a t which time one of us (J. F. T.) worked as a Post Doctoral Fellow of the National Institute of Health. The association of G. H . N. T. with the work made possible by a Lalor Foundation Fellowship tenable a t Cornel1 University. The assistance of Mr. R. Rabson with the quantitative work is ac-
[CONTRIBUTION FROM TIIB
knowledged. The completion of the work a t Corne11 University was assisted by the Grasselli Grant to Cornel1 Cniversity for work being directed by one of us (F. C . S.1. Grateful acknowledgnlent of all this help is now made. ITH.ICA,
RESEARCH LAHORiT
XEIV YORK
lRlOi, IIEIICK
&
C O . , I S C . , R.edI\vA\I']
Isomerization and Displacement Reactions in the Cyanopregnene Series. Cortisone 21-Methyl Ether
Synthesis of
BY H U X S G - ~ I I N L O ROGER S , ~ ~TCLLAND JOHN BABCOCK Lb RECEIVED SOVEMBER 10, 1953 20-Cyaiio-li-prrg1ieiie-:i!a),L'1-:liol-~~i~I-l 1-oiie has been found to undergo geometrical inversion under allinliiic conditions accompanied by ether formation a t positioll 21. This discovery has enabled the synthesis of 17-liydr oxydehydrocorticosterone-21-methyl ether.
20-Cyano-l'i-pregnene-3( a ),2 1-diol-11-one 3,21diacetate (Ia) is deacetylated readily by brief treatment with methanolic alkali to give the corresponding diol (I).2 It has been found that prolonged treatment of Ia with methanolic alkali leads to the formation of two new substances: A,m.p. 234-23.5' and B, m.p. 221-222". Although the ultraviolet and infrared spectra of these substances clearly indicated the retention uf the cyanoethylene system, their non-identity with 20-cyano-17-pregnene-3(a),21-diol-ll-one(I) was demonstrated by mixed melting point comparison. The same two substances, A and B, were formed by reaction of the pure diol I with methanolic alkali. The proportion of the two products formed was found to depend on the length of time the reaction mixture was allowed t o stand. Thus, when a solution of the diol I in 3y0 methanolic alkali was allowed to stand for 21 hours or longer, substance B, m.p 221-922', could be isolated in over 8070 yield in addition to a small amount of substance A , m.p. 231-232". \Vhen the reaction time was shortened to three or four hours, however, a mixture of substances A and B together with some unchanged starting material was obtained in which substance m.p. 234-235", prcdominated. It was suggested from the analytical data alone t h a t substance B might be a configurational or positional isomer of the startinK diol I having structure I1 or XI, respectively. I t was found, however, t h a t substance B formed only a monoacetate, m.p. 158-159" even after prolonged refluxing with acetic anhydride. This monoacetate derivative, moreover, exhibited no free hydroxyl band in the infrared spectrum thereby excluding structures I1 or XI for this substance. The possibility remained that the 21-hydroxyl group of the starting diol I had been replaced by a methoxyl group to give the methyl ether (111). Consistent with this consideration 111 should form a monoacetate possessing no free hydroxyl function. Structure I11 for substance B was coilfinned by active hydrogen and n~ethouvl groul) deteriiiinatioiis performed 011 I1 I :m(1 its ( 1 ) (a) H u ~ t i g - M i n l o n , 178 S i n 1," R o a d , Shaiigliai, C h i n a : Tohn B a b c o c k . U p j o h n C o m p a n y . Kalarnazoo, M i c h . (2) 1,. H . S a r e t t , TEIIS JOUR.V.IL. 70, 1431 (1018).
(111
monoacetate, respectively. Ozonolysis of 111, furthermore, yielded the known 17-ketone (V),3 whereas hydroxylation with osmium tetroxide afforded pregnane-3(a),lT(~u),21-triol-11,20-dione 3monoacetate 21-methyl ether (VII) (see below) in analogy with the conversion of I to dihydrocortisone (VI).* In order to demonstrate that the 21-hydroxyl function of I had become displaced by a methoxyl group from the solvent, methanol, I was in turn treated with ethanolic alkali. I n this case a new substance, m.p. 144-145', was obtained together with the above-mentioned substance A, m.p. 23423.5'. The analysis, ultraviolet and infrared spectra of the new product, m.p. 144-14t50, agreed with the 21-ethoxy structure IV. The latter also formed a monoacetate, m.p. 139.5-140.3', and the presence of an ethoxyl group was confirmed by Zeisel determination. Substance -4, m.p. 234-235', was found to be the A17,20-geometrical isomer of the starting diol I and is therefore to be formulated as I I . 4 Thus when the diol I was treated with alkali in aqueous dioxane, only I1 along with some unchanged I could bc isolated. I1 formed a diacetate, m.p. 1.'3.5', still in possession of the cyanoethylene system and exhibiting no hydroxyl band in the infrared spectrum. Thr. structure of I1 was confirmed by conversion of its 21-inonoacetate through osmylation and oxidation to VI, identical with t h a t obtained from the 21iiionoacetate of I by the same sequence of rcactions.5 It is interesting to note that I1 may be the interrncdiate product in the formation of 111 from I . As mentioned above, I is largely converted to I1 on short standing with methanolic alkali but gives I11 I.. 11. S a r e t t , J . Bioi. C h e m . 162, 601 ilO4Ci). I ) Inspection of t h e m o d e l s of I a n d 11 i n d i c a t e s t h a t in I t h e r e i s a ili>.Rnite c r < , w d i n sof groiips between C i s a n d ( 2 2 1 as well a s i i d i s t i n c t r m I r i c t i < ~ ot u ire? r,ltatiun u f t h e C i , - h y d r o x ~ i ~ i e t l i ~ I egruuii rle almut I n c o n t r a s t t h e r e t o ii*i iipparrril s l e r i c t h e & X I S c b f tlir C:o ( 2 1 1 b u n d friterference is ulfrrrd t,, t h e Iine.ir - - C S yrniip i n 11. I t nidy theref h r h r ~ ~ , r i i l , i ~ l teerrli t s t i v r l v 1h:ii I I rel,resciits (lie inure stiil>le el, 1iiv:t-ic:il 4;)eL.ie.i nnd ccmseqnr-itly is t h e coiiliguration t o h e aeriuiieil tr, t h e pred~,inin:int i s o m e r arising f r o m t h e base-catalyzed inversion. (.j) P r i v a t e c o m m u n i c a t i o n from Dr. R . E . l u n e s a n d Rlr. S. A . Kobinson of t h e s e l a b o r a t o r i e s f:?)
May 5 , 19.54
R1 = R ? R, R,
SSNTHESIS OF CORTISONE 2 1 - l f E T I I Y L
11, R1 =
H CH!CO
11:~ R I 1111, Ri
R: = H
111, R I = R:! = I V , KI = Rz =
RI CHjCO H , RJ CHrCO
CH?OOCCTIg
CH:!OCH?
CO
co
I
NC
0
VI I
I/
in good yield after a longer reaction time. Furthermore, I1 is independently converted t o 111 in over 80% yield on treatment with methanolic alkali under the same conditions for the conversion of I to I11 directly. Both isomers I and I1 as well as the 21-methoxy compound I11 are quite stable to acid since they could be recovered essentially unchanged after refluxing for two hours with hydrochloric acid in methanol. The mechanics of the transformation of I to give I1 and I11 may be visualized reasonably as proceeding via base-catalyzed ,&elimination to a cyano diene (see below) wherein the latter provides a unidirectionally activated system to further reaction with alkoxide ion to provide the configurationally more stable 21-substituted cyanoethylene.
Visualized in this manner path A becomes essentially irreversible with path 13 reversible. The above
CHIOCTIJ
\/ C
os04
x
V , R = H, CFIqCO
--OH
CH COO 1’1
2397
H , CHjCO, CH3 H , CHiCO, C2Hs
I
I
0
ETHER
IX
/
VI11
XI
interpretation gains some measure of support from the fact that I in the presence of hydroxide ion alone undergoes geometrical inversion whereas it is stable to acid; furthermore, the formation of 17-substituted derivatives was never realized. The relatively small differences in optical rotation of 11, I11 and I V and their acetate derivatives in contrast to their marked difference in rotation as compared with I and its acetate derivative strongly indicate that 11, I11 and IV are all in the same geometrical series. The methoxycyanopregnene (111) was converted into 17-11ydroxydehydrocorticosterone21-methyl ether (X) by oxidation with N-bromoacetamide to give the 3-keto derivative VI11 followed by hydroxylation with osmium tetroxide to give the keto1 IX and finally bromination and dehydrobromination. The structure of X was confirmed by analysis, methoxyl determination and spectroscopic examinations. Acknowledgment.-We are indebted to Dr. Max Tishler for his interest and encouragement in the pursuance of this investigation and to Drs. Karl Pfister and N. L. Wendler for many stimulating discussions and suggestions.
2398
HUt\h'G-lfINLOX',
Experimental
ROGERTULLAND
JOHN Eluant
BABCOCK E l u a t e , g.
Vol. 76 X p . range,
O C
Chloroform 0.269 218-221 20-Cyano-17-pregnene-3(~),2l-diol-1l-one 3,2 1-Diacetate Acetone ,021 L'28-2:14 (Ia).'-Five grams of 20-cyanopregnane-3(a) ,20,21-triol-11one 3,21-diacetate' was dissolved in 17.5 cc. of pyridine. Recrystallization (chloroform-petroleum ether) of, the After cooling the solution to O", the solution was treated "chloroform fraction" gave 0.28 g., m.p. 221-222.,2 , dewith 2.9 cc. of phosphorus oxychloride and allowed to stand pressed with starting material but not depressed 011 admixovernight a t room temperature. The mixture was poured ture with 111 from I . Recrystallization of the "acetone into cold, dilute hydrochloric acid. After decantation of fraction" from the solvent pair gave 0.011 g., m.p. 233the dilute acid, the gum was triturated with water and di234"-not depressed with the starting material. solved in methanol. The solution was diluted with water 20-Cyano-17-pregnene-3(~),2l-diol-1l-one (11)from I in until turbid and allowed to stand a t 0-5' for four days. The cyanopregnene was filtered and recrystallized from a Dioxane.-To a solution of 1.0 g. of I in 50 cc. of dioxane mixture of benzene and petroleum ether; weight 2.
