Total synthesis of stephanotic acid methyl ester

Article abstract of DOI:10.1021/ol060153c

The methyl ester of the naturally occurring macrocyclic pentapeptide stephanotic acid, containing an unusual β-substituted α-amino acid with a tryptophan C-6 to leucine β-carbon link, has been synthesized. The key steps include the formation of this amino acid through a thioxo-oxazolidine intermediate and a Horner-Wadsworth-Emmons reaction using a phosphonoglycine, derived by a dirhodium(II)-catalyzed N-H insertion reaction, to give a dehydroamino acid and subsequent rhodium(I)-catalyzed asymmetric hydrogenation to introduce the modified tryptophan residue.

Full text of DOI:10.1021/ol060153c

ORGANIC
LETTERS
2006
Vol. 8, No. 10
1975-1978
Total Synthesis of Stephanotic Acid
Methyl Ester
David J. Bentley,† Alexandra M. Z. Slawin,‡ and Christopher J. Moody*,†,§
Department of Chemistry, UniVersity of Exeter, Stocker Road, Exeter, EX4 4QD, U.K.,
School of Chemistry, UniVersity of St. Andrews, Fife, Scotland, KY16 9ST, U.K., and
School of Chemistry, UniVersity of Nottingham, UniVersity Park, Nottingham,
NG7 2RD, U.K.
c.j.moody@nottingham.ac.uk
Received January 19, 2006
ABSTRACT
The methyl ester of the naturally occurring macrocyclic pentapeptide stephanotic acid, containing an unusual â-substituted r-amino acid with
a tryptophan C-6 to leucine
â
-carbon link, has been synthesized. The key steps include the formation of this amino acid through a thioxo-
oxazolidine intermediate and a Horner
−
Wadsworth Emmons reaction using a phosphonoglycine, derived by a dirhodium(II)-catalyzed N
−
−H
insertion reaction, to give a dehydroamino acid and subsequent rhodium(I)-catalyzed asymmetric hydrogenation to introduce the modified
tryptophan residue.
Natural products of the moroidin family, many of which are
potent inhibitors of tubulin polymerization, are characterized
by the presence of a highly modified tryptophan within a
macrocyclic peptide array. Thus, moroidin 1 itself (Figure
1), originally isolated from the leaves of the Australian rain
forest bush Laportea moroides, and the structure determined
by a combination of molecular modeling and detailed NMR
experiments by the Williams group in Cambridge, contains
the highly unusual direct linkages of the tryptophan C-2 and
C-6 to the imidazole N-1 of histidine and the â-carbon of a
leucine residue, respectively.1 More recently, moroidin has
been reisolated from the seeds of Celosia argentia, along
with the closely related celogentins, for example, celogentin
A 2, which share a similar structural motif based on the same
tryptophan core.2 The simplest member of this family of
cyclic peptides, stephanotic acid 3, isolated from Stephanotis
floribunda, lacks the right-hand histidine-containing ring of
moroidin and has a leucine-isoleucine substitution.3
Despite their fascinating structures, these cyclic peptides
have attracted little attention from synthetic chemists,
although over a decade ago we developed a route to simple
N-(2-indolyl)imidazoles4 and subsequently used this meth-
odology to prepare the right-hand cycle of moroidin.5 Castle
(2) Morita, H.; Shimbo, K.; Shigemori, H.; Kobayashi, J. Bioorg. Med.
Chem. Lett. 2000, 10, 469-471. Kobayashi, J.; Suzuki, H.; Shimbo, K.;
Takeya, K.; Morita, H. J. Org. Chem. 2001, 66, 6626-6633. Suzuki, H.;
Morita, H.; Iwasaki, S.; Kobayashi, J. Tetrahedron 2003, 59, 5307-5315.
Suzuki, H.; Morita, H.; Shiro, M.; Kobayashi, J. Tetrahedron 2004, 60,
2489-2495. Morita, H.; Suzuki, H.; Kobayashi, J. J. Nat. Prod. 2004, 67,
1628-1630.
(3) Yoshikawa, K.; Tao, S.; Arihara, A. J. Nat. Prod. 2000, 63, 540-
542.
(4) Comber, M. F.; Moody, C. J. Synthesis 1992, 731-733.
(5) Harrison, J. R.; Moody, C. J. Tetrahedron Lett. 2003, 44, 5189-
5191.
† University of Exeter.
‡ University of St. Andrews.
§ University of Nottingham.
(1) Leung, T.-W. C.; Williams, D. H.; Barna, J. C. J.; Foti, S.; Oelrichs,
P. B. Tetrahedron 1986, 42, 3333-3348. Kahn, S. D.; Booth, P. M.; Waltho,
J. P.; Williams, D. H. J. Org. Chem. 1989, 54, 1901-1904. Kahn, S. D.;
Booth, P. M.; Waltho, J. P.; Williams, D. H. J. Org. Chem. 2000, 65, 8406-
8406.
10.1021/ol060153c CCC: $33.50
© 2006 American Chemical Society
Published on Web 04/21/2006

use of this methodology in the first total synthesis of
stephanotic acid methyl ester 6.
Our original plan was to use the bromine functionality
incorporated in the model tryptophans 4 and 5 to introduce
a carbon side chain at C-6, for example, by a Pd-catalyzed
coupling reaction. However, this proved unsatisfactory and
therefore an alternative strategy was developed as outlined
in Figure 3. This requires a macrocyclization between the
Figure 3. Retrosynthetic analysis.
isoleucine and valine residues along with other peptide
coupling reactions, the formation of the tryptophan stereo-
center by olefination and asymmetric hydrogenation, and the
use of a thioxo-oxazolidine ring to form the unusual
â-substituted leucine residue. The formation of this sterically
hindered â-substituted R-amino acid is a key step in the
overall synthesis.8
Figure 1. Moroidin family of macrocyclic peptides.
The starting point for our synthesis was N-tert-butoxy-
carbonyl-6-isobutyrylindole 7, readily prepared from 6-cy-
anoindole by reaction with excess isopropylmagnesium
chloride followed by N-protection. Attempts to prepare a
dehydroamino acid derivative by Horner-Wadsworth-
Emmons reaction of ketone 7 with N-protected trimethyl
phosphonoglycine derivatives were unfortunately unsuccess-
ful.9 Further, although in model studies on isobutyrophenone
we had demonstrated the successful application of the
Scho¨llkopf protocol using ethyl isocyanoacetate to give an
N-formyl-protected dehydroamino acid ester,10 this was
unsuccessful when applied to the isopropyl ketone 7, and
therefore, a less direct route was adopted using the meth-
odology for the formation of hindered dehydroamino acids
developed by Hoppe some years ago.11 The method uses the
reactive reagent ethyl isothiocyanatoacetate to give a thioxo-
oxazolidine that, after N-acylation, fragments with loss of
COS upon treatment with base to give the protected
and Srikanth have reported an asymmetric synthesis of
6-(tert-butyldimethylsiloxy)methyl-2-triethylsilyltrypto-
phan as a precursor for the central core of moroidin/celo-
gentins,6 using Cook’s versatile tryptophan synthesis in which
the indole is formed by the Larock methodology but
replacing the original alkylation of a Scho¨llkopf auxiliary
with a phase-transfer-catalyzed alkylation using the chiral
catalyst developed by Park et al.
More recently, we reported an asymmetric synthesis of
the central tryptophan residue of stephanotic acid, using
dirhodium(II)-catalyzed carbene N-H insertion chemistry
in conjunction with rhodium(I)-catalyzed asymmetric hy-
drogenation of dehydro di- and tripeptides to give the core
tryptophan residues 4 and 5 (Figure 2).7 We now report the
(6) Castle, S. L.; Srikanth, G. S. C. Org. Lett. 2003, 5, 3611-3614.
(7) Bentley, D. J.; Moody, C. J. Org. Biomol. Chem. 2004, 2, 3545-
3547.
(8) For other approaches to the asymmetric synthesis of â-substituted
R-amino acids, see the following and references therein: He, L. W.;
Srikanth, G. S. C.; Castle, S. L. J. Org. Chem. 2005, 70, 8140-8147.
(9) For a recent review on dehydroamino acids, see: Bonauer, C.;
Walenzyk, T.; Ko¨nig, B. Synthesis 2006, 1-20.
(10) Scho¨llkopf, U.; Gerhart, F.; Schroder, R.; Hoppe, D. Liebigs Ann.
1972, 766, 116-129.
(11) Hoppe, D.; Follmann, R. Chem. Ber. 1976, 109, 3062-3078.
Figure 2. Previously synthesized tryptophan residues.
1976
Org. Lett., Vol. 8, No. 10, 2006

dehydroamino acid. Thus, ketone 7 was subjected to these
conditions to give a mixture of diastereomeric thioxo-
oxazolidines 8 that were converted into the separable alkenes
(E)- and (Z)-10 in a combined yield of 56% over the three
steps (Scheme 1). Attempts at asymmetric hydrogenation of
Scheme 1
Figure 4. X-ray crystal structure of (()-12.
trimethyl phosphonoglycine. However, we recently reported
a route to dehydrodi- and -tripeptides using a more complex
phosphonoglycine that already incorporated one or more
additional amino acid residues.7,15 Hence, phosphonoglycine
13 was made by dirhodium(II)-catalyzed N-H insertion of
the (presumed) carbene intermediate of trimethyl diazo-
phosphonoacetate into the amide NH of N-Boc-valinamide
(Scheme 2).
the dehydroamino acid derivative (Z)-10 using MeDuPHOS
or MeBPE rhodium(I) catalysis, developed by Burk for â,â-
disubstituted dehydroamino acids,12 gave only poor enantio-
selectivities (up to 28% ee) and yields. Therefore, a less
elegant achiral reduction, followed by a subsequent separa-
tion/resolution, was resorted to. Although the (E/Z)-alkenes
10 could be separated, it was more convenient to reduce the
mixture using magnesium in methanol13 to give a mixture
of racemic diastereomers 11 in good overall yield in a ca.
3:2 ratio. Formylation of the major isomer, with concomitant
N-Boc deprotection, was achieved using titanium(IV) chlo-
ride and dichloromethyl methyl ether14 to give the (()-amino
acid derivative 12 in modest yield. X-ray crystallography
established the relative stereochemistry as (2-S/R, 3-R/S) as
required for the natural product (Figure 4).
Scheme 2
A Horner-Wadsworth-Emmons reaction using phospho-
noglycine 13 directly onto the indole-3-carboxaldehyde 12
would have led to protecting group selectivity problems
between the methyl and ethyl esters. The racemic compound
12 was therefore subjected to hydrolysis, and the free acid
was coupled to isoleucine tert-butyl ester to give a mixture
of inseparable diastereomers 14 in good yield (Scheme 3).
The indole 14 was N-protected as its Boc derivative to
activate the aldehyde for the subsequent Horner-Wad-
sworth-Emmons reaction with phosphonoglycine 13. This
was carried out using Schmidt’s (Z)-selective DBU protocol16
and gave the dehydroamino acid 15. At this stage, the ∼1:1
mixture of diastereomers 15 was readily separable by
To generate the tryptophan stereocenter by asymmetric
hydrogenation, an appropriate dehydrotryptophan was re-
quired. Simple dehydroamino acids can be prepared by
Horner-Wadsworth-Emmons reactions of a phosphonogly-
cine such as the commercially available N-benzyloxycarbonyl
(12) Burk, M. J.; Gross, M. F.; Martinez, J. P. J. Am. Chem. Soc. 1995,
117, 9375-9376.
(13) Lee, G. H.; Youn, I. K.; Choi, E. B.; Lee, H. K.; Yon, G. H.; Yang,
H. C.; Pak, C. S. Curr. Org. Chem. 2004, 8, 1263-1287.
(14) Mattiello, L.; Fioravanti, G. Synth. Commun. 2001, 31, 2645-2648.
Poriel, C.; Ferrand, Y.; Juillard, S.; Le Maux, P.; Simonneaux, G.
Tetrahedron 2004, 60, 145-158.
(15) Buck, R. T.; Clarke, P. A.; Coe, D. M.; Drysdale, M. J.; Ferris, L.;
Haigh, D.; Moody, C. J.; Pearson, N. D.; Swann, E. Chem.-Eur. J. 2000,
6, 2160-2167.
(16) Schmidt, U.; Griesser, H.; Leitenberger, V.; Lieberknecht, A.;
Mangold, R.; Meyer, R.; Riedl, B. Synthesis 1992, 487-490.
Org. Lett., Vol. 8, No. 10, 2006
1977

Scheme 3
standard chromatography thereby completing the necessary
resolution step in the synthesis.
conditions using HATU/HOAt as coupling agents18 to give
the macrocycles D1-17 and D2-17. It was apparent by H
1
NMR spectroscopy that, although D2-17 was formed cleanly,
some epimerization had occurred during the formation of
D1-17. Each of the macrocycles D1-17 and D2-17 was then
individually subjected to deprotection by hydrogenolysis over
Pearlman’s catalyst to give the free amine that was directly
coupled to pyroglutamic acid using HATU/HOAt. The
pyroglutamic acid coupling arising from D2-17 gave a single
compound Diast-6. As expected, the pyroglutamic acid
coupling arising from D1-17 gave the diastereomeric product
6, contaminated by a further diastereomer as a result of the
epimerization referred to above. Purification by preparative
HPLC provided pure stephanotic acid methyl ester 6 whose
1H and 13C NMR spectra were identical to those provided
in the literature.3
Because we did not know which diastereomer of 15, D1,
or D2 was which at this stage, each was subjected to
asymmetric hydrogenation in methanol using Burk’s (S,S)-
EtDuPHOS Rh(I) catalytic system [(+)-(1,2-bis(2S,5S)-2,5-
diethyl phospholano)benzene(1,5-cyclooctadiene)rhodiumI-
trifluoromethanesulfonate] under 90 psi of hydrogen,17 which
gave each product 16 in excellent yield and as a single
1
diastereomer as evidenced by H NMR spectroscopy. The
stereoselectivity was confirmed by hydrogenation of each
diastereomer of 15 (D1-15 and D2-15) using an achiral
rhodium(I) catalyst [1,1′-bis(diisopropylphosphino)ferrocene-
(1,5-cyclooctadiene)rhodiumItetrafluoroborate]. Comparison
of the achiral reaction products with D1-16 and D2-16 by
HPLC on a chiral stationary phase confirmed that the asym-
metric hydrogenation had proceeded with 99% diastereo-
meric excess in each case. The stereochemistry of the new
chiral centers was assigned as (S) on the basis of previous
literature.7,15
Acknowledgment. We thank the EPSRC for support
(DTA Scheme), the EPSRC Mass Spectrometry Center at
Swansea for mass spectra, Dane Toplis for preparative HPLC
work, and Dr. Chris McErlean for helpful discussions.
Macrocyclization of each diastereomer of 16 was initiated
by simultaneous deprotection of the valine N-Boc group and
the isoleucine tert-butyl ester by treatment with trifluoroacetic
acid, with concomitant loss of the tryptophan N-Boc. The
resulting amino acids were cyclized under high dilution
Supporting Information Available: Full experimental
details for compounds 6-12 and 14-17, copies of NMR
spectra, and HPLC analysis of 16. This material is available
free of charge via the Internet at http://pubs.acs.org.
OL060153C
(17) Burk, M. J.; Feaster, J. E.; Nugent, W. A.; Harlow, R. L. J. Am.
Chem. Soc. 1993, 115, 10125-10138. Burk, M. J.; Gross, M. F.; Harper,
T. G. P.; Kalberg, C. S.; Lee, J. R.; Martinez, J. P. Pure Appl. Chem. 1996,
68, 37-44.
(18) Jou, G.; Gonzalez, I.; Albericio, F.; Lloyd Williams, P.; Giralt, E.
J. Org. Chem. 1997, 62, 354-366.
1978
Org. Lett., Vol. 8, No. 10, 2006