Total synthesis of the originally proposed and revised structures of scleritodermin A

Article abstract of DOI:10.1021/ol801419m

(Chemical Equation Presented) The first total synthesis of the originally proposed and revised structure of scleritodermin A, along with an isomer, were achieved by the use of an α-azido carboxyl group serving as the key α-ketoamide precursor, thus leading to a revision of the structure originally proposed for natural scleritodermin A.

Full text of DOI:10.1021/ol801419m

ORGANIC
LETTERS
2008
Vol. 10, No. 17
3765-3768
Total Synthesis of the Originally
Proposed and Revised Structures of
Scleritodermin A
Sheng Liu, Yong-Mei Cui, and Fa-Jun Nan*
Chinese National Center for Drug Screening, State Key Laboratory of Drug Research
Shanghai Institute of Materia Medica, Chinese Academy of Sciences 189 Guoshoujing
Road, Zhangjiang Hi-Tech Park, Shanghai 201203, China
fjnan@mail.shcnc.ac.cn
Received June 23, 2008
ABSTRACT
The first total synthesis of the originally proposed and revised structure of scleritodermin A, along with an isomer, were achieved by the use
of an r-azido carboxyl group serving as the key r-ketoamide precursor, thus leading to a revision of the structure originally proposed for
natural scleritodermin A.
Scleritodermin A is a cyclic depsipeptide isolated from the
lithistid sponge Scleritoderma nodosum.1 Significant in vitro
cytotoxicity of this peptide was demonstrated in human tumor
cell lines. Scleritodermin A (Figure 1), a highly modified
peptide, has an unusual N-sulfated side chain and a novel
conjugated thiazole moiety, as well as an R-ketoamide group.
Intrigued by the architecture of this molecule, we initiated a
research program aimed at the total synthesis of scleritoder-
min A, as well as the modification of its structure for fine-
tuning of the biological activity of this peptide. The synthesis
(1) Schmidt, E. W.; Raventos-Suarez, C.; Bifano, M.; Menendez, A. T.;
Fairchild, C. R.; Faulkner, D. J. J. Nat. Prod. 2004, 67, 475.
Figure 1. Originally proposed and revised structures of sclerito-
(2) (a) Hagihara, M; Schreiber, S. L. J. Am. Chem. Soc. 1992, 114, 6570.
(b) Wipf, P.; Kim, H. J. Org. Chem. 1993, 58, 5592. (c) Deng, J.; Hamada,
Y.; Shioiri, T.; Matsunaga, S.; Fusetani, N. Angew. Chem., Int. Ed. Engl.
1994, 33, 1729. (d) Maryanoff, B. E.; Greco, M. N.; Zhang, H. C.; Andrade-
Gordon, P.; Kauffman, J. A.; Nicolaou, K. C.; Liu, A.; Brungs, P. H. J. Am.
Chem. Soc. 1995, 117, 1225. (e) Bastiaans, H. M. M.; Van der Baan, J.;
Ottenheijm, H. C. J. J. Org. Chem. 1997, 62, 3880. (f) Sowinski, J. A.;
Toogood, P. L. Chem. Commun. 1999, 981. (g) Wasserman, H. H.; Zhang,
R. Tetrahedron 2002, 58, 6277.
dermin A and its isomers.
of naturally occurring marine metabolites bearing R-ketoa-
mide units such as cyclotheonamides and keramamides has
been extensively studied since the early 1990s.2 Here, we
10.1021/ol801419m CCC: $40.75
Published on Web 08/08/2008
2008 American Chemical Society

report a new strategy for assembling these macrocyclic
R-keto lactams by the use of an R-azido carboxyl group as
the key intermediate and the first total synthesis of the
originally proposed and revised structures of scleritoder-
min A.
Scheme 2. Synthesis of Geometric Conjugated Thiazole Acids
Figure 2 depicts our retrosynthetic analysis. The unique
features of our synthesis plan involved the addition of an
A,5 was subjected to Horner-Wadsworth-Emmons olefi-
nation according to the Still-Gennari procedure,6 followed
by saponification to yield the (2Z,4E)-conjugated thiazole
acid 8.
The coupling of fragments 8 and 9 was accomplished using
Yamaguchi’s protocol7 (Scheme 3). The selective removal
Scheme 3. Fragment Coupling and Initial Synthesis
Figure 2. Retrosynthetic analysis of target molecule 1.
azide group to the cyclic backbone of the target molecule
and the application of a biomimic transamination reaction
for the formation of the key R-keto lactam unit. Therefore,
molecule 1 was divided into two parts: the macrocyclic core
5 and the dipeptide side chain 6. Further disconnections
provided three synthetic precursors: R-azido ꢀ-amino acid
7, (2Z,4E)-conjugated thiazole moiety 8, and dipeptide 9.
Our synthesis process started with the preparation of
fragments 7 and 8. Ozonolysis of (R)-10 in MeOH using
Wasserman’s procedure3 yielded an R-keto ester, which was
subsequently converted into the desired alcohol (two isomers,
6:1) after reduction with Zn(BH4)2. The subsequent applica-
tion of the standard Mitsunobu reaction,4 followed by a
saponification step, resulted in the production of an amino
acid (fragment 7) (Scheme 1). In a parallel procedure
of the Boc group with HCl proceeded smoothly in the
presence of the acid-labile tert-butyl ester.8 The resulting
amide was coupled with fragment 7 to produce the cycliza-
tion precursor, fragment 15 (two isomers, 2.6:1). Subse-
quently, the simultaneous unmasking of the amine and acid
groups by treatment with TFA allowed the cyclization to be
effected via activation with PyBop and (iPr)2NEt at high
dilution in CH2Cl2 (54% yield). The Alloc protecting group
Scheme 1. Synthesis of Azido-Bearing Fragments
(Scheme 2), the aldehyde 12, which was previously reported
by Serra et al. in their attempt at synthesizing scleritodermin
(5) Sellanes, D.; Manta, E.; Serra, G. Tetrahedron Lett. 2007, 48, 1827.
(6) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405.
(7) Inanaga, J; Hirata, K; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull.
Chem. Soc. Jpn. 1979, 52, 1989.
(3) Wasserman, H. H.; Ho, W. B. J. Org. Chem. 1994, 59, 4364.
(4) Mitsunobu, O.; Yamada, M. Bull. Chem. Soc. Jpn. 1967, 40, 2380.
(8) Gibson, F. S.; Bergmeier, S. C.; Rapoport, H. J. Org. Chem. 1994,
59, 3216.
3766
Org. Lett., Vol. 10, No. 17, 2008

was removed with Pd(PPh3)4,9 permitting the incorporation
of the side chain 6 and forming fragment 4. Several attempts
to chemoselectively reduce the azide compound 4 to the
corresponding amine via a Staudinger reaction using triph-
enylphosphine or trimethylphosphine were problematic, due
to difficulties in hydrolyzing these stable phosphazine ylides.
However, we were able to circumvent this problem by
hydrogenating the azide group using Lindlar’s catalyst; the
conjugated moiety was better tolerated under this condition.
With this R-amine amide in hand, the stage was now set for
the critical transamination reaction. In Corey’s work toward
the synthesis of Et-743,10 a pyridoxal mimic-mediated
oxidative deamination was used; however a strong base such
as DBU was needed. In our study we found that mild
conditions11 using glyoxylate as an electrophile and cop-
per(II) ions as catalysts in an aqueous acetate buffer at pH
6.0 were more suitable and led to the production of
compound 16 as a single diastereomer (51% yield for both
steps).
required trans double bond was generated from aldehyde
12 (Scheme 2). It seemed reasonable to us that continuing
the synthesis following the established procedure for com-
pound 1 with (R)- or (S)-11 derived from L-allo-Ile or L-Ile
would deliver both 2 and 3. However, our attempts to
implement macrocyclization at the previous position (be-
tween Ile and Pro) gave only dimers because the favorable
conformation for cyclization was changed when the cis
double bond was replaced with a trans one in the precursor
peptide. Using the fragments and synthetic intermediates at
hand, we designed a new route for synthesis of the macro-
cyclic core (Scheme 4; all yields shown were based on (S)-
Scheme 4. Completion of the Synthesis
With a successful route to the production of racemization-
free compound 16 established, the subsequent removal of
both Bn and Boc protective groups with TFA in the presence
of pentamethylbenzene12 and the selective monosulfation
with sulfur trioxide pyridine complex in aqueous MeCN led
to molecule 1 with a good yield. However, 1H and 13C NMR
spectra of molecule 1 did not match those reported for natural
scleritodermin A: there were significant discrepancies in
chemical shifts in the region for conjugated vinylic protons.
It was apparent that both CH-3 and CH-5 signals of the
synthetic sample were located more than 1.0 ppm upfield
from the reported value. Since the NOESY correlation
between CH-3 and CH-5 signals of the natural product in
the original assignment seemed much weaker than the
NOESY correlation of synthetic molecule 1, we postulated
that the correct configuration of the conjugated thiazole
residue might be 2E,4E. Furthermore, the observation of a
difference in chemical shifts between synthetic molecule 1
and the natural sample represented the two -CH3 groups at
the keto-Ile moiety (0.81 and 0.79 ppm for structure 1, versus
0.91 and 0.79 ppm for the natural sample), when taken
together with the fact that the keto-Ile moiety with an (S)-
isobutyl side chain is present in both keramamide F13 and
oriamide,14 led us to propose that the corresponding stereo-
center of scleritodermin A may be in the (S)-configuration,
despite the 14-(R) assignment suggested in the original
isolation paper.
11). Fragments (R)- or (S)-11 and fragment 9 were depro-
tected and coupled to produce the required alcohol 17 in a
high yield. The coupling of molecules 17 and 13 was
subsequently carried out via Yamaguchi esterification,7
giving rise to compound 18. In addition, adoption of the
LiOH condition to chemoselectively hydrolyze methyl ester
groups was successful in recovering small amounts of
fragment 13 (12%) from the reaction. Ultimate closure of
the depsipeptide was readily achieved between the tyrosine
and R-azido-ꢀ-amino acid residues, after deprotecting the
Boc group. The synthesis then proceeded uneventfully
following the previously described conditions and yielded
To test the hypotheses proposed above, compounds 2 and
3, which have an (R)- or (S)-isobutyl at the keto-Ile moiety
and a 2E,4E-conjugated thiazole residue, respectively, were
selected as our new target molecules. Fragment 13 with the
1
molecules 2 and 3. The H NMR spectra of compound 2
(9) Dessolin, M; Guillerez, M. G.; Thieriet, N.; Guibe, F.; Loffet, A.
Tetrahedron Lett. 1995, 36, 5741.
was found to be quite similar to the spectra recorded for the
natural product: the signals for CH-3 and CH-5 found at 7.54
and 7.41 ppm, respectively, were identical to the correct
values, as expected, but persisting discrepancies in chemical
shift at the keto-Ile moiety suggested that molecule 2
represented an epimer of the natural product. Much to our
delight, the 1H and 13C NMR of compound 3 were identical
(10) Corey, E. J.; Gin, D. Y.; Kania, R. S. J. Am. Chem. Soc. 1996,
118, 9202.
(11) Papanikos, A.; Rademann, J; Meldal, M. J. Am. Chem. Soc. 2001,
123, 2176.
(12) Yoshino, H.; Tsuji, M. Chem. Pharm. Bull. 1990, 38, 1735.
(13) Itagaki, F.; Shigemori, H.; Ishibashi, M.; Nakamura, T.; Sasaki,
T.; Kobayashi, J. J. Org. Chem. 1992, 57, 5540.
(14) Chill, L.; Kashman, Y.; Schleyer, M. Tetrahedron. 1997, 53, 16147.
Org. Lett., Vol. 10, No. 17, 2008
3767

with those of natural scleritodermin A.15 Curiously, the
precursor2a-f or forming the R-ketoamide at the early stage,2g
the application of a mild transamination reaction in this
chemical environment toward the formation of the R-ketoa-
mide unit was effective. This, combined with the fact that
the azide group is inert in peptide synthesis made our strategy
flexible and versatile. Since R-ketoamides are substructures
found in a variety of natural products, the methodology
described herein can be applied to other natural targets.
specific rotation value of our synthetic compound 3 ([R]25
D
-108, c 0.1, MeOH) was different from that reported for
natural scleritodermin A ([R]D -41, c 0.1, MeOH); however,
we believe this discrepancy was caused by the impurity of
the natural product in 1H NMR and HPLC analysis.16 Thus,
we revised the two incorrect assignments reported in the
original paper describing the isolation of natural product
scleritodermin A, and we assigned molecule 3 described here
as the correct structure of scleritodermin A.
In summary, we achieved the first total synthesis of
scleritodermin A (product 3) along with two isomers
(products 1 and 2), thus leading to a revision of the structure
originally proposed for natural scleritodermin A. Although
several methods to construct R-keto lactam containing
macrocycles are established, such as using a protected
R-hydroxy carboxyl group as the key R-ketoamide
Acknowledgment. This work was supported by National
Natural Science Foundation of China (Grants 30725049 and
30623008), 863 Hi-Tech Program(grant 2006AA09Z442),
and the Shanghai Commission of Science and Technology
(Grant 05DZ19331). We are grateful to Prof. Eric Schmidt
of Utah University for providing a natural sample of
scleritodermin A.
Supporting Information Available: Experimental details,
characterization data, and copies of 1H and 13C NMR spectra
for synthetic intermediates. This material is available free
of charge via the Internet at http://pubs.acs.org.
(15) Simultaneous injection of natural and synthetic compounds gave
only one major peak at tr ) 30.88 min (A ) 0.05 M KH2PO4 in H2O, B )
CH3OH, gradient ) 0-100% B in 45 min). A sample of Scleritodermin A
was kindly supplied by Professor Eric Schmidt of Utah University.
(16) See Supporting Information for details.
OL801419M
3768
Org. Lett., Vol. 10, No. 17, 2008