Discodermolide/Dictyostatin Hybrids: Synthesis and Biological Evaluation

Article abstract of DOI:10.1021/ol026942l

(Matrix Presented) Two hybrid analogues of discodermolide and dictyostatin (3, 26) have been designed and synthesized. These are the first macrocyclic analogues of discodermolide and biological activities were evaluated and compared with linear discodermolide analogues.

Full text of DOI:10.1021/ol026942l

ORGANIC
LETTERS
2002
Vol. 4, No. 25
4443-4446
Discodermolide/Dictyostatin Hybrids:
Synthesis and Biological Evaluation†
Youseung Shin, Nakyen Choy, Raghavan Balachandran, Charitha Madiraju,
Billy W. Day,* and Dennis P. Curran*
Department of Chemistry and Department of Pharmaceutical Sciences,
UniVersity of Pittsburgh, Pittsburgh, PennsylVania 15260
curran@pitt.edu
Received September 21, 2002
ABSTRACT
Two hybrid analogues of discodermolide and dictyostatin (3, 26) have been designed and synthesized. These are the first macrocyclic analogues
of discodermolide and biological activities were evaluated and compared with linear discodermolide analogues.
Discodermolide 1 is a polyketide natural product isolated
from the deep-water Caribbean sponge Discodermia dissoluta
that possesses potent antimitotic and substantial immuno-
suppressive activity (Figure 1).1 It has a mode of action
reported total syntheses of discodermolide,3 and we have
developed a flexible synthetic route to make discodermolide
analogues that lack the natural product’s C7-OH, C14-Me,
and C16-Me groups. These omissions were based on
conformational and SAR (Structure Activity Relationship)
analyses.4
The discodermolide-like macrocyclic natural product dic-
tyostatin-1 has been isolated by Pettit and assigned the
structure 2.5 It exhibits antiproliferative activity superior to
1 (ED50 0.38 nM, P388 leukemia cells), and the configura-
tions of two stereocenters at C16′ and C19′ and the absolute
configuration are as yet unassigned.5
Structures of 1 and 2 have notable similarities, although
the carbon backbone of dictyostatin 2 is two atoms longer
(1) Gunasekera, S. P.; Gunasekera, M.; Longley, R. E.; Schulte, G. K.
J. Org. Chem. 1990, 55, 4912; (correction) 1991, 56, 1346.
(2) (a) Ter Haar, E.; Kowalski, R. J.; Hamel, E.; Lin, C. M.; Longley,
R. E.; Gunasekera, S. P.; Rosenkranz, H. S.; Day, B. W. Biochemistry 1996,
35, 243. (b) Kowalski, R. J.; Giannakakou, P.; Gunasekera, S. P.; Longley,
R. E.; Day, B. W.; Hamel, E. Mol. Pharmacol. 1997, 52, 613.
(3) Paterson, I.; Florence, G. I.; Gerlach, K.; Scott, J. P.; Sereinig, N. J.
Am. Chem. Soc. 2001, 123, 9535 and references therein.
(4) (a) Hung, D. T.; Nerenberg, J. B.; Schreiber, S. L. J. Am. Chem.
Soc. 1996, 118, 11054. (b) Martello, L. A.; LaMarche, M. J.; He, L.;
Beauchamp, T. J.; Smith, A. B., III; Horwitz, S. B. Chem. Biol. 2001, 8,
843. (c) Isbrucker, R. A.; Gunasekera, S. P.; Longley, R. E. Cancer
Chemother. Pharmacol. 2001, 48, 29.
Figure 1. Structures of discodermolide and dictyostatin-1.
similar to that of paclitaxel (Taxol), arresting the cell cycle
at the M phase, but it remains potent against multidrug
resistant (MDR) carcinoma cell lines.2 Several groups have
† Dedicated to Professor Andrew S. Kende.
10.1021/ol026942l CCC: $22.00 © 2002 American Chemical Society
Published on Web 11/16/2002

and there are stereochemical and other differences. A
hypothetical condensation of the carbamate NH2 group of
discodermolide 1 with the proximate6 lactone carbonyl group
(C1) gives a 22-membered ring (not shown), which is the
same size as the ring in dictyostatin. This suggests that
macrocyclic analogues of discodermolide will be of interest,
and we report herein the synthesis of the first such analogues,
which are in effect hybrids of discodermolide and dictyosta-
tin-1.
Scheme 1
Macrocyclic hybrid 3 with an alkyl chain bridging the
lactone carbonyl group and the alkene was the initial target.
This can be synthesized convergently from three components
(4, 5, and 6) via sequential Wittig couplings7 and a
macrocyclization (Figure 2).
conditions to give 13. This product was converted to the
Weinreb amide, and the secondary hydroxyl group was
protected with p-methoxybenzyl bromide (PMBBr). DIBAL
reduction yielded an aldehyde, which was reacted with 14
in an anti-aldol fashion to yield ester 15.8 The anti relation-
ship was confirmed by comparing 13C NMR chemical shift
data with that of a reference compound.9
Scheme 2
Figure 2. Retrosynthetic analysis of hybrid 3.
Fragment 6 was synthesized from 1,10-decanediol (7, not
shown) by mono-TBS protection (NaH/TBSCl, 42%) fol-
lowed by Dess-Martin oxidation (80%). Fragment 5 was
prepared from methyl (S)-(+)-3-hydroxy-2-methylpropionate
8 (Scheme 1). The aldehyde derived from TBS protection
and DIBAL reduction of 8 was submitted to an Evans aldol
reaction with 9 to produce 10 in excellent overall yield.
Subsequently, 10 was reduced with LiBH4 and protected with
p-anisaldehyde dimethylacetal (PMP is p-methoxyphenyl).
TBAF deprotection followed by conversion of the hydroxy
group to an iodide and exposure of this to excess PPh3 in
refluxing benzene gave the phosphonium salt 5.
Fragment 4 was prepared starting from 1,4-butanediol (11,
Scheme 2). Protection with TBSCl followed by oxidation
yielded an aldehyde that was subjected to Evans syn-aldol
(5) (a) Pettit, G. R.; Linchacz, Z. A.; Gao, F.; Boyd, M. R.; Schmidt, J.
M. J. Chem. Soc., Chem. Commun. 1994, 1111. (b) The stereostructure is
taken from the following patent: Pettit, G. R.; Cichacz, Z. A. US5430053,
1995. Given its similarity to discodermolide, dictyostatin-1 may be the
enantiomer of 2.
(6) For conformational studies of discodermolide, see: (a) Smith, A. B.,
III; LaMarche, M. J.; Falcone-Hindley, M. Org. Lett. 2001, 3, 695. (b)
Monteagodo, E.; Cicero, D. O.; Cornett, B.; Myles, D. C.; Snyder, J. P. J.
Am. Chem. Soc. 2001, 123, 6929.
The secondary hydroxyl group of 15 was then protected
as a TBS ether. DIBAL reduction and subsequent oxidation
(8) Heathcock, C. H.; Pirrung, M. C.; Montgomery, S. H.; Lampe, J.
(7) Furukawa, T.; Curran, D. P. Org. Lett. 2002, 13, 2233.
Tetrahedron 1981, 37, 4087.
4444
Org. Lett., Vol. 4, No. 25, 2002

of the primary alcohol produced an aldehyde that was
subjected to Peterson olefination10 to install the required diene
16. Selective deprotection of the primary TBS group of 16
worked well in a HF-pyridine solution with excess pyridine.
The resulting alcohol was converted to an iodide, which in
turn was treated with PPh3 to give the foamy white
phosphonium salt 4.
by COSY and HMQC NMR experiments. The location of
the macrolactone ring was confirmed by HMBC NMR
experiments.
To gauge the effect of the macrocyclic system on biologi-
cal activity, we also synthesized several acyclic reference
molecules. Compounds 19, 20, and 21 were readily made
from appropriate synthetic intermediates (17 or 18) in
reasonable yields (Figure 3).
The coupling of the three fragments is summarized in
Scheme 3. Generation of the ylide from phosphonium salt 5
Scheme 3
Figure 3. Acyclic compounds for biological testina.
These analogues were tested for antiproliferative activity
in vitro against two human cancer cell lines (Table 1).12
Table 1. Human Cancer Cell Growth Inhibitory and Paclitaxel
Displacing Properties of Macrolactone Discodermolide
Analogues
GI50 (µM)
MDA-MB-231
(breast)
2008
displacement of
[3H]paclitaxel (%)c
(ovary)
3
27 ( 1a
16 ( 1a
18 ( 5
21 ( 2
17 ( 1
16 ( 3
27 ( 8
64 ( 2
19
20
21
26
1
18 ( 1a
22 ( 5a
26 ( 3a
>50a
1.4 ( 0.1b
0.016 ( 0.003b
19 ( 2a
>50a
1.0 ( 0.1b
0.072 ( 0.005b
and NaHDMS followed by addition of aldehyde 6 gave the
Wittig product in good yield (75%) provided that the reaction
was conducted at high concentration (1 M in 5). The
formation of the Z-alkene was confirmed by the 10-Hz
coupling constant between the vinyl protons.
Selective opening of the PMB acetal was accomplished
by addition of 3 equiv of DIBAL to give a primary alcohol.
This was oxidized to an aldehyde under Dess-Martin
conditions. Wittig conditions similar to those above were
then deployed to prepare 17 from this aldehyde and phos-
phonium salt 4.
Selective deprotection of 17 was achieved using HF-
pyridine and the resulting primary alcohol was oxidized to
acid 18. The other TBS group was then removed with TBAF.
Using the Yamaguchi protocol,11 the macrolactone ring was
then constructed. PMB deprotection using DDQ provided
target product 3, whose protons and carbons were assigned
a 50% growth inhibitory concentration after 48 h of continuous exposure
(mean ( standard deviation; N ) 4). b 50% growth inhibitory concentration
after 72 h of continuous exposure (mean ( standard deviation; N ) 4).
c Percent displacement by 4 µM test agent of 2 µM [3H]paclitaxel bound to
microtubules formed from 2 µM tubulin and 20 µM dideoxyGTP (N ) 6,
mean ( SD).
Macrolactone 3 and noncyclized alcohol 19 and ester 20
exhibited similar 50% growth inhibitory concentrations, in
the 15-30 µM range. Carboxylic acid 21 was inactive (>50
µM) possibly due to poor cell membrane penetration. The
modest activity of these compounds is encouraging given
the simplicity and flexibility of their lower chain. We
therefore decided to introduce the more complex bottom part
of dictyostatin-1 2 lacking only the C9′-OH group.
(9) Rychnovsky, S. D.; Rogers, B.; Yang, G. J. Org. Chem. 1993, 58,
3511.
(10) (a) Andringa, H.; Heus Kloos, Y. A.; Brandsma, L. J. Organomet.
Chem. 1987, 336, C41. (b) Paterson, I.; Schlapbach, A. Synlett. 1995, 498.
(11) Inanaga, J.; Kuniko, H.; Hiroko, S.; Katsuki, T.; Yamaguchi, M.
Bull. Chem. Soc. Jpn. 1979, 52, 1989.
Synthesis of the needed aldehyde 23 (Scheme 4) started
from the intermediate 13, which was reduced to an alcohol
(12) Wipf, P.; Reeves, J. T.; Balachandran, R.; Giuliano, K. A.; Hamel,
E.; Day, B. W. J. Am. Chem. Soc. 2000, 122, 9391.
Org. Lett., Vol. 4, No. 25, 2002
4445

Scheme 4
Scheme 5
with LiBH4, followed by PMB acetal protection. Selective
acetal opening produced alcohol 22, which was subjected
to Dess-Martin oxidation to give an aldehyde. Wittig-
Horner reaction and removal of the TBS group with HF-
pyridine gave a primary alcohol, which was oxidized to
aldehyde 23.
Center part Wittig salt 5 was reacted with aldehyde 23 to
give the (Z)-olefin (Scheme 5). This was followed by
selective PMP acetal opening with NaCNBH3-TMSCl13 to
yield a primary alcohol. The aldehyde obtained after Dess-
Martin oxidation was again subjected to Wittig reaction with
4 to generate 24. Ester 24 was reduced to the alcohol with
DIBAL, followed by Dess-Martin oxidation and application
of the Still (Z)-variant14 of the Wittig reaction to afford (E,Z)
doubly unsaturated ester 25. Selective removal of the TBS
groups was problematic; however, we learned from extensive
experimentation that exposure to 3 N HCl-MeOH in THF
(1:1) worked well. The resulting ester was hydrolyzed by
using 1 N KOH in refluxing in EtOH. Finally, the Yamaguchi
lactonization protocol followed by DDQ deprotection gave
macrolactone 26, whose structure was confirmed by HMBC
and other NMR experiments. No isomerization of either of
the dienes was detected.
Compound 26, the most functionalized of the analogues,
proved to be the most potent in terms of antiproliferative
activity against human carcinoma cells (Table 1) showing a
50% growth inhibitory concentration against breast and
ovarian cancer cells of about 1 µM. Furthermore and unlike
the other new analogues, compound 26 displaced [3H]-
paclitaxel stoichiometrically bound to microtubules15 at about
one-third the potency of 1.
In summary, we have designed and synthesized hybrids
of discodermolide and dictyostatin-1. The hybrids exhibit
moderate biological activities with the most “dictyostatin-
like” compound 26 being the most promising. These are the
first macrocyclic analogues of discodermolide, and the
synthetic route paves the way for the synthesis of other
hybrids as well as for dictyostatin-1.
Acknowledgment. Financial support was provided by the
National Institutes of Health (National Cancer Institute)
through grant CA78039. We thank Dr. Kenneth Bair of
Novartis for the gift of 1.
Supporting Information Available: Spectroscopic and
analytical data for compounds 3, 19-21, and 26. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL026942L
(13) Johansson, R.; Samuelsson, B. J. Chem. Soc., Perkin Trans. 1 1984,
2371.
(14) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405.
(15) Hamel, E.; Sackett, D. L.; Vourloumis, D.; Nicolaou, K. C.
Biochemistry 1999, 38, 5490.
4446
Org. Lett., Vol. 4, No. 25, 2002