Studies Directed toward Asymmetric Synthesis of Cardioactive Steroids via Anionic Polycyclization

Article abstract of DOI:10.1021/ol027125o

(Matrix Presented) The use of anionic polycyclization (AP) in constructing the steroidal backbone of cardenolides was investigated. The reaction of 2-carbomethoxy-2-cyclohexenone I with the enolate of Nazarov reagent II gave, after decarboxylation and ald

Full text of DOI:10.1021/ol027125o

ORGANIC
LETTERS
2002
Vol. 4, No. 26
4693-4696
Studies Directed toward Asymmetric
Synthesis of Cardioactive Steroids via
Anionic Polycyclization
Zhixiang Yang, Dean Shannon, Vouy-Linh Truong, and Pierre Deslongchamps*
Laboratoire de synthe`se organique, Institut de pharmacologie de Sherbrooke,
UniVersite´ de Sherbrooke, Sherbrooke, Que´bec J1H 5N4, Canada
pierre.deslongchamps@usherbrooke.ca
Received October 17, 2002
ABSTRACT
The use of anionic polycyclization (AP) in constructing the steroidal backbone of cardenolides was investigated. The reaction of 2-carbomethoxy-
2-cyclohexenone I with the enolate of Nazarov reagent II gave, after decarboxylation and aldol condensation, steroid III with control of
stereochemistry.
In 1988, our research group reported a new stereocontrolled
synthesis of polycyclic compounds.1 We have now found
that the reaction between cyclohexenone 1 and the enolate
of Nazarov reagent 2 affords, after selective decarboxylation,
steroid 5 in good yield and with very good diastereoselec-
tivity (Scheme 1). This tandem double Michael addition/
aldol condensation, which we call the anionic polycycliza-
tion, is a very efficient method for the construction of
polycyclic compounds stereoselectively.
Scheme 1a
Construction of steroid 5 was very encouraging; however,
since most naturally occurring cardioactive steroids contain
a â-hydroxyl at C-14, along with the cis A/B and C/D ring
fusions,2 it would be very interesting to pursue this stereo-
chemistry. One such example is the natural product ouabain
(6a, Figure 1), which is of particular interest because it has
shown the potential for treating hypertension as well as
congestive heart failure.3 This glycosidic steroid, extracted
a Reagents and conditions: (a) Cs2CO3, CHCl3, rt; (b) pTSA,
PhH, reflux, 47% (two steps).
(1) Lavalle´e, J.-F.; Deslongchamps, P. Tetrahedron Lett. 1988, 29, 6033.
(2) Lociuro, S.; Tsai, T. Y. R.; Wiesner, K. Tetrahedron 1988, 44, 35
and references therein.
(3) (a) Thomas, R.; Gray, P.; Andrew, J. AdV. Drug Res. 1990, 19, 312.
(b) Scheiner-Bobis, G.; Schoner, W. Nat. Med. 2001, 7, 1288. (c) Kawamura,
A.; Abrell, L. M.; Maggiali, F.; Berova, N.; Nakanishi, K.; Labutti, J.; Magil,
S.; Haupert, G. T., Jr.; Hamlyn, J. M. Biochemistry 2002, 40, 5835.
from the leaves of Digitalis, has cis A/B and C/D ring
fusions, a trans B/C juncture, and a butenolide at C-17.
Ouabain was therefore seen as an ideal candidate for total
synthesis.4
10.1021/ol027125o CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/06/2002

Scheme 2a
Figure 1. Target compounds ouabain and ouabagenin.
Early studies in our group5 directed toward anionic
polycyclization using Nazarov reagents in which the C-17
carbonyl was protected against aldol condensation were not
as successful as the studies using Nazarov reagent 2 (Scheme
1). The desired tetracyclic compound was not formed in one
pot, but the reaction instead stopped after double Michael
addition, yielding the tricycle. This was due to a lower
reactivity in the intramolecular aldol condensation between
C-8 and C-14, allowing enolization of the â-ketoester to
impede the desired reaction pathway.6 Selective decarboxy-
lation of the tert-butylester was therefore necessary before
aldol condensation between C-8 and C-14 would occur. This
variation furnished the desired tetracycle; however, the low
yields were disappointing.
It was believed that poor regioselectivity in enol formation
was affecting the yields for the intramolecular aldol con-
densation; thus the use of selective enolization by reduction
of an R-bromoketone at C-8 should solve the problem.7 Upon
treatment of tricycle 9 (Scheme 2) with SmI2 at -20 °C,
tetracycle 10 (8R-H) was obtained as the major product in a
much improved yield of 63%. The desired isomer 11 (8â-
H) was formed with a very low yield of 7% but was obtained
in 90% yield via acidic isomerization of 10. The use of lower
temperature for the reduction gave no aldol product, and
temperatures higher than -20 °C led to greater amounts of
the retro-Michael product 12.
a Reagents and conditions: (a) Cs2CO3, CH2Cl2, rt; (b) pTSA,
PhH, reflux, 56% (two steps); (c) SmI2, THF, -20 °C, 63% (10),
7% (11), 25% (12); (d) anhydrous HCl, CH2Cl2, 90%.
Another issue dealt with the choice of functionality at C-17
to allow for efficient access to the butenolide at a later point
in the sequence. A report by Stork et al.8 demonstrated that
the butenolide synthesis could be achieved via a nitrile. We
therefore chose a CH2OTBDPS group as a stable precursor
to the nitrile at position 17. After verifying the overall
strategy and getting a better idea of the synthetic steps
necessary it may be deemed possible to go back and leave
the nitrile unprotected throughout the route (or even to
eventually use a â-furyl group). With these considerations
in mind, compound 13 (Figure 2) became the targeted
Nazarov reagent for exposure to the anionic polycyclization
protocol.
The use of the R-bromoketone was a very effective
solution to improve the yield for the aldol condensation;
however, the need to isomerize 10 to 11 under acidic
conditions was a point of concern since this may limit the
range of functionalities available for the molecule. Also, the
appearance of retro-Michael degradation using this reduction
protocol was worrisome. With our sights set on the synthesis
of (-)-ouabain it was felt that construction of a Nazarov
reagent containing an acetate at C-11 in the R-orientation
was in order.
Figure 2. Targeted Nazarov reagent.
The synthesis began with selective reduction of the known
Hajos-Parrish ketone9 (14, Scheme 3) using NaBH4. This
was followed by tosylation and then cyanation as reported
by Caine et al.10a and Overman et al.10b to give cyanoenone
(4) One synthetic approach toward ouabagenin: Overman, L. E.; Rucker,
P. V. Tetrahedron Lett. 1998, 39, 4643. Overman, L. E., Nasser, T.; Rucker,
P. V. Tetrahedron Lett. 1998, 39, 4647. Overman, L. E.; Rucker, P. V.
Heterocycles 2000, 52, 1297.
(5) (a) Ruel, R.; Deslongchamps, P. Tetrahedron Lett. 1990, 31, 3961.
(b) Ruel, R.; D, P. Can. J. Chem. 1992, 70, 1939.
(6) This can occur by protonation of the newly formed enolate (cf. 3)
followed by in situ protonation and deprotonation of the more acidic
â-ketoester moiety.
(7) Chapdelaine, D.; Belzile, J.; Deslongchamps, P. J. Org. Chem. 2002,
67, 5669.
(8) Stork, G.; West, F.; Lee, H.-Y.; Isaacs, R. C. A.; Manabe S. J. Am.
Chem. Soc. 1996, 118, 10660.
(9) Hajos, Z. G.; Parrish, D. R. Org. Synth. 1985, 63, 26
(10) (a) Caine, D.; Kotian, P. L. J. Org. Chem. 1992, 57, 6587. (b) Deng,
W.; Jensen, M. S.; Overman, L. E.; Rucker, P. V.; Vionnet, J.-P. J. Org.
Chem. 1996, 61, 6760.
4694
Org. Lett., Vol. 4, No. 26, 2002

Scheme 3a
Scheme 4a
a Reagents and conditions: (a) NaIO4, RuCl3, H2O, CCl4,
CH3CN, rt, 80%; (b) EtSH, DCC, DMAP, CH2Cl2, 0 °C to rt, 81%;
(c) Et3SiH, Pd(OAc)2, acetone, sieves, rt; (d) Ph3PCHC(OEt)CHCO2-
All, PhH, reflux, 85% (two steps); (e) 80% AcOH (aqueous), 80
°C, 84%.
a Reagents and conditions: (a) NaBH4, EtOH, -15 °C, 94%;
(b) TsCl, Pyr, 0 °C to rt, 93%; (c) NaCN, DMSO, rt, 85%; (d)
(CH2OH)2, pTSA, PhH, reflux, 95%; (e) DIBAL-H, PhCH3, 0 °C,
91%; (f) NaBH4, EtOH, -20 °C, 96%; (g) 1 M HCl, acetone, rt,
95%; (h) TBDPSCl, imidazole, DMF, rt, 99%; (i) Pb(OAc)4,
PhCH3, 138 °C, 88%, 18:19 ) 8.3:1.
With the stage set for anionic polycylclization, treatment
of Nazarov reagent 13 with cyclohexenone 7, under basic
conditions, afforded tricycle 21 (Scheme 5) in 61% yield.
Selective decarboxylation of the allyl ester in 71% yield,
followed by exposure to base at elevated temperature, gave
the desired tetracyclic aldol product, 22, in a very satisfying
yield of 71%. Very importantly, this showed that remote
functionalities (i.e., the C-11 R-acetate) could be installed
to promote efficient aldol condensation. Furthermore, the
correct stereochemistry at C-8 was directly obtained, and no
competition from the retro-Michael degradation was observed
as was depicted in Scheme 2. The silyl groups were then
removed via TBAF generating diol 23, which was suitable
for X-ray analysis.16
15 in 74% yield over three steps. Selective reduction of the
nitrile using DIBAL-H to give the desired aldehyde was not
successful in the presence of the ketone functionality.
Reduction of the nitrile was therefore preceded by ketaliza-
tion giving aldehyde 16 in 86% yield over two steps.
Reduction of the aldehyde, followed by hydrolysis of the
ketal and then silylation of the alcohol afforded the desired
protecting group at the pro-17 position in 90% yield over
three steps. Acetoxylation using lead tetraacetate11 at high
temperature gave the thermodynamically favored equatorial
product 18 in 78% yield. Lower temperatures and/or shorter
reaction times decreased the isomeric ratio between 18 and
19 to as low as 1.4:1, respectively.
The enone of compound 18 was then oxidatively cleaved,12
and the resulting carboxylic acid was converted to thioester13
20 (Scheme 4) in 65% yield over two steps. Reduction to
the aldehyde,14 followed by Wittig reaction15 and selective
hydrolysis gave the targeted Nazarov reagent 13 in 71% yield
over the last three steps.
In regards to the stereochemical outcome of the aldol
condensation, it could proceed via four stereoelectronically
Scheme 5a
Although the earlier studies5 mentioned above showed that
aldol condensation between C-8 and C-14 was low yielding,
it was felt that this new Nazarov reagent possessing the
R-acetate at C-11 could play a role in facilitating the desired
aldol condensation. Also, even though cyclohexenone 7
(Scheme 5) would probably not be the final ring A precursor
in the synthesis of ouabain, it was chosen for this initial trial
since it was previously found to be a good substrate for
double Michael addition (in terms of yield and facial
selectivity).7
(11) Thompson, A.; Ourisson, G. Tetrahedron 1991, 47, 7045.
(12) Webster, F. X.; Enterrios, J. R.; Silverstein, R. M. J. Org. Chem.
1987, 52, 686.
(13) Keck, G. E.; Boden, E. P.; Mabury, S. A. J. Org. Chem. 1985, 50,
709.
(14) Jutzi-Eme, A.-M.; Nuninger, F.; Eberle, M. K. J. Org. Chem. 1994,
59, 7249.
a Reagents and conditions: (a) Cs2CO3, CH2Cl2, rt, 61%; (b)
Pd(PPh3)4, morpholine, THF, rt, 71%; (c) KHMDS, THF, reflux,
71%; (d) TBAF, THF, rt, 85%.
(15) Chapdelaine, D.; Dube, P.; Deslongchamps, P. Synlett 2000, 1819.
Org. Lett., Vol. 4, No. 26, 2002
4695

8R-H configuration as well as the rigorous basic conditions
used in Scheme 5, it appears plausible that compound 22
was formed via pathway C to give first 26, which was then
isomerized in situ to the more stable isomer 22. It is also
possible that the pathway D produced directly the thermo-
dynamic product 22 as it appears to occur with less steric
hindrance according to molecular models.
With the synthesis of steroid 22 complete, and the
stereoselective generation of five contiguous chiral centers
from anionic polycyclization, the crucial groundwork has
been established for the asymmetric total synthesis of
ouabain. A literature procedure will be used for the buteno-
lide synthesis at C-17, and other ring A cyclohexenones are
currently under investigation to allow for easier access to
ring A in the targeted natural product. Developments in these
regards will be reported in due time. Although ouabain is
the specific target in this line of research, other derivatives
may show interesting biological activity.3 This will be
assessed in the course of our efforts.
Acknowledgment. This research was supported by the
Natural Sciences and Engineering Research Council of
Canada (NSERCC). Also, a research Chair in organic
chemistry to P. Deslongchamps from BioChem Pharma Inc.
is deeply appreciated.
Supporting Information Available: Experimental details
and characterization data for all new compounds, as well as
complete X-ray crystallographic data for compound 23. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL027125O
Figure 3. Comparison of the possible aldol pathways.
(16) The authors thank Dr. A. Decken for X-ray diffraction analysis
(deposited at the Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK. CCDC no. 195519).
(17) For a general reference refer to: Deslongchamps, P. Stereoelectronic
Effects in Organic Chemistry; Pergamon Press: Oxford, England, 1983; p
274.
allowed transition states17 leading to four possible diaster-
eomers (Figure 3). Taking into account the results described
in Scheme 2, which shows that the kinetic product has an
4696
Org. Lett., Vol. 4, No. 26, 2002