Reaction of Peracids with Δ20 (21)-Steroid Enol Acetates

Robert Bruce Moffett, and George Slomp Jr. J. Am. Chem. Soc. , 1954, 76 (14), pp 3678–3682. DOI: 10.1021/ja01643a024. Publication Date: July 1954...
1 downloads 0 Views 600KB Size
3678

ROBERT BRUCEMOFFETT AND GEORGE SLOMP, JR. [CONTRIBUTION FROM

THE

Vol. 76

RESEARCH LABORATORIES O F THEUPJOHN COMPANY]

Reaction of Peracids with A20(21)-Steroid Enol Acetates BY ROBERT BRUCEMOFFETTAND GEORGE SLOMP, JR. RECEIVED JANUARY 28, 1954 A study has been made of the action of perbenzoic (or peracetic) acid on A20(*1)-steroidenol acetates. The AZ0(*l)-enol acetate from pregnenolone (I) by treatment with perbenzoic acid followed by hydrolysis and reacetylation gave the (5,6)01oxido-21-acetoxy derivative (11). The Aa,s,20(21)-dienol acetate (111) from progesterone gave preferentially oxygenation in steroids gives preferentially 16,17-oxides with the the 6 - p position. Addition of peracid t o 20-acetoxy-A1e~*7~~20~z1~-diene A20(21)-eno1 acetate groupings still intact. These were converted t o the corresponding (16,17)-oxido-20-keto-21-acetoxy steroids by addition of bromine followed by treatment with potassium acetate. Under more strenuous conditions both the 16(17) and 20(21) double bonds could be epoxidized with perbenzoic acid.

The preparation of steroid enol acetates by the acid-catalyzed reaction of isopropenyl acetate with 20-keto steroids has been described in a previous article' from these laboratories. Vanderhaeghe, Katzenellenbogen, Dobriner and Gallagher2 have prepared the enol acetate 3P,20diacetoxy-20-allopregnene by a similar procedure and converted it to the 21-acetoxy compound, 3p,21-diacetoxyallopregnan-20-one, by means of perbenzoic acid followed by hydrolysis and reacetylation. We independently carried out this series of reactions in connection with a study of steroid A20(21)-enol acetates. In a similar way the enol acetate from pregnenolonel*z(I) was converted by way of the diepoxide to the known 3P-21-diacetoxy- (5,6)a-oxidopregnan-20-one (11).

1, 2 molar equivalents of CsHjCOOOH 2, S a O H

CH2

II

C-OAC

-

1, CsHsCOOOH 2, NaOH

CH~OAC

3, A c ~ O

C=O

1

*cO

I

Lid ,,

0

1, 1 molar equivalent of C6HECOOOH 2, IiaOH

J. CHs

I1

I c=o

A study was then made of the reaction of perI benzoic acid with 3,20-diacetoxy-3,5,20-pregnatriene (111, the dienol acetate of progesterone). An excess of perbenzoic acid was added a t 0" to a solution of this enol acetate and aliquots were titrated a t intervals with sodium thiosulfate. The results are expressed in Fig. 1 and indicate that one molar equivalent of perbenzoic acid reacted in about seven 0 IV OH V minutes while two molar equivalents had not been completely absorbed after four hours.4 In order to lent of perbenzoic acid a t 0". After one hour the determine which double bond reacted first, epoxi- mixtures were treated with dilute sodium hydroxide dations were carried out using one molar equiva- and worked up giving two products. One was the known6 allopregnane-3,6,20-trione(IV), and the (1) R. B . Mo5ett and D. I. Weisblat, THISJOURNAL, 74, 2183 other was shown to be 6-hydroxyprogesterone (V). (1952). (2) H. Vanderhaeghe, E. R. Katzenellenbogen, K. Dobriner and 6-Hydroxy-Ai4-3-ketonesof this type are known to T. F. Gallagher, zbid., 74, 2810 (1952). rearrange very easily to the saturated 3,6-dike(3) L. Ruzicka, PI. A. Plattner, H. Henner and 0. Ernot. Helw Chim. A d a , 29, 248 (1946). (4) A kinetic study was made frem these data by Dr. Edward Garr e t t of our Department of Physics and will be published separately

(6) (n) M. Ehrenstein and T. 0 . Stevens, J . &E. C k e m . , 5 , 318 (lQ40); (b) S . Lieberman, K,Dobriner, B. R. Hill, L . F . Pieser and C P Rhoads, J Biol. Chem., 171,263 11948)

PERACLDS WITH AN(2 1 ) - S ENOL ~ ~ACETATES ~ ~ ~

July 20, 1954

3679

was established by analysis and infrared absorption spectrum which showed the band a t 1660 cm.-' characteristic of A20(21)-enol acetate.'S2 This was confirmed by hydrolysis and reacetylation to the '3 1 known 3~-acetoxy-(16,17)-a-oxidoallopregnan-20onell (XI). By the addition of bromine to the enol acetate X , the 21-bromo compound XI1 was ob, tained. This was converted to 3@,21-diacetoxy(16,17)a-oxidoallopregnan-20-one (XIII) by the 0 20 40 60 120 180 240 action of sodium iodide followed by potassium aceTime (minutes). tate. This appeared to be identical with that preFig 1.-Reaction of perbenzoic acid with 3,20-diacetoxy- pared previously by Plattner, et aZ.,l2by another 3,5,20-pregnatriene (111) method. When the enol acetate IX was allowed to stand with perbenzoic acid a t room temperature tones.6 The structure of V was established by for five days, two molar equivalents added and the analysis, spectra and oxidation to the known 6-ke- diepoxide XIV was isolated. toprogesterone (VI) The preparaCHi tion of both the 6a- and Gp-hydroxyprogesterone by Balant and Ehrenstein8 confirmed the 6-hydroxy structure of our compound and moreover showed i t to be the 60-isomer. When w ,O slightly over two molar equivalents C6HjCOOOI-I of perbenzoic acid was allowed to stand for 5 days a t room temperature ).,J H with the enol acetate I11 and the O 0 , 20 hr 'XIV IX product was hydrolyzed, two other products were isolated. Analyses CH~OA~ CH2Br CH2 CHI and spectra showed them to be a I I I I/ c=o C-OAC saturated monohydroxy triketone and (f=O C=O a dihydroxy diketone with a,P-unI;,? saturation. This indicated that they \I 1, NaOH \&' might be the 21-hydroxy analogs of 2, KOAc fI IV and V. However, the reporting of these structures by Herz'lg and 'xIII XI1 x XI Ehrensteina showed this was not the I n a similar way, the action of perbenzoic acid case. It was possible that these might be the XV was 17a-hydroxy analogs, VI1 and VIII, formed by on 3P,20-diacetoxy-5,16,2O-pregnatriene the rearrangement of the 20-21 double bond to studied. Titration of aliquots of the reaction mixthe 17(20)-position prior to the addition of per- ture gave the curve shown in Fig. 2 which indicates benzoic acid. This rearrangement has previously that a t 25' two molar equivalents reacted within been demonstrated.a The 17a-hydroxy structures one hour while the third required about 30 hours to were indeed found to be correct by comparison of one of them, the a,p-unsaturated ketone (VIII), with 6,9,17a-dihydroxyprogesterone which Meister, et a1.,'0 obtained by biological oxygenation of 17ahydroxyprogesterone. Since the products (VI1 and VIII) were isolated in only small yields it is highly probable that the expected 21-hydroxy analogs were also present in the reaction mixture. A study was made of the action of perbenzoic II acid on 3~,20-diacetoxy-16,20-allopregnadiene1 (IX) c in the hope that oxygen could be introduced in both the 17- and 21-positions. Even though an excess .3 of perbenzoic acid was allowed to stand with this , , diene a t 0" for 23 hours, titration of an aliquot indiI" cated that only one molar equivalent had reacted. 0 5 10 15 20 25 30 The remainder of the reaction mixture afforded a Tlme (hours). good yield of 3/3,20-diacetoxy-(16,17)a-oxido-20Fig. 2.-Reaction of 3,S,20-diacetoxy-5,16,2O-pregnaallopregnene (X). The structure of this compound

e

1

,'

11

,4 I

,

q-

$ I , ,

(6) M. Ehrenstein. J . Org. Chcm., 13, 214 (1948). (7) (a) M . Ehrenstein, ibid., 4 , 506 (1939); (b) R.B.Moffett, J. E. Stafford, J. Linsk and W. M . Hoehn, THISJOURNAL, 68, 1857 (1946). (8) C.P. Balant and M. Ehrenstein, J . Org. Chem., 17, 1587 (1952). (9) P.T. Herzig and M. Ehrenstein, i b i d . , 16, 1050 (1951). (IO) P. D. Mdster, D. H. Peterson, H. C. Murray, 0.B. Spero, S . H. Eppstein, A. Weintraub, L. M . Reineke and H. M . Leigh, THIS J O U R N A L , 7S, 416 (1953).

,

1

triene ( X V ) with first 3.26 molar equivalents, then 6.53 molar equivalents of perbenzoic acid. (11) P.A. Plattner, L. Ruzicka, H. Heusser and E . Angliker, Hclo. Chim. Acts, 30, 385 (1947); P1. A. Plattner, H. Heusser and M . Fluser, ibid., 31, 2210 (1948). (12) P1. A. Plattner, L. Ruzicka, H . Heusser and E. Angliker, ibid., 30, 395 (1947).

3680

ROBERTBRUCEMOFFETT AND GEORGE SLOMP, O