Fluorous mixture synthesis of (-)-dictyostatin and three stereoisomers

Article abstract of DOI:10.1021/ol0526827

A mixture of four stereoisomers whose configurations are encoded by fluorous silyl protecting groups has been prepared and converted over 22 steps to a mixture of protected dictyostatins. Demixing by fluorous HPLC followed by removal of the fluorous prote

Full text of DOI:10.1021/ol0526827

ORGANIC
LETTERS
2006
Vol. 8, No. 2
301-304
Fluorous Mixture Synthesis of
)-Dictyostatin and Three
Stereoisomers
(−
Yoshikazu Fukui, Arndt M. Bru1ckner, Youseung Shin, Raghavan Balachandran,
Billy W. Day, and Dennis P. Curran*
Departments of Chemistry and of Pharmaceutical Sciences, UniVersity of Pittsburgh,
Pittsburgh, PennsylVania 15260
curran@pitt.edu
Received November 4, 2005
ABSTRACT
A mixture of four stereoisomers whose configurations are encoded by fluorous silyl protecting groups has been prepared and converted over
22 steps to a mixture of protected dictyostatins. Demixing by fluorous HPLC followed by removal of the fluorous protecting groups (detagging)
provides dictyostatin and three C6,C7 stereoisomers. Biological evaluation showed that the monoepimers of the natural product retained
highly potent activity.
(-)-Dictyostatin is a sponge-derived macrolactone that
exhibits potent anticancer activity (Figure 1).1 For a decade,
research moved slowly because of an incomplete (and
ultimately incorrect) stereostructure proposal and because
only tiny quantities were available. Recent total syntheses2
have confirmed the revised structure 13 and provided samples
for further biological testing.
taxoid binding site. With the recent withdrawal of discoder-
molide from clinical development,5 the importance of the
dictyostatin family increases further.
We have recently made several analogues of dictyostatin
including the moderately active C15,16 (Z)-alkene 36a and
the highly active 16-normethyl analogue 4.6b We have also
made a number of stereoisomers of dictyostatin during our
work toward its synthesis and structure confirmation.6c
With the structure of 1 secured, we hypothesized that
stereoisomers along the bottom chain, especially at C6 and
C7, would be among the most interesting to make. This is
because discodermolide7 (and, by inference, dictyostatin) is
tolerant to changes in that region of the molecule.
In vitro testing on synthetic samples shows that dictyostatin
exhibits anti-proliferative potencies, comparable or superior
to its open-chain cousin discodermolide 2.4 It is active against
paclitaxel-resistant cell lines and is one of the best micro-
tubule stabilizers known, potently competing with radiola-
beled paclitaxel, discodermolide, and epothilone B for the
(4) (a) Isbrucker, R. A.; Cummins, J.; Pomponi, S. A.; Longley, R. E.;
Wright, A. E. Biochem. Pharm. 2003, 66, 75-82. (b) Madiraju, C.; Edler,
M. C.; Hamel, E.; Raccor, B. S.; Balachandran, R.; Zhu, G.; Giuliano, K.
A.; Vogt, A.; Shin, Y.; Fournier, J.-H.; Fukui, Y.; Bru¨ckner, A. M.; Curran,
D. P.; Day, B. W. Biochemistry, 2005, 44, 15053-15063.
(1) (a) Pettit, G. R.; Cichacz, Z. A.; Gao, F.; Boyd, M. R.; Schmidt, J.
M. J. Chem. Soc., Chem. Commun. 1994, 1111-1112. (b) Pettit, G. R.;
Cichacz, Z. A. US Patent 5430053, 1995.
(2) (a) Paterson, I.; Britton, R.; Delgado, O.; Meyer, A.; Poullennec, K.
G. Angew. Chem., Int. Ed. 2004, 43, 4629-4633. (b) Shin, Y.; Fournier, J.
H.; Fukui, Y.; Bruckner, A. M.; Curran, D. P. Angew. Chem., Int. Ed. 2004,
43, 4634-4637. Related synthetic studies: (c) O’Neil, G. W.; Phillips, A.
J. Tetrahedron Lett. 2004, 45, 4253-4256. (d) Kangani, C. O.; Bru¨ckner,
A. M.; Curran, D. P. Org. Lett. 2005, 7, 379-382.
(5) Novartis AG, Annual Report to the Securities Exchange Commission,
file number 1-15024, Form 20-F, Jan 28, 2005, p 42.
(6) (a) Shin, Y.; Choy, N.; Turner, T. R.; Balachandran, R.; Madiraju,
C.; Day, B. W.; Curran, D. P. Org. Lett. 2002, 4, 4443-4446. (b) Shin, Y.;
Fournier, J. H.; Balachandran, R.; Madiraju, C.; Raccor, B. S.; Zhu, G.;
Edler, M. C.; Hamel, E.; Day, B. W.; Curran, D. P. Org. Lett. 2005, 7,
2873-2876. (c) Shin, Y. Ph.D. Thesis, University of Pittsburgh, 2005.
(3) Paterson, I.; Britton, R.; Delgado, O.; Wright, A. E. Chem. Commun.
2004, 632-633.
10.1021/ol0526827 CCC: $33.50
© 2006 American Chemical Society
Published on Web 12/23/2005

Figure 1. Structures of dictyostatin, discodermolide, and selected
analogues.
Faced with a 20-25-step synthesis to make each stereo-
isomer, we decided to use fluorous mixture synthesis8 to
prepare all four C6,C7 isomers together in a single set of
operations. The natural product dictyostatin is one of these
isomers and it serves as a control in the fluorous mixture
synthesis. At the same time, additional quantities of dic-
tyostatin were needed for further testing, and these were
provided by the synthesis. We communicate herein an
overview of this synthesis, which is by far the most ambitious
undertaking yet in the nascent field of solution phase mixture
synthesis with separation tagging.
Figure 2. Key bond dissections in the synthetic plan.
12a-d. Anti-configured alcohol (6R,7S)-9 and its (6S,7R)
enantiomer are known compounds10 that were prepared and
silylated with the indicated fluorous TIPS groups (FTIPS )
Si(iPr)2(CH2)2Rf, where Rf is perfluoroalkyl).11 This estab-
lishes the stereochemical code in 10a (Rf ) C3F7) and its
quasienantiomer12 10b (Rf ) C4F9). Standard oxidative
cleavage to give aldehydes 11a,b followed by HWE olefi-
The plan for the fluorous mixture synthesis, summarized
in Figure 2, is similar to that used for dictyostatin and its
normethyl analogue with one significant change. In the earlier
work,2b,6b we opted for a late introduction of the C23-C26
diene, but here we introduced it prior to fragment coupling
to increase convergence.2a,6c
Scheme 1. Premix Stage of the Synthesis
Accordingly, the key intermediate is M-5,9 and this is built
from a top phosphonate 6,2a a middle alkyne 7, and a bottom
Weinreb amide M-8. Fragments 6 and 7 are single com-
pounds, while M-8 is provided as a mixture of all four
isomers coded by fluorous tags.
The premix stage of the synthesis is summarized in
Scheme 1 and entails the synthesis of the four tagged esters
(7) Reviews and selected recent references: (a) Myles, D. C. In Annual
Reports in Medicinal Chem; Doherty, A. M., Ed.; Academic Press: San
Diego, CA, 2002; Vol. 37. (b) Smith, A. B.; Freeze, B. S.; LaMarche, M.
J.; Hirose, T.; Brouard, I.; Rucker, P. V.; Xian, M.; Sundermann, K. F.;
Shaw, S. J.; Burlingame, M. A.; Horwitz, S. B.; Myles, D. C. Org. Lett.
2005, 7, 311-314. (c) Smith, A. B.; Freeze, B. S.; Xian, M.; Hirose, T.
Org. Lett. 2005, 7, 1825-1828. (d) Smith, A. B.; Freeze, S. B.; LaMarche,
M. J.; Hirose, T.; Brouard, I.; Xian, M.; Sundermann, K. F.; Shaw, S. J.;
Burlingame, M. A.; Horwitz, S. B.; Myles, D. C. Org. Lett. 2005, 7, 315-
318. (e) Curran, D. P.; Furukawa, T. Org. Lett. 2002, 4, 2233-2235. (f)
Choy, N.; Shin, Y.; Nguyen, P. Q.; Curran, D. P.; Balachandran, R.;
Madiraju, C.; Day, B. W. J. Med. Chem. 2003, 46, 2846-2864. (g) Minguez,
J. M.; Kim, S.-Y.; Giuliano, K. A.; Balachandran, R.; Madiraju, C.; Day,
B. W.; Curran, D. P. Bioorg. Med. Chem. 2003, 11, 3335-3357.
(8) (a) Luo, Z. Y.; Zhang, Q. S.; Oderaotoshi, Y.; Curran, D. P. Science
2001, 291, 1766-1769. (b) Zhang, W.; Luo, Z.; Chen, C. H. T.; Curran,
D. P. J. Am. Chem. Soc. 2002, 124, 10443-10450. (c) Zhang, W. ArkiVoc
2004, 101-109. (d) Dandapani, S.; Jeske, M.; Curran, D. P. Proc. Nat.
Acad. Sci. U.S.A. 2004, 101, 12008-12012.
(9) The prefix “M” denotes a four-compound mixture.
302
Org. Lett., Vol. 8, No. 2, 2006

nation provided (6R,7S)-12a and (6S,7R)-12b.13 The syn-
configured silyl ethers were made through an Evans aldol
reaction to give both enantiomers of 1314 followed by
silylation as above to provide 14c (Rf ) C6F13) and 14d
(Rf ) C8F17). Cleavage of crude 14c,d with LiBH4 gave
alcohols 15c,d.13 These were purified and then subjected to
TEMPO/bleach oxidation and olefination to provide (6S,7S)-
12c and (6R,7R)-12d.
mixture with NaBH4 as a nonselective control reagent. The
fluorous HPLC traces of these two reaction mixtures are
compared in Figure 3.16 The chromatogram of M-18 from
The synthesis of the bottom fragment was then completed
in a mixture mode, as summarized in Scheme 2. The four
Scheme 2. Synthesis of the Bottom Fragment Mixture
Figure 3. HPLC analyses of the mixtures M-18a-d (ref 16) from
the nonselective (a) and selective (b) reductions at C9. Arrows show
the minor (â) epimers in the Noyori reduction mixture.
esters 12 were blended in a ratio of 1.5:1:1:1.5, and the
resulting mixture M-12a-d was reduced with DIBAL and
then tritylated to provide M-16a-d. Selective desilylation
of the TBS group was accomplished with HF/pyr, and the
resulting alcohol was oxidized with the Dess-Martin reagent
to furnish aldehyde M-17a-d. Further oxidation with
NaClO2 provided an acid, which was coupled with N,O-
dimethylhydroxylamine to provide M-8a-d. The products
of these mixture reactions were purified by standard flash
chromatography without demixing. At key points, intermedi-
ates were isolated by small scale demixing (1-5 mg) and
characterized by NMR spectroscopic analysis.
The fragment coupling and completion of the synthesis
are summarized in Scheme 3. The middle 7 and bottom
M-8a-d fragments were coupled via alkynyllithium addition.
The resulting alkynyl ketone was then reduced by (S,S)-
Noyori catalyst15 to give M-18a-d in 94% yield. To
determine the stereoselectivity of this catalyst-controlled
reduction, we also reduced a small sample of the ketone
the NaBH4 reduction showed four pairs of peaks of about
equal area with the larger separation between pairs corre-
sponding to the demixing based on fluorous tag and the
smaller separation among pairs corresponding to the epimeric
alcohols at C9.
In contrast, the chromatogram of the sample of M-18 from
Noyori reduction showed that the four C9-R epimers were
formed with excellent selectivity (>15:1), with only very
small peaks for the corresponding C9-â epimers (see arrows
in Figure 3b). Thus, with only two reactions and without a
preparative demixing, we learned that the Noyori reduction
exhibited high stereoselectivity that was not affected by the
substrates with assorted configurations at nearby C6 and C7.
The alkyne M-18 was reduced to the cis-alkene by Lindlar
hydrogenation and the resulting secondary hydroxy group
was protected as a TBS ether. The primary TES group was
then selectively cleaved with dichloroacetic acid to give M-19
in 90% yield. Dess-Martin oxidation of the primary alcohol
to the aldehyde was followed by HWE coupling with the
top fragment 6 to give the R,â-unsaturated ketone in 82%
yield for 2 steps. The C17-C18 alkene was selectively
reduced with Stryker’s reagent.2a,6c,17 The reduction of the
C19 ketone was performed by LiAl(Ot-Bu)3H, giving the
â-alcohol M-20â as the major product. The R-isomer was
removed by silica gel chromatography of the mixture.
(10) (a) White, J. D.; Hong, J.; Robarge, L. A. J. Org. Chem. 1999, 64,
6206-6216. (b) Theodorakis, E. A.; Drouet, K. E. Chem. Eur. J. 2000, 6,
1987-2001.
(11) The silane precursors of the FTIPS triflates and the fluorous silica
products were purchased from Fluorous Technologies, Inc. DPC holds an
equity interest in this company.
(12) Zhang, Q. S.; Curran, D. P. Chem. Eur. J. 2005, 11, 4866-4880.
(13) Note the changes in CIP priorities between 10/11, 11/12, and 14/
15.
(14) (a) Phukan, P.; Sasmal, S.; Maier, M. E. Eur. J. Org. Chem. 2003,
1733-1740.
(15) (a) Matsumura, K.; Hashiguchi, S.; Ikariya, T.; Noyori, R. J. Am.
Chem. Soc. 1997, 119, 8738-8740. (b) Haack, K.; Hashiguchi, S.; Fujii,
A.; Ikariya, T.; Noyori, R. Angew. Chem., Int. Ed. Engl. 1997, 36, 285-
287. (c) Marshall, J. A.; Bourbeau, M. J. Org. Lett. 2003, 3197-3199.
(16) Early work was done with a TBS ether on C17 and traces for these
reactions are shown in Figure 3. Similar results were obtained with the
TES ether.
(17) Mahoney, W. S.; Brestensky, D. M.; Stryker, J. M. J. Am. Chem.
Soc. 1988, 110, 291-293.
Org. Lett., Vol. 8, No. 2, 2006
303

Scheme 3. Fragment Coupling and Completion of the Synthesis
The C19 hydroxy group was protected with TBSOTf, the
This work shows that fluorous mixture synthesis can be
used to leverage effort in a complex total synthesis by
trityl group was selectively removed by ZnBr2, and the
resulting allylic alcohol was oxidized with the Dess-Martin
reagent. Still-Gennari reaction18 then provided (E),(Z)-diene
M-21 with good yield and selectivity. PMB removal with
DDQ, hydrolysis of the conjugated ester by 1N KOH, then
macrolactonization under Yamaguchi conditions19 gave a
mixture of major (2Z),(4E) and minor (2E),(4E) macrolac-
tones in 59% yield for the 2 steps.
Demixing of the final mixture was accomplished by
preparative chromatography over FluoroFlash-PF8.11 This
provided the four individual components, which were desi-
lylated with 3N HCl in MeOH to give dictyostatin (6R,7S)-1
and the other three C6,C7-epi-dictyostatin diastereomers in
the indicated yields after HPLC purification.20 The sample
of dictyostatin (6R,7S)-1 was identical to previously syn-
thesized dictyostatin,2b thereby confirming the success of the
fluorous mixture synthesis.
The antiproliferative effects of these compounds were
assayed against human ovarian carcinoma cells (Table 1).
The bis-epi diastereomer, (6S,7R)-22 was less active than
the other isomers, but still exhibited a mid-nanomolar
potency. In contrast, the monoepimers (6S,7S)-23 and
(6R,7R)-24 showed potent antiproliferative effects; (6R,7R)-
24 was about equipotent to dictyostatin (6R,7S)-1, and
(6S,7S)-23 was four times more potent. More detailed
biological characterization of the monoepimers is clearly
warranted by these exciting preliminary results.
Table 1. 50% Growth Inhibitory Concentrations (GI50) of
(C6,C7) Dictytostatin Diastereomers in 1A9/Ptx22 Human
Ovarian Carcinoma Cells
compd
GI50 (nM)
(6R,7S)-1 (dictyostatin)
(6S,7R)-22
3.4 ( 0.7
123 ( 25
(6S,7S)-23
(6R,7R)-24
0.81 ( 0.17
4.7 ( 0.6
providing more compounds without a proportional increase
in work. Here, 22 steps were conducted on fluorous mixtures,
whereas conducting the same syntheses in parallel to provide
the four target products would have required 88 steps. The
use of fluorous HPLC was critical to the success of the
project since it provided the opportunity to analyze inter-
mediate mixtures and characterize their underlying compo-
nents, as well as to demix the final mixture to provide the
pure target products.
Acknowledgment. We thank the National Institutes of
of Health (CA07839) for funding this work. Y.F. thanks
Shionogi & Co., Ltd., and A.M.B. thanks the German
Academic Exchange Organization (DAAD) for postdoctoral
fellowships.
(18) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405-4508.
(19) Inanaga, J.; Kuniko, H.; Hiroko, S.; Katsuki, T.; Yamaguchi, M.
Bull. Chem. Soc. Jpn. 1979, 52, 1989-1993.
(20) Small, variable amounts of the C2-(E), C4-(E) isomers formed during
the macrolactonization were removed at this stage. The variable yields of
the final products may reflect differences in amounts of (E,E)-isomers, but
may also reflect unequal material losses in purifications at the stage of
reduction of the C19 ketone or at other stages.
Supporting Information Available: Full experimental
details and key characterization data of new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL0526827
304
Org. Lett., Vol. 8, No. 2, 2006