Total Synthesis of (-)-Steganone Utilizing a Samarium(II) Iodide Promoted 8-Endo Ketyl-Olefin Cyclization

Article abstract of DOI:10.1021/ja9930059

A six-step synthesis of (±)-steganone from commercially available 3,4,5-trimethoxybenzyl alcohol features a samarium(II) iodide promoted 8-endo ketyl-olefin coupling to install, in a single transformation, the 8,5 ring system common to the lignan lactones. The racemic synthesis provided the basis for the construction of (-)-steganone, which exploited a chromium tricarbonyl moiety both to establish and protect the desired absolute stereochemistry through key transformations, including a SmI₂-promoted 8-endo radical cyclization and two palladium-catalyzed couplings.

Full text of DOI:10.1021/ja9930059

52
J. Am. Chem. Soc. 2000, 122, 52-57
Total Synthesis of (-)-Steganone Utilizing a Samarium(II) Iodide
Promoted 8-Endo Ketyl-Olefin Cyclization
Lauren G. Monovich,† Yvan Le Hue´rou,†,‡ Magnus Ro1nn,† and Gary A. Molander*,†,‡
Contribution from the Department of Chemistry and Biochemistry, UniVersity of Colorado, Boulder,
Colorado 80309-0215, and Roy and Diana Vagelos Laboratories, Department of Chemistry, UniVersity of
PennsylVania, Philadelphia, PennsylVania 19104-6323
ReceiVed August 18, 1999
Abstract: A six-step synthesis of (()-steganone from commercially available 3,4,5-trimethoxybenzyl alcohol
features a samarium(II) iodide promoted 8-endo ketyl-olefin coupling to install, in a single transformation,
the 8,5 ring system common to the lignan lactones. The racemic synthesis provided the basis for the construction
of (-)-steganone, which exploited a chromium tricarbonyl moiety both to establish and protect the desired
absolute stereochemistry through key transformations, including a SmI2-promoted 8-endo radical cyclization
and two palladium-catalyzed couplings.
Introduction
The 1973 article describing the isolation of steganone and
steganacin (Figure 1) marked them as biologically active targets1
and initiated synthetic studies culminating in the first published
total syntheses in 1976.2,3 Steganone continued to be a popular
target through the next two decades because some natural and
synthetic congeners of steganone, steganacin, and stegane inhibit
tubulin polymerization both in vitro and in vivo.4 Structure-
activity relationships of synthetic congeners and new isolates
continue to be investigated on steganone analogues.5 Many
synthetic studies have been conducted on these compounds and
to date, nine total syntheses of steganone have been achieved.
Two syntheses,6,7 based on significant advances in stereoselec-
tive biaryl couplings, provided a fresh perspective on the
construction of (-)-steganone. Both relied on a previously
developed sequence to generate the 8-5 ring system.8 To date,
no synthesis has featured a radical ring closure to form the eight-
membered ring and in all cases formation of the five- and eight-
Figure 1. Cyclooctadiene lignan lactones steganone and steganacin.
membered rings is separated by several synthetic transforma-
tions. Consequently, the 8-5 ring system of steganone presented
itself as an appropriate challenge with which to explore the
SmI2-promoted 8-endo ketyl-olefin radical cyclization in total
synthesis.
Previous racemic syntheses of steganone reveal that the eight-
membered ring can be generated by ring-expansion or ring-
closing methods. Ring expansion from a 6,4 ring system
produced by [2+2] cycloaddition to a phenanthroline unit2,9 or
an imaginative ring expansion from a 7,3 ring system10 both
yield substituted eight-membered rings requiring further elabo-
ration to install the lactone ring of steganone. Ring-closing
methods are represented by an alkylation, which generates the
eight-membered ring after intermolecular biaryl coupling,8,11 and
by an intramolecular biaryl coupling,3,12 which forms the eight-
* To whom correspondence should be addressed at the University of
Pennsylvania.
† University of Colorado.
‡ University of Pennsylvania.
(1) Kupchan, S. M.; Britton, R. W.; Zeigler, M. F.; Gilmore, C. J.;
Restivo, R. J.; Bryan, R. F. J. Am. Chem. Soc. 1973, 95, 1335.
(2) Hughes, L. R.; Raphael, R. A. Tetrahedron Lett. 1976, 1543.
(3) Kende, A. S.; Liebeskind. L. S. J. Am. Chem. Soc. 1976, 98, 267.
(4) (a) Tomioka, K.; Ishiguro, T.; Mizuguchi, H.; Komeshima, N.; Koga,
K.; Tsukagoshi, S.; Tsuruo, T.; Tashiro, T.; Tanida, S.; Kishi, T. J. Med.
Chem. 1991, 34, 54. (b) Wang, R. W.-J.; Rebhun, L. I.; Kupchan, S. M.
Cancer Res. 1977 37, 3071. (c) Zavala, F.; Guenard, D.; Robin, J.-P.; Brown,
E. J. Med. Chem. 1980, 23, 546. (d) Wickramaratne, D. B. M.; Pengsuparp,
T.; Mar, W.; Chai, H. B.; Chagwedera, T. E.; Beecher, C. W. W.;
Farnsworth, N. R.; Kinghorn, A. D.; Pezzuto, J. M.; Cordell, G. A. J. Nat.
Prod. 1993, 56, 2083. (e) Dhal, R.; Brown, E.; Robin, J.-P. Tetrahedron
1983, 39, 2787. (f) Ayres, D. C.; Loike, J. D. Chemistry & Pharmocology
of Natural Products; Lignans: Chemical, Biological, and Clinical Proper-
ties; Cambridge University Press: Cambridge; New York, 1990.
(5) Bradley, A.; Motherwell, W. B.; Ujjainwalla, F. J. Chem. Soc., Chem.
Commun. 1999, 10, 917.
(9) (a) Becker, D.; Hughes, L. R.; Raphael, R. A. J. Chem. Soc., Perkin
Trans. 1 1977, 1674. (b) Krow, G. R.; Damodaran, K. M.; Michener, E.;
Wolf, R.; Guare, J. J. Org. Chem. 1978, 43, 3950. (c) Larson, E. R.; Raphael,
R. A. Tetrahedron Lett. 1979, 5041. (d) Narasimhan, N. S.; Aidhen, I. S.
Tetrahedron Lett. 1988, 29, 2987. (e) Aidhen, I. S.; Narasimhan, N. S. Indian
J. Chem., Sect. B 1993, 32B, 211.
(10) (a) Magnus, P.; Schultz, J.; Gallagher, T. J. Chem. Soc., Chem.
Commun. 1984, 1179. (b) Magnus, P.; Schultz, J.; Gallagher, T. J. Am.
Chem. Soc. 1985, 107, 4984.
(6) Meyers, A. I.; Flisak, J. R.; Aitken, R. A. J. Am. Chem. Soc. 1987,
109, 5446.
(7) Uemura, M.; Daimon, A.; Hayashi, Y. J. Chem. Soc., Chem. Commun.
1995, 19, 1943.
(8) Ziegler, F. E.; Fowler, K. W.; Sinha, N. D. Tetrahedron Lett. 1978,
2767.
(11) (a) Brown, E.; Dhal, R.; Robin, J.-P. Tetrahedron Lett. 1979, 733.
(b) Mervic, M.; Ben-David, Y.; Ghera, E. Tetrahedron Lett. 1981, 22, 5091.
(c) Ziegler, F. E.; Chliwner, I.; Fowler, K. W.; Kanfer, S. J.; Kuo, S. J.;
Sinha, N. D. J. Am. Chem. Soc. 1980, 102, 790. (d) Dhal, R.; Brown, E.;
Robin, J.-P. Tetrahedron 1983, 39, 2787.
(12) Ziegler, F. E.; Schwartz, J. A. J. Org. Chem. 1978, 43, 985.
10.1021/ja9930059 CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/22/1999

Total Synthesis of (-)-Steganone
J. Am. Chem. Soc., Vol. 122, No. 1, 2000 53
membered ring and biaryl linkage simultaneously, but without
stereochemical control.
Both ring-expansion and ring-closing strategies13 have been
successfully applied to enantioenriched syntheses of steganone
and/or steganacin, although the phenanthroline-based ring
expansion requires a late-stage resolution.14 A remote stereo-
center can be utilized in the intramolecular and intermolecular
biaryl couplings to produce the desired biaryl geometry, although
in no case is the initial biaryl formation stereoselective.15 The
absolute stereochemistry of the remote stereocenter is ultimately
relayed to the biaryl geometry by thermal equilibration after
biaryl formation. Alternatively, a stereocontrolled, intermolecular
biaryl coupling can be performed early in the synthesis, followed
by eight-membered ring formation and concluding with the five-
membered ring formation.6,7 We sought to combine the advan-
tages of a stereocontrolled biaryl coupling with the efficiency
afforded by simultaneous construction of the eight- and five-
membered rings of steganone utilizing a SmI2-promoted ketyl-
olefin radical cyclization protocol.
Figure 2. Biaryl intermediates utilized in previous syntheses of (-)-
steganone.
SOMO produced by their reduction.19 In fact, there exists a
correlation between the substitution pattern of the aromatic rings
of aromatic carbaldehydes and their success in reductive
couplings.20 Electron-donating groups, such as the methylene-
dioxy substituent of 3, favor coupling relative to simple
reduction by raising the SOMO of the aldehyde. Despite the
effect of these electron-donating groups on the reduction
potential of the aldehyde, the notable selectivity of SmI2 was
expected to allow exclusive reduction of the aromatic aldehyde
to its ketyl in the presence of the butenolide and of the product
lactone.21 Although SmI2 rapidly reduces R,â-unsaturated esters
to their saturated derivatives,22 forcing conditions (strongly
acidic or basic media) are required to reduce simple esters.23
The conformational bias imposed by the biaryl unit was
anticipated to aid cyclization. The biaryl motif enforces folding
of at least four of the eight carbons destined to be the eight-
membered ring and partially counteracts the negative entropy
of activation leading to slow formation of eight-membered rings,
an effect evident in the success of the alkylation reaction featured
in one steganone synthesis.8,11 Alkylations forming eight-
membered carbocycles are generally unsuccessful or low-
yielding because intermolecular reactions are often faster. At
the outset it was unknown as to whether the propensity of these
biaryl systems to cyclize would be required for the SmI2-
promoted coupling between an aromatic aldehyde and R,â-
unsaturated carbonyl partners, because other carbonyl/alkene
pairings have shown no dependency on this feature.24
Results and Discussion
Excellent yields in SmI2-promoted 8-endo ketyl-olefin
cyclization can be attained in substrates in which the olefin
component is substituted with an electron-withdrawing group.16,17
A high degree of efficiency can be achieved when the electron-
withdrawing group provides a portion of the carbon framework
required in the target structure. Such is the case with butenolide
3 (eq 1), which was envisioned to undergo reductive coupling
The novel late-stage features of the current synthesis made
preliminary investigation of a racemic version prudent. Fur-
thermore, the stereochemical outcome of the 8-endo cyclization
was difficult to predict with regard to the stereocenters at the
8-5 ring juncture relative to the biaryl axis. All three stereo-
chemical elements are potentially alterable after cyclization, but
it was necessary to know which, if any, would require
adjustment before proceeding with an enantioenriched synthesis.
This information was used to determine the absolute stereo-
chemistry of the biaryl juncture needed to produce the natural
series.
with a samarium ketyl radical anion generated from an aromatic
carbaldehyde moiety. To our knowledge, no coupling of this
type has been described previously to generate an eight-
membered ring.18
The skeleton common to steganone, steganacin, and stegane
is well suited to the 8-endo cyclization for reasons addressed
in detail throughout this contribution. First, the electronic
characteristics of the aldehyde deserve comment. Often alde-
hydes, especially aromatic aldehydes, perform less well than
ketones in reductive couplings because of the lower-lying
The racemic biaryl hydroxy aldehyde 4, which bears close
resemblance to enantiopure biaryls 5 and 6 synthesized previ-
ously by Uemura7 and Meyers (Figure 2),6 respectively, was
envisioned as an early intermediate (eq 2). To access 4, we
initially investigated an Ullmann coupling between 6-bromopip-
(13) For reviews of eight-membered-ring syntheses see: (a) Petasis, N.
A.; Patane, M. A. Tetrahedron 1992, 48, 5757. (b) Metha, G.; Singh, V.
Chem. ReV. 1999, 99, 881.
(14) Larson, E. R.; Raphael, R. A. J. Chem. Soc., Perkin Trans. 1 1982,
521.
(15) (a) Robin, J.-P.; Gringore, O.; Brown, E. Tetrahedron Lett. 1980,
21, 2709. (b) Tomioka, K.; Ishiguro, T.; Iitaka, Y.; Koga, K. Tetrahedron
1984, 40, 1303.
(19) Kan, T.; Nara, S.; Ito, S.; Matsuda, F.; Shirahama, H. J. Org. Chem.
1994, 59, 5111.
(16) Molander, G. A.; McKie, J. A. J. Org. Chem. 1994, 59, 3186.
(17) Dinesh, C. U.; Reissig, H.-U. Angew. Chem., Int. Ed. Engl. 1999,
38, 789.
(18) (a) Fukuzawa, S.-I.; Iida, M.; Nakanishi, A.; Fujinami, T.; Sakai,
S. J. Chem. Soc., Chem. Commun. 1987, 920. (b) Fukuzawa, S.-I.;
Nakanishi, A.; Fujinami, T.; Sakai, S. J. Chem. Soc., Perkin Trans. 1 1988,
1669. (c) Enholm, E. J.; Trivellas, A. Tetrahedron Lett. 1989, 30, 1063.
(d) Enholm, E. J.; Trivellas, A. J. Am. Chem. Soc. 1989, 111, 6463. (e)
Enholm, E. J.; Satici, H.; Trivellas, A. J. Org. Chem. 1989, 54, 5841.
(20) Molander, G. A.; McWilliams, J. C.; Noll, B. C. J. Am. Chem. Soc.
1997, 119, 1265.
(21) Reduction of a methyl ketone in the presence of an R,â-unsaturated
ester is known. See ref 20.
(22) Inanaga, J.; Sakai, S.; Handa, Y.; Yamaguchi, M.; Yokoyama, Y.
Chem. Lett. 1991, 2117.
(23) (a) Kamochi, Y.; Kudo, T. Tetrahedron Lett. 1991, 32, 3511. (b)
Kamochi, Y.; Kudo, T. Tetrahedron 1992, 48, 4301.
(24) Monovich, L. G., Ph.D. Thesis, University of Colorado, 1998.

54 J. Am. Chem. Soc., Vol. 122, No. 1, 2000
MonoVich et al.
eronal and 2-iodo-3,4,5-trimethoxybenzyl alcohol protected as
a TBDMS ether. Related reactions between 6-bromopiperonal
and a bis(orthosubstituted) aryl iodide produced acceptable
yields of a biaryl that was carried through to steganone by Robin
and co-workers.11a,d,15a Using 2 equiv of the appropriate aryl
iodide led to a disappointing 28% yield of biaryl 4 protected as
the TBDMS ether based on 6-bromopiperonal. Although protec-
tion of the alcohol was required because free hydroxyl groups
poison the Ullmann coupling, the easily removable TBDMS
ether was capable of withstanding the 230 °C reaction temper-
ature. However, the Ullmann coupling works best with a large
excess (2-5 equiv) of the aryl iodide that is generally
unrecoverable because of extensive homocoupling and reduction
to the parent arene.
ladium-catalyzed coupling with 3-tributylstannyl-(5H)-furan-2-
one (eq 4).31 This route, utilizing a stannylated butenolide,
Although quick and operationally simple, the poor yields in
the Ullmann coupling led us to explore a Suzuki coupling.
Arylboronic acids and their esters perform well in palladium-25
and nickel-catalyzed biaryl syntheses and are generally the best
choice for the organometallic partner when sterically congested
biaryls are desired.26 For reasons that will be made clear in the
ensuing discussion, we chose to convert 6-bromopiperonal into
the arylboronic acid derivative. As is generally noted with
arylboronic acids substituted with electron-withdrawing groups,
the standard Suzuki conditions27 employing aqueous Na2CO3
as a base led to extensive deboronation at the expense of
coupling yields. Premature destruction of the carbon-boron
bond can be slowed by utilizing milder bases28 or nonaqueous
conditions.29 With NaHCO3 as a base, Suzuki coupling pro-
ceeded slowly to the desired biaryl 4 in moderate yield (eq 2).
required no protection/deprotection sequence and resulted in no
regiochemical ambiguity of the double bond. Consequently, it
would likely function in the asymmetric version of the synthesis.
In the event, desulfurative stannylation of the corresponding
phenylthiobutenolide produced 3-tributylstannyl-(5H)-furan-2-
one, which was known to undergo Pd-catalyzed couplings with
aryl iodides.31 Carbon-carbon bond formation occurred between
the stannylated furanone and benzyl bromide 7 at 80 °C when
Pd2(dba)3/trifurylphosphine (Pd:L 1:2) was used as a catalyst.32
A fully conjugated isomer, formed by a thermal- or palladium-
catalyzed mechanism, proved to be a minor component in the
reaction mixture.
Treatment of substrate 3 with 2.2 equiv of SmI2 in THF/
HMPA at 0 °C using t-BuOH as a proton source gave two
diastereomeric cyclized products isolated in a 3:1 ratio (Scheme
1). The minor diastereomer, epipicrosteganol 10,33 was identified
by comparison of its spectral data to that reported in the
literature, as five of the eight possible diastereomers are known.
The major diastereomer was previously unknown. Both dia-
stereomers were of the picro series (8-5 cis ring system),
because neither were oxidized to the known trans compounds,
steganone or isosteganone, by Dess-Martin periodinane. X-ray
crystallographic data generated from the major diastereomer also
support this assignment (Figure 3). Following the nomenclature
for these systems, the major diastereomer 9 is called isopicro-
steganol. The synthesis of isopicrosteganol from an epimer was
attempted previously for structure activity determination because
one member of the picro series had been demonstrated to be as
potent as steganacin in in vitro studies.4a
Conversion of benzyl alcohol 4 to the benzyl bromide 7
utilizing MsCl and Et3N in CH2Cl2 followed by direct conver-
sion of the resulting chloride to the bromide by the addition of
excess LiBr in acetone provided 7, which contained a useful
handle to install the butenolide of 3 (eq 3). Many syntheses of
3-substituted furanones30 require multiple transformations in-
cluding condensations, alkylations, and/or retro-aldol transfor-
mations. Although butenolide 3 could be accessed by alkylation
of the R-phenylselenyl-γ-butyrolactone, we focused on a pal-
(25) Benbow, J. W.; Martinez, B. L. Tetrahedron Lett. 1996, 37, 8829.
(26) For a review see: Stanforth, S. P. Tetrahedron 1998, 54, 263. See
also: Johnson, M. G.; Foglesong, R. J. Tetrahedron Lett. 1997, 38, 7001.
(27) (a) Oh-e, T.; Miyaura, N.; Suzuki, A. Synlett 1990, 221. (b) Miyaura,
N.; Yanagi, I.; Suzuki, A. Synth. Commun. 1981, 11, 513.
(28) (a) Wright, S. W.; Hageman, D. L.; McClure, L. D. J. Org. Chem.
1994, 59, 6095. (b) Gronowitz, S.; Ho¨rnfeldt, A.-B.; Yang, Y.-H. Chem.
Scr. 1986, 26, 383.
(29) Watanabe, T.; Miyaura, N.; Suzuki, A. Synlett 1992, 207.
(30) (a) Demnitz, D. W. J. Tetrahedron Lett. 1989, 30, 6109. (b)
Bonadies, F.; Cardilli, A.; Lattanzi, A.; Pesci, S.; Scettri, A. Tetrahedron
Lett. 1995, 36, 2839. (c) Na¨sman, J. H.; Kopola, N.; Pensar, G. Tetrahedron
Lett. 1986, 27, 1391. (d) Canonne, P.; Akssira, M.; Lemay, G. Tetrahedron
Lett. 1983, 24, 1929. (e) Bloch, R.; Gilbert, L. J. Org. Chem. 1987, 52,
4603.
In both intermediates (e.g., 8), the oxygen of the reacting
ketyl in the intermediates lies pseudoequatorially and the
stereochemistry at the adjacent stereocenter is set by attack on
either of the two diastereotopic faces of the butenolide (Scheme
(31) Hollingworth, G. J.; Sweeney, J. B. Tetrahedron Lett. 1992, 33,
7049.
(32) Farina, V.; Krishnan, B. J. Am. Chem. Soc. 1991, 113, 9585.
(33) Tomioka, K.; Ishiguro, T.; Iitaka, Y.; Koga, K. Chem. Pharm. Bull.
1986, 34, 1501.

Total Synthesis of (-)-Steganone
J. Am. Chem. Soc., Vol. 122, No. 1, 2000 55
Scheme 1
Figure 3. X-ray crystal structure of isopicrosteganol.
which the sense of absolute stereochemistry of the stereocenter
R to the lactone carbonyl was retained.15a
1, only one isomer shown). Protonation from the convex side
of the samarium(III) enolates yields the observed products.
Oxidation of 9 or 10 surprisingly provided an inseparable
pair of diastereomeric ketones (eq 5). Neither ketone matched
the spectral data reported for steganone or isosteganone (8-5
trans diastereomers), so we concluded that 11, picrosteganone,
exists as a mixture of atropisomers.
Picrosteganone 11, when heated to reflux in xylenes, did not
yield steganone and led only to recovered starting material. By
contrast, base-promoted conditions (NaOAc in EtOH) provided
steganone from picrosteganone in 90% yield (eq 7). The
The remaining transformation of picrosteganone 11 to ste-
ganone utilized a well-established driving force.34 Steganone,
the target natural product, is the only diastereomer of the four
possible that allows the ketone carbonyl to achieve planarity
with the nearby arene ring as evidenced by its carbonyl
stretching frequency [1667 cm-1 versus that of isosteganone
(1707 cm-1)] and by the chemical shift of the aromatic proton
ortho to the ketone (δ 7.53 ppm for steganone versus δ 6.71
ppm for isosteganone).34 This thermodynamic driving force has
been frequently used in total syntheses of steganone. Isostega-
none was reported to be converted to steganone by either of
the following two conditions: (1) by heating to reflux a solution
of isosteganone in xylenes or (2) by heating to reflux a solution
of isosteganone in MeOH or EtOH in the presence of NaOAc
(eq 6). Only the first method has been utilized to convert
enantioenriched isosteganone into enantioenriched steganone in
discrepancy between the behavior of picrosteganone and iso-
steganone under the equilibration conditions noted above may
be explained by the requirement of picrosteganone to undergo
enolate formation. The mechanism proposed for the facile
isomerization of isosteganone to steganone under both sets of
conditions involves a retro-Michael reaction followed by
Michael reaction of the resultant carboxylic acid.34 It is possible
that the observed isomerization of isosteganone to steganone
in hot xylenes involves only a simple thermal-promoted rotation
about the biaryl axis, which is known to occur in a related
system6 and is facilitated by the presence of the sp2-hybridized
ortho carbon. For picrosteganone, rotation about the biaryl axis
yields only the starting material: the two possible atropisomers
of the picro series. Consequently, the base-promoted conditions
are required and the significantly lower temperature at which
isomerization occurs suggests that the retro-Michael/Michael
manifold may be operating. Formation of a sodium enolate,
followed by â-elimination to give the sodium carboxylate,
(34) Kende, A. S.; Liebeskind, L. S.; Kubiak, C.; Eisenberg, R. J. Am.
Chem. Soc. 1976, 98, 6389.

56 J. Am. Chem. Soc., Vol. 122, No. 1, 2000
MonoVich et al.
Scheme 2
Figure 4. Uemura’s enantiopure chromium tricarbonyl complex 12.
produces a common intermediate enone from picrosteganone
or isosteganone.
With this information in hand, we were prepared for the
enantioselective synthesis of steganone via either of two routes
for generating an enantioenriched biaryl intermediate using
literature precedent.6,7,35 We were especially intrigued by the
opportunities presented by Uemura’s method.7 Although sensi-
tive to oxidation, Cr0 arene complexes were known to perform
well in the reducing environment of SmI2.36 In some, but not
all, cases the complexation of the chromium results in attack
of the arene ring by radical intermediates. In others, the arene
is preserved and the chromium moiety serves only as a tool for
controlling stereochemistry.35 It was in the second capacity that
we envisioned exploiting the chromium tricarbonyl unit.
The difficulties associated with forming the biaryl unit as a
single atropisomer early in the synthesis lie in the arene-
substitution pattern. The thermal isomerization barrier for a
steganone precursor that carries only three ortho substituents is
highly sensitive to the exact identity of those substituents. When
one of the three possible groups (e.g., a formyl moiety) adds
only a small amount to the inversion barrier and a second
substituent is also relatively small (e.g., a methoxy group), the
barrier to inversion renders the biaryl stereochemically labile
even at 0 °C.35 At least one of three labile intermediates
(including 6) in the Meyers synthesis of steganone resulted in
some loss of biaryl stereochemistry.6 Even with careful handling,
the enantiomeric excess of the product did not reflect the initial
success of the biaryl formation.
of biaryl 5 occurred under the reaction conditions and during
purification, so the alcohol was converted directly to the
corresponding benzyl chloride 14 or bromide 15, either of which
could be purified and stored without incident. In contrast to the
uncomplexed case, the synthesis of bromide 15 required the
use of methanesulfonic anhydride to avoid formation of an
intermediate chloride 14, which could not be separated from or
converted to bromide 15 by treatment with excess LiBr (eq 8).
With the chromium-arene complex intact in structure 5,
however, the lability of the biaryl system would pose no danger
owing to its propensity to exist as the desired diastereomer. In
biaryl 5, the formyl group would position itself distal to the
chromium tricarbonyl moiety. At best, the chromium-complexed
biaryl 5 could be converted to a substrate for the SmI2-promoted
reductive coupling and carried through the synthesis of the 8-5
ring system. At worst, the SmI2-based synthesis would involve
only one labile intermediate. The potential for the chromium
to serve as both a tool for forming and for protecting the desired
biaryl atropisomer led us to explore this route.
Both halides 14 and 15 were successfully converted to the
desired chromium-complexed substrate 16 for SmI2-promoted
coupling (eq 9). However, the more reactive benzyl bromide
Construction of chromium-complexed biaryl 5 proceeded by
Uemura’s method,7 generating enantiomerically pure chromium-
carbonyl complex 12 in five steps from 3,4,5-trimethoxy
benzaldehyde (Figure 4).
Reduction of benzaldehyde 12 with NaBH4 proceeded
smoothly to produce alcohol 13, which was successfully coupled
with 2-formyl-4,5-methylenedioxyboronic acid in the presence
of Pd(PPh3)4 and aqueous Na2CO3 (Scheme 2). The observed
diastereomer 5 is likely produced from the other possible
diastereomer, itself the direct product of the Suzuki coupling.
Rotation about the biaryl axis under the reaction conditions
would provide the desired diastereomer. Some decomplexation
15 provided superior yields. Optimized conditions singled out
the weakly coordinating AsPh3 as the ligand of choice for the
coupling. The facility with which oxidative addition occurs to
these chromium-complexed benzyl halides using a relatively
electron poor palladium complex bears contrast to previously
reported descriptions of palladium couplings with benzyl
halides,37 and to the one described in the present text that require
(35) Kamikawa, K.; Watanabe, T.; Uemura, M. J. Org. Chem. 1996,
61, 1375.
(36) (a) Schmaltz, H.-G.; Siegel, S.; Schwarz, A. Tetrahedron Lett. 1996,
37, 2947. (b) Taniguchi, N.; Hata, T.; Uemura, M. Angew. Chem., Int. Ed.
Engl. 1999, 38, 1232.
(37) Milstein, D.; Stille, J. K. J. Am. Chem. Soc. 1979, 101, 4992.

Total Synthesis of (-)-Steganone
J. Am. Chem. Soc., Vol. 122, No. 1, 2000 57
a highly polar, coordinating solvent such as DMF, DMA, or
NMP to facilitate oxidative addition. This high-yielding coupling
utilizing the chromium-complexed benzyl halides represents a
new application of the special reactivity at the benzylic position
of chromium-complexed arenes.38
Scheme 3
Unlike the SmI2-promoted coupling performed on the un-
complexed, racemic substrate 3 that afforded diastereomers 9
and 10, the chromium-complexed version 16 yielded a single
diastereomer in 65-75% yield with the chromium arene moiety
intact (eq 10). The relative stereochemistry between biaryl and
DBU in THF heated at reflux40 afforded (-)-steganone from
enantiopure picrosteganone 11 in 82% yield with an optical
purity >99%.39
ring juncture stereocenters in the chromium-complexed product
17 fortunately corresponded to the major diastereomer 9
observed in the racemic synthesis. The conditions differed
slightly from the racemic case, as a larger excess of SmI2 was
required.
Conclusions
The total synthesis of steganone via an 8-endo radical
cyclization allows direct construction of an 8-5 fused ring
system without undue manipulation of functionality and trans-
lates to a high degree of synthetic efficiency. The racemic
synthesis of steganone entails only six linear steps from
commercially available compounds. Construction of (-)-ste-
ganone exploits the chromium tricarbonyl moiety both to
establish and protect the desired sense of absolute stereochem-
istry through key transformations, including a SmI2-promoted
8-endo radical cyclization and two palladium-catalyzed cou-
plings. A new means of equilibration from picrosteganone to
steganone suitable for the enantiomeric series of compounds
has been introduced. Finally, this synthesis of steganone
proceeds via a new compound, isopicrosteganol, whose synthesis
had been attempted previously for structure-activity determi-
nation. There is significant interest in this latter compound
because of its potential biological activity.
Oxidation of the chromium-complexed isopicrosteganol 17
using PCC buffered with NaOAc in CH2Cl2 accomplished the
deprotection/decomplexation reaction and the oxidation of the
secondary carbinol to the corresponding ketone in one pot
(Scheme 3). The Dess-Martin reagent used in the racemic
synthesis also performed these two transformations, but gener-
ated an impurity that was not easily removed from the desired
ketone 11. As expected, oxidation to the ketone produced a 2:1
mixture of atropisomers, in which the absolute stereochemical
information contained in alcohol 17 was retained only at the
lactone ring juncture stereocenters in ketone 11.
At this point, equilibration of the stereocenter R to the ketone
carbonyl of enantiopure 11 was required to produce the desired
trans stereochemistry at the ring juncture and the driving force
to correct the stereochemistry of the biaryl. Although enantio-
merically pure isosteganone had been previously converted to
steganone by simply heating neat or in xylenes,15a this process
required only rotation about the biaryl bond. Furthermore, the
use of NaOAc in boiling EtOH had been employed only in the
racemic series, and thus the effect of its application on the
various enantiomerically enriched isomers of steganone was
unknown. In fact, use of NaOAc in boiling EtOH to isomerize
enantioenriched 11 led to an 87/13 mixture of (-)-steganone
to (+)-steganone by chiral HPLC.39 This loss of enantiomeric
excess in the final product signaled a partial equilibration of
the stereocenter R to the lactone carbonyl. The absolute
stereochemistry at this center is set relative to the biaryl during
the SmI2-promoted cyclization and is ultimately necessary to
relay all of the stereochemical information back to the remaining
stereocenters in the final equilibration step. Fortunately, utilizing
Acknowledgment. Dedicated with great respect and admira-
tion to Professor Albert I. Meyers. We gratefully acknowledge
the generous financial support of the NIH (GM35249). M.R.
thanks the Stiftelsen Bengt Lundqvists Minne for a fellowship.
We also thank Professor Motokazu Uemura and co-workers for
providing us with detailed experimental procedures for the
preparation of 12, and Dr. Mazakazu Sono for translating these
procedures from Japanese to English. Finally, we acknowledge
Dr. Bruce Noll for determining the crystal structure of compound
9.
Supporting Information Available: Full experimental
1
details, H and 13C NMR spectra for 1 and 14-17 and X-ray
structural data for 9 (PDF). This material is available free of
(38) Merlic, C. A.; Walsh, J. C.; Tantillo, D. J.; Houk, K. N. J. Am.
Chem. Soc. 1999, 121, 3596.
charge via the Internet at http://pubs.acs.org.
(39) This ratio was determined by comparing the HPLC chromatogram
of our compound with the chromatogram of racemic steganone. Condi-
JA9930059
tions: Chiralcel OD, hexanes/EtOH 95/5, 2 mL/min. (-)-Steganone: tR
)
(40) Still, W. C.; Murata, S.; Revial, G.; Yoshihara, K. J. Am. Chem.
Soc. 1983, 105, 625.
74 min. (+)-Steganone: tR ) 47 min.