Anabolic Agents. 2,3-Epithioandrostane Derivatives P. D. KLI~ISTRA

P. D. KLI~ISTRA,'. E. F. NCTTING, AND R. E. COUNSELL. Divisions of Chemical and Biological Research, G. D. Searle and Company, Chicago, Illinois 60680...
0 downloads 0 Views 600KB Size
September 1966

LkV.%BOLIC 2,3-EPITHIO.iNDROST.INE

DERIVATIVES

693

Anabolic Agents. 2,3-Epithioandrostane Derivatives P. D. KLI~ISTRA,' E. F. NCTTING, AND R. E. COUNSELL Divisions of Chemical and Biological Research, G. D. Searle and Company, Chicago, Illinois 60680 Received dfarch 17. 1966

A number of 2,3-epithioandrostane derivatives were prepared in the hope of obtaining compounds which possessed high anabolic and minimal androgenic activity. h comparijon of the biological activity of the;e episulfide derivatives with the corresponding epoxides is diicuised. The chemical pathway t o iheqe substance. is described in detail.

A recent' report2 from t'he Shionogi laboratories concerning steroidal episulfides has prompt'ed us to describe some of our work in this area. I n previous publication^^,^ me recorded att'empt'smade at preparing compounds possessing high anabolic and low androgenic act'ivity. Because of the interesting activity of compounds of the androstane series possessing a 2,3 double bond,5s6 we felt, it of interest to prepare some 2,3epoxides and episulfides and compare their parenteral and oral anabolic and androgenic act'ivit'ies. The preparation of st'eroidal epoxides (Table I) proceeds in good yields either by peracid treatment of an ~ l e f i nor~via ~ ~base treatment of the corresponding h a l ~ h y d r i n s . ~Specifically, ,~ the 2,3a-epoxides I1 were prepared by treating the olefins I with either perbenzoic or m-chloroperbenzoic acid. On the other hand, the 2,3p-epoxides VI1 were derived from the bromohydrins V I upon treat'ment with base. In most' cases, these epoxides were used as intermediates in the preparation of the episulfides as indicated below. Earlier attempts in these laboratories to prepare t'he episulfides by direct exchange of the epoxide oxygen with sulfur as reported by Van Tamelen for cyclohexene oxidelo were unsuccessful affording only unchanged epoxide. However, in 1962, Lightner and Djerassi" reported on t'he conversion of 2,3a-epoxy-Sa-cholestane to the 2,3&episulfide in very low yield by the direct displacement procedure. A more feasible approach to the 2,3-epithio-Sa-cholestanes was report'ed in 1962 by Takeda and Komeno who treated the corresponding epoxide with thiocyanic acidI2 to obtain t,he thiocyano alcohol which mas converted t,o t'he episulfide with alcoholic potassium h y d r 0 ~ i d e . I ~ In our studies, the r e a d o n scheme outlined in Chart I proved to be the most convenient rout'e t'o the a- and 0-episulfides in the androst.ane series. Briefly, the 2,3a-epoxides I1 were treated with an ether solution of thiocyanic acid prepared in situ12t,o give good yields of (1) T o %,horn inquiries should be addressed. (2) K. Takeda, T. Komeno, J. K a a a n a m i , 8. Ishihara, H. Kadokawa, H. Tokura, a n d H. Itani. Tetrahedron, 2, 329 (1965). (3) R. E. Counsell, P. D. Klimstra, a n d F. B. Colton, J . Org. Chem., 27, 248 (1962). (4) P. D. Klimstra and R . E. Counsell, J . M e d . Chem., 8, 48 (1965). (5) R. E. Counsell, G. Adelstein, P. D . Klimstra, and B. Smith, submitted for publication. (6) R. E. Counsell and P. D. Klimstra, U. S. P a t e n t 3,203,966 (1965); Edwards a n d A. Bowers, Chem. I n d . (London), 1625 (1960). (7) J . Iriarte, J . Org. Chem., 20, 543 (1955). (8) J. Fajkos a n d F. Sorm. Collection Czech. Chem. Commun., 24, 3115 (1959). (9) J. Fried and T. Sabo, J . A m . Chem. Soc., 79, 1130 (1957). (10) E. E. Van Tamelen, i b i d . , 73,3444 (1951). (11) (a) D. A. Lightner and C. Djerassi, Chem. Ind. (London), 1236 (1962); (b) Tetrahedron, 21, 583 (1965). (12) (a) K. Takeda and T. Komeno, Chem. P h a r m . Bull. (Tokyo), 8, 468 (1960); ( h ) i b i d . , 8 , 672, 680 (1960). (13) K. Takeda and T. Komcno. Chem. I n d . (London), 1793 (1962).

H

H I

H '

I1

I11

NCS-0

A

H

IV

V

&

HO Br. -

H

+

H

334 H

0

NCS H

H

IX

X a , R, R I - 0 b, R = H ; R i = O H C,

R=H;R,=OAc

d. R-CH,; Ri=OH

the thiocyano alcohols I11 (Table I). Treatment of 111 with methanolic potassium hydroxide at room temperature for up to 2 hr afforded the 2,3P-episulfides IV (Table I) in good yield. The 2,3@-epoxidesVI1 were treated similarly to give the 2,3a-episulfides IX. An alternate procedure (Chart 11) which can be utilized for the preparation of the 2,3a-episulfides was shown by the preparation of 2,3a-epithio-Sa-androstan17p-01 (IXb). When purified 2a-bromo-5a-androstane3,17-dione was treated with potassium thiocyanate in acetone in a manner similar to that described previously

I1

0.78

I t I. GO ! ) , 79 10.41

10,5!l s.41 S.!H

!I , 1 .i

4.41

s , !I4 !I.

1.5

!l.27 !I.X i

!I ?(ti

I (I.07 9.2; !I.S(i

! I . 2fi

IO.0i

CH \ I { T I1

0

k SI

for the cholestane berieS,l3 the corresponding thiocyanate Val4 was obtained. Reductioii of Vu n-ith lithium tri-t-butoxyalurninum hydride afforded 2athiocyano-5a-androstaiie-3/3,17/3-d1ol(XI). Treatineiit of XI with methanolic potasqiuni hydroxide at room temperature gave IXh. An attempt was made to prepare the 3-lieto-2@thiocyarioandrostaneq by mild oxidation of the thiocyano alcohols 111. Treaiinent of IIIa with clironuc~ acid reagentI5 in acetone, however, gave the 2athiocyano epimer (V) identical with that obtained above. Apparently, because of the nonbonded 1,3dinxial interaction with the 19-methyl group, the axial configuration of the 2@-thiocyano substituent is uiistable and immediately inverts to the equatorial form

+

(1%)T h e nmr spectrum of this substance indicated a quartet ( J a X This ia con = 20 ops) a t 262 cps indlcatiie of a n axial proton a t C-2

sistent n i t h t h a t reported by Takeda and Komeno'x for 2a-thiocyano-5ocholestan-3-one (15) K. Bowden. I. XI. Heilbron E R H Jones, and €3 C L Weedon J ChPm Sor , 99 (1946).

(16) (a) \\'I, v i s i , t u tliaiik L)r. I t . H. liil,lr uf our laborat,ories f u r helpful iiiucusaions concerning the ninr spectrum of these compounds. (b) The i.liemical sliift for llie i'-:+ protons in these cases is similar t o t h a t observed for w - h a l o ketoneti in u h i c h the signal for the epimer a i t h a n equatorial hy,lropen apljears at a h i g h e r tield rhan that for t h e epimer containing a n axiul Iiydrogrn: N. P. Rhacca and D. H. Williams, "Applications of N M R Specrroacopy in Organic Chemistry," Holden-Day, Inc., Ran Franriaco, Calif., 1064, 11 i R .

AXABOIJC2,~-~PITHTO;lNnRoST.4NE DERIVATIVES

Scptcmber 1966

by Fajkoss and for the 2,3-episulfides by the conversion of the ketone I I I b to the alcohol IVb. The nmr data obtained for the epoxides (I1 and VIII) and episulfides (IV and IX) were in agreement with that recently reported by Tori and ~o-workers'~ for various 2,3-epoxides and episulfides in the cholestane series. Biological Results.-The androgenic and niyotrophic activities were determined by the method of Eisenberg and Gordon1* as modified by Saunders and Drill,19 The compounds listed were given to 22-day-old castrated male rats intramuscularly by injection or orally by means of a stomach tube. The relative potencies are given in terms of per cent activity of testosterone propionate (intramuscular) or 17a-methyltestosterone (oral) and were determined from the minimal levels a t which significant increases in ventral prostate and seminal vesicle or levator ani muscles weights mere obtained. Table I1 compares the androgenic and anabolic activities for the compound evaluated in this study. TABLE I1 ANDROGENIC-IIYOTROPHIC ACTIVITIES" --Oralp----Androgenic

-Intramuscular--Androgenic*

(SV +

Compound Testosterone propionate Ib IIb IVb VIIb IXb 17m-Methyltestosterone Id IId IVd VIId IXd

VP) '2 100 0.49 2.2 1.3 42

1.4 0.62 5.6 27

Rfyotrophic M/Ad LAc 100

0.80 13 6.7 308

4.8 2.8 36 154

(SV

+

vp:2

Myotrophic LA AI/.$

1.0 1.6 5.9 5.2 7.3

3.4 4.5 6.4 5.7

I I I I

I I I

I

25

100

4.0

100

100

1.0

I

I

18

78

I

I

91

1100

4.3 12.1

Potencies are given in terms of per cent of the activity of testosterone propionate and 17a-methyltestosterone. b SV = seminal vesicles, VP = ventral prostate; since there is no uniform standard reference for androgenic designation, the values of one-half of the sum of the T'P and S T have been used as the LA = levator ani. d Myotrophic to criteria of androgenicity. androgenic ratios. e The letter I designates inactivity. a

Parenterally, the various epoxides (IIb, d and VIIb, d) (Table 11)showed very lit.tle or no anabolic or androgenic responses when compared to testost'erone propionate. The most active of these substances was VIId, the 2,3P-epoxide. Similarly, these same compounds (IIb, d and VIIb, d) demonstrat,ed very low oral activity when compared to 17a-,methyItestosterone, failing to stimulate growt,h of the sex glands. These results, however, should not be too surprising since there are few if any reports of epoxy st'eroids possessing appreciable endocrine-stimulating properties.20z21 hlthough one would expect the episulfides (IV and IX) to possess biological properties similar to the (17) (1964). (18) (1950). (19) (195i). (20) (1962). (21)

K. Tori, T. Komeno, a n d T. Nakagawa, J . Org. Chem., '29, 1136

E. Eisenberg and G. S.Gordon, J . Pharrnacol. Ezptl. Therap., 99, 38 F. J. Saunders and V. A . Drill, Proc. Soc. E s p t l . B i d . .Wed., 94, 646 R. E. Counsell and P. D. Klimstra, .I. .\fed. Pharm. Chem., 5 , 477 hl. E. Wolff, W. Ho, and R. K n o k , ibid., 7 , ,577 (1964).

695

isosteric epoxides (I1 and VII), pronounced anabolicandrogenic activity was found for some of the episulfide analogs (Table 11). For example, 2,3a-epithio17a-met hyl-5 a-androst an- 17P-01 (IXd) was found to have approximately equal androgenic and 11 times the anabolic activity of methyltestosterone after oral administration. Even IXb, lacking a l7a-methyl group, showed interesting and significant oral activity. In addition, the activity with respect to structure and configuration appeared to be quite specific for the aepisulfide isomer since the 2,3p-episulfides were much less active. The 2,3a-episulfides IXb and IXd also displayed significant activity following parenteral administration. For example, IXb was approximately three times as potent as testosterone propionate anabolically with a myotrophic/androgenic ratio of 7 . 3 . Again, there is an association of high anabolic activity with spatial configuration since the 2,3p-episulfide analogs IVb and IVd had very little parenteral anabolic or androgenic activity. The myotrophic/androgenic ratios of many of the epoxides and episulfides are also shown in Table 11. Many of the values may be misleading and possibly of limited value because of the relatively lorn order of activity of the compounds in both the anabolic and androgenic categories. However, because of the higher order of response observed for compounds IVd, IXb, and IXd, the ratios listed are useful and meaningful. For instance, the 2,3a-episulfide IXd, has one of the best myotrophic/androgenic ratios of those reported so far.22-2b It is impossible at this point to rationalize the high activity of the 2,3a-episulfide over the 2,3p isomers. From a chemial standpoint, one could expect the formation of some of the 2-dehydro analogs from the decomposition of the sulfur However, the 17P-hydroxy-2-dehydro analog (Ib) has no oral anabolic activity while the corresponding 2,3a-episulfide does produce a significant oral response. In addition, while both the 2-dehydro and 2,3a-epithio-17a-methyl analogs (Id and IXd) have similar high anabolic activity, the later substance is only about one-half to one-third as androgenic. Similarly, one might expect the much less active 2,3P-epithio-17a-niethylderivative (IVd) to also be transformed into the 2-dehydro analog ; however, one might speculate that metabolically this apparently is not the case. The intermediate thiocyano alcohols I11 and VI11 and the a-thiocyanoketo derivatives Va and Xd were evaluated for anabolic and androgenic activity and were found to be essentially inactive. While the anabolic-androgenic potencies of the epoxides I1 and VI1 are not very pronounced, the results as listed in Table I1 are of particular interest in considering the hypothesis of Wolff and co-workers2126 regarding the mode of adsorption of the steroids on receptors: "The tendency is for the asymmetrical sp2 system to afford a more active compound when hindrance is greater on the a-face, than when hindrance is greater (22) Most of t h e presently commercially available anabolic agents, and the more potent ones reported in t h e literature, have ratlos of 2-13. (23) K . Junkman and G Suchoasky, Arrnezmlttel-Forsck.. II,214 (1962) (24) F. -1.Kincl and R. I. Dorfman, Sterozds, 1, 116 (19M). (25) E F. Nutting, R. E. Counsell, P. D Klimstra, and U. Cglhaun, submitted for publication. (26) >I E. Wolff and T. Jen, J . .Wed Chem., 6, 726 (1963).

PTERIDIKECARBOXAMIDE DIURETICS.I

September 1966

17~-Methyl-3~-thiocyano-5~-androstan-l'7/3-01-2-one (X).--A solutioii of VIIId (1.0 g) in acet,oiie ( 3 5 ml) was treated dropwise with standard chromic acid solutio11.~5 The excebs reageiit wa': destroyed by a small amomit of isopropyl alcohol. The inorganic salts were removed by filtering through diatomaceous earth and the filtrate was concentrated in zlacuo, t,he residue was diluted with H20,and a crystalliiie product was collected. Recryatallization from methanol-I-120 afforded the an-thiocyanate X (0.7 g), mp 151-153°. An additional recrystallization from methanol gave an analytical sample: mp 15:3-154'; [ , ] ~ S D S13i.5'; x,,~ 292 mp ( e 60); rimr, 232.5-239.5 ( 3 p - R ) , 7 3 (C-17 methyl), 50.5 (C-18 methyl), 49 cps (C-19 methyl). 2~-Thiocyano-5~-androstane-38,17/3-diol (XI).-To a n ice-cold solution of Va (4.0 g) in T H F (100 ml) wvas added a cold solutioii

697

of lithium tri-~-butoxyalumiiiumhydride (20 g ) in T H F (100 ml). ~ for 1 hr a t about 5' and poured into an The reactioii w a stirred ice-cold 5 5 AcOII solution. The proditct was extracted with ether aiid the extracts were washed with HZO, 5% PiaHCOa, and finally H Y Oagain before drying over anhydrous Na2S04. Solvent removal in vacuo left a white solid which was recrystallized from acetone-hexane to give the diol X I (3.6 g), mp 202-205", [ C I ] ~ ~ D +23". A n d . Calcd for. CPOHSINO2S:C, 68.7%; H, 8.94. Found: C, 68.57: I€, 9.23.

Acknowledgment.--We wish to thank Dr. F. B. Coltori for his interest and comments during the course of t,his work.

Pteridinecarboxamide Diuretics. I. Reaction of 4,6-Diamino-5-nitrosopyrimidineswith Substituted Malonamides' T. S. OSDESE,ARTHURA. SANTILLI,~ LEE E. NCCARDLE, ASD MARVIX E. ROYENTHALE Wytth Laboratories, Inc., Research and Development DivisLon, Raclnor, Pennsylvania

Receizled JIarth 12, 1966 The reaction of 4,6-diamino-S-riitrosop~rimidiiieswith a iiumber of ?i,S'-bis-substituted maloriamides in the presence of 1 equiv of sodium iii ethanol afforded mixtures of 4-amino-7-substituted amino-N-substituted 6pteridinecarboxamides and 4-amino-i-hydroxy-N-substituted 6-pteridinecarboxamides. The 7-substituted sminopteridinecarboxamides were found to be effective oral diuretics in rats, whereas the i-hydroxypteridinecarboxamides were inactive at comparable dose levels.

The base-catalyzed reaction of 4,6-dianiino-5-nitroso2-phenylpyrimidine (I) with cyanoacetaniide yields 4,7-diamino-2-phe1iyl-G-pteridinecarboxamide (la) (see Scheme I).3 Variation of substituents on the 2 position SCHENEI

I

NH?

I11

NH2

Ia XCCH C O N I XiOEt

I

", I

I1

of the pyrimidine, as well as substitution on the amide nit'rogen of the cyanoacetamide, permits the preparation of many biologically active substituted 6-pteridinecarboxaniides. These have been the subject of several recent patent'si I n each of these react'ions, ring closure to the pteridine results froiii the elimination of 1 equiv of water between the nitroso group of the pyrimidine and the active methylene group of the cyanoacetamide and the coricomit'ant addition of the amino group of the pyrimidine to the nitrile group of the cyanoacetamide. The reaction of I with diethyl malonate in the presence of 1 equiv of sodium in et'hanol affords ethyl 4aniino-7-hydroxy-2-phenyl-6-pteridinecarboxylate (11). In this reaction, ring closure occurs with elimination of mater and ethanol. Other exampIes of pteridine formation by reactions involving 4-amino-5-nitrosopyrirnidines have been described by Pachter and co-workers.j h recent review has appeared 011 biologically act,ive pteridines derived froni 4-aniiilo-~-riit,rosopyridine~.~ In view of what has been reported concerning these types of reactions an attempt was made to prepare 4-amino -7 - hydroxy-2 -phenyl-6-pteridinecarboxaniide (Ha) in a single step by the reaction of I with malonamide in the presence of an equivalent amount of sodium ethoxide. It was expected t,hat water and ammonia would be eliiiiinated in the reaction, thus (1) A preliminary account of this work u-as presented before the Division of Medicinal Chemistry, 150th National Meeting of t h e American Chemical Society, Atlantic City, N , Sept 1965, Abstracts, p 16. ( 2 ) T o whom inquiries regarding this article should be sent. (3) T. S. Osdene and E. C. Taylor, U. S. Patent 2,975,180 (IQ61). (4) J. Weinstock, U. S.Patent 2.963.478 (1960); T. Y. Osdene and A . .%. Santilli, U. S.Patents 3,138,595,3,138,591, 3,138,592, and 3,122,547 (1964). (5) I. J . Pacliter and P. E. Nemetli, J . Org. Chem., 28, 1187 (1963); I. J. Pacllter, P. E. Nemeth, and A . J. Villani, t h i d . , 28, l l Q i(1963). ( 6 ) T. S.Osdene in "Pteridine Chemistry." \V. Pfieiderer and E. C. Taylor, Ed., The Macmillan Co., S e a York, h-.I-., 1964, p 6 5 .