Synthesis of the DEF-bis-spiroacetal of spirastrellolide a exploiting a double asymmetric dihydroxylation/spiroacetalisation strategy

Article abstract of DOI:10.1039/b612697a

An efficient synthesis of the C₂₆-C₄₀ tricyclic [5,6,6]-bis-spiroacetal segment of the marine macrolide spirastrellolide A has been developed, exploiting a novel double Sharpless asymmetric dihydroxylation/spiroacetalisation sequence. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b612697a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Synthesis of the DEF-bis-spiroacetal of spirastrellolide A exploiting a
double asymmetric dihydroxylation/spiroacetalisation strategy{
Ian Paterson,* Edward A. Anderson, Stephen M. Dalby, Jong Ho Lim, Philip Maltas and Christian Moessner
Received (in Cambridge, UK) 4th September 2006, Accepted 6th September 2006
First published as an Advance Article on the web 20th September 2006
DOI: 10.1039/b612697a
acetonide deprotection/spirocyclisation step. We now report the
development of an improved, second-generation approach to the
challenging northern hemisphere of spirastrellolide, leading to a
novel double Sharpless asymmetric dihydroxylation/spiroacetalisa-
tion sequence to efficiently construct the fully elaborated C26–C40
DEF subunit 4.
An efficient synthesis of the C26–C40 tricyclic [5,6,6]-bis-
spiroacetal segment of the marine macrolide spirastrellolide A
has been developed, exploiting a novel double Sharpless
asymmetric dihydroxylation/spiroacetalisation sequence.
Spirastrellolide A (1, Fig. 1) is a structurally intriguing antimitotic
macrolide isolated by Andersen and co-workers from the
Caribbean sponge Spirastrella coccinea.1,2 It induces premature
cell mitosis and displays potent (IC50 = 1 nM) and selective
inhibition of protein phosphatase 2A, and represents a potential
candidate for the development of antitumour therapeutic agents.3
Despite extensive NMR spectroscopic analysis of its methyl ester
derivative 2,2 the relationship between four stereoclusters (C3–C7,
C9–C24, C27–C38 and C46) and the absolute configuration remain
undefined, leading to 16 possible stereoisomers.
As summarised in Scheme 1, we rationalised that a late-stage
asymmetric dihydroxylation7 of the C26,27 alkene would circum-
vent the problematic elimination of spirocyclisation substrate 5,
leading to competing furan formation in producing DEF subunit
3, by removing the need for diol deprotection. Furthermore, the
selectivity of the spirocyclisation process might benefit from
reversing the direction of acetalisation from F A FED to ED A
FED. Hence, dihydroxylation of olefin 6 and DE ring formation
would be followed by silyl ether cleavage, then F ring formation, to
give the tetracyclic DEF bis-spiroacetal 3. The diketone 6 might
arise from a Horner–Wadsworth–Emmons union of aldehyde 7
with phosphonate 8.
We have adopted a flexible, modular synthetic approach
towards spirastrellolide based around bond disconnections that
isolate regions of known relative configuration. To date, we have
described the preparation of two diastereomeric C1–C25 ABC
subunits4 and the tetracyclic C26–C40 DEF subunit 3.5,6 However,
our route to the latter segment was compromised by an inefficient
The synthetic route adopted for aldehyde 7 (Scheme 2) was
improved from our initial studies, in which olefins of type 9 were
constructed through olefin cross-metathesis. Instead, deconjugative
Knoevenagel condensation of malonic acid with aldehyde 10
Fig. 1 Spirastrellolide A and targeted DEF subunits 3 and 4.
University Chemical Laboratory, Lensfield Road, Cambridge, UK
CB2 1EW. E-mail: ip100@cam.ac.uk; Fax: +44 1223 336362;
Tel: +44 1223 336407
{ Electronic supplementary information (ESI) available: Experimental
details; 1H, 13C NMR data for 4, 13, 15, 16, 22. See DOI: 10.1039/
b612697a
Scheme 1 Revised retrosynthesis towards tetracyclic DEF subunit 3.
This journal is ß The Royal Society of Chemistry 2006
4186 | Chem. Commun., 2006, 4186–4188

View Article Online
the D ring could be isolated from this reaction in moderate yield,
attributed to its lability during chromatographic purification.
Treatment of 13 with HF?py led to silyl ether cleavage and
cyclisation to a mixture of the desired DEF spiroacetal 35 and its
isomer 14, epimeric at the C35 acetal carbon, in a 3 : 1 ratio (67%).
This spirocyclisation process is believed to proceed under kinetic
control, while attempts to isomerise 14 to generate further 3 were
plagued by competing elimination reactions.
Recognising that the sense of asymmetric induction in the two
separate dihydroxylation steps was the same for both olefins, we
decided to remove the need for protecting groups on both diols. By
exploiting a late-stage double asymmetric dihydroxylation to
install the required oxygenation at C26, C27, C37 and C38 directly
(Scheme 4), we speculated that the pivotal bis-spirocyclisation to
generate the modified DEF segment 15 from diene 16 might be
accomplished under especially mild conditions. This approach
again utilises a HWE reaction with phosphonate 8 to assemble the
key dihydroxylation substrate 16.
Scheme 2 Reagents and conditions: (i) CH2(CO2H)2, piperidine (cat.),
xylenes, 136 uC; Me3SiCl, MeOH; (ii) K2OsO2(OH)4, (DHQ)2PHAL,
K3Fe(CN)6, K2CO3, t-BuOH–H2O, 0 uC; (iii) TBSCl, DMF, imid., DCM;
(iv) H2, Pd/C, EtOH; (v) Dess–Martin periodinane, NaHCO3, DCM; (vi)
(E)-dIpc2BCH2CHLCHCH3, 278 uC; (vii) TESOTf, 2,6-lutidine, DCM,
278 uC; (viii) O3, DCM, 278 uC; PPh3.
afforded the E-(b,c)-unsaturated ester 9 (79%, b,c : a,b 10 : 1).8
Sharpless asymmetric dihydroxylation7 of 9, followed by hydroxyl-
differentiating lactonisation and functional group manipulation,
then gave aldehyde 11 (62%, 99% ee). This aldehyde was subjected
to a Brown crotylation,9 TES ether formation, and ozonolysis to
afford aldehyde 7 (52%), the substrate for HWE olefination.
Coupling partner 8 was prepared from allylic chloride 125
(Scheme 3) by cleavage of the silyl ether and subsequent oxidation/
phosphonate introduction (53%). HWE coupling of phosphonate
8 with aldehyde 7 using Ba(OH)2 as base10 provided the
corresponding enone (70%), which was subjected to a 1,4-
reduction using Stryker’s reagent (96%).11 Following TES ether
deprotection, Dess–Martin oxidation of the resulting crude
hydroxy ketone (presumably in equilibrium with the hemiacetal)
provided 1,5-diketone 6 (68%).12 This diketone was subjected to
Sharpless asymmetric dihydroxylation using our optimised condi-
tions with the (DHQ)2PYR ligand.5 Hemiacetal 13 incorporating
The aldehyde 17 required for HWE union was assembled from
the previously prepared methyl ester 9 (Scheme 5) using our lactate
aldol chemistry.13–15 Oxidation state adjustment of 9 delivered
aldehyde 18 (93%), which underwent a highly selective (. 20 : 1 dr)
boron aldol reaction with ethyl ketone 19 to provide the anti
adduct 20 (86%).14 TBS etherification of 20 and cleavage of the
lactate auxiliary then gave aldehyde 17 (86%). Once again, the
Ba(OH)2 protocol10 proved the method of choice for the HWE
olefination with phosphonate 8, and following a similar sequence
Scheme 4 The double AD strategy for tricyclic DEF subunit 15.
Scheme 5 Reagents and conditions: (i) LiAlH4, THF; (ii) Dess–Martin
periodinane, DCM; (iii) 19, c-Hex2BCl, Et3N, Et2O, 0 uC, then 18, 278 uC;
(iv) TBSOTf, 2,6-lutidine, DCM; (v) NaBH4, MeOH then K2CO3; NaIO4,
THF–pH 7 buffer; (vi) Ba(OH)2, THF–H2O (40 : 1); (vii) [(PPh3)CuH]6,
PhMe–H2O (200 : 1); (viii) MeCN–HCl; (ix) Dess–Martin periodinane,
DCM.
Scheme 3 Reagents and conditions: (i) CSA (cat.), MeOH; (ii) Dess–
Martin periodinane, NaHCO3, DCM; (iii) (MeO)2POCH2Li, THF,
278 uC; Dess–Martin periodinane, DCM; (iv) Ba(OH)2, THF–H2O
(40 : 1); (v) [(PPh3)CuH]6, PhMe–H2O (200 : 1); (vi) PPTS, DCM–MeOH
(3 : 1); (vii) Dess–Martin periodinane, DCM; (viii) K2OsO2(OH)4,
(DHQ)2PYR, K3Fe(CN)6, K2CO3, t-BuOH–H2O, 0 uC; (ix) HF?py, THF.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 4186–4188 | 4187

View Article Online
DEF fragment 3, a close match of chemical shift data was found
between 4 and 15 and the spirastrellolide methyl ester (2),2 with all
substituents on the DE-ring spiroacetal equatorial. Additionally, a
diagnostic NOE enhancement was observed between H27 and H38,
consistent with the bis-spiroacetal possessing the doubly anomeric
effect-stabilised configuration at C31 and C35.
In conclusion, we have developed an improved approach to the
construction of the challenging C26–C40 DEF region of the marine
macrolide spirastrellolide, employing a novel double Sharpless
dihydroxylation/spiroacetalisation sequence, as in 16 A 15. This
eliminates the need for protecting groups and enhances the supply
of the northern hemisphere, enabling the initiation of fragment
coupling studies to unravel the remaining stereochemical ambi-
guities and further advance the total synthesis of this potent
bioactive polyketide.
We thank the EPSRC (EP/C541677/1), Merck Research
Laboratories, Homerton College, Cambridge (Research
Fellowship to E.A.A.), SK Corporation (J.H.L.) and the
German Academic Exchange Service (DAAD, Postdoctoral
Fellowship to C.M.) for support and Professor Raymond
Andersen (University of British Columbia) for helpful discussions.
Scheme 6 Reagents and conditions: (i) K2OsO2(OH)4, (DHQ)2PYR,
MeSO2NH2, K3Fe(CN)6, K2CO3, t-BuOH–H2O, 0 uC; (ii) PPTS (cat.),
DCM–MeOH (1 : 1); (iii) TESOTf, 2,6-lutidine, DCM, 278 uC; (iv) PPTS
(cat.), DCM–MeOH (6 : 1), 0 uC.
Notes and references
of reactions to those already described, 1,5-diketone 16 was
obtained on a gram scale in 47% yield.
1 K.-S. Yeung and I. Paterson, Chem. Rev., 2005, 105, 4237;
R. D. Norcross and I. Paterson, Chem. Rev., 1995, 95, 2041;
I. Paterson and E. A. Anderson, Science, 2005, 310, 451.
2 D. E. Williams, M. Roberge, R. Van Soest and R. J. Andersen, J. Am.
Chem. Soc., 2003, 125, 5296; D. E. Williams, M. Lapawa, X. Feng,
T. Tarling, M. Roberge and R. J. Andersen, Org. Lett., 2004, 6, 2607.
3 L. H. Le, C. Erlichman, L. Pillon, J. J. Thiessen, A. Day, N. Wainman,
E. A. Eisenhauer and M. J. Moore, Invest. New Drugs, 2004, 22, 159;
R. E. Honkanen and T. Golden, Curr. Med. Chem., 2002, 9, 2055.
4 I. Paterson, E. A. Anderson, S. M. Dalby and O. Loiseleur, Org. Lett.,
2005, 7, 4125.
We were now ready to perform the crucial double asymmetric
dihydroxylation reaction. Our original route5 utilised the
(DHQ)2PHAL ligand for the internal olefin, whilst the
(DHQ)2PYR ligand was optimal for the terminal olefin. Guided
by this and other precedents,7 we opted for the latter conditions
(Scheme 6). In the event, exposure of diene 16 to Sharpless
conditions led to rapid (1 h) dihydroxylation of the internal olefin,
followed by reaction at the terminal double bond (3 h), where
NMR analysis indicated the formation of a complex mixture of
isomeric products 21. As purification did not prove fruitful,
the crude reaction mixture was exposed to PPTS in DCM–MeOH
(1 : 1), leading to cyclisation to give the targeted DEF subunit 15,
together with other isomers.
5 I. Paterson, E. A. Anderson, S. M. Dalby and O. Loiseleur, Org. Lett.,
2005, 7, 4121.
6 For other synthetic studies, see: J. Liu and R. P. Hsung, Org. Lett.,
2005, 7, 2273; I. Paterson, E. A. Anderson and S. M. Dalby, Synthesis,
2006, 3225; Y. Pan and J. K. De Brabander, Synlett, 2006, 853; C. Wang
and C. J. Forsyth, Org. Lett., 2006, 8, 2997; A. Fu¨rstner, M. D. B.
Fenster, B. Fasching, C. Godbout and K. Radkowski, Angew. Chem.,
Int. Ed., 2006, 45, 5506; A. Fu¨rstner, M. D. B. Fenster, B. Fasching,
C. Godbout and K. Radkowski, Angew. Chem., Int. Ed., 2006, 45, 5510;
J. Liu, J. H. Yang, C. Ko and R. P. Hsung, Tetrahedron Lett., 2006, 47,
6121.
7 H. C. Kolb, M. S. VanNieuwenhze and K. B. Sharpless, Chem. Rev.,
1994, 94, 2483.
8 C. Garc´ıa, T. Mart´ın and V. S. Mart´ın, J. Org. Chem., 2001, 66, 1420;
N. Ragoussis, Tetrahedron Lett., 1987, 28, 93.
9 H. C. Brown and K. S. Bhat, J. Am. Chem. Soc., 1986, 108, 293.
10 I. Paterson, K.-S. Yeung and J. B. Smaill, Synlett, 1993, 774.
11 W. S. Mahoney, D. M. Brestensky and J. M. Stryker, J. Am. Chem.
Soc., 1988, 110, 291.
12 Varying amounts of a dihydropyran side-product i were isolated from
this oxidation; this could be readily recycled to the oxidation substrate
upon exposure to PPTS in THF–water.
The use of PPTS for this equilibration/cyclisation process was
found to be essential, as stronger acids (CSA, HCl) led to
competitive formation of furan-containing by-products. To over-
come material losses in the chromatographic isolation of 15,
protection of the crude mixture of isomeric DEF diols was carried
out (TESOTf, 2,6-lutidine). The resulting silylated products now
proved more stable to purification and the desired DEF bis-
spiroacetal 22 was readily separated from other isomers.
Pleasingly, these other isomers could be recycled via re-exposure
to PPTS in DCM–MeOH (1 : 1), which cleaved both TES ethers
and effected spiroacetal equilibration without competing elimina-
tion side-reactions. Subsequent reprotection afforded more of the
targeted DEF tricycle 22, giving a 65% yield overall from 16 after
one recycle. Finally, to enable investigation of fragment union with
the southern hemisphere,4 the primary TES ether was selectively
cleaved using PPTS (DCM–MeOH, 6 : 1, 0 uC) leading to the fully
functionalised DEF subunit 4 (93%).
13 I. Paterson, D. J. Wallace and C. J. Cowden, Synthesis, 1998, 689.
14 I. Paterson, D. J. Wallace and S. M. Vela´zquez, Tetrahedron Lett., 1994,
35, 9083.
The stereochemistry of the newly formed bis-spiroacetals was
confirmed using 1H NMR spectroscopy. As with our previous
15 I. Paterson and D. J. Wallace, Tetrahedron Lett., 1994, 35, 9087.
4188 | Chem. Commun., 2006, 4186–4188
This journal is ß The Royal Society of Chemistry 2006