Sept., 1950
LACTONES FROM DEHYDROISOANDROSTERONE
4073
[CONTRIBUTION FROM TEE RESEARCH AND BIOLOGICAL LABORATORIES, AYERST, MCKENNA & HARRISON, LTD.]
Lactones Prepared from Dehydroisoandrosterone BY CARLVON SEEMANN AND GORDON A. GRANT The oxidative cleavage of ring D in a 17-ketosteroid with subsequent lactonization of the hydroxy-acid postulated as intermediate has first been described by Westerfeld, who oxidized estrone (I) with alkaline hydrogen peroxide obtaining a lactone ClsHzzOs which formed an acetate and a methyl ether. Westerfeld was unable, however, to formulate an exact structure for his lactone, and he proposed the alternative possibilities of cleavage of ring D between CIS and C17 or between Clc and (217, with lactones of the structure I1 or I11 as the end-products of his reaction. He also advanced some evidence which induced him to prefer to ascribe structure I1 to his lactone.
A most convenient preparation of some functional derivatives of this acid has been described by Miescher and his co-workers,'O who oxidized dehydroisoandrosterone acetate with potassium hypoiodite. I t s constitution as A9r1*-2,13-dimethyl-7(/3)-hydroxydodecahydrophenanthryl-l -acetic acid-2-carboxylic acid (IV) was established by Kuwada.s-8 Pl-COOH
HO IV
0
HO
m
HO
I11
The biological properties of this lactone were subsequently i n v e ~ t i g a t e d . ~ ~ ~ ~ ~ A promising approach to the problem of the synthesis of similar lactones with known structure seemed to be offered through the use as starting materials of the dibasic acids obtained from certain 17-keto-steroids by oxidative opening of ring D. The present investigation relates to the synthesis and characterization of two such lactones derived from esters of 3(p)-hydroxy- Ass6etiobilienic as obtained from dehydroisoandrosterone. (1) Westerfeld, J . Biol. Chem., 148, 177 (1942). (2) 0.W. Smith, Endocrinology, 36, 146 (1944). (3) 0.W. Smith, Proc. SOC.E x p f l . B i d . Mcd., 69, 242 (1945). (4) 0.U'. Smith and G. V. S. Smith, J. Clin. Endocrinol., 6,483 (1046). (5) Kuwada, J. Pharm. SOC.Japan, 66,75 (1936); C.A . , 31, 2224 (1837). (6) Kuwada, J . Pharm. Soc. Japan, 68, 631 (1936); C. A , , 31, 2224 (1937). (7) Kuwada and Nakamura, J. Pharm. SOC.Japan, 68,841(1938); C. A . , 83, 2530 (1939). (8) Kuwada and Nakamura, J . Pharm. SOC.Japan, 68,835(1938); C. A , , 88, 2531 (1939). (9) Butenandt, Schmidt-ThomC, Weiss, Drealer and Meinerts, B ~ .72,417 , ~iwg).
The sequence of reactions we employed to obtain lactones from the above dibasic acid IV consisted essentially of two steps: (1) protection of one of the two carboxylic acid groups by selective esterification and (2) reduction of the remaining free carboxyl to the corresponding alcohol through the half-ester acid chloride, the benzylthiol ester and Raney nickel reduction.'l Selective esterification had been achieved by K u ~ a d a , who ~ J described the two isomeric monomethyl-half-esters of IV, the primary or a-monomethyl ester VIa and the tertiary or 0-monomethyl ester VIIIa. The method of preparing the tertiary or pmonoesters by partial hydrolysis of the diesters as first employed in this series by Kuwada6J follows the general rule relating to the ease of hydrolysis of primary as against tertiary carboxylic acid esters, It has given consistently reliable results in the hands of Miescher and his co-workers12and the constitution of such compounds seems to be well established. The preparation of the primary a-monoester which had hitherto required the esterification of the free dibasic acid I V with the alcohol and a catalyst has been considerably improved by the finding that the crude hypoiodite oxidation mixture of V yielded on acidification a mixture consisting of about four-fifths a-monoester VIa and one-fifth diester VII. From this mixture the amonoester VIa could be isolated by alkali extraction, while the @-monoesterVIIIa could be prepared in quantity by treating the above mixture with diazomethane to form the diester VII, followed by partial hydrolysis to the &monoester VIIIa. (10) Wettstein, Fritzsche, Hunziker and Miescher, Helo. Chim. ~ c t a 24, , 3323 (1941).
(11) Jeger, Norymberski, Szpilfogel and Prelog, ibid., 29, 684 (1946). (12) E . g., Heer and Miescher, ibid., 48, 156 (1945); Heer, Billeter and Miescher, i b i d . , 28, 991 (1945); Heer and Miescher, i b i d . , 29. 1895 (19461,and SO, 786 (1947).
4074
C.4RL VON SEEMANN AND (>ORDON
~ 4GRAKT .
\*01. i?
Condensation of the acid chlorides I X and X in benzene solution with benzyl mercap/Y P '~~-C0OCI-la tan under the influence of pyridine gave the I thiolbenzyl esters XI and XII, respectively. , \ /'\, /' '-cH2CooCH' The ester XI, which was not obtained in KO1 crystalline form, on hydrolysis with one nor1' 4 I VI I mal aqueous methanolic potassium hydrox* c O / v ~ ~ ' I{()/'\ '" ide consumed three equivalents of alkali to vield IV, while XII, because of the tertiary KO1 $ CHiL. 0.5 KOH nature of its carbomethoxy group, was hydrolyzed by only two equivalents of alkali COOH t c ~yield VIIIa. /'2!-COOCH, Treatment of the thiolbenzyl esters X I and XI1 in methanolic solution at room tem,.\ ,-cH2COOH I ,,,-, 4 -CH,COOCH? perature with Raney nickel according to the method mentioned aboveI2 yielded not the VIa R = H = expected 6-hydroxyacid esters C22H3406 (XVa RO '&'\' VIb: R = Ac R d \ -'''\'' VIIIb, R = Ac and XVIa), but rather two neutral compounds of the composition C21H3004, which j. proved to be the acetoxy-lactones XIIIa - -CC)CI p\ -COOCH? and X W a . The neutral substances, which ,/\V/-CH,COOCHq /\,/ -CHnCOCl could not be acetylated, while insoluble in cold dilute alkali, dissolved under reflux, IX s consuming two equivalents of alkali to yield sparingly soluble sodium salts. i4cidification J.I ;--COSCHGH, of the saponification mixtures yielded the A'-COOCH~ unacetylated lactone^'^ X I I I b and XIVb. While reduction of the thiolbenzyl esters / '-CH~COOCH~ I CH2COSCH2CI;II X I and XI1 with Raney nickel may a t first SI1 have produced the 6-hydroxyacid esters XVa and XVIa, one molecule of methanol was apparently eliminated with the concomitant 0 111 1 formation of the lactone ring c 0 c 0 This unusual reaction might be explained /A by the inordinately strong tendency toward c 0 crt lactonization of compounds of the type of A/"\ c c XT'c and XVIc, which seemed to hc only I1 11 stable in the form of their sodium salts C19HdI4Na (XVd and XVId), both of whlch RO/ 'd' HO ' were isolated and identified by analysis. XIIIa, R = Ac XI\'a, R = Ac Acidification of XVd and XVld a t room XIIIb, R = H ~n-i)K , = ri temperature proved to be sufficient to close ' -CHJJIl - COOR' the lactone rings with the formation of X I I I b and XIVb, and the usual heating with stronq CH JCOOII' C H?CH-OH ,'-\ '\, acids to effect lactonization was not required. Finally, we were able to show that thr I double bond between CSand Ca of the origl/ \ R O / Vv KO/ "4 '" nal sterol was still present in the two acetoxyXVa, R = Ac, R' = CHa XVIa, R = Ace K' = CHlactones XIIIa and XIVa, as well as in the XVb, R * Ac, R' = H XVIb, R = Ac, R' = H two lactones XIIIb and XIVb. -111 four XVc, R = H, R' = €I XVIc, R = H, R' = H XVd, R = H, R' = h-a XVId, R = H, R ' = S a compounds gave positive color reactions with tetranitromethane. In addition we The half-esters VIa and VIIIa were acetylated prepared the dibromides C21H3004Br2 of the with acetic anhydride-pyridine and the acetyl acetoxy-lactones XIIIa and XIVa in the usual derivatives VIb and VIlIb allowed to react in manner, which could be debrorninated with zinc benzene solution with oxalyl chloride13 to yield to yield again the parent compounds XIIIa and the corresponding ester acid chlorides I X and X W a . Levy and JacobsenI6 have reported the preparax.14 tion of a lactone obtained by oxidati6n of dehydro(13) Oxelpl chloride t ~ z sempleyd at tixis pomt (see Stallberg0
A
4
if
4
1
/ \
i
' v4
J.
/ %
Stenhagen, THIS JOURNAL, 69, 2568 (1947)), to obviate the isornerization of the ester acid chlorides as found to occur with thionyl chloride by Cason (ibrd., 69, 1548 (1047)). (14) Heer and Miescher, Hclo Chim A d a , 80, 786 (1947)
(15) For reasom of convenience we designated the lactones rived from the a-monomethyl ester VIa as or-lactones, and thO.r rived from the 8-monomethyl ester V I I I a as 8-lactoneq !161 I.ev, and Jacohwn J Erol Chem , 171, 71 f10+71
(It..
