Spirastrellolide studies. Synthesis of the C(1)-C(25) southern hemispheres of spirastrellolides a and B, exploiting anion relay chemistry

Article abstract of DOI:10.1016/j.tet.2010.01.082

Construction of the C(1)-C(25) southern fragments of both spirastrellolide A and B are described. Highlights of the syntheses include effective use of the three component anion relay chemistry (ARC) tactic recently introduced by our laboratory, a stereose

Full text of DOI:10.1016/j.tet.2010.01.082

Tetrahedron 66 (2010) 6597–6605
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Spirastrellolide studies. Synthesis of the C(1)–C(25) southern hemispheres of
spirastrellolides A and B, exploiting anion relay chemistry
*
Amos B. Smith, III , Helmars Smits, Dae-Shik Kim
Department of Chemistry, Monell Chemical Senses Center and Laboratory for Research on the Structure of Matter, University of Pennsylvania, 231 South 34th St.,
Philadelphia, PA 19104, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Construction of the C(1)–C(25) southern fragments of both spirastrellolide A and B are described.
Highlights of the syntheses include effective use of the three component anion relay chemistry (ARC)
tactic recently introduced by our laboratory, a stereoselective spirocyclization via concomitant Ferrier
reaction to elaborate the BC spiroketal and use of two dithiane unions to install the A ring as well as
C(22)–C(25) fragment. The synthesis proceeded with longest linear sequences of 33 and 32 steps,
respectively for spirastrellolide A and spirastrellolide B.
Received 24 November 2009
Received in revised form
11 January 2010
Accepted 21 January 2010
Available online 1 February 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Spirastrellolide
Anion relay chemistry
Brook rearrangement
Stereoselective Spirocyclization
Petasis-Ferrier reaction
Dithiane
Multicomponent reaction
Spiroketal
Natural product
Polyketide
Total synthesis
1. Introduction
turn leads to premature cell entry into mitosis and thereby mitotic
arrest.1a,b Similar biological effects have been observed with Ser/
Spirastrellolides A and B (1 and 2) comprise two architecturally
unique polyketide natural products, isolated by Andersen and co-
workers in 2003 and 2006, respectively from Spirastrella coccinea,
a marine sponge endemic to the Caribbean (Scheme 1).1 In 2004
extensive NMR studies permitted assignment of relative configura-
tions of the three major segments of 1 [cf. C(3)–C(7), C(9)–C(24), and
C(27)–C(38)].1b The stereochemical relationship between the seg-
ments of both spirastrellolides A and B, except for the configuration
at C(46), as well as the assignment of absolute configuration of the
macrolide core followed in 2007 based on an X-ray analysis of
a crystalline derivative obtained from a degradation product of spi-
rastrellolide B (2).1c Assignment of the final sterochemical issue, the
C(46) stereogenic center was achieved in 2007 upon isolation of the
cleaved side-chain fragment derived from a related congener
(Spirastrellolide D).1d
Thr phosphatase inhibitors such as fostriecin, okadaic acid, and
calyculin A.1b,c The biological properties of spirastrellolide B (2)
have not been reported.
The architectural complexity and potential biomedical signifi-
cance, in conjunction with the scarcity of natural materials have
attracted the attention of the chemical community. To date, one
elegant total synthesis of spirastrellolide A methyl ester2 by the
Paterson group, as well as a number of synthetic studies have been
reported.3 Herein we disclose a detailed account of our studies
leading to construction of the C(1)–C(25) southern hemisphere of
both spirastrellolides A and B.4
From the retrosynthetic perspective, disconnection of both A
and B (1 and 2) at three strategic locations, the macrocyclic lactone,
the C(25)–C(26)
s-bond, and the C(40)–C(41) trans p-bond
(Scheme 1), suggests three potential advanced synthetic targets:
side chain 3, DEF bis-spiro ketal 4, and ABC ring fragment 5. Further
disconnection of 5 reveals A-ring dithiane 6, BC spiroketal 7 and
known dithiane (ꢀ)-8. For construction of 7 we envisioned
a 1,3-anti-reduction of the C(13) ketone in diketone 9, followed by
acid-catalyzed spiroketalization. Construction of 9 in turn would
Spirastrellolide A (1) displays selective inhibition of protein
phosphatase PP2A (IC50¼1 nM for PP2A and 50 nM for PP1), that in
* Corresponding author. Tel.: þ1 215 898 4860; fax: þ1 215 898 5129; e-mail
address: smithab@sas.upenn.edu (A.B. Smith III).
0040-4020/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.01.082

6598
A.B. Smith III et al. / Tetrahedron 66 (2010) 6597–6605
Construction of the requisite dithiane (þ)-12 entailed selective
1,2-diol protection of known triol (ꢀ)-136 as an acetonide, the latter
readily prepared in four steps from commercially available (ꢀ)-tri-
OH
41
HO 47
O
O
F
46
E
O-acetyl-
D-glucal, followed by methylation of the remaining hy-
38
OMe
R
40
1
droxy substituent.
D
35
O
O
28
O
27
O
25
3
A
OH
S
S
S
O
O
HO
S
7
acetone, HCl
71%
HO
23
O
OH
11
Spirastrellolide A (1):
R = Cl, X=X = HC=CH
9
21
OH
13
O
17
O
OH
(–)-13
OH
(+)-14
HO
OMe
B
C
Spirastrellolide B (2):
X
X
R = H, X=X = H2C-CH2
S
O
S
i. NaH, Me2SO4
O
ii. MeOH
99%
BnO
OMe
(+)-12
O
F
OTBS
SO2Ar
E
OMe
40
MeO
47
O
D
41
HO
O
Scheme 2.
27
O
3
4
For the ARC tactic, construction of bifunctional linchpin 11 be-
gan with C-silylation of known dithiane (þ)-157 employing TESCl,
followed by acidic hydrolysis of the acetonide to provide diol (ꢀ)-17
(Scheme 3). Numerous methods were explored for the direct con-
version of the diol to the epoxide. Surprisingly, the one pot Fraser-
Reid protocol8 lead to predominant Brook rearrangement, while
use of two step protocols relying on activation of the primary hy-
droxyl group resulted in extensive decomposition, presumably due
to formation of sulfonium ion 20.9 We therefore explored activation
of the secondary hydroxyl instead of the primary.
OPiv
25
23
3
A
O
O
7
O
11
O
9
21
O
O
13
O
17
OMe
B
C
15
5
OTBS
O
OBPS
OMe
3
S
S
21
O
O
13
A
25
S
O
O
S
S
O
2M HCl
THF
S
S
t-BuLi, HMPA
7
17
+
B
C
O
O
THF, −78 °C
TES
15
S
then TESCl
62%
7
S
(−)-8
(–)-16
9
(+)-15
80%
6
S
OH
O
OTESO
O
S
OH
S
O
11
O
BnO
OH
(–)-17
15
21
13
15
17
TES
TES
OMe
11
9
1) AcCl, Py
TES OH
2) MsCl, Py
3) K2CO3, MeOH
S
O
OLG
S
S
S
S
O
S
O
O
BnO
(−)-10
(64% over 3 steps)
19
TES
OMe
11
Scheme 1.
12
S
S
O
TES
15
TES
S
call upon Type II Anion Relay Chemistry (ARC),5 an effective mul-
ticomponent union tactic recently introduced by our laboratory,
that would employ commercially available epoxide (ꢀ)-10, bi-
functional linchpin 11, and dithiane 12.
S+
OH
(–)-18
20
Scheme 3.
2. Results and discussion
Chemoselective acetylation of (ꢀ)-17, followed in turn by
mesylation of the secondary hydroxyl and ring closure employing
potassium carbonate furnished epoxide (ꢀ)-18 in 64% overall yield.
That (ꢀ)-18 is epimeric at C(15) relative to the proposed retron (11)
held little consequence since the C(15) hydroxyl would be elimi-
nated later in the sequence to install the C(15)–C(16) olefin.
2.1. Construction of the C(10)–C(22) spiroketal fragments of
spirastrellolides A and B (1 and 2)
We began the synthesis by focusing on the C(10)–C(22) spi-
rocyclic fragments of spirastrellolides
A and B (Scheme 2).

A.B. Smith III et al. / Tetrahedron 66 (2010) 6597–6605
6599
Unfortunately epoxide (ꢀ)-18 did not display the expected re-
activity with dithiane (þ)-12 required in the ARC coupling protocol.
Even with extensive experimentation, varying the base, solvents
and temperature, the yield of the coupled product (ꢀ)-21 was
modest at best (ca. 20–30%) (Scheme 4). The corresponding linch-
pin (ꢀ)-11, the C(15) diastereomer of (ꢀ)-18, prepared in analogous
fashion, also proved unreactive. In silico conformational analysis
employing MacroModelÒ suggested that the methyl substituent at
C(14) causes the epoxide methylene to become buried within the
dithiane ring.
1) n-BuLi/KO-t-Bu
S
O
S
OTBS
S
S
O
THF, –78 °C
S
O
S
O
2) (–)-25, THF,
–78 °C
OMe
(+)-12
OMe
26
O
BnO
(–)-10
OH
OTBS
S
S
O
S
Hg(ClO4)2
CaCO3, 0 °C
THF/H2O
75%
S
O
THF
–78 to 0 °C
OBn
OMe
(+)-27
77%
S
O
S
OTES
S
S
O
1) n-BuLi, THF
2)
S
S
O
O
OH
O
OTBS
O
O
11
O
OMe
OMe
15
21
13
17
S
S
(+)-12
O
(–)-21
28%
OBn
OMe
TES
THF, rt
(+)-24
Scheme 6.
(–)-18
Scheme 4.
epoxide (ꢀ)-25 to discriminate between 26 and the anion of (þ)-12,
thus avoiding formation of significant amounts of the alkylation
product derived from 26 and (ꢀ)-25. Pleasingly, removal of both
dithiane groups with Hg(ClO4)210 then proceeded without difficulty
to furnish 1,4-bis-ketone (þ)-24 in 75% yield.
Having experienced difficulties with both linchpins in imple-
menting the initially proposed ARC tactic, we were forced to revise
the synthetic plan. The simpler linchpin (ꢀ)-25, developed pre-
viously in our laboratory and shown to be a competent coupling
partner in Anion Relay Chemistry would be employed, making
necessary a late stage introduction of the C(14) methyl group
(Scheme 5).
Selective anti-reduction of the
b-hydroxy ketone in (þ)-24, pos-
sessing the C(17) ketone, exploiting the Gribble–Evans protocol11
was immediately followed by hydrolysis of the acetonide under
acidic conditions. This reaction, as anticipated, proceeded with
concomitant spiroketalization to furnish the desired spiroketal
(þ)-28, as well as spiroketal (þ)-29 and a small amount of hemiketal
(þ)-30 (Scheme 7). The yields were 57, 24, and 3%, respectively.
Presumably spiroketal (þ)-29 arises via dehydration of an in-
termediate hemiketal, followed in turn by Ferrier reaction12 of the
resulting 1,2-glycal leading to spirocyclization (vide infra).
BnO
BnO
BnO
BnO
OBPS
OBPS
21
O
O
13
21
O
O
13
17
OMe
B
C
17
OMe
B
C
15
TBSO
23
22
1) Me4NBH(OAc)3
O
OTBS
OH
O
O
AcOH/CH3CN, −35 °C
O
13
17
2) 3.5% aq HClO4
CH2Cl2
OH
O
OTBS
15
O
O
OBn
OMe
11
(+)-24
O
BnO
21
13
17
OMe
24
HO
HO
BnO
OH
OH
OMe
(+)-28
13
13
BnO
O O
17
O O
17
OMe
57%
TBSO
24%
(+)-29
S
S
O
S
O
S
O
BnO
O
TBS
HO
BnO
O
OMe
(+)-12
13
(−)-25
O
(−)-10
O OH
17
Scheme 5.
OMe
(+)-30
3%
Linchpin (ꢀ)-25, was readily prepared in a single step from epi-
chlorohydrin and 2-TBS-1,3-dithiane.5a With all partners in hand for
the union, the revised three component ARC tactic was executed.
Optimal conditions proved to be addition of linchpin (ꢀ)-25 to the
anion derived from (þ)-12, generated by treatment with the
Schlosser base (Scheme 6). The resultant intermediate, after Brook
rearrangement was then alkylated with epoxide (ꢀ)-10 to furnish
the expected three component product (þ)-27 in 77% yield.
Scheme 7.
Initial studies to install the C(14) methyl group in a stereo-
controlled fashion were performed on a slightly modified substrate,
namely (þ)-31 possessing the C(20) tert-butyldiphenylsilyl (BPS)
ether instead of methyl ether (Scheme 8). This spiroketal was
prepared in large quantity during our early studies on anion relay
chemistry employing an analogous route to that used for prepa-
ration of (þ)-28 (see Supplementary data). Protection of the pri-
mary hydroxyl group as the BPS ether, followed in turn by
benzylation of the remaining secondary hydroxyl, chemoselective
Pleasingly this reaction could be conveniently run on a 5 g scale.
It is important to note that under these conditions the Brook
rearrangement occurs readily at ꢀ78 ꢁC, with monoalkylation as
the major pathway. Presumably the low temperature permits

6600
A.B. Smith III et al. / Tetrahedron 66 (2010) 6597–6605
removal of the TBS group, and Dess–Martin oxidation of the
resulting hydroxyl furnished ketone (þ)-34. Generation of the
enolate with LHMDS at ꢀ78 ꢁC, followed by treatment with MeI
then led to a mixture of two methylation products (6:1 by NMR).
Extensive NMR studies revealed that the desired regioisomer 35
was the minor product.
At this stage we turned to possible optimization of the Ferrier
reaction observed earlier during the spirocyclization (Scheme 7).
Treatment of the reduction product of (þ)-24 first with p-TsOH
followed by aqueous HClO4 pleasingly led to an increase in the yield
of Ferrier reaction product (þ)-29 (Scheme 10). Protection of the
primary hydroxyl as a BPS ether and the secondary hydroxyl as
a benzyl ether then provided olefin (þ)-40. The overall yield of this
sequence (24/40) was 45%.
HO
BnO
HO
BnO
OH
OBPS
OBPS
O
O
O
O
BPSCl, imid.
CH2Cl2
1) Me4NBH(OAc)3
OBPS
OH
O
OTBS
O
AcOH/CH3CN, −35 °
C
O
11
O
87%
2) p-TsOH, acetone
3) 3.5% aq HClO4
15
21
13
17
TBSO
TBSO
(+)-31
(+)-32
OBn
OMe
(+)-24
CH2Cl2
BnO
OBPS
OBPS
1) NaH, BnBr,
DMF
HO
BnO
HO
BnO
OH
OH
OMe
BnO
O
O
DMP, NaHCO3
CH2Cl2
O
O
O
O
2) AcOH,
THF/H2O
OMe
HO
(75% over 2 steps)
84%
(+)-33
TBSO
(+)-29
(+)-28
62%
13%
BnO
BnO
BnO
BnO
OBPS
OMe
1) BPSCl, imid.
OBPS
OBPS
O
O
LiHMDS, THF
MeI, HMPA
DMAP, CH2Cl2, 84%
2) NaH, BnBr, TBAI
O
O
DMF, 87%
29% (69% borsm)
(+)-40
O
O
(+)-34
35:36 = 1:6
Scheme 10.
BnO
BnO
BnO
Turning next to introduction of the C(14) equatorial methyl
group, selective allylic oxidation at C(14) in (þ)-40 with selenium
dioxide14 followed by nucleophilic substitution was now envi-
sioned. The success of this scenario would critically depend upon
the regio and stereoselective allylic oxidation to furnish the axial
alcohol as the major product via an ene reaction mechanism. Given
that the top face of the olefin is sterically more accessible, with the
OBPS
OBPS
OBPS
OBPS
BnO
O
O
O
O
O
35
36
Scheme 8.
axial proton perfectly aligned with the
p-system according to
To enhance access to the desired C(14) isomer 35, kinetic for-
mation of the TMS enol ether was explored, anticipating that the
adjacent ketal might provide the required chemoselective differen-
tiation (Scheme 9). Unfortunately even the best conditions furnished
a mixture of regioisomers (1.5:1) favoring the desired isomer. We
could however hydrolyze the undesired isomer selectively to the
open chain 1,3-diketone (þ)-39.
molecular model studies, such a reaction sequence appeared fea-
sible. Indeed, treatment of olefin (þ)-40 with SeO2 for 1.5 h at
120 ꢁC in the presence of pyridine, employing a microwave reactor,
provided alcohol (ꢀ)-41 with 5:1 diastereoselectivity, along with
some recovered starting material (Scheme 11). Longer reaction
times or higher reaction temperatures resulted in better axial se-
lectivity, albeit with a decrease in yield. From the perspective of
material advancement the undesired equatorial diastereomer could
be converted to the desired isomer exploiting Mitsunobu inversion.
BnO
BnO
TMSOTf,
2,6-lut.
OBPS
OMe
OBPS
OMe
BnO
O
O
BnO
O
O
CH2Cl2
0 °C to rt
99%
O
TMSO
BnO
BnO
(+)-37
1.5:1
BnO
BnO
OBPS
OMe
OBPS
OMe
BnO
BnO
O
O
SeO2, Py,
O
O
OBPS
OMe
HO
MeLi, MeI
dioxane,
O
O
PPTS, CH2Cl2
THF/H2O
MW 120 °C
(−)-41
(+)-40
55%, dr = 5:1
TMSO
(+)-38
BnO
57%
OBPS
OMe
(PhSO2)2CH2
LiDBB, t-BuOH
THF, −78 °C
BnO
O
O
PhO2S
OH
O
OH
OH
DIAD, PPh3, PhH
PhO2S
65% (over 2 steps)
OBPS
(+)-42
OMe
(+)-39
BnO
HO
OBn
32%
O
OBPS
OMe
OBPS
NaH, Trisimid
HO
O
O
O
O
Scheme 9.
OMe
THF
92%
(−)-7
Diketone (þ)-39 could then be recycled to ketone (þ)-37 by
treatment with strong acid. Unfortunately all attempts to alkylate
the anion derived from (þ)-38 lead to an intractable mixture of
methylation products. We also explored cyclopropanation of (þ)-38
according to the method developed by Charette and co-workers,13
but again an inseparable mixture (3:1) of cyclopropane products
resulted.
(+)-43
Scheme 11.
Turning to the introduction of the C(14) methyl, substituent,
Mitsunobu alkylation15 with bis(phenylsulfonyl)-methane pro-
ceeded smoothly, leading to bis-sulfone (þ)-42, which was then

A.B. Smith III et al. / Tetrahedron 66 (2010) 6597–6605
6601
fully reduced with LiDBB at ꢀ78 ꢁC, employing t-BuOH as the
proton source to furnish diol (þ)-43.15 Final application of the
Fraser-Reid protocol8 on (þ)-43 completed construction of (ꢀ)-7,
the C(10)–C(22) fragment of spirastrellolide A (1).
methoxy benzyl protecting group instead of the TBS moiety are not
competent nucelophiles, due to competitive deprotonation at the
benzylic methylene group. Oxidative hydrolysis of the dithiane
group in (ꢀ)-47 employing Hg(ClO4)2,10 followed by reduction of
the resulting
b-hydroxy ketone according to the Gribble–Evans
protocol11 next furnished (ꢀ)-48 in 71% yield in conjunction with
21% of the 1,3-syn diol (dr¼3.4:1). Chemoselective removal of the
TBS group with camphor sulfonic acid, followed by protection of
the primary alcohol as a pivalate ester then led to diol (ꢀ)-49 as
a crystalline solid. Both the connectivity and relative stereochem-
istry of (ꢀ)-49 were established by single crystal X-ray analysis.4
To introduce the final C(23)–C(25) four-carbon side chain, we
called upon the Honda precedent (Scheme 14),17 fully anticipating
known dithiane (ꢀ)-8 to deliver the desired 1,3-syn stereogenicity
at C(22) as observed with dihydro cinnamaldehyde (Scheme 14).
2.2. Completion of the C(1)–C(25) southern hemisphere of
spirastrellolide A (1)
Construction of the requisite dithiane (ꢀ)-6 comprising the C(1)–
C(9) side chain began with known pyran (ꢀ)-443h,16 (Scheme 12).
Reduction of the methyl ester with LiAlH4, followed by protection of
the resulting primary alcohol as the TBS ether led to (ꢀ)-45.
Oxidative cleavage of the double bond to the corresponding
aldehyde employing ozonolysis, followed by dithiane formation
promoted by MgBr2 then furnished dithiane (ꢀ)-6.
OTBS
S
S
S
S
CO2Me
n-BuLi, THF
92%
dr = 9.4:1
CHO
Ph
Ph
1) LiAlH4, Et2O, 96%
O3
O
O
O
OH
CH2Cl2; Ph3P
93%
2) TBSCl, Imid.
CH2Cl2, 97%
(−)-8
(−)-44
(−)-45
Scheme 14.
OTBS
S
OTBS
To this end, the requisite aldehyde (þ)-50 was readily generated
via a three step sequence involving protection of the free diol as the
acetonide, removal of BPS group, and Dess–Martin oxidation
(Scheme 15). Treatment of the latter with the lithium anion derived
from (ꢀ)-817 led to a mixture of adducts (6.5:1) in moderate yield.
Surprisingly, the minor product proved to be the desired isomer, as
established by the Kakisawa modification of the Mosher ester
analysis of both diastereomers (see Supplementary data).18
1,3-propane dithiol
MgBr2, Et2O, 71%
O
O
S
(−)-46
(−)-6
Scheme 12.
With dithiane (ꢀ)-6 in hand, treatment with the Schlosser base
at ꢀ78 ꢁC followed by addition of epoxide (ꢀ)-7 furnished coupled
product (ꢀ)-47 in excellent yield (Scheme 13). Noteworthy here,
dithianes analogous to (ꢀ)-6, but possessing either a benzyl or p-
OPiv
O
OTBS
OTBS
OBPS
HO
1) 2,2-dimethoxypropane
p-TsOH, acetone
1) n-BuLi, KOt-Bu
THF, −78 °C
O O
O
2) TBAF, THF
3) DMP, NaHCO3, CH2Cl2
O
OBPS
HO
OMe
HO
(−)-49
2) (−)-7, THF
−78 to 0 °C
O O
S
75% (over 3 steps)
S
S
OPiv
O
S
OMe
87%
(−)-6
(−)-47
OTBS
O
O
S
(−)-8
S
O
H
O
1) Hg(ClO4)2, CaCO3
THF-H2O, 98%
O O
n-BuLi, −78 °C
56%
dr = 6.3:1
OBPS
HO
OMe
2) Me4NBH(OAc)3
O O
AcOH/CH3CN, -35 oC
71% dr = 3.4:1
(+)-50
OMe
HO
(−)-48
OPiv
OPiv
O
1) CSA, MeOH,
CH2Cl2, 98%
O
O
S
OBPS
HO
O
S
OH
H
OMe
2) PivCl, DMAP, CH2Cl2
87%
O O
O O
OMe
HO
(−)-49
(+)-
51
Scheme 13.
Scheme 15.

6602
A.B. Smith III et al. / Tetrahedron 66 (2010) 6597–6605
Fortunately the desired diastereomer could be obtained via
2.3. Completion of the C(1)–C(25) southern hemisphere of
spirastrellolide B (2)
a three step oxidation–reduction sequence (Scheme 16), involving
reduction of the ketone derived from alcohol (þ)-51 with DIBAL-H to
furnish exclusively the required stereogenicity at C(22), presumably
via chelation control. This process could be carried out on the
diastereomeric mixture derived from (þ)-50.
Having achieved a workable, albeit not highly efficient synthesis
of the C(1)–C(25) southern hemisphere of spirastrellolide A (1), we
turned our attention to the corresponding southern hemisphere of
spirastrellolide B (2), lacking a C(15)–C(16) olefin in ring B (54). Our
synthetic plan was envisioned to diverge from the spirastrellolide A
route (vide supra), not only by reducing the B ring olefin at the most
opportune stage, but also by reordering installation of the right and
left hand side chains (Scheme 18), thus avoiding excessive protecting
group manipulations. In addition, we could explore the previously
problematic introduction of the C(23)–C(26) fragment with correct
stereogenicity of C(22) at an earlier stage in the synthetic sequence,
as well as make overall improvements where appropriate.
OPiv
O
S
O
S
DMP, NaHCO3
CH2Cl2
OH
O O
H
OMe
O
96%
(+)-51
OPiv
O
OH
25
1) DIBALH, CH2Cl2
3
O
S
A
−78 °C
O
O
23
S
7
O
O
11
2) PivCl, DMAP
CH2Cl2
50% (78% borsm)
O O
H
O
9
21
O
O
13
OMe
O
O
17
OMe
B
C
15
(+)-52
54
OPiv
O
O
H
Me
O
S
HO
BnO
BnO
O
S
H-
O
S
O
25
23
Al
OH
OTBS
(β-face attack)
21
O
O
13
O
O O
H
17
OMe
B
C
S
OMe
O
O
15
O
A
O
56
21
(+)-53
O O
17
13
OMe
B
C
+
25
Scheme 16.
S
15
S
S
S
55
23
(−)-8
Next, removal of the dithiane in (þ)-53 employing the Corey
conditions (NCS, AgNO3)19 furnished an
a-hydroxy ketone, which
Scheme 18.
was subjected to anti-reduction with Zn(BH4)2,20 employing
1,2-chelation control (Scheme 17). Acetonide protection of the
resulting 1,2-diol completed the synthesis of the C(1)–C(25) south-
ern hemisphere fragment of spirastrellolide A (1).
The synthesis of 54 began with the previously prepared
bis-sulfone (þ)-42, which upon reduction with LiDBB proceeded
chemoselectively to furnish alkene (þ)-57 (Scheme 19). Impor-
tantly, careful control of the amount of LiDBB permitted selective
reduction of the bis-sulfone with only minor amounts of concom-
itant reduction of the benzyl ethers. Removal of the BPS protecting
group, followed by Parikh–Doering oxidation21 then proceeded
OPiv
BnO
BnO
BnO
OBPS
OMe
OBPS
OMe
O
S
1) NCS, AgNO3, 2,6-lut.
MeCN/H2O, 56%
BnO
PhO2S
PhO2S
O O
LiDBB
O O
O
S
THF, t-BuOH
−78 °C
OH
2) Zn(BH4)2, Et2O
O O
H
3) Me2C(OMe)2, p-TsOH
(+)-57
(+)-42
70%
OMe
O
64% (over 2 steps)
BnO
BnO
(+)-53
OH
SO3·Py, DMSO
i-Pr2NEt, CH2Cl2
99%.
TBAF, THF
96%
O O
OMe
OPiv
(+)-58
O
S
S
S
S
O
O
O O
BnO
BnO
BnO
O
O
OH
H
OMe
H
H
(-)-8
O O
BnO
O O
OMe
O
n-BuLi, THF
OMe
−78 °C
(+)-5
(+)-59
(+)-60
49%, dr = 6.3:1
Scheme 17.
Scheme 19.

A.B. Smith III et al. / Tetrahedron 66 (2010) 6597–6605
6603
smoothly to afford aldehyde (þ)-59. With ample quantities of
(þ)-59 in hand, addition of the lithium anion derived from dithiane
(ꢀ)-8 gave alcohol (þ)-60 in modest yield, as a diastereomeric
Undaunted, we also explored varying the structure of the nu-
cleophilic partner employed in the dithiane union process with both
trityl and BPS protected congeners (ꢀ)-67 and (ꢀ)-68 (Scheme 22).
mixture (6.3:1) albeit again favoring the undesired
b
epimer at
However, in both cases, the larger substituents b to the dithiane
C(22). A small amount (ca.10–15%) of aldehyde (þ)-59 could also be
anion led to an increase in the ratio of the undesired
b
epimer.
recovered.
Numerous other coupling protocols were explored including use
of the Schlosser base, polar additives (HMPA or TMEDA), as well as
chelation controlled conditions employing the cerium anion de-
rived from (ꢀ)-8, the latter tactic employed with great success
during our spongistatin total synthesis.22 However, in no case was
a significant improvement in diastereoselectivity observed. Modi-
fying the dithiane coupling partner by replacing the olefin moiety
in (ꢀ)-8 with a hydroxyl protected as a TBS ether23 was also ex-
plored. In this case, almost exclusive formation of the undesired
OR
S
S
S
S
BnO
BnO
BnO
BnO
O
OR
OH
O O
O O
H
OMe
OMe
n-BuLi, THF
(+)-65
(+)-69
(−)-67, R = Tr
45%, dr = 2.7:1
(−)-68, R = BPS
b
epimer resulted (Scheme 20).
70%, dr = 1.6:1
Scheme 22.
S
S
OTBS
To correct the undesired stereogenicity at C(22) in (þ)-66, an
oxidation–reduction sequence, identical to that employed in our
first generation approach, was put into play (Scheme 23) to pro-
duce (þ)-71 with the desired stereochemistry at C(22). Oxidative
removal of dithiane moiety, again employing the Corey protocol19
OTBS
S
BnO
BnO
BnO
BnO
S
(−)-61
O
OH
H
OMe
O O
O O
n-BuLi, THF
−78 °C
OMe
furnished
a-hydroxy ketone (þ)-72, which was subjected to che-
50%, dr = 9:1
(+)-62
(+)-59
lation controlled 1,2-reduction20 of the C(23) carbonyl followed by
protection of the resulting 1,2-diol as an acetonide to provide
(þ)-73. Removal of both benzyl groups, followed by use of the
Fraser-Reid tactic8 then provided epoxide (þ)-74 ready for union
with the C(1)–C(10) A-ring dithiane-(ꢀ)-6.
Scheme 20.
At this juncture, we turned to reduction of the C(15,16) olefin,
the idea being that a possible conformational change of the spiro-
cycle might lead to an increase of the desired epimer upon
a
dithiane addition. Olefin (þ)-57 was therefore selectively hydro-
genated in the presence of two benzyl ethers employing the Adams
catalyst to furnish (þ)-63 (Scheme 21). Selective removal of the BPS
group, followed by Parikh–Doering oxidation provided access to
spirocyclic aldehyde (þ)-65, now lacking the C(15,16) olefin. Ad-
dition of the anion derived from (ꢀ)-8 provided an improved yield
(ca. 77%) of the coupling product (þ)-66, however virtually no
stereochemical preference was observed. Nonetheless, the increase
in yield in conjunction with the 1:1 ratio constituted an improve-
S
S
BnO
BnO
S
BnO
BnO
S
22
OH
O
DMP, Py
O
O
H
O
O
H
CH2Cl2
93%
OMe
OMe
(+)-66
(+)-70
S
S
BnO
`
ment from the previously observed outcome vis-a-vis material
22
NCS, AgNO3
OH
DIBALH
CH2Cl2
advancement. Also of importance, the epimeric alcohols are sepa-
rable by flash chromatography.
BnO
O
O
H
2,6-lut., CH3CN-H2O
70%
OMe
80%
(+)-71
BnO
BnO
BnO
BnO
OBPS
OMe
O
OBPS
O O
PtO2, H2
23
BnO
BnO
BnO
BnO
O
O
O O
1) Zn(BH4)2,
Et2O
OH
OMe
O
EtOAc-EtOH
84%
O
O
O
H
OMe
(+)-63
(+)-57
OMe
2) Me2C(OMe)2,
TsOH
BnO
BnO
(+)-73
70%
O
(+)-72
OH
OMe
TBAF, THF
97%
O O
SO3·Py, DMSO
i-Pr2NEt, CH2Cl2
O
(+)-64
1) LiDBB, THF, 94%
2) Trisimid, NaH, 89%
O
O
O
OMe
S
S
S
S
BnO
BnO
BnO
BnO
OH
(+)-74
O
O O
H
OMe
O O
(-)-8
Scheme 23.
OMe
n-BuLi, THF
77%
dr = 1.1:1
Toward this end, treatment of epoxide (þ)-74 with the anion
derived from dithiane (ꢀ)-6 proceeded smoothly to furnish alcohol
(þ)-75 in 86% yield (Scheme 24), which upon removal of dithiane
moiety employing MeI/CaCO3, followed by 1,3-anti reduction led to
(+)-65
(+)-66
Scheme 21.

6604
A.B. Smith III et al. / Tetrahedron 66 (2010) 6597–6605
diol (þ)-76. Completion of the synthesis of the C(1)–C(25) southern
hemisphere of spirastrellolide B (þ)-54 was then achieved by
protection of 1,3-diol as an acetonide and removal of the TBS ether.
1103 cmꢀ1; 1H NMR (500 MHz, C6D6)
d
6.17 (ddd, J¼17.3, 10.4, 6.7 Hz,
1H), 5.56 (dd, J¼10.0, 2.6 Hz, 1H), 5.44 (dd, J¼10.0, 1.9 Hz, 1H), 5.11
(app dt, J¼17.3, 1.5 Hz 1H), 5.08 (app dt, J¼10.5, 1.4 Hz, 1H), 4.63 (d,
J¼6.5 Hz, 1H), 4.39–4.26 (m, 4H), 3.95 (dd, J¼10.0, 6.1 Hz, 1H), 3.93
(d, J¼9.3 Hz, 1H), 3.80 (app td, J¼9.5, 2.7 Hz, 1H), 3.61–3.56 (m, 1H),
3.53 (ddd, J¼11.0, 9.4, 5.2 Hz, 1H), 3.31–3.25 (m, 1H), 3.19 (s, 3H),
3.16–3.11 (m, 1H), 2.20 (ddd, J¼14.3, 4.5, 2.8 Hz, 1H), 2.02 (ddd,
J¼13.0, 9.7, 6.3 Hz, 1H), 1.95–1.82 (m, 4H), 1.79–1.72 (m, 3H),
1.70–1.65 (m, 3H), 1.64 (s, 3H), 1.56 (s, 3H), 1.55 (s, 3H), 1.47 (app td,
J¼13.5, 4.1 Hz, 1H), 1.40 (s, 3H), 1.36–1.22 (m, 4H), 1.20 (d, J¼6.7 Hz,
3H), 1.17 (s, 9H), 1.10–1.02 (m, 2H), 0.76 (d, J¼7.1 Hz, 3H); 13C NMR
OTBS
O
S
O
O
OH
(−)-6, n-BuLi,
KOt-Bu
O
O
O
O O
O O
S
OMe
OMe
THF
84%
(+)-74
(125 MHz, C6D6)
d 177.6, 142.3, 133.5, 129.1, 113.5, 108.5, 100.2, 93.5,
(+)-75
81.6, 75.8, 74.6, 74.5, 74.1, 73.4, 72.1, 66.3, 62.9, 61.6, 56.1, 43.6, 42.0,
41.2, 38.7, 37.2, 36.2, 35.1, 34.4, 32.3, 32.0, 27.3, 27.0, 26.5, 26.3, 25.6,
24.0, 23.4, 17.3, 17.2; HRMS (ESI, MþNa) m/z 729.4539 (calcd for
C40H66O10Na: 729.4554).
Alcohol (þ)-54. TBAF in THF (0.1 mL, 1.0 M, 0.10 mmol) was
added to a solution of (þ)-76A (0.006 g, 0.0081 mmol) in THF
(0.4 mL) at 0 ꢁC. After stirring for 1 h aqueous saturated NH4Cl was
added and the resulting mixture was extracted with EtOAc (3ꢂ).
The combined organic layers were dried over anhydrous Na2SO4
and concentrated in vacuo. Flash chromatography on silica gel,
OTBS
O
O
1) MeI, CaCO3
OH
CH3CN-H2O, 88%
2) Me4NBH(OAc)3
CH3CN, AcOH
O
O O
HO
OMe
62%, dr = 4:1
(+)-76
using ethyl acetate/hexanes (1:9/3:7) as eluent, gave 0.005 g
23
(98%) of (þ)-54 as white foam: [
a
]
þ26.0 (c 0.133, CHCl3); IR
D
OH
(neat) 3450, 2933, 1639, 1457, 1377, 1250, 1220, 1092 cmꢀ1; 1H NMR
(500 MHz, CDCl3)
1) Me2C(OMe)2,
TsOH
O
O
d
5.94 (ddd, J¼17.4, 10.4, 7.0, 1H), 5.15–5.05 (m,
O
2H), 4.36 (dd, J¼6.0, 0.8, 1H), 4.19–4.12 (m, 1H), 4.04–3.97 (m, 1H),
3.93 (dd, J¼10.3, 6.0, 1H), 3.82–3.76 (m, 2H), 3.59 (dd, J¼9.3, 0.9,
1H), 3.58–3.52 (m, 2H), 3.47 (td, J¼10.3, 2.1, 1H), 3.41 (ddd, J¼11.3,
9.4, 5.0, 1H), 3.31 (s, 3H), 3.03–2.95 (m, 1H), 2.73 (t, J¼5.4, 1H),
2.02–1.92 (m, 2H), 1.85–1.65 (m, 7H), 1.64–1.50 (m, 7H), 1.49–1.16
(m, 7H), 1.44 (s, 3H), 1.38 (s, 3H), 1.36 (s, 3H), 1.32 (s, 3H), 1.07 (d,
89%
O
O O
2) TBAF, THF
98%
O
OMe
(+)-54
Scheme 24.
J¼6.8, 3H), 0.88 (d, J¼6.6, 3H); 13C NMR (125 MHz, CDCl3)
d 142.1,
113.9, 108.2, 100.5, 95.2, 81.5, 78.6, 76.1, 74.6, 74.4, 72.1, 71.4, 63.2,
62.9, 61.7, 56.0, 42.7, 42.2, 40.3, 38.4, 37.0, 36.3, 35.7, 34.7, 32.0, 31.8,
27.5, 26.7, 26.4, 25.8, 25.1, 23.7, 22.8, 18.4,17.2; HRMS (ES, MþNa) m/
z 647.4110 (calcd for C35H60O9Na: 647.4135).
3. Summary
Effective syntheses of the C(1)–C(25) southern hemispheres of
both spirastrellolide A (1) and B (2) have been achieved. In the case
of spirastrellolide A (1), the synthesis required 33 steps and pro-
ceeded in 0.20% overall yield, whereas the second generation ap-
proach to spirastrellolide B (2) required 32 steps and proceeded in
0.60% overall yield. Both syntheses feature four dithiane unions,
including a highly effective three component anion relay (ARC)
tactic to access the linear precursor of the BC spiroketal. Further
studies toward the total syntheses of spirastrellolides A and B
continue in our laboratory.
Acknowledgements
This paper is dedicated to Professor Steven Ley, outstanding
scientist, scholar and friend, on the occasion of receipt of the 2010
Tetrahedron Prize.
Financial support was provided by the National Institutes of
Health through Grant G.M.-29028. We are also grateful to Eli Lilly
and Company for a graduate fellowship to D.-S.K. Finally we thank
Drs. G.T. Furst, R. Kohli, and P. Carroll at the University of Penn-
sylvania for assistance in obtaining NMR, high-resolution mass
spectra, and X-ray analysis, respectively.
4. Experimental
4.1. General
Supplementary data
Acetonide (þ)-5. To a solution of ketone (þ)-53A (0.0019 g,
2.9 mmol) in Et2O (0.7 mL) at 0 ꢁC was added a solution of Zn(BH4)2
in Et2O (0.07 mL, 0.15 M, 0.011 mmol). The resulting solution was
stirred for 40 min at 0 ꢁC, and a saturated NaHCO3 aqueous solution
(2 mL) was then added. The mixture was extracted with Et2O
(3 mLꢂ3), and the combined organic layers were washed with brine
(2 mL), dried over MgSO4, filtered, and concentrated in vacuo.
Without further purification, the resulting diol was dissolved in
acetone/2,2-dimethoxypropane (0.7/0.1 mL). To the resulting solu-
tion at room temperature was added TsOH (1 mg, 5 mmol). After
30 min at room temperature, triethylamine (0.6 mL) was added, and
the resulting mixture was concentrated in vacuo. Flash chromatog-
Spectroscopic and analytical data and selected experimental
procedures associated with this article can be found in the online
version. Supplementary data associated with this article can be
found in the online version, at doi:10.1016/j.tet.2010.01.082.
References and notes
1. (a) Williams, D. E.; Roberge, M.; Van Soest, R.; Andersen, R. J. J. Am. Chem. Soc.
2003, 125, 5296; (b) Williams, D. E.; Lapawa, M.; Feng, X.; Tarling, T.; Roberge, M.;
Andersen, R. J. Org. Lett. 2004, 6, 2607; (c) Warabi, K.; Williams, D. E.; Patrick, B. O.;
Roberge, M.; Andersen, R. J. J. Am. Chem. Soc. 2007, 129, 508; (d) Williams, D. E.;
Keyzers, R. A.; Warabi, K.; Des-jardine, K.; Riffell, J. L.; Roberge, M.; Andersen, R. J.
J. Org. Chem. 2007, 72, 9842.
raphy on silica gel, using ethyl acetate/hexanes (1:4) as eluent,
20
provided 0.0013 g (64%) of (þ)-5 as a colorless oil: [
a
]
þ9.1 (c 0.15,
D
2. (a) Paterson, I.; Anderson, E. A.; Dalby, S. M.; Lim, J. H.; Genovino, J.; Maltas, P.;
Moessner, C. Angew. Chem., Int. Ed. 2008, 47, 3016; (b) Paterson, I.; Anderson, E. A.;
C6H6); IR (neat) 3078, 2934, 2872, 1730, 1458, 1377, 1221, 1158,

A.B. Smith III et al. / Tetrahedron 66 (2010) 6597–6605
6605
Dalby, S. M.; Lim, J. H.; Genovino, J.; Maltas, P.; Moessner, C. Angew. Chem., Int. Ed.
2008, 47, 3021.
7. Laakso, L. M. Doctoral Thesis; University of Pennsylvania: Philadelphia, PA, USA,
2001.
3. (a) Liu, J.; Hsung, R. P. Org. Lett. 2005, 7, 2273; (b) Liu, J.; Yang, J. K.; Ko, C.;
Hsung, R. P. Tetrahedron Lett. 2006, 47, 6121; (c) Paterson, I.; Anderson, E. A.;
Dalby, S. M.; Loiseleur, O. Org. Lett. 2005, 7, 4121; (d) Paterson, I.; Anderson, E.
A.; Dalby, S. M.; Loiseleur, O. Org. Lett. 2005, 7, 4125; (e) Paterson, I.; Anderson,
E. A.; Dalby, S. M.; Lim, J. H.; Maltas, P.; Moessner, C. Chem. Commun. 2006,
4186; (f) Furstner, A.; Fenster, M. D. B.; Fasching, B.; Godbout, C.; Radkowski, K.
Angew. Chem., Int. Ed. 2006, 45, 5506; (g) Furstner, A.; Fenster, M. D. B.;
Fasching, B.; Godbout, C.; Radkowski, K. Angew. Chem., Int. Ed. 2006, 45, 5510;
(h) Pan, Y.; De Brabander, J. K. Synlett 2006, 853; (i) Wang, C.; Forsyth, C. J. Org.
Lett. 2006, 8, 2997; (j) Paterson, I.; Anderson, E. A.; Dalby, S. M.; Lim, J. H.;
Loiseleur, O.; Maltas, P.; Moessner, C. Pure Appl. Chem. 2007, 79, 667; (k) Wang,
C,; Forsyth, C. J. Heterocycles 2007, 72, 621; (l) Furstner, A.; Fasching, B.; O’Neil,
G. W.; Fenster, M. D. B.; Godbout, C.; Ceccon, J. Chem. Commun. 2007, 3045; (m)
Keaton, K. A.; Phillips, A. J. Org. Lett. 2008, 10, 1083; (n) Yang, J.-H.; Liu, J.; Hsung,
R. P. Org. Lett. 2008, 10, 2525; (o) Chandrasekhar, S.; Rambabu, C.; Reddy, A. S.
Org. Lett. 2008, 10, 4355.
8. (a) Hicks, D. R.; Fraser-Reid, B. Synthesis 1974, 203; (b) Cink, R. D.; Forsyth, C. J.
J. Org. Chem. 1995, 60, 8122.
9. Benechie, M.; Delpech, B.; Khuong-Huu, Q.; Khuong-Huu, F. Tetrahedron 1992,
48, 1895.
10. Fujita, E.; Nagao, Y.; Kaneko, K. Chem. Pharm. Bull. 1978, 26, 3743.
11. (a) Gribble, G. W.; Nutaitis, C. F. Tetrahedron Lett. 1983, 24, 4287; (b) Evans, D. A.;
Chapman, K. T.; Carreira, E. M. J. Am. Chem. Soc. 1988, 110, 3560.
12. Ferrier, R. J.; Prasad, N. J. J. Chem. Soc. C 1969, 1, 570.
13. Charette, A. B.; Lacasse, M.-C. Org. Lett. 2002, 4, 3351.
14. Bernet, B.; Bishop, P. M.; Caron, M.; Kawamata, T.; Roy, B. L.; Ruest, L.; Sauve, G.;
Soucy, P.; Deslongchamps, P. Can. J. Chem. 1985, 63, 2810.
15. Yu, J.; Cho, H.-S.; Falck, J. R. J. Org. Chem. 1993, 58, 5892.
16. Bhattacharjee, A.; Soltani, O.; DeBrabander, J. K. Org. Lett. 2002, 4, 481.
17. Honda, Y.; Morita, E.; Ohshiro, K.; Tsuchihashi, G.-I. Chem. Lett. 1988, 21.
18. Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem. Soc. 1991, 113, 4092.
19. Corey, E. J.; Erickson, B. W. J. Org. Chem. 1971, 36, 3553.
20. Nakata, T.; Tanaka, T.; Oishi, T. Tetrahedron Lett. 1983, 24, 2653.
21. Parikh, J. R.; Doering, W. v. E. J. Am. Chem. Soc. 1967, 89, 5505.
22. Smith, A. B., III; Tomioka, T.; Risatti, C. A.; Sperry, J. B.; Sfouggatakis, C. Org. Lett.
2008, 10, 4359.
4. For an initial communication on the C(1)–C(25) fragment of spirastrellolide A
see: Smith, A. B., III; Kim, D.-S. Org. Lett. 2007, 9, 3311.
5. (a) Smith, A. B., III; Xian, M. J. Am. Chem. Soc. 2006, 128, 66; (b) Smith, A. B., III;
Xian, M.; Kim, W.-S. J. Am. Chem. Soc. 2006, 128, 12368.
6. Abel, S.; Faber, D.; Hu¨ter, O.; Giese, B. Angew. Chem., Int. Ed. Engl. 1995, 33, 2466.
23. Ide, M.; Yasuda, M.; Nakata, M. Synlett 1998, 936.