α-aminoxy peptoids: A unique peptoid backbone with a preference for cis-amide bonds

Article abstract of DOI:10.1002/chem.201605100

α-Peptoids, or N-substituted glycine oligomers, are an important class of peptidomimetic foldamers with proteolytic stability. Nevertheless, the presence of cis/transamide bond conformers, which contribute to the high flexibility of α-peptoids, is conside

Full text of DOI:10.1002/chem.201605100

A Journal of
Accepted Article
Title: α-Aminoxy Peptoids: A Unique Peptoid Backbone with a
Preference for Cis-Amide Bonds
Authors: Viktoria Krieger, Emanuele Ciglia, Roland Thoma, Vera
Vasylyeva, Benedikt Frieg, Nader de Sousa Amadeu, Thomas
Kurz, Christoph Janiak, Holger Gohlke, and Finn Kristian
Hansen
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.201605100
Link to VoR: http://dx.doi.org/10.1002/chem.201605100
Supported by

10.1002/chem.201605100
Chemistry - A European Journal
FULL PAPER
α-Aminoxy Peptoids: A Unique Peptoid Backbone with a
Preference for Cis-Amide Bonds
Viktoria Krieger,[a] Emanuele Ciglia,[a] Roland Thoma,[b] Vera Vasylyeva,[b] Benedikt Frieg,[a] Nader de
Sousa Amadeu,[b] Thomas Kurz,[a] Christoph Janiak,[b] Holger Gohlke,[a] and Finn K. Hansen*[a,c]
Abstract: α-Peptoids, or N-substituted glycine oligomers, are an
important class of peptidomimetic foldamers with proteolytic stability.
Nevertheless, the presence of cis-/trans-amide bond conformers,
which contribute to the high flexibility of α-peptoids, is considered as
a major drawback. A modified peptoid backbone with an improved
control of the amide bond geometry could therefore help to
overcome this limitation. Here we have performed the first thorough
analysis of the folding propensities of α-aminoxy peptoids (or
N-substituted 2-aminoxyacetic acid oligomers). To this end, the
amide bond geometry and conformational properties of a series of
model α-aminoxy peptoids were investigated using 1D and 2D NMR
experiments, X-ray crystallography, NBO analysis, CD spectroscopy,
and MD simulations revealing a unique preference for cis-amide
bonds even in the absence of cis-directing side chains. The
conformational analysis based on the MD simulations revealed that
α-aminoxy peptoids can adopt helical conformations that can mimic
the spatial arrangement of peptide side chains in a canonical α-helix.
Given their ease of synthesis and conformational properties,
α-aminoxy peptoids represent a new member of the peptoid family
capable of controlling the amide isomerism while maintaining the
potential for side-chain diversity.
reduce their binding affinity to proteins. In contrast to peptides, a
high degree of cis-amide bonds is observed in α-peptoids.[8] The
cis- and trans-amide conformations are almost isoenergetic for
N-alkyl α-peptoid monomers, and studies have shown that the
nature of the side chain can significantly modulate the ratio of
cis-/trans-amide bond conformers.[9] Nevertheless, α-peptoids
can form stable, helical secondary structures when the
cis-/trans-amide isomerism is optimally controlled.
O
N
O
N
R
O
R
O
-peptoid
-peptoid
(trans-amide)
(cis-amide)
O
O
O
O
O
N
O
N
R
R
-aminoxy peptoid
-aminoxy peptoid
(trans-alkoxyamide)
(cis-alkoxyamide)
Figure 1. Selected peptoid backbones.
Similarly, it has been hypothesized that β-peptoids (N-alkyl
β-alanines) can exhibit foldamer properties.[10-14] A recent study
by Olsen and co-workers provided a high-resolution X-ray
crystal structure of a β-peptoid hexamer containing strongly
cis-inducing N-(S)-1-(1-naphthyl)ethyl side chains. The crystal
Introduction
structure disclosed all-cis amide bond geometry and
a
right-handed helical conformation with exactly three residues per
turn and a helical pitch of 9.6-9.8 Å between turns.[15] However, it
is important to note that the amide bond geometry in β-peptoids
is also highly dependent on the nature of the side chain.[16]
α-Peptoids, or oligomers of N-substituted glycine, have several
advantages over peptides as potential bioactive compounds
including proteolytic stability and increased cell permeability.[1,2]
Peptoid libraries have been utilized as protein binding agents
and as inhibitors of protein-protein interactions,[3-6] although
primary screening hits identified from peptoid libraries have
usually not displayed high activity or potency.[7] The major
limitation of peptoids is the lack of conformational constraints
due to the absence of internal hydrogen bonding, which may
Thus far, the control of the folding properties of α- and
β-peptoids is mainly achieved by incorporation of specific cis- or
trans-directing side chains. This focus on a small number of side
chains comes at the expense of diversity.[7,17] Accordingly, it is
worthwhile to address this limitation and to aim at the design of
peptoid backbones with an improved control of cis-/trans-amide
isomerism. In 2002, Shin and Park reported the preparation of
α-aminoxy peptoid pentamers, but the folding properties were
not investigated.[18] The backbone-controlled secondary
structures of α-aminoxy peptides[19,20] prompted us to analyze
the conformational properties of α-aminoxy peptoids, or
oxa-analogues of β-peptoids, in detail. In this work, we report the
results of our study revealing a unique cis-amide preference of
α-aminoxy peptoids in comparison to other peptoid backbones.
[a]
[b]
[c]
V. Krieger, Dr. E. Ciglia, B. Frieg, Prof. Dr. T. Kurz, Prof. Dr. H.
Gohlke, JProf. Dr. F. K. Hansen, Institute of Pharmaceutical and
Medicinal Chemistry, Heinrich-Heine-Universität Düsseldorf,
Universitätsstr. 1, 40225 Düsseldorf, Germany
R. Thoma, Dr. V. Vasylyeva, Dr. N. de Sousa Amadeu, Prof. Dr. C.
Janiak, Institute of Inorganic and Structural Chemistry, Heinrich-
Heine-Universität Düsseldorf, Universitätsstr. 1, 40225 Düsseldorf,
Germany
JProf. Dr. F. K. Hansen, Pharmaceutical/Medicinal Chemistry,
Institute of Pharmacy, Leipzig University, Brüderstr. 34, 04103
Leipzig, Germany
Results and Discussion
E-mail: finn.hansen@uni-leipzig.de
Supporting information for this article is given via a link at the end of
the document.
We first designed and synthesized a series of model α-aminoxy
peptoids utilizing an N-terminal acetyl and a C-terminal
This article is protected by copyright. All rights reserved.

10.1002/chem.201605100
Chemistry - A European Journal
FULL PAPER
varying concentrations (0.15-150 mM, Figure S3) which showed
no concentration dependency of the peak ratio. Interestingly, a
VT-NMR experiment with the deacetylated analog of α-aminoxy
peptoid
3 (2-((benzylamino)oxy)-1-(piperidin-1-yl)ethan-1-one)
did not show any additional conformational peaks (Figure S4).
Thus, we reasoned the presence of two sets of NMR signals
might indicate the coexistence of two well-defined species that
equilibrate slowly on the NMR time scale, e.g. cis-/trans-amide
bond rotamers. This hypothesis was supported by detection of
EXSY signals in 2D NOESY experiment on 3 in CDCl3 at –30°C
(Figure S5).[23] This behavior is typical for protons under
significant chemical exchange on the saturation time scale and
1
thus implies the existence of rotamers.[24] Subsequent H NMR
Figure 2. α-Aminoxy peptoids investigated. The relevant backbone dihedral
angles are defined: ω [Cα (i – 1); C (i – 1); N; O], ϕ [C (i – 1); N; O, Cα], θ [N;
O; Cα, C], and ψ [O, Cα; C, N (i + 1)], χ1 [NCβ; NCα; N; O].
experiments with compounds 1,2,4-6 at -30°C provided further
evidence for cis-/trans-amide bond conformers. In the case of
compounds 1,4, and 5 we detected a similar behavior as
observed for 3 with ratios[25] of 84:16 (1), 84:16 (4), and 82:18
(5). Notably, the 1H NMR of peptoid 2 (containing a trans-
directing side chain) revealed an increased amount of the minor
conformer at -30°C in CDCl3 (ratio 65:35), whereas α-aminoxy
peptoid 6 (containing a cis-directing side chain) showed well-
defined peaks with no evidence for a significant population of
another conformer. To study the conformational stability of
longer peptoid sequences, α-aminoxy peptoid oligomers 8-10
were synthesized (see Supporting Information). 1H NMR spectra
conducted in CDCl3, acetonitrile-d3 and methanol-d4 showed
piperidinyl cap group to investigate the influence of several side
chains on the amide conformation (see Supporting Information
for synthetic details). To be consistent with peptoid
nomenclature, we use the terms “cis” and “trans” to describe the
amide dihedral angle omega referring to the relative position of
the backbone atoms (see Figure 1). As a consequence, the
cis-conformer features an E-configured hydroxamate moiety,
while the trans-conformer corresponds to
a Z-configured
hydroxamate. The definition of all other relevant dihedral angles
is outlined in Figure 2. α-Aminoxy peptoids 1,3-5 were designed
to investigate the cis-/trans-isomerism in the absence of strongly
cis- or trans-directing side chains. Compound 2 was included
due to its strongly trans-amide inducing phenyl side chain.[21]
1
well-defined signals (see Supporting Information for details). H
NMR spectra measured at -30°C in CDCl3 revealed one major
conformer in all cases. However, in α-aminoxy dipeptoid 8 a
second conformational isomer was observed (ratio 86:14). In the
case of α-aminoxy tripeptoid 9 and α-aminoxy tetrapeptoid 10
only traces of by-peaks were observed.
α-Aminoxy peptoid
6 served as a cis-promoting model
compound due to the cis-directing triazolium side chain.[22]
Model peptoid 7 containing a tert-butyl ester was incorporated to
analyze an alternative C-terminal cap group. α-Aminoxy
peptoids 8-10 were synthesized as model compounds for
oligomeric α-aminoxy peptoids to study conformational
preference in oligomers (see Supporting Information for
synthetic details).
Since several model monomers and oligomers revealed the
presence of two conformers, we performed comprehensive NMR
experiments on α-aminoxy peptoid 3 in CDCl3 to determine the
Arrhenius activation energy for the rotational barrier. For this
purpose, the interconversion rate constants at the respective
coalescence temperatures were measured by VT-NMR
experiments using different NMR instruments. A plot of the
temperature dependence of the interconversion rate constants
afforded the Arrhenius activation energy (Ea = 42 2 kJ mol-1, see
Supporting Information for details).
We first studied the conformational homogeneity of our model
systems using standard 1D NMR techniques at room
temperature. 1H NMR spectra of peptoid monomers 1-6 were
recorded in CDCl3 at room temperature and revealed only a
single set of NMR signals. However, some signals appeared
1
broad. We therefore collected H spectra of monomers 1-6 in
acetonitrile-d3, methanol-d4, D2O and DMSO-d6 observing again
some relatively broad signals in the case of peptoids 1-5 (see
Supporting Information). These results prompted us to study the
1H NMR spectra in more detail. We performed a variable
temperature (VT)-NMR on α-aminoxy peptoid 3 in CDCl3.
Cooling caused the signals of 3 to split into two sets of sharp
signals at -20°C and -30°C in ratios of 87:13 and 86:14,
respectively (Figure S1). In order to exclude that the second
species arises from the piperidinyl group, we repeated the
VT-NMR experiment with model α-aminoxy peptoid 7 containing
a tert-butyl ester as the C-terminal cap. Compound 7 followed
the same trend as observed for α-aminoxy peptoid 3 (Figure S2).
To study whether the additional NMR signals arise from
aggregation, we recorded 1H NMR spectra of 3 at -30°C at
In order to investigate whether α-aminoxy peptoid monomers
prefer cis- or trans-configured amide bonds, we decided to study
the amide bond geometry of our model compounds in detail.
Based on our observed side chain dependency we assumed a
cis-preference in the case of our model α-aminoxy peptoids. To
confirm a preferred cis-amide bond configuration in solution, the
ω torsion was analyzed by 2D NMR spectroscopy. Compound 6
was chosen for a 2D NOESY NMR spectrum due to the absence
of any additional conformational peaks. We observed a strong
cross peak between the N-terminal acetyl group and the
backbone methylene protons as well as a weak NOE between
the acetyl and the side chain methylene protons (Figure S6). We
reasoned that a cis-configured amide (ω ≈ 0°) would fulfill these
distance requirements, whereas
a trans-configured amide
This article is protected by copyright. All rights reserved.

10.1002/chem.201605100
Chemistry - A European Journal
FULL PAPER
(ω ≈ 180°) should show a stronger NOE between the acetyl and
the side chain methylene protons. A 2D NOESY spectrum of
α-aminoxy peptoid 3 at -30°C revealed for the major conformer a
strong cross peak between the N-terminal acetyl group and the
backbone methylene protons which indicates cis-amide
geometry (Figure S7). Notably, our results are in good
agreement with data reported from Grel and co-workers who
showed that simple alkoxyamides prefer cis-amide bonds in the
absence of hydrogen bonds.[26] The above mentioned data
suggest that the α-aminoxy peptoid monomers 1,3-6 possess a
strong preference for cis-configured amide bonds. In particular, it
is worth noting that the model compounds 1,3-5 are showing a
significant preference for cis-amide bonds (> 80%) even in the
absence of any cis-directing side chains. In contrast, α-peptoids
bearing identical cap groups and side chains revealed
inhomogeneous mixtures of cis/trans rotamers or even a
preference for trans-amide bonds (Figure S8).
others, showed that the secondary structure in α-aminoxy
peptides is primarily stabilized by eight-membered ring hydrogen
bonds between C=Oi and N-Hi+2 which are not possible in the
case of α-aminoxy peptoids.[19,20,27] Thus, in order to further
investigate the cis-amide preference in α-aminoxy peptoids, we
performed a natural bond orbital (NBO) analysis on all-methyl
model peptoids in cis-amide and trans-amide configurations
(Figure 4 A+B) as well as the cis-amide α-aminoxy tripeptoid 9
(Figure 4 C).
Figure 4. Results from natural bond orbital analysis. (A), (B) HF/6-31G*-
optimized structures from an all-methyl model compound in cis (A) and trans
(B) configuration. (C) HF/6-31G*-optimized structure of α-aminoxy tripeptoid 9
in cis configuration. In (A – C) the orange dotted lines depict donor-acceptor
interactions. The labels depict the donor-acceptor stabilization energies (in
kcal mol-1) averaged over the two respective oxygen lone pairs, determined by
the second-order perturbation theory analysis.
For the model peptoid in cis-amide configuration, the NBO
analysis revealed Cα,i–H···O=Ci+1 donor-acceptor interactions,
which results in an eight-membered δ turn according to the
classification by Toniolo and Benedetti[28] (Figure 4 A); second
order perturbation theory analysis in the NBO basis revealed a
mean donor-acceptor stabilization energy of 1.56 kcal mol-1
(Figure 4 A, Figure S9 A). The Cα hydrogen atom involved in the
Cα–H···O=C interaction is more electron deficient than the
geminal, non-involved hydrogen (0.2458 au in Cα–H···O=C vs.
0.2281 au in Cα–H; Figure S9 A), indicating that electrons are
delocalized via the Cα–H···O=C interaction. The stabilization
energy computed here is close to the absolute value of the lower
limit of the electronic association energy De of 2.1 kcal mol-1
computed for N,N-dimethylformamide dimers associated via C–
H···O=C hydrogen bonds.[29] The Cα ··· O distance of 3.1 Å
coincides with expectations of distances of weak hydrogen
bonds.[30] As to the trans configuration, no Cα–H···O=C
interaction was revealed by the NBO analysis but rather a Cω,i–
H···O=Ci interaction within the same residue (Figure 4 B); the
latter interaction (stabilization energy: ~1.00 kcal mol-1) is
weaker than the Cα,i–H···O=Ci+1 one (Figure 4 A+B; Figure S9
A+B).
In agreement with results from the all-methyl model peptoid in
cis configuration, NBO analysis also revealed the eight-
membered δ turn in the cis-amide α-aminoxy tripeptoid 9 (Figure
4 C). Here, the donor-acceptor stabilization energy is stronger
than in the model peptoid (1.85 kcal mol-1 in 9 vs. 1.56 kcal mol-1
in the model peptoid; Figure 4 A+C; Figure S9 A+C). This might
be explained by the electron-pushing effect of the benzyl side
chains, leading to the amide oxygen being more negatively
charged (-0.7269 au for 9 vs. -0.6980 au for the model peptoid;
Figure S9 A+C).
Figure 3. Thermal ellipsoid plots of (A) α-aminoxy peptoid
1 and (B)
α-aminoxy tripeptoid 9 (50% probability for 1, 70% for 9, H atoms with arbitrary
radii). The N and C(=O) atom of the N-methylacetamide group in 1 is
disordered in a C2-type (180°) rotation around the O3-O2 connector with equal
occupancy (see Supporting Information for further details); one of the
disordered positions is shown semi-transparent.
Fortunately, we were able to obtain X-ray quality crystals of
α-aminoxy peptoid monomer
1 and tripeptoid 9 by slow
evaporation from CDCl3. The X-ray structure analysis of
α-aminoxy peptoid 1 proved a cis-amide configuration in the
solid state. Although the N-methylacetamide group in α-aminoxy
peptoid 1 was disordered in the crystal structure, it is important
to note that the amide is cis-configured in both the A and B
component (Figure 3). The X-ray structure of α-aminoxy
tripeptoid 9 revealed an all cis-amide geometry (Figure 3). The
observed torsion angles are summarized in Table S1.
Interestingly, the N-terminal (i) and internal (i+1) monomer
disclosed almost identical backbone torsion angles, whereas the
C-terminal monomer (i+2) showed different torsion angles. This
phenomenon can be explained by packing effects in the solid
state (see Supporting Information for a detailed discussion). The
X-ray structure shows the trimer adopting
a
zig-zag
conformation in which the i and i+2 side chains are oriented on
the same side of the backbone, whereas the i+1 benzyl group
points in the opposite direction.
On the first view the cis-preference of α-aminoxy peptoids is
somewhat surprising keeping in mind that α-aminoxy peptides
preferentially adopt trans-amide bonds.[27] However, we, and
This article is protected by copyright. All rights reserved.

10.1002/chem.201605100
Chemistry - A European Journal
FULL PAPER
Scheme 1. Synthesis of α-aminoxy peptoids 19-21 containing α-chiral aromatic side chains (Ns = 2-nitrobenzenesulfonyl).
Taken together, by means of NBO calculations, we found
stabilizing donor-acceptor interactions for both cis and trans
configurations. The formation of an eight-membered δ turn
between residues i→i+1 in the cis configuration is accompanied
itself.[11,15] To verify this hypothesis we synthesized an acetylated
monomer containing the N-(S)-1-phenylethyl group (see
Supporting Information). The CD spectrum of this α-chiral
monomer disclosed a strong minimum at around 220 nm (Figure
S10). Based on this result it appears unlikely that the minimum
at 220 nm is related to secondary structure formation as simple
monomers would not be expected to adopt an ordered
conformation. On the other hand, starting at the length of a
tetramer a shoulder appeared at 209 nm. It could be speculated
that this minimum at 209 nm might be indicative of
by
a
favorable stabilization energy for Cα,i–H···O=Ci+1
interactions, which could explain the cis-amide preference for
α-aminoxy peptoids. Note that, to the best of our knowledge, no
unambiguous report has become available so far for the
occurrence of a δ turn in a linear peptide.[31]
To gain further insight into potential folding propensities of
oligomers we decided to synthesize oligomers featuring α-chiral
side chains (Scheme 1). Circular dichroism (CD) spectroscopy
has been extensively used to study the secondary structure of
α-peptoids. Figure 5 shows the CD spectra of the α-aminoxy
peptoids 19-21 in acetonitrile.
length-dependent
secondary-structure
formation.
Since
characteristic CD signals do not necessarily need to be
indicative of a folding event,[32] further studies on longer
oligomers will be required to investigate the nature of the
minimum near 209 nm in more detail. However, to study the
thermostability of the observed CD spectral bands, we collected
CD spectra of the hexamer 21 at temperatures in the range
20-70°C. As can be seen in Figure 5 B, the hexamer 21 shows
only a small reduction in ellipticity and no change in shape until
70°C, denoting thermal stability of the CD spectral bands.
It is not possible to identify the secondary structure of novel
peptidomimetics in solution by solely analyzing CD spectra. We
therefore designed and synthesized tetramer 28 and hexamer
29 in order to perform additional conformational studies. Based
on the X-ray analysis of tripeptoid 9, we speculated that
α-aminoxy peptoids can adopt a conformation in which the i, i+2
and i+4 side chains are located on one side of the peptoid face,
whereas the i+1, i+3 and i+5 side chains are located on the
opposite side. Thus, we designed the hexamer 29 containing
α-chiral, aromatic side chains at the i+1, i+3 and i+5 positions to
create an “aromatic face”, which is known to stabilize peptoid
helices.[33] Three structurally diverse side chains were chosen for
the i, i+2 and i+4 in order to minimize overlapping NMR signals.
Figure 5. A) CD spectra of compounds 19-21 in acetonitrile (50 µM) at 25°C;
B) CD spectra of compound 21 at varying temperatures in acetonitrile (50 µM)
(Data are expressed in terms of per-residue molar ellipticity (deg cm2/dmol).
The CD spectra of 19-21 reveal a characteristic minimum
around 220 nm. Somewhat surprisingly, the intensity of the
minimum near 220 nm decreased with increasing chain length of
the α-aminoxy peptoids. This phenomenon might indicate that
this signal is not caused by secondary structure formation but is
related to the N-(S)-1-phenylethyl-containing amide motif
This article is protected by copyright. All rights reserved.

10.1002/chem.201605100
Chemistry - A European Journal
FULL PAPER
The tetramer 28 represents a simplified and truncated analogue
of 29. The synthesis of the oligomers 28 and 29 is outlined in
Scheme 2 (see Supporting Information for synthetic details).
Briefly, we prepared a set of dimeric building blocks (Scheme
2 A), which were subsequently assembled to the desired
oligomers by amide coupling reactions followed by deprotection
and acetylation of the N-terminal aminoxy moiety (Scheme 2 B,
C). We conducted 2D NOESY experiments on 28 and 29 to gain
further insights into the amide bond geometry and secondary
structure of oligomeric α-aminoxy peptoids. When analyzing
tetramer 28 we observed a strong cross peak between the
N-terminal acetyl group and the backbone methylene group of
the N-terminal monomer (i) as well as a strong cross peak
between the backbone protons of the monomers i+2 and i+3
indicating cis-amide geometry for the C- and N-terminal amide
bonds (Figure S11). However, no cross peaks could be
observed for the internal amides. Moreover, no further distinct
inter-residue NOEs could be detected. In the case of hexamer
29, a clear NOE was found for the C-terminal amide suggesting
a cis-amide geometry (Figure S12).
sidechain protons, which can affect cross-relaxation. Notably, a
VT-NMR experiment on compound 28 revealed no indications
for the presence of cis/trans-amide bond rotamers (Figure S13).
The CD spectra of 28 and 29 in acetonitrile (Figure S14) look
very similar when compared to the spectra of compounds 19-21
(Figure 5).
Since our CD and NMR data are insufficient to verify the
presence of distinct secondary structures, we performed
comprehensive computational studies to investigate the
conformational preferences of α-aminoxy peptoids, and their
potential as helix mimetics. The backbone torsion angles ϕ, θ, ω,
and ψ (Figure 2) are critical for an accurate description of the
conformational preferences. We parameterized ϕ and θ for
molecular dynamics (MD) simulations at the molecular
mechanics level based on ab initio calculations (Figure 2 and
Figure S15). Fitting of the molecular mechanics energy against
the ab initio energy for angles ϕ and θ provided very good
correlations (R2 ≥ 0.95 (Figure S15)). For the ω and ψ backbone
torsion angles (Figure 2), molecular mechanics energies using
parameters of the GAFF force field (see Supporting Information)
already yielded a very good agreement with ab initio energies
(R2 = 0.99 (Figure S16)). The energy minima identified for the
torsion potentials of ϕ, θ, ω, and ψ showed good agreement with
the torsion angles found in the crystal structure of α-aminoxy
Unfortunately, no inter-residual NOESY cross-peaks could be
detected. This issue was further investigated by a proton
inversion-recovery experiment (Table S2), revealing significant
differences in longitudinal-relaxation times of backbone- and
Scheme 2. Synthesis of tetramer 28 and hexamer 29.
This article is protected by copyright. All rights reserved.

10.1002/chem.201605100
Chemistry - A European Journal
FULL PAPER
peptoids 1 and 9 (Figures S15 and S16, Table S3). Hence, the
torsion angle parameters accurately reproduce relative energies
of the backbone dihedral angles of α-aminoxy peptoids.
α-aminoxy peptoids 1 and 9 (Figures S17, S18, S19, and Table
S3). Relative free energies of α-aminoxy peptoid 1 as a function
of the backbone dihedral angles ϕ, θ, and ψ computed from the
frequency of occurrence of conformations during MD simulations
allowed us to identify preferred backbone conformations of
α-aminoxy peptoid 1 (Figure S24); given the almost identical
torsion angle distributions (Figures S17, S18, S19), similar
conformations are expected to be favorable also for other
α-aminoxy peptoids. The lack of mirror symmetry in the θ/ϕ and
θ/ψ projections of the relative free energy is compatible with the
observation that the amide nitrogen shows a certain degree of
pyramidalization, in agreement with Jordan and co-workers,
leading to a transient chirality.[34] Superimposing low-energy
conformations of α-aminoxy peptoids 1,2, and 3 extracted from
the respective MD simulations onto the crystal structure of
α-aminoxy peptoid 1 results in backbone atom root mean square
deviations (RMSD) ≤ 0.6 Å (Figure 6 A, B and C). Similarly,
overlaying a low-energy conformation of α-aminoxy peptoid 9
onto the crystal structure of the same compound results in a
backbone atom RMSD = 1.2 Å (Figure 6 D). These results show
that the MD simulations yield low-energy conformations of
α-aminoxy peptoids that are in very good agreement with
experimentally determined structures.
Figure 6. (A), (B), and (C): Overlay of low-energy conformations of α-aminoxy
peptoids 1, 2, and 3 obtained from MD simulations (green) onto a crystal
structure of α-aminoxy peptoid 1 (pink), resulting in RMSD of the backbone
atoms of 0.57 Å, 0.60 Å, and 0.59 Å, respectively. In the lower panel, the
overlays are rotated by 90°. (D): Overlay of a low-energy conformation of
α-aminoxy peptoid 9 obtained from MD simulations (green) onto a crystal
structure of α-aminoxy peptoid 9 (magenta), resulting in an RMSD of the
backbone atoms of 1.2 Å. In the right panel, the overlay is rotated by 90°.
We used the thus validated setup of MD simulations to
investigate the conformational properties of the more complex
tetrameric (28) and hexameric (29) α-aminoxy peptoids. MD
simulations of 28 (3.5 µs length) and 29 (2 µs length) in explicit
chloroform yielded again distributions of backbone torsion
angles that are in very good agreement with values of the crystal
structures of α-aminoxy peptoids 1 and 9, and are similar to
those obtained with α-aminoxy peptoids 1,2, and 3 (data not
shown). The conformational ensembles obtained by the MD
simulations of α-aminoxy peptoid 28 and 29 were clustered
according to the RMSD. The four most populated clusters that
account for 71% (α-aminoxy peptoid 28) and 64% (α-aminoxy
peptoid 29) of the MD ensembles were further analyzed.
Overlaying the Cβ atoms of α-aminoxy peptoid 28 onto Cβ atoms
of a canonical α-helix shows that the α-aminoxy peptoid scaffold
can closely mimic the spatial arrangement of peptide side chains
at positions i, i + 1, i + 3, i + 5. When the N-terminus of the
peptoid is oriented toward the helix C-terminus, the RMSD of the
coordinates of the respective atom pairs is 0.5 Å (Figure 7 A).
When the peptoid is oriented in the opposite direction, the
RMSD is 0.7 Å (Figure 7 B). Overlaying the Cβ atoms of peptoid
29 with Cβ atoms of a canonical α-helix shows that the
α-aminoxy peptoid scaffold can also address an i, i + 2, i + 3, i +
5, i + 7, i + 11 amino acid pattern when the N-terminus of the
peptoid is oriented toward the helix C-terminus (Figure 7 C), and
alternatively an i, i + 4, i + 6, i + 8, i + 9, i + 11 pattern when the
peptoid is oriented in the opposite direction (Figure 7 D). The
RMSD of the coordinates of the respective atom pairs is 1.1 Å in
both cases. In summary, the conformational analysis based on
the MD simulations reveals that α-aminoxy peptoids adopt
conformations that can mimic i, i + 1, i + 3, i + 5; i, i + 2, i + 3, i +
5, i + 7, i + 11; and i, i + 4, i + 6, i + 8, i + 9, i + 11 patterns of
substituent orientations with respect to an α-helix.
Subsequently, we performed MD simulations of α-aminoxy
peptoids 1 (3.5 µs length), 2 and 3 (3 µs length each), and 9
(3.5 µs length) in explicit chloroform. Choosing this solvent
allows for a direct comparison of simulation results with NMR
data (see above). The MD simulations reached equilibrium with
respect to conformational preferences of ϕ, θ, and ψ angles, as
indicated by highly symmetric torsion angle distributions (Figures
S17, S18, S19) and frequent transitions between energetically
preferred states (Figures S20, S21, S22). In contrast, for
α-aminoxy peptoids 1,2, or 3 started either with a cis- or
trans-amide conformation, a transition in the ω torsion angle
between the two conformers was not observed, in line with an
energetic barrier separating the two conformers (Figure S16 for
the N-ethoxyformamide model compound; Figure S23 for
α-aminoxy peptoid 30 (N,N-dimethyl-2-((N-methylacetamido)-
oxy)acetamide) for probing the effect of an acetyl cap group and
an N-methyl side chain) and previous work.[34] In addition, our
calculations show that the cis conformer is preferred over the
trans conformer by ~2.8 kcal mol-1 for peptoid 30, whereas cis
and trans conformers are virtually isoenergetic for the
N-ethoxyformamide model compound (Figures S23 and S16,
respectively), which is consistent with cis/trans energy
differences of up to 1.4 kcal mol-1 in regular (not α-aminoxy)
peptoids, as reported by Yoo and Kirshenbaum.[35] These
findings also agree with the above NMR experiments and crystal
structure analyses. Finally, the distributions of backbone torsion
angles of monomeric α-aminoxy peptoids 1,2, and 3 observed in
the MD simulations showed a very good to good agreement with
the torsion angle values found in the crystal structures of
This article is protected by copyright. All rights reserved.

10.1002/chem.201605100
Chemistry - A European Journal
FULL PAPER
Figure 7. Overlay of Cβ atoms of low-energy conformations of α-aminoxy peptoids 28 (A, B) and 29 (C, D) onto Cβ atoms of a canonical α-helix (indicated by the
pink spheres). In the case of α-aminoxy peptoid 28, the scaffold can closely mimic the spatial arrangement of peptide side chains at positions i, i + 1, i + 3, i + 5,
with an RMSD of the coordinates of the respective atom pairs of 0.5 Å when the N-terminus of the α-aminoxy peptoid is oriented toward the helix C-Terminus,
and 0.7 Å when the α-aminoxy peptoid is oriented in the opposite direction (A and B, respectively). In the case of α-aminoxy peptoid 29, the scaffold can closely
mimic the spatial arrangement of peptide side chains at positions i, i + 2, i + 3, i + 5, i + 7, i + 11 when the N-terminus of the α-aminoxy peptoid is oriented toward
the helix C-terminus (C), and i, i + 4, i + 6, i + 8, i + 9, i + 11 when the α-aminoxy peptoid is oriented in the opposite direction (D), with an RMSD of 1.1 Å in both
cases.
Conclusions
Experimental Section
Taken together, we have performed the first thorough analysis of
the folding propensities of α-aminoxy peptoids. To this end, we
designed and synthesized a series of model α-aminoxy peptoids
utilizing a well-established cap group to investigate the influence
of several side chains on the amide conformation and secondary
structure. Interestingly, for α-aminoxy peptoids we observed a
unique preference for cis-amide bonds in comparison to other
peptoid backbones. The formation of an eight-membered δ turn
that is stabilized by Cα,i–H···O=Ci+1 interactions between
residues i→i+1 in the cis configuration might explain the
cis-amide preference of α-aminoxy peptoids. MD simulations
suggest that α-aminoxy oligopeptoids can fold into a distinct
secondary structure closely mimicking the spatial arrangement
of peptide side chains in a canonical α-helix. It is worth noting
that α-aminoxy peptoid monomers can be easily prepared from
the readily available starting material 11 either by classical
alkylation with alkyl halides or by Mitsunobu reaction. This
efficient monomer synthesis may therefore allow for the
preparation of α-aminoxy peptoid libraries with a large chemical
diversity in the future. Given their ease of synthesis, structural
diversity, and conformational properties, α-aminoxy peptoids
represent a promising new class of peptidomimetic foldamers
that may be highly useful in a variety of novel applications such
as material sciences and medicinal chemistry. In conclusion,
α-aminoxy peptoids were introduced as a new peptoid backbone
capable of controlling the amide isomerism while maintaining the
potential for side-chain diversity.
See the Supporting Information for experimental details. The
structural data has been deposited with the Cambridge
Crystallographic Data Center (CCDC-No. 1508156 and
1508157).
Acknowledgements
This work was supported by funds from the Fonds der
Chemischen
Industrie
(to
FKH).
The
Deutsche
Forschungsgemeinschaft (DFG) is acknowledged for funds used
to purchase the UHR-TOF maXis 4G, Bruker Daltonics, Bremen
HRMS instrument used in this research. Financial support by
Deutsche
Forschungsgemeinschaft
(DFG)
for
funds
(INST 208/704-1 FUGG to HG) to purchase the hybrid computer
cluster used in this study is gratefully acknowledged.
Computational support and infrastructure was provided by the
“Center for Information and Media Technology” (ZIM) at the
Heinrich Heine University Düsseldorf (HHU). We thank the
Institute of Physical Biology at HHU for providing us access to
the CD spectrometer.
Keywords: peptidomimetics • foldamers • peptoids • α-aminoxy
peptoids • helix mimetics
[1]
[2]
S. M. Miller, R. J. Simon, S. Ng, R. N. Zuckermann, J. M. Kerr, W. H.
Moos, Bioorg. Med. Chem. Lett. 1994, 4, 2657 – 2662.
Y.-U. Kwon, T. Kodadek, J. Am. Chem. Soc. 2007, 129, 1508 – 1509.
This article is protected by copyright. All rights reserved.

10.1002/chem.201605100
Chemistry - A European Journal
FULL PAPER
[3]
[4]
[5]
[6]
M. M. Reddy, K. Bachhawat-Sikder, T. Kodadek, Chem. Biol. 2004, 11,
1127 – 1137.
P. G. Alluri, M. M. Reddy, K. Bachhawat-Sikder, H. J. Olivos, T.
Kodadek, J. Am. Chem. Soc. 2003, 125, 13995 – 14004.
T. Hara, S. R. Durell, M. C. Myers, D. H. Appella, J. Am. Chem. Soc.
2006, 128, 1995 – 2004.
I. M. Mándity, F. Fülöp, Expert Opin. Drug Discov. 2015, 10, 1163 –
1177.
[7]
[8]
[9]
S. Suwal, T. Kodadek, Org. Biomol. Chem. 2013, 11, 2088 – 2092.
J. A. Hodges, R. T. Raines, Org. Lett. 2006, 8, 4695 – 4697.
B. Yoo, K. Kirshenbaum, Curr. Opin. Chem. Biol. 2008, 12, 714 – 721.
[10] C. Baldauf, R. Günther, H.-J. Hofmann, Phys. Biol. 2006, 3, 1 – 9.
[11] A. S. Norgren, S. Zhang, P. I. Arvidsson, Org. Lett. 2006, 8, 4533 –
4536.
[12] C. A. Olsen, M. Lambert, M. Witt, H. Franzyk, J. W. Jaroszewski, Amino
Acids 2008, 34, 465 – 471.
[13] C. A. Olsen, H. L. Ziegler, H. M. Nielsen, N. Frimodt-Møller, J. W.
Jaroszewski, H. Franzyk, Chem. Bio. Chem. 2010, 11, 1356 – 1360.
[14] O. Roy, S. Faure, V. Thery, C. Didierjean, C. Taillefumier, Org. Lett.
2008, 10, 921 – 924.
[15] J. S. Laursen, P. Harris, P. Fristrup, C. A. Olsen, Nat. Comm. 2015, 6,
7013.
[16] J. S. Laursen, J. Engel-Andreasen, P. Fristrup, P. Harris, C. A. Olsen, J.
Am. Chem. Soc. 2013, 135, 2835 – 2844.
[17] C. Caumes, O. Roy, S. Faure, C. Taillefumier, J. Am. Chem. Soc. 2012,
134, 9553 – 9556.
[18] I. Shin, K. Park, Org. Lett. 2002, 4, 869 – 872.
[19] X. Li, Y.-D. Wu, D. Yang, Acc. Chem. Res. 2008, 41, 1428 – 1438.
[20] X. Li, D. Yang, Chem. Commun. 2006, 32, 3367 – 3379.
[21] N. H. Shah, G. L. Butterfoss, K. Nguyen, B. Yoo, R. Bonneau, D. L.
Rabenstein, K. Kirshenbaum, J. Am. Chem. Soc. 2008, 130, 16622 –
16632.
[22] C. Caumes, O. Roy, S. Faure, C. Taillefumier, J. Am. Chem. Soc.
2012, 134, 9553 – 9556.
[23] a) B. H. Meier, R. R. Ernst, J. Am. Chem. Soc. 1979, 101, 6441 – 6442;
b) C. L. Perrin, T. J. Dwyer, Chem. Rev. 1990, 90, 935 – 967.
[24] D. X. Hu, P. Grice, S. V. Ley, J. Org. Chem. 2012, 77, 5198 – 5202.
[25] All ratios were calculated by integrating and averaging at least three
1H NMR signals at 15 mM concentration.
[26] P. L. Grel, A. Salaün, C. Mocquet, B. L. Grel, T. Roisnel, M. Potel, J.
Org. Chem. 2011, 76, 8756 – 8767.
[27] a) D. Diedrich, A. J. Rodrigues Moita, A. Rüther, B. Frieg, G. J. Reiss, A.
Hoeppner, T. Kurz, H. Gohlke, S. Lüdeke, M. U. Kassack, F. K. Hansen,
Chem. Eur. J. 2016, DOI: 10.1002/chem.201602521; b) B. Draghici,
F. K. Hansen, A.-M. Buciumas, B. E.-D. M. El-Gendy, E. Todadze, A. R.
Katritzky, RSC Adv. 2011, 1, 602 – 606; c) D. Yang, G. J. Liu, Y. Hao,
W. Li, Z. M. Dong, D. W. Zhang, N. Y. Zhu, Chem. Asian. J. 2010, 5,
1356 – 1363.
[28] C. Toniolo, E. Benedetti, Crit. Rev. Biochem. 1980, 9, 1 – 44.
[29] R. Vargas, J. Garza, D. A. Dixon, B. P. Hay, J. Am. Chem. Soc. 2000,
122, 4750 – 4755.
[30] C. Bissantz, B. Kuhn, M. Stahl, J. Med. Chem. 2010, 53, 5061 – 5084.
[31] C. Toniolo, M. Crisma, A. Moretto, C. Peggion, F. Formaggio, C.
Alemán, C. Cativiela, C. Ramakrishnan, P. Balaram, Chem. Eur. J.
2015, 21, 13866 – 13877.
[32] A. Glättli, X. Daura, D. Seebach, W. F. van Gunsteren, J. Am. Chem.
Soc. 2002, 124, 12972 – 12978.
[33] C. W. Wu, T. J. Sanborn, K. Huang, R. N. Zuckermann, A. E. Barron, J.
Am. Chem. Soc. 2001, 123, 6778 – 6784.
[34] P. A. Jordan, B. Paul, G. L. Butterfoss, P. D. Renfrew, R. Bonneau, K.
Kirshenbaum, Biopolymers (Peptide Science) 2011, 96, 617 – 626.
[35] B. Yoo, K. Kirshenbaum, Curr. Opin. Chem. Biol. 2008, 12, 714 – 721.
This article is protected by copyright. All rights reserved.

10.1002/chem.201605100
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
V. Krieger, Dr. E. Ciglia, R. Thoma, Dr.
V. Vasylyeva, B. Frieg, Dr. N. de Sousa
Amadeu, Prof. Dr. T. Kurz, Prof. Dr. C.
Janiak, Prof. Dr. H. Gohlke, JProf. Dr. F.
K. Hansen*
α-Aminoxy peptoids: The first
thorough analysis of the folding
propensities of α-aminoxy peptoids
has been performed. This novel class
of foldamers shows
a
unique
preference for cis-amide bonds. The
Page No. – Page No.
conformational analysis based on the
MD
simulations
revealed
that
α-Aminoxy Peptoids:
Peptoid Backbone with a Preference
for Cis-Amide Bonds
A
Unique
α-aminoxy peptoids can adopt helical
conformations that can mimic the
spatial arrangement of peptide side
chains in a canonical α-helix.
This article is protected by copyright. All rights reserved.