Cyclopropane-Based Peptidomimetics Mimicking Wide-Ranging Secondary Structures of Peptides: Conformational Analysis and Their Use in Rational Ligand Optimization

Article abstract of DOI:10.1002/chem.201605312

Detailed conformational analyses of our previously reported cyclopropane-based peptidomimetics and conformational analysis-driven ligand optimization are described. Computational calculations and X-ray crystallography showed that the characteristic features of cyclopropane function effectively to constrain the molecular conformation in a three-dimensionally diverse manner. Subsequent principal component analysis revealed that the diversity covers the broad chemical space filled by peptide secondary structures in terms of both main-chain and side-chain conformations. Based on these analyses, a lead stereoisomer targeting melanocortin receptors was identified, and its potency and subtype selectivity were improved by further derivatization. The presented strategy is effective not only for designing non-peptidic ligands from a peptide ligand but also for the rational optimization of these ligands based on the plausible target-binding conformation without requiring the three- dimensional structural information of the target and its peptide ligands.

Full text of DOI:10.1002/chem.201605312

A Journal of
Accepted Article
Title: Cyclopropane-Based Peptidomimetics Mimicking Wide-Ranging
Secondary Structures of Peptides: Conformational Analysis and
Their Use in Rational Ligand Optimization
Authors: Akira Mizuno, Tomoshi Kameda, Tomoki Kuwahara, Hideyuki
Endoh, Yoshihiko Ito, Shizuo Yamada, Kimiko Hasegawa,
Akihito Yamano, Mizuki Watanabe, Mitsuhiro Arisawa, and
Satoshi Shuto
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201605312
Link to VoR: http://dx.doi.org/10.1002/chem.201605312
Supported by

10.1002/chem.201605312
Chemistry - A European Journal
FULL PAPER
Cyclopropane-Based Peptidomimetics Mimicking Wide-Ranging
Secondary Structures of Peptides: Conformational Analysis and
Their Use in Rational Ligand Optimization
Akira Mizuno,[a] Tomoshi Kameda,[c] Tomoki Kuwahara,[a] Hideyuki Endoh,[a] Yoshihiko Ito,[d] Shizuo
Yamada,[d] Kimiko Hasegawa,[e] Akihito Yamano,[e] Mizuki Watanabe,[a] Mitsuhiro Arisawa[a] and
Satoshi Shuto*[a,b]
Abstract: Detailed conformational analyses of our previously reported cyclopropane-based peptidomimetics and conformational analysis-
driven ligand optimization are described. Computational calculations and X-ray crystallography showed that the characteristic features of
cyclopropane function effectively to constrain the molecular conformation in a three-dimensionally diverse manner. Subsequent principal
component analysis revealed that the diversity covers the broad chemical space filled by peptide secondary structures in terms of both main-
chain and side-chain conformations. Based on these analyses, a lead stereoisomer targeting melanocortin receptors was identified, and its
potency and subtype selectivity were improved by further derivatization. The presented strategy is effective not only for designing non-peptidic
ligands from a peptide ligand, but also for the rational optimization of these ligands based on the plausible target-binding conformation without
requiring the three-dimensional structural information of the target and its peptide ligands.
Introduction
target is a G protein-coupled receptor (GPCR).[3] Therefore, an
effective methodology for designing and optimizing
peptidomimetics without 3D structural information of the target
and its peptide ligands is strongly needed.
Peptidomimetics, in which natural bioactive peptides are
converted into non-peptidic drug-like molecules, are an
important concept in modern drug development.[1] The essence
of peptidomimetic design is to mimic the secondary structure of
a key amino acid sequence in the parent peptide ligand
interacting with its target.[2] The rational design and optimization
of peptidomimetics, however, are challenging due to the inherent
flexibility of peptides. This is especially true when the defined
three-dimensional (3D) structural information on the peptide-
target interaction is not available, as is often the case when the
One of the most well-investigated approaches for
developing peptidomimetics targeting peptide-protein and
peptide-receptor interactions is to mimic the folded secondary
structures, such as β-turns, of peptide ligands.[4] This is mainly
because a number of bioactive peptides are recognized by the
target proteins while in their folded conformations.[5] While
promising, however, this approach is often unsuccessful due to
the difficulty of mimicking the exact conformations recognized by
the target, due to the high conformational diversity of peptide
secondary structures.[6] For example, there are more than eight
types of β-turn structures, which makes it difficult to predict
which β-turn structure is important for binding.[7] Further, in some
cases, key amino acid residues interacting with the target
protein might assume a more extended conformation than the
turn structures.[8] The design of peptidomimetics, therefore,
requires the incorporation of “3D structural diversity” to cover the
broad array of peptide secondary structures, i.e., from folded to
extended forms, including their halfway structures, rather than
focusing on mimicking only one of them.
Cyclopropane effectively restricts the conformation of a
molecule in the cis (folded)- or trans (extended)-form as shown
in Figure 1a. This structural property, cis/trans-restriction, is
widely used in medicinal chemistry,[9] including the field of
peptidomimetics.[10] In addition, cyclopropane has two other
characteristic features. First, cyclopropanes directly connected
to an sp2 carbon exist preferentially in the bisected conformation,
as shown in Figure 1b, due to the π-donating stereoelectronic
effect of the cyclopropane ring.[11] Second, cis-configured
substituents on a cyclopropane ring mutually exert marked steric
repulsion, because they are fixed in the eclipsed form, referred
to as cyclopropylic strain.[12] Consequently, the rotational
conformation of the substituents on cyclopropane can be
[a]
Dr. A. Mizuno, T. Kuwahara, H. Endoh, Dr. M. Watanabe, Dr. M.
Arisawa, Prof. Dr. S. Shuto
Faculty of Pharmaceutical Sciences
Hokkaido University
Kita-12, Nishi-6, Kita-ku, Sapporo, Hokkaido 060-0812 (Japan)
E-mail: shu@pharm.hokudai.ac.jp
[b]
[c]
[c]
[d]
Prof. Dr. S. Shuto
Center for Research and Education on Drug Discovery
Hokkaido University
Kita-12, Nishi-6, Kita-ku, Sapporo, Hokkaido 060-0812 (Japan)
Dr. T. Kameda
Computational Biology Research Center
National Institute of Advanced Industrial Science and Technology
2-4-7, Aomi, Koutou-ku, Tokyo 135-0064 (Japan)
Dr. Y. Ito, Prof. Dr. S. Yamada
School of Pharmaceutical Sciences
University of Shizuoka
Yata, Suruga-ku, Shizuoka, Shizuoka 422-8526 (Japan)
Dr. K. Hasegawa, Dr. A. Yamano
Rigaku Corporation, Application Laboratories
3-9-12, Matsubara-cho, Akishima, Tokyo 196-8666 (Japan)
Supporting information for this article including full analytical and
experimental procedures as well as copies of 1H and 13C NMR data
is available on the WWW under http://xxxx
This article is protected by copyright. All rights reserved.

10.1002/chem.201605312
Chemistry - A European Journal
FULL PAPER
restricted to minimize the steric repulsion due to the strain, as
shown in Figure 1c.
conformational analyses. In this optimization process, we
investigated the 3D structural information important for the
receptor binding, which led to the efficient identification of a
more potent and subtype-selective MCR ligand.
Results
Conformational Analyses of the Cyclopropane-Based
Peptidomimetics
1) Computational calculations
The designed cyclopropane-based peptidomimetics comprised
eight stereoisomers because of the three consecutive
asymmetric carbons (Figure 3). The direction of the
cyclopropane ring is either up or down. The peptidomimetic
backbone (indicated in red in Figure 3) can be fixed to the cis- or
trans-configuration by the ring, and is also constrained to either
the extended or folded form by the cyclopropylic strain
depending on the stereochemistry of the asymmetric center
adjacent to the ring (C1’). The eight stereoisomers were thus
named I (up/trans-folded), II (up/trans-extended), III (up/cis-
extended), IV (up/cis-folded), ent-I (down/trans-folded), ent-II
(down/trans-extended), ent-III (down/cis-extended), and ent-IV
(down/cis-folded).
Figure 1. Steric and stereoelectronic features of cyclopropane. (a) cis/trans-
restriction; (b) bisected conformational-preference; (c) cyclopropylic strain.
We hypothesized that taking advantage of all of these
characteristic features of cyclopropane would provide a useful
peptidomimetic scaffold with sufficient 3D structural diversity to
mimic various peptide secondary structures. As shown in Figure
2, when the central peptide bond moiety of a tetrapeptide motif
is replaced with a cyclopropane ring, the conformation of the i+2
and i+3 moieties can be controlled by the cis/trans-restriction
and the bisected-conformational preference, respectively, and
the spatial arrangement of the i and i+1 moieties can be
restricted by the cyclopropylic strain, depending on the
stereochemistries of the stereogenic carbon centers. This
hypothesis was examined by designing peptidomimetics
targeting melanocortin receptors (MCRs), a family of GPCRs.
Some of the designed mimetics exhibited the desired binding
affinity to the receptor.[13] Whether or not the cyclopropane-
based peptidomimetics actually have 3D structural diversity as
hypothesized, however, was not experimentally verified, which
limited the utility of these mimetics to some extent. For this
viewpoint, the 3D structures of the peptidomimetics should be
investigated by an appropriate methodology.[14]
Figure 3. Diagram of the structural difference of the eight stereoisomers in the
cyclopropane-based peptidomimetics.
Structural analyses of the four stereoisomers I–IV were
carried out by computational calculations with MacroModel using
the simplified models A (for I, up/trans-folded), B (for II, up/trans-
extended), C (for III, up/cis-extended), and D (for IV, up/cis-
folded), and their most stable conformations are shown in Figure
4. The cyclopropylic strain was observed to be effective in all
four stereoisomers. Their backbone conformation was evaluated
based on the dα value, which is the distance between the two
Figure 2. Design of cyclopropane-based peptidomimetics by replacing the
middle peptide bond moiety in tetrapeptide motifs with a cyclopropane ring.
terminal carbons corresponding to Cαi and Cαi+3 in
a
tetrapeptide motif. The dα values obtained were 7.2 Å for A, 9.5
Å for B, 9.1 Å for C, and 6.2 Å for D; demonstrating that the
backbone conformations were remarkably varied with dα values
ranging from 6.2 to 9.5 Å. Thus, the conformational diversity of
the peptidomimetic backbone could cover a wide range of
peptide secondary structures from β-turns to β-strands.
Thus, in the present study, we performed conformational
analyses of the cyclopropane-based peptidomimetics to clarify
whether the mimetics actually cover the broad chemical space
filled by the diverse peptide secondary structures in terms of
both main-chain and side-chain conformations. Furthermore, we
also performed rational ligand optimization based on the
This article is protected by copyright. All rights reserved.

10.1002/chem.201605312
Chemistry - A European Journal
FULL PAPER
A' (down/trans-folded, ent-I)
cyclopropylic strain
Ph
αi
αi
O
NH
1'
1'
H
N
Ph
αi+3
H
O
bisected conformation
1'
αi+3
dα = 7.6 Å
cyclopropylic strain
1'
B' (down/trans-extended, ent-II)
Figure 4. Most stable conformations for models A–D. The distance between
the two terminal carbons (indicated by αi and αi+3) are shown as dα in each
model structure.
O
H
H
N
Ph
dα = 9.6 Å
N
H
Ph
αi+3
1'
αi
O
2) X-ray Crystallography
bisected conformation
cyclopropylic strain
1'
To corroborate the above calculation results by X-ray
crystallography, the simplified mimetics A’–D’ were designed
and synthesized as racemates, considering the crystallinity. As
shown in Scheme 1, synthesis was based on stepwise
incorporation of the four side-chain moieties (Ri to Ri+3) into
cyclopropane unit 1, which we previously established as a
general synthetic route.[13] For trans-type mimetics, the Ri+2
moiety was introduced into the lactone carbonyl group of 1,
while the ethoxy group of the ester was replaced with the amine
bearing the Ri+3 moiety. For cis-type mimetics, the Ri+2 moiety
was introduced to the ester carbonyl group of 1, while the
lactone was opened with the amine bearing the Ri+3 moiety. The
C1’ stereogenic center was constructed via the Grignard
reaction, introducing the Ri+1 moiety to N-tert-butanesulfinyl
imines 2 and 3. Finally, the amino group of the Grignard
products was coupled with the carboxylic acid bearing the Ri
moiety to give the desired mimetics.[15]
αi+3
αi
C' (up/cis-extended, III)
O
αi
αi
HN
O
1'
1'
H
HN
αi+3
Ph
1'
bisected conformation
αi+3
dα = 8.5 Å
D' (up/cis-folded, IV)
cyclopropylic strain
1'
1'
αi+3
Ph
αi
bisected conformation
CH3
CH3
H
tBu
HN
H3C
C' (cis-extended)
O
H3C
O
O
O
NH
CH3
O
1'
CH3
1'
CH3
O
1'
1'
1'
HN
O
O
NH
Ph
HN
Ph
αi+3
O
NH
HN
Ph
Ph
HN
Ph
CH3
O
HN
Ph
HN
Ph
αi
Ph
dα = 5.5 Å
A' (trans-folded)
B' (trans-extended)
D' (cis-folded)
Grignard reaction
Figure 5. X-ray crystallographic structures of compounds A’-D’. The
conformational restriction due to the cyclopropylic strain and the bisected-
conformational preference are highlighted, and the dα value is shown for each
crystal structure.
incorporate
converted to
sulfinyl imine
O
+1Ri
with Ri +1-CH2MgX
Ri +2 moiety
1'
Ri +2
O
S
O
O
HN
O
Ri +2
tBu
N
O
O
HN
Ri +3
Ri +3
Ri
OEt
HN
coupled with
coupled with
Ri +3-CH2NH2
trans-folded
trans-extended
Ri -CH2CO2H
1
2
As shown in Figure 5, the X-ray crystal structures showed
that both the bisected-conformational preference and the
cyclopropylic strain worked effectively to restrict the
O
Cl
epichlorohydrin
incorporate
Grignard reaction
Ri +2 moiety
with Ri +1-CH2MgX
conformation.
Accordingly,
the
structures
were
well
O
Ri +1
Ri +2
O
Ri +2
OEt
superimposed on the corresponding calculated structures, i.e., A’
on the trans-folded model A, B’ on the trans-extended model B,
C’ on the cis-extended model C, and D’ on the cis-folded model
D, while, in compound C’, the amide bond adjacent to the
cyclopropane ring rotated by approximately 45º from the ideal
bisected conformation, probably due to the steric repulsion
against the spatially close ethyl group. The dα values of the
crystal structures of the synthesized mimetics A’-D’ were almost
O
1'
O
NH
HN
Ri +3
O
Ri +3
O
O
N
HN
S
coupled with
Ri +3-CH2NH2
tBu
converted to
sulfinyl imine
Ri
coupled with
Ri -CH2CO2H
cis-folded
cis-extended
1
3
Scheme 1. Structures of compounds A’–D’ designed for X-ray crystallography,
and the general synthetic scheme of the cyclopropane-based peptidomimetics.
This article is protected by copyright. All rights reserved.

10.1002/chem.201605312
Chemistry - A European Journal
FULL PAPER
consistent with those calculated for models A-D, respectively. In
addition, the dα value of crystalline D’ was shorter than that
tetrapeptide motifs. In addition, the trans-folded scaffolds (I and
ent-I), which had middle PC1 values, mimic the main-chain
conformations between β-turns and β-strands. The PC2 values
of the eight scaffolds differed from each other, indicating that the
two side chains (Ri+1 and Ri+2) are oriented in different and
diverse 3D positions. These analyses demonstrated that the
cyclopropane-based peptidomimetics cover the spatial
arrangement of the side chains differently, even when the main-
chain conformations of the mimetics are analogous.
calculated for model
D by 0.7 Å, suggesting that the
peptidomimetic backbone can span to an even broader extent
than predicted by the computational calculations.
3) Principal Component Analysis
Peptidomimetic structures must be evaluated not only for the
main-chain (backbone) conformation, but also for the side-chain
(Ri to Ri+3) positioning, due to its major role in the interaction with
targets. Accordingly, using the obtained X-ray crystal structures,
the peptidomimetics were compared with the diverse secondary
structures of tetrapeptide motifs by principal component analysis
(PCA), a mathematical method generally used to evaluate the
diversity of molecular characteristics.[16]
(a)
βi +2
Ri +2
Ri +1
Ri +2
βi +2
Ri +1
βi +1
O
αi +2
HN
O
βi +1
O
O
αi +1
N
αi +2
HN
αi +1
NH
H
Ri +3
αi +3
O
O
NH
Ri +3
αi +3
αi
Ri
Ri
NH
Considering the range where the conformational restriction
of cyclopropane can be effective, the six carbon atoms were
selected as the comparison points, which comprised the four α-
carbons of the residues i to i+3 (Cαi to Cαi+3) and the two β-
carbons of the central two residues (Cβi+1 and Cβi+2) indicated by
the magenta points in Figure 6a. To build a chemical space that
describes the conformational diversity of tetrapeptide motifs, 3D
structural data of tetrapeptide motifs (12,410 in total) were
extracted from the X-ray crystal structures of peptides deposited
in PepX, a peptide-protein complex database.[17] PCA was
performed on the 3D coordinates of the six carbon atoms in the
extracted tetrapeptide motifs, and the first and second principal
components (PC1 and PC2) were used to build the chemical
space. As shown in Figure 6b, PC1 (horizontal axis) and PC2
(vertical axis) of each extracted tetrapeptide motif are plotted as
red crosses. PC1 clearly correlated with the distance between
Cαi and Cαi+3 (dα in Figure 7; correlation coefficient r = 0.95),
while PC2 was weakly to moderately correlated with the
distance between Cβi+1 and Cβi+2 (dβ in Figure 7; r = -0.30) and
the dihedral angle of Cβi+1-Cαi+1-Cαi+2-Cβi+2 (θ in Figure 7; r = -
0.55). Based on these structural meanings, PC1 can describe
the folded/extended character of the main chain, while PC2
indicates the 3D positioning of the side chains of the central two
residues i+1 and i+2. Therefore, the broadly distributed red plots
in the chemical space shown in Figure 6b clearly demonstrate
that the conformations of the tetrapeptide motifs in PepX are
diverse in terms of both main-chain conformation and side-chain
positioning.
αi
cyclopropane-based
peptidomimetics
tetrapeptide motifs
(b)
ent-III
ent-II
III
IV
ent-IV
I
II
ent-I
PC1
Figure 6. (a) The six carbon atoms are indicated by the magenta points, which
were selected to build the chemical space. (b) Tetrapeptide motifs in PepX
(red crosses) and X-ray structures of cyclopropane-based peptidomimetic
scaffolds I–IV and ent-I–IV (blue filled squares) in the chemical space based
on the 3D coordinates of the selected six carbon atoms. The horizontal axis
and vertical axis represent PC1 and PC2, respectively.
Ri +2
Cβi+2
Ri +1
O
dβ
θ
Cαi+2
Cβi+1
O
N
H
Cαi+1
HN
HN
Ri +3
O
Cαi+3
Ri
dα
Cαi
The corresponding six carbon atoms in the cyclopropane-
based peptidomimetics were also selected for the PCA (Figure
6a, magenta points). PC1 and PC2, calculated based on the X-
ray crystal structures of the compounds A’–D’, corresponding to
the eight scaffolds I–IV and ent-I–IV, are plotted as blue filled
squares in Figure 6b. As we speculated, the 3D structures of the
peptidomimetics distributed diversely along the PC1 and PC2
axes to mimic both main-chain and side-chain conformations of
the various tetrapeptide motifs in PepX. The cis-folded scaffolds
(IV and ent-IV) had low PC1 values to mimic the folded
conformations, such as β-turns, while the trans-extended
scaffolds (II and ent-II) and the cis-extended scaffolds (III and
ent-III) had high PC1 values to mimic the extended
conformations, such as β-strands, of the main chain of the
Figure 7. Descriptors for the peptide secondary structures (dα, dβ, θ).
Conformational Analysis-Driven Ligand Optimization
1) Identification of the Lead Stereoisomer
In a previous study,[13] the eight stereoisomers were designed as
MCR ligands based on the tetrapeptide sequence (His6-L/D-Phe7-
Arg8-Trp9), which is known as the core sequence of peptidic
MCR ligands.[18] Among them, down/trans-folded mimetic 4 and
down/trans-extended mimetic 5 (Figure 8a) showed definite
affinity for the human MCR subtype 4 (hMC4R). Because the
This article is protected by copyright. All rights reserved.

10.1002/chem.201605312
Chemistry - A European Journal
FULL PAPER
two mimetics showed similar binding affinity for hMC4R (4, Ki =
0.38 µM; 5, Ki = 0.37 µM), however, it was required to determine
which of these mimetics could be the lead stereoisomer for
further ligand optimization.
Based on the above-mentioned conformational analyses,
stereoisomers 4 and 5 should be apart in the chemical space
(corresponding to ent-I and ent-II, respectively, in Figure 6b),
due to the distinct backbone conformation and side-chain
positioning. As depicted in Figure 8a, the two mimetics are
opposite at the C1’ stereochemistry, and thus the 3D
orientations of the i and i+1 moieties are differently constrained
due to the strong cyclopropylic strain effect, as determined by
diastereoselective Grignard reaction[19] to chiral N-tert-
butylsulfinyl imines 12 and 13. The diastereoselectivity and yield
were sufficient under normal conditions (0.1
concentration at rt). This observation was in contrast to the
same reaction with PhCH2CH2MgCl, in which low
M substrate
a
concentration and high temperature (0.01 M at 110 ºC) were
effective, and only low yield (30%) was produced when the
substrate was 13.[13] The major Grignard adducts were
deprotected with HCl/AcOEt, followed by coupling with
carboxylic acid 20 or 21. The resulting crude amides were
treated with K2CO3/MeOH. For the synthesis of 24-27, the Fmoc
group was exchanged with an Ac group in one-pot by the
the calculations (Figure 4) and X-ray crystal structures (Figure 5). addition of excess amounts of Ac2O after treatment with K2CO3
We speculated that the two functional groups of the i and i+1
moieties, i.e., the imidazolyl and phenyl groups, in mimetics 4
and 5 might be ambiguously recognized by the receptor, as both
are planar aromatic rings with π-binding ability tethered via a
conformationally flexible “-CH2-CH2-” chain (indicated with dotted
blue circles in Figure 8a). Therefore, to provide more
conformational rigidity, we designed down/trans-folded mimetics
6, 8, and 10 (derivatives of 4) and down/trans-extended
mimetics 7, 9, and 11 (derivatives of 5), in which the phenethyl
group was shortened to a benzyl group and/or an AcNH- group
was introduced at the i moiety (Figure 8b).
in MeOH (condition B). The primary hydroxyl group of 22-27 was
converted to a guanidino group to give the desired compounds
6-11.
Scheme 2. Synthetic scheme for modified mimetics 6-11.
All of the newly synthesized mimetics were evaluated for
their binding affinity for hMC4R, and the results are summarized
in Table 1. Unfortunately, the binding affinities for hMC4R of
these modified mimetics were lower than those of the parent
mimetics 4 and 5. The difference in the affinities between each
epimeric pair (6 vs. 7, 8 vs. 9, and 10 vs. 11), however, was
larger than that between the parent epimeric pair 4 and 5, as
expected. In these three epimeric pairs, the down/trans-folded
mimetics derived from 4 were significantly more potent than the
corresponding down/trans-extended mimetics (Ki values: 6 < 7,
Figure 8. (a) Diagram of the structural difference between mimetics 4 and 5.
(b) Designed mimetics 6-11, which are modified at the conformationally
flexible methylene chains of 4 and 5 indicated with dotted blue circles.
The synthetic scheme for the modified mimetics 6-11 is
shown in Scheme 2. The benzyl group was introduced by a
This article is protected by copyright. All rights reserved.

10.1002/chem.201605312
Chemistry - A European Journal
FULL PAPER
8 < 9, 10 < 11). These results showed that the down/trans-folded
structure is more preferable for binding to hMC4R, whose turn-
like, but relatively extended, form could effectively mimic the
conformation of the core tetrapeptide sequence (His-L/D-Phe-
Arg-Trp) in its interaction with the receptor.
The binding affinities for the three MCR subtypes (hMC3R,
hMC4R, and hMC5R) are shown in Table 2. As expected, the
binding affinity for the three subtypes increased in all mimetics
except for 36 at hMC3R. Among them, mimetic 34 had the
highest affinity for hMC4R (Ki = 0.032 µM), which was 10 times
as potent as the lead 4. As is the case with mimetic 4, mimetic
34 showed antagonistic activity to hMC4R (IC50 = 0.22 µM),
which was also approximately 10 times more potent. In addition,
34 appeared to be highly selective for hMC4R; the selectivity
indices of hMC3R/hMC4R and hMC5R/hMC4R were 36 and 28,
respectively, while those of 4 were 13 and 5.5. The stability of
the mimetic 34 in human serum at 37 ºC was also tested, as the
additional amide moiety might have made it susceptible to
proteolysis. As a result, 34 had high proteolytic stability (93%
remained after 24-h incubation), compared to the core sequence
tetrapeptide, Ac-His-Phe-Arg-Trp-NH2 (t1/2 < 1 h).
Table 1. Binding affinities of the modified mimetics 6–11 for hMC4R.
mimetic
scaffold
Ki (µM)[a]
6
down/trans-folded
down/trans-extended
down/trans-folded
down/trans-extended
down/trans-folded
down/trans-extended
1.70 ± 0.11
3.57 ± 0.07
0.64 ± 0.17
1.57 ± 0.19
4.00 ± 0.33
6.89 ± 0.66
7
8
9
10
11
Table 2. Effect of mimetics 34–36 for hMC3R, hMC4R, and hMC5R.
mimetic
Ki (µM)[a]
hMC3R
IC50 (µM)[c]
[a] Mean value ± S.E. (n = 3).
hMC4R
hMC5R
hMC4R
34%
4
4.90 ± 0.27
0.38 ± 0.05
2.09 ± 0.11
at 1µM
98%
at 10µM
2) Development of More Potent and Selective ligands
Based on the identification of the down/trans-folded mimetic 4 as
the lead stereoisomer, the development of more potent ligands
was investigated. Capping the terminal amino group of the
34
35
36
1.16 ± 0.11
0.77 ± 0.08
50.7%[b]
0.032 ± 0.001
0.138 ± 0.024
0.235 ± 0.022
0.894 ± 0.288
0.233 ± 0.012
0.395 ± 0.037
0.22
nd
tetrapeptide MCR ligand (H-His-D-Phe-Arg-Trp-NH2) with
hydrophobic group increased the binding affinity for the MCR
subtypes,[20] suggesting that
hydrophobic pocket existed
a
a
nd
around the MCR binding site of the terminal amino group of the
core sequence. Therefore, if the down/trans-folded structure
mimics the receptor-binding conformation of the core sequence,
introducing a hydrophobic group onto the i moiety of 4 would
increase the MCR binding affinity, hopefully with higher subtype
selectivity. Accordingly, derivatives 34-36 with an N-terminal
hydrophobic group were designed (Figure 9) and synthesized
following the analogous procedure for the synthesis of 8 with the
use of the corresponding carboxylic anhydride instead of Ac2O
at the K2CO3/MeOH treatment step.
[a] Mean value ± S.E. (n = 3). [b] Inhibition rate of 125I-labeled [Nle4,D-
Phe7]-α-MSH binding at 10 µM. [c] Inhibitory effect on the agonistic
activity of 4.7 nM of MT-II. For mimetic 4, inhibition rates at 1 µM and 10
µM are shown. nd, not determined.
Discussion
A combination of computer-aided and organic synthesis-aided
methods was applied to conformational analysis in this study. In
the computational calculations (Figure 4) and X-ray
crystallography (Figure 5), the backbone conformations of the
mimetics were assessed based on the dα value to show their
high diversity spanning folded to extended forms. The distance
between α-carbons is easy to calculate and often used in
structural studies of peptides and peptidomimetics;[4] however, it
only describes the folded/extended character of the main-chain
conformation. Although the main-chain conformation is important,
it is essential to estimate the 3D positioning of the side chains,
which often play a major role in the interaction with target
molecules.
Figure 9. Designed mimetics 34-36 with a hydrophobic group at the N-
terminal moiety.
To estimate the side-chain positioning, Garland and Dean
used the Cα-Cβ bond vector as a descriptor.[14c] While it is an
This article is protected by copyright. All rights reserved.

10.1002/chem.201605312
Chemistry - A European Journal
FULL PAPER
excellent descriptor, it is not easy to use a vector parameter to
changes, e.g., SHU9119 and SHU8941, in which the phenyl
group of D-Phe7 is replaced with a 2-naphthyl or p-iodophenyl
group, respectively.[22] The conformations of MT-II and SHU-
9119 are reportedly quite similar except for some difference in
the 3D positions of the side-chain functional groups.[23]
Interestingly, a single mutation of Leu133 to Met in hMC4R
converts SHU9119 from an antagonist to an agonist without
affecting its binding affinity.[24] Considering these functional
fluctuations among peptidic MCR ligands, it can be deduced that
the antagonistic activity of mimetics 4 and 34 is due to the 3D
positions of the side-chain functional groups rather than the
overall structure of the molecule. This is also supported by the
effect of the introduction of an N-terminal hydrophobic group in
34, which increased MCR binding affinity, as similarly observed
in the peptidic MCR agonists.
evaluate the structural analogy between
a peptidomimetic
scaffold and a peptide secondary structure. Instead, Burgess et
al. used the distance between the atoms corresponding to Cβ as
a scalar descriptor,[14b] which is presented as dβ in our study.
Because dβ does not reflect the orientation of the side chains,
however, it works as a "rough cut" parameter, as stated in their
paper.[14b] Therefore, an alternative analysis was required for
more accurate fitting of the 3D positioning of the side chains.
We used PCA to generate the desired parameters
correlating with the structural information of both the main chain
and side chains (Figure 6). PC2 was moderately correlated with
the θ value; therefore, the 3D positioning of side chains with
similar dβ values can be evaluated separately depending on the
θ value, which indicates the orientation of the side chains.
Notably, as depicted in Figure 10, side chains Ri+1 and Ri+2 in the
cyclopropane-based peptidomimetics are oriented opposite from
each other in space in the enantiomeric pairs, i.e., the up-series
vs. down-series, based on a combinational steric effects of the
cis/trans-restriction and the cyclopropylic strain. This 3D
structural difference between the enantiomers leads to a variety
of θ values, resulting in diversity along the PC2 axis and making
it possible to mimic a wide range of spatial arrangements of
tetrapeptide motifs in terms of the side-chain positioning.
O
αi +3
βi +1
αi +1
O
αi +3
N
H
Ph
αi +2
H
N
αi
N
H
Ph
βi +2
αi +2
Ph
O
NH
βi +2
αi +1
βi +1
O
αi
Ph
A' (down/trans -folded)
B' (down/trans -extended)
RMSD = 0.70
RMSD = 2.10
D
-Phe
D
-Phe
Arg
αi +3
Arg
Ri +2
Ri +2
Ri +1
Ri +1
O
O
O
NH
HN
Ri +3
O
NH
HN
Ri +3
Trp
αi
αi
αi +3
Trp
αi +3
αi +3
αi
Ri
Ri
αi
His
up-series
down-series
His
Figure 10. The opposite spatial arrangements of the side chains Ri+1 and Ri+2
between the up-series and down-series.
dα = 7.6 Å for A', 7.9 Å for MT-II
dα = 9.6 Å for B', 7.9 Å for MT-II
HN
D
-Phe
NH2
While the overall conformation is restricted, the 3D
positions of the functional groups themselves showed some
flexibility in the designed MCR ligands. This flexibility might have
made the ligand-receptor interaction loose, as observed in 4 and
5. The two mimetics showed similar binding affinity for hMC4R,
despite having a distinct 3D structure. This contradiction was
addressed by the conformational analysis-driven strategy
(Figure 8). The methylene chains tethering Ri and Ri+1 in 4 and 5
were modified to differentiate their interaction with the receptor,
employing the strong cyclopropylic strain effect. As a result, it
was indicated that the down/trans-folded mimetic 4 could mimic
the receptor-binding conformation of the core sequence.
Accordingly, mimetic 4 was successfully derivatized to the 10-
times more potent hMC4R ligand 34 (Ki = 32 nM) with good
subtype selectivity and proteolytic stability.
Arg
NH
βi +2
O
βi +1
H
N
αi +2
H
N
αi +1
His
N
H
O
NH
Trp
αi +3
NH
O
N
αi
O
O
NH
NH
H
N
O
NH2
O
NH
O
HN
MT-II
O
Figure 11. Superimposition of the reported conformation of MT-II (green) and
the X-ray crystallographic structures of compounds A’ (orange) and B’
(magenta). The superimposing points for root-mean-square deviation (RMSD)
calculations are Cαi, Cαi+1, Cαi+2, Cαi+3, Cβi+1, and Cβi+2 in MT-II and the
corresponding carbons in compounds A’ and B’ (indicated by red points).
The mimetics 4 and 34 had antagonistic activity to hMC4R.
This antagonistic activity can be explained by a difference of the
3D positions of the side-chain functional groups from those
required for agonistic activity. The well-known peptidic MCR
agonist MT-II[21] is converted to an antagonist by subtle structural
This article is protected by copyright. All rights reserved.

10.1002/chem.201605312
Chemistry - A European Journal
FULL PAPER
The information on the receptor-binding conformation of
the core sequence obtained by our organic chemistry-based
approach was indeed consistent with that obtained by an
independent spectroscopic and computational approach.
Carotenuto et al previously reported the NMR-based
conformational calculations of MT-II.[23b] The X-ray crystal
structures of the simplified mimetics A’ and B’ (corresponding to
4 and 5, respectively) were superimposed on the calculated
different calculated models on the β-turn structure were used for
the linear peptidic MCR ligand, NDP-MSH,[27] i.e., type II’ β-turn
composed of the core sequence itself (His6-D-Phe7-Arg8-Trp9)[28]
and type I’ β-turn composed of the tetrapeptide shifted from the
core sequence by one residue (Glu5-His6-D-Phe7-Arg8).[29] The
PC1 value of MT-II distributes in the middle to high range,
supporting that the conformation of the core sequence would be
relatively extended rather than forming
a β-turn structure.
structure of the core sequence in MT-II. As depicted in Figure 11, Therefore, an effective peptidomimetic backbone targeting the
the X-ray crystal structure of A’ was well superimposed on the
calculated MT-II structure with the almost same dα values,
whereas that of B’ was not. This difference was quantified by the
MCRs could be extended or halfway between the folded and the
extended forms. This atypical preference for the extended
conformation might have been covered by the PC1 diversity of
the cyclopropane-based peptidomimetics, resulting in the
identification of the turn-like but more extended structure
(halfway between the folded and the extended forms) of ent-I.
The negative PC2 value of MT-II is mainly attributed to the D-
configuration of the D-Phe7 residue. The D-Phe7 residue is
important for the increased MCR affinity of many peptidic MCR
ligands, including MT-II and NDP-MSH. Thus, this unique side-
chain positioning arising from the D-configuration was covered
by the PC2 diversity of the designed mimetics.
root-mean-square deviation values calculated for Cαi, Cαi+1
,
Cαi+2, Cαi+3, Cβi+1, and Cβi+2 in MT-II and the corresponding
carbon atoms in A’ and B’ (indicated by red points in Figure 11),
which were 0.70 for A’ and 2.10 for B’, respectively. The 3D
structures of our peptidomimetics were also compared with the
reported low-energy conformations of MT-II[23b] in the chemical
space based on PC1 and PC2. As shown in Figure 12, the plots
for the low-energy conformations of the core sequence in MT-II
(green squares) distributes near the plot for the scaffold ent-I
rather than that for scaffold ent-II. These results support that the
down/trans-folded structure of 4, identified as the lead
stereoisomer, would be similar to the conformation of the core
sequence in MT-II from the viewpoint of both the main-chain
conformation and the side-chain positioning.[25]
scaffold ent-II
(down/trans-extended)ꢀ
scaffold ent-I
(down/trans-folded)ꢀ
low-energy simulated structures of MT-IIꢀ
Figure 12. Comparison of the cyclopropane-based peptidomimetics with the
NMR-based calculated conformations of the core sequence in MT-II in the
chemical space. x-axis, PC1; y-axis, PC2; green squares, the low-energy
simulated conformations of the core sequence in MT-II; blue squares, the eight
scaffolds of the cyclopropane-based peptidomimetics; red crosses,
tetrapeptide motifs in the PepX database.
Figure 13. Examples of other approaches to obtain MCR ligands.
Examples of MCR ligands obtained by other approaches
are shown in Figure 13. There are several reports focusing on β-
turn mimetics. Although successful examples such as 38 were
reported,[26b] the potency and subtype selectivity were not high,
and β-turn mimetic scaffold 37 was not so effective to obtain
MCR ligands.[30] These results might be due to the preference of
MCRs for extended conformation of peptidomimetics as
described above. As another strategy different from
peptidomimetics, high-throughput screening of a GPCR-focused
small molecule library (2025 compounds)[31] and following
The comparison in the chemical space (Figure 12) also
visualizes how the 3D structural diversity-oriented strategy
worked to identify ent-I as a lead scaffold. Because the
secondary structure around the core sequence in the peptidic
MCR ligands is thought to be a β-turn,[18] a peptidomimetic
strategy targeting MCRs often focuses on β-turn mimetic
scaffolds,[26]
as
typically
used
for
GPCR-targeting
peptidomimetics.[5] It is unclear, however, whether the core
tetrapeptide sequence itself forms the β-turn. For example, two
This article is protected by copyright. All rights reserved.

10.1002/chem.201605312
Chemistry - A European Journal
FULL PAPER
optimization developed hMC4R antagonist 39.[32] Its potency and
subtype selectivity are comparable to those of 34, indicating that
the only eight stereoisomers designed in this study could work
as effectively as such larger library.
Filippov, J. H. van Boom, R. Grisshammer, M. Baldus, Proc. Natl. Acad.
Sci. USA 2003, 100, 10706-10711.
[9]
a) P. D. Armstrong, J. G. Cannon, J. P. Long, Nature 1968, 220, 65-66;
b) K. Shimamoto, Y. Ohfune, J. Med. Chem. 1996, 39, 407-423; c) T.
Onishi, C. Mukai, R. Nakagawa, T. Sekiyama, M. Aoki, K. Suzuki, H.
Nakazawa, N. Ono, Y. Ohmura, S. Iwayama, M. Okunishi, T. Tsuji, J.
Med. Chem. 2000, 43, 278-282; d) K. Vervisch, M. D'hooghe, K. W.
Tornroos, N. De Kimpe, Org. Biomol. Chem. 2009, 7, 3271-3279; e) K.
Yoshida, K. Yamaguchi, A. Mizuno, Y. Unno, A. Asai, T. Sone, H.
Yokosawa, A. Matsuda, M. Arisawa, S. Shuto, Org. Biomol. Chem.
2009, 7, 1868-1877; f) H. Zhang, W. Tückmantel, J. B. Eaton, P.-w.
Yuen, L.-F. Yu, K. M. Bajjuri, A. Fedolak, D. Wang, A. Ghavami, B.
Caldarone, N. E. Paterson, D. A. Lowe, D. Brunner, R. J. Lukas, A. P.
Kozikowski, J. Med. Chem. 2012, 55, 717-724; g) M. Boztaş, Y.
Çetinkaya, M. Topal, İ. Gülçin, A. Menzek, E. Şahin, M. Tanc, C. T.
Supuran, J. Med. Chem. 2015, 58, 640-650.
Conclusions
Focusing on the importance of the 3D structural diversity of
molecules in searching for and mimicking a key conformation,
computational calculations and X-ray crystallography confirmed
that the steric and stereoelectronic features of cyclopropane
work effectively to constrain the peptidomimetic molecules into
diverse 3D structures. This 3D structural diversity was depicted
by the PCA, which showed that the cyclopropane-based
peptidomimetics cover the broad chemical space filled by
peptide secondary structures in terms of both main-chain and
side-chain conformations. These detailed conformational
analyses allowed us to obtain the 3D structural information that
[10] a) K. I. Varughese, C. H. Wang, H. Kimura, C. H. Stammer, Int. J. Pept.
Protein Res. 1988, 31, 299-300; b) K. Burgess, K.-K. Ho, B. Pal, J. Am.
Chem. Soc. 1995, 117, 3808-3819; c) P. Wipf, J. Xiao, Org. Lett. 2005,
7, 103-106; d) J. E. DeLorbe, J. H. Clements, M. G. Teresk, A. P.
Benfield, H. R. Plake, L. E. Millspaugh, S. F. Martin, J. Am. Chem. Soc.
2009, 131, 16758-16770; e) S. S. Bhella, M. Elango, M. P. S. Ishar,
Tetrahedron 2009, 65, 240-246; f) L. Sernissi, M. Petrović, D. Scarpi, A.
Guarna, A. Trabocchi, F. Bianchini, E. G. Occhiato, Chem. Eur. J. 2014,
20, 11187-11203.
is important for rational ligand optimization. As
a result,
down/trans-folded mimetic 4 was selected, leading to more
potent and subtype-selective MCR ligand 34. The process
described in this study was carried out without knowing the 3D
structural information of the target and its peptide ligands.
Therefore, this peptidomimetic strategy can be applied even
when the ligand-target interaction is poorly documented, as is
often the case in the early stage of drug development.
[11] a) Y. Kazuta, H. Abe, T. Yamamoto, A. Matsuda, S. Shuto, J. Org.
Chem. 2003, 68, 3511-3521; b) A. J. Cruz-Cabeza, F. H. Allen, Acta
Crystallogr. Sect. B: Struct. Sci. 2011, 67, 94-102.
[12] a) S. Shuto, S. Ono, Y. Hase, N. Kamiyama, H. Takada, K. Yamashita,
A. Matsuda, J. Org. Chem. 1996, 61, 915-923; b) M. Watanabe, T.
Hirokawa, T. Kobayashi, A. Yoshida, Y. Ito, S. Yamada, N. Orimoto, Y.
Yamasaki, M. Arisawa, S. Shuto, J. Med. Chem. 2010, 53, 3585-3593.
[13] A. Mizuno, S. Miura, M. Watanabe, Y. Ito, S. Yamada, T. Odagami, Y.
Kogami, M. Arisawa, S. Shuto, Org. Lett. 2013, 15, 1686-1689.
Acknowledgements
[14] Methods for defining peptidomimetic structure have been reported. For
example, a) using distance between Cα atoms, U. Rosenström, C.
Sköld, B. Plouffe, H. Beaudry, G. Lindeberg, M. Botros, F. Nyberg, G.
Wolf, A. Karlén, N. Gallo-Payet, A. Hallberg, J. Med. Chem. 2005, 48,
4009-4024; b) using distance between Cβ atoms, E. Ko, J. Liu, K.
Burgess, Chem. Soc. Rev. 2011, 40, 4411-4421; c) using Cα-Cβ bond
vector, S. L. Garland, P. M. Dean, J. Comput. Aided Mol. Des. 1999, 13,
485-498; d) using dihedral angle, P. Gillespie, J. Cicariello, G. L. Olson,
Biopolymers 1997, 43, 191-217. For recent review of the classification
of peptidomimetics, e) M. Pelay-Gimeno, A. Glas, O. Koch, T. N.
Grossmann, Angew. Chem. Int. Ed. 2015, 54, 8896-8927.
We are grateful to Prof. Dr. A. Carotenuto (University of Naples)
for providing the calculated structural data of MT-II and also to
Sanyo Fine Co., Ltd. for the gift of the chiral epichlorohydrins.
This investigation was supported by Grant-in-Aids for Scientific
Research (21390028 to S.S. and 2310550201, 2179000300 to
M.A.) from the Japan Society for the Promotion of Science.
Keywords: Peptidomimetics • Drug design • Three-dimensional
structural diversity • Chemical space • Conformation analysis
[15] For the synthesis of compounds A'-D', see Supporting Information. The
supplementary crystallographic data in CIF format can be obtained from
The
Cambridge
Crystallographic
Data
Centre
via
[1]
[2]
[3]
I. Avan, C. D. Hall, A. R. Katritzky, Chem. Soc. Rev. 2014, 43, 3575-
3594.
http://www.ccdc.cam.ac.uk/products/csd/request/ with the CCDC
numbers: CCDC 865592, CCDC 865593, CCDC 865594 and CCDC
865595.
R. M. J. Liskamp, D. T. S. Rijkers, J. A. W. Kruijtzer, J. Kemmink,
ChemBioChem 2011, 12, 1626-1653.
[16] a) C. M. Dobson, Nature 2004, 432, 824-828; b) T. Kameda, S. Takada,
Proc. Natl. Acad. Sci. USA 2006, 103, 17765-17770; c) J.-L. Reymond,
M. Awale, ACS Chem. Neurosci. 2012, 3, 649-657.
a) D. Chatenet, B. Folch, D. Feytens, M. Létourneau, D. Tourwé, N.
Doucet, A. Fournier, J. Med. Chem. 2013, 56, 9612-9622; b) L. Karhu,
A. Turku, H. Xhaard, BMC Struct. Biol. 2015, 15, 9.
[17] P. Vanhee, J. Reumers, F. Stricher, L. Baeten, L. Serrano, J.
Schymkowitz, F. Rousseau, Nucleic Acids Res. 2010, 38, D545-D551.
[18] E. M. Haslach, J. W. Schaub, C. Haskell-Luevano, Bioorg. Med. Chem.
2009, 17, 952-958.
[4]
[5]
[6]
L. R. Whitby, Y. Ando, V. Setola, P. K. Vogt, B. L. Roth, D. L. Boger, J.
Am. Chem. Soc. 2011, 133, 10184-10194.
G. Ruiz-Gómez, J. D. A. Tyndall, B. Pfeiffer, G. Abbenante, D. P. Fairlie,
Chem. Rev. 2010, 110, PR1-PR41.
[19] M. T. Robak, M. A. Herbage, J. A. Ellman, Chem. Rev. 2010, 110,
3600-3740.
R. Haubner, D. Finsinger, H. Kessler, Angew. Chem. Int. Ed. Engl.
1997, 36, 1374-1389.
[20] a) L. N. Koikov, F. H. Ebetino, M. G. Solinsky, D. Cross-Doersen, J. J.
Knittel, Bioorg. Med. Chem. Lett. 2003, 13, 2647-2650; b) J. R. Holder,
F. F. Marques, Z. Xiang, R. M. Bauzo, C. Haskell-Luevano, Eur. J.
Pharmacol. 2003, 462, 41-52; c) L. N. Koikov, F. H. Ebetino, M. G.
[7]
[8]
K. C. Chou, Anal. Biochem. 2000, 286, 1-16.
a) M. A. Ceruso, D. F. McComsey, G. C. Leo, P. Andrade-Gordon, M. F.
Addo, R. M. Scarborough, D. Oksenberg, B. E. Maryanoff, Bioorg. Med.
Chem. 1999, 7, 2353-2371; b) S. Luca, J. F. White, A. K. Sohal, D. V.
This article is protected by copyright. All rights reserved.

10.1002/chem.201605312
Chemistry - A European Journal
FULL PAPER
Solinsky, D. Cross-Doersen, J. J. Knittel, Bioorg. Med. Chem. Lett.
2004, 14, 3997-4000.
[21] F. Al-Obeidi, M. E. Hadley, B. M. Pettitt, V. J. Hruby, J. Am. Chem. Soc.
1989, 111, 3413-3416.
[22] V. J. Hruby, D. S. Lu, S. D. Sharma, A. L. Castrucci, R. A. Kesterson, F.
A. Alobeidi, M. E. Hadley, R. D. Cone, J. Med. Chem. 1995, 38, 3454-
3461.
[23] a) M. Cai, C. Cai, A. V. Mayorov, C. Xiong, C. M. Cabello, V. A.
Soloshonok, J. R. Swift, D. Trivedi, V. J. Hruby, J. Peptide Res. 2004,
63, 116-131; b) P. Grieco, D. Brancaccio, E. Novellino, V. J. Hruby, A.
Carotenuto, Eur. J. Med. Chem. 2011, 46, 3721-3733.
[24] Y. Yang, M. Chen, Y. Lai, I. Gantz, K. E. Georgeson, C. M. Harmon, J.
Biol. Chem. 2002, 277, 20328-20335.
[25] The plot for up/trans-extended scaffold II is also close to the plots for
MT-II in the chemical space (Figure 12). However, more detailed
conformational analysis showed that II occupied an area apart from
MT-II, while the area occupied by ent-I significantly overlapped with
that by MT-II. For details, see section VII and Figure S6 in Supporting
Information.
[26] a) C. Haskell-Luevano, A. Rosenquist, A. Souers, K. C. Khong, J. A.
Ellman, R. D. Cone, J. Med. Chem. 1999, 42, 4380-4387; b) T. R.
Webb, L. Jiang, S. Sviridov, R. E. Venegas, A. V. Vlaskina, D. McGrath,
J. Tucker, J. Wang, A. Deschenes, R. Li, J. Comb. Chem. 2007, 9, 704-
710.
[27] T. K. Sawyer, P. J. Sanfilippo, V. J. Hruby, M. H. Engel, C. B. Heward,
J. B. Burnett, M. E. Hadley, Proc. Natl. Acad. Sci. USA 1980, 77, 5754-
5758.
[28] K. L. Chapman, G. K. Kinsella, A. Cox, D. Donnelly, J. B. C. Findlay, J.
Mol. Biol. 2010, 401, 433-450.
[29] M.-K. Cho, C.-J. Lee, C.-H. Lee, S.-Z. Li, S.-K. Lim, J.-H. Baik, W. Lee,
J. Struct. Biol. 2005, 150, 300-308.
[30] B. Wels, J. A. W. Kruijtzer, K. M. Garner, R. A. H. Adan, R. M. J.
Liskamp, Bioorg. Med. Chem. Lett. 2005, 15, 287-290.
[31] K. Lavrador, B. Murphy, J. Saunders, S. Struthers, X. Wang, J. Williams,
J. Med. Chem. 2004, 47, 6864-6874.
[32] W. Jiang, F. C. Tucci, C. W. Chen, M. Arellano, J. A. Tran, N. S. White,
D. Marinkovic, J. Pontillo, B. A. Fleck, J. Wen, J. Saunders, A. Madan,
A. C. Foster, C. Chen, Bioorg. Med. Chem. Lett. 2006, 16, 4674-4678.
This article is protected by copyright. All rights reserved.

10.1002/chem.201605312
Chemistry - A European Journal
FULL PAPER
FULL PAPER
Akira Mizuno, Tomoshi Kameda, Tomoki
Kuwahara, Hideyuki Endoh, Yoshihiko
Ito, Shizuo Yamada, Kimiko Hasegawa,
Akihito Yamano, Mizuki Watanabe,
Mitsuhiro Arisawa and Satoshi Shuto*
Cyclopropane-Based
Detailed conformational analyses revealed that cyclopropane provides
conformationally restricted peptidomimetics with high three-dimensional (3D)
structural diversity, mimicking broad peptide secondary structures. The presented
strategy is effective not only for designing non-peptidic ligands, but also for rational
optimization of these ligands based on the plausible target-binding conformation
without requiring the 3D structural information of the target and its peptide ligands.
Peptidomimetics Mimicking Wide-
Ranging Secondary Structures of
Peptides: Conformational Analysis
and Their Use in Rational Ligand
Optimization
This article is protected by copyright. All rights reserved.