Enantiospecific formal total synthesis of the tumor and GSK-3β inhibiting alkaloid, (-)-agelastatin A

Article abstract of DOI:10.1021/ol035036l

(Matrix presented) An enantiospecific total synthesis of Weinreb's advanced intermediate 2 for (-)-agelastatin A has been achieved from the Hough-Richardson aziridine 8. Noteworthy reactions in our sequence include the highly regioselective trans-diaxial

Full text of DOI:10.1021/ol035036l

ORGANIC
LETTERS
2003
Vol. 5, No. 16
2927-2930
Enantiospecific Formal Total Synthesis
of the Tumor and GSK-3â Inhibiting
Alkaloid, (−)-Agelastatin A
,†
Karl J. Hale,* Mathias M. Domostoj,† Derek A. Tocher,† Ed Irving,§ and
Feodor Scheinmann§
The Christopher Ingold Laboratories, UniVersity College London, 20 Gordon Street,
London WC1H 0AJ, England, and The Medicinal Chemistry Department, Ultrafine,
Synergy House, Guildhall Close, Manchester Science Park,
Manchester M15 6SY, England
k.j.hale@ucl.ac.uk
Received June 9, 2003
ABSTRACT
An enantiospecific total synthesis of Weinreb’s advanced intermediate 2 for (−)-agelastatin A has been achieved from the Hough−Richardson
aziridine 8. Noteworthy reactions in our sequence include the highly regioselective trans-diaxial ring-opening of 8 with azide ion to set up the
vicinal diamido functionality present within (−)-2 and the Grubbs−Hoveyda ring-closing metathesis (RCM) reaction that was used to construct
its cyclopentene core.
In 1993, Pietra and co-workers at the University of Trento
reported their isolation of the novel oroidin alkaloid, (-)-
agelastatin A, from a collection of the axinellid sponge
Agelas dendromorpha.1 They deduced its absolute stereo-
structure through a combination of spectral methods that
included, inter alia, CD spectroscopy and a new application
of the exciton-coupling technique to diamides.1a,b
tions (IC50 ) 0.075 µg/mL) and that it inhibited the growth
of an L1210 murine tumor cell line.1c (-)-Agelastatin A was
also found to prolong the life-expectancy of mice with L1210
leukemia when repeatedly administered intraperitoneally at
doses of 2.6 mg/kg, although no antitumor effects were noted
when it was given intravenously.1c The antitumor mechanism
of (-)-agelastatin A has yet to be elucidated.
Of special interest were their observations that (-)-
agelastatin A potently inhibited the proliferation of a human
KB nasopharyngeal cancer cell line at low drug concentra-
Although (-)-agelastatin A does not inhibit casein kinase
1, CDK1/cyclin B, or CDK5/p25, it has been found to
selectively inhibit GSK-3â (glycogen synthase kinase-3â)
with an IC50 of 12 µM.2
GSK-3â hyperphosphorylates the microtubule-binding
protein, tau. Hyperphosphorylated tau proteins are major
components of the neurofibrillary tangles observed in Alzhe-
imer’s disease (AD) and are now believed to be key players
in the onset of this disease.3
† University College London.
§ Ultrafine.
(1) (a) D’Ambrosio, M.; Guerriero, A.; Debitus, C.; Ribes, O.; Pusset,
J.; Leroy, S.; Pietra, F. J. Chem. Soc., Chem. Comm. 1993, 1305. (b)
D’Ambrosio, M.; Guerriero, A.; Chiasera, G.; Pietra, F. HelV. Chim. Acta
1994, 77, 1895. (c) D’Ambrosio, M.; Guerriero, A.; Ripamonti, M.; Debitus,
C.; Waikedre, J.; Pietra, F. HelV. Chim. Acta 1996, 79, 727.
10.1021/ol035036l CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/12/2003

GSK-3â also plays a central role in the Wnt/â-catenin/
TCF cell-signaling pathway.3b,4 Wnts are secreted glycopro-
teins that bind to the Frizzled/Frizzled-2-type receptors on
the surfaces of cells to activate the protein Dishevelled (Dv1),
which inhibits GSK-3â.4 Active Wnt signaling causes the
up-regulation and accumulation of free cytosolic â-catenin,
which subsequently translocates to the nucleus where it
complexes with members of the TCF (T-cell factor) family
of transcription factors, to activate the expression of genes
involved in cell growth and proliferation. Target genes for
â-catenin/TCF-mediated transcription include include c-myc,5
cyclin D1,6,7 matrilysin,8 TCF1,9 the multidrug resistance 1
(MDR1) gene,10 c-jun, fra-1, the urokinase-type plasminogen
activator receptor,11 and osteopontin.12 Another recently
confirmed target is the peroxisome proliferator-activated
receptor (PPAR) δ-gene.13
Scheme 1. Retrosynthetic Planning for (-)-Agelastatin A
It is now well established that functionally inactivating
mutations to GSK-3â cause an accumulation of â-catenin
and that this in turn activates certain tumorigenic promoters
involved in melanoma and colon cancer such as TCF4.4 The
observation that (-)-agelastatin A can selectiVely inhibit
GSK-3â and yet still function as a powerful antitumor agent
is therefore quite remarkable and makes this a molecule of
enormous biological interest.
Importantly, this observation suggests that it might be
possible to design new and more potent small-molecule
GSK-3â inhibitors that will be highly selective and non-
tumorigenic. Inhibitors of deregulated GSK-3â activity could
potentially serve as drugs14 for treating neurodegenerative
diseases such as AD or for preventing neuronal apoptosis
after stroke. They might also function as novel insulin
mimetics, because insulin activates a protein cascade (PI3-
Kinase/PKB) that inhibits GSK-3â.14b It was with the
preparation of novel GSK-3â inhibitors in mind that we
recently embarked on an enantioselective total synthesis of
(-)-1 and its analogues.
At the outset of our studies, Weinreb’s group had already
published an elegant total synthesis of racemic agelastatin
A15 that employed a novel hetero Diels-Alder cycloaddition
reaction and a Sharpless/Kresze allylic amination16 sequence
for assembly of the cyclopentane core. Feldman and Saunders
subsequently reported an enantioselective route to (-)-
agelastatins A and B that exploited an unusual vinylcarbene
C-H insertion reaction for carbocycle construction.17
O’Brien’s group at the University of York18 have also been
active in this area, having reported a concise asymmetric
synthesis of an N-tosylated C-ring fragment that could prove
to be useful for a future total synthesis of the natural product.
(2) Meijer, L.; Thunnissen, A.-M. W. H.; White A. W.; Garnier, M.;
Nikolic, M.; Tsai, L.-H.; Walter, J.; Cleverley, K. E.; Salinas, P. C.; Wu,
Y.-Z.; Biernat, J.; Mandelkow, E.-M.; Kim, S.-H.; Pettit, G. R. Chem. Biol.
2000, 7, 51.
(3) (a) Lovestone, S.; Reynolds, C. H.; Latimer, D.; Davis, D. R.;
Anderton, B. H.; Gallo, J.-M.; Hanger, D.; Mulot, S.; Marquardt, B. Curr.
Biol. 1994, 4, 1077. (b) Harwood, A. J. Cell 2001, 105, 821.
(4) For some useful reviews and references on â-catenin/Wnt signaling,
see: (a) Morin, P. J. BioEssays 1999, 21, 1021. (b) Plakis, P. Genes DeV.
2000, 14, 1837. (c) Sharpe, C.; Lawrence, N.; Martinez Aris, A. BioEssays
2001, 23, 311. (d) Chung, D. C. Gastroenterology 2000, 119, 854.
(5) He, T. C.; Sparks, A. B.; Rago, C.; Hermeking, H.; Zawel, L.; da
Costa, L. T.; Morin, P. J.; Vogelstein, B.; Kinzler, K. W. Science 1998,
281, 1509.
In our approach to (-)-agelastatin A (Scheme 1), we hoped
to prepare one of Weinreb’s advanced intermediates (2) in
(6) Tetsu, O.; McCormick, F. Nature 1999, 398, 422.
(7) Shutman, M.; Zhurinsky, J.; Simcha, I.; Albanes, C.; D’Amico, M.;
Pestell, R.; Ben-Ze’ev, A. Proc. Nat. Acad. Sci. U.S.A. 1999, 96, 5522.
(8) (a) Crawford, H. C.; Fungleton, B. M.; Rudolph-Owen, L. A.; Goss,
K. J.; Rubinfeld, B.; Polakis, P.; Matrisian, L. M. Oncogene 1999, 18, 2883.
(b) Brabletz, T.; Jung, A.; Dag, S.; Hlubek, F.; Kirchner, T. Am. J. Pathol.
1999, 155, 1033.
(9) Roose, J.; Huls, G.; van Beest, M.; Moerer, P.; van der Horn, K.;
Goldschmeding, R.; Lotgenberg, T.; Clevers, H. Science 1999, 85, 1923.
(10) Yamada, T.; Takaoka, A. S.; Naishiro, Y.; Hayashi, R.; Maruyama,
K.; Maesawa, C.; Ochiai, A.; Hirohashi, S. Cancer Res. 2000, 60, 4761.
(11) Mann, B.; Gelos, M.; Siedow, A.; Hanski, M. L.; Gratchev, A.;
Ilyas, M.; Bodmer, W. F.; Moyer, M. P.; Rieken, E. O.; Buhr, H. J.; Hanski,
C. Proc. Natl. Acad. Sci U.S.A. 1999, 96, 1603.
(14) For some recent reviews and pertinent references on the potential
of GSK-3 inhibitors as drugs, see: (a) Martinez, A.; Castro, A.; Dorronsoro;
Alonso, M. Med. Res. ReV. 2002, 22, 373. (b) Eldar-Finkelman, H. Trends
Mol. Med. 2002, 8, 126. (c) Witherington, J.; Bordas, V.; Haigh, D.; Hickey,
D. M. B.; Ife, R. J.; Rawlings, A. D.; Slingsby, B. P.; Smith, D. G.; Ward,
R. W. Bioorg. Med. Chem. Lett. 2003, 13, 1581.
(15) (a) Stein, D.; Anderson, G. T.; Chase, C. E.; Koh, Y-h.; Weinreb,
S. M. J. Am. Chem. Soc. 1999, 121, 9574. (b) Anderson, G. T.; Chase, C.
E.; Koh, Y.-h.; Stein, D.; Weinreb, S. M. J. Org. Chem. 1998, 63, 7594.
(16) (a) Sharpless, K. B.; Hori, T. J. Org. Chem. 1976, 41, 176. (b)
Bussas, R.; Kresze, G. Liebigs Ann. Chem. 1980, 629.
(17) (a) Feldman, K. S.; Saunders, J. C. J. Am. Chem. Soc. 2002, 124,
9060. (b) Feldman, K. S.; Saunders, J. C.; Laci Wrobleski, M. J. Org. Chem.
2002, 67, 7096.
(12) El-Tanani, M. K.; Barraclough, R.; Wilkinson, M. C.; Rudland, P.
S. Oncogene 2001, 20, 1793.
(13) He, T. C.; Chan, T. A; Vogelstein, B.; Kinzler, K. W. Cell 1999,
99, 335.
(18) Baron, E.; O’Brien, P.; Towers, T. D. Tetrahedron Lett. 2002, 43,
723.
2928
Org. Lett., Vol. 5, No. 16, 2003

optically pure form and, thereafter, to use Weinreb’s synthetic
end-game to access the target itself. Weinreb’s team had
previously converted racemic 2 into (()-agelastatin A by a
six-step protocol that looked highly amenable to analogue
construction. It seemed logical therefore to capitalize on their
past successes.
Scheme 2. Synthetic Route to Oxazolidinone (-)-3a
Our strategy for accessing the enantiopure pyrroloamide
2 would selectively N-acylate the oxazolidinone 3 with
Boc2O prior to attaching the TMS-pyrrole-carboxylic acid
residue (Scheme 1). A ring-closing metathesis reaction would
be used to construct the cyclopentene ring system of 3, while
diene 4 would be assembled from the aldehyde 5 by Wittig
methylenation. A Vasella reductive ring-opening19 was
envisaged for accessing 5 from 6; the latter, in turn, would
be derived from aziridine 820 by a multistep pathway that
would involve aziridine acylation with methyl chloroformate
and Furst-Plattner ring-opening21 of the product with azide
ion. Potentially, the latter sequence could completely control
the stereochemistry of the two vicinal amido groupings in 7
and nicely set the stage for the chemistry that would follow.
With this in mind, we now present our synthesis of (-)-2 in
full.
Our route to diene 4 set off from the known aziridine 820
(Scheme 2). The latter is readily available on a large scale,
without recourse to chromatography, through the procedure
of Hough and Richardson.20 Significantly, N-acyl derivatives
of 8 show a marked preference for undergoing trans-diaxial
ring-opening reactions with strong nucleophiles to give
3-amino sugars with the R-D-altro-configuration.21 As a
result, we predicted that when the N-methylcarbamate 10
was reacted with sodium azide in hot DMF, we would obtain
11 as the major reaction product. In the event, a smooth and
highly regioselective aziridine ring-opening occurred at 140
°C using just 4 equiv of NaN3. Compound 11 was isolated
as the sole reaction product in 88% yield.
The next two steps of the synthesis were hydrogenolysis
of the azide group in 11 and protection of the crude amine
12 with trimethylsilylethylsulfonyl chloride (SES-Cl)22 and
silver cyanide in benzene at 75 °C. Significantly, this new
protocol for introducing the SES group gave significantly
higher yields of 7 (58-61% from 11) than more traditional
amine-catalyzed protocols, which were all much slower and
more problematic with respect to the occurrence of side-
reactions.
The most satisfactory means of detaching the O-ben-
zylidene acetal from 7 treated it with anhydrous HCl in
MeOH for 1.5 h. Diol 13 was thereafter selectively O-
tosylated in 79% yield with tosyl chloride, DMAP, and Et3N
(19) (a) Bernet, B.; Vasella, A. HelV. Chim. Acta. 1979, 62, 1990. (b)
Bernet, B.; Vasella, A. HelV. Chim. Acta. 1979, 62, 2400.
(20) (a) Gibbs, C. F.; Hough, L.; Richardson, A. C. Carbohydr. Res.
1965, 1, 290. (b) Ali, Y.; Richardson, A. C.; Gibbs, C. F.; Hough, L.
Carbohydr. Res. 1968, 7, 255.
(21) (a) Buss, D. H.; Hough, L.; Richardson, A. C. J. Chem. Soc. 1965,
2736. (b) Guthrie, R. D.; Murphy, D. J. Chem. Soc. 1965, 3828.
(22) We have found that SESCl is most conveniently prepared by the
following literature method: (a) Huang, J.; Widlanski, T. S. Tetrahedron
Lett. 1992, 33, 2657. Other protocols for SESCl preparation include: (b)
Weinreb, S. M.; Demko, D. M.; Lessen, T. A.; Demers, J. P. Tetrahedron
Lett. 1986, 27, 2099. (c) Weinreb, S. M.; Chase, C. E.; Wipf, P.;
Venkatraman, Org. Synth. 1997, 75, 161.
in CH2Cl2. The secondary alcohol of 14 was then O-silylated
with triethylsilyl chloride and DMAP in pyridine to obtain
15, which was converted into the iodide 6 by nucleophilic
displacement with NaI in acetone at reflux. A Vasella
Org. Lett., Vol. 5, No. 16, 2003
2929

reductive ring-opening19 on 6 (see Scheme 1) with zinc dust
in aqueous THF subsequently procured aldehyde 5 (Scheme
2).
Scheme 3. Formal Total Synthesis of (-)-Agelastatin A
To our great dismay, the Wittig methylenation of 5 with
Ph3PdCH2 was unsuccessful under all the conditions that
we studied. Tebbe and Peterson olefinations with Cp2Tid
CH2 and Me3SiCH2MgCl, respectively, also failed to olefi-
nate 5. Eventually, we found that Kocienski’s modification
of the Julia olefination protocol was effective in this
capacity;23 this reacted the anion of tetrazolyl sulfone 16 with
aldehyde 5 to produce the diene 4 directly. Although it was
not usually possible to remove all of the tetrazole by-product
from diene 4 on a preparative scale, this impurity did not
adversely affect the subsequent RCM reaction to obtain 18.
The recently developed Hoveyda-Grubbs ruthenium
alkylidene 1724 was by far the most convenient and effective
catalyst for effecting this ring-closure. It very cleanly
converted 4 into 18, notwithstanding the presence of the
urethane and sulfonamido groupings within 4, which often
deactivate RCM catalysts of earlier vintage. Yet again, it
still proved to be difficult to remove all of the tetrazole
byproduct from 18 by chromatography. Slightly impure 18
was therefore heated with K2CO3 in MeOH at reflux for 2 h
to obtain the pure oxazolidinone 3 in 36% overall yield for
the three steps from aldehyde 5.
We were then faced with the challenging issue of having
to chemoselectively N-acylate the oxazolidinone nitrogen of
3 in the presence of the sulfonamido functionality to obtain
(-)-19 (Scheme 3). Fortunately, this could be achieved
reasonably cleanly and efficiently (in 63% yield) using Boc2O
and DMAP in CH2Cl2. Racemic 19 had previously featured
as an advanced intermediate in Weinreb’s total synthesis of
1
(()-agelastatin A, and the H and 13C NMR spectra of our
optically pure (-)-19 matched those recorded by these
workers for (()-19 in the same solvent. Our enantiopure 19
also had a large negative [R]D (-88° at c 0.22 in CH2Cl2),
and its relative and absolute stereostructure were further
confirmed by X-ray crystallography (see Scheme 3).
As expected from Weinreb’s work on (()-19, the N-
acylation of (-)-19 with acid chloride 20 proceeded satis-
factorily, delivering (-)-2 in 80% yield. Our new route to
(-)-2 thus constitutes a fully stereocontrolled enantiospecific
formal total synthesis of (-)-agelastatin A.
as a host of analogues for further biological testing. The
chemical biological results of these efforts will be the subject
of future publications.
Acknowledgment. K.J.H. and M.M.D. thank the Board
of Ultrafine for generously providing the studentship for
M.M.D. and funding for chemicals. Merck, Sharpe & Dohme
(Harlow), Pfizer (Sandwich), and Novartis Pharma AG
(Basel) are also thanked for additional financial support.
Currently, we are attempting to use the Weinreb end-game
to prepare multigram quantities of (-)-agelastatin A, as well
Supporting Information Available: Full experimental
procedures and detailed spectral data of all new compounds,
copies of 500 MHz 1H and 125 MHz 13C NMR spectra, and
HRMS spectra and X-ray crystallographic analysis/data for
(-)-19. This material is available free of charge via the
Internet at http://pubs.acs.org.
(23) (a) Blakemore, P. R.; Cole, W. J.; Kocienski, P. J.; Morley, A. Synlett
1998, 26. (b) Blakemore, P. R.; Kocienski, P. J.; Morley, A.; Muir, K. J.
Chem. Soc., Perkin Trans. 1 1999, 955. (c) Baudin, J. B.; Hareau, G.; Julia,
S. A.; Ruel, O. Tetrahedron Lett. 1991, 32, 1175.
(24) (a) Kingsbury, J. S.; Harrity, J. P. A.; Bonitatebus, P. J., Jr.; Hoveyda,
A. S. J. Am. Chem. Soc. 1999, 121, 791. (b) Garber, S. B.; Kingsbury, J.
S.; Gray, B. L.; Hoveyda, A. M. J. Am. Chem. Soc. 2000, 122, 8168. (c)
Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54, 4413.
OL035036L
2930
Org. Lett., Vol. 5, No. 16, 2003