[CONTRIBUTION FROM
THE
DIVISIONOF CHEMISTRY AND DIVISIONOF PATHOLOGY UNIVERSITYOF TEXNESSEE]
Seroflocculating Steroids. 111.
AUD MICROBIOLOGY,
?vfr:DrcAr,
17\1r'r,
Chloro and Other Bile Acid Derivatives2
c.
B Y FREDERICC H A I i G , ROBERT T. BLICKETZSTAFF, A A R O N FELDSTEIS, 3a JANE R.AlvSOllf G R A Y , G E O R G E MCCALEB3b AND
DOUGLAS H.
s.
SPRUNT
RECEIVED KOVEMBER 20, 1956 For study of their seroflocculating activity, a number of 30-chloro bile acid derivatives were prepared by a new method involving replacement of a tosylate group in pyridine with pyridinium chloride, Several new acetates, tosylates and other related compounds are reported,
When the synthesis of ethyl 3~-chloro-ll-chole- For the preparation of ethyl 3,f3-chloro-9(11)nate (28) by the method previously described5 cholenate, because of the low melting points and was attempted on a larger scale, the yield and qual- poor crystallizing behavior of the corresponding 3aity of product were found to be considerably Iow- hydroxy and 3a-tosyloxy compounds which are inered. Various other direct methods for preparing termediates in the synthesis, conversion from the halides also proved unsatisfactory for this com- methyl ester is preferred. The sequence of reacpound. The presence of 28 among the products of a tions, part of which is also illustrative of other tosyl chloride reaction in pyridine described in the chloro compounds mentioned above, is shown. preceding paper' furnished a clue for a satisfactory Puzzling changes in melting point were encounnew method for preparing the compound. tered during crystallizations of several of the ethyl Since 28 obviously resulted from an intermediate ester tosylates when methanol was used as solvent tosylate, the commonly used reaction of a tosylate a t 50". Since the fractions continued to give corwith lithium chloride in acetone (and in dioxane) rect analyses, replacement of tosylate group was was tried on the tosylate of ethyl 3a-hydroxy-11- not taking place, and ester interchange was not cholenate. Intractable mixtures of unsaturated suspected, because numerous other ethyl esters and chloro compound resulted. However, by sub- in this related group of compounds can be crystalstituting pyridine and pyridine hydrochloride for lized unchanged from methanol. However, ester the solvent-halide system, a good quality chloro interchange did take place, as was evident when product was obtained in 51%. yield. The yield is the supposed ethyl ester tosylate 24 was converted only moderate, and the superiority of the method into methyl 3P-chloro-11-cholenate. Apparently a lies in the purity of the product, which is very little trace of 9-toluenesulfonic acid resulting from hycontaminated by dienic impurity. A mixture of drolysis of the 3-tosylate group (it is known that chloro compound containing more than traces of esters of the 3-hydroxy steroids are easily hydrodiene is separated only with much difficulty; the lyzed) catalyzes the ester interchange. By adding inferior products obtained in other halogenation 9-toluenesulfonic acid to the refluxing methanol methods are such mixtures. The reaction tempera- solution during crystallization, conversion of ethyl ture is an important factor, as previous studies on to methyl ester is fairly rapid and complete. The replacement reactions would suggest,e and a t ele- facile ester interchange of the bile acid esters might vated temperatures undesirable mixtures of chloro have been expected from the marked ease with and dienic compounds are obtained. At tempera- which cholenic and substituted cholanic acids are tures between 78 and 90' the substitution reaction esterified. For instance, 3a-hydroxy-11-cholenic proceeds a t a satisfactory rate, with little competi- acid is completely esterified in 40 minutes by methation from the elimination reaction. nol containing small amounts of aqueous hydroMethyl 3P-chlorocholanate, methyl and ethyl chloric acid a t room temperature.* Yamasaki, 3/?-chloro-ll-cholenate, methyl 3P-chloro-9(11)- Rosnati, Fieser and Fieserg recently encountered cholenate and methyl 3~-chloro-12a-hydroxychola-an ester interchange during the reduction of nate have all been prepared by this method, methyl 4P-bromodehydrolithocholate by sodium which m7e are studying further with the view of its borohydride in the presence of ethanol. They also converted ethyl 3,4-dibromocholanate into the use as a general method of preparing halides.' methyl ester by refluxing in methanol containing (1) Paper I 1 of this series, THISJOURNAL, 79, 2161 (1957). concentrated hydrochloric acid. Several of our es(2) This investigation was supported in p a r t by grants (CS-9053 ters underwent interchange similarly when reand C-2249) from t h e Kational Cancer Institute, of t h e National Institutes of Health, Public Health Service. fluxed, but a purer product usually resulted when ( 3 ) (a) Worcester Foundation for Experimental Biology, Shrewsthe reaction was left a t room temperature for 10 to bury, hlass. ; (b) Department of Chemistry, University of Arkansas, 18 days. Ethyl 3~-chloro-ll-cholenate(28) preFayetteville, .4rk. (-1) Numbered according t o t h e consecutive order in which t h e pared in this way crystallized directly from the compounds are described in t h e Experimental sections of this (111) and t h e accompanying papei ( I V ) . ( 5 ) Paper I of this series, T H I S J O U R N A L , 76, 3213 (1954). (6) E. D. Hughes and C. K. Ingold, T Y ~ P IFavaday S. soc., 37, 657 (1941). (7) Organic halides have been obtained often as a result of reaction between alcohols and tosyl (and mesyl) chlorides in pyridine [e.g., C. R . Noller, C . A . Luchetti, E. M . Acton a n d R.A. Bernhard, THIS J O U R N A L , 75, 3831 (19.5311, b u t t h e use of pyridine base and pyridinium halide on a tosylate appears t o be novel. F. Drahowzal and D.
Klamann [Monatsh., 82, 460 (1951)l record failure t o obtain %-butyl a n d n-octyl chlorides from the tosylates under conditions similar t o ours except for a considerably smaller proportion of pyridine, b u t they describe successful preparations of halides using pyridinium chloride without solvent and with carbon tetrachloride a s solvent. ( 8 ) L. L. Engel, V. R . hlattox, B. F. hfcKenzie, W. F. hlcGuckin and E. C. Kendall. J . Riol. Chem., 162, 36.5 (1946). (9) K. I'amasaki. V. Rosnati, 51. Ficser a n d 1,. F . Fieser, Trirs JOURNAL, 77, 3308 (1956).
May 5 , 1957
CHLORO AND BILEACIDDERIVATIVES OF SEROFLOCCULATING STEROIDS
A,,'-'
i\
1. KOH. M K J H 2, HCl
c1
TsO
2165
-1
/EtOH(HC1)
32
reaction mixture in a state of highest purity in near quantitative yield. This ester interchange constitutes a good method for preparing higher esters of the bile acids from the often more readily available methyl esters. The extreme ease of esterification of the &bile acids is in contrast to the considerable resistance exhibited by the bisnorcholanic acids to this reaction. 3~-Hydroxy-5-bisnorcholenicacid forms the methyl and ethyl esters in poor yield even after several days of reflux with the respective alcohols. Perhaps for this reason, only the methyl ester prepared with diazomethane has been reported. The expected difference in the rates of reaction between the Go-secondary and (223-primary carboxyl groups is doubtless intensified considerably by the interference offered by the steroid nucleus. Experimental lo, ESTERS OF HYDROXY ACIDS. Esterification of Hydroxy Acids.-All of the Cn4-acids were esterified readily by using the proportion of 1 g. of acid:lO ml. of absolute (methyl or ethyl) alcoho1:O.l ml. of concd. HC1. (The alcoholic solution is approximately 0.12 N in HC1.) With the 10: 1 ratio of solvent to acid, the reaction is rapid and complete. Methyl lithocholate (1) , I 2 was obtained from lithocholic acid in quantitative yield as large needles, m.p. 129-130"; recrystallized from methanol-water and dried a t 100" (1 mm.) for 24 hr. for analysis, m.p. 125-127.5", [ a ]+22". ~ Anal. Calcd. for C25H4203 (390.59): C, 76.87; H, 10.84. Found: C, 76.6; H, 10.8. This conflicts with the Reindel report'? that the 125-127" melting product contains 1.5 moles of methanol as solvent of crystallization. Fieser and EttoreI3reported m.p. 125-127"; Fieser and Rajagopalan,14 129-130°, [a]D +29 1 2 " (diox.). Ethyl Lithocholate (2) . I 2 By ester interchange from 1 by refluxing 1.5 hr. in ethanol and concentrated hydrochloric acid; 7470 of 92-95' melting product. Recrystal(10) Microanalyses by the Microchemical Laboratory of New York University, and by Galbraith Microanalytical Laboratories, Knoxville, Tenn. (11) Melting points were taken on an electrical micro-hotstage and are uncorrected. Optical rotations were determined in 1-2Y0 chloroform solutions at about 2 5 ' , except where noted, using a Schmidt and Haensch polarimeter, and values are accurate to & l o . The values marked with an asterisk (*) were obtained with a Keston polarimeter attachment (Standard Polarimeter Co., 225 E.54th Street, X. Y. C.) to a Beckman DU spectrophotometer with accuracy estimated to be better than *2'. (12) F.Reindel and A. Niederliinder, Ber., 68B,1969 (1935). (13) L.F.Fieser and R . Ettore, THISJOURNAL, 76, 1700 (1953). (14) L. F. Fieser and S. Rajagopalan, ibid., 73,5530 (1950).
31 lized from methanol-water, m.p. 94.596' (lit. m.p. 9293'), [ a ]+28.8'*. ~ Anal. Calcd. for CZH& (404.60): C, 77.18; H, 10.96. Found: C,77.18; H , 10.84. Methyl 12a-hydroxycholanate (3) was prepared according to Barnett and Reichstein,'; m.p. 119.5-120.0" (lit. 120121'), [a]D+39". Ethyl 12a-hydroxycholanate (4) was prepared by the same method from ethyl 12a-hydroxy-3-cholenate (54)16in quantitative yield; crystallized from absolute ethanol, m.p. 135-136.5", [ a ] ~+46". Anal. Calcd. for CZBH4403 (404.60): C, 77.2; H, 11.0. Found: C, 77.1; H , 10.9. Compound 4 was also prepared by esterification of 12ahydroxycholanic acid, m.p. 135-1365'. Ethyl 3a-hydroxy-9(1l)-cholenate(5) was prepared by esterification of 3a-hydroxy-9(1l)-cholenic acid," m p. [ a ] ~ +35.6". Anal. Calcd. for C26H4208 75.0-76.5', (402.60): C, 77.56; H , 10.52. Found: C,77.25; H, 10.44. Compound 5 also was obtained from methyl 3a-acetoxy9(11)-cholenate (12) by treatment with ethanol and hydrochloric acid. Methyl 3p-hydroxy-5-bisnorcholenate( 6 ) was prepared by refluxing the corresponding acid in methanol and concd. hydrochloric acid for two days; 547, yield of product melting 140-144" out of ligroin (63-70'). A chromatographed ~ sample for analysis melted a t 142-143.7', [ a ] -55.4"*. (This compound was previously prepared by use of diazomethane by Fernholz,ls m.p. 1 4 0 O . ) Anal. Calcd. for C23H3603 (360.52): C, 76.62; H , 10.07. Found: C, 76.95; H. 9.97. Ethyl 36-Hydroxy-5-bisnorcholenate (7) .-Two grams of 3p-hydroxy-5-bisnorcholenic acid (Nutritional Biochemical Corp.) in 150 ml. of absolute ethanol and 2 ml. of concd. sulfuric acid was refluxed for 45 hr. After usual processing a product melting a t 162-164" was obtained in 34y0 yield, ~ after recrystallization from ligroin (63-70'), [ a ] -65.6". Anal. Calcd. for 624H&(374.64): C, 76.96; H , 10.23. Found: C, 77.2; H , 10.5. ACETATES.-The3-acetates were made by room temperature acetylation using acetic anhydride and pyridine; the 12-acetates required heating to loo", with the exception of those prepared using 9-toluenesulfonic acid as catalyst (see compounds 18 and 19). Ethyl 3~-acetoxycholanate(S)12 was obtained from ethyl lithocholate, out of ethanol, m.p. 95-98" (lit. 90-91"), [a]D $38.2'. Anal. Calcd. for C28H4604 (446.65): C, 75.29; H , 10.38. Found: C, 75.48; H, 10.56. Methyl 12a-acetoxycholanate (9) from 3, heavy prisms ~ Anal. Calcd. out of methanol, m.p. 95.5-96.5", [ a ]f69'. for C,,H4404 (432.62): C, 74.95; H, 10.25. Found: C, 75.03; H, 10.14. Ethyl 12~~-acetoxycholanate (10) from 4, out of absolute ~ ethanol as short, thick prisms, m.p. 63-64", [ a ] +69.9". (15) J. Barnett and T. Reichstein, Helo. Chim. Acta, 21, 926 (1938). (16) Paper IV of this series, THISJOURNAL, 79, 2167 (1957). (17) L. F. Fieser and S. Rajagopalan, ibid., 79, 121 (1951). (18) E. Fernholz, A n n . , 607, 128 (1933); A. Butenandt and G. Fleischer. Bcr., 70, 98 (1937).
2166 Anal.
F.CHANG,R.BLICKENSTAFF, A. FELDSTEIN, J. GRAY,G. MCCALEB AND D. SPRUNT Vol. 79 Calcd. for C&4604
(446.65): C, 75.29; H , 10.38.
104.5-107', ( O J D $-30°. - 4 ~ ~ 1Cnlcrl. . for C a r I b % S (542.69): C, 70.81; 11, 6 . X . Found: C, 70.9; 11, 8.9. Ethyl 3a-acetoxy-11-cholenate (11 ) from ethyl 3a-hy- ( 0 ) By ester interchange: IVhen theethyl ester (24) was fir.it tlroxy-1 l-~liolenate,~ nut of methanol as fine needles, n1.p. prepared, cryst2llizatiori from methanol a t 50" brought 91 5-93.7", [a]D4-42'. Anal. Ca!ctl. for C ~ ~ H U (444.63): O~ out coloriess crystals, the niclting point of Tvhich O n conC , 75.63; H, 9.97. Found: C, 75.6; H, 9.9. tinued crystallizatii,il rose from 65 to IO-t.5". Tiint all Ethyl 30-acetoxy-9(11)-cholenate (12) from 5 , out of t:ster interchange f w r n thi, ecliyl to triethyl ester had taketi lcgroin (35-60"), m.p. 96.5-97.5', [DID f64.8". Arzal. 1)incewas proved by thc conversion of the product to methyl Calcd. for C~~H4404 (444.63): C, 75 63; H, 9.97. Found: ~3~-chloro-ll-cholcnatc (27). C, 75.44; H , 10.35. Ethyl 3a-tosyloxy-l l-cholenate (24) b y to Methyl 12a-acetoxy-3-cholenate (13) from 53,18 flat ethyl 3ru-hydroxy-11-ehoicnate.5 U'ell-shaped Anal. prisms out of methanol, m.p. 107-108", [DID +63.9'. absolute ctlia.no1; has double melting point 5 Calcd. for C27fl4.04 (430 61): C, 75.31; H, 9.83. Found: ,>-96..i", i a ] o +3(7.2"*.
