Oct. 5, 1960
COMMUNICATIONS TO THE ED~TOR
5251
sis showed t o consist of 87y0 methyl trans-1,2dimethylcyclopropane - 1 - carboxylate (IV), nZ6D 1.4218 [Anal. Found: C, 65.86; H, 9.501, 2% 11, 7% methyl 2,3-dimethyl-2-butenoate (V) (identified by infrared spectrum and gas chromatographic retention time identical with those of an authentic sample), and 4y0 methyl angelate (identified in the same manner as V). At 33-35’ irradiation of I gave a mixture of esters found (5) Since our interests when we started this work were not centered by gas chromatography to consist of 73y0 11, around the intermediacy of protonated cyclopropanes in carbonium ion rearrangements, no attempt was made to identify any product 3y0 V, and 2OYO methyl tiglate (identified in the of cyclopropane skeleton. We wish to emphasize that our arguments same manner as V). The methyl angelate and do not necessarily apply to reactions done under basic conditions,l8 methyl tiglate formed in these irradiations renor do they exclude protonated cyclopropanes as intermediates in the sulted from the apparent reversal of pyrazoline formation of cyclopropane compounds.ab In addition we wish to point 77, out that the work of J. D. Roberts and J. A. Yancy, THISJOURNAL, formation, a reaction which has not been previously 5558 (1955), on the reaction of 2.3,3-trimethyl-2-butanol-l-C14 with observed. concentrated hydruchloric acid also excludes any protonated cycloproThe structures of the products were established pane intermediates prior to formation of classical carbonium ions, as cyclopropanes by their infrared, ultraviolet, and ur before reaction of classical carbonium ions with chloride ions. n.m.r. spectra. The infrared spectra of I1 and I V KEDZIECHEMICAL LABORATORY DEPARTMENT OF CHEMISTRY GERASIMOS J. KARABATSOS(in carbon tetrachloride) contain no olefinic bands MICHIGAN STATEUNIVERSITY JOHND. GRAHAM in the 1700-1600 or 950-880 cm.-’ regionsJ5 EASTLANSING, MICHIGAN while their ultraviolet spectra show only weak end RECEIVED AUGUST10, 1960 absorption ( 6 ca. 200). Olefinic hydrogen peaks are absent from their n.m.r. spectra, while cyclopropane hydrogens appear in the region T g = 8.6LIGHT-INDUCED DECOMPOSITION OF
topic enrichment of the alcohol (30% excess C 1 3 ) suggests that detection of C 1 3 in carbon-4 would have been achieved, had mechanism (1) occurred t o the extent of 20% or more. The above arguments lead t o the conclusion that under strong acid conditions carbonium ion rearrangements of the neopentyl system do not occur mainly via protonated cyc1opropanes.j
PYRAZOLINES. AN IMPROVED ENTRY INTO THE CYCLOPROPANE SERIES
Sir:
9.8.1°
Steric assignments of I1 and IV were made on the basis of (a) competitive saponification of a mixture of I1 and IV in which the less hindered ester moiety of I1 was hydrolyzed more rapidly than that of IV, and (b) their formation from pyrazolines, in which stereospecific inversion is considered to be unlikely. Acknowledgment.-This investigation was supported in part by a grant (No. RG-5883) from the Division of Research Grants, National Institutes of Health.
Thermal decomposition of pyrazolines is a wellknown route to cyclopropanes.1,2,3.4The synthetic value of the reaction is reduced considerably, however, by the extensive formation of olefinic produ c t ~ , by ~ ~a ~lack ~ ~of ~stereospecificity,6 ~ - ~ and often by extensive tar formation.2 We now wish to report that light-induced decomposition of stereoisomeric pyrazolines has led to the formation of cyclopropanes stereospecifically, and with(8) L. J. Rellamy, “Infrared-red Spectra of Complex Molecules,” out olefin formation. 2nd ed., John Wiley and Sons, New York, N.Y., 1958. When 3-carbomethoxy-cis-3,4-dimethyl-l-pyra- (9) G.V. D. Tiers, J . Phys. Chem., 62. 1151 (1958). zoline (I), prepared by treatment of methyl tiglate (10) L. M.Jackman, “Applications of Nuclear Magnetic Resonance with d i a ~ o m e t h a n e , was ~ , ~ irradiated with a sun- Spectroscopy in Organic Chemistry,” Pergamon Press, New York, Y., 1959. lamp a t ca. 15’, the sole product (by gas-liquid N.(11) Lubrizol Fellow, 1959-1960. chromatographic analysis) was methyl cis-1,2- DEPARTMENT OF CHEMISTRY AND dimethylcyclopropane - 1 - carboxylate (11), n2% CHEMICAL ENGINEERING KENNETH L. RINEHART, JR. 1.4289 [Anal. Found: C, 65.26; H, 9.441. UNIVERSITY OF ILLINOIS THOMAS V. VAN ATJKEN~~ URBANA, ILLINOIS RECEIVED A41JGUST22, 1960
THE METABOLISM OF ALDOSTERONE : ISOLATION AND CHARACTERIZATION OF TWO NEW METABOLITES’
Sir:
Irradiation a t ca. 30-35’ of 3-carbomethoxytrans-3,4-dimethyl-l-pyrazoline (111),5 prepared from methyl angelate and diazomethane, gave a mixture of esters which gas chromatographic analy(1) E. Biichner and L. Perkel, Bdr., 36, 3774 (1903). (2) K . von Auwers and F. Konig, A n n . , 496, 252 (1932). (3) D . E.McGreer, J . Org. Chem., 25, 852 (1960). (4) W. M. Jones, THISJOURNAL, 82, 3136 (1960). and preceding papers. ( 5 ) K.L. Rinehart, Jr., and T. V. Van Auken, paper in preparation. (0) H. L. Slates and N. L. Wender, TXISJOURNAL, 81, 5472 (1939). (7) K . von Auwers and F. Konig, Ann., 496,27 (1932).
I n this report we describe the isolation of two new metabolites of d-aldosterone, 5~~-(4,5)-dihydroaldosterone (Ia) and 3POH15a-(4,5)-tetrahydroaldosterone (IIa), from the incubate of d-aldosterone with rat liver homogenates. I n addition, the synthetic preparation of 5 CY- (43) -dihydroaldosterone 21-acetate (IIIb), 380H15a-(4,5)-tetrahydroaldosterone (IIb), the 3-keto etiolactone (Ivb), and the 3-hydroxy etiolactone (VIb) are recorded. Romani, et aLI2have suggested the formation of (1) This work was supported in part by a grant (P. H. S. A-1156) from the National Institute of Arthritis and Metabolic Diseases of the National Institutes of Health, Education and Welfare. (2) J. D . Romani. C. Bessard, I. %sa-Castellanos and A. Keller, A n n . Endocrinol., 20, 209 (1959).
5252
~ O R l h I U N I C A T I O N ST O THE
CH,OH
HO,
CHzOIi
I
EDITOR
Vol. 82
tograms and further purified by chromatography over A1203. Compound Ia was crystallized from CHClri-Et20, m.p. (needles) 154-156", [ a ]23D 53 (CHC13, C = 0.362), vE::'~ 3700, 3500, 1710, 1060, 982, 963, 911 cm.-'. Compound I I a was purified by formation of the digitonide, [a]2 2 ~ 48" (CHC13, C = 0.165). Compound Ia was acetylated with Ac20 in pyridine and gave after crystallization from moist C H 3 0 H the monoacetate (IIIa), m.p. (needles) 173-1'74". CHC13 Raldosterone 21-acetate = 2.15; Vmsx 3700, 3500, 1745, 1710, 1135, 1100, 995, 980, 903 cm.-'. For comparison, the dihydroaldosterone 21-acetate (IIIb) was prepared by reduction of synthetic d-aldosterone 21-acetate over Pt02 in neutral solution, m.p. (needles) (analytical sample) 176184", mixed m.p. with I I I a 174-184", [ O ~ ] * ~51D f. 3' (CHCh, C = 0.652), vE;:'3 3700, 3500, 1745, 1710, 1135, 1100, 995, 980, 903 cm.-'. (Anal. Calcd. for C23H3206: C, 68.29; H, 7.94. Found: C, 68.20; H , '7.54.) Raidosterone 21-acetate = 2.0. Compounds IIIa and I I I b were identical. The dihydroaldosterone (Ia) was oxidized with NaIO, to the 3-ketoetiolactone (IVa), m.p. (prisms from moist EtOH) 264-275", vE?:'~ 1'780, 1710, 1135, 1095, 965-995 (multiple), 910 cm.-'. Synthetic d-aldosterone was oxidized with K a I 0 4 to the etiolactone (V), m.p. 323-826", 3':::v 1780, 1670, 1618, 1115, 1100, 1080, 1060, 980, 960, 910 cm.-l. (Although the reported m.p. for V is 308-313°,11 the infrared spectra are identical.) The etiolactone (V) was reduced over PtO? in acidic solution to the 3-ketoetiolactone (IVb), m.p. (prisms from moist acetone) 282-2%", (needles from moist EtOH) (analytical sample) 278-285", mixed m.p. with IVa 2Gt5-28Z0, [a]2 3 1 ) 23 f. 4" (CHC13, C = 0.45), (Anal. Calcd. for Cz0H?604: C, 72.70; H , 7.93. Found: C, 72.48; H , 7.82.) ; and the 3-hydroxyetiolactone compound (VIb), purified by formation of the digitonide, m.p. (prisms from moist acetone) 230-235'. Compounds IVa and IVb were identical. Compound VIb was oxidized with chromic acid in pyridine solution to Compound 1%. Since the %acetate of Compound VIb previously had beeii shown to be of the 5 a H series,l1 the configuration of IVa and IVb a t 5 is thus determined. The structure of Compound I a is established by direct coniparison with synthetically prepared 5a-(4>5) dihydroaldosterone and, further, by conversion to the 3-ketoetiolactone (IVa) and direct comparison with the synthetically prepared 3-ketoetiolactone (11%). Compound IIa. the other metabolite isolated, was identical with a sample of tetrahydroaldosterone (IIb) prepared by reduction of synthetic daldosterone 21-acetate over PtO? in acid solution. ~. 5 =t This compound had m.p. 1 15-12Xo, [ a ]2 1 4 6" (CHCl3, C = 0.291), v,"::,"'3 3700, 3500, 1705 (weak), 1135, 1060, 1025, 998, 960, 904 cm-l. Compound I I a was oxidized with NaIOl to the 8-hydroxyetiolactone (VIa) m.p. (needles from 3700, 1780, moist acetone) 232-238", v::;jx3 1095, 1085, 1022, 998, 980-965 (multiple), $114,
+
0
u .
w
€io
.
Ia,R=H IIIa, IIIb,R=Ac
IIa, I I b
w
0
I v a . IYb
VIa. VIb
0 V
a tetrahydro derivative during incubation of dlaldosterone with rat liver homogenates. Ulick and Lieberman3 have reported the isolation from human urine of a C2105 pregnane derivative possessing hydroxyl groups a t C3, CISand CZl,and two carbonyl groups, one a t Go; however, definite characterization and assignment of structure was not possible. Synthetic d-aldosterone 21-acetate, m.p. 195.5201') [ a ] ~ 130°,4was incubated with a homogenate of rat liver, enzymatically reduced T P N H and a T P N H generating system5 for three hours a t 37" under nitrogen. The incubate was extracted with organic solvents, and the extracts were chromatographed on paper using a new paper chromatography system of aqueous diethylacetamide.6 Compound Ia migrated with an RE = 0.60' and gave positive color tests with blue tetrazolium chloride,8 with SbC1:lYand with Zimmermann's dinitrobenzene reagent.I0 There was no absorption in the ultraviolet. Compound IIa migrated with an RE = 0.21 and gave positive color tests with blue tetrazolium chloride and with SbC13, a negative test with Zimmermann's reagent, and no absorption in the ultraviolet. These compounds were eluted from paper chroma( 3 ) s. Click and s. Lieberman, THISJ O U R N A L , 79, 6567 (1957).
+
(4) Melting points (uncorrected) were determined on a Kofler type h o t stage. Infrared spectra were determined using an Infracord Model 137 Spectrophotometer. Microanalyses were performed b y Dr. Alfred Bernhardt, Max Planck Institute, Mulheim, Germany. ( 5 ) "Methods in Enzymology," Edited by S. P. Colowick and X . 0. Kaplan, Academic Press, New York. S.Y., 1955, p . 323. (C) I n this report, System S o . 7 , diethylacetamide: water:isooctane: tuluene (1 : 2 : 3 : 31, was used This is one of a series of aqueous dimethyl- a n d diethylacetamide systems which have been in use in our laboratory during the past two years. These systems will be described in full in a forthcoming paper. (7) RE = distance traveled by sample relative t o E (cortisone) on paper chromatograms scanned with blue tetrazolium. (8) K. Savard, J . BioL, Chern., 202, 457 ( l W 3 ) . (9) G. M. Shull, J. L. Sardinos and 12. C. Nubel, A r c h . Biochem. and Biofiiiys., 37, 186 ( l Y 5 2 ) . ( 1 0 ) R . B. Burton, A . Zaffaroni and E. H. Keutmano, J . B i d C h e m . , 88, 7GY (l!i5l)
+
(11) S. A. Simpson, J . F. T a i t . A . Wettstein, K. Xeher, J . V. E U W , 0. Schindler and T . Reichstein. Hela. Chim A ( l 1 7 .3 7 , 1200 (1954).
Oct. 5, 1960
COMMUNICATIONS TO THE EDITOR
5253
creased, since the molar volume of the group -C(OH)=CHis smaller than that of the group -COCH2by about 4.0 ml./mole (estimated from the densities of unsaturated alcohols and isomeric carbonyl compounds) ; if the internnl Hbond is taken into account, this difference would undoubtedly be somewhat greater. If the ring formation has the positive volume requirement RESEARCH INSTITUTE FOR M. M . PECHET described above, the molar volume of the enol MEDICINEAND CHEMISTRY R . H. HESSE form should be larger than that of the keto form H . KOHLER by about 5 ml./mole. The effect of pressure on CAMBRIDGE 42, MASS., A N D MASSACHUSETTS GENERALHOSPITAL this equilibrium was studied by Kabachnik, BOSTON,MASS. Yakushkima and Kislyakova,’ who reported no RECEIVEDAUGUST4, 1960 significant change in K for the pure substance. Since the rate of interconversion is extremely T H E KETO-ENOL EQUILIBRIUM OF ETHYL variable (depending on traces of impurities”) ACETOACETATE UNDER H I G H PRESSURE i t appeared desirable to repeat this work. The Sir : new data (Table I) show that K in the pure liquid It has been demonstrated in recent years that increases with pressure; the molar volume of the studies of the effect of pressure on reaction rates enol form exceeds that of the keto form by 1.0in the liquid phase often permit certain conclusions 1.5 ml./mole. This would suggest that it may be possible to recognize reactions involving a cyclic to be drawn regarding the mechanism involved. For instance, reactions are accelerated if they transition state by a retarding effect of pressure. proceed through a transition state that has a An experimental program to test this statement greater separation of charge than the reactant(s) is about to start in this laboratory. and vice versa, since the intense electric field of an TABLE I ion causes local compression of solvent molecules.2 K AV I n the expression d In kldp = -AV*/RT, the fi(atm ) voenolb keto/enol (ml /mole) c volume of activation is negative in such cases. 1 7.7 12.0 -1.5 On the other hand, a reaction is decelerated 1350 7.2 13.0 -1.0 ( A V* is positive) if the formation of the transition 2500 7.0 13.2 -1.3 state depends on homopolar bond breaking and 3700 6.4 14.6 vice versa,3 since the sum of the van der Waals The apparatus will be described a t a later date. At radii of the fragments exceeds that of the re- 25’. Samples of the ester were withdrawn without releasing the pressure, collected in a quartz vessel and analyzed actant. a t once in the usual way ( R . H.Meyer and P. KappelIt appeared that another feature often of interest meyer, Ber., 44, 2718 (1911)) Care was taken t o insure in mechanistic studies, i e . , formation of a cyclic the system was a t equilibrium and that interconversion transition state, could be subject to a pressure during analysis was negligible. Calculated from (In K, effe~t.A ~ comparison of the densities of straight In K 1 ) = - A V / R T . chain hydrocarbons shows that the molar volume of Acknowledgment.-This work was done a t the n-CmHPm+2 exceeds that of n-Cm--6H2,,-10 by 96 Rohm and Haas Research Laboratories. The f 1 ml./mole (m ranging from 11 to 17). This author had the benefit of many helpful discussions difference is a measure of the volume of hexa- with Mr. 0. H. Loeffler and Drs. C. Huggett and methylene, -(CH2)6-. This value is 12 ml./ R . Iwanciow. mole smaller than the molar volume of cyclohexane ( 7 ) br. I . Kahachnik, S. E. Yakushkima and N. V . Kislyakova, (108 ml./mole). Parachor data5 on ring structures Doklady A k a d . X a u k . , S.S.S.R., 96, 1169 (1954); rf. C . A . , 49, 8815 similarly suggest that ring structures have greater (1955). The authors stated t h a t the system was allowed 4 hours to volume requirements than straight chains. Pre- reach equilibrium; in this work i t was found t h a t a t least 20 hours sumably the core of such doughnut-like molecules was necessary. is too small to be accessible to other molecules. STATE UNIVERSITYOF NEW YORK CENTER %-ILLIAMJ. LE NOBLE I n this work, the keto-enol equilibrium was LONGISLAKD studied as an example. If the process of forming OYSTERBAY,NEW YORK RECEIVEDAUGUST23, 1960 a cyclic structure such as the enol form of ethyl acetoacetateB does not contribute to the change in molar volume, the equilibrium constant ( K T H E STEREOCHEMISTRY OF JACOBINE keto/enol) should decrease as the pressure is in- Sir. (1) For an excellent review, see S. D. Hamann, “Physico-Chemical Recent investigations‘ have shown all previous Effects of Pressure,” Academic Press, Inc., New York, N. Y., 1957. structures proposed for jacobine, jaconecic acid (2) H . C . David and S. D. Hamann, Trans. Fav. SOC.,SO, 1188 and isojaconecic acid to be incorrect. The struc(1952!. (3) A. E . Nicholson and R. G. W. Norrish, Disc. Far SOC.,97 tures of these compounds are correctly represented (1956). by I, I1 and 111, respectively. We wish now to (4) C . Walling aod J. Peisach have considered this possibility present evidence which permits assignment of recently in the dimerization of isoprene t o cyclic products ( T m s J O U R N A L 80, 5819 (1958)). stereochemistry to the above compounds, as ( 5 ) S. Sugden, “The Parachor and Valency,” Geo. Rutledge & shown in l a , IIa and IIIa.
SO0 cm.-l, w-hich was identical with VIb, the 3hydroxyetiolactone prepared from the synthetic etiolactone (V). The isolation and characterization of 5a-(4,5)dihydroaldosterone (Ia) and 3$0H,Sa-(4,5)tetrahydroaldosterone (Ira) indicates that the ring A reduced products with A/B trans ( 5 a ) in configuration are favored.
’
Q
0
Sons, L t d . , London, 1929. (6) G. W. Wheland, Ch. X I V , “Advanced Organic Chemistry,” 2nd Edition. John Wiley &Sons, Inc., h-ew York, N. Y.,1949.
(1) R. E. Bradbury and S. Masamune, T i m J O U R N A L81, , 5201 (19591, andalso see T. A. Geissman. Ausl. J. Chcm., 18, 247 (1959).