Total Synthesis of Apoptolidin: Construction of Enantiomerically Pure Fragments

Article abstract of DOI:10.1021/ja0304953

A general strategy for the total synthesis of the antitumor agent apoptolidin (1) is proposed, and the chemical synthesis of the defined key building blocks (4, 5, 6, 8, and 9) in their enantiomerically pure forms is described. The projected total synthes

Full text of DOI:10.1021/ja0304953

Published on Web 11/21/2003
Total Synthesis of Apoptolidin: Construction of
Enantiomerically Pure Fragments
K. C. Nicolaou,* Konstantina C. Fylaktakidou, Holger Monenschein, Yiwei Li,
Bernd Weyershausen, Helen J. Mitchell, Heng-xu Wei, Prasuna Guntupalli,
David Hepworth, and Kazuyuki Sugita
Contribution from the Department of Chemistry and The Skaggs Institute for Chemical Biology,
The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037, and
Department of Chemistry and Biochemistry, UniVersity of California, San Diego,
9500 Gilman DriVe, La Jolla, California 92093
Received August 18, 2003; E-mail: kcn@scripps.edu
Abstract: A general strategy for the total synthesis of the antitumor agent apoptolidin (1) is proposed, and
the chemical synthesis of the defined key building blocks (4, 5, 6, 8, and 9) in their enantiomerically pure
forms is described. The projected total synthesis calls for a dithiane coupling reaction to construct the
C20-C21 bond, a Stille coupling reaction to form the C11-C12 bond, and a Yamaguchi macrolactonization
to assemble the macrolide ring, as well as two glycosidation reactions to fuse the carbohydrate units onto
the molecule. First and second generation syntheses to the required fragments for apoptolidin (1) are
described.
Introduction
to modulate tumor cell resistance to drug-induced apoptosis.
One important goal of the new approaches is the identification
Apoptosis, the programmed cell death process, is a morpho-
logical phenomenon characterized by nuclear chromatin con-
densation and fragmentation, endoplasmic reticulum dilation,
and cell shrinkage. This cell suicide mechanism, tightly coupled
with cell replication, ensures a constant and controlled flux of
fresh cells and has important roles in a wide range of
physiological processes, including fetal development, aging,
tissue homeostasis, and the immune response. Deviations from
the natural pathways of apoptosis result in pathologies for which
adequate therapies or prevention are lacking; for example,
unscheduled apoptosis may cause neurodegeneration disorders
such as Alzheimer’s and Parkinson’s diseases, whereas the
failure of dividing cells to initiate apoptosis may contribute to
cancer.1
of novel agents that selectively target molecular abnormalities
in tumor cells, which may provide an opportunity to kill such
cells selectively in the presence of normal cells.
First reported by Hayakawa and co-workers,3 apoptolidin (1)
is a cytotoxic agent found during the course of a screening
program directed toward the discovery of novel apoptosis
inducers. Isolated from the cultivation broth of an actinomycete
identified as Nocardiopsis sp., apoptolidin was found to
selectively induce cell death via apoptosis in rat glia cells
transformed with adenovirus E1A and E1A/E1B19K oncogenes
with considerable potency. Using a series of molecular and cell-
based tools and techniques, Khosla and co-workers later
identified the mitochondrial F0F1-ATPase as the cellular target
of apoptolidin.4a These investigators further suggested that
apoptosis may be initiated in response to a shift in ATP
biosynthesis caused by this compound.4b The selectivity of
apoptolidin against transformed cells is striking. In a recent test
of 37 000 compounds against the National Cancer Institute’s
60 human cancer cell line panel, apoptolidin was found to be
among the top 0.1% of most selective cytotoxic agents.5
Interest in apoptosis has grown exponentially in the past
decade. Within the cancer research community in particular,
this explosion of interest has been fueled by the hope that a
greater mechanistic understanding of apoptotic cell death would
lead to improvements in cancer chemotherapy.2 The traditional
chemotherapeutic approaches suffer from two main problems,
namely, the indiscriminating toxicity to normal cells and drug
resistance. Benefiting from information regarding the in vivo
functions of individual protein components of the apoptotic
machinery, the emerging strategies have been focused on ways
The constitution and relative stereochemistry of apoptolidin
(1) were elucidated via a combination of NMR spectroscopic
and computer modeling techniques, and its absolute stereo-
chemistry was proposed on the basis of a correlation between
(1) For selected reviews on apoptosis, see: (a) Hickman, J. A. Eur. J. Cancer
1996, 32, 921-926. (b) Reed, J. C. Am. J. Pathol. 2000, 157, 1415-1426.
(c) Thornberry, N. A.; Lazebnik, Y. Science 1998, 281, 1312-1316. (d)
Reed, J. C. Cell 1997, 91, 559-562. (e) Green, D. R.; Reed, J. C. Science
1998, 281, 1309-1312.
(2) For selected reviews on apoptosis and cancer chemotherapy, see: (a) Makin,
G.; Dive, C. Trends Cell Biol. 2001, 11, S22-S26. (b) Reed, J. C. Drug
DiscoVery 2002, 1, 111-121.
(3) Kim, J. W.; Adachi, H.; Shin-Ya, K.; Hayakawa, Y.; Seto, H. J. Antibiot.
1997, 50, 628-630.
(4) (a) Salomon, A. R.; Voehringer, D. W.; Herzenberg, L. A.; Khosla, C.
Chem. Biol. 2001, 8, 71-80. (b) Salomon, A. R.; Voehringer, D. W.;
Herzenberg, L. A.; Khosla, C. Proc. Natl. Acad. Sci. U.S.A. 2000, 97,
14766-14771.
(5) Developmental Therapeutics Program NCI/NIH. http://dtp.nci.nih.gov.
9
10.1021/ja0304953 CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 15433-15442
15433

A R T I C L E S
Nicolaou et al.
easily modified to accommodate the chemical sensitivity and
complexity of the molecule; (2) a convergent route that would
allow convenient access to designed analogues in order to probe
structure-activity relationships. The chemical sensitivity of this
target molecule is primarily a reflection of the labile nature of
its macrocyclic ring with its all-trans polyene systems, the
susceptibility of the disaccharide domain toward acid hydrolysis,
and the migratory tendencies of the acyl chain of the macro-
lactone from C19 to the other hydroxyl groups under basic
conditions.9 In light of these potentially problematic issues, we
had to exercise special prudence in drafting a suitable road map
toward the target molecule (1) that would avoid both acidic and
basic media in the final stages. We also had to rely on silicon
protecting groups which could, in principle, be removed under
mild and neutral conditions at the end of the campaign.10
Figure 1. Molecular structure of apoptolidin (1).
its macrocyclic system and the 6-deoxy-glucose residue.6 As
shown in Figure 1, the structure of apoptolidin (1) contains a
20-membered macrolide ring equipped with a side chain
carrying a six-membered hemiketal system. The molecule also
carries a disaccharide moiety made of a D-oleandrose unit and
an L-olivomycose ring, as well as a novel 6-deoxy-glucose
residue. The molecular architecture of apoptolidin is distin-
guished by a total of 25 stereocenters and 5 geometrical sites
and features several unsaturated sites within its macrocyclic
scaffold, a densely substituted hemiketal ring, and two 2-deoxy-
pyranoses. The promising biological activity and daunting
molecular architecture of apoptolidin (1) have stimulated a
considerable body of synthetic work directed toward its total
synthesis. Thus, besides our own, investigations have been
reported by the Koert, Sulikowski, Toshima, Fuchs, and Loh
groups, among others.7 In addition, the Wender group has
engaged in semisynthetic studies directed toward the preparation
of analogues for biological screening purposes.8
The synthetic challenge posed by the apoptolidin structure
offers a unique opportunity for discovery and invention in the
area of new synthetic strategies and technologies. Furthermore,
such an endeavor, if successful, may provide validation to
apoptolidin’s structure and open entries into analogues for
biological investigations, leading to useful structure-activity
relationships (SAR). In this and the following article, we
describe a full account of our synthetic investigations in this
area, which culminated in the first total synthesis of apoptolidin
(1) and biological evaluation of a series of relevant analogues.
1. Retrosynthetic Analysis. The overall retrosynthetic analy-
sis of the target molecule, apoptolidin (1), is shown in Figure
2. Thus, given the high sensitivity of the DE disaccharide unit,
this domain was excised first, leading, upon appropriate
protecting group installment, to advanced intermediate 2 as a
potential precursor. Note that 2 is suitably protected to serve as
a substrate for glycosidation with glycosyl donor 4, in the
synthetic direction, a process that can also be applied to the
construction of analogues with varying sugar residues at C27.
The stereochemistry of the glycoside bond linking this disac-
charide group (DE) to the mainframe of the molecule is R and,
thereby, should pose no particular challenge as it is the one
anomerically favored.
While disaccharide 4 can easily be traced to readily available
carbohydrate units (10 and 11), the macrolide portion 2 still
needs considerable trimming before readily accessible fragments
can be recognized. From all possible (and there are many)
retrosynthetic disconnections for the rupture of the macrocycle,
we chose the Yamaguchi lactone-forming reaction11 for its
reliability and proven record in highly complex settings, a
scission that together with the indicated retro-glycosidation led
to trihydroxy acid 3 and carbohydrate unit 5. The sulfoxide
glycosyl donor 5 was chosen in order to ensure the desired
R-glycoside bond formation at low temperature and under mild
conditions as expected from a Kahne protocol.12 With regards
to seco acid 3, it was anticipated that the conformational rigidity
and constrain conferred to the backbone by its olefinic bonds,
Results and Discussion
(9) During the course of this synthetic endeavor, Khosla reported the stability
problem with apoptolidin. We subsequently reported, in our total synthesis
communications, the labile nature of apoptolidin and its tendency toward
isomerization. Wender and Sulikowski later isolated and identified the major
isomer as isoapoptolidin. For detailed information, see: (a) Salomon, A.
R.; Voehringer, D. W.; Herzenberg, L. A.; Khosla, C. Chem. Biol. 2001,
8, 71-80. (b) Nicolaou, K. C.; Li, Y.; Fylaktakidou, K. C.; Mitchell, H.
J.; Sugita, K. Angew. Chem., Int. Ed. 2001, 40, 3854-3857. (c) Wender,
P. A.; Gulledge, A. V.; Jankowski, O. D.; Seto, H. Org. Lett. 2002, 4,
3819-3822. (d) Pennington, J. D.; Williams, H. J.; Salomon, A. R.;
Sulikowski, G. A. Org. Lett. 2002, 4, 3823-3825. For information on its
sensitivity toward acidic conditions, see also: (e) Salomon, A. R.; Zhang,
Y.; Seto, H.; Khosla, C. Org. Lett. 2001, 3, 57-59. (b) Hayakawa, Y.;
Kim, J. W.; Adachi, H.; Shin-ya, K.; Fujita, K.; Seto, H. J. Am. Chem.
Soc. 1998, 120, 3524-3525.
In contemplating a synthetic strategy for the total synthesis
of apoptolidin (1), we kept in mind the following objectives:
(1) a stereocontrolled, yet flexible approach which could be
(6) Hayakawa, Y.; Kim, J. W.; Adachi, H.; Shin-ya, K.; Fujita, K.; Seto, H. J.
Am. Chem. Soc. 1998, 120, 3524-3525.
(7) (a) Schuppan, J.; Ziemer, B.; Koert, U. Tetrahedron Lett. 2000, 41, 621-
624. (b) Nicolaou, K. C.; Li, Y.; Weyershausen, B.; Wei, H.-X. Chem.
Commun. 2000, 4, 307-308. (c) Sulikowski, G. A.; Lee, W.-M.; Jin, B.;
Wu, B. Org. Lett. 2000, 2, 1439-1442. (d) Schuppan, J.; Wehlan, H.;
Keiper, S.; Koert, U. Angew. Chem., Int. Ed. 2001, 40, 2063-2066. (e)
Nicolaou, K. C.; Li, Y.; Fylaktakidou, K. C.; Mitchell, H. J.; Wei, H.-X.;
Weyershausen, B. Angew. Chem., Int. Ed. 2001, 40, 3849-3854. (f)
Nicolaou, K. C.; Li, Y.; Fylaktakidou, K. C.; Mitchell, H. J.; Sugita, K.
Angew. Chem., Int. Ed. 2001, 40, 3854-3857. (g) Toshima, K.; Arita, T.;
Kato, K.; Tanaka, D.; Matsumura, S. Tetrahedron Lett. 2001, 42, 8873-
8876. (h) Chen, Y.; Evarts, J. B., Jr.; Torres, E.; Fuchs, P. L. Org. Lett.
2002, 4, 3571-3574. (i) Chng, S.-S.; Xu, J.; Loh, T.-P. Tetrahedron Lett.
2003, 44, 4997-5000.
(10) For an excellent review on selective deprotection of silyl ethers, see: Nelson,
T. D.; Crouch, R. D. Synthesis 1996, 1031-1069.
(11) (a) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull.
Chem. Soc. Jpn. 1979, 52, 1989-1993. (b) Hikota, M.; Tone, H.; Horita,
K.; Yonemitsu, O. J. Org. Chem. 1990, 55, 7-9. (c) Hikota, M.; Sakurai,
Y.; Horita, K.; Yonemitsu, O. Tetrahedron Lett. 2003, 44, 6367-6370.
(d) Nicolaou, K. C.; Finlay, M. R. V.; Ninkovic, S.; Sarabia, F. Tetrahedron
1998, 54, 7127-7166.
(8) (a) Wender, P. A.; Jankowski, O. D.; Tabet, E. A.; Seto, H. Org. Lett.
2003, 5, 487-490. (b) Wender, P. A.; Jankowski, O. D.; Tabet, E. A.;
Seto, H. Org. Lett. 2003, 5, 2299-2302.
(12) Yan, L.; Kahne, D. J. Am. Chem. Soc. 1996, 118, 9239-9248.
9
15434 J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003

Total Synthesis of Apoptolidin
A R T I C L E S
Figure 2. Retrosynthetic analysis of apoptolidin (1).
and the severe steric hindrance around the C20 hydroxyl group,
would favor a C1-C19 ring closure over the other options.13, 14
Following the concept of convergence, seco acid 3 was
disconnected by a retro Stille coupling reaction15 to reveal
stannane methyl ester 6 and vinyl iodide 7 as potential
precursors. The latter fragment (7) was then dissected further,
employing retro dithiane coupling technology,16 to afford
acetylenic aldehyde 8 and dithiane 9 as starting points. The
remainder of this article describes the construction of these five
building blocks (4, 5, 6, 8, 9) and coupling of 10 and 11 to
produce 4 in preparation for the final assembly of apoptolidin
(1).
pinocampheylborane18 in the presence of boron trifluoride
etherate at -78 °C provided the acetylenic alcohol 13 in 82%
yield as a single stereoisomer. Protection of the free hydroxyl
group of 13 as its TBS ether (TBSOTf, 2,6-lutidine, 97% yield)
was followed by ozonolysis of the terminal olefin in the presence
of an indicator dye (Sudan red 7B)19 to afford aldehyde 15 via
silyl ether 14. The latter compound (15) was then employed in
a Wittig reaction20 with (carbethoxymethylene)triphenyl-
phosphorane in toluene at 100 °C, furnishing the desired trans-
olefin 16 in high yield and high stereoselectivity (90%, E:Z ratio
>9:1). Successive DIBAL-H reduction of 16 (90%) and TPAP-
NMO oxidation of the resulting primary alcohol (17) afforded
R,â-unsaturated aldehyde 18 which was converted, via a
Horner-Wadsworth-Emmons reaction21 with the appropriate
phosphonate, to diene ester 19 with high efficiency and in
excellent stereoselectivity (81% yield from 17, E:Z ratio >10:
1). After reduction (DIBAL-H, 89% yield) of the diene ester
19 to its alcohol counterpart (20) and oxidation of the latter
compound to the corresponding aldehyde with TPAP-NMO,
2. Construction of Vinylstannane Methyl Ester 6 (C1-
11 Fragment). Our devised synthetic sequence leading to 6 is
C
shown in Scheme 1. Thus, asymmetric crotylation of the known
propargylic aldehyde 1217 with Brown’s (Z)-crotyl-(+)-diiso-
(13) For studies on the regioselective Yamaguchi lactonization of apoptolidin-
like model compounds, see: (a) Nicolaou, K. C.; Li, Y.; Weyershausen,
B.; Wei, H.-X. Chem. Commun. 2000, 4, 307-308. (b) Schuppan, J.;
Wehlan, H.; Keiper, S.; Koert, U. Angew. Chem., Int. Ed. 2001, 40, 2063-
2066.
(14) For additional reports on the regioselective Yamaguchi lactonization and
its application in total synthesis, see: (a) Kageyama, M.; Tamura, T.; Nantz,
M. H.; Roberts, J. C.; Somfai, P.; Whritenour, D. C.; Masamune, S. J. Am.
Chem. Soc. 1990, 112, 7407-7408. (b) Evans, D. A.; Trotter, B. W.;
Coleman, P. J.; Coˆte´, B.; Dias, L. C.; Rajapakse, H. A.; Tyler, A. N.
Tetrahedron 1999, 55, 8671-8726.
(15) (a) Stille, J. K. Angew. Chem., Int. Ed. Engl. 1986, 25, 508-524. (b) Farina,
V.; Krishnamurphy, V.; Scott, W. J. Org. React. 1997, 50, 1-652. (c)
Betzer, J.-F.; Lallemand, J.-Y.; Pancrazi, A. Synthesis 1998, 522-534.
(16) Corey, E. J.; Seebach, D. Angew. Chem., Int. Ed. Engl. 1965, 4, 1075-
1077.
(17) Sharma, A.; Chattopadhyay, S. J. Org. Chem. 1998, 63, 6128-6131.
(18) Brown, H. C.;Bhat, K. S. J. Am. Chem. Soc. 1986, 108, 293-294.
(19) Veysoglu, T.; Mitscher, L. A.; Swayze, J. K. Synthesis 1980, 10, 807-
810.
(20) For a review of the Wittig reaction, see: Kelly, S. E. In ComprehensiVe
Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press: New
York, 1991; Vol. 1, pp 755-781.
(21) For reviews of the Horner-Wadsworth-Emmons reaction, see: (a)
Wadsworth, W. S., Jr. Org. React. 1977, 25, 73-254. (b) Maryanoff, B.
E.; Reitz, A. B. Chem. ReV. 1989, 89, 863-927.
9
J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003 15435

A R T I C L E S
Nicolaou et al.
Scheme 1. Synthesis of Vinylstannane 6 (Olefination Approach)a
Figure 3. Alternative retrosynthetic tracing of triene ester 22 to boronic
acid 23 and vinyl bromide 24.
terpart via a tin-iodide exchange reaction, underscoring the
potential sensitivity of this polyunsaturated system.
An alternative route to vinylstannane methyl ester 6 based
on a Suzuki coupling reaction23 (see Figure 3) was developed
as outlined in Scheme 2. This shorter, and more efficient,
sequence began with one-carbon homologation of the known
aldehyde 15 with the Ohira-Bestmann reagent24 to afford
diacetylene 25 (85% yield). After methylation (nBuLi-MeI,
98% yield) of the terminal acetylene within 25, the resulting
system (26) was chemoselectively hydroboronated with cat-
echolborane in the presence of catalytic amounts of 9-BBN25
to furnish vinyl boronate ester 27 in 85% yield. This highly
regio- and stereoselective reaction was best performed neat at
80 °C. Boronate ester 27 was then hydrolyzed to its boronic
acid counterpart (23). The other required coupling partner,
bromodiene ester 24, was conveniently prepared from the known
vinyl bromide 28 via a one-pot oxidation (MnO2)-Wittig
olefination26 followed by ester exchange (LiOH; CH2N2), a
sequence that proceeded through carboxylic acid 29. Finally,
the two partners, 23 and 24, were united through a Suzuki
coupling reaction brought about by the catalytic action of
bistriphenylphosphinedichloropalladium(II) in the presence of
sodium acetate in methanol at 70 °C to afford conjugated ester
21 (81% yield), which provides ready access to the desired
segment 6 as outlined in Scheme 2. Overall, this highly
convergent approach led to 6 from 15 in 35% overall yield and,
as such, it is the preferred route to this fragment.
a (a) (Z)-(+)-Crotyldiisopinocampheylborane (2.0 equiv), BF3‚Et2O (2.0
equiv), THF, -78 °C, 6 h; then NaBO3‚4H2O (6.7 equiv), THF:H2O (2:1),
25 °C, 12 h, 82%; (b) TBSOTf (1.3 equiv), 2,6-lutidine (2.0 equiv), CH2Cl2,
0 °C, 2 h, 97%; (c) O3, Sudan red 7B (0.02 equiv), CH2Cl2, -78 °C; then
PPh3 (2.1 equiv), -78 f 25 °C, 12 h; (d) Ph3PdC(CH3)CO2Et (5.0 equiv),
toluene, 100 °C, 12 h, E:Z > 9:1, 90% over two steps; (e) DIBAL-H (2.5
equiv), toluene, -78 °C, 2 h, 90%; (f) TPAP (0.05 equiv), NMO (4.0 equiv),
4Å MS, CH2Cl2, 25 °C, 30 min; (g) NaH (4.5 equiv), (EtO)2P(dO)CH(CH3)-
CO2Et (5.0 equiv), THF, 0 f 25 °C, 1 h, E:Z >10:1, 81% over two steps;
(h) DIBAL-H (2.5 equiv), toluene, -78 °C, 2 h, 89%; (i) TPAP (0.05 equiv),
NMO (6.0 equiv), 4Å MS, CH2Cl2, 25 °C, 30 min; (j) NaH (5.0 equiv),
(EtO)2P(dO)CH(CH3)CO2Me (6.0 equiv), THF, 0 f 25 °C, 1 h, E:Z >10:
1, 95% over two steps; (k) TBAF (4.0 equiv), THF, 0 f 25 °C, 1 h, 98%;
(l) nBu3SnH (4.0 equiv), Pd(PPh3)2Cl2 (0.05 equiv), THF, 0 °C, 30 min,
69%. THF ) tetrahydrofuran; TBAF ) tetra-n-butylammonium fluoride;
TPAP ) tetra-n-propylammonium perruthenate; NMO ) N-methylmor-
pholine N-oxide; DIBAL-H ) diisobutylaluminum hydride; TBSOTf ) tert-
butyldimethylsilyl trifluoromethanesulfonate; MS ) molecular sieves.
3. Construction of Acetylenic Aldehyde 8 (C12-C20 Frag-
ment). Two versions of this fragment were synthesized, 8a,
equipped with a p-methoxybenzyl group at C19, and 8, which
carried a 3,4-dimethoxybenzyl group at that position. The initial
studies were carried out with 8a, but as the project evolved we
adopted version 8, whose synthesis was shorter and more
efficient than the one developed for 8a. Here we describe both
constructions.
a second Horner-Wadsworth-Emmons homologation produced
the desired all-trans triene ester 21 (95% yield from 20).
The final stages of this sequence entailed exposure of 21 to
TBAF at room temperature, conditions which resulted in the
concomitant removal of both silicon groups (98% yield)
followed by palladium-catalyzed [Pd(PPh3)2Cl2] hydrostan-
nylation22 to afford, in 69% isolated yield, vinylstannane 6.
Although relatively stable at refrigerate temperature, stannane
6 decomposed upon attempted conversion to its iodide coun-
The first generation synthesis, for the preparation of building
block 8a, is summarized in Scheme 3. The known lactone 30
(23) For selected reviews on the Suzuki coupling reaction, see: (a) Miyaura,
N.; Suzuki, A. Chem. ReV. 1995, 95, 2457-2483. (b) Suzuki, A. J.
Organomet. Chem. 1999, 576, 147-168.
(24) (a) Ohira, S. Synth. Commun. 1989, 19, 561-564. (b) Mu¨ller, S.; Liepold,
B.; Roth, G. J.; Bestmann, H. J. Synlett 1996, 6, 521-522.
(25) For a discussion concerning the effects of 9-BBN on the hydroboration
reactions, see: Arase, A.; Hoshi, M.; Mijin, A.; Nishi, K. Synth. Commun.
1995, 25, 1957-1962.
(22) Zhang, H. X.; Guibe, F.; Balavoine, G. J. Org. Chem. 1990, 55, 1857-
1867.
(26) Wei, X.; Taylor, R. J. K. Tetrahedron Lett. 1998, 39, 3815-3818.
9
15436 J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003

Total Synthesis of Apoptolidin
A R T I C L E S
Scheme 2. Alternative Construction of Vinylstannane 6 (Suzuki
Scheme 3. Construction of Aldehyde 8aa
Coupling Strategy)a
a (a) (MeO)2P(dO)C(dN2)C(dO)Me (4.0 equiv), NaOMe (4.0 equiv),
THF, -78 f 25 °C, 1 h, 85%; (b) nBuLi (2.0 equiv), MeI (7.0 equiv),
THF, -78 f 25 °C, 2 h, 98%; (c) catecholborane (1.1 equiv), 9-BBN (0.10
equiv), neat, 80 °C, 15 h, 85%; (d) phosphate buffer (pH ) 7), THF, 25
°C, 2 h, 90%; (e) MnO2 (10.0 equiv), Ph3PdC(CH3)CO2Et (1.2 equiv),
CH2Cl2, 25 °C, 42 h, 91%; (f) LiOH (2.0 equiv), THF:H2O (2:1), 25 °C,
12 h, 93%; (g) MNNG (5.0 equiv), 40% KOH aq, Et2O, 0 °C, 30 min,
99%; (h) Pd(PPh3)2Cl2 (0.05 equiv), NaOAc (5.0 equiv), MeOH, 70 °C, 5
h, 81%. 9-BBN ) 9-borabicyclo[3.3.1]nonane; MNNG ) 1-methyl-3-nitro-
1-nitrosoguanidine.
a (a) Me2C(OMe)2 (41 equiv), TsOH (0.02 equiv), CH2Cl2, 25 °C, 3 h;
(b) 2-methoxypropene (5.8 equiv), CH2Cl2, 25 °C, 7 h; (c) morpholine (13.6
equiv), 65 °C, 15 h, 97% over three steps; (d) Me2SO4 (7.1 equiv),
BnEt3N+Cl- (0.03 equiv), 50% NaOH aqueous, CH2Cl2, 0 f 25 °C, 12 h,
86%; (e) PPTS (0.2 equiv), MeOH, 25 °C, 3 h, 90%; (f) SEMCl (2.0 equiv),
nBu4N+I- (1.1 equiv), DIPEA (4.0 equiv), CH2Cl2, 40 °C, 7 h, 99%; (g)
LiBHEt3 (3.0 equiv), THF, 0 °C, 3 h, 81%; (h) PMBCl (2.1 equiv), NaH
(2.0 equiv), nBu4N+I- (0.5 equiv), DMF, 25 °C, 24 h, 83%; (i) 60% AcOH
aq (excess), 25 °C, 15 h, 88%; (j) trimethyl orthoacetate (1.2 equiv), PPTS
(0.008 equiv), CH2Cl2, 25 °C, 45 min; then AcCl (1.2 equiv), CH2Cl2, 25
°C, 5 h; then K2CO3 (1.3 equiv), MeOH, 25 °C, 12 h, 88%; (k)
allenylmagnesium bromide (3.0 equiv), Et2O, -78 °C, 1 h, 94%; (l) 1.5%
HCl in MeOH (excess), MeOH, 25 °C, 30 min, 95%; (m) DDQ (1.2 equiv),
4Å MS, CH2Cl2, 0 f 25 °C, 2 h, 78%; (n) TBSOTf (2.0 equiv), 2,6-lutidine
(2.5 equiv), CH2Cl2, 0 f 25 °C, 1 h, 98%; (o) nBuLi (2.2 equiv), MeI (7.0
equiv), THF, -78 f 25 °C, 2 h, 99%; (p) DIBAL-H (3.0 equiv), CH2Cl2,
-78 °C, 30 min, 72%; (q) SO3‚py (5.0 equiv), Et3N (6.0 equiv), DMSO:
CH2Cl2 (2:1), 0 f 25 °C, 30 min, 95%. SEM ) 2-(trimethylsilyl)ethoxy-
methyl; PMB ) p-methoxybenzyl; DIPEA ) diisopropylethylamine; MIP
) 1-methyl-1-methoxyethyl; PPTS ) pyridinium p-toluenesulfonate; Tf )
trifluromethane sulfonyl; DDQ ) 2,3-dichloro-5,6-dicyano-1,4-benzo-
quinone; DMSO ) methyl sulfoxide; py ) pyridine.
(originating from L-ascorbic acid)27 was chosen as a suitable
starting material for this sequence on the basis of the recognition
that it contains all three stereogenic centers embedded in
fragment 8a in their correct configuration. Exposure of 30 to
dimethoxy acetone in the presence of p-toluenesulfonic acid
gave acetonide 31, whose further reaction with 2-methoxypro-
pene under neutral conditions furnished fully protected deriva-
tive 32. Reaction of 32 with morpholine at 65 °C led to amide
33 via a nucleophilic attack which opened the lactone ring (97%
yield, these three steps were carried out in a one-pot process).
Methylation of the free hydroxyl group in 33 with dimethyl
sulfate under phase-transfer conditions (BnEt3N+Cl-) furnished
compound 34 (86% yield), whose protecting group at C19 was
removed under mildly acidic conditions (PPTS in MeOH, 35,
90% yield) and replaced with a SEM group (SEMCl,28 99%
yield), leading to intermediate 36. Reduction of the latter
compound (36) with SuperHydride (LiBHEt3,29 81% yield) at
(27) (a) Andrews, G. C.; Crawford, T. C.; Bacon, B. E. J. Org. Chem. 1981,
46, 2976-2977. (b) Bock, K.; Lundt, I.; Petersen, C. Acta Chem. Scand.
B 1981, 35, 155-157.
(28) Lipshutz, B. H.; Pegram, J. J. Tetrahedron Lett. 1980, 21, 3343-3346.
9
J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003 15437

A R T I C L E S
Nicolaou et al.
Scheme 4. Construction of Aldehyde 8a
0 °C, followed by protection of the resulting primary alcohol
(37) as a p-methoxybenzyl ether, led to 38. The last protection
was necessary in order to allow epoxide formation30 on the other
end of the molecule, a process that began with the acid-induced
cleavage of the acetonide ring (AcOH, 88% yield) to afford
diol 39, and which was completed with the treatment of the
latter substrate with trimethylorthoacetate and pyridinium p-
toluenesulfonate followed by sequential addition of acetyl
chloride and potassium carbonate (40, 88% yield).
Nucleophilic opening of epoxide 40 with allenylmagnesium
bromide31 proved problematic at first but was finally ac-
complished in reproducible 90-95% yields by the use of freshly
prepared Grignard reagent in ether (0.1 M), affording hydroxyl
acetylene 41. Initial attempts to carry out this reaction in THF
or THF/ether solvent systems led to mixtures of the desired
product (41) and the corresponding bromohydrin derived from
nucleophilic attack of bromide at the terminus of epoxide 40.
With compound 41 in hand, the road to the targeted
intermediate 8a was straightforward. Thus, removal of the SEM
group (HCl-MeOH, 95% yield) gave diol 42, whose exposure
to 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) resulted
in the formation of p-methoxybenzylidene acetal 43 in 78% yield
(mixture of stereoisomers, ca. 1:1). Protection of the remaining
hydroxyl group within 43 as a TBS ether (98% yield) was
followed by methylation of the terminal acetylene (44), leading
to compound 45 (99% yield). The DIBAL-induced opening of
the benzylidene32 in 45 proceeded regioselectively, and in 72%
yield, to afford a primary alcohol, which was oxidized (SO3‚py
and DMSO33) to afford aldehyde 8a in 95% yield.
The second generation and more efficient route to a C12-
20 aldehyde fragment started with commercially available (+)-
C
a (a) PMBCl (2.1 equiv), NaH (2.0 equiv), nBu4N+I- (0.5 equiv), DMF,
0 f 25 °C, 4 h, 71%; (b) allenylmagnesium bromide (2.5 equiv), Et2O,
-78 °C, 1 h, 90%; (c) TBSOTf (2.0 equiv), 2,6-lutidine (2.5 equiv), CH2Cl2,
0 f 25 °C, 2 h, 97%; (d) nBuLi (2.0 equiv), MeI (7.0 equiv), THF, -78
f 25 °C, 2 h, 95%; (e) DDQ (2.0 equiv), CH2Cl2:H2O (18:1), 0 f 25 °C,
3 h, 97%; (f) SO3‚py (5.0 equiv), Et3N (5.0 equiv), DMSO:CH2Cl2 (2:1),
0 °C, 3 h, 96%; (g) (+)-allyldiisopinocampheylborane (1.6 equiv), Et2O,
-100 °C, 2 h; then 30% H2O2 aqueous/NaOH aqueous (3.0 M), 25 °C, 15
h, dr ca. 10:1, 95%; (h) MeOTf (2.2 equiv), 2,6-di-tert-butyl-4-methyl-
pyridine (3.0 equiv), CH2Cl2, 40 °C, 24 h, 85%; (i) K3Fe(CN)6 (3.0 equiv),
K2CO3 (3.0 equiv), (DHQ)2-PYR (0.01 equiv), OsO4 (0.005 equiv), tBuOH:
H2O (1:1), 0 °C, 12 h, dr ca. 6:1, 85%; (j) DMBA (3.0 equiv), CSA (0.05
equiv), toluene, 110 °C, 8 h; 99%; (k) DIBAL-H (3.0 equiv), CH2Cl2, -78
°C, 30 min, 70%; (l) SO3‚py (5.0 equiv), Et3N (5.0 equiv), DMSO:CH2Cl2
(2:1), 0 f 25 °C, 1.5 h, 90%. DMF ) N,N-dimethylformamide; MS )
molecular sieves; (DHQ)2-PYR ) 2,5-diphenyl-4,6-bis(9-O-dihydroquinyl)
pyrimidine, DMBA ) 3,4-dimethoxybenzaldehyde, DMB ) 3,4-dimethoxy-
benzyl.
glycidol (47) and targeted the 3,4-dimethoxy aldehyde 8 as
shown in Scheme 4. Thus, protection of 47 as its p-methoxy-
benzyl ether led to 48, whose reaction with freshly prepared
allenylmagnesium bromide in ether afforded acetylenic alcohol
49 in 90% yield. Silylation of the latter compound followed by
methylation of the terminal acetylene (nBuLi-MeI, 95% yield)
afforded acetylene 51. The DDQ-facilitated deprotection of the
primary position of 51 proceeded smoothly (97% yield) to afford
alcohol 52, which was oxidized with SO3‚py-DMSO, furnish-
ing aldehyde 53 (96% yield). The latter compound was reacted
with (+)-allyldiisopinocampheylborane34 to afford, after oxida-
tive workup, homoallylic alcohol 54 in high yield (95%) and
diastereoselectivity (ca. 10:1). Upon much experimentation, the
next desired intermediate, methyl ether 55, was generated by
treatment of 54 with methyl triflate in the presence of excess
2,6-di-tert-butyl-4-methylpyridine (85% yield). The next step
involved asymmetric dihydroxylation of the terminal olefin
within 55, a process that was brought about by applying the
Sharpless procedure (AD-mix-R,35 85% yield, ca. 6:1 ratio of
diastereoisomers) and led to diol 56. Formation of the corre-
sponding 3,4-dimethoxybenzylidene36 derivative 57 (DMBA,
CSA, 99% yield, mixture of isomers) followed by regioselective
DIBAL-induced ring opening of the latter compound furnished
the desired primary alcohol 58 (70% yield). Oxidation of 58
with SO3‚py and DMSO in CH2Cl2 then generated the targeted
acetylenic aldehyde 8 in 90% yield. The high efficiency of this
route coupled with the ease of removal of the DMB group as
compared to the PMB group made this building block (8) the
substrate of choice for subsequent operations.
(29) Brown, H. C.; Kim, S. C.; Krishnamurthy, S. J. Org. Chem. 1980, 45,
1-12.
(30) Kolb, H. C.; Sharpless, K. B. Tetrahedron 1992, 48, 10515-10530.
(31) Brandsma, L.; Verkruijisse, H. In PreparatiVe Polar Organometallic
Chemistry I; Springer: Berlin, 1987; p 63.
(32) (a) Maruoka, K.; Yamamoto, H. Tetrahedron 1988, 44, 5001-5032. (b)
Takano, S.; Akiyama, M.; Sato. S.; Ogasawara, K. Chem. Lett. 1983, 1593-
1596.
(33) Parikh, J. R.; Doering, W. v. E. J. Am. Chem. Soc. 1967, 89, 5505-5507.
(34) Racherla, U. S.; Brown, H. C. J. Org. Chem. 1991, 56, 401-404.
(35) For a review, see: Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B.
Chem. ReV. 1994, 94, 2483-2547. (b) Crispino, G. A.; Jeong, K. S.; Kolb,
H. C.; Wang, Z. M.; Xu, D.; Sharpless, K. B. J. Org. Chem. 1993, 58,
3785-3786.
4. Construction of Dithiane 9. The construction of the
required building block 9 was conducted as outlined in Schemes
(36) Wanner, M. J.; Willard, N. P.; Koomen, G.-J.; Pandit, U. K. Tetrahedron
1987, 43, 2549-2556.
9
15438 J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003

Total Synthesis of Apoptolidin
A R T I C L E S
Scheme 6. Construction of Dithiane Building Block 9a
Scheme 5. Unsuccessful Attempt To Construct the C21-C28
Fragment. A Surprising Reversal of Stereoselectivitya
a (a) (R)-3-(1-Oxopropyl)-4-(phenylmethyl)-2-oxazolidinone (A, 1.15
equiv), nBu2BOTf (1.4 equiv), Et3N (1.60 equiv), CH2Cl2, -78 f 0 °C, 2
h; then 30% H2O2 aqueous, 0 °C, 2 h, dr ca. 10:1, 83%; (b) HNMe(OMe)‚HCl
(2.5 equiv), AlMe3 (2.5 equiv), CH2Cl2, -20 f 25 °C, 12 h, 90%; (c)
TMSOTf (2.0 equiv), 2,6-lutidine (4.0 equiv), CH2Cl2, -30 °C, 1 h; (d)
DIBAL-H (3.0 equiv), CH2Cl2, -78 °C, 2 h, 89% two steps; (e)
HS(CH2)3SH (2.0 equiv), BF3‚Et2O (1.0 equiv), CH2Cl2, -30 °C, 15 min,
78%; (f) TBSOTf (4.0 equiv), 2,6-lutidine (8.0 equiv), CH2Cl2, 0 f 25
°C, 12 h, 97%; (g) TBAF (2.0 equiv), THF, 0 f 25 °C, 1 h; (h)
PhI(OCOCF3)2 (1.5 equiv), MeCN:phosphate buffer (pH ) 7, 4:1), 0 °C,
12 min; (i) Ac2O (5.0 equiv), Et3N (10 equiv), 4-DMAP (catalytic), CH2Cl2,
25 °C, 5 h, 70% over three steps; (j) TBAF (2.0 equiv), THF, 0 f 25 °C,
1 h; (k) 2-methoxypropene (5.0 equiv), TsOH (0.10 equiv), THF, 0 °C, 2
h, 85% over two steps.
a (a) 1,3-Dithiane (1.5 equiv), nBuLi (1.5 equiv), THF, -78 f 25 °C,
30 min, 91%; (b) PMBCl (1.3 equiv), NaH (1.3 equiv), DMF, 25 °C, 4 h,
99%; (c) MeI (excess), K2CO3 (1.0 equiv), MeCN:H2O (6:1), 45 °C, 5 h,
92%; (d) (Z)-(+)-crotyldiisopinocampheylborane (4.0 equiv), BF3‚Et2O (4.0
equiv), THF, -78 °C, 6 h; then NaBO3‚4H2O (15 equiv), THF:H2O (1:1),
25 °C, 12 h, 98%; (e) TBSOTf (1.5 equiv), 2,6-lutidine (2.0 equiv), CH2Cl2,
0 f 25 °C, 2 h, 97%; (f) OsO4 (0.03 equiv), NMO (2.0 equiv), tBuOH:
H2O:THF (10:1:2), 25 °C,12 h; then NaIO4 (5.0 equiv), phosphate buffer
(pH ) 7), 25 °C, 2 h, 94%; (g) DDQ (4.0 equiv), CH2Cl2:H2O (18:1), 0 f
25 °C, 1.5 h, 97%; (h) 2-methoxypropene (5.0 equiv), TsOH (0.10 equiv),
THF, 0 f 25 °C, 2 h, 75%; (i) (Z)-(-)-crotyldiisopinocampheylborane (4.0
equiv), BF3‚Et2O (4.0 equiv), THF, -78 °C, 6 h; then NaBO3‚4H2O (15
equiv), THF:H2O (1:1), 25 °C, 12 h, dr ca. 9:1; or (Z)-(+)-crotyldiisopi-
nocampheylborane (4.0 equiv), BF3‚Et2O (4.0 equiv), THF, -78 °C, 6 h;
then NaBO3‚4H2O (15 equiv), THF:H2O (1:1), 25 °C, 12 h, dr ca. 8:1; (j)
TBAF (1.5 equiv), THF, 0 f 25 °C, 1 h, 85% over two steps for (Z)-(+)-
crotylborane; 75% over two steps for (Z)-(-)-crotylborane; (k) 2,2-
dimethoxypropane (3.0 equiv), CSA (0.10 equiv), THF, 0 f 25 °C, 1.5 h,
93%. TsOH ) p-toluenesulfonic acid.
aldehyde 64 in 94% yield. Before proceeding further, however,
it was necessary to confirm the C25-C27 anti stereochemical
relationship within structures 63 and 64. To this end, the
acetonide 65 was prepared from 63 by removal of the p-
methoxybenzyl group and ring formation with 2-methoxypro-
pene in the presence of p-toluenesulfonic acid (75% yield). The
13C NMR spectrum of 65 (150 MHz, CDCl3) exhibited chemical
shifts at δ 100.5, 24.6, and 24.4 ppm corresponding to the three
acetonide carbons in accord with literature values for the desired
1,3-anti relationship.38
The C23-C25 syn stereochemical outcome of the crotylborane
addition to aldehyde 64 was surprising in that the “matched”
nature of the aldehyde and borane components was expected
to assist the chiral Ipc units in dictating an anticipated anti
stereochemistry between C23 and C25 substituents.39,40 Interest-
ingly, however, both the (+) and the (-) enantiomers of the
5 and 6. Thus, starting with the commercially available (-)-
methylglycidol (59, Scheme 5), addition of lithiodithiane
resulted in the formation of hydroxy dithiane 60, whose
protection as a p-methoxybenzyl ether (99% yield) led to
compound 61. Unmasking the aldehyde functionality within the
latter intermediate (MeI-K2CO3,37 92% yield) then furnished
aldehyde 62, whose reaction with Brown’s (Z)-crotyl-(+)-
diisocampheylborane furnished hydroxy olefin 63 in 98% yield
1
and as a single stereoisomer (by H NMR spectroscopy). In
preparation for a second crotylation, the hydroxy olefin 63 was
protected as a TBS ether (97% yield) and its double bond was
cleaved with osmium tetroxide and sodium periodate to afford
(38) (a) Rychnovsky, S. D.; Skalitzky, D. J. Tetrahedron Lett. 1990, 31, 945-
948. (b) Rychnovsky, S. D.; Rogers, B.; Yang, G. J. Org. Chem. 1993, 58,
3511-3515. (c) Evans, D. A.; Rieger, D. L.; Gage, J. R. Tetrahedron Lett.
1990, 31, 7099-7100.
(37) Takano, S.; Hatakeyama, S.; Ogasawara, K. J. Chem. Soc., Chem. Commun.
1977, 68.
9
J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003 15439

A R T I C L E S
Nicolaou et al.
Scheme 7. Construction of Carbohydrate Unit A (5)a
(Z)-crotyldiisopinocampheylborane gave a similar result (ca. 8:1
ratio of isomers in favor of 67). The unexpected C23-C25 syn
stereochemical relationship within 67 was detected again through
acetonide formation (treatment of TBAF, followed by 2-meth-
oxypropene) and NMR spectroscopic analysis (13C signals of
the acetonide carbons in 68 at δ 98.9, 30.0, 19.7 ppm).
In search for a solution to this stereochemical problem to
form the C22-C23 bond, and after considerable experimentation,
we were pleased to observe that the Evans oxazolidinone
technology41 provided us with an avenue to the desired anti
C23-C24 diastereoisomer (Scheme 6). Thus, reaction of aldehyde
64 with the boron enolate of Evans’ propionyl chiral auxiliary
A (Scheme 6) furnished the desired aldol product (69) in
excellent yield (83%) and diastereoselectivity (ca. 10:1).
Generation of the Weinreb amide42 from imide 69 (MeNHOMe‚
HCl-AlMe3) then led to intermediate 70, which, upon reaction
with TMSOTf and 2,6-lutidine, furnished derivative 71. DIBAL-
induced reduction of the Weinreb amide functionality within
71 led to aldehyde 72 (89% overall yield from 70), whose
exposure to 1,3-propanedithiol in the presence of boron tri-
fluoride etherate furnished the targeted dithiane 73 (78% yield,
from which the labile TMS group had fallen off). The selection
of the trimethylsilyl protecting group for the C23 hydroxy group
of intermediate 71 was far from arbitrary. Rather, it was a
consequence of earlier experiments in which a TBS group at
the C23 position caused epimerization problems (>25% epimer-
ization) at C22 during the dithiane formation step (72 to 73)
due to its steric bulk. A subsequent experiment with an aldehyde
of type 72 with a free hydroxyl group at C23 revealed no
epimerization, underscoring the importance of the nature of the
protecting group at this position on the ability to epimerize C22
during dithiane formation. Required for the DIBAL reduction
step that generated the aldehyde moiety from the Weinreb
amide, the trimethylsilyl group proved ideal in that it was
promptly cleaved by the Lewis acid during the dithiane-forming
reaction.
a (a) MeI (5.0 equiv), NaH (2.5 equiv), DMF, 25 °C, 1.5 h, 92%; (b)
ethylene glycol (1.2 equiv), TsOH (0.1 equiv), MeOH, 25 °C, 12 h, 91%;
(c) TBSOTf (1.1 equiv), 2,6-lutidine (1.5 equiv), CH2Cl2, 0 f 25 °C, 3 h,
95%; (d) (COCl)2 (2.0 equiv), DMSO (2.5 equiv), Et3N (4.0 equiv), CH2Cl2,
-78 °C, 4.5 h, 100%; (e) NaBH4 (1.2 equiv), MeOH, 0 °C, 5 min, 70%;
(f) TBSOTf (2.0 equiv), 2,6-lutidine (2.5 equiv), CH2Cl2, 0 f 25 °C, 1 h,
100%; (g) mCPBA (1.0 equiv), CH2Cl2, -78 °C, 5 h, 67%. DMF ) N,N-
dimethylformamide; mCPBA ) 3-chloroperoxybenzoic acid.
desired, fully protected dithiane building block 9 was obtained
from hydroxy compound 73 by treatment with TBSOTf and
2,6-lutidine in 97% yield.
5. Construction of the Carbohydrate Building Blocks. The
special challenges posed by apoptolidin’s carbohydrate frag-
ments included proper functionalization of the rings as well as
provisions for efficient and stereocontrolled glycoside bond
formation. Of particular interest with regard to the latter aspect
was the construction of the 2-deoxy-â-glycoside bonds associ-
ated with carbohydrate units E and D. As we shall see below,
we called upon our previously developed methodology43 involv-
ing 1,2-phenylsulfeno migration to solve this problem, and we
utilized the Kahne glycosyl sulfoxide technology to ensure the
proper attachment of carbohydrate unit A onto the aglycon.
Scheme 7 summarizes the construction of sulfoxide 5
representing carbohydrate moiety A. Thus, the readily available
L-rhamnose-derived thioglycoside 1044 was methylated (NaH-
MeI, 92% yield) to afford methyl ether 76, whose acetonide
group was removed by exposure to ethylene glycol and
p-toluenesulfonic acid, leading to diol 77 (91% yield). Silylation
of 77 [TBSOTf (1.5 equiv), 2,6-lutidine, 95% yield] followed
by Swern oxidation [(COCl)2, DMSO, Et3N, 100% yield]
furnished ketone TBS ether 79 via monosilylated compound
78. The desired inversion at C-2 was realized when the latter
compound (78) was subjected to NaBH4 reduction in MeOH at
0 °C, leading to the 6-hydroxy compound 80 (L-glucose
derivative) exclusively and in 70% yield. The hydride attack
To confirm the desired 1,3-anti stereochemical relationship
within the skeleton of these intermediates, acetonide 75 (Scheme
6) was prepared from 73 by the action of TBAF to liberate the
C
25 hydroxyl group, followed by treatment of the resulting diol
with 2-methoxypropene and p-toluenesulfonic acid. Furthermore,
dithiane 73 was converted to carbohydrate system 74 by the
action of (i) TBAF, (ii) PhI(OCOCF3)2, and (iii) Ac2O (70%
yield over three steps). Indeed, while the 13C signals for 75 (δ
100.5, 25.1, 24.0 ppm) supported the C23-C25 anti stereochem-
1
ical relationship, the H NMR spectroscopic analysis (NOE
studies) of 74 revealed the syn-anti-syn relationships of the
substituents at C22-C23, C23-C24, and C24-C25. Finally, the
(39) For selected reviews on double asymmetric induction, see: (a) Roush, W.
R. In ComprehensiVe Organic Synthesis; Trost, B. M., Fleming, I., Eds.;
Pergamon Press: New York, 1991; Vol. 1, pp 1-53. (b) Mengel, A.; Reiser,
O. Chem. ReV. 1999, 99, 1191-1223. For a discussion on gauche-gauche
pentane interactions in the transition state of the crotylboration reaction,
see: Roush, W. R. J. Org. Chem. 1991, 56, 4151-4157 and references
therein.
(40) (a) Cram, D. J.; Wilson, D. R. J. Am. Chem. Soc. 1963, 85, 1245-1249.
(b) Cram, D. J.; Elhafez, F. A. A. J. Am. Chem. Soc. 1952, 74, 5828-
5835. (c) Che´rest, M.; Felkin, H.; Prudent, N. Tetrahedron Lett. 1968, 9,
2199-2204. (d) Anh, N. T.; Eisenstein, O. Tetrahedron Lett. 1976, 17,
155-158.
(41) (a) Evans, D. A. Aldrichimica Acta 1982, 15, 23-32. (b) Gage, J. R.; Evans,
D. A. Organic Syntheses; Wiley & Sons: New York, 1993; Collect. Vol.
VIII, pp 339-343.
(42) (a) Basha, A.; Lipton, M.; Weinreb, S. M. Tetrahedron Lett. 1977, 18,
4171-4172. (b) Levin, J. I.; Turos, E.; Weinreb, S. M. Synth. Commun.
1982, 12, 989-993.
(43) (a) Nicolaou, K. C.; Ladduwahetty, T.; Randall, J. L.; Chucholowski, A.
J. Am. Chem. Soc. 1986, 108, 2466-2467. (b) Nicolaou, K. C.; Fylak-
takidou, K. C.; Mitchell, H. J.; van Delft, F. L.; Rodr´ıguez, R. M.; Conley,
S. R.; Jin, Z. Chem. Eur. J. 2000, 6, 3095-3115.
9
15440 J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003

Total Synthesis of Apoptolidin
A R T I C L E S
Scheme 8. Construction of Carbohydrate Unit D (88)a
Figure 4. Chelation-controlled addition of MeMgBr (A) and sterically
controlled addition of MeLi (B) to carbohydrate ketone 87.
MeMgBr in ether at -78 °C, presumably acting via a chelation-
controlled mechanism (see A, Figure 4), could deliver the
desired tertiary alcohol 88 in 91% yield (dr ca. 7.5:1). It is
noteworthy that addition of MeLi to ketone 87 in ether resulted
in the exclusive formation of the opposite diastereoisomer (3-
epi-88, not shown), presumably via the alternative, sterically
controlled arrangement B (1,3-diaxial interaction between
incoming nucleophile and PhS group, Figure 4).
The stereoselective synthesis of the required DE disaccharide
fluoride donor 4 proceeded from tertiary alcohol 88 via glycosyl
acceptor D (93) and glycosyl donor E (95) as shown in Scheme
9. Thus, 88 was silylated (excess TBSOTf-2,6-lutidine, 93%
yield) to afford bis-TBS ether 89 which was debenzylated with
DDQ, leading to 90 in 97% yield. Exposure of 2-hydroxyphen-
ylthioglycoside 90 to DAST resulted in the expected 1,2-
migration with inversion of configuration at C-2, furnishing
2-phenylthioglycosyl fluoride 91 in quantitative yield (R:â
anomers, ca. 1:10 ratio). A benzyl group was then chosen to
guard the anomeric position (BnOH, SnCl2, ether), leading to
92 (R-anomer, 90% yield). Exposure of the latter compound
(92) to excess TBAF then generated diol 93 ready for coupling
with glycosyl donor 95.
The desired substrate 95 was produced from phenylthiogly-
coside 11 in two steps, namely, selective C-3 methylation (nBu2-
SnO, MeI, 83% yield) to furnish 94 and treatment with DAST
to induce 1,2-migration within the latter compound, a process
that was accompanied by inversion of configuration of C-2 as
expected. With both coupling partners (93 and 95) in hand, we
were poised for the crucial glycosidation event in which we
expected the 2-phenylthio group within ring E (95) to play the
dominant role in directing the formation of the desired â-gly-
coside bond. Thus, mixing 93 and 95 in the presence of SnCl2
in ether at room temperature for 12 h resulted in the formation
of the desired disaccharide 96, in 45% yield, together with its
3-O-regioisomer (26% yield). Glycosidation attempts with C-3-
protected derivatives (TMS, TBS, OAc) failed, presumably due
to steric shielding exerted by the protecting group over the
neighboring C-4 position. Arrival at the targeted building block
4 required (i) installment of a TES group onto the tertiary alcohol
(TESOTf-2,6-lutidine, 84% yield) to afford 97, (ii) exposure
of 97 to Raney Ni (to remove the two thiophenyl groups and
the benzyl moiety, 92% yield), and (iii) treatment of the resulting
lactol (98) with DAST (4, 100% yield, R:â anomers, ca. 3:1
ratio).
a (a) TBSOTf (1.5 equiv), 2,6-lutidine (2.0 equiv), CH2Cl2, -78 °C, 0.5
h, 95%; (b) BCl3‚SMe2 (1.2 equiv), CH2Cl2, 0 °C, 30 min, 92%; (c)
nBu2SnO (1.3 equiv), toluene, reflux, 6 h; then allyl bromide (1.5 equiv),
CeF (1.2 equiv), DMF, 70 °C, 12 h, regioisomers ca. 3:1, 90%; (d) PMBCl
(1.5 equiv), nBu4N+I- (0.5 equiv), NaH (1.6 equiv), DMF, 0 °C, 1.5 h,
83%; (e) RhCl(PPh3)3 (0.05 equiv), DABCO (1.5 equiv), MeOH:H2O 10:
1, reflux, 2 h; then OsO4 (0.05 equiv), NMO (1.2 equiv), acetone:H2O 10:
1, 25 °C, 12 h, 92%; (f) DMP (1.5 equiv), NaHCO3 (2.0 equiv), CH2Cl2,
25 °C, 1 h, 96%; (g) MeMgBr (2.0 equiv), Et2O, -78 °C, 10 min, dr ca.
7.5:1, 91%; DABCO ) 1,4-diazabicyclo[2,2,2]octane. DMP ) Dess-Martin
periodinane.
apparently occurs from the opposite side of the bulky anomeric
substituents occupying an axial position. The last two steps of
the sequence involved TBS protection (TBSOTf-2,6-lutidine,
100% yield) and mCPBA oxidation (67% yield) of the sulfur
residue, leading to the targeted building block 5 via intermediate
81.
The same thioglycoside (10) was employed as a starting
material for the synthesis of building block 88, a suitably
functionalized intermediate to become carbohydrate residue D
within apoptolidin’s structure. Thus, and as shown in Scheme
8, substrate 10 was silylated (TBSOTf-2,6-lutidine, 95% yield)
to afford TBS ether 82, from which the acetonide protecting
group was removed through the action of BCl3‚SMe2, leading
to diol 83 in 92% yield. The selective conversion of diol 83 to
its 2-protected derivative 86 required three steps, namely,
selective C-3 allylation as accomplished by the tin acetal
technology45 (nBu2SnO-allyl bromide, 90% yield), introduction
of the PMB group onto the C-2 hydroxy group (PMBCl, NaH,
nBu4NI, 83% yield), and cleavage of the allyl ether [RhCl-
(PPh3)3, DABCO; OsO4-NMO, 92% yield). Ketone 87 was
then generated from 86 by treatment of the latter compound
with DMP46 (96% yield). The crucial installment of the methyl
group at C-3 was then attempted, leading to the discovery that
(44) Bajza, I.; Lipta´k, A. Carbohydr. Res. 1990, 205, 435-439.
(45) For an excellent review on the tin acetal technology and its applications in
carbohydrate synthesis, see: Grindley, T. B. AdV. Carbohydr. Chem.
Biochem. 1998, 53, 17-142.
(46) (a) Dess, D. R.; Martin, J. C. J. Org. Chem. 1983, 48, 4155-4156. (b)
Meyer, S. D.; Schreiber, S. L. J. Org. Chem. 1994, 59, 7549-7552.
Conclusion
Within this article, a strategy for the total synthesis of
apoptolidin (1) is laid out, and the first phase of its implementa-
9
J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003 15441

A R T I C L E S
Nicolaou et al.
Scheme 9. Construction of Disaccharide Domain DE (4)a
a (a) TBSOTf (2.0 equiv), 2,6-lutidine (3.0 equiv), CH2Cl2, 0 f 40 °C, 24 h, 93%; (b) DDQ (1.5 equiv), CH2Cl2:H2O (5:1) 0 °C, 1.5 h, 97%; (c) DAST
(2.0 equiv), CH2Cl2, 0 °C, 20 min, 100%, R:â ca. 1:10; (d) BnOH (10.0 equiv), SnCl2 (1.0 equiv), Et2O, 0 f 25 °C, 5 h, 90%; (e) TBAF (2.5 equiv), THF,
25 °C, 6 h, 98%; (f) nBu2SnO (1.0 equiv), toluene, reflux, 6 h; then MeI (1.5 equiv), CeF (1.0 equiv), DMF, 55 °C, 1 h, 83%; (g) DAST (3.0 equiv), CH2Cl2,
0 °C, 0.5 h, 100%, R:â ca. 1:7; (h) 93 (0.73 equiv), SnCl2 (3.0 equiv), Et2O, 0 f 25 °C, 12 h, 45% of 4-O glycoside relative to 95, 26% of 3-O glycoside;
(i) TESOTf (1.5 equiv), 2,6-lutidine (2.0 equiv), CH2Cl2, 0 f 25 °C, 6 h, 84%; (j) Raney Ni (ca. 4.0 equiv), EtOH, 55 °C, 5 h, 92%; (k) DAST (1.5 equiv),
CH2Cl2, 0 °C, 15 min, 100%, R:â ca. 15:1. Bn ) benzyl; DAST ) (diethylamino)sulfur trifluoride.
500, AMX-500, or AMX-400 instruments and calibrated using residual
undeuterated solvent as an internal reference. The following abbrevia-
tions were used to explain the multiplicities: s ) singlet, d ) doublet,
t ) triplet, q ) quartet, m ) multiplet, quin ) quintuplet, sext ) sextet,
sep ) septet, b ) broad. IR spectra were recorded on a Perkin-Elmer
1600 series FT-IR spectrometer. Electrospray ionization mass spec-
trometry (ESIMS) experiments were performed on an API 100 Perkin-
Elmer SCIEX single quadrupole mass spectrometer at 4000 V emitter
voltage. High-resolution mass spectra (HRMS) were recorded on a VG
ZAB-ZSE mass spectrometer under fast atom bombardment (FAB)
conditions with NBA as the matrix or using MALDI.
tion is described. Specifically, stereocontrolled routes to the
defined key building blocks (4, 5, 6, 8, 9) have been developed
and optimized; second generation schemes allowing for their
gram scale synthesis have been chartered. With these intermedi-
ates readily available, their assembly and elaboration to apop-
tolidin (1) and analogues thereof became possible. The following
article47 describes the investigations that led to the coupling of
the synthesized building blocks and the completion of the total
synthesis of apoptolidin (1).
Experimental Section
General Procedures. All reactions were carried out under an argon
atmosphere with dry solvents under anhydrous conditions, unless
otherwise noted. Dry tetrahydrofuran (THF), toluene, diethyl ether
(ether), and methylene chloride (CH2Cl2) were obtained by passing
commercially available pre-dried, oxygen-free formulations through
activated alumina columns. Yields refer to chromatographically and
spectroscopically (1H NMR) homogeneous materials, unless otherwise
stated. Reagents were purchased at the highest commercial quality and
used without further purification, unless otherwise stated. Reactions
were monitored by thin-layer chromatography (TLC) carried out on
0.25 mm E. Merck silica gel plates (60F-254) using UV light as
visualizing agent and an ethanolic solution of phosphomolybdic acid
and cerium sulfate and heat as developing agents. E. Merck silica gel
(60, particle size 0.040-0.063 mm) was used for flash column
chromatography. Preparative thin-layer chromatography (PTLC) separa-
tions were carried out on 0.25 or 0.50 mm E. Merck silica gel plates
(60F-254). NMR spectra were recorded on Bruker DRX-600, DRX-
Acknowledgment. We thank Drs. D. H. Huang, G. Siuzdak,
and R. Chadha for NMR spectroscopic, mass spectrometric, and
X-ray crystallographic assistance, respectively. This work was
financially supported by the National Institutes of Health
(U.S.A.), The Skaggs Institute for Chemical Biology, American
Bioscience, Inc., predoctoral fellowships from Boehringer
Ingelheim, Eli Lilly, and The Skaggs Institute for Research (all
to Y.L.), postdoctoral fellowships from the George Hewitt
Foundation (to K.C.F.) and the Alexander von Humboldt
Foundation (Feodor Lynen Fellowships) (to B.W. and H.M.),
and grants from Abbott, Amgen, ArrayBiopharma, Boehringer
Ingelheim, DuPont, Glaxo, Hoffmann-La Roche, Merck, Pfizer,
and Schering Plough.
Supporting Information Available: Experimental procedures
and compound characterization. This material is available free
of charge via the Internet at http://pubs.acs.org.
(47) Nicolaou, K. C.; Li, Y.; Sugita, K.; Monenschein, H.; Guntupalli, P.;
Mitchell, H. J.; Fylaktakidou, K. C.; Vourloumis, D.; Giannakakou, P.;
O’Brate, A. J. Am. Chem. Soc. 2003, 125, 15443-15454.
JA0304953
9
15442 J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003