Toward the synthesis of peloruside A: Fragment synthesis and coupling studies

Article abstract of DOI:10.1021/ol034035q

(Matrix presented). The asymmetric synthesis of building blocks 3, 4, and 5, corresponding to C₁₂-C₁₉, C₇-C ₁₁, and C₁-C₆ segments of peloruside A, is reported, along with boron-mediated al

Full text of DOI:10.1021/ol034035q

ORGANIC
LETTERS
2003
Vol. 5, No. 4
599-602
Toward the Synthesis of Peloruside A:
Fragment Synthesis and Coupling
Studies
Ian Paterson,* M. Emilia Di Francesco, and Toralf Ku1hn
UniVersity Chemical Laboratory, Lensfield Road,
Cambridge CB2 1EW, United Kingdom
ip100@cus.cam.ac.uk
Received January 8, 2003
ABSTRACT
The asymmetric synthesis of building blocks 3, 4, and 5, corresponding to C −C , C −C , and C −C segments of peloruside A, is reported,
12
19
7
11
1
6
along with boron-mediated aldol coupling studies directed toward the assembly of the complete carbon skeleton of this microtubule-stabilizing
macrolide.
Peloruside A (1) is a novel cytotoxic polyketide, isolated by
Northcote and co-workers,1a from a New Zealand marine
sponge, Mycale hentscheli. Elucidation of its structure and
relative stereochemistry by extensive NMR studies revealed
a polyoxygenated 16-membered macrolide, containing a
pyranose ring, with a branched unsaturated side chain at C15.
Acting as a potent antimitotic agent, peloruside A inhibits
the growth of a range of cancer cell lines at nanomolar
concentrations.1
peloruside A to the 16-membered macrolide epothilone B
(2),3 currently in clinical trials as an anticancer drug, hints
that it may act as a Taxol/epothilone surrogate by binding
in a common site on â-tubulin.
Like paclitaxel (Taxol), recent studies1b have demonstrated
that peloruside functions by promoting tubulin polymeriza-
tion and interfering with microtubule dynamics, inducing
apoptosis following arrest of the cell cycle in the G2-M
phase. Thus, peloruside now joins an elite group of nontaxane
microtubule-stabilizing agents (including the epothilones,
discodermolide, eleutherobin, and laulimalide) that have
potential as drug candidates for the treatment of solid
tumors.2 Notably, the apparent1b structural resemblance of
(1) (a) West, L. M.; Northcote, P. T.; Battershill, C. N. J. Org. Chem.
2000, 65, 445. (b) Hood, K. A.; West, L. M.; Rouwe´, B.; Northcote, P. T.;
Berridge, M. V.; Wakefield, S. J.; Miller, J. H. Cancer Res. 2002, 62, 3356.
(c) Hood, K. A.; Ba¨ckstro¨m, B. T.; West, L. M.; Northcote, P. T.; Berridge,
M. V.; Miller, J. H. Anti-Cancer Drug Des. 2001, 16, 155.
As the current supply of peloruside A from the sponge
source is limited, an efficient total synthesis is required to
10.1021/ol034035q CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/24/2003

Scheme 1
enable further biological and preclinical evaluation, as well
via a highly stereoselective HWE homologation to give 8
as to initiate analogue chemistry. To this end, we now report
the asymmetric synthesis of three peloruside subunits (3, 4,
and 5), along with studies directed toward the assembly of
the complete carbon skeleton.
(Scheme 2). Installation of the 1,2-syn diol was then
Scheme 2a
Allowing for the uncertainty over the absolute configura-
tion,1a,4 we planned a flexible strategy (Scheme 1) to
introduce the 10 stereocenters in 1. We envisaged an
endgame based on a selective macrolactonization of a
suitable seco acid derivative such as 6, whereby, after
oxidation of the remaining C9 hydroxyl group, final depro-
tection would induce hemiacetal formation and thus generate
peloruside A. Retrosynthetic analysis of advanced intermedi-
ate 6, involving disconnections at C6-C7 and C11-C12,
revealed the three subunits 3, 4, and 5, selected as building
blocks of comparable complexity. Stereoselective aldol
couplings involving methyl ketones 3 and 5 might then be
used to assemble 6 in a convergent manner and install the
elaborate polyol sequence. To establish the correct 1,3- and
1,5-diol stereorelationships, as indicated in 6 and 7, respec-
tively, we planned to make use of our 1,5-anti aldol
methodology5,6 for implementing the key coupling steps, in
combination with 1,3-anti reductions of the resulting â-hy-
droxy ketones. The required 1,2-syn diol relationships
embedded in subunits 4 and 5 would be installed by
appropriate Sharpless asymmetric dihydroxylations,7 while
subunit 3 should be available with use of suitable aldol
methodology.
a Conditions: (a) PMBBr, NaH, Bu4NI, THF, 0 °C; (b) (COCl)2,
DMSO, CH2Cl2, -78 °C; NEt3; (c) (EtO)2P(O)CH2CO2Et, NaH,
PhMe/THF, 0 °C; (d) (DHQ)PHN, K2CO3, K3Fe(CN)6, K2OsO4,
MeSO2NH2, t-BuOH/H2O, 4 °C; (e) DDQ, 4 Å MS, CH2Cl2; (f)
CSA, 4 Å MS, CH2Cl2; (g) TBSOTf, 2,6-lutidine, CH2Cl2, 0 °C;
(h) DIBAL, CH2Cl2/hexane, -50 °C.
accomplished by catalytic Sharpless dihydroxylation meth-
odology.8 When the (E)-enoate 8 was treated with AD-mix-
R, containing the (DHQ)2PHAL ligand, the desired diol 9
was obtained in high yield, albeit in moderate enantiomeric
purity (82% ee).9 From a screening of structurally related
chiral ligands, the monomeric (DHQ)PHN10 was found to
improve the enantioselectivity of the dihydroxylation step,
providing 9 in 95% yield and with 96% ee. DDQ-mediated
First, an efficient and scaleable synthesis of the C7-C11
subunit 4 was developed by starting from neopentylglycol
(2) For recent reviews, see: (a) Altmann, K. H. Curr. Opin. Chem. Biol.
2001, 5, 424. (b) He, L. F.; Orr, G. A.; Horwitz, S. B. Drug DiscoVery
Today 2001, 6, 1153. (c) Stachel, S. J.; Biswas, K.; Danishefsky, S. J. Curr.
Pharm. Des. 2001, 7, 1277.
(3) Bollag, D. M.; McQueney, P. A.; Zhu, J.; Hensens, O.; Koupal, L.;
Liesch, J.; Goetz, M.; Lazarides, E.; Woods, C. M. Cancer Res. 1995, 55,
2325.
(4) The absolute configuration of peloruside A has not yet been
determined.
(5) (a) Paterson, I.; Gibson, K. R.; Oballa, R. M. Tetrahedron Lett. 1996,
37, 8585. (b) Paterson, I.; Collett, L. A. Tetrahedron Lett. 2001, 42, 1187.
(6) Evans, D. A.; Coleman, P. J.; Coˆte´, B. J. Org. Chem. 1997, 62, 788.
(7) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. ReV.
1994, 94, 2483.
(8) For related studies on gem-dimethyl-containing substrates, see:
Ohmori, K.; Nishiyama, S.; Yamamura, S. Tetrahedron Lett. 1995, 36, 6519.
(9) The enantiomeric purities of 9 and 14 were determined by chiral
HPLC, using the racemic diol as the reference. The absolute configurations
were determined by the advanced Mosher method (ref 18 and Supporting
Information).
(10) Sharpless, K. B.; Amberg, W.; Beller, M.; Chen, H.; Hartung, J.;
Kawanami, Y.; Lu¨bben, D.; Manoury, E.; Ogino, Y.; Shibata, T.; Ukita, T.
J. Org. Chem. 1991, 56, 4585.
600
Org. Lett., Vol. 5, No. 4, 2003

oxidative cyclization11 of this mono-PMB protected triol
resulted in an unexpected mixture of PMP acetals. The
intermediate oxocarbenium ion is trapped by either of the
two hydroxyl groups, resulting in a ca. 1:1 mixture of seven-
and six-membered cyclic acetals, 10 and 11. Advantageously,
this crude product mixture could be subjected to equilibrating
conditions (CSA, CH2Cl2) to afford solely the desired, and
thermodynamically more stable, six-membered cyclic acetal
11 in 70% overall yield. Silylation of alcohol 11 and DIBAL
reduction of the ester group then completed the synthesis of
the C7-C11 subunit 4.
desired methyl ketone 5 and the cyclic silylated acetal 16.
Upon treatment with catalytic TBAF, the latter was converted
cleanly into 5.
Next, the remaining peloruside subunit 3, containing 1,4-
related stereocenters at C15 and C18 and the trisubstituted (Z)-
alkene of the side chain, was prepared by application of our
asymmetric boron aldol methodology, making use of the (S)-
lactate-derived ketone 17 (Scheme 4).13,14 Here an aldol
Scheme 4a
The preparation of the C1-C6 methyl ketone 5 commenced
from methyl acetoacetate (Scheme 3). Formation of the
Scheme 3a
a Conditions: (a) c-Hex2BCl, Me2NEt, Et2O; HCHO, -78 °C;
MeOH, H2O2, pH 7 buffer; (b) TIPSCl, imidazole, DMAP, CH2Cl2;
(c) NaBH4, MeOH; (d) K2CO3, MeOH; (e) Pb(OAc)4, Na2CO3,
CH2Cl2, 0 °C; (f) (CF3CH2O)2P(O)CHMeCO2Me, 18-crown-6,
KHMDS, THF, -78 °C; (g) DIBAL, CH2Cl2, -40 °C; (h) DMP,
CH2Cl2; (i) (-)-Ipc2BCl, Me2CO, Et3N, Et2O, -78 °C; MeOH,
H2O2, pH 7 buffer; (j) PMBTCA, TfOH, Et2O.
reaction of 17 with formaldehyde when using c-Hex2BCl/
Me2NEt (Et2O, -78 °C) gave a separable mixture of
diastereomers, favoring the expected13,15 adduct 18 (92:8 dr,
82% yield of 18). TIPS ether formation, followed by a
sequence13a involving ketone reduction with NaBH4, benzoate
hydrolysis, and glycol cleavage with Pb(OAc)4, provided the
enantiomerically pure aldehyde 19 (88%). Using the Still-
Gennari HWE variant,16 homologation of 19 gave the desired
(Z)-enoate 20 exclusively (94%). Following conversion into
aldehyde 21, an aldol reaction with acetone required reagent
control to achieve a good level of diastereoselectivity. Thus,
(-)-Ipc2BCl/Et3N in Et2O was employed17 to give a separable
mixture (83:17 dr) from which the desired (15R)-adduct 2218
was isolated in 65% yield. Finally, PMB ether formation led
to the C12-C19 methyl ketone 3.
a Conditions: (a) (MeO)3CH, CSA, MeOH; (b) LiAlH4, THF, 0
°C; (c) PDC, 4 Å MS, CH2Cl2; (d) (MeO)2P(O)CH2CO2Me, NaH,
THF, 0 °C; (e) DIBAL, CH2Cl2, -78 °C; (f) PMBBr, NaH, THF,
0 °C; (g) (DHQ)2PYR, K2CO3, K3Fe(CN)6, K2OsO4, MeSO2NH2,
t-BuOH/H2O, 4 °C; (h) PPTS, MeOH; (i) NaH, MeI, THF, 0 °C;
(j) HClaq, CH2Cl2, 0 °C; (k) TBSOTf, 2,6-lutidine, CH2Cl2, -78
°C; (l) cat. TBAF, THF.
methyl acetal, followed by a 3-step homologation sequence,
afforded enoate 12 (10:1 E/Z; 71%). Reduction and PMB
ether formation then provided the allylic ether 13 in 85%
yield. Again, the Sharpless asymmetric dihydroxylation
reaction on 13 required an extensive screening of ligands to
enhance the enantioselectivity from 54% ee when using AD-
mix-R, containing (DHQ)2PHAL, to 92% ee in 89% yield
when using (DHQ)2PYR. Upon treatment of the resulting
diol 14 with mild acid (PPTS, MeOH), the adjacent hydroxyl
groups were differentiated by engaging one of them in
formation of a tetrahydrofuranyl acetal, while the other was
subsequently methylated (NaH, MeI) to provide 15 (92%).
Careful acid-mediated hydrolysis of the acetal,12 followed
by silylation with TBS triflate, afforded a mixture of the
(12) This reaction was accompanied by the formation of the elimination
product 2-(4-methoxybenzyloxymethyl)-5-methylfuran.
(13) (a) Paterson, I.; Wallace, D. J.; Cowden, C. J. Synthesis 1998, 639.
(b) Paterson, I.; Wallace, D. J.; Vela´zquez, S. M. Tetrahedron Lett. 1994,
35, 9083.
(14) Ketone 17 was prepared from ethyl (S)-lactate in 62% yield by an
identical 3-step sequence to that described in ref 13a for the enantiomeric
series.
(15) See the Supporting Information for a proof of stereochemistry of
aldol adduct 18.
(16) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405.
(17) (a) Paterson, I.; Goodman, J. M.; Lister, M. A.; Schumann, R. C.;
McClure, C. K.; Norcross, R. D. Tetrahedron 1990, 46, 4663. (b) Paterson,
I.; Florence, G. J. Tetrahedron Lett. 2000, 41, 6935. (c) Paterson, I.; Oballa,
R. M.; Norcross, R. D. Tetrahedron Lett. 1996, 37, 8581.
(18) The configurations of 22, 24, 26, and 28 were established by 1H
NMR analysis of the corresponding (R)- and (S)-MTPA esters, using the
advanced Mosher method, see: Kusumi, T.; Hamada, T.; Ishitsuka, M. O.;
Ohtani, I.; Kakisawa, H. J. Org. Chem. 1992, 57, 1033.
(11) Oikawa, Y.; Yoshioka, T.; Yonemitsu, O. Tetrahedron Lett. 1982,
23, 889.
Org. Lett., Vol. 5, No. 4, 2003
601

Scheme 5
With all three building blocks 3, 4, and 5 in hand, we set
between ketone 5 and aldehyde 25, triple asymmetric
induction should amplify this selectivity.17c
Having constructed the â-hydroxy ketone 24 in an efficient
manner by employing 1,5-anti stereoinduction in the aldol
coupling step, we turned to achieving a suitable reduction
to set in place the 1,3,5-triol sequence (Scheme 6). An
out to investigate the two key aldol couplings required to
assemble the carbon skeleton of peloruside (i.e. aldols #1
and #2, Scheme 1). The results are presented in Scheme 5.
With c-Hex2BCl/NEt3 the coupling of methyl ketone 3 with
the achiral aldehyde 23 proceeded, as expected,5 with an
excellent level of remote 1,5-anti induction (>95:5 dr) in
favor of the desired (11R)-adduct 24 (88%).18 Hence, it
should be possible to exploit this high level of substrate-
based induction arising from the boron enolate of 3 in
combination with a suitable C11 aldehyde derived from 4.
Scheme 6
We next sought to explore the potential influence of the
stereogenic centers contained in ketone 5 and aldehyde 25,
obtained by Dess-Martin oxidation of 4, in the planned C6-
C7 coupling (i.e. aldol #2). The c-Hex2BCl-mediated aldol
reaction between ketone 5 and isobutyraldehyde served to
confirm the anticipated role5,19 of the â-methoxy group in
securing the desired 1,5-anti stereoinduction, giving ketone
26 with moderate selectivity of 75:25 dr (Scheme 5). The
π-facial selectivity of aldehyde 25 was then evaluated in aldol
reactions with acetone. Not only did we observe low
diastereoselectivity with c-Hex2BCl (Table 1, entry 1), but
the undesired all-syn product 28 was preferred under a variety
of other conditions, particularly using the Mukaiyama
protocol (entry 3), thus indicating that 1,2-stereoinduction
follows the Felkin-Anh model, where the steric effect from
the large alkyl group overrides any electronic control from
the R-oxygen substituent in the aldehyde 25. Fortunately, it
proved possible by using (+)-Ipc2BCl17 to favor the desired
(7S)-configuration in 27 with 75:25 dr (entry 5). We
anticipate that in the (+)-Ipc2BCl-mediated aldol coupling
Evans-Tishchenko reduction20 on 24 with SmI2 and EtCHO
gave the alcohol 29 exclusively in 88% yield. This 1,3-anti
reduction differentiates the C11 and C13 hydroxyls and
provides the C9-C19 subunit of peloruside.
In summary, we have achieved highly stereoselective
syntheses of several peloruside subunits (3, 4, 5, and 29),
and established that they can be coupled together in the
desired manner. Studies toward completing a total synthesis
of peloruside A are underway.
Acknowledgment. We thank the EPSRC (GR/S19929),
EC (HPRN-CT-2000-00018), Merck Sharp and Dohme, and
DAAD (Fellowship to T.K.) for support.
(19) The reduced level of 1,5-anti induction is attributed to the bulky C2
TBS ether adversely affecting the conformation of the stereodirecting methyl
ether at C3 (for a related example, see ref 5b) and/or an opposing influence
from the C2 stereocenter. For a situation where more remote stereocenters
win out over the 1,5-effect, see: Paterson, I.; Chen, D. Y.-K.; Coster, M.
J.; Acen˜a, J. L.; Bach, J.; Gibson, K. R.; Keown, L. E.; Oballa, R. M.;
Trieselmann, T.; Wallace, D. J.; Hodgson, A. P.; Norcross, R. D. Angew.
Chem., Int. Ed. 2001, 40, 4055.
Supporting Information Available: Spectroscopic data
for new compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
(20) Evans, D. A.; Hoveyda, A. H. J. Am. Chem. Soc. 1990, 112, 6447.
OL034035Q
602
Org. Lett., Vol. 5, No. 4, 2003