A new strategy for stereocontrol of cation-olefin cyclization. The first chemical emulation of the A/B-trans-9,10-syn-folding pathway of steroid biosynthesis from 2,3-oxidosqualene

Full text of DOI:10.1021/ja9632926

11982
J. Am. Chem. Soc. 1996, 118, 11982-11983
A New Strategy for Stereocontrol of Cation-Olefin
Cyclization. The First Chemical Emulation of the
A/B-trans-9,10-syn-Folding Pathway of Steroid
Biosynthesis from 2,3-Oxidosqualene
E. J. Corey* and Harold B. Wood, Jr.
Department of Chemistry and Chemical Biology
HarVard UniVersity, Cambridge, Massachusetts 02138
ReceiVed September 19, 1996
The enzymic cyclization of (S)-2,3-oxidosqualene (1) to the
protosterol cation (2)1,2 has never been emulated in a chemical
system, in part because the A/B-trans, 9,10-syn, B/C-trans
stereochemistry of 2, which requires boat geometry for the B
ring, is several kcal/mol less stable than the alternative trans-
anti-trans A/B/C arrangement. In general, chemically conducted
Figure 1. X-ray crystal structure of TBS-enol ether 7.
cation-olefin cyclizations have led to trans-anti-trans type fused
ring systems (triterpene folding) exclusively.1a,3 We report
herein on a substrate design which favors cyclization to A/B-
trans, 9,10-syn product (steroid numbering) and which suggests
a new approach to the control of stereochemistry and olefinic
face selectivity in cation-olefin cyclizations.
equiv of tert-butyldimethylsilyl triflate, TBSOTf, and 2.3 equiv
of 2,6-lutidine in THF at -78 °C for 1 h), (2) selective
hydrogenation of the 14,15-double bond (H2, Pd-C, EtOH, 25
°C, 3 h), (3) oxidative cleavage of the C(7) exocyclic vinyl ether
to form 6 (0.05 equiv of RuCl3, 4 equiv of NaIO4 in CCl4-
CH3CN-H2O at 25 °C for 1.5 h), and (4) transformation of
ketone 6 to the enol ether 7 (1.3 equiv of TBSOTf and 2 equiv
of KHMDS in THF at -78 °C for 45 min).
A crystal of 7 was subjected to X-ray crystallographic analysis
which revealed the three-dimensional structure shown in Figure
1.7 The structure of 7 involves the syn-relationship between
the H at C(9) and the CH3 at C(10) and A/B-trans geometry in
common with the protosterol system. The formation of 7
therefore must involve the same olefinic face selectivity (i.e.,
substrate folding) which is involved in forming the A- and
B-rings during sterol biosynthesis. To the best of our knowledge
this is the first time that this stereochemical pathway has been
demonstrated for a non-enzymic cation-olefin cyclization. The
cis-fusion of the B- and C-rings, which provides a clear
indication that C-ring formation is not concerted with B-ring
closure, reflects the preferred face selectivity for cation-olefin
closure to form the C-ring in this particular system. Evidently,
the bicyclic B-ring cation 8 is a discrete intermediate, and attack
by the terminating π-system occurs preferentially at the R-face
of C(8) of the bicyclic oxaallylic cation. That mode of closure
also controls the C/D-ring fusion of 7 to be cis.
The specific substrate that was the focus of this investigation
was the chiral oxirane 4, which was synthesized as a 1:1 mixture
of E- and Z-isomers4,5 at the vinyl trifluoroethyl ether subunit
by Wittig olefination of the corresponding ketone (3) (Ph3P+-
CH2OCH2CF3 Br-, potassium tert-amyl oxide, C7H8, 25 °C,
4.5 h). Treatment of the 1:1 mixture of E- and Z-4 with
MeAlCl2 (2 equiv) in CH2Cl2 at -95 °C for 1 h produced a
single tetracyclic hydroxy R,â-enone (5) which was obtained
pure in 40% yield after flash column chromatography on silica
gel. The structure of 5 was determined unambiguously as
described below.6 It is likely that 5 arises from the E-form of
4 because of the correspondence of vinyl ether geometry. It
may also be that tetracyclic products are not formed from the
Z-form of 4 because such structures would be severely
destabilized by steric repulsion between C(15) of the D-ring
and the trifluoroethoxy group. On the basis of this assumption,
the yield of the cyclization product would be 80%.
The oily cyclization product 5 was converted to the crystalline
tetracyclic ketone bis-silyl ether 7, mp 145.5-146 °C, by the
following sequence: (1) silylation of the 3-hydroxyl group (1.3
(1) (a) Johnson, W. S. Tetrahedron 1991, 47, xi. (b) Abe, I.; Rohmer,
M.; Prestwich, G. D. Chem. ReV. 1993, 93, 2189.
(2) (a) Corey, E. J.; Virgil, S. C. J. Am. Chem. Soc. 1991, 113, 4025.
(b) Corey, E. J.; Virgil, S. C.; Sarshar, S. J. Am. Chem. Soc. 1991, 113,
8171. (c) Corey, E. J.; Virgil, S. C.; Cheng, H.; Baker, C. H.; Matsuda, S.
P. T.; Singh, V.; Sarshar, S. J. Am. Chem. Soc. 1995, 117, 11819.
(3) (a) Sutherland, J. K. In ComprehensiVe Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 3, p 341. (b)
Taylor, S. K. Org. Prep. Proced. Intl. 1992, 24, 245. (c) Corey, E. J.; Lee,
J. J. Am. Chem. Soc. 1993, 115, 8873. (d) Corey, E. J.; Lee, J.; Liu, D. R.
Tetrahedron Lett. 1994, 35, 9149. (e) Corey, E. J.; Lin, S. J. Am. Chem.
Soc. 1996, 118, 8765.
(4) The 1:1 mixture of E- and Z-isomers of 4 could not be separated
chromatographically. Further research is planned toward the objective of
the stereoselective conversion of 3 to the E-isomer of 4.
We believe that there is a logical mechanistic basis for the
realization of the biomimetic A/B-trans, 9,10-syn cyclization
(5) The methyl vinyl ether analog of 4 was not a satisfactory substrate
for cyclization because the basicity of the methoxy group interfered by
coordinating to the Lewis acidic reagent.
(7) The coordinates of 7 and other X-ray data can be obtained, on request,
from the Director, Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge, CB2 1EZ, U.K.
(6) The E-geometry of the C(7) exocyclic double bond of 5 was also
1
revealed by H NMR NOE data.
S0002-7863(96)03292-1 CCC: $12.00 © 1996 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 118, No. 47, 1996 11983
Scheme 1
as the use of the epoxide-initiated cyclization of a polyene to
achieve direct formation of the protosterol system.
The synthesis of the intermediate 3 which was required for
this study was accomplished by the route outlined in Scheme
1. δ-Keto ester 9, prepared by enantioselective Michael addition
according to Pfau et al. (90% ee, 70% yield),9 was converted
via the corresponding R,â-enone (10)10 to the cyclopentadiene
ester 11. Selective reduction of the ester function of 11 gave
the aldehyde 12 which upon Horner-Emmons-Wittig chain
extension produced the E-R,â-unsaturated ester 13. Dibal
reduction of the ester 13 followed by Griffith-Ley oxidation
with N-methylmorpholine N-oxide and catalytic tetra-N-propyl-
ammonium perruthenate11 gave the aldehyde 14 which was
converted to the TMS ether of the corresponding cyanohydrin
15. Alkylation of 15 with (S)-6,7-oxidogeranyl bromide12
provided the coupling product 1613 which by base-catalyzed
deprotection afforded ketone 3.
Figure 2. Stereopair representations of alternative transition states in
the cyclization of 4. Top: 9,10-syn. Bottom: 9,10-anti.
Conclusion. A trifluoroethoxymethylene group placed in
conjugation with an existing double bond, as in substrate 4,
dramatically alters the face selectivity of cation-initiated cy-
clization at that double bond. This phenomenon, the basis of
which is clarified by the analysis of nonbonded steric repulsion
in a transition state model, allows synthetic access to polycyclic
products which are not readily accessible by other means.
Cation-olefin polyannulation, already an unusually powerful
synthetic construction, thereby acquires additional versatility.
pattern in the chemical conversion of 4 to 5. Figure 2 presents
stereopair representations of possible transition states for the
sterol folding (9,10-syn) (top) and triterpene folding (9,10-anti)
(bottom) pathways for the closure of ring B. In both transition
states there is planar (i.e., most stable) geometry of the
conjugated diene and a 3 Å separation of the carbons which
bond during B-ring cyclization.8 The transition state for 9,10-
syn-cyclization (top) is clearly more favorable energetically than
the transition state for 9,10-anti-ring closure (bottom), since the
latter involves relatively severe steric repulsion, especially
between the methyl group on the cationic center and the methyl
substituent on the conjugated diene reaction partner. Thus, the
strategic introduction of an additional olefinic linkage into a
substrate for cation-olefin cyclization can cause an alteration
of transition state geometries and a dramatic shift in the preferred
stereochemical course of cyclization. Experiments are in
progress to demonstrate the effectiveness of this new strategy
in the solution of hitherto formidable synthetic problems, such
Acknowledgment. We are grateful to the National Institutes of
Health and the National Science Foundation for generous financial
support. The X-ray crystallographic analysis of 7 was carried out by
Dr. Marcus Semones.
Supporting Information Available: Experimental procedures,
spectroscopic data for compounds 3-7 and 9-16 and X-ray data for
7 (19 pp). See any current masthead page for ordering and Internet
access information.
JA9632926
(8) The reaction barrier in cation-olefin addition is mainly due to solvent
reorganization (cation desolvation-resolvation) and steric effects. In the
gas phase there is essentially no barrier to the addition of the tert-butyl
cation to isobutylene which is 6.5 kcal/mol exothermic at a C-C separation
of 3 Å (Prof. William L. Jorgensen, Yale University, personal communica-
tion, March, 1996). For such highly exothermic (ca. 20 kcal/mol), low
activation enthalpy cyclizations, an early transition state is likely and a
bonding C-C distance of 3 ( 0.3 Å seems reasonable.
(9) Pfau, M; Revial, G.; Guingnat, A.; d’Angelo, J. J. Am. Chem. Soc.
1985, 107, 273.
(10) Minami, I.; Takahashi, K.; Shimizu, I.; Kimura, T.; Tsugi, J.
Tetrahedron 1986, 42, 2971.
(11) Griffith, W. P.; Ley, S. V. Aldrichimica Acta 1990, 23, 13.
(12) Corey, E. J.; Noe, M. C.; Shieh, W.-C. Tetrahedron Lett. 1993, 34,
5995.
(13) Stork, G.; Maldonado, L. J. Am. Chem. Soc. 1971, 93, 5286.