Total synthesis of (+)-calyculin A and (-)-calyculin B: Asymmetric synthesis of the C(9-25) spiroketal dipropionate subunit

Article abstract of DOI:10.1021/ja992134m

An asymmetric synthesis of the stereochemically fully endowed C(9-25) spiroketal fragment (+)-BC of the calyculins (1-8) is described. Highlights of the synthesis include: a highly diastereoselective IBr-induced iodocarbonate cyclization to introduce the

Full text of DOI:10.1021/ja992134m

10468
J. Am. Chem. Soc. 1999, 121, 10468-10477
Total Synthesis of (+)-Calyculin A and (-)-Calyculin B:
Asymmetric Synthesis of the C(9-25) Spiroketal Dipropionate
Subunit
Amos B. Smith, III,* Gregory K. Friestad, Joseph Barbosa, Emmanuel Bertounesque,
Kenneth G. Hull, Makoto Iwashima, Yuping Qiu, Brian A. Salvatore, P. Grant Spoors, and
James J.-W. Duan
Contribution from the Department of Chemistry, the Laboratory for Research on the Structure of Matter,
and the Monell Chemical Senses Center, UniVersity of PennsylVania, Philadelphia, PennsylVania 19104
ReceiVed June 22, 1999
Abstract: An asymmetric synthesis of the stereochemically fully endowed C(9-25) spiroketal fragment (+)-
BC of the calyculins (1-8) is described. Highlights of the synthesis include: a highly diastereoselective IBr-
induced iodocarbonate cyclization to introduce the C(21) stereocenter in epoxide (+)-18, fragment unions
exploiting the reaction of acyl anion equivalents with epoxides to construct masked advanced aldols (-)-35
and (+)-71 as single diastereomers, chelation-controlled addition of the C(14-15) vinyl group to aldehyde
(+)-38 to set the stereogenicity at C(16), selective reduction of the C(13) ketone via 1,3-induction, and
development of an orthogonal protection scheme permitting both convenient installation of the C(17) phosphate
group and flexibility in subsequent fragment couplings.
In 1986 Fusetani and co-workers reported the isolation and
X-ray structural analysis of (-)-calyculin A (1, Figure 1), an
architecturally novel marine metabolite derived from the Pacific
sponge Discodermia calyx.1 Seven additional congeners, B-H
(2-8) isolated from the same source were disclosed in 19882
and 1990.3 More recently, calyculin J (9), the calyculinamides
(10-12), and dephosphonocalyculin A (13) were reported.4 The
absolute stereochemistry of calyculin A, assigned initially via
the circular dichroism of the C(33-37) amino acid fragment
obtained by degradation,5 was subsequently confirmed via
asymmetric synthesis of the antipode.6 Structural assignments
of (-)-calyculins B-H were based on spectral comparison with
those of (-)-calyculin A;1 calyculin J, the calyculinamides, and
dephosphonocalyculin A were characterized by chemical cor-
relation.4
The calyculins display a wide variety of biological activities.
Most, including dephosphonocalyculin A, are potent inhibitors
of protein phosphatases 1 and 2A.3,7,8 Calyculins A-D also
exhibit potent cytotoxicity against L1210 leukemia cells,2 and
significant inhibition of cell division in the fertilized egg assay
of both the starfish Asterina pectinifera and Hemicentrotus
pulcherrimus.2 (-)-Calyculin A also stimulates contraction of
Figure 1. The absolute stereochemistry depicted represents the
antipodes of the natural products.
smooth muscles,9 promotes tumor growth in mice,10 and displays
in vivo antitumor activity against Ehrlich and P388 leukemia
in mice.1 Given the potent phosphatase inhibitory activity in
conjunction with the remarkable cell membrane permeability,
(-)-calyculin A has recently found application in the study of
intracellular signal transduction.8,11
(1) Kato, Y.; Fusetani, N.; Matsunaga, S.; Hashimoto, K.; Fujita, S.;
Furuya, T. J. Am. Chem. Soc. 1986, 108, 2780.
(2) Kato, Y.; Fusetani, N.; Matsunaga, S.; Hashimoto, K.; Koseki, K. J.
Org. Chem. 1988, 53, 3930.
(3) Matsunaga, S.; Fujiki, H.; Sakata, D.; Fusetani, N. Tetrahedron 1991,
47, 2999.
(4) (a) Matsunaga, S.; Wakimoto, T.; Fusetani, N.; Suganuma, M.
Tetrahedron Lett. 1997, 38, 3763. (b) Dumdei, E. J.; Blunt, J. W.; Munro,
M. H. G.; Pannell, L. K. J. Org. Chem. 1997, 62, 2636.
(5) Matsunaga, S.; Fusetani, N. Tetrahedron Lett. 1991, 32, 5605.
(6) Hamada, Y.; Tanada, Y.; Yokokawa, F.; Shioiri, T. Tetrahedron Lett.
1991, 32, 5983.
(7) Ishihara, H.; Martin, B. L.; Brautigan, D. L.; Karaki, H.; Ozaki, H.;
Kato, Y.; Fusetani, N.; Watabe, S.; Hashimoto, K.; Uemura, D.; Hartshorne,
D. J. Biochem. Biophys. Res. Commun. 1989, 159, 871.
(8) Sheppeck, J. E., II; Gauss, C.-M.; Chamberlin, A. R. Bioorg. Med.
Chem. 1997, 5, 1739 and references therein.
(9) (a) Ishihara, H.; Ozaki, H.; Koichi, S.; Masatoshi, H.; Karaki, H.;
Watabe, S.; Kato, Y.; Fusetani, N.; Hashimoto, K.; Uemera, D.; Hartshorne,
D. J. J. Pharmacol. Exp. Ther. 1989, 250, 388. (b) Hartshorne, D. J.;
Ishihara, H.; Karaki, H.; Ozaki, H.; Sato, K.; Hori, M.; Watabe, S. AdV.
Prot. Phosphatases; Merlevede, W., Di Salvo, J., Eds.; Leuven University
Press: Louvain, Belgium, 1988; Vol. 5, pp 219-231.
(10) Fujiki, H.; Suganuma, M.; Yoshizawa, H.; Sugimura, T.; Manam,
S.; Kahn, S. M.; Jiand, W.; Hoshina, S.; Weinstein, I. B. Mol. Carcinog.
1989, 2, 184.
10.1021/ja992134m CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/30/1999

Synthesis of (+)-Calyculin A and (-)-Calyculin B
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10469
Not surprisingly, the calyculins have also attracted the
attention of the synthetic community, culminating in recent
notable total syntheses of the unnatural and natural antipodes
of calyculin A [i.e., (+)-1 and (-)-1, respectively] by Evans12
and Masamune,13 a formal total synthesis of (-)-calyculin A
by Shioiri exploiting an advanced Masamune intermediate,14
and a synthesis of calyculin C by Armstrong.15 A number of
synthetic approaches have also been reported.6,16
Scheme 1
Intrigued by the diverse biological activities, as well as the
interesting architectural features, in particular those of the [5.6]
spiroketal, a central theme in our phyllanthostatin and breynolide
synthetic ventures, we initiated work on the total synthesis of
both calyculin A and B.17 Elucidation of the absolute stereo-
chemistry of (-)-calyculin A after the start of this synthetic
program revealed that the natural products were enantiomeric
to our targets. Our strategy is, of course, viable for the natural
calyculins with minor modification.
Synthetic Analysis. From the outset, we envisioned an
approach (Scheme 1) which would provide both calyculins A
and B (1 and 2) from a common advanced intermediate, avoid
extensive manipulations of the light-sensitive3 C(1-9) cyano-
tetraene, and permit flexibility in fragment coupling. Accord-
ingly, disconnections at the C(2) and C(8) olefins led to
phosphonate A, possessing a latent C(3) carbonyl for penultimate
Peterson olefination to access both 1 and 2, after construction
of a C(8,9) trisubstituted olefin. Disconnection of BCDE at the
(11) Cohen, P.; Holmes, C. F. B.; Tsukitani, Y. Trends Biochem. Sci.
1990, 15, 98.
(12) Evans, D. A.; Gage, J. R.; Leighton, J. L. J. Am. Chem. Soc. 1992,
114, 9434.
(13) Tanimoto, N.; Gerritz, S. W.; Sawabe, A.; Noda, T.; Filla, S. A.;
Masamune, S. Angew. Chem., Int. Ed. Engl. 1994, 33, 673.
(14) For a formal synthesis of (-)-1, see: Yokokawa, F.; Hamada, Y.;
Shioiri, T. J. Chem. Soc., Chem. Commun. 1996, 871.
(15) Ogawa, A. K.; Armstrong, R. W. J. Am. Chem. Soc. 1998, 120,
12435.
C(25) olefin revealed subtargets BC, envisioned to arise via
coupling of an acyl anion equivalent (e.g., B) with epoxide C,
and DE, available from oxazole D and lactam E. In this the
first of two full accounts, we present the asymmetric synthesis
of the stereochemically fully endowed C(9-25) BC fragment
corresponding to the enantiomer of the calyculins.17a,18 In the
accompanying paper we present the synthesis of fragments A,
D, and E, their union, and completion of the total syntheses of
(+)-calyculin A and (-)-calyculin B.
The C(9-25) BC subtarget, the core of the calyculins,
encompasses three stereochemical triads [i.e., C(10-13), C(15-
17), and C(21-23)] comprising 11 of the 15 stereogenic centers
and a heptasubstituted [5,6] spiroketal. From the retrosynthetic
perspective, any successful synthesis of this fragment would
require a highly stereoselective strategy, in concert with a
carefully designed orthogonal functional group protection
scheme.
(16) (a) Evans, D. A.; Gage J. R. Tetrahedron Lett. 1990, 31, 6129. (b)
Zhao, Z.; Scarlato, G. R.; Armstrong, R. W. Tetrahedron Lett. 1991, 32,
1609. (c) Hara, O.; Hamada, Y.; Shioiri, T. Synlett 1991, 283. (d) Hara,
O.; Hamada, Y.; Shioiri, T. Synlett 1991, 285. (e) Koskinen, A. M. P.;
Chen, J. Tetrahedron Lett. 1991, 32, 6977. (f) Yokokawa, F.; Hamada, Y.;
Shioiri, T. Synlett 1992, 149. (g) Yokokawa, F.; Hamada, Y.; Shioiri, T.
Synlett 1992, 151. (h) Evans, D. A.; Gage J. R. J. Org. Chem. 1992, 57,
1958. (i) Evans, D. A.; Gage J. R.; Leighton, J. L.; Kim, A. S. J. Org.
Chem. 1992, 57, 1961. (j) Evans, D. A.; Gage J. R.; Leighton, J. L. J. Org.
Chem. 1992, 57, 1964. (k) Vaccaro, H. A.; Levy, D. E.; Sawabe, A.; Jaetsch,
T.; Masamune, S. Tetrahedron Lett. 1992, 33, 1937. (l) Matsubara, J.;
Nakao, K.; Hamada, Y.; Shioiri, T. Tetrahedron Lett. 1992, 33, 4187. (m)
Barrett, A. G. M.; Edmunds, J. J.; Horita, K.; Parkinson, C. J. J. Chem.
Soc., Chem. Commun. 1992, 1236. (n) Barrett, A. G. M.; Edmunds, J. J.;
Hendrix, J. A.; Horita, K.; Parkinson, C. J. J. Chem. Soc., Chem. Commun.
1992, 1238. (o) Barrett, A. G. M.; Edmunds, J. J.; Hendrix, J. A.; Malecha,
J. W.; Parkinson, C. J. J. Chem. Soc., Chem. Commun. 1992, 1240. (p)
Barrett, A. G. M.; Malecha, J. W. J. Chem Soc., Perkin Trans. 1 1994,
1901. (q) Ogawa, A. K.; DeMattei, J. A.; Scarlato, G. R.; Tellew, J. E.;
Chong, L. S.; Armstrong, R. W. J. Org. Chem. 1996, 61, 6153. (r) Pihko,
P. M.; Koskinen, A. M. P. J. Org. Chem. 1998, 63, 92.
Our strategy for BC relied on the successful attachment of
the C(9-13) dipropionate side chain to the C(14-25) spiroketal
(Scheme 1). Consistent with concurrent investigations on the
utility of dithiane-epoxide coupling reactions for the union of
complex fragments, as achieved in our syntheses of FK-506,
rapamycin and discodermolide,19 disconnection of the BC
carbon framework at C(13-14) led to spiroketal epoxide C and
an acyl anion equivalent. Although we focused first on exploit-
ing a C(13) dithiane (vide infra), the eventual successful strategy
employed a cuprate derived from vinyl bromide B.
(17) (a) Smith, A. B., III; Duan, J. J.-W.; Hull, K. G.; Salvatore, B. A.
Tetrahedron Lett. 1991, 32, 4855. (b) Smith, A. B., III; Salvatore, B. A.;
Hull, K. G.; Duan, J. J.-W. Tetrahedron Lett. 1991, 32, 4859. (c) Smith, A.
B., III; Duan, J. J.-W.; Hull, K. G.; Salvatore, B. A.; Bertounesque, E.
Abstracts of Papers, 203rd National Meeting of the American Chemical
Society, San Francisco, CA.; American Chemical Society: Washington,
DC, 1992, ORGN 216. (d) Salvatore, B. A.; Smith, A. B., III. Tetrahedron
Lett. 1994, 35, 1329. (e) Smith, A. B., III; Iwashima, M. Tetrahedron Lett.
1994, 35, 6051. (f) Iwashima, M.; Kinsho, T.; Smith, A. B., III. Tetrahedron
Lett. 1995, 36, 2199. (g) Smith, A. B., III; Friestad, G. K.; Duan, J. J.-W.;
Barbosa, J.; Hull, K. G.; Iwashima, M.; Qiu, Y.; Spoors, P. G.; Ber-
tounesque, E.; Salvatore, B. A. J. Org. Chem. 1998, 63, 7596. (h) Smith,
A. B., III; Friestad, G. K.; Duan, J. J.-W.; Barbosa, J.; Hull, K. G.; Iwashima,
M.; Qiu, Y.; Spoors, P. G.; Salvatore, B. A. Abstracts of Papers, 216th
National Meeting of the American Chemical Society, Boston, MA.;
American Chemical Society: Washington, DC, 1998, ORGN 456. (i) For
complete details and additional examples, see: Smith, A. B., III; Cho, Y.
S.; Friestad, G. K. Tetrahedron Lett. 1998, 39, 8765.
(18) Taken in part from the Ph.D dissertation of J. J.-W. Duan, University
of Pennsylvania, 1992.
(19) For related examples, see: Smith, A. B., III; Condon, S. M.;
McCauley, J. A. Acc. Chem. Res. 1998, 31, 35.

10470 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Smith et al.
Scheme 2
Treatment of the derived tert-butyl carbonate (+)-21 (Scheme
3) with iodine in acetonitrile (-20 °C) furnished the desired
isomer (+)-22 (79%), albeit with only modest selectivity (5.7:
1). We therefore embarked on a program to improve the Bartlett
protocol. Best conditions were IBr in acetonitrile.27 For example,
treatment of (+)-21 with IBr (toluene, -80 to -85 °C)
significantly increased the selectivity (R/â )13.9:1) while
maintaining the yield at 85%. Similar improvements in diaste-
reoselectivity employing IBr proved general.27
Conversion of the major carbonate (+)-22 to epoxide (-)-
23 was then achieved by treatment with potassium carbonate
(3 equiv) in dry methanol. Although standard conditions for
tert-butyldimethylsilyl (TBS) protection of alcohol (-)-23
(TBSCl, imidazole, or TBSOTf, 2,6-lutidine) were unsatisfac-
tory, the combination of TBSCl and TMEDA cleanly afforded
(+)-18.
Dithiane Metalation: Allylic Ether Interference. To con-
struct the C(19-20) bond via union of epoxide (+)-18, we first
examined dithianes of general structure 24. Conversion of
commercially available 2,2-dimethyl-3-hydroxypropanal (25,
Scheme 4) to the corresponding dithiane 26 and subsequent
Spiroketal epoxides of general structure 14 (Scheme 2) were
envisioned to derive from substrate-directed diastereoselective
epoxidation of vinyl spiroketal 15, the latter arising from acyclic
precursor 16 via an acid-catalyzed spiroketalization under
thermodynamic control. That the desired C(19) S configuration
would result was based on steric, anomeric, and hydrogen-
bonding considerations, in conjunction with molecular mechan-
ics calculations for a closely related model.20 Precursor 16 in
turn would be obtained by coupling epoxide 18 with dithiane
17; this strategy held the promise of considerable flexibility in
the introduction of the C(14-15) vinyl group (i.e., either before
or after dithiane coupling).
Scheme 4
Construction of Epoxide 18: Development of an Improved
Iodocarbonate Cyclization. The synthesis of epoxide (+)-18
began with 3-benzyloxypropanal (19, Scheme 3), prepared either
via the Kozikowski three-step synthesis21 or more directly by
monobenzylation of 1,3-propanediol (89% yield) and Swern
oxidation. Reaction with the Brown allylboration agent, Z-
crotyldiisopinocampheylborane,22 furnished alcohol (+)-20 both
in good yield (78%) and enantiopurity (90% ee).23
Swern oxidation provided aldehyde 27. Asymmetric allylbora-
tion with Z-(γ-methoxymethoxyallyl)diisopinocampheylborane,28
Scheme 3
(20) MM2 calculations of the spiroketal shown below revealed that the
desired C(19) S isomer was 6.00 kcal/mol lower in strain energy than the
corresponding C(19) R isomer.
(21) Kozikowski, A. P.; Stein, P. D. J. Org. Chem. 1984, 49, 2301.
(22) (a) Brown, H. C.; Bhat, K. S. J. Am. Chem. Soc. 1986, 108, 293.
(b) Brown, H. C.; Bhat, K. S. J. Am. Chem. Soc. 1986, 108, 5919.
(23) The optical purity of alcohol (+)-20 was determined by comparison
of the optical rotation ([R]25 +12.1°, c 1.096, CHCl3) with the literature
value of its antipode ([R]25 D -13.2°, c 0.972, CHCl3) of known optical
D
purity (98.5% e.e.). Kobayashi, Y.; Uchiyama, H.; Kanbara, H.; Kusakabe,
M.; Sato, F. Heterocycles 1987, 25, 549.
(24) For a review on epoxidation of olefins, see: Rao, A. S.; Paknikar,
S. K.; Kirtane, J.G. Tetrahedron 1983, 39, 2323.
(25) (a) Fukuyama, T.; Vranesic, B.; Negri, D. P.; Kishi, Y. Tetrahedron
Lett. 1978, 2741. (b) Mihelich, E. D.; Daniels, K.; Eickhoff, D. J. J. Am.
Chem. Soc. 1981, 103, 7690.
(26) (a) Bartlett, P. A.; Meadows, J. D.; Brown, E. G.; Morimoto, A.;
Jernstedt, K. K. J. Org. Chem. 1982, 47, 4013. (b) For cyclizations of lithium
carbonates, see: Cardillo, G.; Orena, M.; Porzi, G.; Sandri, S. J. Chem.
Soc., Chem. Commun. 1981, 465. Bongini, A.; Cardillo, G.; Orena, M.;
Porzi, G.; Sandri, S. J. Org. Chem. 1982, 47, 4626.
(27) (a) Duan, J. J.-W.; Sprengeler, P. A.; Smith, A. B., III. Tetrahedron
Lett. 1992, 33, 6439. (b) Duan, J. J.-W.; Smith, A. B., III. J. Org. Chem.
1993, 58, 3703.
At this point, stereoselective conversion of alcohol (+)-20
to R-epoxide (-)-23 was required. As anticipated,24 epoxidation
with t-BuOOH/VO(acac)225 provided a mixture of epoxides (R/â
) 3:2). We therefore turned to the Bartlett iodine-induced
carbonate cyclization26 as a means of 1,3-stereoinduction.

Synthesis of (+)-Calyculin A and (-)-Calyculin B
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10471
generated in situ from methoxymethyl allyl ether by successive
treatment with s-butyllithium, (+)-B-methoxydiisopinocam-
pheylborane, and boron trifluoride etherate then provided alcohol
(+)-24a in 47% yield.29 Simple protection of alcohol (+)-24a
as a TBS ether furnished (+)-24b.
Cognizant that protecting groups can often prove critical in
the metalation of dithianes,30,31 we prepared three additional
dithianes (24c-e). Selective removal of the methoxymethyl-
protecting group in (+)-24b with bromodimethylborane (Me2-
BBr)32 at -78 °C afforded dithiane (-)-24c (Scheme 4);
hydroxyl protection as a tetrahydropyranyl ether gave dithiane
24d as a diastereomeric mixture ( ∼1:1). A fourth dithiane, (-)-
24e, was obtained by removal of the methoxymethyl moiety
from (+)-24a and conversion to the acetonide.
role in promoting dithiane acidity is now well established.39 In
particular, the unfilled σ* orbital has been invoked to explain
both the enhanced acidity of the equatorial proton and the
preferential equatorial conformation of the lithium in metalated
dithianes.39 For 24a-e, the Thorpe-Ingold effect would
promote conformations i and ii with the alkene in close
proximity to the dithiane.40
Figure 2.
Dithianes 24b-e were then examined for their susceptibility
to metalation. These studies were performed by treatment of
the dithiane with one of a variety of bases, followed by addition
of methyl alcohol-d or deuterium oxide; the extent of deuterium
In conformation i, metalation would be expected to be slow
due to inefficient overlap of the axial C-H bond with C-S
σ*. In conformation ii the filled alkene π orbital could interact
with C-S σ*, and thereby interfere with the role of the C-S
σ* orbital in promoting dithiane C-H acidity. Thus, neither
conformation favors deprotonation. Danishefsky recently in-
voked a similar conformation and π-donor effect to explain the
anomalous stereoselectivity in an enolate aldol.41
Dithiane (-)-31: Precursor to Spiroketal 15. To avoid the
above problem, while maintaining the same overall strategy for
C(19-20) bond construction, we selected the simpler dithiane
31 (Scheme 6); introduction of the C(14-15) vinyl group would
take place at a latter stage.
1
incorporation was determined by H NMR. Various solvents,
temperatures,33 and additives [hexamethylphosphoramide
(HMPA), tetramethylethylenediamine (TMEDA), 1,4-diaza-
bicyclo[2.2.2]octane (DABCO),34 or 1,3-dimethyl-3,4,5,6-tet-
rahydro-2(1H)-pyrimidinone (DMPU)35] were explored. In
general, the dithianes either decomposed or were recovered with
little or no deuterium incorporation. The only successful case
entailed treatment of (-)-24e with n-butyllithium and potassium
t-butoxide in THF/pentane at -78 °C. Addition of D2O afforded
∼60% deuterium incorporation; however recovery of (-)-24e
was poor.
Scheme 6
Although dithianes have been widely used as acyl anion
equivalents, few reported examples include allylic ether
functionality.30,36,37d In the successful cases, the allylic ether
π-systems were not terminal. Our results suggest that the
terminal allylic ether and not the protecting groups nor steric
hindrance is the source of the difficulty30,31,37 since the simpler
dithiane 28 with both similar protection and steric hindrance
was readily metalated (Scheme 5).38
Scheme 5
The synthesis of dithiane 31 began with known alcohol (+)-
32 (Scheme 7), which was prepared from the dimethyl ester of
(36) During our successful rapamycin and FK506 syntheses, the dithianes
i and ii, each containing nonterminal allylic ethers, were successfully
metalated (t-BuLi, HMPA/THF, -78 °C). (a) Smith, A. B., III; Condon, S.
M.; McCauley, J. A.; Leazer, J. L., Jr.; Leahy, J. W.; Maleczka, R. E., Jr.
J. Am. Chem. Soc. 1997, 119, 947. (b) Smith, A. B., III; Chen, K.; Robinson,
D. J.; Laakso, L. M.; Hale, K. J. Tetrahedron Lett. 1994, 35, 4271.
A possible explanation can be found in the interaction
between the alkene π-system and the unfilled C-S σ* orbital
(Figure 2). That the C-S σ* orbital plays a key stereoelectronic
(28) Brown, H. C.; Jadhav, P. K.; Bhat, K. S. J. Am. Chem. Soc. 1988,
110, 1535.
(29) The modest yield of (+)-24a is partially due to dithiane oxidation
during oxidative workup. The alternative oxidant trimethylamine N-oxide
might potentially avoid this problem; see: Kabalka, G. W.; Hedgecock, H.
C., Jr. J. Org. Chem. 1975, 40, 1776.
(30) Corey, E. J.; Hua, D. H.; Pan, B.-C.; Seitz, S. P. J. Am. Chem. Soc.
1982, 104, 6818.
(31) Park, P.; Broka, C. A.; Johnson, B. F.; Kishi, Y. J. Am. Chem. Soc.
1987, 109, 6205.
(32) Guindon, Y.; Morton, H. E.; Yoakim, C. Tetrahedron Lett. 1983,
24, 3969.
(33) The temperature of the deprotonation study was limited to 0 °C or
below due to the observation that SN2′ displacement by the n-butyl group
begins to occur at 0 °C or above. For example, when 24d was treated with
n-BuLi/TMEDA/THF for 10 min at -20 °C and then at 1 h at 0 °C, the
SN2′ product was isolated in 24% yield as well as recovered starting material
with no deuterium incorporation.
(37) Dithianes bearing acetonides, tert-butyldimethylsilyl ethers, tetra-
hydropyranyl ethers, and methoxymethyl ethers have been successfully
metalated. (a) Toshima, H.; Suzuki, T.; Nishiyama, S.; Yamamura, S.
Tetrahedron Lett. 1989, 30, 6725. (b) Khandekar, G.; Robinson, G. C.;
Stacey, N. A.; Steel, P. G.; Thomas, E. J.; Vather, S. J. Chem. Soc., Chem.
Commun. 1987, 877. (c) Mori, Y.; Suzuki, M. Tetrahedron Lett. 1989, 30,
4383. (d) Jones, A. B.; Villalobos, A.; Linde, R. G., II; Danishefsky, S. J.
J. Org. Chem. 1990, 55, 2786.
(38) We and others have shown that hindered dithianes can be success-
fully deprotonated.31,36,37a
(39) Seebach, D.; Gabriel, J.; Ha¨ssig, R. HelV. Chim. Acta 1984, 67,
1083 and references therein.
(40) (a) Beesley, R. M.; Ingold, C. K.; Thorpe, J. F. J. Chem. Soc. 1915,
107, 1080. (b) Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry of
Organic Compounds; Wiley: New York, 1994; pp 682-683.
(34) Screttas, C. G.; Eastham, J. F. J. Am. Chem. Soc. 1965, 87, 3276.
(35) Mukhopadhyay, T.; Seebach, D. HelV. Chim. Acta 1982, 65, 385.

10472 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Smith et al.
Scheme 7
single experiment (81% yield). Importantly, the dithiane-
epoxide coupling tactic generates a single masked aldol.
With the C(19-20) linkage established, we turned to
introduction of the C(14-15) vinyl substituent. Protection of
alcohol (-)-35 as the TBS ether [(-)-36; Scheme 8] followed
by DIBAL reductive cleavage of the p-methoxybenzylidene
acetal45 furnished alcohol (+)-37 in 94% yield, nicely exposing
the primary alcohol with concomitant orthogonal protection of
the C(17) secondary alcohol as the p-methoxybenzyl (PMB)
ether. Parikh-Doering oxidation (DMSO, SO3‚pyr, Et3N)46
followed by addition of vinylmagnesium bromide furnished
alcohol (-)-30 in 81% yield (two steps) with excellent diaste-
reoselectivity (>20:1, 1H NMR). The C(16) S configuration of
(-)-30, initially assigned on the basis of a chelation-control
1
model,47 was confirmed by H NMR analysis of the corre-
sponding mandelate esters48 as illustrated in Scheme 9.
(S)-malic acid.42 Swern oxidation43 and treatment with 1,3-
propanedithiol and boron trifluoride etherate provided the
dithiane along with concomitant removal of the pentylidene
moiety; the yield of diol (+)-34 was 86% (two steps). Diol
protection proceeded smoothly to furnish 31 as a 3:2 mixture
of diastereomeric p-methoxybenzyl acetals (85% yield). Acid-
catalyzed equilibration with camphorsulfonic acid (CSA) in
hexanes led to crystallization of one diastereomer; reequilibration
of the mother liquors effected complete conversion to a single
diastereomer in 81% yield (acetal configuration not determined).
In contrast with dithianes 24b-e, metalation of (-)-31 (t-
BuLi, HMPA/THF, -78 °C, 15 min or n-BuLi, TMEDA, -20
°C, 1.5 h) proceeded without complication, as judged by
deuterium incorporation (1H NMR; >90%) This result is again
in accord with our speculation that the allylic ether moiety in
dithianes 24b-e prevents successful metalation.
Scheme 9
Metalation of 31a,b (3:2 mixture of diastereomers, 1.5 equiv)
with n-butyllithium (1.5 equiv) in THF and TMEDA (6 equiv)
followed by addition of DMPU (6 equiv) and epoxide (+)-18
(1 equiv) at -20 °C led to alcohol 35 in 90% yield. In the
absence of DMPU, the yield of 35 decreased to 75%.44 Upon
scale-up, best results entailed deprotonation of (-)-31 (single
diastereomer) with n-BuLi in 10% HMPA/THF (Scheme 8),
followed by introduction of a slight excess (1.1 equiv) of epoxide
(+)-18; these conditions afforded 120 g of alcohol (-)-35 in a
Scheme 8
Spiroketalization. To our delight treatment of (-)-30 with
a mixture of 48% aqueous hydrogen fluoride, acetonitrile, and
dichloromethane (1:9:50) for 12 h at room temperature provided
spiroketal (+)-44a as the only diastereomer in 88% yield
(Scheme 10). Interestingly, Evans and co-workers obtained a
5:1 diastereomeric mixture with a closely related substrate.12
(43) (a) Mancuso, A. J.; Swern, D. Synthesis 1981, 165. (b) Tidwell, T.
T. Synthesis 1990, 857. (c) Tidwell, T. T. Org. React. 1991, 39, 297.
(44) DMPU is known to facilitate opening of epoxides with dithiane
anions. Lipshutz, B. H.; Moretti, R.; Crow, R. Tetrahedron Lett. 1989, 30,
15.
(45) Takano, S.; Akiyama, M.; Sato, S.; Ogasawara, K. Chem. Lett. 1983,
1593.
(46) Parikh, J. R.; Doering, W. E. J. Am. Chem. Soc. 1967, 89, 5505.
(47) The selectivity can be rationalized via the chelation-controlled model
illustrated below.
(41) Harris, C. R.; Kuduk, S. D.; Balog, A.; Savin, K. A.; Danishefsky,
S. J. Tetrahedron Lett. 1999, 40, 2267.
(42) (a) Seebach, D.; Aebi, J.; Wasmuth, D. Organic Syntheses; Wiley
& Sons: New York, 1990; Collect. Vol. VII, p 153. (b) Lavalle´e, P.; Ruel,
R.; Grenier, L.; Bissonnette, M. Tetrahedron Lett. 1986, 27, 679.
(48) Trost, B. M.; Belletire, J. L.; Godleski, S.; McDougal, P. G.;
Balkovec, J. M.; Baldwin, J. J.; Christy, M. E.; Ponticello, G. S.; Varga, S.
L.; Springer, J. P. J. Org. Chem. 1986, 51, 2370.

Synthesis of (+)-Calyculin A and (-)-Calyculin B
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10473
Scheme 10
desired â-epoxides 46a-c,51 whereas 44d gave a complex
mixture. Metal-catalyzed epoxidation with t-butyl hydroperoxide
and VO(acac)2 or Mo(CO)6 resulted only in decomposition.
We next explored catalytic osmium tetroxide dihydroxylation
of 44c. Although osmylation and peracid epoxidation often
afford complementary stereoselectivities,52 dihydroxylation of
44c provided the R and â diols 47c and 48c in a 1:8 ratio53
(85% yield, Scheme 12). To surmount the apparent intrinsic
substrate diastereofacial bias via reagent control, we turned to
the Sharpless asymmetric dihydroxylation (AD) process.54 In
no case however were we able to override the intrinsic substrate
bias. For a detailed discussion of these studies see our earlier
publication.17f
Scheme 12
The stereochemistry of spiroketal (+)-44a was assigned on the
basis of both proton coupling constants and nOe measure-
ments.49
To retain the useful PMB ether at C(17) an alternate protocol
was developed (Scheme 10). Treatment of (-)-30 with tetrabu-
tylammonium fluoride furnished the corresponding triol in 95%
yield; removal of the dithiane with concomitant spiroketalization
(HgCl2, CaCO3) then afforded a mixture of spiroketal epimers
(44d and 43) in a combined yield of 89%. Although the
undesired spiroketal epimer predominated (48% yield), presum-
ably as a result of kinetic control, it could be readily separated
and converted entirely to (+)-44d upon exposure to TsOH in
CH2Cl2 (73% yield). Thus, the total yield of (+)-44d for the
two steps was 76%. The structure of (+)-44d was subsequently
confirmed by crystallographic analysis of the corresponding
â-epoxide (vide infra).
Epoxidation of the C(15-16) Olefin: A Difficult Maneu-
ver. Although our analysis of vinyl spiroketals such as 44a did
not yield a clear-cut stereochemical prediction for epoxidation,
Ohmura and co-workers50 observed a modest preference in the
m-CPBA epoxidation of two related 2-alkenyltetrahydrofurans.
In our case however, treatment of vinylspiroketals 44a-c18
(Scheme 11) with m-CPBA furnished predominantly the un-
Fortunately, after considerable experimentation we discovered
that Payne epoxidation (peroxybenzimidic acid)55 of 44d
occurred with complete reversal of the intrinsic diastereofacial
bias (9.5:1; 89% yield; Table 1).17e Despite the structural
similarities between peracids and peroxybenzimidic acid, peracid
and Payne epoxidations have previously expressed opposite
diastereofacial preferences.56 Interestingly, differential protection
of the two secondary hydroxyl groups (e.g., 44b) led to a
dramatic reversal of diastereoselectivity. Taken together these
results suggest an unusually complex interplay of conformational
effects and steric and hydrogen bonding interactions in the Payne
epoxidation process.
Table 1. Payne Epoxidation of Vinyl Spiroketals
alkene
R/â (45:46)a
yield (%)b
82
Scheme 11
44a (R1 ) R2 ) H)
3:1
1:18
44b (R1 ) H, R2 ) BPS)
44c (R1 ) TBS, R2 ) BPS)
44d (R1 ) PMB, R2 ) H)
44e (R1 ) Ac, R2 ) H)
44f (R1 ) TBS, R2 ) H)
44g (R1 ) Ac, R2 ) BPS)
86
no reactionc
9.5:1d
4:1
8.6:1
89
85
80
no reactionc
a Determined by 1H NMR spectroscopy. b Combined isolated yield.
c No reaction occurred at 0 °C or at room temperature. d Ratio of (+)-
C:46d.
(49) Except for the large geminal coupling between H7 and H8 (14.1
Hz), all other coupling constants were small, falling in the range of 1.9 Hz
to 3.3 Hz, suggesting no diaxial coupling in the pyran ring; large coupling
constants (5-10 Hz) would be expected for diaxial hydrogens in the C(19)
R chair isomer. A twist boat conformation would also result in coupling
constants larger than the observed 3.3 Hz. NOE enhancements of the H4
(13.8%) and H8 (8.4%) hydrogens when the Me2 was irradiated support
assignment of the C(19) S isomer.
Importantly, epoxides (+)-C and (+)-46a formed good
quality single crystals for X-ray crystallography, thus permitting
(51) The fact that the major products in all three cases [(+)-46a, (+)-
46b and (+)-46c] possess the same epoxide configuration was confirmed
by chemical conversion of (+)-46c to (+)-46b and (+)-46b to (+)-46a
with tetrabutylammonium fluoride.
(52) For discussion, see: Chao, T.-M.; Baker, J.; Hehre, W. J.; Kahn,
S. D. Pure Appl. Chem. 1991, 63, 283.
(53) The major isomer was readily converted to â-epoxide 46c (TsCl,
pyridine; K2CO3, MeOH; 86%). The configurations of the oxidation products
were correlated with the known epoxide 46b or the tetraol. The latter
structures were determined via X-ray crystallographic analyses.
(54) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. ReV.
1994, 94, 2483.
(55) Payne, G. B. Tetrahedron 1962, 18, 763.
(56) (a) Carlson, R. G.; Behn, N. S. J. Org. Chem. 1967, 32, 1363. (b)
Woodward, R. B.; Gosteli, J.; Ernest, I.; Friary, R. J.; Nestler, G.; Raman,
H.; Sitrin, R.; Suter, C.; Whitesell, J. K. J. Am. Chem. Soc. 1973, 95, 6853.
(50) Kawahata, Y.; Takatsuto, S.; Ikekawa, N.; Murata, M.; Ohmura, S.
Chem. Pharm. Bull. 1986, 34, 3102.

10474 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Smith et al.
stereochemical assignment not only of the epoxide but also the
spiroketal.57 Having established a viable sequence to epoxide
C,58 we turned to the attachment of a C(9-13) subunit.
determined via D2O treatment and NMR analysis ( ∼90% d).
Moreover, the metalated dithiane reacted smoothly with both
1,2-epoxybutane (76% yield) and methyl iodide (90% yield).
Scheme 14
Figure 3. ORTEP Drawing of Epoxide (+)-C
Fragment Coupling via a Dithiane. Dithiane (+)-49 was
prepared beginning with the Roush crotylboration product (+)-
5059 (Scheme 13). Protecting-group interchange to acetonide
(+)-51, ozonolysis, and conversion of the resultant aldehyde
to the corresponding dithiane (1,3-propanedithio-bis-trimethyl-
silane, ZnI2)60 afforded (+)-49 in 50% yield.
Unfortunately, lithiated 49 (n-butyllithium, TMEDA, HMPA/
THF) failed to react with epoxides 45b,c and 45h,i61 required
for the calyculin synthesis (Scheme 15). The use of other bases
(t-BuLi or n-BuLi/t-BuOK) and/or additives (DMPU or HMPA)
provided no improvement. In most cases, both starting materials
were recovered.
Scheme 13
Scheme 15
Metalation of (+)-49 (Scheme 14) was successfully effected
with n-butyllithium (TMEDA, HMPA/THF, 1.5 h, 0 °C) as
(57) The epoxides described in Table 1 were chemically correlated with
(+)-C or (+)-46a, thereby confirming the structural assignments.
(58) We also explored an alternative which would (a) address the C(15)
stereochemistry by chelation-controlled addition of acetylene i to aldehyde
ii and (b) further elucidate the scope of the IBr-induced iodocarbonate
cyclization, in this context to be used to install and define the C(13) oxygen
stereogenicity after reductive removal of the iodide. We successfully
achieved the former objective via reaction of aldehyde ii with the acetylenic
Grignard reagent derived from i to afford adduct iii (R/â ) 15:1),
presumably through magnesium chelation. Unfortunately, the iodocarbonate
cyclization could not be realized; under a variety of conditions, no reaction
or decomposition occurred. We speculate that steric hindrance at the olefin
of carbonates iv and a related C(9) acetate precludes the reaction under
mild conditions; the lability of the spiroketal functionality and protecting
groups towards iodine monobromide prevented more forcing conditions.
For complete details and for preparation of i and ii, see: Duan, J. J.-W.
Ph.D. Thesis, University of Pennsylvania, 1992.
In an attempt to circumvent the lack of reactivity we reversed
the functionality, attaching the dithiane moiety to the spiroketal
with the expectation that metalation of 56a would lead to
successful coupling with epoxide 57. This alternative benefited
from the Nakata precedent62 that reaction of lithiated 1,3-dithiane
with a similar epoxide proceeded in 99% yield. In the event,
the reactivity of the epoxide (45) proved to be sensitive to the
nature of protecting groups; with a TBS group at C(21) no
reaction occurred. However, the PMB-protected epoxide 45i
afforded an encouraging 65% yield of 56a after in situ
O-methylation at C(15).
Also encouraging, metalation of dithiane 56a with n-BuLi
in 10% HMPA/THF, followed by treatment with D2O afforded
the deuterated dithiane (56b) in 90% yield with 95% incorpora-
tion of one deuterium (Scheme 16). However, exposure of the
metalated dithiane to epoxide 57 led to no coupling; only modest
recovery of the dithiane was possible (48-59%).63,64
Fragment Coupling via a Vinyl Cuprate. Thwarted by the
unusually recalcitrant dithiane-epoxide coupling, we shifted
(61) Protected spiroketals 45h and 45i were prepared from (+)-C by
reaction with PMBOCH2Cl (iPr2NEt, CH2Cl2) and p-methoxybenzyl chloride
(KHMDS, THF), respectively. The former reagent was obtained by known
methods. (a) Corey, E. J.; Bock, M. G. Tetrahedron Lett. 1975, 3269. (b)
Benneche, T.; Strande, P.; Undheim, K. Synthesis 1983, 762.
(62) Nakata, T.; Fukui, M.; Ohtsuka, H.; Oishi, T. Tetrahedron 1984,
40, 2225.
(63) Coupling experiments with the TBS-protected analogue of 57 were
likewise unsuccessful.
(64) A side product derived from dithiane 56a exhibiting 1H NMR signals
associated with a monosubstituted alkene was isolated in low yield (<10%);
this material evidently formed by elimination of the benzyloxy group from
C(25). For a similar side reaction in our penitrem synthetic studies, see:
Smith, A. B., III; Haseltine, J. N.; Visnick, M. Tetrahedron 1989, 45, 2431.
(59) Roush, W. R.; Palkowitz, A. D.; Ando, K. J. Am. Chem. Soc. 1990,
112, 6348.
(60) Evans, D. A.; Truesdale, L. K.; Grimm, K. G.; Nesbitt, S. L. J.
Am. Chem. Soc. 1977, 99, 5009.

Synthesis of (+)-Calyculin A and (-)-Calyculin B
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10475
Scheme 16
epoxides. For example, fully protected epoxide (+)-46c failed
to react with lithium di-n-butylcyanocuprate in the presence of
boron trifluoride etherate or lithium n-butyl-2-thienylcyanocu-
prate (Scheme 19). In contrast, a diastereomeric mixture of
dihydroxyl epoxides (+)-45a and (+)-46a reacted smoothly with
lithium di-n-butylcyanocuprate at -78 °C to yield 65. With this
encouraging result, we proceeded with the coupling employing
cuprate 59.
Scheme 19
our focus to an alternate acyl anion equivalent. The reaction of
a mixed vinyl cuprate with an epoxide65 appeared to be an
attractive option; oxidative cleavage of the resulting exometh-
ylene unit would be equivalent to removal of a dithiane to
unmask the aldol linkage. For this strategy, we required a vinyl
halide for in situ generation of 59 (Scheme 17).
Initially, we planned to generate vinyl lithium 66 from vinyl
stannane (-)-64 via transmetalation.70 However, when a mixture
of (-)-64 and n-butyllithium was stirred at -78 °C, no reaction
occurred (Scheme 20). At -40 °C or above, transmetalation
was accompanied by decomposition. The reasons responsible
for the sluggish tin-lithium exchange remain unknown, al-
though Collins reported that exchange reactions can be affected
by olefin stereochemistry.71 Attempts to convert (-)-64 directly
to cuprate 67 with lithium dimethylcyanocuprate72 resulted in
decomposition.
Scheme 17
Scheme 20
We first employed vinyl iodide (-)-60 (Scheme 18), prepared
from Roush crotylboration product (+)-50 (Scheme 13). Protec-
tion as a p-methoxybenzylidene acetal (Scheme 18), followed
by Wacker oxidation66 led to methyl ketone (-)-62 in modest
yield. Conversion to the kinetic enol triflate67 and treatment with
in situ-generated Bu2CuLi‚LiCN,68 furnished vinyl stannane (-)-
64 which was then transformed to vinyl iodide (-)-60 (I2,
CCl4).69,70 Although the overall yield for this five-step process
was not suitable for large-scale material advancement, we
proceeded with initial attempts to effect the coupling reaction.
We next explored the formation of vinyl lithium 66 via
lithium-iodine exchange73 (Scheme 21). Treatment of (-)-60
in tetrahydrofuran with tert-butyllithium at -78 °C for 30 min
generated the desired vinyl lithium (66) without incident. The
mixed cuprate (6 equiv) derived from 66 and lithium 2-thien-
ylcyanocuprate reacted with epoxide (+)-45a (1 equiv) smoothly
to furnish triol (+)-68 in 95% yield. This gratifying result,
although requiring a large excess of vinyl cuprate due to the
free hydroxyls in epoxide (+)-45a, validated our approach. We
Scheme 18
(66) (a) Tsuji, J.; Nagashima, H.; Nemoto, H. Organic Syntheses; Wiley
& Sons: New York, 1990; Collect. Vol. VII, p 137. (b) Tsuji, J. Synthesis
1984, 369.
(67) McMurry, J. E.; Scott, W. J. Tetrahedron Lett. 1983, 24, 979.
(68) Gilbertson, S. R.; Challener, C. A.; Bos, M. E.; Wulff, W. D.
Tetrahedron Lett. 1988, 29, 4795.
(69) Davies, J.; Roberts, S. M.; Reynolds, D. P.; Newton, R. F. J. Chem.
Soc., Perkin Trans. 1 1981, 1317.
(70) Hanson, R. N.; El-Wakil, H. J. Org. Chem. 1987, 52, 3687.
(71) Collins, P. W.; Jung, C. J.; Gasiecki, A.; Pappo, R. Tetrahedron
Lett. 1978, 3187.
Model studies with n-butyl cuprates revealed that protecting
groups again play an important role in the reactivity of spiroketal
(65) (a) Bartlett, P. A.; Meadows, J. D.; Ottow, E. J. Am. Chem. Soc.
1984, 106, 5304. (b) Lipshutz, B. H.; Kozlowski, J. A.; Parker, D. A.;
Nguyen, S. L.; McCarthy, K. E. J. Organomet. Chem. 1985, 285, 437.
(72) Behling, J. R.; Babiak, K. A.; Ng, J. S.; Campbell, A. L.; Moretti,
R.; Koerner, M.; Lipshutz, B. H. J. Am. Chem. Soc. 1988, 110, 2641.
(73) Corey, E. J.; Beames, D. J. J. Am. Chem. Soc. 1972, 94, 7210.

10476 J. Am. Chem. Soc., Vol. 121, No. 45, 1999
Smith et al.
next sought both to improve the preparation of the vinyl halide,
in particular the Wacker oxidation, and to minimize the amount
of the cuprate. During this refinement, vinyl bromide (+)-B
emerged as a more viable cuprate precursor (vide infra).
TLC), pure 70 [(86% yield, [R]23 0° (c 2.93, CHCl3)] was
D
obtained following aqueous workup and chromatography.
Gratifyingly, no evidence of acetonide hydrolysis was detected
1
either by TLC or H NMR.
Subsequent experimentation provided further insight and
improvement. The utilization of copper(II) acetate could be
reduced to substoichiometric levels (20 mol % Cu(OAc)2, 2 d,
84% yield of 70; Scheme 23). In addition, since the Cu(OAc)2
Scheme 21
Scheme 23
Wacker Oxidation: Improved Conditions for Acid-Sensi-
tive Substrates.17i Although the palladium-catalyzed Wacker
oxidation of terminal olefins to methyl ketones occupies a
prominent position among methods for synthesis of natural
products,74 oxidation of (+)-51 employing the standard Tsuji
conditions (10 mol % PdCl2, 2 equiv CuCl, O2, DMF, H2O)66
proved both sluggish, and on larger scales led both to partial
acetonide hydrolysis and problematic emulsions.75
redox shuttle begins with Cu(II), preoxidation of the catalyst
system by O2 is not required.77 The amount of insoluble material
is also reduced; thus, the reactions are more easily stirred,
perhaps facilitating oxygen uptake. Removal of soluble Cu-
(OAc)2, even in stoichiometric amounts, is also greatly facilitated
without formation of emulsions upon workup. Finally scale-up
(63 mmol) afforded a similar yield of ketone 70 (82%). Methyl
ketone 70 (Scheme 23) was next transformed to the kinetic enol
triflate, followed by reaction with the mixed stannyl thienyl
cuprate derived from lithium 2-thienylcyanocuprate.78 Titration
of the resultant vinyl stannane with N-bromosuccinimide
completed the preparation of vinyl bromide (+)-B. The opti-
mized route provided (+)-B in 54% yield (six steps) from the
Roush crotylboration product (+)-61.
Vinyl Cuprate-Epoxide Coupling: Optimization. To
minimize consumption of vinyl bromide (+)-B, as well as to
facilitate subsequent manipulations, we explored the coupling
reaction with the C(21) hydroxyl in spiroketal (+)-C protected.
As observed previously with vinyl iodide (-)-60, the protection
scheme proved to have a dramatic effect on the coupling
reaction. Although initial experiments suggested the use of a
2-trimethylsilylethoxymethyl (SEM) ether for C(21) protection,
removal of this group presented significant challenges later in
the synthesis. We therefore chose a TBS group; accordingly,
we proceeded with epoxide 45j (Scheme 24), prepared in 98%
yield by silylation of (+)-C. Optimal conditions79 entailed
generation of the vinyl lithium species (2 equiv t-BuLi, Et2O,
-78 f 0 °C, 30 min), introduction of freshly prepared lithium
2-thienylcyanocuprate (-78 f -45 °C, 30 min), and addition
of the epoxide at -78 °C, followed by stirring for 2 d at -5
°C. Under these conditions, the desired coupling product (+)-
71 formed in 83% yield on ∼2 g scale. Importantly, this
coupling tactic again generated a single diastereomeric masked
aldol.
Examination of the accepted Wacker catalytic cycle66 (Scheme
22) reveals that 1 equiv of HCl is extruded during the formation
of the organopalladium intermediate (69) and a second is
formally lost from palladium hydride prior to regeneration of
PdCl2 for a new catalytic cycle. Although 2 equiv of HCl are
taken up in the copper redox shuttle, and thus net production
of HCl is not expected, transient accumulation of HCl must
occur. Not unexpectedly, addition of base (Na2CO3) shuts down
the catalytic cycle. We reasoned that replacing CuCl with Cu-
(OAc)2 (i.e., replacing HCl in the catalytic cycle with HOAc)
might suppress the hydrolysis of the acetonide while maintaining
a weakly acidic medium to permit catalytic turnover.76
Scheme 22
(75) The major side product had 1H NMR properties consistent with
hemiketal i.
To this end, alkene (+)-51 (29 mmol) was added to PdCl2
(10 mol %) and Cu(OAc)2‚H2O (58 mmol) in AcNMe2/H2O
(7:1, 57 mL) and the ambient atmosphere replaced with oxygen
(Scheme 23). After complete consumption of (+)-51 (30 h,
(74) (a) For an application to synthesis of pseudomonic acid C, see:
Balog, A.; Yu, M. S.; Curran, D. P.; Yu, G.; Carcanague, D. R.; Shue,
Y.-K. Synth. Commun. 1996, 26, 935. (b) For an application to synthesis
of swinholides, see: Paterson, I.; Cumming, J. G.; Smith, J. D.; Ward, R.
A.; Yeung, K.-S. Tetrahedron Lett. 1994, 35, 3405. (c) For other recent
synthetic applications, see: Pellissier, H.; Michellys, P.-Y.; Santelli, M.
Tetrahedron Lett. 1994, 35, 6481; Money, T.; Wong, M. K. C. Tetrahedron
1996, 52, 6307.
(76) Cupric acetate has been employed in industrial Wacker oxidations,
see: Hrusovsky, M.; Vojtko, J.; Cihova, M. Hung. J. Ind. Chem. 1974, 2,
137; Chemical Abstracts 1975, 82, 139205.
(77) With CuCl as the redox shuttle reagent, a mixture of PdCl2 and
CuCl is typically stirred under O2 for 1-2 h prior to introduction of the
alkene substrate. It has been noted that the use of CuCl2 leads to chlorinated
byproducts, see ref 66.
(78) Piers, E.; Tillyer, R. D. J. Org. Chem. 1988, 53, 5366.

Synthesis of (+)-Calyculin A and (-)-Calyculin B
J. Am. Chem. Soc., Vol. 121, No. 45, 1999 10477
Scheme 24
Completion of the C(9-25) subtarget entailed conversion of
triol (+)-75 to the corresponding primary monopivaloate ester
(Scheme 25), silylation of the two secondary hydroxyl groups
with TBSOTf to afford (+)-77 in excellent yield, and orthogonal
deprotection by oxidative hydrolysis of the PMB ether with
DDQ to furnish the C(17) alcohol in 92% yield. Phosphorylation
using the three-stage protocol introduced by Evans12 completed
construction of the fully elaborated C(9-25) subtarget (+)-BC
in 91% yield.
Scheme 25
Elaboration of the C(9-25) BC Subtarget. We next
addressed introduction of both the C(13) stereogenic center and
the C(17) phosphate, along with functional group manipulations
required to prepare BC for union with subtargets A and DE
(see Scheme 1). O-Methylation installed the C(15) O-methyl
group (Scheme 24); oxidative cleavage of the alkene using
catalytic osmylation and sodium periodate exposed the C(13)
carbonyl. Selective DIBAL reduction then afforded â alcohol
(+)-74 (â/R >12:1), presumably by internal delivery of hydride
via prior coordination with the C(15) methoxyl group.80
In summary, assembly of (+)-BC, the C(9-25) spiroketal
dipropionate fragment for the unnatural antipodes of the
calyculins, has been achieved in a convergent, completely
stereocontrolled fashion. The longest linear sequence spans 26
steps from commercially available 1,3-propanediol, providing
(+)-BC in 3.8% overall yield (87% yield per step). The
accompanying paper describes the preparation of subtargets A
and (+)-DE, their union with (+)-BC, and completion of the
total synthesis of (+)-calyculin A and (-)-calyculin B.
To facilitate manipulations leading to the anticipated Horner-
Emmons olefination at C(9), the acetonide was removed to
afford triol (+)-75. On a small scale this transformation
proceeded in 64% yield; on scale-up the yield decreased to 51%.
A mitigating feature of this transformation was the isolation of
acetonide migration product (+)-76 (10% yield),81 which upon
13C NMR spectroscopic analysis of the acetonide signals
permitted assignment of the stereochemical course of the DIBAL
reduction.82
Acknowledgment. Support was provided by the National
Institutes of Health (National Cancer Institute) through Grant
CA-19033 and postdoctoral fellowships to G.K.F. and K.G.H.
We also thank DuPont Pharmaceuticals, Inc. and Zeneca
Pharmaceuticals, Inc. for support of this work. In addition, we
thank Drs. George T. Furst and Patrick J. Carroll and Mr. John
Dykins of the University of Pennsylvania Spectroscopic Service
Center for assistance in securing and interpreting high-field
NMR spectra, X-ray crystal structures, and mass spectra,
respectively.
(79) Coupling in THF resulted in modest yields (6-37%), consistent
with observations by others of decreased cuprate reactivity in THF; for
example, see: Bertz, S. H.; Eriksson, M.; Miao, G.; Snyder, J. P. J. Am.
Chem. Soc. 1996, 118, 10906. Analysis by TLC could distinguish between
cuprate protonolysis product (+)-51 and the corresponding oxidative
homocoupling product (identified by 1H NMR only); both were cleanly
and selectively formed upon quenching of an aliquot of the vinyl cuprate
with deoxygenated water and air, respectively. The vinyl lithium gave no
homocoupling product. The occurrence of oxidative homocoupling provided
a convenient TLC marker to indicate both formation and persistence of the
vinyl cuprate during the coupling reaction. The vinyl cuprate proved quite
stable in Et2O, persisting for at least 2 days at -5 °C.
(80) Solladie´ reported the anti reduction of â-arylsulfinyl ketones with
DIBAL, see: Solladie´, G.; Demailly, G.; Greck, C. Tetrahedron Lett. 1985,
26, 435. Although Solladie´ proposed an alternative mechanism, internal
delivery of hydride has been proposed to explain these results; see: Evans,
D. A.; Chapman, K. T.; Carreira, E. M. J. Am. Chem. Soc. 1988, 110, 3560.
(81) Acetonide migration product (+)-76 and the product of desilylation
of triol (+)-75 (isolated from larger scale deprotection experiments) could
each be readily redirected to the synthetic sequence.
Supporting Information Available: Spectroscopic and
analytical data and experimental procedures for all synthetic
intermediates (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
JA992134M
(82) (a) Rychnovsky, S. D.; Skalitzky, D. J. Tetrahedron Lett. 1990, 31,
945. (b) Evans, D. A.; Rieger, D. L.; Gage, J. R. Tetrahedron Lett. 1990,
31, 7099.