X I I a ) Anal. Calcd. for Ci9H3003: C , 74.47; H, 9.87; C, 15.66. derived from experiments b, c and d above led t o the same Found: C, 74.32; H,9.96; 0,15.25. enol acetate XIa. Further elution with ether-acetone (90: 10, 1 1.) afforded Potassium Cyanide Cleavage of 5~~,6a-Oxidopregnane- 6~-cyanoandrostane-3j3,5~~,li~-triol (VIc) (760 mg.), n1.p. 3,20-bis-cycloethylene-ketal(XIIIa) .-(a) Potassium cyan- 228-250°, raised by one crystallization from methanol t o ide (6.0 g.) was added t o a solution of 5~,6~-oxidopregnane- 246-250'. The analytical sample from methanol had m.p. 3,20-biscycloethylene keta16e(XII)(3.0 g.) in ethylene glycol 254-256', [ a ] D 4-8"; 3350and2220 ern.-'; VIcdid not (100 cc.) and heated under reflux for 35 min. After imme- exhibit selective absorption in the ultraviolet. diate cooling, water was added and the product isolated with Anal. Calcd. for C Z ~ H ~ I O &C: , 72.03; H , 9.37; K, ethyl acetate and then adsorbed from benzene onto alumina 4.20. Found: C, 71.68; H, 9.40; N,3.90. (150 g.). Elution with benzene (3.5 1.) afforded 6-cyano6p-Cyanoandrostane-3p,50r, 17p-triol 3,17-Diacetar e pregnane-5~u-ol-3,20-biscycloethylene-ketal ( X I V ) (2.17 g.), m.p. 218-221", raised by several crystallizations from ace- (VIId).-Acetic anhydride (1 cc.) was added t o a solution of (VIIc) (200 mg.) in tone-hexane t o 227-228", [CY]D 10'; 222-228 m&, 6p-cyanoandrostane-3p,~a,l'i~-triol pyridine (10 cc.) and heated a t 95" for 90 min. Addition of E , 99; ; : A : 3450 and 2230 em.-'. ice-water and isolation with ethyl acetate afforded the diaceAnal. Calcd. for C26H3906S: C, 70.08; H , 8.82; N, tate VIId, m.p. 195-197' raised by crystallizations from 3.14. Found: C,70.15; H,8.68; h-,3.27. 3500, 2250, acetone-hexane t o 202-204", [CY]D -68"; 1740 and 1715 em.-'. ( b ) Potassium cyanide (2.0 9.) was added to a solution of the bis-ketal epoxide X I I I a (1.0 g . ) in ethylene glycol (40 Anal. Calcd. for CZ~HX,O~N: C, 69.03; H, 8.45; 0, cc.) and heated under reflux for 1.25 hours. Water was 10.16. Found: C, 69.13; H, 8.31; 0, 19.64. added to the cooled solution and the product isolated with Treatment of 6/3-Cyanoandrostane-3p,5a,17P-triol (VIIc) ethyl acetate and then adsorbed from benzene-hexane (50 : with Hydrogen Chloride in Acetic Acid.-A solution of 6P50, 100 cc.) onto alumina (40 g.). Elution with benzenecyanoandrosta1ie-3~,5a,17~-triol (VIIc) (200 mg.) in acetic hexane (50:50, 300 cc.) and benzene (100 cc.) afforded 6- acid (15 cc.) a t room temperature was saturated wlth dry cyano-As pregnene-3,20-biscycloethylexie-ketal (XVa) (200 hydrogen chloride a n d then kept a t room temperature for 3 mg.), m.p. 175-181", raised by several crystallizations from hours. Addition of water and filtration afforded 6Pcpaiio224 mu, E androstane-3P ,5a,1'TP-triol 3,17-diacetate ( V I I d ) , ni .p. acetone-hexane t o 182-184', [a]n -74", 9,590; X",: 2,200 and 1,640 cm.-l. 189-194', raised by one crystallization from acetone-hexane Anal. Calcd. for C2sH3,0cN: C , 73.02; H, 8.72; Tu', to 200-202", undepressed on admixture with a sample pre3.28; 0, 14.97. Found: C, 72.69; H , 8.87; S,3.31; 0, pared as in the preceding experiment; [ a ]-68". ~ The in15.05. frared spectra of the two products were identical. Treatment of 6~~-Cyanoandrostane-3P,5a,l7@-triol (VIc) Further elution with benzene-ether (90: 10, 600 cc.) afof forded 3-(2'-hydroxyethyl)-6-cyano-A3~6-pregnadiene-20-cy- with Hydrogen Chloride in Acetic Acid.-Repetition cloethylene-ketal (XVIa) (460 mg.), m.p. 167-170°, r?ised by the previous experiment with VIc led smoothly t o 6a-cyanoandrostane-3P,5cu, 17P-triol 3,li-diace:ate ( VId), m .p. 230[CY] D crystallizations from acetone-hexane t o 175-177 -96", XEz" 282-284 mp, E 23,990; 3550,2206,1620 232" (from acetone-hexane), [ a ] D = k O and 1590 c m . 7 . Anal. Calcd. for Cd&O&: C, 69.03; H, 8.45; 0 , Anal. Calcd. for C&S;OJ-: C, 73.02; H, 8.72; 0, 19.16. Found: C,68.94; H,8.31; 0, 18.69. 14.97. Found: C,73.00; H,8.63; 0, 15.18. 6~~-Cyanoandrostane-5~1-01-3,17-dione (IXb) .-A solution of Ga-c~-anoandrostane-3P,5a,17P-triol( V I c ) (200 mg.) in Perchloric Acid Hydrolysis of 6-Cyano-As-pregnen-3,20acetone (20 cc.) a t 0" was oxidized for 1-2 minutes with an bis-cycloethylene-ketal (XVa) .-Perchloric acid (8.4 cc., 70%) was added t o a solution of 6-cyano-As-pregnene-3,20- excess of 8 N chromic acid.:' Additioii of water and isolation with ether afforded Ga-cyanoandrostane-5a-ol-3,17biscycloethylene ketal (XVa) (500 mg.) in tetrahydrofuran (8.4 cc.). After 3 hours a t room temperature water (100 dione ( I X b ) (185 mg.), m.p. 276-280°, raised by crystalliza-

.

.

Oct. 5, 1959

C-fj-CYANO STEROID

HORMONES

5241

hEtoH 230 and 288-290 mp, E 4,900 and 11,500, respectively; tions from acetone-hexane to 2&1-286', [a]D +82'; 286-292 mn, E 54; ::A: 3,450, 2220, 1747 and 1700 cm?. h E:'-""" 340-342 mu, E 18,200. 6-Cyan0-A~~~-androstadieae-3,17,9-diol Diacetate (XIc).Anal. Calcd. for C&~703N: C, 72.92; H , 8.26; N, Crude 6a-cyanotestosterone acetate (XIIc) (1.0 g.), pre4 - 2 5 . Found: C, 73.39; H , 8.47; h ' , 4.57. 6P-Cyanoandrostane-5~-01-3,17-dione (VIIIb).-A solu- pared as in t h e preceding experiment,'was dissolved in ace(VIIc) ( 1 .O g . ) tyl chloridracetic anhydride (1: 1, 20 cc.) and heated under tion of BP-c~~anoandrostane-5a,3P,178-triol reflux in an atmosphere of nitrogen for 2 hours. Removal in acetone (200 cc.) a t 0" was oxidized for 1-2 minutes with :in rxceys of 8 N chromic a ~ i d . 5 ~ Addition of water and ex- of the solvent in vacuo and crystallization of the residue from dimethanol afforded 6-cyan0-A~~~-androstadiene-3,17~-diol traction with ether afforded 6B-cyanoandrostane-5a-ol-3,il7-dione (VIIIb) (990 mg.), m.p. 283-287", unchanged after acetate (XIc) (320 mg.), m.p. 200-202", raised by several A mix- crystallizations from methanol to 216-217', [ a ] D - 125"; crystallizations from acetone-hexane, [ a ] D +43'. 262-264 mu, E 17,400; ::A: 2200,1770,1730and 1660 :,"A 3500, ture of I X b and VIIIb had m.p. 258-276'; cm.-l. 2215, 1735 and 1715 em.-'. Anat. Calcd. for C21H3104N: C, 72.51; H, 7.86; 0, Anal. Calcd. for C ~ O H ~ , O ~C, N : 72.92; H , 8.26; N, 4.25; 0, 14.57. Found: C, 73.02; H , 8.29; iX,4.25; 0 , 16.10; N, 3.52. Found: C, 72.31; H , 8.10; 0, 15.87; N, 3.76. 14.78. Potassium Cyanide Cleavage of Sa,6a-Epoxidopregnane6-Cyano-A4-androstene-3,17-dione (XIIb ) .-Dry hydrocygen chloride was bubbled for 10 minutes through a solution 17a-01 3,lO-Biscycloethylene-ketal(XIIIc).-Potassium of 6B-cyanoandrostane-5a-ol-3,20-dione(VIIIb) (50 mg.) anide (1.0 g.) was added to a solution of 5a,6a-epoxidopregin acetic acid. The solution was then kept a t room tempera- nane-17a-ol-3,20-bis-cycloethylene-ketal(XIIIc) (500 g.) in ture for 18 hours. Addition of ice-water and filtration af- ethylene glycol (20 cc.) and heated underreflux for 1.25 hours. (XIIb) as an amor- Addition of water and isolation with ethyl acetate gave a prodforded 6-c,-ano-A4-androstene-3,17-dione phous solid (38 mg.), m.p. 107-112", [a]D - 3 2 ° ~ , 0 2 , ~ ~uct which was adsorbed from benzene (40 cc.) onto alumina (30 g.). Elution with benzene-ether (90:10, 300 cc.) 228-230 and 288-290 mu, e 5,120 and 11,750. Amax 338-340 mp, E 19,950; ;:A: 2220, 3550, 1750, 1670, 1640 afforded 6-cyano-A~-pregnene-17~-ol-3,20-biscycloethyleneketal (XVc) (130 mg.), m.p. 250-255", raised by several and 1600(sh) em.-'. Sodium Borohydride Reduction of 6,3-Cyanoandrostane- crystallizations from acetone-ether to 255-256', [ a ] D- 96'; 3,l'I-dione (VIIIb).-Sodium borohydride (300 mg.) in diox- h?::."," 224 mu, E 10,200; A",: 3500,2210 and 1105 cm.-l. Anal. Calcd. for C26H3705N:C, 70.40; H , 8.41; N, ane-water (5: 1 , 6 cc.) was added to a solution of the diketone VIIIb (550 mg.) in dioxane (15 cc.). After 3 hours a t room 3.16. Found: C,70.36; H,8.54; N,3.27. temperature the solution was neutralized with acetic acid, Further elution with benzene-ether (70:30, 400 cc.) gave poured onto ice-water and the product extracted with ethyl 6 cyano-3- (2 '-hydroxyethyl) -A3*5-pregnadiene-l7a-01-20-cyacetate. I t was then suspended in benzene (75 cc.) and cloethylene-ketal (XVIc) (210 mg.), m.p. 165-170°, raised heated under reflux for 10 minutes when it was cooled t o by several crystallizations from acetone-ether to 178-179", 20". Filtration then afforded GP-cpanoandrostane-3P,178[ a ] D -103'' "::h: 282-284 mp, E 19,500; 3500, diol (VIIc) (180 mg.), n1.p. 295-303", raised by one crystal- 3400,22OO,lk27 and 1588crn.-l. lization from aqueous methanol to 305-308", undepressed Anal. Calcd. for C26H3705N: C, i0.40; H, 8.41; N, on admixture with an authentic sample; [ a ] D -28" (diox- 3.16. Found: C, 70.02; H, 8.2; N, 3.39. ane). The benzene filtrate was then adsorbed onto alumina 6-Cyan0-A~~~-pregnadiene-3,17a-diol-20-0ne 3,17-Diace(25 g.). Elution with benzene-ether (70:30, 350 cc.) afforded f$-cyanoandrostane-3a,li8-diol (X). (150 (mg.) tate (XId).-(a) Perchloric acid (10 cc., 3 N ) was added to a solution of 6-cyano-A~-pregnene-17a-ol-3,2O-biscycloethyl(28%)), m.p. 220-225', raised by crystallizations from aceene-ketal (XVc) (500 mg.) in tetrahydrofuran (15 cc.). ~ tone-hexane to 223-225', [ a ]-27". After 3 hours a t room temperature, addition of water and isoAnal. Calcd. for CzoHl103N: C, 72.03; H , 9.37; 0, lation with ethyl acetate afforded 6a-cyano-17a-hydroxy14.39; X,4.20. Found: C, 71.76; H , 9.04; 0, 14.18; N, progesterone ( X I I d ) (non-crystalline), A%$H 230 and 288 4.66. mu, E 4,680 and 9,775. The total crude product in acetic Further elution with ether (150 cc.) afforded an additional anhydride (5.0 cc.) and acetyl chloride (5.0 cc.) was heated 60 mg. of the 3P-alcohol VIIc, m.p. 298-303' (total yield under reflux for 2 hours in an atmosphere of nitrogen. The (52%). reaction mixture then was evaporated to dryness in high Potassium Cyanide Cleavage of Testosterone Ketal a- v?cuum and the product crystallized from methanol t o afEpoxide Acetate (XIIIb).-Potassium cyanide (2.0 g.) was ford 6-cyano-A3~6-pregnadiene-3,17a-diol-20-one-3, l7diaceadded to a solution of testosterone ketal a-epoxide acetate3I tate (XId) (220 mg.), m.p. 185-187', raised by several ( X I I I b ) (1.0 g.) in ethylene glycol (40 cc.) and heated under crystallizations from methanol to 202-203', [a]D - 112', reflux for 90 minutes. Addition of water and isolation with 262-264 mu, e 18,620; ::A: 2200, 1770, 1740, 1700 ethkl acetate afforded a product which was adsorbed from and 1655 cm.-'. benzene onto alumina 70 g. Elution with benzene-ether : H, 7.57; N, (90: 10, 600 cc.) gave 6-cyano-A6-androstene-17~-ol-3-cyclo- Anal. Calcd. for C Z ~ H ~ ~ Oc ~, N71.04; ethylene ketal (XVb) (320 mg.), 224 mp, E 10,470, 3.19. Found: C, 70.75; H, 7.60; N,3.00. ( b ) Perchloric acid hydrolysis of 6-~yano3-(2'-hydroxym.p. 173-186", raised by several crystallizations from ace(XVIc) (1.55 9.) folI ",".::A 224 mp, E ethyl)-A3 s-pregnadiene-17a-ol-20-one tone-hexane to 213-214", [ ~ ] D -107"; lowed by enol acetylation was carried out exactly as de13,180; XE.":3,600,2220 and llOOcm.-i. Anal. Calcd. for Cz2H3103N: C, 73.91; H, 8.74; 0, scribed for XVc in the preceding experiment and afforded 13.43; N, 3.92. Found: C, 74.36; H , 8.58; 0, 13.31; XId (800 mg.), m.p. 186-191 , raised by one crystallization from methanol to 197-200°, undepressed on admixture N, 3.81. 2 : " 264 mp, e a sample prepared as in method ( a ) ; 1 Further elution with benzene-ether (50: 50) and ether af- with 18,620. forded 3-(2'-h~'droxyethyl)-6-cyano-A~~~-androstadiene-17~6a-Cyan0-17a-acetoxyprogesterone(XIIe) .-A solution of 01 (XVIb) (450 mg.), 282-284 mp, E 16,000, m.p. 6-cyano-A3s6-pregnadiene-3,17a - diol- 20 -one -3,17-diacetate 143-147', raised by crystallizations from acetone-hexane ( X I d ) (200 mg.) in methanol (6 cc.) containing potassium to 154-156', [ a ] D -119'; 284 mp, E 18,620; hydroxide (60 mg.) was kept a t room temperature for 45 3500,2180, 1710 (acetone of crystallization) 1630 and 1580- minutes when the solution was acidified with acetic acid. (sh) em.-'. Addition of water and filtration afforded 6a-cyano-17aAnal. Calcd. for CtzHai08N.CH~COCH3: C, 72.25; H, acetoxyprogesterone (XIIe) (145 mg.) (mainly in the enol 8.98; 0, 15.40; N, 3.37. Found: C, 72.58; H , 8.63; 0, form) as an amorphous solid, m.p. 126-128", [ a ] D -120'; 14.99; N, 3.58. 288-290 mp, e 14,790, plateau 220-230 mp, E 2,700. 6a-Cyanotestosterone Acetate (XIIc).-Concentrated hyA n d . Calcd. for C24H3104hT: C, 72.51; H, 7.86; iY, drochloric acid (0.2 cc.) was added to a solution of 3-(2'-hy(XVIb) (100 3.52. Found: C,72.01; H,7.94; N,3.67. droxyethq 1)-6-cyano-A~~6-androstadiene-17~-ol An attempt to crystallize the above product (m.p. 126mg.) in acetic acid (5.0 cc.). After 3 hours a t room tempera128') from acetone-hexane afforded a semi-solid product, ture addition of water and filtration afforded Ga-cyanotesto[ a ]-73", ~ ",A.:: 230 and 288 m p , e 4,365 and 10,000; sterone acetate (85 Omg.) as an illdefined non-crystalline ::A: 2175, 1725 and 1630 cm.-'. solid, m.p. 115-127 , which resisted crystallization;

-

Az:"

R. XORMAN JONES

5242

AND

BEATRICE S. GALLAGHER

3-Cycloethylene Ketal of Cortisone B.M.D.(XVIII).Ethylene glycol (40 cc.) was added t o a solution of cortisone B.M.D. (XVII) (3.0 g.) in benzene (400 cc.) containing p toluenesulfonic acid monohydrate (200 mg.) and heated under reflux with a water separator for 8 hours. Sodium bicarbonate solution (20 cc., 5%) and water (250 cc.) were added and the product isolated with benzene. Removal of the solvent and crystallization of the product from benzenehexane afforded the 3-cycloethylene ketal of cortisone A.M.D. (XVIII) (1.75 g.), m.p. 188-192', raised by several D crystallizations from benzene-hexane to 200-202', [a] -88'; XVIII exhibited no selective absorption in the ultraviolet. A n d . Calcd. for C2jH3107:C , 67.24; H, 7.67; 0,25.09. Found: C,66.79; H , 7 . 7 3 ; 0,25.26. Permonophthalic Acid Oxidation of the 3-Cycloethylene Ketal of Cortisone B .M .D. (XVIII).-Permonophthalic acid ( 3 . 8 g . ) in ether (50 cc.) was added dropwise over 15 minutes to a solution of t h e cycloethylene ketal of cortisone B.M.D. (XI'III) (4.75 g.) in chloroform (50 cc.) at -10". After keeping at 0' for 16 hours the solution was washed nith cold 5yb sodium bicarbonate solution until i t was acid free and then with water t o neutrality. Removal of the wlvent after dr5 ing over sodium sulfate afforded a product which was adsorbed from benzene onto alumina (200 g.). Elution with benzeneether (80:20, 1.5 1.) afforded the 3cycloethylene ketal of cortisone B.M .D.-Sa,Ga-epoxide (XIXa) (2.55 g.), m.p. 293-297'. The analytical sample from ethyl acetate-hexane had m.p. >300°, [ a ] D -90". A n a l . Calcd. for C?&408: C , 64.92; H , 7.41; 0 , 2 7 6 7 . Found: C,64.76; H,7.63; 0 , 2 7 3 8 . Potassium Cyanide Cleavage of the 3-Cycloethylene Ketal of Cortisone B.M.D.-Sa,6a-epoxide (XIXa).-Potassium cyanide ( 2 . 5 9.) was added to a solution of the epoxide XIXa (1.23 g.) in ethylene glycol (50 cc.) and heated under reflux for 45 minutes. Addition of water and isolation with ethyl acetate afforded a product which wis adsorbed from benzene onto alumina (50 g.). Elution with benzene (750 cc.) afforded the A6-6-cya;o-3-cycloethj lene ketal (XXa) (250 mg.), m.p. 262-264 , raised by several crystallizations 224 from benzene-hexane to 267-269", ID -138", ":,A: mp, E 10,073. .4nal. Calcd. for C&&x: C, 66.22; H , 7.05; N, 2.97. Found: C, 66.59; H , 7.15; N, 3.34. Further elution with benzeneether (90: 10, 750 cc.) afforded t y the 6-cyano enol ether XXIa (420 mg.), m.p. 192-197 , raised by several crystallizations from ben-

[CONTRIBVTION FROM

THE

lrol. 81

zene-hexane to 209-21lo, [,ID -13no, k",t,"," 284-286 m p , e 18,200. Anal. Calcd. for C26H3307N: C, 66.22; H , 7.05; X , 2.97. Found: C,66.iG; H , 7.02; N,3.14. Potassium Cyanide Cleavage of Cortisone Bis-ketal-5a,6aepoxide (XIXb).-Potassium cyanide (4.0 g.) was added to a solution of cortisone bis-ketal-joc,6a-epoxide~3 ( X I X b ) (2.0 9.) in ethylene glycol (80 cc.) and heated under reflux for 1 hour. Addition of water to the cooled solution and extraction with ethyl acetate gave a product which was adsorbed from methylene dichloride-benzene (1 :30, 500 cc.) onto alumina (80 g.). Elution with ether-acetone (90: 10, 1 1.) afforded after one crystallization from acetone-hexane 6-cyano-A6-pregnene-17a,2 1-diol-1 1-one - 3,2O - bis - cycloethylene ketal acetate (XXb) (440 mg.), m.p. 273-275", raised by several crystallizations from acetone-hexane to 277-279", [ a ] ~ -56'; " : :A 22-1 mp, E 11,000; ::A: 3450, 2200 and 1705 cm.-'. Anal. Calcd. for C26H3507K: C, 65.94; H , 7.45; N, 2.96. Found: C,65.64; H,7.66; S , 3 . 0 7 . Further elution with ether-acetone (70:30, 800 cc.) afforded 6-cyano-3-(2 '-hydroxyethyl)-A3~5-pregnadiene-I7a,21diol-11-one-20-cycloethylene ketal ( X X I b ) (370 mg.), m.p. 2~l7-2-19"~ raised by several crystallizations from ethyl ace~ X",t."," 284-286 mp, tate-hexane to 248-250°, [ a ] -59'; e 17,800; ;:A: 3450,2200, 1690 and 1625 cm.-'. Anal. Calcd. for C2&607x: C , 65.94; H , 7.45; K, S , 2 . 9 6 . Found: C,65.64; H,7.66; S,3.06. 6~Cyanocortisone(XXII).-Perchloric acid (3.7 cc., 35%) was added to a solution of 6-cyano-3-(2'-hydrox\-ethyl)A3pS-pregnadiene-170,21-diol - 11,20- dione -20 - cycloethylene ketal (XXIb) (220 mg.) in tetrahydrofuran ( 7 . 4 cc.). After 3 hours at room temperature water was added and the product extracted with ethyl acetate. The combined extracts were washed with sodium bicarbonate solution ( j o b ) , mater and then dried over sodium sulfate. Removal of the solvent afforded 6a-cyanocortisone ( X X I I ) as an amorphous solid, 1n.p. 152-162', which resisted crystallization; [ a ]D -22' 220-222 and 286-288 mp, E 4,300 and (dioxane); " : :A 3500, 2200, 1720(sh), 1700, 10,800, respectively; ::A: 1670 and 1625 cm.-l. The analysis for XXII was unsattsfactory. Anal. Calcd. for C Q ~ H ~ ~ O C,~ 68.55; K: H, 7.06; N , 3.63. Found: C,67.56; H , 7 . 3 1 ; N,2.95.

POSTAL 2679, MEXICO,D . F. APARTADO

DIVISION OF PURECHEMISTRYOF THE NATIONALRESEARCHCOUKCILOF CANADAAND SLOAN-KETTERING INSTITUTE FOR CAXCERRESEARCH']

THE

The Infrared Spectra of Steroid Lactones BY R. NORMAN JONES

AND

BEATRICES . GALLAGHER

RECEIVEDFEBRUARY 20, 1959 Characteristic features of the infrared spectra of twenty-six types of steroid lactones are summarized and discussed. For saturated -1- and &lactones it is observed that, in the absence of direct perturbing influences such as the inductive effects of vicinal substituents, the C=O stretching frequency in solution is negligibly influenced by the location of the lactone group with respect to the steroid ring system. The ranges previously assigned to the positions of the C=O stretching bands of saturated y- and &lactone systems are confirmed, and other absorption specific to different types of steroid lactones is noted between 1450 and 1350 cm. -1. The unsaturated lactone systems of A20[z2)-cardenolides and A20~22-bufadienolides exhibit two bands in the C=O stretching region of the spectrum; the relative intensities of the two bands show large solvent effects, but they are not significantly influenced b y the formal extension of the conjugated system into ring D.

Numerous publications during the past few years have served to focus attention on the prevalence of lactone groups in naturally occurring compounds; typical examples are the bitter principles, such as pictrotoxin and limonin, the santonins, nepetalactone, and gibberellic acid. Lactones are also encountered in steroids; unsaturated y - and 6(1) Published a s Contribution KO.5303 from the Laboratories of the National Research Council of Canada, and No, XXXII in the series "Studies in Steroid hletabolism."

lactone ring systems characterize the cardiac, squill and bufalin aglycones m-hile various types of saturated steroid lactones have been obtained in the course of synthetic and degradative studies. Infrared spectrometrv has been used extensively to distinguish betm-een y- and &lactone ring systems in natural products; a range of 1780-l7GO cm.-' is commonly reported for the C=O stretching frequency in saturated y-lactones and 1750-1'735 cm.-'