Steroids and steroid analogues from Stille-Heck coupling sequences

Article abstract of DOI:10.1002/anie.200352162

Tricyclic 1,3,5-hexatrienes were prepared from 4-substituted 2-bromocyclohexenyl triflate, enantiomerically pure hexahydroindenylstannanes, and tert-butyl acrylate in a sequence of Stille and Heck reactions. In the example shown, an enantiomerically pure

Full text of DOI:10.1002/anie.200352162

Angewandte
Chemie
Oligocyclic Dienes
Steroids and Steroid Analogues from Stille–Heck
Coupling Sequences**
Hans Wolf Sünnemann and Armin de Meijere*
Steroids have been synthesized in many different ways,[1] and
to a certain extent they have become one of the preferred
testing grounds for the development of more efficient
methods for organic synthesis. A maximum increase of
structural complexity in a minimum number of operational
steps in synthesis can be achieved either by completely new
transformations or by new sequential combinations of pre-
viously established reactions. A striking example is the total
synthesis of estrone by Vollhardt et al., in which a cobalt-
mediated intramolecular cocyclization of two alkynes and an
alkene is applied followed by an intramolecular Diels–Alder
reaction.[2] Our previously reported selective Stille–Heck
coupling sequences of 1-bromo-2-trifluoromethanesulfonyl-
oxycyclohexenes with alkenylstannanes and alkyl acrylates to
yield unsymmetrically 1,6-disubstituted 1,3,5-hexatrienes,
which can be transformed by 6p-electrocyclizations into
oligocyclic dienes,[3] appeared to be suitable to access steroids
and steroid analogues efficiently by a convergent A + CD!
ACD!ABCD strategy.
Scheme 1. A) NH3, Li (2.30 equiv), PhNH2 (0.75 equiv), THF, ꢀ338C,
2 h; B) CuBr·SMe2 (1.00 equiv), tBuLi (1.10 equiv), DIBAH
(1.50 equiv), HMPA (8.00 equiv), TMSCl (2.00 equiv), THF, ꢀ100!
ꢀ408C, 6 h; MeLi (1.50 equiv), THF, 08C, 45 min; C) PhN(Tf)2 (1.10–
2.40 equiv), THF, ꢀ78!258C, 22–24h; D) LDA (2.60 equiv), Bu3SnH
(2.20 equiv), CuCN (1.10 equiv), THF, ꢀ308C, 2.5 h; AgOAc
(3.00 equiv), EtOAc, 258C, 1 h; E) [Pd2(dba)3·CHCl3 (10 mol%), AsPh3
(8 mol%), CuI(5 mol%), LiCl (3.00 equiv), NMP, 65 8C, 5 h; F) 7
(8 mol%), nBu4NOAc (2.00 equiv), tert-butyl acrylate (5.00 equiv),
DMF, MeCN, H2O (5:5:1), 1058C, 4 h; G) decalin, 2158C, 45 min.
DIBAH=diisobutylaluminum hydride, HMPA=hexamethylphosphor-
amide, TMS=trimethylsilyl, Tf=trifluoromethanesulfonyl, LDA=
lithium diisopropylamide, NMP=N-methylpyrrolidone.
The sequence (Scheme 1) starts with a Stille coupling[4] of
the enantiomerically pure hexahydroindenylstannanes cis-2
and trans-2 onto the 4-substituted 2-bromocyclohexenol
triflates 3a,b to yield the tricyclic bromobutadienes cis-4a,b
and trans-4a,b, respectively. The bicycloalkenylstannanes cis-
2 and trans-2 were prepared in high yields (82–97%) with a
tributylstannyl cuprate[5] from the corresponding cis-[6] and
trans-configurated[7] enol triflates, which were obtained in
55–79% yield[8] by reductive enolate formation from the a,b-
unsaturated bicyclic ketone 1.[9] The bromocyclohexenol
triflate 3a was assembled from p-methoxyphenol by Birch
reduction,[10] and bromination,[11] (74% yield) regioselective
formation of the lithium enolate, and trapping with N,N-
bis(trifluoromethanesulfonyl)aniline (69% yield).[3] The
starting compound for the synthesis of 3b was 1,4-cyclo-
hexanedione monoethyleneacetal, which was converted into
the trimethylsilyl enol ether.[12] Since the bromination product
2-bromo-1,4-cyclohexanedione monoethyleneacetal is not
stable, it was transformed directly into the bromocyclohex-
enol triflate 3b (53% yield over two steps from the
trimethylsilyl enol ether).[13] Of the reaction conditions
tried, the protocol for Stille couplings developed by Farina
et al.[14] with triphenylarsane[15] as a ligand and a copper(i)
cocatalyst proved to be best and gave yields in the range of
70–86% (Table 1). From the racemic 4-methoxy derivative 3a
the products cis-4a and trans-4a were obtained as 1:1
mixtures of diastereomers. The tricyclic bromodienes cis-
4a,b and trans-4a,b were immediately subjected to subse-
quent Heck reactions with tert-butyl acrylate[16] to form the
Table 1: Chemoselective Stille and Heck coupling reactions of 4-
substituted 2-bromocyclohexenol triflates leading to tricyclic 1,3,5-
hexatrienes (see Scheme 1).
[*] Dipl.-Chem. H. W. Sünnemann, Prof. Dr. A. de Meijere
Institut für Organische Chemie der
Starting cmpd.
Alkene
Product
Yield [%][a]
d.r.[b]
Georg-August-Universität Göttingen
Tammannstrasse 2, 37077 Göttingen (Germany)
Fax: (+49)551-399475
E-mail: Armin.deMeijere@chemie.uni-goettingen.de
3a
3b
3a
3b
cis-2
cis-2
trans-2
trans-2
cis-4a
cis-4b
trans-4a
trans-4b
cis-5a
80
83
70
86
74
1:1
–
1:1
–
[**] This work was supported by the Fonds der Chemischen Industrie as
well as by the companies BASF AG and Chemetall GmbH
(chemicals). The authors are indebted to Dr. E. Ottow, Schering AG,
Berlin, for a generous gift of the Hajos-Wiechert ketone 1, and to Dr.
Burkhard Knieriem, Göttingen, for his careful proofreading of the
final manuscript.
cis-4b
1:1
cis-4b
cis-5b
76
79
78
–
trans-4b
trans-4b
trans-5a
trans-5b
1:1
–
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
[a] Yield of isolated product. [b] The diastereomeric ratio was determined
by 1H NMR spectroscopy.
Angew. Chem. Int. Ed. 2004, 43, 895 –895
DOI: 10.1002/anie.200352162
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
895

Communications
unsymmetrically substituted tricyclic 1,3,5-hexatrienes cis-
5a,b and trans-5a,b, respectively.
The best catalyst for this step turned out to be the
palladacycle 7[17] in the presence of tetrabutylammonium
acetate as a base. For higher yields (74–79%), a solvent
mixture consisting of DMF, MeCN, and water[18] was required
(Table 1). The methoxy-substituted hexatrienes trans-5a and
cis-5a were also isolated as 1:1 mixtures of diastereomers.
The tricyclic 1,3,4,5,6-pentasubstituted hexatrienes cis-
5a,b and trans-5a,b are reasonably well set up for an
additional ring closure by thermal 6p-electrocyclization
(Scheme 1).[19,20] The best conditions for this turned out to
be by heating in decalin solution at temperatures in the range
215–2208C.[21] After 30 min at 2158C, the hexatriene cis-5b
had been completely converted into a mixture of the expected
product of the 6p-electrocyclization and that arising from a
subsequent 1,5-hydrogen shift, the apparently thermodynami-
cally more stable isomer cis-6b. Upon extended heating, cis-
6b was the sole product. Heating the hexatrienes cis-5a,b in
decalin at 215–2208C for 45 min provided the steroid
analogues cis-6a,b cleanly with yields in the range of 71–
75% (Table 2). Thus, the new sequence provides interesting
new 7-substituted 19-nor-steroid analogues with an unnatural
cis CD-ring fusion.
Scheme 2. A) Decalin, 2158C, 30 min.
ylation without and with subsequent aromatization, respec-
tively, of the B ring (Scheme 2).
It is noteworthy that all the 6p-electrocyclizations occur
with a high degree of outward disrotational selectivity,[22] as
the tetracycles cis-6b, trans-6b, and trans-8b were all isolated
as single diastereomers.[23] These steroidal products ought to
be enantiomerically pure as well, since the bicycloalkenyl-
stannanes cis-2 and trans-2 were enantiomerically pure
starting materials and any racemization along the route
would not be probable.
Table 2: Thermal 6p-electrocyclizations of tricyclic 1,3,4,5,6-pentasub-
stituted 1,3,5-hexatrienes.
Starting cmpd.
Product
Yield [%][a]
d.r.[b]
[a]2D0 [c]
cis-5a
cis-5b
trans-5b
cis-6a
cis-6b
trans-8b
75
1:1
–
–
+54.5 (10.0)
+49.7 (6.00)
+27.0[d] (10.0)
71
63[d]
Received: June 18, 2003 [Z52162]
[a] Yield of isolated product. [b] The diastereomeric ratio was determined
1
by H NMR spectroscopy. [c] Concentration c [mgmLꢀ1] in parentheses.
Keywords: cross-coupling · homogeneous catalysis ·
pericyclic reactions · steroids · total synthesis
.
[d] Plus 15% of trans-6b, which has an optical rotation of [a]2D0 =+24.4
(12.4 mgmLꢀ1, C6H6).
[1] a) S. N. Ananchenko, I. V. Torgov, Tetrahedron Lett. 1963, 1553 –
1558; b) N. Cohen, B. L. Banner, W. F. Eichel, D. R. Parrish, G.
Saucy, J.-M. Cassal, W. Meier, A. Fürst, J. Org. Chem. 1975, 40,
681 – 685; c) Y. Zhang, E.-i. Negishi, J. Am. Chem. Soc. 1989,
111, 3454 – 3456; d) H. Nemoto, N. Matsuhashi, M. Imaizumi, M.
Nagai, K. Fukumoto, J. Org. Chem. 1990, 55, 5625 – 5631;
e) C. D. Dzierba, K. S. Zandi, T. Möllers, K. J. Shea, J. Am.
Chem. Soc. 1996, 118, 4711 – 4712; f) K. C. Nicolaou, D. Vour-
loumis, N. Winssinger, P. S. Baran, Angew. Chem. 2000, 112, 47 –
126; Angew. Chem. Int. Ed. 2000, 39, 44 – 122.
[2] a) R. L. Funk, K. P. C. Vollhardt, J. Am. Chem. Soc. 1980, 102,
5253 – 5261; b) R. L. Funk, K. P. C. Vollhardt, J. Am. Chem. Soc.
1979, 101, 215 – 217.
[3] P. v. Zezschwitz, F. Petry, A. de Meijere, Chem. Eur. J. 2001, 7,
4035 – 4046.
[4] For reviews on the Stille reaction see: a) T. N. Mitchell in Metal-
Catalyzed Cross-Coupling Reactions (Eds.: F. Diederich, P. J.
Stang), Wiley-VCH, Weinheim, 1998, pp. 167 – 202; b) V. Farina,
V. Krishnamurthy, W. J. Scott, Org. React. 1997, 50, 1 – 652;
c) T. N. Mitchell, Synthesis 1992, 803 – 815; d) M. Kosugi, K.
Fugami in Handbook of Organopalladium Chemistry for
Organic Synthesis (Eds.: E.-i. Negishi, A. de Meijere), Wiley,
New York, 2002, pp. 263 – 284.
Heating the isomer trans-5b with trans CD-ring fusion at
2208C for 30 min led to a separable mixture of the primary
6p-electrocyclization product trans-8b and the product of a
subsequent 1,5-hydrogen shift, trans-6b, in a ratio of 4:1
(Scheme 2). The overall yield is 78%, and the 63% yield of
the main product trans-8b is preparatively satisfying. When a
sample of trans-6b was heated in decalin at 2158C for 45 min,
trans-8b was again the main product, which proves reversi-
bility for the 1,5-hydrogen shift. In view of this, complete
transformation of trans-6b into trans-8b may be possible.
Because of the trans relationship of the hydrogens on C8 and
C14 and the trans CD-ring fusion, this product is more closely
related to the natural steroids than cis-6a,b.
Variation of the temperature (190–2608C) and extended
heating (up to 90 min) of trans-5b (Table 2) did not signifi-
cantly affect the ratio of two regioisomers trans-8b and trans-
6b, unlike the observation for the hexatrienes cis-5a,b.
However, at 2608C, the hexatriene trans-5b furnished not
only a mixture of the regioisomeric cyclohexadienes trans-8b
and trans-6b but also significant amounts (10–15%) of the
products 9 and 10, which are formed by de-tert-butoxycarbon-
[5] S. R. Gilbertson, C. A. Challener, M. E. Bos, W. D. Wulff,
Tetrahedron Lett. 1988, 29, 4795 – 4798.
896
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 895 –895

Angewandte
Chemie
[6] J. E. McMurry, W. J. Scott, Tetrahedron Lett. 1983, 24, 979 – 982.
[7] U. Groth, T. Köhler, T. Taapken, Tetrahedron 1991, 47, 7583 –
7592.
[8] For the trans-configurated enol triflate the yield was determined
over two steps since the diisobutylaluminum enolate was first
trapped as the trimethylsilyl enol ether, which was cleaved with
MeLi to form the more reactive lithium enolate.
[9] a) Z. G. Hajos, D. R. Parrish, J. Org. Chem. 1974, 39, 1612 – 1614;
b) Z. G. Hajos, D. R. Parrish, J. Org. Chem. 1974, 39, 1615 – 1621;
c) R. A. Micheli, Z. G. Hajos, N. Cohen, D. R. Parrish, L. A.
Portland, W. Sciamanna, M. A. Scott, P. A. Wehrli, J. Org. Chem.
1975, 40, 675 – 681.
[10] J. A. Marshall, G. A. Flynn, Synth. Commun. 1979, 9, 123 – 127.
[11] K. A. Parker, H.-J. Kim, J. Org. Chem. 1992, 57, 752 – 755.
[12] W. J. Kerr, M. McLaughlin, A. J. Morrison, P. L. Pauson, Org.
Lett. 2001, 3, 2945 – 2948.
[13] P. L. Stotter, K. A. Hill, J. Org. Chem. 1973, 38, 2576 – 2578.
[14] V. Farina, B. Krishnan, D. R. Marshall, G. P. Roth, J. Org. Chem.
1993, 58, 5434 – 5444.
[15] More recent experiments showed that triphenylarsane does not
play any significant role in the Stille coupling of 3b with trans-2.
Even without the presence of AsPh3 the product trans-4b was
obtained in a comparable yield of 84%.
[16] Reactions with tert-butyl acrylate proved to give consistently
better yields those with methyl acrylate, from which early
precipitation of palladium black was observed. This may be
attributed to a better stabilization of the palladium(o) catalyst by
the less electron-deficient tert-butyl acrylate. Cf. a) M. Buback,
T. Perkovic, S. Redlich, A. de Meijere, Eur. J. Org. Chem. 2003,
2375 – 2382; b) T. Perkovic, Dissertation, Universität Göttingen,
1999.
[17] The palladacycle was prepared from Pd(OAc)2 and (oTol)3P:
a) W. A. Herrmann, C. Brossmer, K. Öfele, C.-P. Reisinger, T.
Priermeier, M. Beller, H. Fischer, Angew. Chem. 1995, 107,
1989 – 1992; Angew. Chem. Int. Ed. Engl. 1995, 34, 1844 – 1848;
b) W. A. Herrmann, C. Brossmer, C.-P. Reisinger, T. Riermeier,
K. Öfele, M. Beller, Chem. Eur. J. 1997, 3, 1357 – 1364.
[18] Cf. a) F. Ozawa, A. Kubo, T. Hayashi, Chem. Lett. 1992, 2177 –
2180; b) S. Lemaire-Audoire, M. Savignac, C. Dupuis, J.-P. GenÞt
Tetrahedron Lett. 1996, 37, 2003 – 2006; c) F. Y. Zhao, M. Shirai,
M. Arai, J. Mol. Catal. A 2000, 154, 39 – 44; d) L. F. Tietze, T.
Nöbel, M. Spescha, Angew. Chem. 1996, 108, 2385 – 2386;
Angew. Chem. Int. Ed. Engl. 1996, 35, 2259 – 2261.
[19] Cf. a) T. L. Gilchrist, R. J. Summersell, J. Chem. Soc. Perkin
Trans. 1 1988, 2595 – 2601; b) A. Lansky, Dissertation, Universi-
tät Göttingen, 1992.
[20] K. Voigt, P. von Zezschwitz, K. Rosauer, A. Lansky, A. Adams,
O. Reiser, A. de Meijere, Eur. J. Org. Chem. 1998, 1521 – 1534.
[21] When the hexatrienes were heated at lower temperatures (150–
1608C), the reaction required significantly longer times. For
example, even after 12 h traces of the starting materials could be
detected and, besides the desired products, several side products
were formed.
[22] a) B. M. Trost, P. G. McDougal, J. Org. Chem. 1984, 49, 458 – 468.
This rotational selectivity has also been termed torquoselectivity.
Cf. b) E. A. Kallel, Y. Wang, D. C. Spellmeyer, K. N. Houk, J.
Am. Chem. Soc. 1990, 112, 6759 – 6763; c) J. D. Evanseck, B. E.
Thomas IV, D. C. Spellmeyer, K. N. Houk, J. Org. Chem. 1995,
60, 7134 – 7141.
[23] The structures were confirmed by 1H–1H NOESY NMR experi-
ments in which interactions within the steroid analogue cis-6b
between the protons of the methyl group at C13 and the proton
at C14 as well as between the proton at C14 and the proton at C7
were detected. For the steroid trans-8b interactions between the
methyl group at C13 and the proton at C8 could be observed, and
also interactions between the proton at C7 and the proton at C8
as well as the proton at C14 and one proton at C15 were found.
Angew. Chem. Int. Ed. 2004, 43, 895 –897
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
897