Convergence leads to success: Total synthesis of the complex nonribosomal peptide polytheonamide B

Article abstract of DOI:10.1002/anie.201001917

Beyond the ribosome: The highly cytotoxic 48 residue nonribosomal peptide polytheonamide B, proposed to act as an ion channel, was synthesized by the coupling of four building blocks. The assembly of these peptide segments required the synthesis of eight nonproteinogenic amino acids, including a unique sulfoxide. This first synthetic route towards polytheonamide B (see scheme) demonstrates the potential of state-of-the-art peptide chemistry.

Full text of DOI:10.1002/anie.201001917

Highlights
DOI: 10.1002/anie.201001917
Natural Products
Convergence Leads to Success: Total Synthesis of the
Complex Nonribosomal Peptide Polytheonamide B**
Christian Ducho*
amino acids · ion channels · peptides ·
peptide synthesis · sulfoxides
B
eyond the limited set of amino acids employed for
cations across lipid bilayers, and its voltage-gating properties,
polytheonamide B has been proposed to act as an ion
channel,[4] probably accounting for the highly cytotoxic effect
of the peptide. While, for instance, gramicidin A is a rather
thoroughly studied peptidic transmembrane channel,[5] mo-
lecular details of polytheonamide action remain to be
elucidated.
ribosomal protein biosynthesis, numerous nonproteinogenic
amino acids are used in nature for the generation of complex
peptidic structures. In many cases, such nonribosomal pep-
tides are assembled by the action of nonribosomal peptide
synthetases (NRPS), large enzymes with modular organiza-
tion. The interesting biological properties of nonribosomal
peptides make them attractive targets for total synthesis, thus
making studies of structure–activity relationships (SAR)
possible.[1]
Because of their interesting structural and biological
properties, the polytheonamides should be considered attrac-
tive new targets for SAR investigations. SAR data might
potentially broaden the molecular understanding of the
polytheonamide mechanism and of channel formation and
function in biological membranes in general. However, the
total synthesis of complex peptide structures such as the
polytheonamides still represents a significant challenge. The
numerous unusual amino acid motifs make the synthetic
assembly by standard solid-phase peptide synthesis (SPPS)
difficult, and biosynthetic approaches often do not permit
structural variations sufficient for detailed SAR studies.
Inoue and co-workers have now reported the first total
synthesis of polytheonamide B (1), which is based on a
convergent synthetic strategy, and also the stereochemical
assignment of the sulfoxide moiety.[6] They have dissected the
structure of 1 into four peptide segments, each composed of
7–16 amino acids. These building blocks were designed in such
a way that their late-stage coupling would occur through the
activation of their C-terminal glycine residues, thus avoiding
epimerization in the coupling steps (Figure 1). Eight of the 13
nonproteinogenic amino acids found in 1 were not commer-
cially available. Consequently, their prepraration represented
the initial synthetic challenge and was followed by assembly
of the four peptide segments by SPPS.
The stereoselective synthesis of a 9-fluorenylmethoxy-
carbonyl(Fmoc)-protected derivative of the characteristic
sulfoxide amino acid for SPPS was a key achievement in the
total synthesis of 1. Starting from suitably protected aspartate
2, methylation in the b position gave 3, which could be
transformed into sulfide 4 in six steps. The diastereoselective
oxidation of 4 was then performed under Katsuki conditions[7]
with urea–hydrogen peroxide in the presence of the chiral
Ti(salen) catalyst 5. Thus, the Fmoc-protected sulfoxide
amino acid ester 6 was obtained in an initial diastereoselec-
tivity of 85% de, which could be improved to 96% de by
column chromatography. Acidic deprotection then furnished
target compound 7 (Scheme 1). The absolute configuration at
The polytheonamides A and B are possibly the largest and
most complex nonribosomal peptides reported so far.[2] They
were isolated from the marine sponge Theonella swinhoei,
though they may potentially be produced by an unknown
symbiotic microorganism. Polytheonamides A and B com-
prise 48 amino acid residues plus an N-terminal 5,5-dimethyl-
2-oxohexanoyl cap (Figure 1). The amino acids display
alternating l and d configuration, with the exception of eight
glycine residues. Furthermore, 13 of the 19 different amino
acid components constituting the polytheonamides are non-
proteinogenic, including some b-methylated derivatives. One
unique amino acid structure can be found: a sulfoxide amino
acid bearing a stereogenic sulfur atom in its side chain
(Figure 1). The only structural difference between polytheo-
namides A and B is the absolute configuration of this
sulfoxide moiety, and so far, an assignment of the config-
uration (RS/SS) to the corresponding polytheonamide peptide
has not been feasible.
Polytheonamides display pronounced cytotoxicity (e.g.
polytheonamide B (1): EC50 = 79 pm). NMR studies indicate
that 1 folds into a b helix stabilized by hydrogen bonds, thus
forming a 30ꢀ long tubular structure.[3] Owing to its three-
dimensional structure, the ability to conduct monovalent
[*] Prof. Dr. C. Ducho
Department of Chemistry
Institute of Organic and Biomolecular Chemistry
Georg-August-University Gꢀttingen
Tammannstrasse 2, 37077 Gꢀttingen (Germany)
Fax: (+49)551-39-9660
E-mail: cducho@gwdg.de
Homepage: http://www.ducho.chemie.uni-goettingen.de
[**] The author thanks the Deutsche Forschungsgemeinschaft (SFB 803
“Functionality controlled by organization in and between mem-
branes”) and the Fonds der Chemischen Industrie (Sachkostenzu-
schuss) for financial support.
5034
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 5034 – 5036

Angewandte
Chemie
Figure 1. Structure of polytheonamides. Nonproteinogenic amino acids are highlighted in green, and those which required preparation in the
course of the total synthesis of polytheonamide B (1) are highlighted in red. The sulfoxide amino acid is shown in blue. The four peptide
segments conceived and prepared for the convergent total synthesis of 1 are indicated.
vergent approach. Segments 1–3 (Figure 1) were synthesized
on a Wang resin, and after cleavage from the resin, they were
converted into thioester derivatives. The resultant peptide
thioester building blocks could be isolated in overall yields
from 9 to 13%. The C-terminal segment 4 was synthesized on
a 2-chlorotrityl resin and obtained in 7% overall yield after 31
reaction steps. Thus, all reactants for the final convergent
assembly of the complete peptide were prepared. Starting
from N-terminally unprotected segment 4, coupling with the
neighboring thioester segment 3 was achieved in the presence
of silver(I) nitrate as introduced by Aimoto.[9] After basic
Fmoc deprotection, the same coupling conditions with sub-
sequent basic deprotection were applied for the reaction with
thioester segments 2 and 1. The yields for each coupling
procedure were in the range of 83 to 91%. This resulted in the
stepwise formation of protected polytheonamide B, and
global acidic deprotection finally furnished target peptide 1.
As the obtained compound was spectroscopically and chro-
matographically identical with naturally occurring polytheo-
namide B, it was proven for the first time that polytheonami-
de B displays R configuration at the sulfoxide moiety, and
polytheonamide A was therefore assigned to be S-configured
in the side chain of amino acid 44.
Scheme 1. Diastereoselective synthesis of the Fmoc-protected sulfoxide
amino acid 7 for solid-phase peptide synthesis. KHMDS=potassium
hexamethyldisilazanide, PhFl=9-phenylfluoren-9-yl, TFA=trifluoroace-
tic acid.
the stereogenic sulfur atom of 7 was proven to be R by
application of the NMR-based methodology of Kusumi and
Yabuuchi after derivatization of the sulfoxide.[8]
In conclusion, the recently reported first total synthesis of
the nonribosomal peptide polytheonamide B (1) is a prime
example of the potential of state-of-the-art peptide chemistry.
It is the latest in the series of impressive achievements in
peptide synthesis, for example, the preparation of the
membrane-pore-forming antibiotic alamethicin[10] and of the
DNA-gyrase inhibitor microcin.[11] In the case of polytheon-
amide B (1), the stereoselective synthesis of nonproteino-
genic amino acids in combination with solid-phase peptide
synthesis and the coupling of peptide segments under mild
conditions provided access to a complex peptidic structure of
high biological relevance. The developed synthetic route also
The preparation of the four peptide segments was then
feasible using standard SPPS chemistry. The main challenges
at this stage, resulting from the amino acid sequence of 1,
were the numerous sterically hindered b-tetrasubstituted
amino acids displaying low reactivity as well as the presence
of many asparagine and glutamine derivatives, which tend to
form interstrand aggregates in the peptide–resin matrix. Since
the sequence length accessible by SPPS was limited to 16, the
Inoue group developed the described segment-based con-
Angew. Chem. Int. Ed. 2010, 49, 5034 – 5036
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5035

Highlights
[4] S. Oiki, I. Muramatsu, S. Matsunaga, N. Fusetani, Folia
Pharmacol. Jpn. 1997, 110, 195P – 198P.
demonstrates the versatility of convergent approaches in the
preparation of longer oligopeptides. The stage is now set for
the synthesis of polytheonamide analogues, SAR studies, and
further structural as well as biophysical investigations.
[5] a) D. W. Urry, Proc. Natl. Acad. Sci. USA 1971, 68, 672 – 676;
b) D. A. Langs, Biopolymers 1989, 28, 259 – 266; c) R. R. Ketch-
em, W. Hu, T. A. Cross, Science 1993, 261, 1457 – 1460.
[6] M. Inoue, N. Shinohara, S. Tanabe, T. Takahashi, K. Okura, H.
Itoh, Y. Mizoguchi, M. Iida, N. Lee, S. Matsuoka, Nat. Chem.
2010, 2, 280 – 285; this work is also discussed in a News and
Views article: C. J. Forsyth, Nat. Chem. 2010, 2, 252 – 254.
[7] B. Saito, T. Katsuki, Tetrahedron Lett. 2001, 42, 3873 – 3876.
[8] T. Yabuuchi, T. Kusumi, J. Am. Chem. Soc. 1999, 121, 10646 –
10647.
Received: March 31, 2010
Published online: June 22, 2010
[1] D. Schwarzer, R. Finking, M. A. Marahiel, Nat. Prod. Rep. 2003,
20, 275 – 287.
[2] T. Hamada, S. Matsunaga, G. Yano, N. Fusetani, J. Am. Chem.
Soc. 2005, 127, 110 – 118.
[3] T. Hamada, S. Matsunaga, N. Fusetani, M. Fujiwara, K. Fujita,
Tennen Yuki Kagobutsu Toronkai Koen Yoshishu 1995, 37, 695 –
700.
[9] S. Aimoto, Biopolymers 1999, 51, 247 – 265.
[10] H. Schmitt, G. Jung, Liebigs Ann. Chem. 1985, 321 – 344.
[11] G. Videnov, D. Kaiser, M. Brooks, G. Jung, Angew. Chem. 1996,
108, 1607 – 1609; Angew. Chem. Int. Ed. Engl. 1996, 35, 1506 –
1508.
5036
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 5034 – 5036