Diaminodiacid-based solid-phase synthesis of peptide disulfide bond mimics

Article abstract of DOI:10.1002/anie.201302197

The antimicrobial peptide tachyplesin I was used as a model to apply the title strategy, which was developed for the preparation of peptidic macrocycles with double disulfide surrogates. The folding and activity of the tachyplesin I analogues were found to be sensitive to the structure of the disulfide surrogates, thus underlining the necessity of a flexible synthetic route for generating disulfide bond surrogates with high structural diversity. Copyright

Full text of DOI:10.1002/anie.201302197

.
Angewandte
Communications
DOI: 10.1002/anie.201302197
Cyclic Peptides
Diaminodiacid-Based Solid-Phase Synthesis of Peptide Disulfide Bond
Mimics**
Hong-Kui Cui, Ye Guo, Yao He, Feng-Liang Wang, Hao-Nan Chang, Yu-Jia Wang, Fang-
Ming Wu, Chang-Lin Tian,* and Lei Liu*
Dedicated to the Bayer company on the occasion of its 150th anniversary
Intensive research has recently been conducted on the
development of conformation-rigidified peptidic macrocycles
for diagnostic and therapeutic applications.[1] These com-
pounds usually exhibit extraordinary thermal and metabolic
stability owing to their defined tertiary folding constrained by
one or multiple disulfide bridges.[2] A potential problem for
the application in vivo is that the disulfide bridges are
unstable towards reducing agents, such as thiols, as well as
towards disulfide isomerases; the reduction then leads to
structural rearrangement and loss of activity.[3] As a result,
a number of approaches have been examined to replace the
disulfide bonds with bioisosteric and metabolically stable
surrogates such as thioether, diselenide, triazole, or hydro-
carbon-based bridges.[4] These disulfide surrogates were
usually synthesized through thiol alkylation, azide–alkyne
cycloaddition, or olefin metathesis reactions occurring at the
peptide side chains after the peptide skeletons are fully
assembled (Figure 1). Limited types of disulfide surrogates
can be generated with the above post-chain-assembly cycli-
zation strategy, because many bridge formation reactions are
not readily compatible with the peptide backbones. More-
over, when more than one disulfide surrogate need to be
installed, the post-chain-assembly cyclization method may
sometimes encounter difficulties in controlling the regiose-
lectivity of the cross-linking reactions.[5]
Figure 1. Two strategies for making disulfide-bond surrogates.
[*] H.-K. Cui,[+] Y. Guo,[+] F.-L. Wang, H.-N. Chang, Y.-J. Wang,
Prof. Dr. L. Liu
In our studies on bioactive peptidic macrocycles[6] we need
to generate disulfide surrogates with a higher degree of
structural diversity. Thus we became interested in the use of
Bayer-Tsinghua University Joint Research Center for Innovative Drug
Discovery
Department of Chemistry, Tsinghua University
Beijing 100084 (China)
pre-prepared diaminodiacids in the synthesis of peptide
disulfide bond mimics (Figure 1). In this strategy all cycliza-
tion steps are achieved through peptide coupling reactions, so
that the difficulty for making nonpeptidic bridges on peptides
can be avoided. Indeed, a few previous studies showed that
the diaminodiacid strategy was useful for the synthesis of
bioactive disulfide-bond mimics, albeit hitherto containing
only one disulfide bridge.[7] To explore this strategy for
peptides bearing two or more disulfide surrogates, we now
describe the synthesis and bioactivities of thioether and
biscarba analogues of tachyplesin I (TPI-1), which is a 17-
residue bicyclic peptide with high antimicrobial activity.[8]
TPI-1 adopts a b-hairpin conformation stabilized by two
cross-strand disulfide bonds. Meldal and Holland-Nell
recently showed that TPI-1 analogues with triazole bridges
E-mail: lliu@mail.tsinghua.edu.cn
Y. He,[+] F. M. Wu, Prof. Dr. C.-L. Tian
School of Life Sciences
University of Science and Technology of China
High Magnetic Field Laboratory, Chinese Academy of Sciences
Hefei 230031 (China)
E-mail: cltian@ustc.edu.cn
[+] These authors contributed equally to this work.
[**] This work was supported by the national “863” and “973” grants
from the Ministry of Science and Technology (No. 2012AA02A700,
2011CB965300, 2011CB911104) and NSFC (No. 20932006,
91013007, 31170817). We also thank Bayer Schering Pharma for the
financial support. We thank Prof. Jingren Zhang for the help in the
bioassays.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201302197.
9558
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 9558 –9562

Angewandte
Chemie
can largely maintain the biological activity.[4i] Here, we found
that the folding and activity of TPI-1 analogues were rather
subtly sensitive to the structure of the disulfide surrogates,
thus underlining the necessity of a flexible synthetic route for
generating disulfide bond surrogates with high structural
diversity.
To implement the diaminodiacid approach for the syn-
thesis of TPI-1 analogues with double disulfide surrogates, we
took inspiration from the recent successful total synthesis of
lantibiotics pioneered by the research groups of Vederas[9]
and van der Donk.[10] Two sets of orthogonally protected
diaminodiacids compatible with Fmoc (9H-fluorenyl-meth-
oxycarbonyl) solid-phase peptide synthesis (SPPS) were
prepared first (Scheme 1). The allyl/allyloxycarbonyl (Alloc)
oxide hexafluorophosphate) as the coupling reagent. Subse-
quently we removed the Alloc and allyl groups on 5 by using
[Pd(PPh3)4]/PhSiH3. After the N-terminal Fmoc group was
removed with 20% piperidine in DMF, peptide 6 was
obtained by lactamization using PyBOP ((benzotriazole-1-
yloxy)-tripyrrolidino-phosphonium hexafluorophosphate),
HOBt (1-hydroxy-1H-benzo-triazole), and NMM (N-meth-
ylmorpholine). Next, three more amino acids (i.e. Val, Arg,
Phe) were assembled onto the peptide leading to 7. The pNb
and pNz groups on 7 were removed by using SnCl2 (6m)/HCl
(1.6 mm). After removal of the N-terminal Fmoc group,
peptide 8 was obtained by lactamization using PyBOP/HOBt/
NMM. Finally, Trp and Lys were attached to 8 producing the
bicyclic peptide 9. After acidic cleavage from the resin with
concomitant removal of all the protecting groups, we obtained
crude TPI-2.
According to HPLC analysis, the crude product contained
a single major component (Figure 2), which can be easily
purified to the desired product (TPI-2) with an overall yield of
isolated product of 7%. Thus, the diaminoacid strategy can
readily afford the desired disulfide mimics with fairly good
efficiency. By using the same approach we also prepared the
thioether-containing TPI-3 with an overall yield of 5% after
isolation (Table 1). Note that in the recent preparation of
a triazole-bridged TPI-1 analogue through CuI-mediated
Scheme 1. Pre-prepared orthogonally protected diaminodiacids.
and p-nitrobenzyl (pNb)/p-nitrobenzyloxycarbonyl (pNz)
protecting groups were employed, because they can be
selectively removed by using [Pd(PPh3)4]/PhSiH3 and SnCl2/
HCl, respectively. From protected l-glutamic acids we
synthesized the biscarba diaminodiacids (1 and 2) through
Kolbe electrolytic decarboxylative cross-dimerization.[11]
From protected l-homoserine and l-cystine we made the
thioether-containing diaminodiacids (3 and 4).[7]
With the diaminodiacid building blocks in hand, we
commenced the synthesis of biscarba TPI-1 analogue TPI-2
(Scheme 2) by using the Rink amide AM resin (loading =
0.3 mmolgÀ1). A linear octapeptide 5 containing both 1 and 2
was first assembled onto the resin by using HCTU (5-chloro-
Figure 2. a) HPLC traces of crude and purified TPI-2. b) ESI-QTOF-MS
1-[bis(dimethylamino)methylene]-1H-benzotriazolium
3-
spectrum of TPI-2 (expected mass: 2190.2551).
Scheme 2. Solid-phase synthetic route for the preparation of TPI-2 through the diaminodiacid strategy. a) [Pd(PPh3)4]/PhSiH3, b) 20% piperidine/
DMF, c) PyBOP/HOBt/NMM, d) SnCl2 (6 m)/HCl (1.6 mm), e) 20% piperidine/DMF, f) PyBOP/HOBt/NMM, g) trifluoroacetic acid/triisopropylsi-
lane/water (95:2.5:2.5; v/v/v). The following protecting groups for amino acid side chains were used: tert-butyl (for Tyr), 2,2,4,6,7-
pentamethyldihydrobenzofurane-5-sulfonyl (Pbf; for Arg), and tert-butyloxycarbonyl (Boc; for Trp and Lys).
Angew. Chem. Int. Ed. 2013, 52, 9558 –9562
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9559

.
Angewandte
Communications
Table 1: Structures of different TPI analogues and their minimal inhibitory concentrations (MIC) against
inhibitory concentration (MIC) was
determined by using a standard
several bacterial strains.[a]
two-fold
dilution
protocol
(Table 1).[14] It was found that TPI-
1 exhibited comparable bactericidal
activities against both Gram-posi-
tive and Gram-negative bacteria, an
observation consistent with the pre-
vious studies.[15] Although the struc-
tures of TPI-2 and TPI-3 deviated
from the ordered b-pleat sheet
structure, they were still effective
inhibitors of bacterial growth with
MIC values about 2- to 16-fold
lower than that of TPI-1. Note that
some slightly better MIC values
were reported by Meldal and Hol-
land-Nell for the triazole-contain-
ing TPI-1 analogues (TPI-triazole
and TPI-triazole’, Table 1).[4i] They
also reported that neither the linear
Compound
TPI-1
X1
S
X2
S
Y1
S
Y2
S
B. subtilis S. epidermidis S. enterica P. aeruginosa E. coli
4
4
(2.3[15b]
32
16
16
16
8
4
16
4
8
(16.2[15a]
64
64
32
64
64
32
32
16
8
(>200[15a]
)
)
)
(8.2[15a]
)
(11.5[15a]
)
TPI-2
TPI-3
TPI-4
TPI-5
TPI-6
TPI-7
TPI-8
TPI-9
CH2 CH2 CH2 CH2
S
CH2 CH2
S
S
S
S
CH2
8
16
8
8
8
8
16
4
256
16
16
16
16
64
8
CH2
S
S
CH2
S
128
64
64
64
16
64
16
S
CH2
S
S
S
CH2 CH2
S
S
CH2
S
S
CH2
S
32
8
S
16
TPI-triazole[4i]
5.5[4i] 8.0[4i]
–
–
–
–
10.0[4i]
TPI-triazo-
4.5[4i] 10.5[4i]
7.0[4i]
le’[4i]
[a] MIC values were obtained by using the standard two-fold dilution protocol and are reported in
mgmLÀ1
.
cycloaddition, two cyclic products were obtained with a regio-
selectivity of 1:1.5 (correct/incorrect).[4i] This problem may
cause complications in both the isolation and structure
determination of the desired product, thus manifesting the
advantage of the diaminodiacid strategy. In contrast, a high
degree of selectivity for the intramolecular cyclization versus
intermolecular amidation in the diaminodiacid strategy may
result from the pseudo-high dilution of the resin-bound
peptides.
The circular dichroism (CD) spectrum of TPI-1 in water
showed a positive peak at 198 nm, a negative minimum at
208 nm, and a strong positive band at 228–230 nm (Fig-
ure 3a). These observations were consistent with the well-
defined b-pleat sheet structure reported for TPI-1.[12] In
contrast, for TPI-2 we only observed a small negative
minimum at 212 nm indicating a less-ordered b-turn con-
formation. More strikingly, for TPI-3 there was no negative
minimum in the CD spectrum, thus suggesting a significant
deviation from the b-turn structure. To corroborate the above
conclusions, we conducted solution NMR spectroscopy
(including DQF-COSY, TOCSY, and NOESY) for TPI-2
and TPI-3 (The NMR structure for TPI-1 was reported[13]).
The determined solution structures revealed that while TPI-
1 adopted an ordered b-pleat sheet structure (Figure 3b),
almost all the carbonyl oxygen atoms of TPI-2 and TPI-3
pointed outward, disfavoring the formation of any interstrand
hydrogen bonds (Figure 3c,d). Thus both CD and NMR
spectroscopy studies indicated that TPI-2 and TPI-3 did not
adopt the authentic b-hairpin structure.
~
&
*
Figure 3. a) CD spectra of TPI-1 ( ), TPI-2 ( ), and TPI-3 ( ) in water
and b–d) structures for TPI-1 (b), TPI-2 (c), and TPI-3 (d) determined
by NMR spectroscopy in aqueous solution.
nor the incorrectly cyclized analogues of TPI-1 showed any
significant antimicrobial activity.[4i] Collectively, these results
indicated that the overall hairpin-like shape, rather than the
details of the b-sheet hydrogen bonding patterns, played a key
role in determining the bioactivity of TPI-1 analogues. It is
important to note that the installation of biscarba or thioether
surrogates in TPI-1 improved its stability. Indeed, TPI-1 was
completely reduced in one hour in aqueous dithiothreitol
(DTT, 1 mm) solution, while TPI-2 and TPI-3 showed no
degradation in DTT solution after 48 h. Furthermore, the
serum stability experiments revealed that 69% of native TPI-
1 was left after 24 h, whereas during the same time 85% of
biscarba analogue TPI-2 remained intact.
The biological activity of the TPI-1 analogues was studied
with antimicrobial assays. Several bacterial strains were
grown in the presence of increasing concentrations of the
TPI-1 analogues. These strains included two Gram-positive
bacteria (Bacillus subtilis and Staphylococcus epidermidis)
and three Gram-negative bacteria (Salmonella enterica,
Pseudomonas aeruginosa, and Escherichia coli). The minimal
9560 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 9558 –9562

Angewandte
Chemie
To scrutinize why the change of disulfide bonds to the
thioether and biscarba surrogates caused significant devia-
tions from the b-hairpin structure, we measured the dihedral
angles of the cross-strand bridges in the NMR structures. For
TPI-1, the dihedral angles between the C–S–S–C atoms were
94.28 (inner bridge) and 97.28 (outer bridge) degrees,
respectively. This observation was consistent with the notion
that the dihedral angle of the disulfide bond in proteins is
usually close to 908.[16] The dihedral angles of the two bridges
in TPI-2 were 59.08 (inner) and 174.78 (outer), whereas for
TPI-3 the values were 49.48 (inner) and 143.18 (outer). Such
large changes of dihedral angles may indicate that the
À
À
different patterns of torsional strain among the S S, C C,
À
and S C bonds led to the variation of the global conforma-
tion.
One possible solution to alleviate the above problem was
to use only one disulfide surrogate in the TPI-1 analogues. It is
also interesting to learn which of the two disulfide bonds in
TPI-1 was more important for maintaining the b-hairpin
structure. Thus we designed and synthesized TPI-1 analogues
with only the inner or outer disulfide bond replaced by one
biscarba or thioether bridge (TPI-4 to TPI-7, Table 1). To test
whether or not the orientation of the thioether bond exerted
any significant effect, we also prepared TPI-8 and TPI-9
(Table 1). Our measurements revealed that the MIC values of
TPI-4 to TPI-9 were fairly close to each other. These values
were more resembling to those of TPI-1, thereby suggesting
that TPI-4 to TPI-9 were more active than TPI-2 or TPI-3. In
particular, the activity of the thioether-bridged TPI-9 was
identical to that of TPI-1 for four strains of bacteria (i.e.
Bacillus subtilis, Staphylococcus epidermidis, Pseudomonas
aeruginosa, and Escherichia coli). Thus the installation of only
one disulfide surrogate indeed caused less activity change
than the replacement of two disulfide bonds. Unfortunately,
our data did not give a clear hint concerning the relative
importance of the inner versus outer disulfide bridges.
To examine the structural distortion owing to the instal-
lation of disulfide surrogates, we measured the backbone Ha
and amide H(N) chemical shifts of TPI-1 to TPI-9. Because
the proton chemical shift value is sensitive to the chemical
environments around the proton, the change of these values
from the parent compound to its analogues may reflect the
extent of structural variation. In our NMR experiments,
proton assignments were achieved for seven peptides (TPI-1,
2, 3, 4, 5, 7, and 9). The root-mean-square deviations of the
backbone Ha chemical shifts of 16 residues for TPI-2, 3, 4, 5,
7, and 9 (Figure 4a; note: Lys1 was exposed to the solvent
molecules and subject to fast proton exchange) were found to
be 0.58, 0.49, 0.40, 0.51, 0.15, 0.37 ppm. Moreover, the root-
mean-square deviations of the backbone amide H(N) chem-
ical shifts of 16 residues for TPI-2, 3, 4, 5, 7, and 9 (Figure 4b)
were 0.56, 0.56, 0.32, 0.52, 0.13, 0.31 ppm, respectively. Thus,
the structures of TPI-4, 7, and 9 were more similar to the
structure of TPI-1 than to TPI-2 and 3. This observation
confirmed our expectation that the installation of only one
disulfide surrogate caused less structural changes than the
replacement of two disulfide bonds.
Figure 4. The chemical shift differences of backbone Ha (a) and
amide H(N) (b) between the wild-type peptide (TPI-1) and modified
peptides (TPI-2, 3, 4, 5, 7, and 9). DAD=diaminodiacid.
of peptide disulfide bond mimics with two cross-linking
bridges. With the b-hairpin antimicrobial peptide tachyple-
sin I as a model, we showed that the diaminodiacid strategy
can readily generate the desired disulfide surrogates with
good yields and reliable structural control. Both the CD and
NMR spectroscopy studies indicated that the biscarba and
thioether-containing TPI-1 analogues did not adopt the
authentic b-hairpin structure owing to the disruption of
interstrand hydrogen bonding. Nonetheless, these analogues
were still effective antimicrobial peptides, thus indicating that
the overall hairpin-like shape, rather than the details of the b-
sheet conformation, dominated the bioactivity of these
compounds. More detailed analysis showed that the dihedral
angles of the disulfide, biscarba, or thioether bridges in TPI-
1 analogues varied dramatically, thereby indicating that the
À
À
À
different torsional strains of the S S, C C, and S C bonds
may cause the structural distortions. Thus, to pinpoint the
optimal peptide disulfide bond mimics, it is important to make
and screen an array of disulfide bond surrogates ideally with
high structural diversity. Although the present study has only
examined the biscarba- and thioether-based bridges, we
expect that the preparation and use of diverse structures of
diaminodiacids would be practical.
Received: March 15, 2013
Revised: April 28, 2013
Published online: June 26, 2013
Keywords: biological activity · cyclic peptides · disulfide bonds ·
.
b-hairpin structure · solid-phase synthesis
[1] a) D. Obrecht, E. Chevalier, K. Moehle, J. A. Robinson, Drug
Discovery Today Technol. 2012, 9, e63; b) G. L. Verdine, G. J.
Hilinski, Methods Enzymol. 2012, 503, 3; c) V. Lavergne, R. J.
Taft, P. F. Alewood, Curr. Top. Med. Chem. 2012, 12, 1514.
To summarize, we examined for the first time the use of
a diaminodiacid-based strategy for the solid-phase synthesis
Angew. Chem. Int. Ed. 2013, 52, 9558 –9562
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9561

.
Angewandte
Communications
[2] a) H. Kolmar, Curr. Opin. Pharmacol. 2009, 9, 608; b) N. L. Daly,
K. J. Rosengren, D. J. Craik, Adv. Drug Delivery Rev. 2009, 61,
918.
[3] a) J. Gehrmann, P. F. Alewood, D. J. Craik, J. Mol. Biol. 1998,
278, 401; b) D. L. Rabenstein, K. H. Weaver, J. Org. Chem. 1996,
61, 7391; c) M. C. A. Laboissiere, S. L. Sturley, R. T. Raines, J.
Biol. Chem. 1995, 270, 28006.
2005, 80, 85; e) S. Fiori, S. Pegoraro, S. R. Bohner, J. Cramer, L.
Moroder, Biopolymers 2000, 53, 550.
[6] Our recent studies on bioactive peptidic macrocycles: a) J.-S.
Zheng, S. Tang, Y. Guo, H.-N. Chang, L. Liu, ChemBioChem
2012, 13, 542; b) J.-S. Zheng, H.-N. Chang, J. Shi, L. Liu, Sci.
China Chem. 2012, 55, 64; c) S. Tang, J. Zheng, K. Yang, L. Liu,
Acta Chim. Sin. 2012, 70, 1471; d) H.-K. Cui, B. Zhao, Y.-H. Li,
Y. Guo, H. Hu, L. Liu, Y.-G. Chen, Cell Res. 2013, 23, 581.
[7] a) R. F. Nutt, D. F. Veber, R. Saperstein, J. Am. Chem. Soc. 1980,
102, 6539; b) R. F. Nutt, R. G. Strachan, D. F. Veber, F. W. Holly,
J. Org. Chem. 1980, 45, 3078; c) Y. Kambayashi, S. Nakajima, M.
Ueda, K. Inouye, FEBS Lett. 1989, 248, 28; d) P. K. Bhatnagar,
E. K. Agner, D. Alberts, B. E. Arbo, J. F. Callahan, A. S.
Cuthbertson, S. J. Engelsen, H. Fjerdingstad, M. Hartmann, D.
Heerding, J. Hiebl, W. F. Huffman, M. Hysben, A. G. King, P.
Kremminger, C. Kwon, S. LoCastro, D. Løvhaug, L. M. Pelus, S.
Petteway, J. S. Takata, J. Med. Chem. 1996, 39, 3814; e) P. J.
Knerr, A. Tzekou, D. Ricklin, H. Qu, H. Chen, W. A. van der
Donk, J. D. Lambris, ACS Chem. Biol. 2011, 6, 753.
[8] T. Nakamura, H. Furunaka, T. Miyata, F. Tokunaga, T. Muta, S.
Iwanaga, M. Niwa, T. Takao, Y. Shimonishi, J. Biol. Chem. 1988,
263, 16709.
[9] W. Liu, A. S. H. Chan, H. Liu, S. A. Cochrane, J. C. Vederas, J.
Am. Chem. Soc. 2011, 133, 14216.
[10] P. J. Knerr, W. A. van der Donk, J. Am. Chem. Soc. 2012, 134,
7648.
[11] J. Hiebl, M. Blanka, A. Guttman, H. Kollmann, K. Leitner, G.
Mayrhofer, F. Rovenszky, K. Winkler, Tetrahedron 1998, 54,
2059.
[12] A. G. Rao, Arch. Biochem. Biophys. 1999, 361, 127.
[13] a) A. Laederach, A. H. Andreotti, D. B. Fulton, Biochemistry
2002, 41, 12359; b) Y. Su, W. F. DeGrado, M. Hong, J. Am.
Chem. Soc. 2010, 132, 9197.
[4] a) J. Rudinger, K. Jost, Experientia 1964, 20, 570; b) S. Sakaki-
bara, S. Hase, Bull. Chem. Soc. Jpn. 1968, 41, 22816. More recent
examples: c) A. Brik, Adv. Synth. Catal. 2008, 350, 1661; d) D. J.
Derksen, J. L. Stymiest, J. C. Vederas, J. Am. Chem. Soc. 2006,
128, 14252; e) C. A. MacRaild, J. Illesinghe, B. J. van Lierop,
A. L. Townsend, M. Chebib, B. G. Livett, A. J. Robinson, R. S.
Norton, J. Med. Chem. 2009, 52, 755; f) A. J. Robinson, B. J.
van Lierop, R. D. Garland, E. Teoh, J. Elaridi, J. P. Illesinghe,
W. R. Jackson, Chem. Commun. 2009, 4293; g) M. Roice, I.
Johannsen, M. Meldal, QSAR Comb. Sci. 2004, 23, 662; h) M.
Empting, O. Avrutina, R. Meusinger, S. Fabritz, M. Reinwarth,
M. Biesalski, S. Voigt, G. Buntkowsky, H. Kolmar, Angew.
Chem. 2011, 123, 5313; Angew. Chem. Int. Ed. 2011, 50, 5207;
i) K. Holland-Nell, M. Meldal, Angew. Chem. 2011, 123, 5310;
Angew. Chem. Int. Ed. 2011, 50, 5204; j) S. Pegoraro, S. Fiori, S.
Rudolph-Bçhner, T. X. Watanabe, L. Moroder, J. Mol. Biol.
1998, 284, 779; k) L. Moroder, J. Pept. Sci. 2005, 11, 187; l) M.
Muttenthaler, S. T. Nevin, A. A. Grishin, S. T. Ngo, P. T. Choy,
N. L. Daly, S.-H. Hu, C. J. Armishaw, C.-I. A. Wang, R. J. Lewis,
J. L. Martin, P. G. Noakes, D. J. Craik, D. J. Adams, P. F. Ale-
wood, J. Am. Chem. Soc. 2010, 132, 3514; m) Z. Dekan, I. Vetter,
N. L. Daly, D. J. Craik, R. J. Lewis, P. F. Alewood, J. Am. Chem.
Soc. 2011, 133, 15866; n) A. D. de Araujo, B. Callaghan, S. T.
Nevin, N. L. Daly, D. J. Craik, M. Moretta, G. Hopping, M. J.
Christie, D. J. Adams, P. F. Alewood, Angew. Chem. 2011, 123,
6657; Angew. Chem. Int. Ed. 2011, 50, 6527; o) A. D. de Araujo,
M. Mobli, G. F. King, P. F. Alewood, Angew. Chem. 2012, 124,
10444; Angew. Chem. Int. Ed. 2012, 51, 10298.
[14] C. Ramalingan, I.-S. Lee, Y.-W. Kwak, Chem. Pharm. Bull. 2009,
57, 591.
[5] Several studies have been reported to overcome the problem of
cross-linking regioselectivity in the post-chain assembly strategy:
a) M. Okumura, S. Shimamoto, Y. Hidaka, FEBS J. 2012, 279,
2283; b) L. Moroder, D. Besse, H. J. Musiol, S. R. Bohner, F.
Siedler, Biopolymers 1996, 40, 207; c) J. Bondebjerg, M.
Grunnet, T. Jespersen, M. Meldal, ChemBioChem 2003, 4, 186;
d) L. Moroder, H. J. Musiol, M. Gotz, C. Renner, Biopolymers
[15] a) A. Ramamoorthy, S. Thennarasu, A. Tan, K. Gottipati, S.
Sreekumar, D. L. Heyl, F. Y. P. An, C. E. Shelburne, Biochem-
istry 2006, 45, 6529; b) Y. Imura, M. NIshida, Y. Ogawa, Y.
Takakura, K. Matsuzaki, Biochim. Biophys. Acta Biomembranes
2007, 1768, 1160.
[16] J. M. Thornton, J. Mol. Biol. 1981, 151, 261.
9562 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 9558 –9562