~ ultrabeen observed.' In the preceding communication2 to occur u n d x these conditions. A \ kthe we have described the first 26-hydroxylation of violet spectrum of I11 was strikingly different from Compound S with Streptomyces sp. 1%-ewish now that of a 2-hydroxy-3-keto-A4-steroid in alkaline - ~ presence of the dienone among to report the microbiological hydroxylation of 4- s o l u t i ~ n . ~The pregnene- 1 7 41-dio1-3,20-dione (I, Reichstein's the products from the alkaline treatment of TI1 Compound S)in the 1t and 2p positions by Rhizoc- was further substantiated by the measurement tonia ferrugena (CBS, Holland). The fermenta- of the polarographic reduction potential of the X shift in the half-wave potential from tion of I (3.0 g.) by R. ferrugena was carried out in a s o l ~ t i o n . ~ medium containing corn steep liquor, ammonium 1.46volts (111) t o 1.31 volts (after alkaline treatdihydrogen phosphate, yeast extract, dextrose, ment) corresponds well to the shifts observed for soybean oil and calcium carbonate. In the fer- cortisone vs. prednisone (0.16 volt) and cortisol z's. mentation stage a forty-eight hour culture was prednisolone (0.17 volt). Furthermore, pyrolysis incubated with the steroid substrate (introduced in of a sample of I11 a t the melting point for ten ethanolic solution) on a rotary shaker for five minutes afforded a mixture of products whose indays at 23'. The culture filtrate was extracted frared spectrum contained the characteristic A1s4 with chloroform and chloroform-methanol (4 :1) diene-3-one bands. Finally, a solution of I11 in glacial acetic acid and the extracts were concentrated to dryness in vacuo below room temperature. The residue was was refluxed for one hour. Upon concentration of redissolved in dry, acid and alcohol-free chloroform the solution ,in vacuo, chromatography of the resand the solution was chromatographed on a silicic idue on Florisil and recrystallization from acetone . acid adsorption column. A methanol-chloroform hexane there was obtained 1,4-pregnadiene-lia,21whose infrared gradient elution3 was utilized to separate the two diol-3,20-dione, m.p. 237-240', steroidal transformation products and residual spectrum was identical with that of an authentic substrate. The steroid substrate, I, was collected sample.6 From this evidence we conclude that the in the first fractions, the 26-hydroxylated product, structure of I11 is clearly &-pregnene-l(,1 7 ~ ~ ~ 2 1 11, in the middle fractions and the 1-hydroxylated diol-3,20-dione. The configuration of the l-hyproduct, 111, in the last fractions. Homogeneity droxyl group remains to be fixed. was indicated by paper chromatography. With cortisol, cortisone, corticosterone, desoxyCrystallization of the middle fractions from corticosterone, progesterone and testosterone transchloroform-ether mixture afforded 101 mg. of 4- formations occur upon incubation in the presence of pregnene-26, 1Ta,2 I-triol-3,20-dione (11), m.p. 2 11- R.ferrugenu as determined by paper chromato220' dec. Further recrystallization from acetone- graphic examination of the product. The structure hexane raised the m.p. to 220-222' dec. The in- of the transformation products in these instances frared spectrum of I1 was identical with that of an remains to be elucidated. Other species of Rhizoctonia have effected the authentic sample,2 and the mixed m.p. showed no depression. transformation of I into cortisol, cortisone and 11The crystallization of the last group of fractions epicortisol; a more detailed account of these (Product 111) afforded 163 mg. of crystalline solid, findings will he published subsequently. hddim.p. 180-200" dec. Cautious recrystallization tional conversion products obtained with organisms from acetone-hexane without application of heat, of this genus are now under investigation. LVe wish to make the following acknowledgments. raised the m.p. to 193-207" dec. (with a transition of about 170"), [ o ( ] ? ~ D +80 (dioxanej (corrected From the Schering Corporation we acknowledge thc for acetone of solvation), A$:,"" 241 mp ( E = 16,500 helpful advice of Drs. Xndr6 Meyer and .%lexander -corrected for acetone of solvation), A:$;' 2.83 p Nussbaum, the assistance of Dr. P. Kabasakaiian (OH), 5.81 p (20-carbonyl), 6.04 p (3-carbonyl), and Mr. J. hIcGlotten who carried out the polaro6.18 p (A4) (one additional polymorphic variety graphic reductions, of Dr. J. Y . Chen who interhas been observed.) Calcd. for C21H3~O5.C3H6O : preted the infrared spectra, and of Slrs. 11. J . C, 68.54; H , 8.63. Found: C, 68.41; H, 8.66. Gentles. From the Institute of 1Zicrobiology we The identity of 111 was established in the fol- acknowledge the technical assistance of 1Zr. A. lowing manner. Integration of the hydroxyl Chernov. We also wish to thank Dr. C. 11.Haenbands of the infrared spectrum confirmed the pres- seler for his helpful advice and for various cultures ence of three hydroxyl groups. Measurement of of Rhizoctonia. the ultraviolet spectrum of I11 in alkaline solution4 ( 5 ) P. Kabasakalian. S . De1,orenzo and J. hIcC.lotten. A n u l . Chriii.. showed a shift in the maximum t o 245 mp but no 28, 1669 (1956). ( 6 ) A . Nobile, W. Charney, P. I,. Perlman, H. I,. Herzog. C . C peak in the 380 mp region after two hours at 60'. Payne, SI. E: Tully, 11 A Jevnik and E. R . Hershlierg, T H I SJ < ) I i R x A l r This is consistent with the conversion of a l-hy- 77, 4184 (19.551. droxy-3-keto-A4-steroid into a A104-diene-3-keto- TSSTITUTE O F ~ ~ I C ! R O R I O I , O C Y GEOR(:EGREEKSPAS CARL P. SCHAFFNER steroid, which transformation would be expected RVTGERSUSIVERSITY (1) S . Kushinsky (Abstracts of Papers, 131st Meeting of the Am. Chem. Soc., Miami, Florida, April 7-12, 1957, p. 36.0) mentions t h e isolation of a 1- or 2-hydroxy intermediate from conversion of 19nortestosterone 17-acetate t o estradiol 17-acetate h y the action of Covynehacteriimt simb1r-c. (2) H . L. Herzog. 11.J. Gentles, F. Carvajal, D. Sutter, W. Charney, E. R . Hershberg and C . P. Schaffner, THISJOURNAL, 79, 3921 (19.37). (,'3) T. K . I,ak.;hmnnun and S . T,ieherman. 'Arrir A i o r h r w Aiophgc , 5 3 , 288 ( 1 9 5 4 ) . 1 t i .< s hipypr,
.r o I y .
/3rj,,.,
2 0 , 12411(iw:),
SEW BRLJSSTVICK, S.J , \ \ r CHARXEY ~ ~ ~ SCHERISG CORPORATIVK HERSHEL L. HERZOG RI.OOMFIEI.~,NEIVJERSEY E. B. HERSHBERG RECEIVEDJ U N E :i, 1057 ____ ~.. ~
CYCLOPENTENE-3,5-DIONE ,Sir:
'The synthesis o f many substituted a n d uti saturated c~yclopeniarie derivati\rPS of thenretic-n 1
~
~
~
COMMUNICATIONS TO THE EDITOR
3924
and practical importance is hindered by the lack of readily available intermediates. With few exceptions, substituted cyclopentadienes are too unstable to be isolated, and these exceptions have not proved useful for further synthetic work.1,2 We wish to report the synthesis of cyclopentene-3,5-dione (I),
"\pi'"I a yellow, stable, crystalline, highly reactive molecule which promises to provide an entry into many new synthetic fields. Cyclopentene-3,5-dione can be prepared easily from cyclopentadiene by bromination, a and treatment of the resultant 3,5-dibromocyclopentene with tetraethylammonium acetate in dry acetone.4 Hydrolysis affords cyclopentene-3,5-diol which can be oxidized to I in 50% yield with chromic acid in acetic acid. After crystallization from ether-pentane and two sublimations it showed m.p. 36-37'. Anal. Calcd.: mol. wt., 96. Found: C, 62.3; H , 4.0; mol. wt., 96; ultraviolet: Xmax 222 (4.16), 322 (1.30), 367 (1.30); infrared: 5.73, 5.83. We shall report on the details of several of its reactions later, but the following will indicate their scope. (1) It is readily reduced to 1,3-cyclopentanedione (m.p. 150-151', reported5 151-152') and this sequence may provide the best synthesis of this useful compound. ( 2 ) The enedione is highly reactive in the Diels-Alder reaction, and readily gives adducts with cyclopentadiene, anthracene and other dienes. (However, it is less reactive than maleic anhydride. We consider this difference significant, and will discuss it in due course.) (3) Cyclopentene-3,5-dione condenses rapidly with aldehydes and ketones in the presence of weak bases. These and other reactions are being explored. The dione is an acid (PH of a 0.1 aqueous solution, 4.5) which is rapidly destroyed by base. Despite its acidity, the molecule appears to be completely ketonic. It does not give a ferric chloride test. I n its enolic form it would be a derivative of cyclopentadienone and as such perhaps partake of the unfavorable electronics which have been suggested to explain the lack of existence of that molecule. I n addition to its interest for its own sake, we believe that this unique molecule, containing as it does three different types of functional groups and with each of its carbon atoms potentially reactive, will be a useful intermediate in the synthesis of many hitherto unavailable molecules.
VOl. 79
A NEW TYPE OF PRODUCT FROM THE WILLGERODT REACTION
Sir: We wish t o report the migration of a keto group to an adjacent carbon atom during a Willgerodt reaction. Methyl 0-methyl-7-propionylpodocarpage (I) under Willgerodt conditions is converted (10% yield) to methyl 0-methyl-7-acetonylpodocarpate (11). Since this new ketone is also converted to the thiomorpholide (111) under Willgerodt conditions, it is probable that a t least part of the material goes through I1 as an intermediate. OCHi
I
A / - = H Sulfur ,'\ C'H3 COOCHI Sulfur ____f
rMorp1ioline 0
1
S
0
/I
11
R C H ~ C C H+-RCHXH~CN-O ~ I1
1:
RCCH2CH3 CHJUgBr Sulfur
0
IV
\
S Y
RCCH3
The reaction of methyl 0-methylpodocarpate' and propionyl chloride in the presence of aluminum chloride gave methyl O-methyl-i-propionylpodocarpate [I, m.p. 103.5-105', X max 255 mp ( E = 10,500)2and 315 mp ( E = 4,090), 3.52 and 5.9s1.1 p(CHCI3); [ c ~ ] ~+134' ~ D (ethanol); calcd. for C, 73.71; H , 5.44. Found: c, 74.04; C&f&: H, 8.541. Chromatography of I on silica under conditions which were shown to separate I and I1 failed to reveal the presence of any 11. X mixture of methyl 0-methyl-7-propionylpodocarpate (107.5 g.), sulfur (14.4 g.), and morpholine (39.2 g.) was heated a t reflux for 10 hr. The reaction mixture was dissolved in benzene, filtered, and the solvent was removed. The expected thiomorpholide, (111, 85.6 g., m p . 172-177') crystallized when an ether solution was concentrated. Further recrystallization from a mixture of chloroform and methanol gave the pure thiomorpholide [m.p. 177-178.5, max 281 p ( E = 18,000), [aIz5D 110' (chloroform) ; calcd. for C26H3,N04S: DEPARTMENT OF CHEMISTRY N, 3.05; S, 6.98; OCH3, 13.50. Found: n', IOWASTATECOLLEGE C. H. DEPUY 3.08; S, 6.04; OCH3 13.021. The material in the AMES,IOWA E:. F ZAWESKI mother liquors from this reaction was chromatographed over silica (1700 g.). Among the com(1) C. H. DePuy and W. von E. Doeting, THISJOURNAL, 75, 5955 pounds eluted with 2y0 ethyl acetate in benzene (1953). (2) F. Ramirez and S. Levy, ibid., 1 9 , 67 (1957). was methyl 0-methyl-7-acetonylpodocarpate (11) (3) W. G. Young, H. K. Hall, Jr., and S. Winstein, ibid., 1 8 , 4336 which crystallized from aqueous methanol as
+
(1956).
(4) L. N. Owen and P. N. Smith, J . Chem. Soc., 4043 (1952). ( 5 ) J. H.Booth, R. G. Wilkinson, S. Kushner and J. H. Williams. THISJOTJRNAT.. 75, 1732 (1953).
(1) I R Shermood and W. F. Short, J Chem S O L ,1006 (1938) (2) The analytlcal determmations were made bv the Analvtlcal Department under Dr Robert T Dillon
