Steroidal Sapogenins. XLVIII. Four Routes to Cortisone Acetate from

May 1, 2002 - Four Routes to Cortisone Acetate from Gentrogenin Acetate2a,b. Edward S. Rothman, and Monroe E. Wall. J. Am. Chem. Soc. , 1959, 81 (2), ...
2 downloads 0 Views 698KB Size
Jan. 20, 1959

CORTISONE ACETATE FROM GENTROGENIN ACETATE

411

[ COKTRIBUTION FROM THE EASTERN REGIOXAL RESEARCH LABORATORY1]

Steroidal Sapogenins. XLVIII. Four Routes to Cortisone Acetate from Gentrogenin AcetateZash BY EDWARD S. ROTHMAN AND MONROE E. WALL RECEIVED MAY19, 1958 Side chain degradation of 11-keto diosgenin acetate (11) gave the key intermediate 3P-acetoxy-5,16-pregnadiene-11,20dione (111), which was converted by conventional methods to 21-acetox)--3~,17or-dihydroxy-~-pregnen-ll,2O-dione (VI). T h e latter compound after oxidation with chromium trioxide in acetone gave the As-3-ketone isomer of cortisone acetate which on equilibration with acid or base catalyst isomerized to cortisone acetate. Several alternate routes to cortisone acetate are also presented.

For several years we have been carrying out paper we wish to present laboratory scale results researches in the field of steroidal sapogenins, showing two completed conversions of gentrogenin particularly the C-ring oxygenated sapogenins to cortisone and two formal syntheses carried t o hecogenin and the more recently discovered gentro- the 1la-hydroxyprogesterone and 16a,17a-epoxygenin, with the view in mind of testing their po- 11a-hydroxyprogesterone stage. tential usefulness for cortisone synthesis. I n preThe conversion of gentrogenin acetate (I) t o vious papers we have described the d i s ~ o v e r y , ~11-ketodiosgenin acetate (11) was effected in six isolation and properties of gentrogenin, certain of steps in 35 to 48% yield as previously describede6 its reactions and degradation products5 and the The subsequent standard method side chain degradation? of the 11-keto sapogenin (11) required H H four further steps and gave a 55% yield of 3Pacetoxy-5,16-pregnadiene-l1,20-dione (111) (19.3% yield from I). Introduction of the 17a-hydroxyl group by Julian's method of hydrogen peroxide epoxidation, hydrogen bromide oxirane opening and Raney nickel dehalogenation8 led to a 69% yield (11% from I) of 3/3,17a-dihydroxy-5-pregnene-11,20-dione (V) . CH

cd

Y-

CHzOAc

ncoo

CHtOAc OVERALL Y I E L D S FROM COMPOUND

1

0

nn

SHEET

&-

-

I,

conversion of gentrogenin and its C-25 epimer to 11-ketodiosgenin and 11-ketoyamogenin.6 In this

HCOO

CO &H r

HCOO

Ix

-

o&:cc

Pm FLOW

U&

CHzOAC

Vlll

m

XI

Utilization Research and Development Division, FLOW SHEET2 Agricultural Research Service, U. S. Department of Agriculture. ArThis compound was converted to cortisone aceticle not copyrighted. (2) (a) Presented a t the 134th National Meeting of the American tate by two independent routes. The method we Chemical Society, Chicago, Ill., September 7-12, 1958; (b) paper flrst tried (see flow sheet 2) paralleled the experiXLVII, Kenney, et al., THISJ O U R N A L , 80, 5568 (1958). in the preparation of (3) C. Djerassi, H. J. Ringold and G. Rosenkranz, i t i d . , 73, ments of Ringold, et Reichstein's Substance S. This procedure in5513 (1951), have reported a cortisone synthesis from hecogenin by a route whose key step involved bismuth oxide oxidation of a C-ring volved formylation of the 3P-hydroxyl group, ketol. More recently significant improvements in the utilization of introduction of the 2 1-acetoxyl group by brominahecogenin have been made by the Glaxo group. Cf.J. Elks, G. H . Phillipps, T. Walker and I.. J. Wyman, J. C h e w Soc., 4330 (1956); tive procedures followed by the protective acetylation of the 17a-hydroxyl group. The 17aJ. H.Chapman, J. Elks, G. H. Phillips and L. J. Wyman, ibid., 4344 (105G); R . M . Evans, J. C. Hamlet, J . S. H u n t , P. G. Jones, A. G. hydroxyl acetylation reaction was necessary in Long, J. F. Oughton, L. Stephanson, T. Walker and B. M. Wilson, order to stabilize the side chain and prevent Dibid., 4366 (1956). homo rearrangement during Oppenauer oxida(4) H. A. Walens, S. Serota and M. E. Wall, THIS JOURNAL, 77, tion of the 3-formate-5-ene system. I n our hands 5196 (1955): J . Org. Chem., 22, 182 (1957); M.E. Wall, J . J . Willaman, T. Perlstein, D. S. Correll and H. S. Gentry, J . A m . Pharm., this 17a-acetylation step worked badly and gave Sci. Ed., 46, 155 (1957). rise to much destruction of material and produc(5) E. S. Rothman and M. E. Wall, J . Org. Chem., 22, 223 (1957). (6) (a) E. S. Rothman and M. E. Wall, THISJOURNAL, 79, 3228 (?) (a) R . E. Marker, el al., iizter alia, ibid., 62, 3350 (1940); (h) (1957). (h) The step of conversion of 5,6,11,23?-tetrabromohecogenin M.E. Wall, H. E. Kenney and E. S. Rothman, ibid., 77,5665 (1955); acetate to 11,23?-dihromogentrogenin acetate has been improved (c) hl. E. Wall and S. Serota, ibid., 79, 6481 (1957). over that reported in reference 6a to give an 86% yield by modifying (8) P. L. Julian, E. W. Meyer, W. J. Karpel and I. R. Waller, i b i d . , the work-up procedure. The cooled reaction mixture is diluted with an 72, 5145 (1950). equal volume of water and the crystalline product is directly filtered (9) 1%.J . Ringold, G. Rosenkranz and F. Sondheimer. i b i d . , 78, 820 off and warhcd wit11 dilute sodium iodide aud water. (19X).

(1) Eastern

UL.,~

412

EDWARD S. ROTHMAN AND MONROE E. WALL

Vol. 81

tion of highly colored products seriously affecting I n one such conversion, 3,R-hydroxy-5,16-pregthe eventual yield of cortisone, nadiene-1lJ20-dione (IIIb), n1.p. 224-228', aZ% Our improved route from 3P,17a-dihydroxy-5- -9' (flow sheet 3) was converted to the 20pregnene- 11,20-dione (V), (see flow sheet 1) dioxolane XIII. The 20-dioxolane XI11 was reavoided this 17a-hydroxyl acetylation step and duced with sodium in ethanol to form 3/3,1lcugave cortisone acetate in five steps in 35-5570 dihydroxy-5,16-pregnadiene-20-one 20-dio~olaiie'~ yield (5 to 7Yo yield from I for nineteen re- (XIV). action steps.) The al-acetoxyl group was introduced as before by brominative procedures starting with the free 3P-hydroxyl compound V to give 21-acetoxy-3j3,17a-dihydroxy-5-pregnenell,?O-dione (VI) in 55y0 yield for the three steps no HO for bromination, reaction with sodium iodide ma xm XE and metathetical acetoxylation. It was observed, after addition of the first molar equivalent of bromine, that infrared unsaturation bands near 830 ern.-* survived. This observation suggests that substitutive bromination a t C-21 occurred, a t least in part, prior to addition a t no C5-C6. These bands disappeared completely after XY XPI xpu addition of the second equivalent of bromine. FLOW SHEET 3 I n the preceding route where the corresponding 3formate (IX), aide supra, was treated with a single Removal of the 20-dioxolane group of XIV molecular equivalent of bromine, isolation a t produced 30,lla-dihydroxy-5,16-pregnadiene-20a later stage of 21-acetoxylated material (X) again one (XV) . I 7 Palladium-catalyzed hydrogenation indicated that the substitution reaction occurred, of XV to form the monoketone XVI followed by a t least in part, preferentially over the addition Oppenauer oxidation gave 11a-hydroxyprogesterraection. one (XVII). This compound has been previously Djerassi, Engle and Bowers'O had shown that the obtained by direct microbiological oxidation of use of chromium trioxide-acetone-sulfuric acid progesterone12a or by chemical methods13 and reagent" converted A5-3p-hydroxy steroids to subsequently converted to cortisone. 14a.b I n a Aj-3-ketones. These workers had not submitted second such conversion (see flow sheet 4) 3 p , l l a cortical side chain steroids to oxidative attack by dihydroxy-5,lG-pregnadiene-20-one (XV) was epthis reagent, but we found that cortisone acetate itself was completely unaffected when so treated. We did not observe any D-homo rearrangement, H O .. HO., oxidative side chain fission or 21-aldehyde formation. On this basis we subjected 21-acetoxy-3/3,HO 17a-dihydroxy-5-pregnene-l1,2O-dione (VI) to the mm XLX chromium trioxide-acetone reagent and obtained a very high yield of a cortisone acetate unsaturaFLOW SHEET 4 tion isomer, namely 21-acetoxy-17a-hydroxy-5pregnene-3,11,20-trione (VII) . Isomerization of oxidized with alkaline hydrogen peroxide to forin 3 P , l l a - dihydroxy - 16a,17a- epoxy - 5 - pregnenethis compound with a trace of ammonium hydroxide 20-one (XVIII). Oppenaber oxidation of XVIII in methanol caused a shift of the double bond into gave 16a,l7a-epoxy-l1a-hydroxy-4-pregnene-3,20conjugation with the 3-keto group to produce dione (XIX) ("lla-hydroxy-16a,17a-oxidoprogescortisone acetate. terone"), identical with the material produced by I n addition to the above syntheses of cortisone microbiological hydroxylation of 16a,17a-epoxyfrom gentrogenin acetate, we have prepared from 4-pregnene-3,20-di0ne~~b and whose conversion this source two 11a-hydroxylated derivatives of to cortisone was described by Ercoli and Kugprogesterone. These latter compounds had pre- gieri.15 viously been prepared from C-ring desoxy steroids traub, L. M. Reineke, S. H. Eppstein, H. C. Murray and H . hI. Leigh by microbiologica112a.bor chemicalla procedures Osborn, ibid., 77,4428 (1955). (13) 0.Mancera, J. Romo, F. Sondheimer, G. Rosenkrauz and C . and subsequently converted to cortisone. 14a,b~15

&-

-

___)

=-W-oW

( I O ) C. Djerassi, R. R. Engle and A. Bowers, J . Org. Chem., 21, 1647 (1956). (11) This reagent w a s first used by the Manchester group for acetylenic carbinol and eburicoic acid oxidation studies. See, for example, R. Boaden, I. M. Heilbron, E. R . H . Jones and B. C. L. Weedon, J . Chem. SOC.,39 (1946); P. Bladon, J. M. Fabian, H. B. Henbest, H . P. Koch and G. W. Wood, i b i d . , 2402 (1951): R. G. Curtis, 1. Heilbron, E . R . H. Jones and G. F. Woods, ibid., 457 (1953); A. Bowers, T. G. Halsall, E. R . H . Jones and A. J. Lemin, i b i d . , 2548 (1053); T. G. Halsall, R . Hodges and E. R . H. Jones, ibid., 3019

(1953). (13) (a) D. H. Peterson, II. C. Murray. S. 11. Eppstein, 1,. X I . Reinrke, A . Wdntraub, P. D . Mcistrr and H . A I . I&h, ' h 3 JOURN m , , 74, 60:%3(1952); (b) I). H . Pctcrson, P. D. hlcister, A. Wriu-

Djerassi, J. Org. Chcm., 11, 1066 (1962). (14) (a) 0. Mancera, H. J , Rinyold, C. Djerassi, G.Rosenkranz and F. Sondheimer, THISJOURNAL. 75, 1286 (1953); (b) J. A. Hogg, P. F. Beal, A. H . Nathan, F. H. I,incoln, W. P. Schneider, B. J. Magerlein, A. R . Hanze end R . W.Jackson, ibid., 7 7 , 4436 (1955). (15) A. Ercoli and P. de Ruggieri, G a m chim. i f a l . , 85, 628, 1304

(19553. (16) R. Antonucci, S. Bernstein, M. Heller, R . Lenhard, R . Littell and J. H. Williams, J . Org. Chcm., 18,7 0 (1953),used similsr dioxolane protection of carbonyl groups to transform the 11-keto group of cortisone alcohol t o Kendall's Compound F by means of lithium aluminum hydride reduction. (17) The diacetate of this compound was identical with the product obtained by direct degradation of 11 a-hydroxydiosgenin or 1 1 ahydroxyyainownin.

Jan. 20,1959

CORTISOKE

ACETATEFROM GENTROGENIN ACET1ITE

413

two hours. The solution was poured into a large volume of Experimental water containing sodium chloride and sodium formate Melting points were determined on the Kofler block but cold and the precipitated solids were filtered off and washed are otherwise not corrected. Optical rotations were deterwith water. Recrystallization from methanol gave 3pmined in chloroform, unless otherwise noted, in a 2-dm. tube formoxy-17a-hydroxy-5-pregnene-l1,20-dione( I X ) , as thick using a solution of 25 mg. of steroid in 1.5 ml. of solvent. scales. The compound melted from 255 t o 260", after Infrared measurements were determined with a Perkintransition on the Kofler block to overlapping thin scales Elmer model 21 instrument using 1.5 t o 2.5 mg. of steroid in covering the entire field. The yield was 80%. The infra0.35 g. of KBr; 10 mg. of steroid in 1 ml. of CSz, 1-mm. red spectrum showed (KBr disk) bands a t 3440 (17a-hycell; or 25 mg. of steroid in 1 ml. of CHC13, 0.5-mm. cell. droxyl), 1693 (C-20-bonded carbonyl), 1710 (ll-carbonyl), Degradation of 11-Keto diosgenin Acetate; 3P-Acetoxy1'725 (formate ester carbonyl), 1180 (formate ester) and 813 5,16-pregnadiene-l1 ,20-dione (III).-In a 250-ml. standard- cm.-1 (C-5 olefin). taper, round-bottomed flask fitted with a wired-on gas2 l-Acetoxy-3~-formoxy-17~-hydroxy-5-pregnene-11 ,20-ditight ground glass stopper7owere placed 14.5 p. of I, 37 ml. one (X).-The formate I S (1.253 g., 0.00334 mole) was of acetic anhydride and a trace of acetic acid. The flask dissolved in 25 ml. of dry methylene chloride and the solucontents after heating for four hours a t 180' were cooled, tion was treated with a solution of bromine (0.00334 mole) in diluted with 74 ml. of acetic acid and 6 ml. of water. There13.32 ml. of methylene chloride. The volume was reduced after the conditions of procedure A, ref. 7b, were followed by distillation i n vacuo a t 30'; the solution was washed with to give 111, 6.16 g., 4970, m.p. 183-186", needles from hex1% sodium hydrogen carbonate solution; and the solvents ane transforming t o spicules on heating, [ a I z 5 ~ -1.7'; removed in vucuo. The residue was dissolved in 25 ml. of 234 mp, log e 3.93. The mother liquors from this acetone, and 2 g. of sodium iodide was added. The mixpreparation were also used since they gave an additional 6y0 ture was let stand overnight and was then poured into a cold of crystalline 16a,l7a-epoxide I V when treated as below. solution of 2 g. of sodium thiosulfate in 70 ml. of water. T h e total yield was therefore considered to be 55'%. The steroid was filtered off and dried in vacuo a t room temAnal. Calcd. for C28H300d: C, 74.56; H , 8.16. Found: perature. The crude iodo compound was dissolved in 25 C , 74.36; H , 8.33. ml. of dry acetone, 2.5 g. of potassium acetate was added and the mixture was heated under reflux and with stirring 16a, 17a-Epoxy-3P-hydroxy-5-pregnene-11 ,20-dione (IV). for 18 hours. The reaction mixture was cooled, poured --4 5.95-g. sample of the preceding preparation I11 was disinto water and the steroid was isolated by extraction with solved in 800 ml. of methanol a t lo",5 ml. of 30% hydrogen ether. Evaporation of the ether i n uacuo gave a colorless peroxide and 2.3 ml. of 4 A' sodium hydroxide were added solid foam. Chromatography of this material with chloroand the mixture was stored overnight. A crystalline waterform elution gave only 540 nig. of the desired crlstalline 21insoluble precipitate formed and was filtered off. The filtrate was diluted with a n equal volume of water and a new crop acetoxy-3P-formoxp- 17a- hydroxy- 5-pregnene- 11,20-dione (X), m.p. 192.5-193.5', red melt, [ a ] * ;+20.8", ~ but the of crystals formed and were filtered off. The filtrate was colorless glassy residues eluted just after the crystalline maneutralized with hydrochloric acid and was well extracted with methylene chloride which was evaporated to give a terial gave 100 mg. of the 17u-acetoxyl derivative XI when glassy residue. To simplify the isolation procedures all treated as below. solids were combined and converted to acetates by disAnal. Calcd. for C~4H3201: C, 66.65; H , i .45. Found: solving in 30 ml. of pyridine and 10 nil. of acetic anhydride C, 66.41; H , 7.44. and heating 0.75 hour on the steam-bath. On dilution with 17a,2 1-Diacetoxy-38-fonnoxy-5-pregnene-l l ,ZO-dione water and isolation by ether extraction, 5.1 g. of ethanol(XI).-The 17a-01 X, 99 mg., in 0.71 ml. of acetic anhydride recrystallized product was obtained (84y0 yield). The product was freed of 100 nig. of a n unidentified extremely was treated with 31.4 mg. of p-toluenesulfonic acid a t room temperature overnight. On decomposing with water an insoluble material (in .p. 318') by solution in ethyl acetate, filtering and crystallizing by evaporation to drj-ness. The amorphous solid was obtained which on chromatography on Florisil gave 80 mg. of crvstallizable material. The earlier product, 3~-acetoxy-16a,17a-epi~xy-5-pregnene-l1,2~~-di~~ne, benzene-eluted 17a,21-diacetoxy-3P-formoxy-5-pregnenemelted from 205-206' forming dense polyhedra undergoing 11,20-dioiie (XI), [ a ]z 5 ~+40° melts with decomposition transition on the Kofler block t o refractile, tetragonal from 228 to 22Q0,occasionally from 233 to 235', and shows prisms, [ a I z 5 D f14.8. no hydroxyl band in the infrared spectrum, a broad multiple Anal. Calcd. for C~H3005: C , 71.48; H, 7.82. Found: ketonic band and a series of ester bands a t 1237, 1250 and C, 71.46; H , 8.08. 1260 cm.-l. Saponification of the intermediate acetate with 5y0methAnal. Calcd. for ClsHdaOs: C , GS.80; H, 7.22. Found: anolic potassium hydroxide gave 4.46 g. of 16u,lTa-epuxyC , 65.50; H , 7.21. 3P-hydroxy-5-pregnene-ll,20-dione (IY), m.p. 111O to The chloroform eluate gave 17cu,21-diacetoxy-3~-hydrox~~about 114' when crystallized from acetone or ethanol. Per5-pregnene-11,20-dione, m . p . %03-205", [ a ]*$D +83" s h o ~ haps the molecule thus crystallized is solvated but the lower ing infrared hydroxvl bands a t 3410 and 3500 cm.-l ( I i B r indicated melting temperature is characteristic. disk), broad ketone bands centered a t 1705 and 1731 cm.-l Anal. Calcd. for C21H2804: C, 73.22; H, 8.19. Found: and a strong, sharp ester band centered a t 1258 cm.-r with a C, 73.98; IT, 8.01. small band a t 1245 cm.-I To augment the yield of cortisone 3~,17~-Dihydroxy-5-pregene-11 ,20-dione (V).-To a so- diacetate non-crj-stallizable mother liquor residues were lution of 2.22 g. of 16a,l7a-epoxy-5-pregnene-ll,20-dione saved and carried through the Oppenauer oxidation reaction ( I V ) in 22 nil. of acetic acid was added 4.3 ml. of a 40% indicated in part b of the cortisone diacetate preparation. solution of hydrogen bromide in acetic acid. After standing Anal. C a h d . for C25H340,: C, 67.24; H, 7.68. Found: for 25 minutes the reaction mixture mas poured into 140 ml. C, 67.02; H , t .BO. of cold water and the precipitated bromohydrin was filtered Cortisone 17~,21-Diacetate and Cortisone.-(a) Two off and air-dried. Ten grams of Raney nickel was heated hundred milligranis of crystalline, 17a,2 1-diacetoxy-3p-formwith 100 ml. of refluxing acetone. Ten ml. of water, 2 ml. oxy-5-pregnene-ll,20-dione( X I ) in 70 ml. of dry toluene, of acetic acid and the crude brornohydrin were added, and stirring and heating under reflux were maintained for 4 200 mg. of aluminum isopropoxide and 2 rnl. of cyclohexanone were refluxed for X5 minutes. \\.ater, 100 ml., w a s hours. The acetone solution was decanted and evaporated added and the two-phase mixture was boiled until all organic to dryness in vucuo. Crystallization of the residue from methylene chloride and from ethyl acetate gave 1.57 g. solvents had been carried off in the vapors. The residue was collected as a pasty mass by decantation of the water. Crys(71yG yield) of 3P,17a-dihydroxy-5-pregnene-ll,20-dione tallization from methanol gave a is70 yield of cortisone (V). The compound formed dense polyhedral crystals, 17a,21-diaceta te (XI1). r1i.p. 277-278" after transition over 220' to lance forms ( b ) Eight hundred tnilligranis of thc uncrystallized mother [a]"D + G o (hIeOH). liquor residue from the :thove preparation consisting esseiitiully c f :i mixture of the 3-liydroxy und 3-forrnoxyl conipounds W d S dissolved in 25 ml. of dry toluene. Eight ml. of 3P-Formoxy-l7~-hydroxy-5-pregnene-l1 ,20-dione (IX).cyclohexanone and 840 mg. of aluminum isopropoxide were The diol \' of the preceding preparation, 1.45 g . , \vasodis- added and the mixture l r a s refluxed for 0.75 hour, cooled and solved in 35 nil. of 90yo formic acid and heated to 60 for water was added. The contents of the flask were heated

414

EDWARD S. ROTHMAN A N D MONROE E. ITALL

1701. 81

until organic solvents were lost, and the gummy solid was 3p,1 l~~Dihydroxy-5,17-pregnadiene-2O-one (XV).--=1 collected by extraction with ether. Evaporation of the sample of X V was prepared from 3P-hydroxy-3,16-pregtiadiether in vucuo and crystallization of the residue from methene-11,20-dione ( I I I b ) using the general dioxolanation proanol gave 400 mg. of cortisone 17a,21-diacetate ( X I I ) , m.p. cedure20 of Antonucci, et el. The dione I I I b , 1 g., 70 nil. of 220-221°,"",::A: 238 nip, c 15,000, identical with an authentic ethylene glycol, 35 i d . of toluene and 35 mg. of p-toluenespecimen. Hydrolysis of cortisone diacetate in 5yosodium sulfonic acid were refluxed for 4 hours under the usual Lvatcrhydroxide in methanol gave an 807, yield of cortisone, iden- removal conditions. The resulting solution was cooled and tical with a n authentic specimen.18 the mixture was diluted with dilute aqueous sodium bicar21-Acetoxy-3P, 17a-dihydroxy-5-pregnene-11,20-dione bonate. The steroid was isolated with ether and solvents (VI).-Five grams of 3P,l7a-dihydroxy-5-pregnene-l1,20-were removed in z'acz~o. The residue w i s dissolved in 301) dioiie (1;) vias dispersed in 30U ml. of chloroform and, during nil. of 1-propanol and 11 g. of sodium i n s added to the the course of 1.5 hours, one molar equivalent of bromine in boiling solution.21 The reduction product was isolated with 61.35 nil. of carbon tetrachloride was added. Xt this point ether and the protectire dioxolane group was removed by the entire quantity of steroid had passed into soluticn. It heating on the steam-bath in 1: 1 aqueous acetic acid soluwas sometimes necessary to add a little more chloroform in tion. Seedles of tlic prciduct separated readily from the reactions where reprecipitation occurred or where turbidity hydrolysis medium even before cooling. The prciduct XI-, developed. X second molar equivalent of bromine was recrystallized from ether and from ethyl acetate-n-licxnnc formed wedges with transition on the Kofler stage to forms added although it was taken up with some reluctance. ;It the end of the bromination solvents were removed in uacuo. with a rhombic profile, m . p . 210-218', 238 nip, e The residue was dissolved in a liter of acetone and treated 8,230, [ a I 2 j -14.8'. ~ with 20 g. of sodium iodide with stirring a t room teniperaAnal. Calcd. for CIIH, 0,: C, 76.32; €1, 9.1,5. Found: ture overnight. The steroids were isolated with ether and C , 75.49: H, 9.11. freed of iodine by very cautious treatment with a minimum of sodium thiosulfate solution. The solvents were again re3 8 , l la-Dihydroxy-5,16-pregnadiene-20-one1la-Monomoved in vucuo and the steroids dissolved in a liter of acet- acetate and 38,l la-Diacetate.--X sample of 1la-acetox)-one containing 4 mi. of acetic acid and 20 g. of dry potassium diosgenin acetate as carried through our usual degradation acetate. The mixture was stirred a t reflux for 12 hours and p r o c e d ~ r e . ' ~ ,Xfter ~ the tn-o-phase t-butyl alcohol-water-again steroidal matter was isolated with ether.'O The potassium h>-droxide, saponification stage it was found that product was recrystallized from aqueous acetone to give the product was 1la-acetoxy-3~-11~-droxy-5,16-pregnadieneneedles de-solvating a t 97",and showing a double m . p . 112', 20-one, ni.p. 183-185', C;:,OH23;imK, e 8,180, [ a ] 2 5-44.5". ~ 208-213", [ a ]z 5 ~+33". The analytical results were indicThe infrared spectrum showed the presence of both hyative of a solvate. droxyl (3640 and 3430 cm.-l) and acetoxyl groups (3.733 Anal. Calcd. for C2iH320e.H20:C, 65.38; H, 8.11. cm.-l). Found: C, 65.37; H , 8.02. Anal. C d k d . for CL3H1104: C, 74.16; H , 8.66. Fuund: 21-Acetoxy-17~-hydroxy-5-pregnene-3,11,20-trione (VII). C, 74.01; H , 8.53. -The 3p-01 1'1, 1.35 g., was dissolved in 250 ml. of acetone (redistilled from potassium permanganate), cooled to 12' Acetylation n.ith acetic anhydride-pyridine overnight a t and treated with 1 ml. of an aqueous solution of 0.267 g. of room temperature gave 3 p , 11aY-diacetoxy-5,16-pregnadiene20-one, m.p. 144-145', X:$," 234mp, €8,875, [ L Y ] ~-40.3". ~D chrornium trioxide and 0.23 cc. of sulfuric acid. The reaction proceeded rapidly and within a minute a gray precipi- The infrared spectrum showed no hydroxyl bands, strong 1730 c i ~ i . -and ~ 123S1255 cm-1 band of relative intensities tate formed. After a n additional two minutes passed the reconsistent with two ester functions, 1679 cm.-' (unsaturated action was stopped by dilution with 600 nil. of saturated ketone) and 807,825 cm.-l bands (olefin). aqueous sodium chloride solution. The green solution gradually deposited a residue of pure-white flattened needles Anal. Calcd. for CajHaaOs: C, 69.74; €I, 7.96. Found: which were filtered off,washed with saline solution and with C, 69.60, H, 8.02. Both the mono- and diacetate resisted water. Aisample was recrystallized from aqueous acetone to saponification of the 1la-acetoxyl group, the infrared specgive 2 l-acetoxy-l7a-hydroxy-5-pregnene-3,11,20-trione. trum showing persistent acetate bands after overnight, room The melting point depended somewhat on the rate of heattemperature reaction in the above-indicated butanolic iug. Rapid determination on a preheated Kofler stage gave a caustic solution. Elevation of temperature resulted in res1-alue of 180-183". A slower determination gave melting a t 175" to a viscid mass t h a t resolidified incompletely to short inification. quadrangular forms reddening (decomposition) a t 200" with 3,13,1la-Dihydroxy-5-pregnene-2O-one(XVI).-The diene final crystal disappearance a t 220'. The sample was nearly XV, 200 mg. in 100 nil. of ether, was shaken with 1.5 g . of transparent t o ultraviolet 239 nip radiation, E 300 (meth- palladium-on-barium sulfate catalyst with hydrogen a t 3 anol). atmospheres pressure. The catalyst !vas filtered off, a n d the product was obtained as fibrous filaments on concentration Anal. Calcd. for C43H3d06: C, 68.63; H, 7.51. Found: of the ether, m.p. 181.5-182', [ L X ] ~ ~+YO". D The product C, 68.84; H , 7.13. was transparent to ultraviolet radiation in the 230-245 I I I ~ Isomerization to Cortisone Acetate .-A sample of 350 region, and in CS2 solution showed a single strong band in the trig. of 21-acetoxy-17a-hydroxy-5-pregnene-3,11,20-trione ketone region a t 1712 cm.-'. I n K B r , in addition to a ( V I I ) was dissolved in 50 ml. of methanol and 0.4 nil. of strong peak a t 1710 cm.-', a weaker peak appeared a t 1683 concentrated ammonia water "28y0" was added. The mixcm.-'. ture was let stand 100 minutes and the steroid was pre1la-Hydroxyprogesterone (XVII).-Oppenauer oxidation cipitated by addition of saturated sodium chloride. The of 200 mg. of the diol-one X V I in 20 ml. of dry toluene with product after crystallization as flattened blades from hexane 3 ml. of cyclohexanone and 600 mg. of aluminum isopropcontaining a very small amount of acetone was identical oxide during a 0.75-hour reflux period gave after stenin disin every respect with authentic cortisone acetate. tillation of solvents and recrystallization from aqueous methanol 180 mg. of 110-hydroxyprogesterone identical in every (18) Huang-Minion, E , \Vilson, he.I,. IVendler and RZ. Tishler, respect with :in authentic specimen obtained by microbioTHISJOURNAL, 74, 5394 (1952). Our thanks are due these authors for logical hydroxlation of progesterone. a comparison specimen. 3p, 11~-Dihydroxy-16~,17~-epoxy-5-pregnene-20-one (XV(19) A careful adsorption chromatography or the reaction product on III).-The A'6-20-ketone XS', 3 g. in 100 ml. of methanol silica gel revealed the presence of two contaminants. One as yet unia t loo, was treated with 26 ml. of 307% hJ-drogeii peroxide dentified, relatively nun-polar component, 0 25 g., m.11. 208-214' and 12 nil. of 4 N sodium hydroxide. After standing overs1mwrcI i i positive formaznn reaction 8400 (0111,1710 (c-I1, night a t 10' 100 ml. of water was added and the volume was ketone), l i 2 3 (ketone and acetate), 1730 (acetate), 1230 (acetate) reduced in half by concentration in VUCZLOa t 25'. The mixand characteristic sharp b a n d s a t 730 and 7 7 j cm.-'. A second ture wiis extracted with ether; the organic layer was wished sumewhat more polar component was the 3-acetate of the desired ~~~~~~~

compound, namely 34.21-diacetoxg-17a-hydroxy-~-pregnene-l1 2 0 dione, m.p. 187.5-188.5O, [ ~ ] ? S D f 2 1 ° , prismatic needles from ether. One sample showed the abo1.e melting point b u t underwent resolidilication of the melt to refractile sheaves with a second mcltiny point a t 03".

(20) R . Antonucci. S. Bernstein, R. Lenhard, IC. J . Sax and J . H IViIliams, J . O r & . C h e w . , 17, 1389 (1053). ( 2 1 ) (,'f. H. Honsser, R. Anliker, and 0. Jeger, IIcir. Ciiirii. . 4 [ l o , 36, 1337 (19.52).

ACID-BASEEQUILIBRIA OF WCORTICOTROPIN

Jan. 20, 1959

with water, dried, and evaporated t o a small volume, whereupon 2.75 g. of a crystalline product, m.p. 196-202", separated. T h e analvtical sample recrystallized from ether as prisms, m.p. 203-204", [ a I Z 5 D-18.4', transparent to ultraviolet. Anal. Calcd. for C21H3004:c, 72.80; H, 8.73. Found: C , 72.61; H , 8.62.

[CONTRIBUTIOS FROM

THE

415

16a,l7a-Epoxy-lla-hydroxy-4-pregnene-3,ZO-dione(160,17a-Epoxy-lla-hydroxy-progesterone) (XIX).-The preceding preparation was submitted t o Oppenauer cxidation under the same conditions used to prepare XVII. The product was XIX identical in all respects with the prcduct obtained by microbiological

PHILADELPHIA 18,PEXNA,

HORMONE RESEARCHLABORATORY A S D THE DEPARTMELT OF BIOCHEMISTRY, GhIVERSITY CALIFOR VIA, BERKELEY]

OF

Corticotropins (ACTH). XII. Acid-Base Equilibria of a-Corticotropin and Bovine Corticotropin1 BY Josh L ~ O N IAND S ~ CHOHH.40 LI RECEIVED JULY 16, 1958 Titration curves for a-corticotropin and bovine corticotropin have been obtained in 0.1 M KC1 solutions at 24.5'. The degree of ionization of phenolic groups has also been determined spectrophometrically. From these data, intrinsic ionization constants ( p K , ) of all acidic and basic groups in the corticotropins, as well as values for the electrostatic parameter ( w ) , have been estimated. The relationship of the amino acid sequence in the corticotropin molecule to the estimated values for p K , and ZB is discussed.

a-Corticotropin from sheep glands, which has been characterized as a polypeptide of relatively low molecular weight4 and whose structure has been established,j has recently been compared with a seemingly identical compound isolated in pure form from bovine glands.6 Titration curves for both the ovine and bovine hormones, together with details of an investigation of the ionization of phenolic hydroxyl groups, will be presented here.

Experimental Titration Assembly.-The cell used for the acid-base titration is shown in Fig. I; its design permitted the investigation of the rather limited amount of material available. The compact combination glass and reference electrode is supported by the cork stopper of the cell. Standard acid or base is delivered from a n Agla micrometer syringe, the tip of the needle (glass, or stainless steel) being immersed in the sample only during the time required for additions. Stirring of the sample is performed by a small magnetic stirrer, rotating simultaneously with the main magnetic stirrer of the thermostatic bath. Proper precautions are applied to exclude carbon dioxide while the sample is being dissolved or transferred to the cell, a slow flow of purified nitrogen heins maintained above the solution while the titration is being performed; a saturator, inserted in front of the cell, prevents changes in the volume of the simple resulting from evaporation. I n the course of these experiments, each sample was titrated throughout the entire p H range under investigation, in a semi-continuous fashion. Measurement of the pH.-The electrode train used was a combination glass electrode/Aig-.igC1 reference electrode, of the type described by Cannon'; its small size and excellent concentric shielding make it especially suitable for semi-continuous titration of small volumes.8 The pH measurements were made with a Beckman model G pH meter. __-___

(1) Presented in part a t the 1 3 1 s meeting of the Amprican Chemical Society, Miami, Florida, April 1957. This u-ork is supported in part by a grant from the Xational Institutes of Health of the Vnited States Public Health Service (Xo. KG-2907). (2) Fulbright Grantee 1958-1937, on leave of absence from the University of Brussels, Belgium. (3) C . H. Li, I. I. Geschwind, J. S. Dixon, A . I.. Levy and J . I. Harris, J . Diol. Chenz., 213, 171 (1953). (-1) A . L. Levy, I. I. Geschwind and C. H. I . i , i b i d . , 213, 187 fl9R.i). ( 5 ) C. H . Li, I. I. Geschwind, R. D . Cole, I. D. Raacke, J. I. Harris and 5 . S. Dixon, X a f i i r e , 176, 687 (195.5). (6) C. H. Li and J. S. Dixon, S c i e i z c t , 1 2 4 , 934 (1956). (7) A I . D. Cannon, ibid., 106,597 (1947). ( 8 ) Electrodes of this type. free from a n y significant sodium ion error, were purchased from Radiometer, Denmark (catalogue number OK 2021-B).

In order to standardize the p H scale, five reference buffers were prepared according to directions given by Bates9 (0.05 Jl K tetroxalate; saturated KH tartrate; 0.05 X K H biphthalate; 0.025 Jf K H ~ P O I 0.025 , AI iYa2HPOA; 0.01 -11 borax). During calibration of the Beckman instrument, all these buffers afforded very consistent readings (among all 5 buffers, discrepancies higher than 0.01 of a p H unit were never observed). Since the pH readings of the buffers showed no drift over a period of several hours, it would appear that no significant leakage of saturated KC1 from the liquid junction of the electrode occurred. -1fter each addition of acid or base during a titration experiment, equilibration of the sample to constant pH was usually reached within 2-3 minutes; only when some insoluble material separated out was i t necessary to allow longer equilibration, although the time interval never had to be extended beyond 10 minutes.

-_

_ _

..

- -

to pH

-

I >c Fig. 1.-Titration cell: a, reference electrode; b, glass electrode; c, magnetic stirrer; d , thermostatic bath. Acid-Base Titration Curves.-All titrations were carried out at 24.5', with samples of 8-10 mg. (about 2 micromoles of a-corticotropin in 1 ml. of COZ-free 0.1 AYKC1. Standard 0.2 X HC1 was added first, and the pH was read after each addition. When a p H around 2 had been reached, the sample was titrated in the other direction with standard C02free 0.2 X KOH; finally, i t was titrated again from a pH around 12 back t o the initial pH value. By this time, the total volume in the cell has increased by approximately one half; however, the mutual neutralization of acid and base has restored the concentration of KCl to what it was a t the start. The protein concentration was averaged from the results of two types of analysis: namely, determination of total nitrogen and measurement of the ultraviolet absorption spectrum. The correction for free acid or base at extreme p H (S) 12. G. Bates, "13lectrometric p H Determinations,' John Wiley and Sons, Inc., Sew York, S . Y . , 1954.