Toward the total synthesis of spirastrellolide A. Part 1: Strategic considerations and preparation of the southern domain

Article abstract of DOI:10.1002/anie.200601654

North and South: The unique biological activity of the natural product spirastrellolide A renders it an attractive lead for anticancer agents. The southern hemisphere (C1-C25) and the northern hemisphere (including the chlorinated [5,6,6]-bis-spiroacetal

Full text of DOI:10.1002/anie.200601654

Communications
Natural Products (1)
The initial assignment[1] of this intriguing natural product,
based on extensive NMR studies of its methyl ester derivative
1a (R = Me), was corrected shortly after publication.[2] Even
the revised structure 1 remains tentative; although it certainly
depicts the correct constitution of spirastrellolide A and the
proper relative stereochemistry within the individual domains
embedded into the complex macrocylic frame, it must be
emphasized that the stereochemical relationships between
the segments color-coded in Scheme 2 have yet to be
determined.Moreover, the absolute stereochemistry of 1 is
still unknown.[2]
DOI: 10.1002/anie.200601654
Toward the Total Synthesis of Spirastrellolide A.
Part 1: Strategic Considerations and Preparation
of the Southern Domain**
Alois Fürstner,* Michal D. B. Fenster,
Bernhard Fasching, CØdrickx Godbout, and
Karin Radkowski
As part of a program aimed at the discovery of new
antimitotic natural products, Andersen and co-workers
recently isolated spirastrellolide A (1; Scheme 1) from the
Scheme 1. One of the 16 possible stereostructures that might repre-
sent spirastrellolide A.
Caribbean sponge Spirastrella coccinea.[1] Unlike many other
antimitotic macrolides of marine origin, 1 does not affect
tubulin polymerization in vitro but was shown to be a very
potent (IC50 = 1 nm) and surprisingly selective inhibitor of
protein phosphatase PP2A, with the ability to drive cells
directly from the S phase into mitosis before causing cell-
cycle arrest.[1,2] In view of the central regulatory role of PP2A,
spirastrellolide A represents a potential lead for the develop-
ment of novel therapeutic agents for the treatment of cancer
as well as various neurological and metabolic disorders.[3]
[*] Prof. A. Fürstner, Dr. M. D. B. Fenster, Dipl.-Ing. B. Fasching,
Dr. C. Godbout, K. Radkowski
Max-Planck-Institut für Kohlenforschung
45470 Mülheim/Ruhr (Germany)
Fax: (+49)208-306-2994
Scheme 2. Retrosynthetic analysis of spirastrellolide based on discon-
nections into four stereoclusters of known relative configuration and
further analysis of the spiroacetal fragment C.
E-mail: fuerstner@mpi-muelheim.mpg.de
[**] Generous financial support from the MPG, the Fonds der
Chemischen Industrie, the Alexander-von-Humboldt Foundation
(fellowship to M.D.B.F.), and the Fonds de Recherche sur la Nature
et les Technologies (QuØbec; fellowship to C.G.) is gratefully
acknowledged. We thank Mrs. B. Gabor, Mrs. C. Wirtz, and Dr. R.
Mynott for their invaluable help with the structural analysis of
several key intermediates.
In view of the 16 possible combinations of these stereo-
clusters, any investigation directed toward this complex
natural product, which contains no less than 21 chiral centers,
a 38-membered macrolactone, a skipped diene, five pyranose
rings, and a tetrahydrofuran moiety which flanks a chlori-
nated [5,6,6]-bis-spiroacetal entity faces considerable chal-
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
5506
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 5506 –5510

Angewandte
Chemie
lenges.[4] Only a convergent, flexible, and modular approach
might eventually be successful; however, any such plot is
confined to disconnections at or close to the boundaries of the
individual stereoclusters as long as the internal relationship
between them has not been established.This notion is
reflected in our global retrosynthetic plan depicted in
Scheme 2, which envisages cross-coupling, metathesis[5] (or
alternative alkene formations), aldol, and esterification trans-
formations for the deconvolution of the target into fragments
A–D.Outlined below is our approach to the fully functional
southern C1–C25 domain of spirastrellolide A, whereas the
accompanying Communication reports the conquest of the
complementary northern hemisphere.[6]
was amenable to scale-up and afforded multigram quantities
of 8 in high overall yield.
The tartrate-derived aldehyde 9[10] served as a convenient
starting material for the preparation of the C17–C25(26)
segment through
a “two-directional” synthetic strategy
(Scheme 4).[11] A reagent-controlled allylation set the R con-
As indicated in Scheme 2, the preparation of the spiroke-
tal segment C envisaged a thermodynamic acetalization of a
dihydroxyketone precursor F.Although it was tempting to
have the Z alkene at C15/C16 in place at that stage, we opted
against this possibility for strategic reasons;[7] rather, we chose
to encode this future olefin as a protected carbonyl group (E),
which in turn facilitates the assembly of the carbon backbone
of the spirocyclization precursor F from segments G and H.
To this end, preparation of the required dithiane com-
menced with monoprotection of 1,3-propanediol 2 followed
by oxidation of the resulting product 3 to the corresponding
[8]
aldehyde (Scheme 3).Brown crotylation and deprotection
of the TBS group then gave the known diol 4[9] (93% ee),
which was elaborated into acetonide 5 prior to ozonolysis of
the double bond followed by reductive work-up with excess
NaBH4.The resulting alcohol 6 was converted into iodide 7,
which could be alkylated with lithio-1,3-dithiane at low
temperature.To obtain high yields, however, syringe-pump
addition of 7 was necessary to minimize elimination of HI.
Under these conditions, the route summarized in Scheme 3
Scheme 4. Preparation of the C17–C25(26) segment through a “two-
directional” synthetic strategy. Reagents and conditions: a) (+)-Ipc2B-
(allyl), Et2O, ꢀ1008C (81%); b) NaH, MeI, THF; c) BH3·THF, THF,
then H2O2; d) TBDPSCl, imidazole, CH2Cl2 (78%; 3 steps); e) 1. CSA
cat., MeOH; 2. SO3·pyridine, Et3N, DMSO, CH2Cl2; f) (+)-[(E)-crotyl]B-
(ipc)2, THF, ꢀ788C (56%; 3 steps); g) TBAF, THF; h) TESCl, imid-
azole, CH2Cl2; i) PDC, CH2Cl2, 08C!RT (80%; 3 steps). TBAF=tetra-
n-butylammonium fluoride, TBDPSCl=tert-butyldiphenylsilyl chloride,
TES=triethylsilyl, PDC=pyridinium dichromate.
figured secondary alcohol that resides at C20; of the many
procedures surveyed, we settled on the classical Brown
protocol,[12] which was the only method that reliably gave 10
as a single diastereomer in consistently high yields of up to
81% on a 20-g scale.The symmetry-related chain extension at
the C23 terminus of aldehyde 14 derived from 10 by standard
protecting-group and oxidation-state management was anal-
ogously performed by Brown crotylation,[8] which was once
more superior to alternative procedures in terms of scalability
and stereoinduction.The terminal TBDPS ether in product 15
thus formed was then cleaved with TBAF; however, all
attempts at a selective oxidation of the primary alcohol in the
resulting diol 16 remained unsuccessful.Therefore, 16 was
bis-silylated with TESCl/imidazole to give 17, which served as
the substrate for a subsequent oxidation with PDC, which
nicely remedied the selectivity issue, thus converting only the
primary TES ether into the desired aldehyde.Overall, this
unambiguous route establishes the challenging stereopentad
that extends from C20–C24 and was successful in providing
multigram amounts of building block 18.
Scheme 3. Preparation of the required dithiane unit. Reagents and
conditions: a) TBSCl, Et3N, CH2Cl2 (97%); b) SO3·pyridine, DMSO,
Et3N, CH2Cl2, ꢀ78!08C (95%); c) (+)-[(E)-crotyl]B(ipc)2, THF, ꢀ788C
(80%; 93% ee); d) TBAF, THF (94%); e) 2,2-dimethoxypropane, CSA
cat., acetone, MS 4Š (92%); f) O3, CH2Cl2/MeOH, ꢀ788C, then
NaBH4, RT (97%); g) I2, PPh3, imidazole, THF, 08C (97%); h) 1,3-
dithiane, tBuLi, ꢀ788C, THF/DMPU (10% v/v; 91%). CSA=camphor-
sulfonic acid, DMPU=1,3-dimethyl-3,4,5,6-tetrahydro-2(1H)-pyrimidi-
none, DMSO=dimethyl sulfoxide, IPC=isopinocampheyl, MS=mo-
lecular sieves, TBS=tert-butyldimethylsilyl.
Angew. Chem. Int. Ed. 2006, 45, 5506 –5510
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5507

Communications
Envisaging the use of the dithiane ring as a linchpin, 8 was
deprotonated with the aid of a mixed organometallic reagent
formed from nBuLi and (nBu)2Mg in THF.[13] The addition of
aldehyde 18 to the resulting anion followed by a Dess–Martin
oxidation[14] furnished ketone 19 (Scheme 5).Gratifyingly,
Scheme 5. Preparation of the polyfunctional fragment 21. Reagents
and conditions: a) 1. nBuLi/(nBu)2Mg (4:1), THF, RT, then aldehyde
18, ꢀ788C; 2. Dess–Martin periodinane, CH2Cl2 (69%; 2 steps);
b) PTSA(1 equiv), MeOH (87%); c) 1. TESOTf, 2,6-lutidine, CH 2Cl2;
2. DMSO, (COCl)2, Et3N, CH2Cl2, ꢀ408C (96%; 2 steps). OTf=triflate,
PTSA=p-toluenesulfonic acid.
Scheme 6. Preparation of segment D. Reagents and conditions:
a) NaH, nBuLi, then bromide 23, THF/HMPA(81%); b) [RuCl 2-
(binap)]2·NEt3 (1 mol%), HCl (2 mol%), H2 (5 bar), MeOH, 408C
(95%; 98% ee); c) O3, MeOH, then Me2S, ꢀ788C!RT; d) 1. ylide 27,
toluene, reflux; 2. CSAcat., CH 2Cl2 (78%; 3 steps; d.r.=8.5:1).
binap=(1,1’-binaphthalene)-2,2’-diylbis(diphenylphosphine), HMPA=
hexamethyl phosphoramide.
treatment of this compound with PTSA in MeOH engen-
dered cleavage of the isopropylidene acetals followed by
spontaneous formation of spiroacetal 20, which was obtained
as a single diastereomer in 87% yield of the isolated product.
Extensive 1D and 2D NMR spectroscopic investigations
unequivocally established the constitution and stereochemis-
try of this intricate product, which could be elaborated into
aldehyde 21 by the convenient persilylation/selective mono-
oxidation tandem process mentioned above.In this particular
case, recourse to Swern oxidation[15] proved optimal, thus
delivering the polyfunctional fragment 21 in virtually quanti-
tative yield.
the “ester end” of 28 would engender an erosion in the optical
purity of this building block, the enantiomeric excess of 28
was carefully checked but found to be unaltered (98% ee,
determined by HPLC).Moreover, the
cis relationship
between C3 and C7 was confirmed by NOE interaction
experiments (Scheme 6).Repeating this route with ( S)-
BINAP rather than (R)-BINAP as the ligand during the b-
ketoester reduction also provided us with multigram quanti-
ties of ent-28 in a respectable 60% yield over four readily
scalable operations.
Stereoselective aldol reactions of either enantiomer with
aldehyde 21 established the required 1,3-anti relationship
between C13 and C11, as found in the natural product
(Scheme 7).Satisfactory solutions for this critical segment
coupling were secured when isomer ent-28 was treated with 21
under modified Mukaiyama conditions[20] with BF3·Et2O as
The synthesis of the tetrahydropyran fragment D relied on
an asymmetric hydrogenation followed by an intramolecular
Michael addition to set the two stereocenters at C3 and C7
[16]
(Scheme 6).Weiler dianion alkylation
of 22 with bromide
23 was followed by a Noyori reduction of the resulting
ketoester 24,[17] which had to be performed in the presence of
a catalytic amount of HCl at moderate temperature and
hydrogen pressure (5 bar, 408C).[18,19] Under these conditions,
product 25 was obtained in an almost quantitative yield and
excellent optical purity (ꢁ 98% ee), without any hydrogena-
tion of the trisubstituted olefin or cleavage of the tert-butyl
ester interfering.
the optimal promoter.BF ·Et2O should be added slowly to
3
Ozonolysis of the alkene group in 25 and reaction of
aldehyde 26 thus formed with the stabilized ylide 27 followed
by exposure of the resulting enone to catalytic amounts of
CSA in CH2Cl2 afforded the Michael addition product 28 in
78% yield over three steps.Compound 28 was readily
separated from its C7 isomer (d.r. = 8.5:1), which could be
re-equilibrated to give a second crop of 28.Apprehensive that
a similar acid-catalyzed retro-Michael/Michael manifold at
the reaction mixture to avoid complications with the acid-
sensitive groups in both reaction partners.This reliable
protocol afforded product 29 in 62% yield as a single
isomer.In contrast, the reaction that employed the enantio-
meric ketone 28 was carried out using the boron–enolate
methodology,[21] thus delivering the diastereomeric anti-aldol
30 together with its separable C11 isomer in 94% yield of the
combined products and a diastereomeric ratio of 4:1.[22]
5508
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 5506 –5510

Angewandte
Chemie
isomers.Progress along these lines will be
reported in due course.
Received: April 26, 2006
Published online: July 5, 2006
Keywords: antitumor agents · macrolides ·
.
natural products · phosphatase inhibitors ·
total synthesis
[1] D.E. Williams, M. Roberge, R. Van Soest,
R.J.Andersen, J. Am. Chem. Soc. 2003, 125,
5296 – 5297.
[2] D.E. Williams, M. Lapawa, X. Feng, T.
Tarling, M.Roberge, R.J.Andersen,
Org.
Lett. 2004, 6, 2607 – 2610.
[3] Recent reviews on phosphatases as targets
for medicinal chemistry: a) L.Bialy, H.
Waldmann, Angew. Chem. 2005, 117, 3880 –
3906; Angew. Chem. Int. Ed. 2005, 44, 3814 –
3839; b) A.McCluskey, A.T.R.Sim, J.A.
Sakoff, J. Med. Chem. 2002, 45, 1151 – 1175;
c) R.E. Honkanen, T. Golden, Curr. Med.
Chem. 2002, 9, 2055 – 2075.
[4] For studies toward 1 reported by other
groups, 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. 2005, 7,
4125 – 4128; c) J.Liu, R.P.Hsung, Org. Lett.
2005, 7, 2273 – 2276; d) Y.Pan, KJ..
De Brabander, Synlett 2006, 853 – 856;
e) for an approach toward the originally
proposed structure, see: I.Paterson, E.A.
Anderson, S.M. Dalby, Synthesis 2005,
3225 – 3228.
Scheme 7. Preparation of two possible diastereomers of the southern hemisphere of
spirastrellolide A. Reagents and conditions: a) Cy2BCl, (iPr)2NEt, CH2Cl2, ꢀ788C (94%;
d.r.=4:1); b) TMSOTf, (iPr)2NEt, CH2Cl2, then BF3·Et2O, ꢀ788C (62%); c) Me4NBH(OAc)3,
MeCN, HOAc, ꢀ25!08C (85%); d) (CH3)2C(OMe)2, acetone, PPTS cat. (87%);
e) 1. dibal-H, CH2Cl2, ꢀ788C; 2. NaClO2, NaH2PO4, 2-methyl-2-butene, tBuOH/H2O (69%;
2 steps). Cy=cyclohexyl, dibal-H=diisobutylaluminum hydride, PPTS=pyridinium
p-toluenesulfonate, TMS=trimethylsilyl.
[5] a) T.M. Trnka, R.H. Grubbs, Acc. Chem.
Res. 2001, 34, 18 – 29; b) A.Fꢀrstner, Angew.
Chem. 2000, 112, 3140 – 3172; Angew. Chem. Int. Ed. 2000, 39,
3012 – 3043.
The major products of either series were then subjected to
a 1,3-anti reduction with Me4NHB(OAc)3 to set the proper
stereochemistry at C9 (30!31),[23] followed by acetonide
formation and conversion of the ester terminus into the
required acid function by reduction/oxidation (Scheme 7).
Not unexpectedly,[24] the dithiane moiety in 31 was concom-
itantly cleaved by the excess of NaClO2 used in the latter step.
This practical maneuver afforded product 32, which consti-
tutes the properly functionalized southern domain of 1, in
high overall yield.The equally conceivable isomer 33,
characterized by a different internal stereochemical relation-
ship between the A ring and the BC stereocluster, was formed
analogously from aldol 29.Detailed spectroscopic analyses
confirmed the structures of these advanced compounds, with
the NOE interaction patterns and the acetal resonances in the
13C NMR spectra being particularly diagnostic.[22,25]
In summary, we have devised a concise, efficient, and
scalable route to the southern C1–C25 domain of spirastrel-
lolide A (1), which was prepared in two possible stereochem-
ical formats.Together with the conquest of the northern
hemisphere reported in the accompanying Communication,[6]
we are now in a position to tackle the final assembly of this
intricate natural product with the hope of unraveling its
stereostructure by total synthesis of various conceivable
[6] A.Fꢀrstner, MD. B. .Fenster, B.Fasching, C.Godbout, K.
Radkowski, Angew. Chem. 2006, 118, 5636 – 5641; Angew. Chem.
Int. Ed. 2006, 45, 5510 – 5515.
[7] Note that a late-stage fragment coupling between the northern
and the southern domain at C25/C26 by metathesis (or other
suitable olefination reactions) is envisaged (cf.Scheme 2); a
selective hydrogenation of this newly formed alkene in the
presence of a pre-existing Z olefin at C15/C16 might raise
serious selectivity issues.
[8] H.C.Brown, K.S.Bhat, J. Am. Chem. Soc. 1986, 108, 293 – 294.
[9] UP. .Dhokte, VV. .Khau, DR. .Hutchison, MJ..Martinelli,
Tetrahedron Lett. 1998, 39, 8771 – 8774.
[10] H.Iida, N.Yamazaki, C.Kibayashi,
3337 – 3342.
J. Org. Chem. 1987, 52,
[11] a) C.S. Poss, S.L. Schreiber, Acc. Chem. Res. 1994, 27, 9 – 17;
b) for an example from our group, see: A.Fꢀrstner, M.Albert, J.
Mlynarski, M.Matheu, E.DeClercq, J. Am. Chem. Soc. 2003,
125, 13132 – 13142.
[12] a) U.S.Racherla, H.C.Brown, J. Org. Chem. 1991, 56, 401 – 404;
b) U.S.Racherla, Y.Liao, H.C.Brown,
J. Org. Chem. 1992, 57,
6614 – 6617; c) H.C. Brown, P.K. Jadhav, J. Am. Chem. Soc.
1983, 105, 2092 – 2093.
[13] a) M.Ide, M.Nakata, Bull. Chem. Soc. Jpn. 1999, 72, 2491 –
2499; b) reviews: D.Seebach, Synthesis 1969, 17 – 36; c) M. Yus,
C.Nꢁjera, F.Foubelo, Tetrahedron 2003, 59, 6147 – 6212; d) A.B.
Angew. Chem. Int. Ed. 2006, 45, 5506 –5510
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5509

Communications
Smith, S.M.Condon, J.A.McCauley, Acc. Chem. Res. 1998, 31,
35 – 46.
[14] S.D.Meyer, S.L.Schreiber, J. Org. Chem. 1994, 59, 7549 – 7552.
[15] A.J.Mancuso, D.Swern, Synthesis 1981, 165 – 185.
[16] 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.
[17] R.Noyori, T.Okhuma, M.Kitamura, H.Takaya, N.Sayo, H.
Kumobayashi, S.Akutagawa, J. Am. Chem. Soc. 1987, 109,
5856 – 5858.
[18] a) S.A.King, A.S.Thompson, A.O.King, T.R.Verhoeven,
J.
Org. Chem. 1992, 57, 6689 – 6691; b) A.Fꢀrstner, T.Dierkes,
O.R.Thiel, G.Blanda, Chem. Eur. J. 2001, 7, 5286 – 5298, and
references therein.
[19] Review: T.Ohkuma, R.Noyori in Comprehensive Asymmetric
Catalysis, Vol. 1 (Eds.: E.N.Jacobsen, A.Pfaltz, H.Yamamoto),
Springer, Berlin, 1999, pp.199 – 246.
[20] 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. 1982, 28, 203 – 331.
[21] For pertinent examples of 1,5-anti stereoinduction in boron aldol
chemistry as required in the present case, see: a) D.A.Evans, B.
Cꢂtꢃ, P.J.Coleman, B.T.Connell, J. Am. Chem. Soc. 2003, 125,
10893 – 10898; b) D.A. Evans, P.J. Coleman, B. Cotꢃ,
J. Org.
Chem. 1997, 62, 788 – 789; c) I. Paterson, K.R. Gibson, R.M.
Oballa, Tetrahedron Lett. 1996, 37, 8585 – 8588; d) for examples
using b-tetrahydropyranyl methyl ketones, see: D.A. Evans,
D.M.Fitch, T.E.Smith, V.J.Cee,
10033 – 10046; e) MT. .Crimmins, P.Siliphaivanh,
J. Am. Chem. Soc. 2000, 122,
Org. Lett.
2003, 5, 4641 – 4644; f) S.A. Kozmin, Org. Lett. 2001, 3, 755 –
758.
[22] A series of control experiments and detailed analyses of the
Mosher esters confirmed the proposed stereochemical assign-
ments; details will be reported in a forthcoming report.
[23] D.A.Evans, K.T.Chapman, E.M.Carreira,
J. Am. Chem. Soc.
1988, 110, 3560 – 3578.
[24] T.Ichige, A.Miyake, N.Kanoh, M.Nakata, Synlett 2004, 1686 –
1690.
[25] Comparison of the NMR data of the individual isomers (see the
Supporting Information) with the reported data of 1, however,
do not allow us to decide which relative configuration is present
in the natural product.
5510
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 5506 –5510