New strategy for convergent steroid synthesis

Article abstract of DOI:10.1021/jo026676p

We published recently our results on a new and convergent synthesis of natural steroids. The strategy was based on a cycloaddition reaction of Nazarov reagents 2 and 5 with cyclohexenones 1 and 4. In this paper we report results that deal with the synthes

Full text of DOI:10.1021/jo026676p

New Str a tegy for Con ver gen t Ster oid Syn th esis
Olivier Lepage and Pierre Deslongchamps*
Laboratoire de synthe`se organique, Institut de pharmacologie de Sherbrooke,
Universite´ de Sherbrooke, Sherbrooke, Que´bec J 1H 5N4, Canada
pierre.deslongchamps@usherbrooke.ca.
Received November 8, 2002
We published recently our results on a new and convergent synthesis of natural steroids. The
strategy was based on a cycloaddition reaction of Nazarov reagents 2 and 5 with cyclohexenones
1 and 4. In this paper we report results that deal with the synthesis of two new bicyclic Nazarov
reagents (13 and 19) and their cycloaddition with two cyclohexenones (1 and 4). These new results
constitute an important improvement concerning the versatility of the strategy since tetracycles
having the stereochemistry found in natural steroids are now available.
SCHEME 1 a
In tr od u ction
Steroids constitute an important class of natural
products, which has attracted the attention of organic
chemists worldwide. These compounds have numerous
therapeutic effects,1 and new strategies for their synthe-
sis are always welcomed. A lot of strategies have been
developed for the synthesis of steroids; only a few start
with a CD bicyclic building block, but are not convergent.2
Recently, we reported our results regarding a new and
efficient strategy for steroid synthesis,3 involving the
cycloaddition of bicyclic Nazarov reagents 2 and 5 and
the cyclohexenones 1 and 4, respectively, followed by a
regioselective decarboxylation (Scheme 1). This new
strategy gave tetracyclic compounds 3 and 6 with com-
plete diastereoselectivity at four new chiral centers. The
stereochemistries obtained at C-5 and C-10 (steroid
numbering) were opposite in the two series. It was
concluded that the configuration of the angular methyl
group in 2 directed the cycloaddition approach in the first
case, while the stereochemistry of the second case was
controlled by the chirality of the OTBDPS group of
cyclohexenone 4. Nazarov reagent 2 was then condensed
with 4 to give exclusively 7. Since none of the other
isomer (8) was observed, we showed that the configura-
tion of the angular methyl group of 2 prevailed over that
of cyclohexenone 4 (Figure 1; approach B is favored over
approach A). Since the stereochemistry normally found
in natural steroids was not observed (cf. 8), we have
continued our study with new Nazarov reagents to
improve the versatility of our strategy. We report the
synthesis of two new bicyclic Nazarov reagents (13 and
19) and the results of their cycloaddition with the two
cyclohexenones 1 and 4.4,5
a
Reagents and conditions: (a) Cs2CO3, CH2Cl2; (b) Pd(PPh3)4,
morpholine, THF, 87% (two steps); (c) Pd(PPh3)4, morpholine, THF,
60% (two steps); (d) Pd(PPh3)4, morpholine, THF, 50% (two steps).
Resu lts a n d Discu ssion
(1) Zeelen, F. J . Medicinal Chemistry of steroids; Elsevier: Amster-
dam, 1990.
(2) (a) Akhrem, A. A.; Titov, Y. A. Total Steroid Synthesis; Plenum
Press: New York, 1970. (b) Blickenstaff, R. T.; Ghoosh, A. C.; Wolf, G.
C. Total Steroid Synthesis; Academic Press: New York, 1974. (c) J iang,
X.; Covey D. F. J . Org. Chem. 2002, 67, 4893.
The Hajos-Parrish ketone 96 was selected as the
starting material for the synthesis of the Nazarov reagent
(3) Lepage, O.; Stone, C.; Deslongchamps, P. Org. Lett. 2002, 4, 1091.
(4) Lavalle´e, J .-F.; Spino, C.; Ruel, R.; Hogan, K.; Deslongchamps,
P. Can. J . Chem. 1992, 70, 1406.
(5) Chapdelaine, D.; Belzile, J .; Deslongchamps P. J . Org. Chem.
2002, 67, 5669.
(6) Hajos, Z. G.; Parrish, D. R. Org. Synth. 1985, 63, 26.
10.1021/jo026676p CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/28/2003
J . Org. Chem. 2003, 68, 2183-2186
2183

Lepage and Deslongchamps
SCHEME 3 a
F IGURE 1. Comparison of the cycloaddition approaches for
the reaction of 2 with 4. When R ) OTBDPS, approach B is
favored over approach A.
SCHEME 2 a
a
Reagents and conditions: (a) NaBH4, TFA, CH3CN, 73%; (b)
MOMCl, (i-Pr)2EtN, CH2Cl2, 82%; (c) DMDO/acetone, CH2Cl2,
71%; (d) KOH, H2O, DMSO, 120 °C, 81%; (e) TMSCl, Et3N, THF;
(f) MOMCl, (i-Pr)2EtN, DMAP, CH2Cl2, 100% (two steps); (g)
TBAF, THF, 86%; (h) TPAP, NMO, 4 Å molecular sieves, CH2Cl2,
100%; (i) (i) LDA, THF, -78 °C, (ii) Comins’ reagent, -30 °C, 92%;
(j) CH2dC(SnBu3)OEt, PdCl2(PPh3)2, LiCl, DMF, 60 °C; (k) oxalic
acid, DMF, 50% (two steps); (l) (i) LiHMDS, THF, -78 °C, (ii)
NCCO2-allyl, -30 °C, 87%.
a
Reagents and conditions: (a) (i) LDA, THF, -78 °C, (ii)
Comins’ reagent, -30 °C, 73%; (b) CO (1 atm), PdCl2(PPh3)2,
K2CO3, MeOH, DMF, 60%; (c) DIBAL, CH2Cl2, -78 °C, 86%; (d)
Dess-Martin periodinane, CH2Cl2, 86%; (e) (i) LDA, allyl acetate,
THF, -78 °C, (ii) 12, -78 °C, 89%; (f) Dess-Martin periodinane,
CH2Cl2, 100%.
ketone was trapped with Comins’ reagent8,9 to afford the
enol triflate 17. A Stille coupling reaction was per-
formed,15,16 and the resulting enol ether was hydrolyzed
to the methyl ketone 18. Finally the lithium enolate of
this methyl ketone reacted with allyl cyanoformate17 to
afford the Nazarov reagent 19.
having an unsaturation in the five-membered ring (13)
(Scheme 2). The transformation of the diketone 9 to the
enone 10 was already reported in the literature.7 The R,â-
unsaturated ester 11 was obtained by trapping the
kinetically favored enolate with the Comins reagent (N-
(5-chloro-2-pyridyl)triflimide)8,9 to form the enol triflate,
followed by a carbonylation reaction.10 The R,â-unsatu-
rated ester 11 obtained was then reduced to the alcohol
and oxidized to the R,â-unsaturated aldehyde 12.11 Allyl
acetate lithium enolate was added to 12, and the result-
ing alcohol was oxidized with the Dess-Martin periodi-
nane,11 affording Nazarov reagent 13.
To compare the cycloaddition, we also synthesized the
Nazarov reagent 19 with a protected hydroxyl group at
the ring junction (Scheme 3). The synthesis of 19 involved
reduction12 of ketone 96 followed by protection, affording
the MOM-protected alcohol 14.3 The â-epoxide was
obtained using dimethyldioxirane in acetone,13 which was
then opened under basic conditions to give the diol 15.13
A sequence of protection and deprotection furnished
compound 16 in which the more hindered tertiary hy-
droxyl group was protected. The secondary alcohol was
then oxidized to the ketone,14 and the enolate of the
Having both Nazarov reagents 13 and 19, we studied
their cycloaddition reaction with cyclohexenones 1 and
4.4,5 The reaction of Nazarov reagent 13 and cyclohex-
enone 1 (Scheme 4) afforded, after selective decarboxyl-
ation,18 only the cis-anti tetracyclic compound 20. It
should be noted that the unsaturation in the five-
membered ring isomerized to become conjugated with the
ketone. The structure of this tetracyclic compound 20 was
proved by X-ray diffraction analysis19 of compound 22, a
derivative of 20. We believe that the anionic cyclization
process takes place either via a highly asynchronous
Diels-Alder reaction or by two consecutive Michael
additions where the first step would be reversible. The
cycloaddition thus proceeds from the less hindered face
of the Nazarov reagent (the R face) probably due to the
strong steric interaction created by the tertiary methyl
group (Figure 2; approach B is favored over approach A).
We then tried to force the cycloaddition to proceed on the
other face of the Nazarov reagent (the â face) by using
the chiral cyclohexenone 45 (Scheme 4). The protected
hydroxyl group of compound 4 blocks the â face of the
(7) Fernandez, B.; Martinez Pe´rez, J . A.; Granja, J . R.; Castedo, L.;
Mourino, A. J . Org. Chem. 1992, 57, 3173.
(8) Castedo, L.; Mourino, A.; Sarandeses, L. A. Tetrahedron Lett.
1986, 27, 1523.
(14) Griffith, W. P.; Ley, S. V. Aldrichimica Acta 1990, 23, 13.
(15) Soderquist, J . A.; J i-Ho Hsu, G. Organometallics 1982, 1,
830.
(9) Comins, D. L.; Dehghani, A.; Foti, C. J .; J oseph, S. P. Org. Synth.
1997, 74, 77.
(10) Tius, M. A.; Kannangara, G. S. K. J . Org. Chem. 1990, 55, 5711.
(11) Dess, D. B.; Martin, J . C. J . Org. Chem. 1983, 48, 4155.
(12) Di Filippo, M.; Izzo, I.; Vece, A.; De Riccardis, F.; Sodano, G.
Tetrahedron Lett. 2001, 42, 1155.
(16) Kwon, H. B.; McKee, B. H.; Stille, J . K. J . Org. Chem. 1990,
55, 3114.
(17) Donnelly, D. M. X.; Finet, J .-P.; Rattigan, B. A. J . Chem. Soc.,
Perkin Trans. 1 1993, 15, 1729.
(18) Tsuji, J .; Nisar, M.; Shimizu, I. J . Org. Chem. 1985, 50, 3416.
(19) We thank Mr. Andreas Decken for the X-ray diffraction
analysis.
(13) Frigerio, M.; Santagostino, M.; Sputore, S. Synlett 1997, 833.
2184 J . Org. Chem., Vol. 68, No. 6, 2003

New Strategy for Convergent Steroid Synthesis
SCHEME 4 a
SCHEME 5 a
a
Reagents and conditions: (a) Cs2CO3, CH2Cl2; (b) Pd(PPh3)4,
morpholine, THF, 55% (two steps); (c) concentrated HCl, MeOH,
reflux, 85%.
a
Reagents and conditions: (a) Cs2CO3, CH2Cl2; (b) Pd(PPh3)4,
morpholine, THF, 60% (two steps); (c) concentrated HCl, MeOH,
reflux, 85%; (d) Dess-Martin periodinane, CH2Cl2, 50%; (e)
Pd(PPh3)4, morpholine, THF, 17% (two steps); (f) concentrated HCl,
MeOH, reflux, 85%.
F IGURE 3. Comparison of the cycloaddition approaches for
the reaction of 19 with 4. When R ) OTBDPS, approach A is
favored over approach B.
of the cyclohexenone 1 and the low stability of the
Nazarov reagent 19 could explain this result. In another
attempt, when we tried the cycloaddition of 19 with the
more reactive cyclohexenone 4,5 we obtained only the cis-
anti tetracyclic compound 26. The structure of tetracyclic
compound 26 was proven by X-ray diffraction analysis19
of its derivative 30 (Scheme 6). It should be noted that,
instead of obtaining the desired tetracycle with a pro-
tected tertiary hydroxyl group, we ended up with enone
26 due to an elimination reaction. The fact that cyclo-
addition proceeded only on the â face of the Nazarov
reagent 19 can be explained by the steric interaction
created by the OTBDPS group on the cyclohexenone 4
and less steric interaction from the angular methyl group
in 19 by comparison with that of 13 or 2 (Figure 3;
approach A is favored over approach B). In fact, Nazarov
reagent 19 reacts like Nazarov reagent 5 having the same
cis CD ring junction. Selective deprotection on compounds
26 (Scheme 5) and 23 (Scheme 4) afforded exactly the
same tetracyclic compound (24).
F IGURE 2. Comparison of the cycloaddition approaches for
the reaction of 13 with 1 or 4. When R ) H, approach B is
favored over approach A. When R ) OTBDPS, approach A is
slightly favored over approach B.
cyclohexenone and favors an approach of the â face of
the Nazarov reagent for the cycloaddition. When the
cycloaddition was carried out, a mixture of compound 23
and another minor compound (approximately 4:1) was
obtained with low yields due to important decomposition.
We have assumed that the minor compound is 25 on the
basis of the result of the cycloaddition between 1 and 13.
This difficult and noncompletely stereoselective cycliza-
tion can be explained by the strong steric interactions
created in both approaches. The major compound (23)
results from the cycloaddition on the â face of the
Nazarov reagent to avoid the steric interaction created
by the protected alcohol in 4 (Figure 2; approach A is
slightly favored over approach B). The structure of the
tetracyclic compound 23 was proved by X-ray diffraction
analysis19 of its derivative compound 30 (Scheme 6).
The cycloaddition of the Nazarov reagent 19 was
then studied (Scheme 5). When the cycloaddition of the
Nazarov reagent 19 with the cyclohexenone 1 was carried
out, only decomposition was observed. The low reactivity
Other reactions were also performed on compound 24
to obtain a crystalline compound for X-ray diffraction
analysis (Scheme 6). We first tried to reduce the double
bond under the usual conditions. Instead of hydrogenat-
ing the double bond under the reaction conditions used,
the ketone function of the enone moiety was hydrogeno-
lyzed to a methylene group. Protection of the free alcohol
followed by selective desilylation of the TBDPS group
followed by oxidation furnished crystalline compound 30
whose structure was established by X-ray diffraction
analysis.19
J . Org. Chem, Vol. 68, No. 6, 2003 2185

Lepage and Deslongchamps
SCHEME 6 a
at C-10 and the methyl at C-13 corresponds to the
chirality found in various natural steroids. The results
presented here broaden the versatility of this synthetic
approach and demonstrates that this methodology could
be used for the synthesis of various natural products.
Tetracyclic compound 6 is suitable for the synthesis of
steroids with a CD cis ring junction, and tetracyclic
compound 26 could be used for steroids having a trans
or a functionalized CD ring junction.20 For example,
withaferin A (31)21 (Scheme 6) is a natural steroid that
could be achieved using our methodology. The total
syntheses of natural products involving this novel pro-
tocol are presently in progress in our laboratory and will
be reported in due course.
Ack n ow led gm en t. A research chair in organic
chemistry to P.D. from BioChem Pharma Inc. is deeply
appreciated. Fellowships FCAR-Quebec and NSERC-
Canada to O.L. are highly appreciated.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and characterization data for all new compounds, H
a
Reagents and conditions: (a) H2 (1 atm), Pd/C*, EtOH, 76%;
1
(b) 4-nitrobenzoyl chloride, pyridine, DMAP, CH2Cl2, 94%; (c)
HF‚pyridine, THF, 90 °C, 65%; (d) Dess-Martin periodinane,
CH2Cl2, 81%.
NMR spectra for all compounds, and X-ray crystal structure
data for compounds 22 and 30. This material is available free
of charge via the Internet at http://pubs.acs.org. The following
crystal structures have been deposited at the Cambridge
Crystallographic Data Centre: 22 (CCDC 196332) and 30
(CCDC 196331).
Con clu sion
In conclusion, we have executed successfully the cyclo-
addition reactions using the Nazarov reagents 13 and 19.
Utilizing this novel protocol, the synthesis of various
steroidal backbones has been achieved. The cycloaddition
of Nazarov reagent 19 and cyclohexenone 4 allowed us
to obtain with complete stereocontrol the important
tetracyclic compound 26. The â configuration of the ester
J O026676P
(20) (a) Ruel, R.; Deslongchamps, P. Tetrahedron Lett. 1990, 31,
3961. (b) J ung, M. E.; J ohnson, T. W. Org. Lett. 1999, 1, 1671. (c)
Patterson, D. G.; Haley, M. J .; Midgley, I.; Djerassi, C. Org. Mass
Spectrom. 1984, 19, 531.
(21) Glotter, E. Nat. Prod. Rep. 1991, 415.
2186 J . Org. Chem., Vol. 68, No. 6, 2003