A total synthesis of spirastrellolide a methyl ester

Article abstract of DOI:10.1002/chem.201203965

A concise total synthesis of spirastrellolide A methyl ester (1 a, R ¹=Me) as the parent compound of a series of highly cytotoxic marine macrolides is disclosed, which exploits and expands the flexibility of a synthesis plan previously developed by our group en route to the sister compound spirastrellolide F methyl ester (6 a, R¹=Me). Key to success was the masking of the signature Δ^15,16-bond of 1 a as a C16-carbonyl group until after the stereogenic center at C24 had been properly set by a highly selective hydrogenation of the C24 exo-methylene precursor 66. Conformational control over the macrocyclic frame allowed the proper stereochemical course to be dialed into this reduction process. The elaboration of the C16 ketone to the C15-C16 double bond was accomplished by a chemoselective alkenyl triflate formation followed by a palladium-catalyzed hydride delivery. The role of the ketone at C16 as a strategic design element is also evident up-stream of the key intermediate 66, the assembly of which hinged upon the addition of the polyfunctionalized dithiane 37 to the similarly elaborate aldehyde fragment 46. Other crucial steps of the total synthesis were an alkyl-Suzuki coupling and a Yamaguchi lactonization that allowed the Northern and the Southern sector of the target to be stitched together and the macrocyclic perimeter to be forged. The lateral chain comprising the remote C46 stereocenter was finally attached to the core region by a modified Julia-Kocienski olefination. The preparation of the individual building blocks led to some methodological spin-offs, amongst which the improved procedure for the N-O-bond cleavage of isoxazolines by zero-valent molybdenum and the ozonolysis of a double bond in the presence of other oxidation-prone functionality are most noteworthy. Preliminary biological data suggest that the entire carbon framework, that is the macrocyclic core plus the lateral chain, might be necessary for high cytotoxicity. Copyright

Full text of DOI:10.1002/chem.201203965

DOI: 10.1002/chem.201203965
A Total Synthesis of Spirastrellolide A Methyl Ester
[a]
Alexander Arlt, Stefan Benson, Saskia Schulthoff, Barbara Gabor, and Alois Fꢀrstner*
Abstract: A concise total synthesis of
spirastrellolide A methyl ester (1a,
R =Me) as the parent compound of a
C15–C16 double bond was accomplish-
ed by a chemoselective alkenyl triflate
formation followed by a palladium-cat-
alyzed hydride delivery. The role of the
ketone at C16 as a strategic design ele-
ment is also evident up-stream of the
key intermediate 66, the assembly of
which hinged upon the addition of the
polyfunctionalized dithiane 37 to the
similarly elaborate aldehyde fragment
46. Other crucial steps of the total syn-
thesis were an alkyl-Suzuki coupling
and a Yamaguchi lactonization that al-
lowed the Northern and the Southern
sector of the target to be stitched to-
gether and the macrocyclic perimeter
to be forged. The lateral chain compris-
ing the remote C46 stereocenter was fi-
nally attached to the core region by a
modified Julia–Kocienski olefination.
The preparation of the individual
building blocks led to some methodo-
logical spin-offs, amongst which the im-
proved procedure for the N–O-bond
cleavage of isoxazolines by zero-valent
molybdenum and the ozonolysis of a
double bond in the presence of other
oxidation-prone functionality are most
noteworthy. Preliminary biological data
suggest that the entire carbon frame-
work, that is the macrocyclic core plus
the lateral chain, might be necessary
for high cytotoxicity.
1
series of highly cytotoxic marine mac-
rolides is disclosed, which exploits and
expands the flexibility of a synthesis
plan previously developed by our
group en route to the sister compound
spirastrellolide F methyl ester (6a,
1
R =Me). Key to success was the mask-
1
5,16
ing of the signature D -bond of 1a as
a C16-carbonyl group until after the
stereogenic center at C24 had been
properly set by a highly selective hy-
drogenation of the C24 exo-methylene
precursor 66. Conformational control
over the macrocyclic frame allowed the
proper stereochemical course to be
dialed into this reduction process. The
elaboration of the C16 ketone to the
Keywords: antitumor agents · mac-
rolides · natural products · phospha-
tase inhibitors · total synthesis
Introduction
PP2A, which is inhibited by the parent compound 1 with an
IC50 of ꢀ1 nm. Because of this remarkable potency, the
[5]
An innovative cell-based assay for antimitotic agents al-
lowed the Anderson group to perform a high-throughput
evaluation of more than 24000 extracts from marine and ter-
restrial organisms of the animal and plant kingdoms. One
of the hits of this screening campaign was the methanol ex-
tract of the marine invertebrate Spirastrella coccinea collect-
ed in the Atlantic off Dominica, which elicited an unusual
phenotypic response. A total of seven closely related macro-
lides, dubbed spirastrellolides A–G (1–7), were isolated
from these extracts in form of the corresponding methyl
Table 1. Structures of the spirastrellolides; the color-coded sectors indi-
cate the stereoclusters, the relative stereochemistry of which could origi-
nally be assigned in 2004, but the mutual relationship between them re-
[1]
[7]
mained unknown. The full stereostructure of the natural products was
[3,4]
established as shown only in late 2007.
1
esters (R =Me) (Table 1). The individual family members
were found similarly potent in triggering premature cell
[2–4]
cycle arrest.
Unlike many other antimitotic agents, how-
ever, they do not interfere with tubulin dynamics but rather
seem to target protein phosphatases (PP), in particular
[
15,16]
1
2
~
R
R
X
Y
Z
[
a] Dipl.-Chem. A. Arlt, Dr. S. Benson, S. Schulthoff, B. Gabor,
Prof. A. Fꢀrstner
Max-Planck-Institut fꢀr Kohlenforschung
p
spirastrellolide A (1)
spirastrellolide B (2)
spirastrellolide C (3)
spirastrellolide D (4)
spirastrellolide E (5)
spirastrellolide F (6)
spirastrellolide G (7)
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
H
H
H
Cl
H
H
H
Cl
H
H
Cl
H
Cl
Cl
H
H
OH
H
H
–
–
4
5470 Mꢀlheim/Ruhr (Germany)
p
Fax : (+49)208306-2994
E-mail: fuerstner@kofo.mpg.de
p
–
H
H
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201203965.
p
3596
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3596 – 3608

FULL PAPER
spirastrellolides might qualify as tools for chemical biology
and eventually even as lead structures in the quest for novel
therapeutic agents for the treatment of metabolic disorders
and/or cancer, if the supply problem can be overcome. In
shifted away from the lactone to the C16–C17 bond flanking
[14]
the conspicuous spiroketal center at C17. This ambitious
plan could be successfully reduced to practice by recourse to
[15,16]
the power of ring-closing alkyne metathesis
with a gold-catalyzed activation of the resulting triple bond
toward transannular spiroacetalization reaction
(Scheme 1).
in concert
this context, it is important to note that the free acids 1 and
1
2
(R =H) have lately been re-isolated from an Epipolasis
a
[6]
[17]
sponge harvested in the East China Sea. This finding
strongly suggests that the spirastrellolides might not be
innate secondary metabolites of the sponges themselves but
rather be of bacterial origin. With the re-extracted materials,
it was also possible to show that the free acids (R =H) and
the derived methyl esters (R =Me) exhibit similarly high
In the present publication we expand our investigations
into this enticing family of marine natural products beyond
spirastrellolide F. Although the total synthesis of spirastrel-
lolide A methyl ester (1a) outlined below has certainly bene-
fitted from our previous work (Scheme 2), care was taken to
refine the preparation and assembly of the required building
blocks while solving the additional selectivity problems
posed by this exigent target.
1
1
[6]
levels of cytotoxicity against the tested HeLa cell line.
The very intriguing biological profile of the spirastrello-
lides is encoded in a molecular framework of exquisite com-
plexity. The elucidation of their intricate topology and the
unusual stereochemical features was highly challenging, not
least because of the small amounts of material available
from the natural sources. While advanced NMR spectrosco-
py had allowed the relative configuration of the color-coded
segments within the polyketide frame to be established
Results and Discussion
Retrosynthetic analysis: As previously discussed in some
detail, our original synthesis plan had hinged upon the for-
mation of the macrocycle of the spirastrellolides at the non-
stereogenic C25–C26 bond by olefin metathesis, which—un-
fortunately—was unsuccessful, despite considerable experi-
[7]
early on (Table 1), the relationship between these domains
could be unraveled only more than four years after the first
report. At this point, a re-isolation campaign had provided
enough material for chemical derivatization. Specifically, a
truncated derivative of 2 could be crystallized, allowing the
absolute stereochemistry of the macrocyclic perimeter to be
[18]
mentation (Scheme 2, blue route). Prompted by this fail-
ure, the approach toward spirastrellolide F was reprogram-
[19]
med toward the use of an alkyl-Suzuki reaction. This re-
vised tactics allowed us to utilize the rather sensitive
“Northern” sector 8 unchanged, while effecting the critical
CꢁC-bond formation one carbon atom further away from
[3]
determined by X-ray diffraction; the configuration of the
lateral stereocenter at C46 had to be elucidated via chemical
[4]
correlation.
Although the spirastrellolides aroused considerable inter-
est immediately after their disclosure and prompted remark-
able chemical creativity, the lasting uncertainty about their
identity imposed severe constraints upon any synthetic en-
deavor. As long as the relationship between the color-coded
stereoclusters of 1 (or congeners) was unclear, no less than
the very bulky and rigid DEF-bisspiroketal unit (Scheme 2,
[12,14]
red and green routes).
The price to be paid for this amendment was the stereo-
chemical uncertainty during the subsequent hydrogenation
of the resulting C24 exo-methylene group in 12 to the
methyl branch. Careful conformational analysis, however,
had suggested that protection of the adjacent hydroxyl
groups at C22 and C23 in the form of an isopropylidene
acetal provides a handle to control the course of the reduc-
tion. This conjecture turned out to be correct: the properly
configured compound was obtained as the only detectable
product, when 12 was hydrogenated with the aid of a modi-
fied Crabtree catalyst. Furthermore, X-ray diffraction data
confirmed our original design strategy by revealing the con-
16 possible isomers had to be considered as the potential
target and all approaches until late 2007 could not help but
envisage fragment coupling processes at or close to the bor-
ders of these domains. This notion is evident from the ex-
ploratory studies published by a sizeable number of research
[
8–10]
groups.
Even the first total syntheses of spirastrelloli-
1
de A methyl ester (1a, R =Me) published by Paterson and
co-workers in 2008 and of spirastrellolide F methyl ester
1
(
6a, R =Me) disclosed by our group in 2009 still reflect this
formational features that account for the exquisite sub-
[11,12]
[12]
limited freedom to operate.
Once the uncertainty had
strate-control over this transformation.
Therefore, the
vanished and the structures were established and confirmed,
both teams followed up with “second-generation” total syn-
theses of the very same targets, which took advantage of the
acquired greater strategic flexibility. Notably, the Paterson
group postponed the macrolactonization event until after
completion of the entire carbon framework. This tactics
was clearly more productive than the originally pursued at-
tachment of the lateral chain to a truncated macrocycle,
which had turned out to be more challenging than anticipat-
ed. Our own second-generation synthesis of 6a features an
even more profound change, as the site of ring closure was
Suzuki reaction/hydrogenation sequence is considered a
stronghold of our approaches to 6a, which converge at this
point (cf., Scheme 1).
If one were to retain this favorable strategic maneuver en
route to spirastrellolide A (1), the timing of events becomes
[13]
[15,16]
a critical issue: specifically, the D
unsaturation within
the B-ring of this target can only be revealed at a late stage
of the synthesis after the exo-methylene group has been re-
duced (Scheme 2). Reasoning that a carbonyl group at C16
might serve as an adequate orthogonal double bond surro-
gate, we opted for a dithiane-based fragment coupling for
Chem. Eur. J. 2013, 19, 3596 – 3608
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3597

A. Fꢀrstner et al.
dized and the resulting alde-
hyde 21 immediately subjected
[21]
to a Brown crotylation
to
yield compound 22 in high
enantiomeric and diastereomer-
ic excess. Cleavage of the
double bond by ozonolysis fol-
lowed by a reductive workup
using sodium borohydride was
effective but furnished a crude
product that contained consid-
erable amounts of boron impur-
1
1
ities as determined by B NMR
spectroscopy. Gratifyingly, the
desired diol was isolated in ex-
cellent purity without the need
for column chromatography by
simply stirring a solution of the
material in THF with aqueous
NaOH (15% w/w). Treatment
with anisaldehyde dimethylace-
tal and catalytic amounts of
pyridinium p-toluenesulfonate
(
PPTS) in the presence of mo-
lecular sieves yielded the p-me-
thoxyphenyl acetal 23 as a 3:1
mixture of epimers at the ben-
zylic position. For the sake of
convenience of characteriza-
tion, the mixture can be equili-
brated to the more stabile
isomer with an equatorially-dis-
posed phenyl group upon treat-
ment with camphorsulfonic acid
[
22]
(
CSA).
The acetal ring of 23 was
then reductively opened on
[23]
treatment with DIBAl-H
to
Scheme 1. Summary of the approaches to the spirastrellolides published by our group: original but unsuccess-
[
18]
give the secondary PMB-ether
ful RCM-based route toward spirastrellolide A methyl ester (1a, blue); first successful total synthesis of spi-
1
[12]
rastrellolide F methyl ester (6a, R =Me) by a Suzuki/Yamaguchi route (red); second generation synthesis 24; this reaction had to be per-
of 6a comprising a Suzuki coupling, an RCAM-based macrocyclization and a Au-catalyzed transannular spiro-
formed at low temperature to
[
14]
ketalization (green), cf. text.
avoid cleavage of the primary
[24]
TBS group.
Alcohol 24 was
transformed into the corre-
sponding iodide 25, which served to alkylate lithiated 1,3-di-
thiane. Recourse to the previously developed slow addition
protocol furnished the desired product 26 in good yield
the assembly of the targetꢂs “Southern” domain. While this
rationale echoes our own exploratory studies in this field,
[9]
the preparation of the required fragments 16–20 would draw
from the knowledge gained since then. Provided that this
route brings the macrocyclic array 15 into reach, the lateral
chain could then be attached via a modified Julia–Kocienski
[9a]
while minimizing the competing elimination of HI.
A
standard desilylation using TBAF gave the corresponding
primary alcohol, which was oxidized under Parikh–Doering
[20]
[25]
olefination that had distinguished our second-generation
conditions to aldehyde 19 in readiness for chain elonga-
[14]
synthesis of the sister compound 6a.
tion.
The required coupling partner 18 was best prepared by a
[
9]
The C1–C16 dithiane fragment: For the construction of the
short dithiane fragment 19 we adapted our previous synthe-
sis of a similar fragment to the current needs (Scheme 3).
Specifically, mono-TBS-protected 1,3-propanediol was oxi-
modification of our original route (Scheme 4). Specifically,
[26]
Weiler-dianion alkylation of 27 with bromide 28 followed
by a Noyori-type reduction of the resulting b-ketoester 29 in
the presence of catalytic amounts of HCl gave product 30 in
[9a]
3598
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3596 – 3608

Spirastrellolide A
FULL PAPER
[
15,16]
Scheme 2. Retrosynthetic analysis, accounting for the need to unravel the D
-bond of the target late in the synthesis, only after the chiral center at
C24 has been properly set by substrate-controlled hydrogenation of the exo-methylene group formed in the projected cross coupling.
the silyl ether and cause a spontaneous oxa-Michael addi-
tion with formation of the desired pyrane derivative 18. Al-
though the same product can also be reached by processing
the free alcohol 30 directly, the temporary TES protection
was found beneficial in terms of yield, selectivity and mate-
[29]
rial throughput.
Under soft enolization conditions, ketone 18 was cleanly
transformed into silyl enol ether 33, which underwent a pro-
ductive Mukaiyama aldol reaction with aldehyde 19 in the
[30]
presence of BF ·Et O as promoter. In line with our previ-
3
2
ous experiences, the subsequent 1,3-anti reduction of the
carbonyl group in the resulting product 34 was best effected
[31]
by the Evans–Tishchenko protocol, without the dithioace-
Scheme 3. a) (ꢁ)-[(E)-crotyl]B
de; b) i) O , MeOH, CH Cl
, ꢁ788C; ii) NaBH
aq. NaOH, THF, 93%; c) pMeOC CH(OMe)
9%, d.r. 3:1; d) DIBAl-H, CH Cl
, ꢁ608C, 82%; e) I
THF, 08C, quant.; f) 1,3-dithiane, nBuLi, THF, DMPU, ꢁ788C!RT,
1%; g) TBAF, THF, 08C ! RT, 98 %; h) SO ·pyridine, DMSO, Et N,
CH Cl , 08C, 85%; RT=room temperature, PPTS=pyridinium p-tolue-
nesulfonate, ipc=isopinocampheyl, MS=molecular sieves, DIBAl-H=
diisobutylaluminiumhydride, DMPU=N,N’-dimethylpropylenurea,
A
H
U
G
R
N
U
G
2
, THF, ꢁ788C, 82%, 94% ee, 95%
, ꢁ788C ! RT, then 15%
, PPTS, CH Cl , MS 4 ꢃ,
, PPh , imidazole,
[32]
tal interfering.
Cleavage of the concomitantly installed
3
2
2
4
acetate moiety and simultaneous reduction of the methyl
ester terminus of 35 was accomplished with DIBAl-H in
preparation for the upcoming lithiation of the dithiane
during the envisaged fragment coupling. Prior to this critical
event, the three hydroxyl groups in 36 had to be protected
upon exposure of the crude material from the DIBAl-step
to TBSOTf/lutidine under standard conditions.
4
A
H
U
G
R
N
N
2
2
2
8
2
3
8
3
3
2
2
TBAF=tetra-n-butylammonium fluoride, TBS=tert-butyldimethylsilyl;
PMB=p-methoxybenzyl.
The C17–C24 fragment: The challenging stereotetrad com-
prised within the C17–C24 sector of the target was built-up
starting from epoxide 38, which in turn derives from divinyl-
carbinol via Sharpless asymmetric epoxidation and O-pro-
excellent optical purity, leaving the trisubstituted alkene un-
[27,28]
touched;
this compound was O-silylated to give the
TES-ether 31 prior to ozonolytic cleavage of the double
bond. The resulting aldehyde reacted cleanly with the stabi-
[33]
tection as previously described. BF ·Et O-mediated open-
3 2
lized ylide Ph P=CC(O)Me to give the corresponding enone
ing of the oxirane ring by lithiated TMS-acetylene followed
3
32, which was treated with camphorsulfonic acid to cleave
by methylation of the resulting hydroxyl group with nBuLi/
Chem. Eur. J. 2013, 19, 3596 – 3608
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3599

A. Fꢀrstner et al.
Scheme 5. a) TMSCꢃCH, nBuLi, THF, ꢁ788C, then 38, BF
, MeOH,
S, ꢁ788C!RT, 80%, see text; d) 41, MgBr ·Et O,
, ꢁ788C!RT, d.r. ꢂ16:1; e) TBAF, THF, 08C, 73% (2 steps);
Cl
08C!RT, 82%; g) HN-
(OMe)Me·HCl, MeMgBr, THF, ꢁ88C!RT, 91%; h) L-selectride,
NaOH/H Cl , 08C,
(2 mol%), 47 (4 mol%), MeCN, 608C; ii)
O, acetone, 508C, 95%; L-selectride=lithium tri-sec-butyl-
2
·OEt ,
3
ꢁ
788C, 94%; b) nBuLi, MeOTf, THF, ꢁ78!08C, 95%; c) i) O
3
CH
CH
2
Cl
Cl
2
, ꢁ788C; ii) Me
2
2
2
2
2
f) 2,2-dimethoxypropane, CSA, CH
AHCTUNGTRENNUNG
2
2
,
2
O
2
, THF, ꢁ788C, 89%; i) TESOTf, 2,6-lutidine, CH
5%; j) i) [CpRu(MeCN) ]PF
5,
2
2
9
4
A
H
U
G
R
N
N
3
6
H
2
Scheme 4. a) NaH, nBuLi, then 28, THF, 08C!RT, 55%; b) [Ru
binap) [Et NH ], HCl cat., H (5 bar), MeOH, 458C, 93 % (>98% ee);
c) TESCl, imidazole, CH Cl , CH Cl /MeOH/
, 08C!RT, quant.; d) i) O
pyridine 5:5:1, ꢁ788C; ii) Me S, 78 8C!RT, 83 %; e) Ph P=C(O)Me, tol-
uene, 1108C, 79%; f) CSA, CH Cl
/MeOH 4:1, ꢁ308C, 85% (>99% ee);
g) TMSOTf, iPr NEt, CH Cl ·OEt , ꢁ788C, 88%, d.r.
, ꢁ788C; h) 19, BF
3:ꢀ7; i) SmI
, MeCHO, THF, ꢁ118C, 95%, d.r. 90/ꢀ10; j) DIBAL-H,
toluene, ꢁ788C!RT, quant.; k) TBSOTf, 2,6-lutidine, CH Cl , 08C!RT,
0%; TES=triethylsilyl; Tf=trifluoromethanesulfonyl.
2 5
Cl ((R)-
AHCTUNGERTNNUNG( hydrido)-borate; CSA=camphorsulfonic acid, Cp=cyclopentadienyl.
2
]
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
2
2
2
2
2
3
2
2
2
3
lem when passing a stream of ozone through the reaction
mixture. We reasoned that the use of a saturated stock solu-
tion of ozone in dichloromethane might solve the selectivity
2
2
2
2
2
3
2
9
2
[36]
2
2
problem. In fact, this change in the experimental setup al-
9
lowed the olefin site of 39 to be selectively cleaved to give
the desired aldehyde 40 in high yield. The ozone solution
was slowly added via a jacketed addition funnel cooled to
ꢁ788C to a solution of the substrate at the same tempera-
ture and the conversion closely monitored by thin-layer
chromatography. Under these conditions, well reproducible
results were obtained even upon scale up of the reaction.
Delighted with the efficiency of this transformation, we
went on to set the yet missing two stereocenters by a
MgBr ·OEt -mediated glycolate aldol reaction of aldehyde
MeOTf afforded product 39 in good yield, which displays an
alkyne and an alkene terminus meant to serve as orthogonal
aldehyde surrogates (Scheme 5). First, the olefinic site had
to be oxidatively cleaved as prelude for a subsequent glyco-
late aldol reaction. In our previous campaign, a similar
transformation had been accomplished by dihydroxylation
under Sharpless conditions followed by cleavage of the re-
2
2
[12]
[37,38]
sulting diol with Pb
A
H
U
G
E
U
G
Despite some optimization,
40 with the known silyl ketene acetal 41.
Changing the
4
however, the overall yield remained moderate and the use
of expensive and/or hardly benign reagents was deemed sub-
optimal, particularly on a larger scale.
solvent from toluene to dichloromethane increased the dia-
stereoselectivity from 3:1 to ꢂ16:1 (determined by HPLC
after global desilylation of the product), while retaining the
excellent productivity of this transformation.
We therefore sought to replace this two-step procedure by
ozonolysis, even though we were apprehensive that the elec-
tron-rich p-methoxybenzyl ether as well as the TMS-protect-
ed alkyne present in 39 might be prone to overoxidation.
Although the reactions were closely monitored by using dif-
In preparation for the fragment coupling by addition of
the lithiated dithiane 37, it was originally planned to convert
[39]
the ester site of 42 into the corresponding Weinreb amide
and the alkyne terminus into the required aldehyde. In so
doing, however, it was observed that treatment of 42 with
HN AHCTUNGTERUNNN(G OMe)Me·HCl and iPrMgCl furnished substantial
[34]
ferent dies as indicators (Sudan red 7B, Sudan III), the re-
[35]
sults remained erratic and the yields rather disappointing.
In all runs, a methyl ester derivative formed by cleavage of
the alkyne moiety was isolated as a major byproduct, sug-
gesting that the dosing of the oxidant was a significant prob-
amounts of the corresponding isopropyl ketone. Taking ad-
vantage of this unexpected course, iPrMgCl was replaced by
[40]
an excess of MeMgBr;
in fact, its use furnished the
3600
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3596 – 3608

Spirastrellolide A
FULL PAPER
methyl ketone 44 directly and in good yield without the
need for the isolation of the Weinreb amide 43 that must
have been transiently formed. This smooth conversion com-
pleted the carbon backbone of the C17–C24 synthon. To
avoid any selectivity issue during the dithiane addition, how-
ever, we opted to reduce the ketone group in 44 with L-se-
lectride in THF. The syn-configured secondary alcohol was
exclusively formed under these conditions in line with a
sired secondary alcohol 48 (mixture of diastereomers at
C17) was isolated in 75% yield, together with 19% of the
product formed by butyl addition. Although the yield of 48
obtained with nBuLi alone was definitely lower (62%), the
loss of advanced material by a competing carbonyl addition
process was not ideal. The subsequent optimization showed
that an increased amount of Bu Mg favored the parasitic
2
butyl addition, whereas a ratio of nBuLi/Bu Mg 8.4:1 al-
2
[41]
Felkin–Anh transition state model. In accord with this ra-
tionale, a substantially lower selectivity (76:24) was ob-
served when the reaction was carried out in dichlorome-
thane, presumably because a-chelation becomes competitive
in the non-coordinating solvent. The relative configuration
of the alcohol product could be proven by X-ray analysis of
lowed the desired product 48 to be obtained in 83% yield
and the side reaction to be suppressed. Even higher Li/Mg
ratios, however, turned out to be less productive, presuma-
bly due to the fading positive effect of the magnesium addi-
tive. It was suggested in the literature that the Bu Mg in-
2
creases the stability of the deprotonated dithiane by forming
[42]
[46]
a derivative (see the Supporting Information).
a magnesium ate-complex.
We have reasons to believe
After protection as a TES-ether, the stage was set for a
that it also reduces the basicity of the anion and hence
favors nucleophilic addition over deprotonation of the alde-
hyde partner.
ruthenium-catalyzed anti-Markovnikov hydration of the ter-
[43]
minal alkyne 45 thus obtained. The reaction required only
mol% of the catalyst, which
was prepared from commercial
CpRu(MeCN) PF ] and the
2
[
ACHTUNGTRENNUNG
3
6
bulky ARPYPhos-ligand 47 in
MeCN. This convenient proce-
dure afforded the desired
product 46 in excellent yields,
without any cleavage of the
TES-ether interfering when
the amount of water was limit-
ed to five equivalents. The ef-
fective use of the alkyne as an
aldehyde equivalent epito-
mized by this conversion com-
pleted the second building
block required for the prepara-
tion of the southern sector of
spirastrellolide A methyl ester
(
1a). The described route was
found high yielding and readily
scalable.
Elaboration of the “Southern”
domain: The decisive forma-
tion of the [6,6]-spiroketal
motif began with the addition
of dithiane 37 to aldehyde 46
(
Scheme 6).
A
number of
[
44,45]
bases and conditions
were
screened but a protocol using a
mixed metallic reagent ulti-
mately gave the best results. In
the event, 37 was deprotonated
Scheme 6. a) nBuLi/Bu
2
Mg, THF, 08C!RT then RT!ꢁ788C, 46 in THF, 83%; b) DMP, CH
2 2
Cl , 08C, 93%;
with a mixture of Bu Mg and
c) DDQ, CH Cl , H O, 0 8C; d) PPTS, CH Cl , MeOH, 26% (51) + 43% (52); e) TBSCl, imidazole, CH Cl ,
2
46]
2
2
2
2
2
2
2
[
nBuLi in a 1:4 ratio and the
solution of the resulting meta-
lated species cooled to ꢁ788C
prior to the addition of the al-
dehyde component. The de-
93%; f) TPAP, NMO, MeCN, CH
THF, ꢁ118C, 75% (87% brsm); i) NCS, AgNO
NaH PO , isopentene, tBuOH, H O, 94% (over two steps); DMP=Dess–Martin periodinane, DDQ=2,3-di-
2
Cl
2
, MS 4 ꢃ, 82%; g) 53, KHMDS, THF, ꢁ788C, 89%; h) HF·pyridine,
3
, 2,6-lutidine, MeCN, H O, 77%; j) DMP, CH Cl ; k) NaClO ,
2
2
2
2
2
4
2
chloro-5,6-dicyano-1,4-benzoquinone, TPAP=tetra-n-propylammonium perruthenate, NMO=N-methylmor-
pholine N-oxide, KHMDS=potassium bis(trimethylsilyl)amide, NCS=N-chloro succinimide, PPTS=pyridini-
um p-toluenesulfonate.
Chem. Eur. J. 2013, 19, 3596 – 3608
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3601

A. Fꢀrstner et al.
Next, the diastereomeric alcohols 48 were oxidized to
The Northern fragment: The preparation of the required
[47]
ketone 49 using Dess–Martin periodinane without affect-
ing the oxidation-prone dithiane. The cleavage of the p-me-
thoxybenzyl ethers with DDQ in a biphasic medium deliv-
ered a mixture of two compounds of distinctly different po-
larity, which are believed to be the open chain diol-ketone
Northern fragment closely followed the sequence previously
[9b]
described by our group,
except that the benzyl ether at
the primary hydroxyl group was replaced by a PMB group.
This change was deemed necessary to ensure selective de-
protection at this site in the final stages of the synthesis,
once the signature C15–C16 double bond embedded in the
B-ring of the target has been unveiled.
5
0 and a hemiacetal form thereof; in any case, the more
polar component converted into the less polar product upon
separation by flash chromatography on silica. A spontane-
ous spiroketalization, as previously noticed for related com-
pounds in the spirastrellolide F series, was not observed,
most likely because of the steric hindrance about the ketone
exerted by the “neopentylic” dithiane neighborhood. Grati-
fyingly though, treatment of this difficult-to-characterize
product with PPTS in CH Cl /MeOH resulted in the cleav-
[53]
To this end, the known epoxide 58 was chosen as the
point of departure, which is accessible in excellent optical
[54]
purity by hydrolytic kinetic resolution. On treatment with
[55]
dimethylsulfonium methylide,
this compound was trans-
formed into the allylic alcohol 59 as the substrate for a dia-
[56]
stereoselective 1,3-dipolar cycloaddition with nitrile oxide
[9b]
57 derived from oxime 56 (Scheme 7). The reaction was
highly diastereoselective, affording diol 60 as the only prod-
uct in high yield after complete desilylation. Selective trans-
formation of the primary alcohol into the corresponding
iodide followed by TES protection of the secondary alcohol
gave fragment 61 in readiness for alkylation of the lithium
2
2
age of the secondary TES-ether followed by clean spirocycli-
[48]
zation.
Despite the only mildly acidic conditions, some
concomitant deprotection of the primary TBS-ether was
also noticed; this side reaction, however, was inconsequen-
tial as the primary alcohol 51 could be selectively re-protect-
ed. The desired spirocyclic product 52 was thus obtained in
respectable yield as a single isomer.
[9b]
enolate derived from the known cyanohydrin 62. The de-
sired product 63 was obtained in excellent yield as an incon-
sequential mixture of diastereomers (d.r. 87:13).
In preparation for the upcoming Suzuki fragment cou-
pling, the remaining free OH group in 52 had to be oxidized
to the corresponding ketone without affecting the dithiane
moiety. Of the different reagents screened, tetra-n-propyl-
ammonium perruthenate (TPAP) in combination with N-
methylmorpholine-N-oxide (NMO) gave by far the best
result, when the reaction was performed at high concentra-
tion; to reach full conversion, however, acetonitrile had to
[49]
be added as a co-solvent. Subsequent deprotonation with
the aid of KHMDS and trapping of the resulting enolate
with triflimide 53 afforded the required alkenyl triflate 54
[12]
without incident. In line with our expectations, this prod-
uct was stable enough to allow for the selective deprotection
of the primary TBS-ether on exposure to HF·pyridine to
give 55, although it was best to quench the reaction before
full conversion had been reached and to recycle the remain-
ing starting material.
At this stage, the dithiane had fulfilled its double role as a
handle for the fragment coupling and an orthogonal protect-
ing group for a carbonyl moiety at C16. Taking the sensitivi-
ty of dithioacetals towards oxidative cleavage into account,
it was originally hoped that the oxidation of the primary al-
cohol in 55 to the required carboxylic acid and the deprotec-
tion of the dithiane could be accomplished in one opera-
[50}
tion.
Although a variety of reagents were screened, the
yields of the desired product 16 remained impractically low.
For the sake of material throughput, we therefore pursued a
stepwise approach, commencing with the well reproducible
[51]
cleavage of the dithioacetal with NCS/AgNO3.
The result-
Scheme 7. a) [Me S]I, nBuLi, THF, ꢁ108C!RT, 88 %; b) tBuOCl,
3
ing product was then effectively transformed by a sequence
CH Cl , ꢁ788C; c) iPrOH, EtMgBr, CH Cl , 08C, then add 57, d) TBAF,
2
2
2
2
[52]
of Dess–Martin- and Pinnick oxidation into acid 16, which
represents properly functionalized surrogate of the
Southern” sector of spirastrellolide A (1).
THF, 76% (over two steps); e) I
2
,
3
imidazole, PPh , THF, 67%;
f) TESOTf, 2,6-lutidine, THF, 93%; g) 62, LDA, THF, ꢁ788C, then 61,
5% (d.r. 87:13); h) [Mo(MeCN) (CO) ], MeCN, O, 50 8C, then
Me NO; i) TASF, DMF, H O; j) PPTS, CH Cl /MeOH, 48% (after one
recycle, over three steps); LDA=lithium diisopropylamide, TASF=
(Me N) S][F Si(Me) ].
a
9
A
H
U
G
R
N
N
3
3
H
2
“
3
2
2
2
[
2
3
A
H
U
T
E
N
N
2
3
3602
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3596 – 3608

Spirastrellolide A
FULL PAPER
The procedure for the subse-
quent NꢁO bond cleavage was
subject to optimization. Rather
than relying on [Mo(CO) ] as
6
[9b,57]
previously described,
found the use of [Mo-
(MeCN) (CO) ] beneficial, as
we
ACHTUNGTRENNUNG
3
3
it allows the reaction to be run
more cleanly at lower temper-
[58]
ature. To facilitate the work-
up, trimethylamine-N-oxide
was added to the crude mix-
ture. This simple means im-
proved the mass balance and
delivered the product virtually
free of metal contaminants.
Since the removal of metal res-
idues is a known issue in the
cleavage of isoxazolines with
zero-valent molybdenum com-
Scheme 8. a) 9-H-9-BBN dimer, THF; b) i) aq. NaOH; ii) 16, [PdCl2
(20 mol%), 65%; c) i) 2,4,6-trichlorobenzoyl chloride, Et N, toluene; ii) DMAP, toluene, reflux, 88%;
[3.3.1]nonane, dppf=1,1’-bis(diphenylphosphi-
2 2 3
ACHTUNGTRENNGU( dppf)·CH Cl ] (20 mol%), Ph As
3
[58,59]
DMAP=4-dimethylaminopyridine, 9-H-9-BBN=9-borabicycloACHTUNGTRENNUNG
plexes,
this simple modifi-
no)ferrocene.
cation should be of wider rele-
vance. The product was isolat-
ed as a mixture of partially desilylated compounds, which
was of no concern as global desilylation with concomitant
break-down of the cyanohydrin was to follow. Treatment
with TASF followed by exposure of the resulting sensitive
material to PPTS in a mixture of CH Cl /MeOH gave the
pulsive transannular interactions within the incipient macro-
cycle.
As expected, the hydrogenation of the C24 exo-methylene
group on the macrocyclic frame of 66 was challenging but
exquisitely selective in favor of the desired stereoisomer
(Scheme 9). This outcome is ascribed to the rigid BC-ring
system that effectively shields one of the p-faces of the
2
2
desired thermodynamically favored [5,6,6]-bis-spiroketal 17
as the major compound, together with three diastereomers
that could be separated and re-equilibrated to further the
olefin, allowing hydrogen to be delivered only from the
[9]
[12b]
material supply. Overall, this sequence afforded the north-
ern fragment in a respectable yield, ready for coupling with
the Southern sector.
open Re-face.
However, considerable transannular strain
has to be overcome during the concurrent rehybridization of
2
3
the carbon atom from sp to sp , during which the methyl
branch clashes into the congested 6,6-spirobicyclic rear as
long as the adjacent isopropylidene acetal is in place. In line
with our previous observations, only a highly active Crab-
tree catalyst escorted by the non-coordinating BARF coun-
teranion was able to affect this transformation (for the struc-
ture of this complex in the solid state, see the Supporting In-
Formation of the macrolactone and completion of the total
synthesis: With all required building blocks in hand, the
project entered the critical phase of macrocycle formation
followed by elaboration of the final target.
In concord with our previous experiences, the Northern
and the Southern sectors could be stitched together by an
[42,61,62]
formation).
Moreover, high hydrogen pressure was
[19]
alkyl-Suzuki coupling followed by a Yamaguchi macrolac-
mandatory, although no significant difference was noticed
between 80 and 200 bar. As catalyst decomposition is fast
under such forcing conditions, it was better not to insist on
complete conversion but rather recycle the remaining sub-
strate. This allowed the desired product 67 to be obtained in
58% yield (73% brsm). Small amounts (ca. 5%) of a by-
product could be separated, which turned out to be the eno-
lether 68 that originates from an iridium-catalyzed isomeri-
[60]
tonization (Scheme 8). To this end, the terminal alkene of
7 was hydroborated with 9-H-9-BBN dimer to give the al-
1
kylborane 64 bearing a borinate ester at the C37 position.
Reaction of this compound with the polyfunctionalized al-
kenyl triflate 16 under the aegis of [(dppf)PdCl ] as precata-
2
lyst and aqueous sodium hydroxide as promoter delivered
the desired product 65 in respectable 65% yield after an ox-
idative work up to cleave the C37-borinate moiety off. This
seco-acid was equally cleanly converted to the macrolactone
[63]
zation of the double bond into the ring.
At this stage, the C16-ketone group of 67 had to be trans-
[60]
[15,16]
6
6 by a Yamaguchi reaction, provided that the preformed
formed into the characteristic D
-alkene of spirastrelloli-
mixed anhydride was slowly added to a refluxing solution of
DMAP in toluene. The need for such harsh conditions is as-
cribed to an unfavorable conformation imposed on the seco-
acid by the isopropylidene group at C22–C23 and/or to re-
de A. Amongst the different choices, we opted for an adven-
turous selective enolization of the carbonyl in the presence
of the equally enolizable lactone group, since the protons at
C15 seemed sterically more accessible and, at the same
time, also more acidic than the protons of the methylene
Chem. Eur. J. 2013, 19, 3596 – 3608
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3603

A. Fꢀrstner et al.
compound to the required olefin was achieved with
the aid of [Pd ACHTGNUTERNN(UNG OAc) ACTHUNGTRNENU(G PPh ) ] and tributylammoni-
2 3 2
[64,65]
um formate,
which afforded the fully functional
macrolactone core 70 of spirastrellolide A in almost
quantitative yield.
The now missing side chain was readily installed
by cleavage of the PMB group with DDQ followed
by oxidation of the resulting primary alcohol 71 to
the corresponding aldehyde 72 on exposure to
Dess–Martin periodinane. A subsequent Julia–Ko-
[20]
[14]
cienski olefination with the known sulfone 73
delivered fully protected spirastrellolide A as single
geometric isomer at the newly formed double bond.
Global deprotection with PPTS in a mixed solvent
system completed our total synthesis of spirastrello-
lide A methyl ester (1a). In line with our own ob-
servations made with the sister compound spira-
[12]
strellolide F methyl ester (6a) as well as with the
[11d]
later report of Paterson and co-workers,
the
NMR spectra of our samples of 1a, especially the
1
H NMR spectra, showed a significant and charac-
teristic time- and solvent dependent behavior. They
prove the integrity of our synthetic samples and
their identity with the natural product (see the Sup-
porting Information).
Surprisingly, however,
a
careful inspection
showed that 1a seems to be in equilibrium with a
minor diastereomer. After separation by HPLC, the
pure components convert back to the same equili-
brium composition during the time needed for a
complete NMR characterization. The same unusual
phenomenon was also observed for some other
macrocyclic derivatives passed through en route to
1
. Whereas the primary alcohol 71 and the corre-
sponding aldehyde 72 were single components each
within the limits of detection), a second isomer
(
built up in case of 67 (d.r. ꢀ87:13), 70 (d.r. ꢀ94:6)
and the fully protected spirastrellolide A (d.r.
ꢀ
90:10). We gather from the literature that a simi-
lar observation was made by Matsunaga and co-
workers during the re-isolation of the spirastrello-
lides from a sponge of the Epipolasis genus collect-
[66]
ed in the East China Sea.
Scheme 9. a) [Ir
73% brsm); b) 53, KHMDS, THF, ꢁ1008C, 73%; c) [Pd
HCO H, nBu N, DMF, 608C, 97%; d) DDQ, CH Cl , H
CH Cl
, 08C!RT, 88 %; f) 72, KHMDS, THF, ꢁ788C, then 71, THF, ꢁ78!ꢁ658C,
9%; g) PPTS, MeOH/Et O/H 7:2:1, 508C, 51%; cod=cycloocta-1,5-diene,
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(cod)
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(PCy
3
)
A
H
U
G
R
N
U
G
A
H
U
G
R
N
U
G
2
(80 bar), 1,2-dichloroethane, 58%
(OAc) (PPh ] (35 mol%),
O, 0 8C, 89%; e) DMP,
(
A
H
N
T
E
N
N
2
A
T
N
T
E
N
N
3
)
2
Due to the complexity and time-dependence of
the spectra, the small concentration of the second
component, and the signal overlap with the major
constituent, the identity of the respective second
isomer in the equilibrium could not be fully estab-
lished. Yet, for compound 67, the spectral disper-
sion was sufficiently large, allowing us to deduce
that the constitution of the minor component is un-
2
3
2
2
2
2
2
6
2
BARF=tetrakis(3,5-bis-trifluoromethylphenyl)borate, brsm=based on recovered
starting material.
group at C2 (Scheme 9). A test reaction, in which KHMDS
was added to a mixture of the substrate and triflimide 53 in
THF at ꢁ788C, afforded an encouraging 55% of the desired
alkenyl triflate 69 together with several minor byproducts.
Pleasingly, the yield could be improved to 73% upon further
lowering of the temperature to ꢁ1008C. Reduction of this
altered, but the stereochemistry about the F-ring seems to
1
3
be affected. For example, the C signals assigned to this
region show distinctive shift differences relative to those of
the major isomer, which are particularly substantial for C36
(Dd=8.1 ppm) and C38 (Dd=4.1 ppm). We tentatively as-
cribe these observations to a reversible epimerization of the
3604
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3596 – 3608

Spirastrellolide A
FULL PAPER
C35-spiroacetal center: al-
though the DEF-ring subunit
Table 2. Assessment of the cytotoxicity (IC50) of 1a and two truncated variants against three different cancer
cell lines after incubation for 4 d.
[
a]
Compound
MDA-MB-231 [nm]
LoVo [nm]
4T1 [nm]
features
a double anomeric
effect, all large substituents on
the five-membered F-ring are
cis-oriented, thus rendering its
[67]
a-side very crowded.
Any
9
.6
14.2
<0.051
bulky group at C38—as pres-
ent in 1a, 67 and 70—may well
destabilize the [5,6,6]-bisspiro-
ketal to the extent that it can
equilibrate with an acetal
epimer even under very mild
[68,69]
conditions.
>
1000
1000
974
>1000
Preliminary biological assess-
ment: As a very first foray
into a more detailed biological
assessment of the spirastrello-
lides, compound 70 represent-
ing the entire macrocyclic pe-
rimeter was globally deprotect-
ed to give the pentaol 74 as a
truncated analogue of 1. Like-
wise, we ventured to attach the
correct side chain to com-
pound 17 in order to obtain
compound 75 representing the
>
1000
>1000
[
a] MDA-MB-231 breast cancer cell line; human LoVo colon carcinoma cell line; 4T1 mouse carcinoma cell
line; the data refer to the standard MTS endpoint assay; MTS=5-(3-carboxymethoxyphenyl)-2-(4,5-dime-
thylthiazolyl)-3-(4-sulfophenyl)tetrazolium, inner salt.
“Northern half” of the natural
product. Together with the synthetic samples of spirastrello-
lide A methyl ester (1a), these compounds were screened
for their cytotoxicity against the human MDA-MB-231
breast cancer cell line, the human LoVo colon carcinoma
cell line, and the 4T1 mouse carcinoma cell line.
cules at the outset of our project, it was clear then that only
a flexible synthesis plan could possibly be successful. The
chemical suppleness inherent to our blueprint paid first im-
portant dividends after the originally projected ring-closing
metathesis reaction had failed to forge the core region of
these targets. The projected assembly process could be
amended without the need to redesign all individual frag-
ments and hence spirastrellolide F methyl ester (6a) be
reached by two different routes that converge in their late
As can be seen from the data compiled in Table 2, 1a
itself was found very cytotoxic, with the 50 picomolar (!)
level of activity against the 4T1 cell line being particularly
impressive. In contrast, however, neither compound 74 with
a truncated side chain nor compound 75 as the upper half of
the natural product exerted any noticeable activity in the
tested concentration range. Although we are certainly ap-
prehensive that no final conclusions can be drawn from such
a limited set of compounds, these data suggest that the
entire carbon framework might be necessary to reach mean-
ingful levels of cytotoxicity. This notion will be subject to
further scrutiny in the future, together with studies into the
phosphatase inhibition properties of these and related com-
[12,14]
stages.
The work presented herein illustrates yet another facet of
the flexible design: the project could be redirected, without
undue effort, toward the total synthesis of the parent com-
pound spirastrellolide A methyl ester (1a). The additional
selectivity issue posed by the signature C15–C16 double
bond of this specific target was addressed by masking this
olefinic site in the form of a carbonyl group until after the
methyl branch at C24 had been properly set by a substrate-
controlled hydrogenation reaction. This tactics, in turn, in-
spired an assembly process based on dithiane chemistry,
which is necessarily different in its details from the routes
leading to 6a, yet shares with them a common fuzzy chemi-
cal logic. Therefore we believe that this venture, in its totali-
ty, meets Woodwardꢂs verdict, that chemical synthesis
always has to have some element of planning, but this ele-
[70]
pounds.
Conclusion
The work outlined above is the latest chapter of our groupꢂs
long engagement with the spirastrellolides, a family of
highly unusual marine natural products with enticing struc-
tures and promising biological properties. Because of serious
stereochemical uncertainties surrounding these target mole-
[71]
ment should never be too rigid.
Chem. Eur. J. 2013, 19, 3596 – 3608
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3605

A. Fꢀrstner et al.
Ed. 2006, 45, 5506–5510; b) A. Fꢀrstner, M. D. B. Fenster, B. Fasch-
ing, C. Godbout, K. Radkowski, Angew. Chem. 2006, 118, 5636–
Experimental Section
5
641; Angew. Chem. Int. Ed. 2006, 45, 5510–5515.
All experimental details can be found in the Supporting Information.
The material includes compound characterization, the crystal structures
of a p-bromobenzoate ester proving the stereochemistry of 45 and of [Ir-
[
[
10] Reviews: a) I. Paterson, S. M. Dalby, Nat. Prod. Rep. 2009, 26, 865–
8
4
73; b) I. Paterson, S. M. Dalby, P. Maltas, Isr. J. Chem. 2011, 51,
06–419; c) M. V. Perkins, Angew. Chem. 2008, 120, 2963–2967;
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(cod)
3
ACHTUNGTRENNUNG( PCy ) ACHTUNGTRENNUNG( pyridine] ACHTUNGTRENNUNG[ BARF] used for the hydrogenation of compound
Angew. Chem. Int. Ed. 2008, 47, 2921–2925.
6
6, and copies of spectra for all new compounds.
11] a) I. Paterson, E. A. Anderson, S. M. Dalby, J. H. Lim, J. Genovino,
P. Maltas, C. Moessner, Angew. Chem. 2008, 120, 3058–3062;
Angew. Chem. Int. Ed. 2008, 47, 3016–3020; b) I. Paterson, E. A.
Anderson, S. M. Dalby, J. H. Lim, J. Genovino, P. Maltas, C. Moess-
ner, Angew. Chem. 2008, 120, 3063–3067; Angew. Chem. Int. Ed.
Acknowledgements
2
008, 47, 3021–3025; c) I. Paterson, E. A. Anderson, S. M. Dalby,
Generous financial support by the MPG and the Fonds der Chemischen
Industrie (Kekulꢄ stipends to A.A. and S.B.) is gratefully acknowledged.
We express our sincere gratitude to all co-workers who helped initiate
the spirastrellolide project and laid a solid basis for the current success
by developing the general strategic and methodological framework for
our studies; their names appear in references [9], [12], [14], [18], and
J. H. Lim, P. Maltas, O. Loiseleur, J. Genovino, C. Moessner, Org.
Biomol. Chem. 2012, 10, 5861–5872; d) I. Paterson, E. A. Anderson,
S. M. Dalby, J. H. Lim, P. Maltas, Org. Biomol. Chem. 2012, 10,
5873–5886.
[12] a) G. W. OꢂNeil, J. Ceccon, S. Benson, M.-P. Collin, B. Fasching, A.
Fꢀrstner, Angew. Chem. 2009, 121, 10124–10129; Angew. Chem. Int.
Ed. 2009, 48, 9940–9945; b) S. Benson, M.-P. Collin, G. W. OꢂNeil, J.
Ceccon, B. Fasching, M. D. B. Fenster, C. Godbout, K. Radkowski,
R. Goddard, A. Fꢀrstner, Angew. Chem. 2009, 121, 10130–10134;
Angew. Chem. Int. Ed. 2009, 48, 9946–9950.
[
29]. Furthermore, we thank the analytical departments of our Institute
for excellent support and Dr. Andreas Maderna and Dr. Frank Loganzo,
Pfizer Inc., for the cytotoxicity data.
[
13] I. Paterson, P. Maltas, S. M. Dalby, J. H. Lim, E. A. Anderson,
Angew. Chem. 2012, 124, 2803–2807; Angew. Chem. Int. Ed. 2012,
51, 2749–2753.
[
1] M. Roberge, B. Cinel, H. J. Anderson, L. Lim, X. Jiang, L. Xu,
C. M. Bigg, M. T. Kelly, R. J. Andersen, Cancer Res. 2000, 60, 5052–
5
058.
[14] S. Benson, M.-P. Collin, A. Arlt, B. Gabor, R. Goddard, A. Fꢀrstner,
Angew. Chem. 2011, 123, 8898–8903; Angew. Chem. Int. Ed. 2011,
50, 8739–8744.
[
[
[
2] D. E. Williams, M. Roberge, R. Van Soest, R. J. Andersen, J. Am.
Chem. Soc. 2003, 125, 5296–5297.
3] K. Warabi, D. E. Williams, B. O. Patrick, M. Roberge, R. J. Ander-
sen, J. Am. Chem. Soc. 2007, 129, 508–509.
4] D. E. Williams, R. A. Keyzers, K. Warabi, K. Desjardine, J. L. Rif-
fell, M. Roberge, R. J. Andersen, J. Org. Chem. 2007, 72, 9842–
[15] a) J. Heppekausen, R. Stade, R. Goddard, A. Fꢀrstner, J. Am.
Chem. Soc. 2010, 132, 11045–11057; b) J. Heppekausen, R. Stade,
A. Kondoh, G. Seidel, R. Goddard, A. Fꢀrstner, Chem. Eur. J. 2012,
1
1
8, 10281–10299; c) A. Fꢀrstner, G. Seidel, Angew. Chem. 1998,
10, 1758–1760; Angew. Chem. Int. Ed. 1998, 37, 1734–1736.
9
845.
[
5] For reviews on phosphatase inhibitors, see: a) A. McCluskey,
A. T. R. Sim, J. A. Sakoff, J. Med. Chem. 2002, 45, 1151–1175;
b) R. E. Honkanen, T. Golden, Curr. Med. Chem. 2002, 9, 2055–
[
[
16] A. Fꢀrstner, P. W. Davies, Chem. Commun. 2005, 2307–2320.
17] A. Fꢀrstner, P. W. Davies, Angew. Chem. 2007, 119, 3478–3519;
Angew. Chem. Int. Ed. 2007, 46, 3410–3449.
2
3
075; c) L. Bialy, H. Waldmann, Angew. Chem. 2005, 117, 3880–
906; Angew. Chem. Int. Ed. 2005, 44, 3814–3839.
[
[
18] A. Fꢀrstner, B. Fasching, G. W. OꢂNeil, M. D. B. Fenster, C. God-
AHCTUNGTRENNGNUb out, J. Ceccon, Chem. Commun. 2007, 3045–3047.
[
[
6] M. Suzuki, R. Ueoka, K. Takada, S. Okada, S. Ohtsuka, Y. Ise, S.
Matsunaga, J. Nat. Prod. 2012, 75, 1192–1195.
7] For the correction of the originally proposed constitution (ref. [2])
and the elucidation of the relative stereochemistry of the color
coded stereoclusters of 1a, see: D. E. Williams, M. Lapawa, X. Feng,
T. Tarling, M. Roberge, R. J. Andersen, Org. Lett. 2004, 6, 2607–
19] a) N. Miyaura, T. Ishiyama, H. Sasaki, M. Ishikawa, M. Satoh, A.
Suzuki, J. Am. Chem. Soc. 1989, 111, 314–321; b) reviews: S. R.
Chemler, D. Trauner, S. J. Danishefsky, Angew. Chem. 2001, 113,
4
676–4701; Angew. Chem. Int. Ed. 2001, 40, 4544–4568; c) G.
Seidel, A. Fꢀrstner, Chem. Commun. 2012, 48, 2055–2070; for early
applications in total synthesis from this laboratory, see: d) A. Fꢀrst-
ner, I. Konetzki, Tetrahedron 1996, 52, 15071–15078; e) A. Fꢀrstner,
G. Seidel, J. Org. Chem. 1997, 62, 2332–2336; f) A. Fꢀrstner, I. Ko-
netzki, J. Org. Chem. 1998, 63, 3072–3080.
2
610.
[
8] For studies toward the spirastrellolides, see: a) I. Paterson, E. A.
Anderson, S. M. Dalby, O. Loiseleur, Org. Lett. 2005, 7, 4121–4224;
b) I. Paterson, E. A. Anderson, S. M. Dalby, O. Loiseleur, Org. Lett.
[
20] P. R. Blakemore, W. J. Cole, P. J. Kocie n´ ski, A. Morley, Synlett 1998,
2
2
005, 7, 4125–4128; c) J. Liu, R. P. Hsung, Org. Lett. 2005, 7, 2273–
276; d) Y. Pan, J. K. De Brabander, Synlett 2006, 853–856; e) I. Pa-
2
6–28.
[
[
21] H. C. Brown, K. S. Bhat, J. Am. Chem. Soc. 1986, 108, 293–294.
22] It should be noted that the direct use of CSA in the acetalization re-
action resulted in significantly lower yields of 23.
terson, E. A. Anderson, S. M. Dalby, J. H. Lim, P. Maltas, C. Moess-
ner, Chem. Commun. 2006, 4186–4188; f) C. Wang, C. J. Forsyth,
Org. Lett. 2006, 8, 2997–3000; g) J. Liu, J. H. Yang, C. Ko, R. P.
Hsung, Tetrahedron Lett. 2006, 47, 6121–6123; h) I. Paterson, E. A.
Anderson, S. M. Dalby, J. Genovino, J. H. Lim, C. Moessner, Chem.
Commun. 2007, 1852–1854; i) A. B. Smith, D.-S. Kim, Org. Lett.
[
23] S. Takano, M. Akiyama, S. Sato, K. Ogasawara, Chem. Lett. 1983,
1
2, 1593–1596.
[
[
24] E. J. Corey, G. B. Jones, J. Org. Chem. 1992, 57, 1028–1029.
25] J. R. Parikh, W. v. E. Doering, J. Am. Chem. Soc. 1967, 89, 5505–
2007, 9, 3311–3314; j) C. Wang, C. J. Forsyth, Heterocycles 2007, 72,
621–632; k) K. A. Keaton, A. J. Phillips, Org. Lett. 2008, 10, 1083–
1086; l) S. Chandrasekhar, C. Rambabu, A. S. Reddy, Org. Lett.
2008, 10, 4355–4357; m) J.-H. Yang, J. Liu, R. P. Hsung, Org. Lett.
2008, 10, 2525–2528; n) J. L.-Y. Chen, M. A. Brimble, Chem.
5
507.
[
[
26] a) L. Weiler, J. Am. Chem. Soc. 1970, 92, 6702–6704; b) S. N.
Huckin, L. Weiler, J. Am. Chem. Soc. 1974, 96, 1082–1087.
27] T. Ohkuma, R. Noyori in Comprehensive Asymmetric Catalysis,
Vol. 1 (Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer,
Berlin, 1999, pp. 199–246.
Commun. 2010, 46, 3967–3969; o) A. B. Smith, H. Smits, D.-S. Kim,
Tetrahedron 2010, 66, 6597–6605; p) G. Sabitha, A. S. Rao, J. S.
Yadav, Synthesis 2010, 505–509; q) J. L.-Y. Chen, M. A. Brimble, J.
Org. Chem. 2011, 76, 9417–9428; r) X. Wang, T. J. Paxton, N. Li,
A. B. Smith, Org. Lett. 2012, 14, 3998–4001.
[28] a) S. A. King, A. S. Thompson, A. O. King, T. R. Verhoeven, J. Org.
Chem. 1992, 57, 6689–6691; see also: b) A. Fꢀrstner, T. Dierkes,
O. R. Thiel, G. Blanda, Chem. Eur. J. 2001, 7, 5286–5298.
[29] See also: G. W. OꢂNeil, A. Fꢀrstner, Chem. Commun. 2008, 4294–
4296.
[
9] a) A. Fꢀrstner, M. D. B. Fenster, B. Fasching, C. Godbout, K. Rad-
kowski, Angew. Chem. 2006, 118, 5632–5636; Angew. Chem. Int.
3606
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3596 – 3608

Spirastrellolide A
FULL PAPER
[
[
30] a) D. A. Evans, M. J. Dart, J. L. Duffy, M. G. Yang, J. Am. Chem.
Soc. 1996, 118, 4322–4343; b) review: T. Mukaiyama, Org. React.
[55] L. Alcaraz, J. J. Harnett, C. Mioskowski, J. P. Martel, T. Le Gall, D.-
S. Shin, J. R. Falck, Tetrahedron Lett. 1994, 35, 5449–5452.
[56] a) S. Kanemasa, M. Nishiuchi, A. Kamimura, K. Hori, J. Am. Chem.
Soc. 1994, 116, 2324–2339; b) see also: J. W. Bode, N. Fraefel, D.
Muri, E. M. Carreira, Angew. Chem. 2001, 113, 2128–2131; Angew.
Chem. Int. Ed. 2001, 40, 2082–2085.
1
982, 28, 203–331.
31] a) D. A. Evans, A. H. Hoveyda, J. Am. Chem. Soc. 1990, 112, 6447–
6
2
449; b) review: K. J. Ralston, A. N. Hulme, Synthesis 2012, 2310–
324.
[
[
32] A. B. Smith, D. Lee, C. M. Adams, M. C. Kozlowski, Org. Lett. 2002,
, 4539–4541.
33] a) M. Nakatsuka, J. A. Ragan, T. Sammakia, D. B. Smith, D. E. Ueh-
ling, S. L. Schreiber, J. Am. Chem. Soc. 1990, 112, 5583–5601; b) for
a use in this laboratory, see: A. Fꢀrstner, O. R. Thiel, J. Org. Chem.
[57] P. G. Baraldi, A. Barco, S. Benetti, S. Manfredini, D. Simoni, Synthe-
sis 1987, 276–278.
[58] a) A. Guarna, A. Guidi, A. Goti, A. Brandi, F. De Sarlo, Synthesis
1989, 175–178; b) J. A. Bull, E. P. Balskus, R. A. J. Horan, M.
Langner, S. V. Ley, Chem. Eur. J. 2007, 13, 5515–5538.
4
2
000, 65, 1738–1742.
34] a) T. Veysoglu, L. A. Mitscher, J. K. Swayze, Synthesis 1980, 807–
10; b) see also: A. Fꢀrstner, D. De Souza, L. Turet, M. D. B. Fen-
[59] J. W. Bode, E. M. Carreira, J. Org. Chem. 2001, 66, 6410–6424.
[60] a) J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull.
Chem. Soc. Jpn. 1979, 52, 1989–1993; b) review: A. Parenty, X.
Moreau, J.-M. Campagne, Chem. Rev. 2006, 106, 911–939.
[61] R. Crabtree, Acc. Chem. Res. 1979, 12, 331–337.
[
8
ster, L. Parra-Rapado, C. Wirtz, R. Mynott, C. W. Lehmann, Chem.
Eur. J. 2007, 13, 115–134 and references therein.
[
35] The same outcome was recorded when 39 was replaced by the anal-
ogous substrate bearing a sterically more hindered TIPS moiety.
36] a) M. B. Rubin, J. Chem. Educ. 1964, 41, 388; for applications in syn-
thesis, see: b) P. A. Wender, B. A. DeChristopher, A. J. Schrier, J.
Am. Chem. Soc. 2008, 130, 6658–6659; c) F. Li, S. S. Tartakoff, S. L.
Castle, J. Org. Chem. 2009, 74, 9082–9093; d) D. L. J. Clive, X. He,
M. H. D. Postema, M. J. Mashimbye, J. Org. Chem. 1999, 64, 4397–
[
62] a) B. Wꢀstenberg, A. Pfaltz, Adv. Synth. Catal. 2008, 350, 174–178;
b) S. P. Smidt, A. Pfaltz, E. Martꢆnez-Viviente, P. S. Pregosin, A. Al-
binati, Organometallics 2003, 22, 1000–1009.
[
[
63] a) J. M. Bueno, J. M. Coterꢇn, J. L. Chiara, A. Fernꢅndez-Mayoralas,
J. M. Fiandor, N. Valle, Tetrahedron Lett. 2000, 41, 4379–4382; b) D.
Solꢄ, X. Urbaneja, J. Bonjoch, Org. Lett. 2005, 7, 5461–5464;
c) E. M. Phillips, J. M. Roberts, K. A. Scheidt, Org. Lett. 2010, 12,
4
410.
2
2
830–2833; d) M. Krel, J.-Y. Lallemand, C. Guillou, Synlett 2005,
043–2046.
[
[
37] K. Hattori, H. Yamamoto, J. Org. Chem. 1993, 58, 5301–5303.
38] M. S. M. Timmer, A. Adibekian, P. H. Seeberger, Angew. Chem.
[
64] S. Cacchi, E. Morera, G. Ortar, Tetrahedron Lett. 1984, 25, 4821–
824.
65] Investigations of this key reaction on a model substrate had shown
this combination to be superior to the use of Et SiH and Bu SnH as
2
005, 117, 7777–7780; Angew. Chem. Int. Ed. 2005, 44, 7605–7607.
4
[
[
39] S. Nahm, S. M. Weinreb, Tetrahedron Lett. 1981, 22, 3815–3818.
40] J. M. Williams, R. B. Jobson, N. Yasuda, G. Marchesini, U.-H. Dol-
ling, E. J. J. Grabowski, Tetrahedron Lett. 1995, 36, 5461–5464.
[
3
3
the hydride sources under a variety of experimental conditions.
66] Matsunaga et al. mentioned that an “unidentified equilibrium” visi-
ble in the signals of the C12–C34 region of spirastrellolide B (2) was
[
41] H. Chikashita, T. Nikaya, H. Uemura, K. Itoh, Bull. Chem. Soc. Jpn.
[
1
989, 62, 2121–2123.
[
42] CCDC 907987 and 908599 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
5
observed upon treatment of the free acid 2 with [D ]pyridine, cf.
ref. [6].
[
67] The crowed a-face of the F-ring is also the likely reason why the
acid-catalyzed spiroketalization of the dihydroxyketone derived
from 63 affords bisspiroketal 17 featuring a double anomeric effect
as the major but not as the only isomer, cf. Scheme 8. Furthermore,
the C36 signals of the resulting major isomer 17a (R=Bn) and of
the minor isomer epi-17a with an inverted C35 acetal center show
the same pronounced shift difference (Dd=10.1 ppm), which is
likely caused by the lone pairs of the other ketal oxygen atoms.
[
43] a) A. Labonne, T. Kribber, L. Hintermann, Org. Lett. 2006, 8, 5853–
5
856; b) L. Hintermann, T. T. Dang, A. Labonne, T. Kribber, L.
Xiao, P. Naumov, Chem. Eur. J. 2009, 15, 7167–7179; c) L. Hinter-
mann in C-X Bond Formation, Top. Organomet. Chem., Vol. 31
(
Ed.: A. Vigalok), Springer, Berlin, 2010, pp. 123–155.
[
44] a) B. H. Lipshutz, E. Garcia, Tetrahedron Lett. 1990, 31, 7261–7264;
b) D. Seebach, E. J. Corey, J. Org. Chem. 1975, 40, 231–237.
45] a) D. Seebach, Synthesis 1969, 17–36; b) M. Yus, C. Nꢅjera, F. Fou-
belo, Tetrahedron 2003, 59, 6147–6212; c) A. B. Smith, S. M.
Condon, J. A. McCauley, Acc. Chem. Res. 1998, 31, 35–46.
[
[
[
46] M. Ide, M. Nakata, Bull. Chem. Soc. Jpn. 1999, 72, 2491–2499.
47] a) D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155–4156;
b) S. D. Meyer, S. L. Schreiber, J. Org. Chem. 1994, 59, 7549–7552.
48] Spirocyclization did not take place as long as the TES-ether was in
place. This observation is ascribed to the steric hindrance imposed
by this group, which is forced by the flanking cis-configured dioxo-
lane at C22-C23 into close proximity to the alcohol at C21 which
has to attack the ketone.
[
[
49] S. V. Ley, J. Norman, W. P. Griffith, S. P. Marsden, Synthesis 1994,
6
39–666.
[
[
[
50] J. S. Yadav, C. S. Reddy, Org. Lett. 2009, 11, 1705–1708.
51] E. J. Corey, B. W. Erickson, J. Org. Chem. 1971, 36, 3553–3560.
52] a) B. S. Bal, W. E. Childers, Jr., H. W. Pinnick, Tetrahedron 1981, 37,
2
1
2
091–2096; b) G. A. Kraus, M. J. Taschner, J. Org. Chem. 1980, 45,
175–1176; c) B. O. Lindgren, T. Nilsson, Acta Chem. Scand. 1973,
7, 888–890.
[68] For a relevant discussion of another natural product containing a
[5,6]-spiroketal, in which destabilization caused by two eclipsing
substituents on the five membered ring overrides a stabilizing
anomeric effect, see: a) P. Buchgraber, T. N. Snaddon, C. Wirtz, R.
Mynott, R. Goddard, A. Fꢀrstner, Angew. Chem. 2008, 120, 8578–
8582; Angew. Chem. Int. Ed. 2008, 47, 8450–8454; b) T. N. Snaddon,
P. Buchgraber, S. Schulthoff, C. Wirtz, R. Mynott, A. Fꢀrstner,
Chem. Eur. J. 2010, 16, 12133–12140.
[
[
53] A. Fꢀrstner, M. Albert, J. Mlynarski, M. Matheu, E. DeClercq, J.
Am. Chem. Soc. 2003, 125, 13132–13142.
54] a) S. E. Schaus, B. D. Brandes, J. F. Larrow, M. Tokunaga, K. B.
Hansen, A. E. Gould, M. E. Furrow, E. N. Jacobsen, J. Am. Chem.
Soc. 2002, 124, 1307–1315; b) L. P. C. Nielsen, C. P. Stevenson, D. G.
Blackmond, E. N. Jacobsen, J. Am. Chem. Soc. 2004, 126, 1360–
[69] A similar epimerization process accompanied by large chemical
1
3
1
362.
shift differences in the C NMR spectra of the equilibrating isomers,
Chem. Eur. J. 2013, 19, 3596 – 3608
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3607

A. Fꢀrstner et al.
was described for the spiroketal center in the pectenotoxins, see: K.
Sasaki, J. L. C. Wright, T. Yasumoto, J. Org. Chem. 1998, 63, 2475–
mann, ChemBioChem 2004, 5, 1575–1579; e) M. Manger, M.
Scheck, H. Prinz, J. P. von Kries, T. Langer, K. Saxena, H. Schwalbe,
A. Fꢀrstner, J. Rademann, H. Waldmann, ChemBioChem 2005, 6,
1749–1753.
2
480.
[
70] For studies from this laboratory on the preparation and evaluation
of natural-product-based phosphatase inhibitors, see: a) A. Fꢀrstner,
F. Feyen, H, Prinz, H. Waldmann, Angew. Chem. 2003, 115, 5519–
[71] R. B. Woodward in Proceedings of The Robert A. Welch Foundation
Conferences on Chemical Research. XII. Organic Synthesis (Ed.:
W. O. Milligan), The Robert A. Welch Foundation, Houston, 1968,
pp. 3–6.
5
522; Angew. Chem. Int. Ed. 2003, 42, 5361–5364; b) A. Fꢀrstner, F.
Feyen, H. Prinz, H. Waldmann, Tetrahedron 2004, 60, 9543–9558;
c) A. Fꢀrstner, J. Ruiz-Caro, H. Prinz, H. Waldmann, J. Org. Chem.
Received: November 6, 2012
2
004, 69, 459–467; d) A. Fꢀrstner, K. Reinecke, H. Prinz, H. Wald-
Published online: February 18, 2013
3608
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3596 – 3608