Steroidal Thiazenones and Related Compounds1 - Journal of

Philip E. Shaw, F. W. Gubitz, K. F. Jennings, G. O. Potts, A. L. Beyler, and Robert L. Clarke. J. Med. Chem. , 1964, 7 (4), pp 555–560. DOI: 10.1021...
1 downloads 0 Views 824KB Size
.July. 1964

STEROIDAL THIAZENOXES

Steroidal Thiazenones and Related Compounds’ PHILIPE. s H . k ~ v , F. Jy.GUBITZ,K. F. JENNINGS, G. 0. P O T T S , AND ROBERT L. CLARKE

A\.

L. BEYLER,

Sterling-Winthrop Research Institute, Rensselaer, New York Receiued Soveniber 8, 1963 Steroids with a thiazenone ring fused to the 2,3-positions have been prepared. Also, the reartion of rnetliyl rnercaptoaretate with a ~a-hronio-3-ketosteroidhas led to a series of 2-rnercapto-3-osygenated derivatives, snnie of whirh were cyrlized. Evidence for the configuration of the mercapto substituent at C-2 is presented.

Xndrostaiio[3,2-e]thiazenones have been prepared by the condensation of mercaptoacetamide with 2 a-bromo-3-ketoandrostanes. Since this new class of steroidal heterocycles contains a heterocyclic ring fused to carbons 2 and 3 of the steroid nucleus, it is similar to the androstano [3,2-c ]pyrazoles* and androstano [2,3-d]isoxazoles3 which were previously prepared in these laboratories and found to possess a favorable ratio of anabolic to androgenic activity. Sokol and Ritter4 observed that simple a-halomethyl ketones reacted with mercaptoacetamide to form products which were later characterized by deStevens, et d . , j as having the 1,4-thiaz-3-en-3-one structure (1).

2a,

R R

= =

OAc, R’

OH, R’

(1) Steroidal Heterocycles I S . For preceding paper, see D. K . Phillips and A. .J. Mason, J . O r g . Chem., 28, 2886 (1963). ( 2 ) R. 0. Clinton, A. J. Manson. F. W. Stonner, H. C. Neumann, R . G . Christiansen, R . L. Clarke, J. H. Ackerman, D. F. Page, J. W.Dean. W,B. Diekinson, and C. Carabateas. J . A m . Chem. Soc., 83, 1178 (1961). 13) .1.J. Manson, F. W. Stonner, H. C. h-euniann, R . G. Christiansen, R. L. Clarke, J. H. .Ickerman, D. F. Paae, .J. IT, Dean, D. K. Phillips, C;. 0. Potts, .i.Irnold, . .\. L. Beyler, and R . 0. Clinton, J . .Wed. Chem., 6, 1 (1963). (1) H. Sokol and .J. J. Ritter, J . A m . Chem. S o c . , 7 0 , 3517 (1948). ( 5 ) G . rIeStevens, A . Halainandaris, and L. Dorfman, ihid., 80, 5198

(1958).

3a, R = OBc, R’ = H

C.

OR /3a\

CH3CONH

In this laboratory, the condensation of mercaptoacetamide with steroidal 2 a-bromo 3-ketones (2a-d and 12) afforded in each case the corresponding thiazenone (3a-d and 13). The reaction mas carried out in either methanol or acetonitrile a t reflux temperature, usually under nitrogen with a reaction period of 24-36 hr. Curiously, in the reaction between 2a and mercaptoacetamide using methanol as the solvent, potentiometric titration of aliquots of the reaction mixture after 1.25, 3.5, and 48 hr. reveaIed that no acidity developed during the course of the reaction. Addition of 2 molar equiv. of collidine to the reaction mixture resulted in only a slight improvement in yield of the thiazenone. The 170-acetoxyl group in 3a could be hydrolyzed by aqueous methanolic potassium carbonate to give 3e without attack on the thiazenone ring. Compound 3e was converted to its 17~-cyclohexylpropionate (3f) and 17p-hemisuccinate (3g) esters. The ultraviolet spectrum of each of the thiazenones showed an expected; absorption maximum a t 293299 mp ( e 2300-2600) and a shoulder a t 230-232 mp ( E 3800-4300). One derivative (3d) was so insoluble in most solvents that its ultraviolet spectrum had to be

= H = CH,

b, R = OH, R ’ = CH, R = COCHS, R’ = H C. R = COCH,. R’ = H d;R, R ‘ = 0 d; R, R’ = 0 e, R = OH, R’ = H f , R = OOCCH?CH,C,H,,, R’ = H g, R = OOCCH,CH,COOH,R’ = H b,

OAc

JYP o L ; l P H

CHJ

4a, R

= Ac b, R = H

5

0Ac

1 22 hr.

I

O

-

8

:$Y7

measured in dimethylformamide, thereby permitting detection of only the absorption maximum a t 29.5 mp ( E 2270). There has been no unequivocal evidence that thiaeenones were formed through replacement of halogen by sulfur and of carbonyl oxygen by nitrogen in a-haloketones, rather than the reverse attachment.5 We have established that nitrogen in the thiazenone ring is indeed attached to the steroid nucleus a t C-3. Thus, desulfurization of 3a with Raney nickel6 gave the saturated acetylamino derivative 4a, which has been described p r e v i ~ u s l y . ~Selective hydrolysis of 4a by (6) C Djerassi, N Crossley, and ,\I 4 Kielczeuski [ J O r g Chem , 27, 1112 (1962)) reported t h a t t h e production of saturated i s unsaturated enamine u a s dependent upon the age of the R a n e l nlckel 1% e ha\ e no data on the ape of the R a n e l nickel used in our desulfuriLatlon reaction (7) R 1 Oletta, Belgian Patent 596 561 ( S o l e m b e r 15 1960)

II

Ei loa, R = Ac

lla, R=Ac

b, R = H

b,R=H

,

QAc

H 13

J

12

0 H

14

lkductioii ot keto ester 15 with lithium tri-t-butos~-alumiiiohydridc follo\red by chrornatography resultcld in the isolatioii of two crystalline products (18 aiicl 19). The cyclized product (18) could tic ohtaiiicd hy deliberate arid-catalyzed cyclization of 19. The mercapto group at C-2 was showii to havc the. a-coiifiguratioii in this serics of compounds. Opeiiiiig of the mol lactoiic ring of 16 by alkaline h.vdrolysib gave a keto acid (20) which, by its mode of formation, can be coiicluded t o h a w the more stable 20-configuratioii for itb carboxyniethylniercapto group. Keto acid 20 \\a. also obtaiiird by alkaline hydrolysis of keto wtcr 15. Reduction of keto acid 20 r i t h litliiurri tri-t-biitoxyaluniiiiohydrideafforded 2a-carboxymethylriiercapto-.‘,cu-aiidroctarie-38,17p-diol(21). Thc bani(* obtaiiied by fapoiiificatioii of hydrouy ice kcto ester 15 was reduced to hydroxy ester 19 under conditions12which would not be expectcd to epimerize an axial substituent O( to a k c h i ( ’ , a t d siiice hydroxy ester 19 was hydrolyzed to a conipoiitid (21) iii which thc iifiguratioii a t C-2 had hwti itueiit a t C-2 iii kcto estw 15 demonstrated, the s and in the conipou~idsderived from it must o c ~ u p ytho more stable ccluatorial ( a ) position. A-ksigiimeiit 01 the a-configuratioii to thc C-2 substituent in keto amidr 6 was based on the fact that it was formed in a m a i i i i ~ r analogous to that of keto ester 15. Biological Results.--The myotrophic atid aiidrogclriic* ’ 1.2) J I ajhos. ~ ‘ o l l ~ ~ l C~ oz 7~1i i

(’i,rm

(’ommun , 24, 228-1 l l ‘ l 5 ( l )

July, 1964

STERoIDaL

d

C H30 OCC H29 A c

O

H

THIAZESOKEY

5.57 TABLEI

?VfYOTROPHIC A S D . i N D R O G E S I C P O T E S C I E S RE:r,.4TITE T O

TEST~STEROSE PROPIONATE Compound

W

Xyotrophic

.Inrhogenic

Separation index"

3e 1 /4 l/T> S 3a '/s 1/3% 4 3b '/8 1/32 4 3g '/0 ' /?n 4 a Separation index is tlie ratio of the relative niyot,rophic to relative androgenic potencies wit'li testost,erone propionate as the reference st'andard.

17

n.3

HOOCCH,S.,#

I

HOW

H 21

activities of 3a, 3b, 3e, and 3g were determined by a modification of the method of Hershberger, et a l l 3 Immature male rats of the Sprague-Dawley strain, 22 days of age (41-14 g.), were castrated and maintained on laboratory chow and tap water ad libitum in air conditioned quarters. Each compound was administered subcutaneously daily except Sunday for 9 days, starting 7 days after castration. The animals were autopsied on the 17th post-castration day, 24 hr. after the last medication. The levator ani muscle and ventral prostate mere excised, blotted, and weighed on a microtorsion balance. The myotrophic and androgenic activities relative to testosterone propionate are presented in Table I, together with indices of separation.

Experimentall4 17p-Acetoxy-5a-androstanoj3,2-e]-1 ',4 'dhiaz-5 '-en-3'-one (3a).-A solution of 22.3 g. (0.054 mole) of 2a-bromo-17p(13) L. G. Hershberger, E. G . Shipley, and R. K. Meyer, Proc. Soc. Ezptl. B i d . M e d . , 83, 175 (1953). ( 1 4 ) Melting points are corrected unless otherwise stated. Specific rotations were measured on 17csolutions in chloroform except a s noted. Ultraviolet spectra were measured in 95% ethanol with the one exception noted where diniethylfornianiide (DhIF) was used. \Ye are grateful t o Dr. F. C . Nachod and staff for the spectral d a t a and to I I r . K. D. Fleischer and staff for analytical servires.

acetoxy-Fa-androstan-3-one (2a),15 7.4 g. (0.081 mole) of nierc a p t o a ~ e t a m i d eand , ~ 13.1 g. (0.108 mole) of ycollidine in 500 nil. of acetonitrile was refluxed under nitrogen for 36 hr. About 300 ml. of solvent was removed in uacuo a t