Concise, asymmetric total synthesis of spirotryprostatin A

Article abstract of DOI:10.1021/ol0351910

(Matrix presented) The structurally intriguing cell-cycle inhibitor spirotryprostatin A has been synthesized utilizing an azomethine ylide dipolar cycloaddition reaction as the key step. This pentacyclic alkaloid contains a prenylated tryptophan-derived o

Full text of DOI:10.1021/ol0351910

ORGANIC
LETTERS
2003
Vol. 5, No. 17
3135-3137
Concise, Asymmetric Total Synthesis of
Spirotryprostatin A
Tomoyuki Onishi, Paul R. Sebahar, and Robert M. Williams*
Department of Chemistry, Colorado State UniVersity, Fort Collins, Colorado 80523
rmw@chem.colostate.edu
Received June 26, 2003
ABSTRACT
The structurally intriguing cell-cycle inhibitor spirotryprostatin A has been synthesized utilizing an azomethine ylide dipolar cycloaddition
reaction as the key step. This pentacyclic alkaloid contains a prenylated tryptophan-derived oxindole moiety that has been created in a
regiocontrolled and stereocontrolled manner in a single step.
The spirotryprostatins,1 tryprostatins,2 and cyclotryprostatins3
represent a promising class of antimitotic arrest agents.
Isolated in 1996, from Aspergillus fumigatus, spirotrypros-
tatin A (1, Figure 1) and spirotryprostatin B (2) were shown
spirotryprostatins are characterized by a unique spiro-
oxindole-substituted cis-prolyl-tryptophanyl-diketopiperazine
that is prenylated at C-18. The detailed mechanism of action
by which these substances inhibit microtubule assembly is
presently not known, and studies to discover the target of
these natural products have been hampered by the small
quantities of these substances that can be conveniently
isolated from the producing organism. Despite their relatively
modest biological activity relative to other members of this
family, the spirotryprostatins have nonetheless garnered the
most attention due to their intriguing molecular structures.
Although total syntheses of spirotryprostatin B (2) have been
reported by several groups, including us,4,5 only one total
synthesis using the classical halohydrin to oxindole spiro-
(3) (a) Cui, C. B.; Kakeya, H.; Okada, G.; Ubukata, M.; Takahashi, I.;
Isono, K.; Osada, H. J. Antibiot. 1995, 48, 1382. (b) Cui, C. B.; Kakeya,
H.; Okada, G.; Onose, R.; Osada, H. J. Antibiot. 1996, 49, 527. (c) Cui, C.
B.; Kakeya, H.; Osada, H. J. Antibiot. 1996, 49, 534.
Figure 1. Structures of spirotryprostatins A and B.
(4) (a) Sebahar, P. R.; Williams, R. M. J. Am. Chem. Soc. 2000, 122,
5666. (b) Sebahar, P. R.; Osada, H.; Usui, T., Williams, R. M. Tetrahedron
2002, 58, 6311.
(5) (a) von Nussbaum, F.; Danishefsky, S. J. Angew. Chem., Int. Ed.
2000, 39, 2175. (b) Overman, L. E.; Rosen, M. D. Angew. Chem., Int. Ed.
2000, 39, 4596. (c) Wang, H.; Ganesan, H. J. Org. Chem. 2000, 65, 4685.
(d) Bagul, T. D.; Lakshmaiah, G.; Kawabata, T.; Fuji, K. Org. Lett. 2002,
4, 249. (e) Meyers, C.; Carreira, E. M. Angew. Chem., Int. Ed. 2003, 42,
694. For a review, see: (f) Lindel, T. Nach. Chem. 2000, 48, 1498.
to completely inhibit the progression of cells at concentra-
tions greater than 253 and 34.4 µM, respectively.1 The
(1) (a) Cui, C. B.; Kakeya, H.; Osada, H. J. Antibiot. 1996, 49, 832. (b)
Cui, C. B.; Kakeya, H.; Osada, H. Tetrahedron 1996, 51, 12651. (c) Cui,
C. B.; Kakeya, H.; Osada, H. Tetrahedron 1997, 53, 59.
(2) Cui, C. B.; Kakeya, H.; Osada, H. Tetrahedron 1997, 53, 59.
10.1021/ol0351910 CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/22/2003

ring-forming contraction sequence has been reported for
spirotryprostatin A so far by Danishefsky.6 We previously
described the total synthesis of spirotryprostatin B (2) using
a stereochemically distinct three-component asymmetric
[1,3]-dipolar cycloaddition.4 Herein, we report a concise
asymmetric total synthesis of spirotryprostatin A (1).
Scheme 2. Synthesis of Spirotryprostatin Aa
In contemplating the synthesis of spirotryprostatin A (1),
it was envisioned that the core pyrrolidine ring could be
formed through an asymmetric [1,3]-dipolar cycloaddition
similar to that employed in our spirotryprostatin B synthesis.4
Spirotryprostatin A (1) differs from spirotryprostatin B (2)
in that it is saturated at C-8 and C-9 and substituted at C-6
by a methoxy group, whereas spirotryprostatin B (2) is absent
of functionality in the aromatic ring and contains the
characteristic C-8, C-9-enamide moiety. The enamide of
spirotryprostatin B was installed via a Barton-modified
Hunsdiecker reaction through an oxidative decarboxylation
of a carboethoxy group that was introduced at C-8 in the
initial dipolar cycloaddition reaction. Since it is not necessary
to install a carboalkoxy group to address the spirotryprostatin
A structure, our strategy revolved around formation of
6-methoxy-3-methylene-1,3-dihydro-indol-2-one (3, Scheme
1). If the synthesis of this dipolarophile could be ac-
Scheme 1. Azomethine Ylide Dipolar Cycloaddition
a Reaction conditions: (a) H2, Pd(OH)2, THF-MeOH (quant.);
(b) nPrCHO, AcOH, 65 °C; (c) L-Pro-OBn HCl, BOP, Et3N, MeCN;
(d) H2, Pd/C, EtOH-MeOH; (e) WSC, Et3N, MeCN; (f) p-TsOH
H2O, 3 Å sieves, toluene, 110 °C. Abbreviations: BOP )
benzotriazol-1-yloxy-tris(dimethylamino)phosphonium hexafluo-
rophosphate; WSC ) 1-[3-(dimethylamino)propyl]-3-ethylcarbo-
diimide hydrochloride (also known as EDCI).
yleneindolin-2-one react to give the corresponding cycload-
duct in 70% yield.7 However, the dipolarophile was generated
by flash vacuum pyrolysis and did not seem compatible with
the synthesis of the methoxy-substituted derivative we
required. After extensive exploration, we found that the
Peterson olefination,8 which has proven to be an efficient
method for the generation of terminal olefins, afforded a
suitable method for the formation of 3.
As shown in Scheme 1, addition of trimethylsilyl meth-
yllithium to 6-methoxy-isatin9 afforded tertiary alcohol 8 in
80% yield. The olefin 3 could be prepared in situ by the
treatment of 8 with trifluoroacetic acid at 0 °C. Compound
3 proved to be an unstable species that was not isolable due
to polymer formation upon concentration. Thus, after neu-
complished, [1,3]-dipolar cycloaddition with 4 and 5 would
generate a cycloadduct 6 that would have the correct con-
figuration at the adjacent C-3 quaternary and C-18 stereo-
genic centers. However, on the basis of the established facial
selectivity of azomethine ylide reactions derived from 4, the
R-proton (C-9) would need to be epimerized before elabora-
tion to the diketopiperazine 7 (Scheme 2). This was antic-
ipated to be a nontrivial operation since, as our spirotrypros-
tatin B synthesis had shown, the thermodynamic instability
of trans-diketopiperazines in this structural family resulted
in the facile epimerization of the prolyl stereogenic center.4
We report here a successful realization of this strategy.
Recently, Horvath and co-workers reported that azome-
thine ylides generated from silylaminonitriles and 3-meth-
(7) Bell, S. E. V.; Brown, R. F. C.; Eastwood, F. W.; Horvath, J. M.
Aust. J. Chem. 2000, 53, 183.
(6) (a) Edmonson, S. D.; Danishefsky, S. J. Angew. Chem., Int. Ed. 1998,
37, 1138. (b) Edmonson, S.; Danishefsky, S. J.; Sepp-Lorenzino, L.; Rosen,
N. J. Am. Chem. Soc. 1999, 121, 2147.
(8) (a) Chan, T.-H.; Chang, E.; Vinokur, E. Tetrahedron Lett. 1970, 11,
1137. (b) Hudrlik, P. F.; Peterson, D. Tetrahedron Lett. 1972, 13, 1785.
(9) Giovannini, E.; Portmann, P. HelV. Chim. Acta 1948, 31, 1381.
3136
Org. Lett., Vol. 5, No. 17, 2003

tralization with triethylamine, the reaction mixture was
washed with water by phase separation and crude 3 was
directly used for the cycloaddition. In the event, addition of
4 and 5 to the resultant crude mixture of 3 resulted in rapid
[1,3]-dipolar cycloaddition yielding a mixture of cycloadducts
(6 and 9). We were unable to detect the generation of
alternate regio- or diastereoisomers. In initial attempts
performed at room temperature, the ratio of products
unfortunately heavily favored the methanol elimination
product 9. Although a strategy utilizing 9 as a potential
intermediate was explored, the olefin and oxindole func-
tionalities were incompatible with the conditions required
to remove the chiral auxiliary.
1H NMR studies were then conducted to decipher at what
stage methanol was being eliminated. After mixing aldehyde
5 and 0.83 equiv of 4 for 5 min in C6D6 at room temperature,
we observed the generation of a significant amount (>50%)
of 3-methyl-2-butenal. This indicated that elimination of
methanol from 5 proceeds rapidly at room temperature. In
an attempt to obviate the elimination before the dipolar
cycloaddition reaction, the reaction was then performed at
0 °C. Under these conditions, the desired cycloadduct (6)
was obtained in 44% yield as a major product along with
20% of 9.
step sequence.4 First, 10 was directly coupled with L-proline
benzyl ester in the presence of BOP to give a diastereomeric
mixture of dipeptides that was used for the next reaction
without additional purification. Reduction of the benzyl ester
followed by WSC-mediated cyclization afforded the cis-fused
product 7 (9% from 10), which has the natural relative and
absolute configuration for the synthesis of spirotryprostatin
A, plus the trans-fused substance (12, 10% from 10), which
was converted into 9-epi-spirotryprostatin A. Finally, 7 was
subjected to treatment with TsOH in refluxing toluene to
give spirotryprostatin A (1) in 43% yield along with tertiary
alcohol 13 (31%). The spectroscopic data for the synthetic
material were in excellent agreement with that of the natural
product kindly provided to us by Dr. Hiroyuki Osada. 9-epi-
Spirotryprostatin A (14) was also prepared by subjecting 12
to the same conditions. The relative configuration of 14 was
confirmed by NOESY to be a trans-fused diketopiperazine.13
Curiously, only a trace amount (2%) of the cis-fused
substance (spirotryprostatin A, 1) was generated from 12. It
is well-known that cis-diketopiperazines are thermodynami-
cally more stable than the corresponding trans isomers for
cyclic anhydrides of proline.14
In summary, a concise asymmetric total synthesis of
spirotryprostatin A utilizing the asymmetric [1,3]-dipolar
cycloaddition reaction of methylene indolinone 3 has been
achieved. The synthesis recorded herein requires only 12
steps (7 steps in the longest linear sequence) from com-
mercially available reagents.4,9 Future efforts in this area,
directed at the preparation and biological evaluation of
synthetic analogues of the spirotryprostatins, are currently
underway and will be reported in due course.
The regiochemistry of 6 was ascertained by the doublet
1
of doublets observed in the H NMR spectrum for the
R-proton (to become C-9), and the relative configuration was
confirmed by NOESY.10 As anticipated, these data indicated
that the cycloadduct possesses the correct relative configu-
ration at C-3 and C-18 but possesses the incorrect relative
stereochemistry at position C-9 (spirotryprostatin numbering).
Amino acid 10 was cleanly produced from catalytic
hydrogenation of 6 using Pd(OH)2 as a catalyst in quantitative
yield (Scheme 2). At this juncture, we investigated the
epimerization of the R-proton of 10 in the presence of an
aldehyde and acid.11 Butyraldehyde (0.5 equiv) and 10 were
dissolved in CD3COOD, and the mixture was heated to 65
Acknowledgment. This work was supported by the
National Science Foundation (Grant CHE 0202827). We are
grateful to Ajinomoto Co. Inc., Japan, for financial support
(to T.O.). Mass spectra were obtained on instruments
supported by the National Institutes of Health Shared
Instrumentation Grant GM49631. We are grateful to Dr.
Hiroyuki Osada of the RIKEN Institute, Japan, for providing
spectra of natural spirotryprostatin A.
1
°C. It was observed by H NMR that the R-proton of the
amino acid was gradually exchanged for deuterium and that
thermodynamic epimerization had occurred.12 When acetic
acid was substituted for CD3COOD, 10 was epimerized to
give an inseparable diastereomeric mixture of amino acids
(11). Separation by PTLC was possible only after conversion
to the pentacyclic substances 12 and 7 by the following three-
Supporting Information Available: Spectroscopic data
for all new compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
OL0351910
(10) NOEs were observed between the proton at position 4 of the
oxindole and the proton at 8′a, between the proton at 6′ and the proton at
3′, and between the proton at 6′ and the proton at 4′.
(13) A NOE was observed between the proton at position 4 of the
oxindole and the olefinic proton of the 2-methyl-1-propenyl group. A NOE
was also observed between the proton at position 4 of the oxindole and the
proton at position 9.
(11) Yamada, S.; Hongo, C.; Chibata, I. Agric. Biol. Chem. 1977, 41,
2413.
(12) After conversion into the corresponding methyl ester by the treatment
with TMSCHN2, these diastereomers could be isolated by PTLC.
(14) Eguchi, C.; Kakuta, A. J. Am. Chem. Soc. 1974, 96, 3985.
Org. Lett., Vol. 5, No. 17, 2003
3137