Structural characterizations of oligopyridyl foldamers, α-helix mimetics

Article abstract of DOI:10.1021/ci200424a

Protein-protein interactions are central to many biological processes, from intracellular communication to cytoskeleton assembly, and therefore represent an important class of targets for new therapeutics. The most common secondary structure in natural proteins is an α-helix. Small molecules seem to be attractive candidates for stabilizing or disrupting protein-protein interactions based on α-helices. In our study, we assessed the ability of oligopyridyl scaffolds to mimic the α-helical twist. The theoretical as well as experimental studies (X-ray diffraction and NMR) on conformations of bipyridines in the function of substituent and pyridine nitrogen positions were carried out. Furthermore, the experimental techniques showed that the conformations observed in bipyridines are maintained within a longer oligopyridyl scaffold (quaterpyridines). The alignment of the synthesized quaterpyridine with two methyl substituents showed that it is an α-helix foldamer; their methyl groups overlap very well with side chain positions, i and i + 3, of an ideal α-helix.

Full text of DOI:10.1021/ci200424a

Article
pubs.acs.org/jcim
Structural Characterizations of Oligopyridyl Foldamers, α-Helix
Mimetics
Jana Sopkova-de Oliveira Santos,*,† Anne Sophie Voisin-Chiret,† Gregory Burzicki,† Laure Sebaoun,†
Muriel Sebban,‡ Jean-Francois Lohier,§ Remi Legay,† Hassan Oulyadi,‡ Ronan Bureau,†
̧
́
and Sylvain Rault†
†Centre d’Etudes et de Recherche sur le Med
Pharmaceutiques, Universite de Caen, bd Becquerel, 14032 Caen Cedex, France
‡Equipe Analyse et Model
́ ̀
isation Moleculaire, IRCOF UMR6014 CNRS-COBRA, Universite de Rouen, rue Tesniere, 76821
́
icament de Normandie, UPRES EA-4258, FR CNRS INC3M, UFR des Sciences
́
́
́
Mont-Saint-Aignan, France
§Laboratoire de Chimie Molec
Caen, 14050 Caen, France
́ ́
ulaire et Thio-organique, UMR CNRS 6507, FR CNRS 3038 INC3M, ENSICAEN - Universite de
S
* Supporting Information
ABSTRACT: Protein−protein interactions are central to many biological processes, from
intracellular communication to cytoskeleton assembly, and therefore represent an important class
of targets for new therapeutics. The most common secondary structure in natural proteins is an
α-helix. Small molecules seem to be attractive candidates for stabilizing or disrupting protein−protein
interactions based on α-helices. In our study, we assessed the ability of oligopyridyl scaffolds to mimic
the α-helical twist. The theoretical as well as experimental studies (X-ray diffraction and NMR) on
conformations of bipyridines in the function of substituent and pyridine nitrogen positions were
carried out. Furthermore, the experimental techniques showed that the conformations observed in
bipyridines are maintained within a longer oligopyridyl scaffold (quaterpyridines). The alignment of
the synthesized quaterpyridine with two methyl substituents showed that it is an α-helix foldamer;
their methyl groups overlap very well with side chain positions, i and i + 3, of an ideal α-helix.
Small molecular scaffolds are often limited in their affinity and
selectivity toward these shallow topological features and large
surface areas. However, synthetic scaffolds that mimic only the
key elements of a protein surface (named “Hot Spots”) can
potentially lead to small molecules having almost the full
activity of a protein domain, only a fraction of the molecular
weight, and no peptide bonds.6 To date, the majority of small
molecule drug design successes correspond to bimolecular
complexes built by analyzing the interactions of regular protein
secondary structure elements, such as surface-exposed extended
β-strand segments and β-turns. In these cases, the designed
nonpeptide molecules retained the orientation of the amino
acid side chains in a strand or a turn.7 However, α-helices are
the most common protein secondary structures, accounting for
over 40% of polypeptide amino acids in natural proteins.8 One
strategy for the design of agents able to disrupt α-helix
interactions therefore involves the mimicry of side-chain
residues on one face of an α-helix. Surprisingly, there are only
few examples of small molecules mimicking a protein α-helix,
and a majority of them are based on an approach in which
hydrogen bonding propensities that control its folding into a
desired conformation are built into the molecule. Among the
previously reported α-helix mimetic scaffolds, we can mention,
INTRODUCTION
■
A majority of processes in living cells are controlled by proteins,
sometimes alone, but more often in concert with other partners.
The formation of protein−protein complexes is essential for
numerous regulatory processes in cells and paramount for
survival. Thus, modulation of protein−protein interactions
(PPIs) is a valuable tool in understanding biochemical pathways
and may lead to the development of new therapeutic candidates.
Indeed, there are numerous examples in the literature of using
antibodies1 or medium-size peptides to inhibit particular protein−
protein assemblies. Although antibodies offer the advantage of
having very high affinities for their target proteins, they are
expensive to produce, have low oral bioavailability, and have
undesired side effects.2 Medium-size peptides exhibit several
disadvantages compared to small molecules. They are generally
inactive by the oral route, which is the most commonly accepted
entry, because they show unfavorable pharmacokinetic properties:
low capacity to cross membranes, particularly from the digestive
tract to the bloodstream, and low systemic stability.3 Medium-size
peptides cannot lead to delayed or long-acting effects.4
An alternative approach involves the design and the synthesis
of small molecules able to disrupt PPIs. However, to date, this
remains an issue.5 The structures of many oligomeric proteins
and protein−protein complexes show them to have generally
large and flat interfaces, making the rational design of small
molecules able to mimic these surfaces a difficult prospect.
Received: September 9, 2011
Published: December 24, 2011
© 2011 American Chemical Society
429
dx.doi.org/10.1021/ci200424a | J. Chem. Inf. Model. 2012, 52, 429−439

Journal of Chemical Information and Modeling
Article
a
a
a
Scheme 1. Synthesis of 5-Bromo-2.4′-bipyridine (1), 3′-Bromo-3.4′-bipyridine (2), 5-Bromo-2.3′-bipyridine (3), 3′-Methyl-
b
b
3.2′:5′.2″:5″.3‴-quaterpyridine (4), and 3′.3″-Dimethyl-3.2′:5′.2″:5″.3‴-quaterpyridine (5)
a
Reaction conditions and reagents: pyridylboronic acid or ester 1.2 equiv dihalopyridine or dihalobipyridine 1.0 equiv, Na2CO3aq. 2.5 equiv,
b
Pd(PPh3)4 0.05 equiv, 1.4-dioxane, reflux. Reaction conditions and reagents: 3-pyridylboronic acid 2.5 equiv, dihalobipyridine 1.0 equiv, Na2CO3aq.
5.0 equiv, Pd(PPh3)4 0.1 equiv, 1.4-dioxane, reflux.
for example, indane,9 biphenyl,7 benzamides,10 polycyclic ether,11
scan (PES) studies of all bipyridines substituted with a bromine
or methyl group consisted of a (−180°, +180°) geometry
optimization with the specified coordinate freezing and a 5°
increment in order to obtain the internal energy barrier to
rotation at the HF/6-31G level.
General Procedure for Suzuki−Miyaura Cross-Coupling
Reactions. Commercial reagents were used as received without
further purification. A mixture of pyridylboronic acid or ester (1.2
equiv), dihalopyridine (1.0 equiv), tetrakis(triphenylphosphine)
palladium (5 mol %), and aqueous Na2CO3 (2.5 equiv) in 1,4-
dioxane was heated to 80 °C for 1 h, then under reflux until the
complete consumption of dihalopyridine (TLC). Crude products
were purified on silica gel column chromatography to give 5-bromo-
2,4′-bipyridine19 (1), 3′-bromo-3,4′-bipyridine (2), and 5-bromo-
2,3′-bipyridine20,21,19 (3) with, respectively, 37%, 42%, and 78%
yields (see Scheme 1).
The same procedure was used with 3-pyridylboronic acid
(2.5 equiv), dihalobipyridine22 (1.0 equiv), tetrakis(triphenyl-
phosphine) palladium(0) (10 mol %), and aqueous Na2CO3
(5.0 equiv) in 1,4-dioxane at reflux (see Scheme 1) to produce
3′-methyl-3,2′:5′,2″:5″,3‴-quaterpyridine (4) and 3′,3″-dimethyl-
3,2′:5′,2″:5″,3‴-quaterpyridine (5) with, respectively, 73% and
76% yield.
Melting points were determined on a Kofler heating bench. IR
spectra were recorded on a Perkin-Elmer BX FT-IR spectropho-
tometer. 1H NMR (400 MHz) and 13C NMR (100 MHz) spectra
were recorded on a JEOL Lambda 400 spectrometer. Mass spectra
were recorded on a JEOL JMS GCMate spectrometer at 70 eV
ionizing potential (EI) and with pfk as an internal standard for the
high-resolution procedure or were performed using a LC-MS
Waters alliance 2695 (ESI+) spectrometer. Chromatography was
carried out on a column using flash silica gel 60 Merck (0.063−
0.200 mm) as the stationary phase. Thin layer chromatography
(TLC) was performed on 0.2 mm precoated plates of silica gel
60F-264 (Merck), and spots were visualized using an ultraviolet-
light lamp.
and the ABT-737 family,12 promising inhibitors disrupting α-helix/
α-helix interactions occurring in apoptosis.
Hamilton and co-workers have designed a preorganized
α-helix mimetic terphenyl13,14 by synthesizing the first entirely
nonpeptidic scaffold from 3,2′,2″-trisubstituted terphenyl
derivatives. This scaffold mimics substituents i, i + 3, and i + 7
one ridgeof an α-helical dodecapeptide.15
Our efforts in the design of novel α-helix mimetics that
disrupt PPIs were prompted by the terphenyl α-helix mimetic
concept. Hamilton and co-workers observed that the terphenyl
scaffolds are not rigid. For example, they could adopt both
right- and left-handed twists, and the possibility of mutual
phenyl orientations is limited. Various other organic scaffolds
have therefore been proposed, including a terpyridyl scaffold.16
In our study, we assessed the ability of oligopyridyl scaffolds
to mimic the α-helical twist. The first main objective was to
define all possible conformations associated with oligopyr-
idines. We first carried out detailed studies using quantum
mechanics calculations of bipyridine conformations as a
function of nitrogen and substitutent positions. We then
compared the theoretical results with experimental 3D data
obtained on the synthesized bi- and quaterpyridines using two
methods, X-ray diffraction (solid state) and NMR (liquid state).
These experimental structural data were then correlated with an
α-helix side chain distribution.
MATERIALS AND METHODS
■
Conformational Analysis. 3D models for every possible
bipyridines as a function of the pyridyl nitrogen and substituent
positions were built using the Discovery Studio software.17 This
corresponds to 30 bipyridine models substituted with a
bromine or methyl group. The Cartesian coordinates of each
model were used as input for the ab initio simulation.
All ab initio calculations reported in the present study were
carried out using the Gaussian 98 software.18 Potential energy
430
dx.doi.org/10.1021/ci200424a | J. Chem. Inf. Model. 2012, 52, 429−439

Journal of Chemical Information and Modeling
Article
X-Ray Diffraction. Single crystal X-ray analysis was carried
out using graphite-monochromatized Mo Kα radiation on a
Bruker-Nonius Kappa II diffractometer equipped with a CCD
area detector. The crystal structure was solved by direct
methods using the SHELX97 package (SHELXS23) and refined
using SHELXL.24 The refinement was based on F2 for all
reflections, and all non-hydrogen atoms were refined
anisotropically. Hydrogen atom positions were determined
either via difference Fourier maps and refined with isotropic
atomic displacement parameters or were calculated and fixed in
the ideal geometry, depending on data quality.
RESULTS AND DISCUSSION
■
Since the critical interactions for α-helix recognition often
involve the side chains of residues on one helical face (i, i + 3
and/or i + 4 and i + 7), it is possible to design an appropriate
scaffold with limited conformations that can orient the attached
functional groups to correctly resemble an α-helical surface
(Figure 1). In the case of the oligophenyl scaffold, the adjacent
NMR Measurements. All NMR experiments were carried
out using a Bruker AVIII 600 spectrometer (Bruker, Wissembourg,
France) equipped with a 10 A gradient amplifier and a 5 mm
CPTXI{1H, 13C,15N} including shielded z-gradients. Solutions
with concentrations in the range 50−80 mg·ml−1 in CDCl3
1
were used. Experiments were carried out at 600 MHz for H
and 150 MHz for 13C. 1H and 13C chemical shifts were
referenced according to the CDCl3 solvent signal used as a
secondary internal reference (1H, δ = 7.26 ppm; 13C, δ = 77 ppm,
with respect to TMS, 0 ppm).
The 1D and 2D NMR spectra were obtained at 298 K.
Complete assignment of all protons and carbons was carried out
using conventional 2D experiments: COSY (1H−1H), TOCSY
(1H−1H), NOESY (1H−1H), HSQC (1H−13C), and HMBC
(1H−13C). For all 2D spectra, a total of 2048 points in F2 and 512
experiments in F1 were recorded. For the HMBC experiment, an
evolution delay of 65 ms was chosen in such a way that
correlations involving long-range J coupling around 8 Hz could be
observed. The NOESY and the TOCSY experiments were
performed using mixing times of 1.5 s and 200 ms, respectively.
Processing and analysis of the NMR spectra was performed with
the Topspin 2.1 software from Bruker on a PC workstation.
Molecular Modeling with NMR Constraints. Three
dimensional NMR structures of quaterpyridines were refined using
the CHARMM program with potential function parameter set 22
from Discovery Studio. The Discovery Studio program was also
used to derive CHARMM force field parameters for the
quaterpyridines applying MMFF partial charges. The ring twist
angles in the two studied quaterpyridines are the torsions with the
lowest energy barriers according to our theoretical calculations
(five of them with ΔE ∼ 3 kcal/mol and one with ΔE ∼ 4.5 kcal/
mol). For this reason, we used only the force field derived by the
Discovery Studio without the introduction of supplementary
dihedral force field parameters based on our mechanic quantum
simulations. To verify the molecular-mechanics force fields
obtained, the structures were energy minimized, and their
molecular geometry was compared with the X-ray structures.
The root-mean-square deviation (RMSD) between the X-ray and
the optimized NMR structure was <0.1 Å for both structures.
During all CHARMM simulations, NOE (distance) restraints
were applied with a force constant of 25 kcal/mol. Starting
from the energy minimized X-ray structure, a dynamics
simulation of 50 ns was carried out for each quaterpyridine at
300 K, with a time step of 1 fs. The dynamics were preceded by
a heating (during 6 ps) and an equilibration (50 ps) step.
During the production phase, configurations were saved every
5 ps and energy minimized to a root-mean-square gradient of less
than 0.001 kcal/(mol·Å2). The 10 lowest-energy configurations
obtained for each quaterpyridine were used in the subsequent
analysis. The selected structures were analyzed and displayed using
Discovery Studio.
Figure 1. The geometry of side chains on one face of an α-helix.
phenyl rings are balanced by two competing factors: (i) a
symmetric interaction between the π orbitals of the aromatic
rings and (ii) a steric repulsion between overlapping ortho
hydrogen atoms and substituents.16 In the case of oligopyr-
idines, in addition to these two effects, the twist between the
adjacent rings will also be influenced by the position of the
nitrogen atoms in the rings. We therefore carried out this study
to evaluate the advantages and disadvantages of the oligopyridyl
compared to the oligophenyl scaffold.
What Is the Influence of Nitrogen and Substituent
Positions on the Twist Angle between Adjacent Pyridine
Rings? This analysis was started by theoretical simulations
carried out on a simple system constituted of two pyridine rings
and substituted by a bromine atom. In total, 30 models were
built, 12 with bromine in the ortho and meta positions and 6 in
the para position (see Table 1). For each model, the internal
barrier of potential energy was calculated by ab initio simulation
as a function of the torsion angle between the two rings. The
results are summarized in Table 1 and Figure 2, showing that
the 30 models can be divided into six groups depending on the
preferential twist angle between adjacent pyridine rings and the
height of the energy barrier associated with the bipyridine twist.
Thus, the bipyridine system substituted by bromine can adopt
six different conformations compared with only two possible in
the biphenyl system (Figure 2). The bromine substituent
represents a relatively important steric hindrance related to its
bond length (dCar−Br ∼ 1.91 Å) and the van der Waals radius
(195 pm) due to its three lone pairs. The nitrogen has a lone
pair, and it introduces a negative partial charge in the ring. So in
431
dx.doi.org/10.1021/ci200424a | J. Chem. Inf. Model. 2012, 52, 429−439

Journal of Chemical Information and Modeling
Article
Table 1. Structural Parameters of Minimum Energy Conformers of Br Substituted Bipyridine Obtained from PES at the HF/2-
a
31G Level
a
Values in parentheses correspond to a 0.5 kcal/mol limit.
our bipyridine system, a competition of three phenomena
occurs: (i) symmetric interaction between π orbitals of rings,
(ii) steric repulsion between overlapping ortho hydrogens and
the ortho substituent, and (iii) electrostatic repulsion.
When the four atoms in the direct neighborhood of the
junction are the carbons, the rings twist angle is about 50°
(ΔE ∼ 3 kcal/mol, Table 1). In this case, the steric repulsion
dominates somewhat on the π-orbitals’ symmetric interaction.
The introduction of a bromine atom at the ortho position
causes a more important steric repulsion (ΔE ∼ 10.5 kcal/mol,
Table 1) and introduces therefore a greater twist angle of about
60°−90°.
When the nitrogen atoms of both rings lie in the direct
neighborhood of the bipyridine junction, independent of the position
of the bromine substituent, the pyridine rings are always strongly
constrained in the coplanar position (ΔE ∼ 10−12 kcal/mol),
and the nitrogen lone pair repulsion as well as weak electro-
static interactions between N and CH cause the nitrogen atoms
to be situated on the opposite site of the junction.
In the case that only one nitrogen of the two rings is in the
ortho position of the junction and the bromine substituent is
in the meta or para position, the two rings are also coplanar
(ΔE ∼ 4.5−5.5 kcal/mol). With the bromine substituent in the
ortho position, the situation is more complex: (i) When the
432
dx.doi.org/10.1021/ci200424a | J. Chem. Inf. Model. 2012, 52, 429−439

Journal of Chemical Information and Modeling
Article
Figure 2. Evolution of potential energy as a function of torsion angle between the ring: (A) biphenyl system substituted by bromine atom and (B, C)
bipyridine system with bromine substituent.
nitrogen of the pyridine without a bromine substituent is in the
ortho position, the twist angle is about 50° due to steric and
lone pair (electrostatic) repulsion between bromine and
nitrogen (ΔE ∼ 6−8 kcal/mol), and the pyridine nitrogen in
this case is always on the opposite site of junction with respect
to the bromine. (ii) When the nitrogen of the pyridine with the
substituent is in the ortho position of the junction, there is steric
repulsion of the substituent and the ring’s twist angle is about
50°, but in this case the lone pair repulsion does not occur and
ΔE ∼ 3 kcal/mol (Table 1, Figure 2).
To mimic a side chain distribution of an α-helix, a twist angle
between two rings close to 40° (i + 3) or 60° (i + 4) is appropriate
(see Figure 1). The simulations showed that we are able to
introduce a twist angle of about 50°, with an energy barrier that is
more or less high (ΔE ∼ 3−8 kcal/mol) and a twist angle of about
60° with a higher energy barrier (Δ E ∼ 10.5 kcal/mol). However,
the theoretical simulations showed that all bipyridines can rotate
left and right with equal probability, as in the biphenyles.
All bromobipyridines were synthesized in our laboratory with
good yields. One disadvantage of the terphenyl scaffolds described
by Hamilton is the rather lengthy, multistep synthetic approach
needed for its assembly.15,25 Therefore, our oligopyridines are
simply assembled by a cross-coupling reaction. Our laboratory
specializes in the preparation of pyridyl boronic species26−29 and in
the study of their ability to be good cross-coupling partners.30−32
An iterative, regioselective, and highly flexible synthesis of new
halo-oligopyridines using the Suzuki−Miyaura cross-coupling
reaction in a strategy called the Garlanding concept was recently
reported by our laboratory.33
leading to a regioselective coupling. On the one hand, this
strategy takes into account the nature and the position of the
halogen on the pyridine ring (especially if there are several
halogen atoms) and, on the other hand, the nature and the
position of the boronic species on the pyridine ring.
Besides that, several trials of cross-coupling procedures using
different reaction conditions (base, catalyst, solvent, temper-
ature, and reaction time) were attempted. Accordingly, we have
determined the optimal conditions, where Na2CO3 was used as
a base, Pd(Ph3)4 as a catalyst, and 1,4-dioxane as a solvent. The
completion of the reaction was determined by the total
consumption of the halide (checked by thin layer chromatog-
raphy). The synthesis of bipyridines 1 and 3 is known in the
literature19−21 but the bipyridine 2 is a new compound.
Of all bromobipyridines synthesized, suitable crystals were
obtained for three derivatives (Table 2). The crystal visualization
under the microscope showed for each derivative only one kind of
crystal (needles), and they are probably no polymorphism. We
were thus able to compare our theoretical results with experimental
values obtained from the crystal structures (Table 2). For each
compound, the crystal space group contained a symmetric
center, and consequently two axial enantiomers are present in
each crystal: one has a right-handed twist (Ring1−Ring2−P),
while the second is left-handed (Ring1−Ring2−M). According
to the theoretical result, the rings of 5-bromo-2,4′-bipyridine (1)
should adopt a coplanar position, and they are indeed coplanar in
the crystal structure (twist angle between the ring planes is 1.27°),
while the rings of 3′-bromo-3,4′-bipyridine (2) should adopt an
orientation between 55° and 90°, and the twist plane angle is
about 55°. For 5-bromo-2,4′-bipyridine (3), two conformations
were observed in the asymmetric unit of 3, and both have a twist
The regioselective control of the pyridine−pyridine linkage is
complex to handle, but our cross-coupling strategy is efficient,
433
dx.doi.org/10.1021/ci200424a | J. Chem. Inf. Model. 2012, 52, 429−439

Journal of Chemical Information and Modeling
Article
Table 2. ORTEP View of the Crystal Structures with the Predicted and Experimental Twist Angle Values
angle of about 15° ( 15° and 13°, respectively), which is in
agreement with the theoretical value of 0°−30°. Therefore, we can
conclude that the theoretical and experimental values are consistent
for three compounds.
Will the Energy Barrier Change if the Bromine, an
Electronegative Substituent, Is Replaced by a Methyl
One? As the aim of our study is to synthesize oligopyridines
substituted by various side chains, the next step was an evaluation
of what changes when the substituent is a hydrophobic methyl
group. The theoretical calculation showed that there are no
differences for para and meta substituent positions, but some
differences were observed for the ortho position (Table 3), due to
the absence of a lone pair on the methyl substituent.
The simulation results showed first that, contrary to brome
bipyridines, all methyl bipyridines in which one nitrogen is
situated in the direct neighborhood of the bipyridine junction
yield the same result, a preferential twist angle of 50° with an
energy barrier of about 3 kcal/mol, which is due to the absence
of a lone pair (electrostatic) repulsion between methyl and
pyridine nitrogen. The absence of a lone pair on the methyl
group influences also the profile of the energy barrier of the
methyl bipyridines in which the nitrogen lies further from the
bipyridine junction, the energy minimum is more narrow (from
50° to 90°), and its height is lower (ΔE ∼ 9.5 kcal/mol, Table 3).
In conclusion, the bipyridine system substituted by a methyl
group has generated five different ring orientations, and
therefore it allows us to play more subtly with the distribution
of substituents on the oligopyridyl compared to oligophenyl
scaffolds (the biphenyl system generate only two different
orientationsFigure 2). However, the highest energy barrier in
biphenyl as in bipyridine is the same level.
Will the Energy Barrier Observed for a Bipyridine Be
the Same if It Is Included in an Oligopyridyl Scaffold?
The implementation of the Garlanding concept22,33,34 was used to
prepare two new terpyridines with good reactivity to introduce the
two pyridin-3-yl moiety.22 3′-Methyl-3,2′:5′,2″:5″,3‴-quaterpyridine
(4) and 3′,3″-dimethyl-3,2′:5′,2″:5″,3‴-quaterpyridine (5) were syn-
thesized and successfully crystallized (Figure 3). As in the case of
bipyridine, only one kind of crystal was observed under the
microscope for both quaterpyridines.
The crystal structure of 4 and 5 shows that the pyridines in
the scaffold turn as expected, and once more as in the case of
bipyridine, the space groups of both crystals are center-
symmetric, and so two axial enantiomers are presents in the
crystals. Therefore, we compared first the theoretical prediction
of twist angles from the bipyridyl model with the twist angles
observed in the crystal structures 4 and 5 at the absolute value
level. Together, we saw only one small difference in 4 with
respect to the theoretical value. The middle twist angle should
be of about 0°−35° according to the simulation, but in the
crystal structure, we observed an angle of about 38°.
To verify if the bipyridyl model is adequate to use for the
prediction in the quaterpyridine, the energy barrier for each
junction in the complete, four-unit model of 4 and 5 was
calculated anew. The simulation results showed only minor
differences in the height of energy barriers and no differences in
the position of energy minima with respect to the predictions
from bipyridyl model. Thus, we can conclude that the bipyridyl
model is applicable for longer pyridyl chains.
Another fact that needs to be considered is that the final
structure in the crystal is also influenced by intermolecular
interactions. Indeed, in the crystal packing of 4 and 5, weak
electrostatic interaction between the pyridine nitrogen and
434
dx.doi.org/10.1021/ci200424a | J. Chem. Inf. Model. 2012, 52, 429−439

Journal of Chemical Information and Modeling
Article
Table 3. Structural Parameters of Minimum Energy Conformers of Methyl Substituted Bipyridine Obtained from PES at the
a
HF/2-31G Level
a
Values in parentheses correspond to a 0.5 kcal/mol limit.
aromatic carbon atom were detected. Therefore, the conforma-
tions of our oligopyridines found by the X-ray analysis were
compared with their conformations in solution. In order to
characterize the three-dimensional structure of 4 and 5 in solution,
we carried out a structural NMR study. In the first step,
spectrum. Quantitative analysis of the interproton distances was
carried out by volume quantification of the correlation peaks
observed in the NOESY spectrum. For the conversion of the
NOE measurement to distance restraints, the usual relationship
between the NOE intensity and the distance between a pair of
nuclei for an isolated pair of spins was used:
1
assignment of the H NMR spectra of 4 and 5 was done by
1H−1H correlation spectroscopy (COSY and TOCSY) and
(1H−13C) heteronuclear HSQC and HMBC, as well as (1H−1H)
nuclear Overhauser effect spectroscopy (NOESY). The proton
chemical shift values and coupling constants of 4 and 5 are
summarized in Table 4.
6
⎛
⎝
⎞
NOEkl
NOEij
⎜
⎜
⎟
⎟
r = rkl
ij
⎠
where NOEkl and NOEij are the NOEs between atoms k and l
and atoms i and j, respectively. In the NOESY spectra of 4 and 5,
the NOE between atoms B3 and B6 of the second pyridine ring
The most important information regarding the structures of
4 and 5 was given by the two-dimensional (1H−1H) NOESY
435
dx.doi.org/10.1021/ci200424a | J. Chem. Inf. Model. 2012, 52, 429−439

Journal of Chemical Information and Modeling
Article
Figure 3. ORTEP view of the two conformers from crystal structures and the twist angle values of compound 4 (A) and compound 5 (B).
Displacement ellipsoids are drawn at the 50% probability level, and H atoms are shown as small spheres of arbitrary radii.
(see atom numbering in Figure 3) corresponding to a distance
of 0.248 nm was used for interproton distance calibration.
Thus, we have highlighted for each compound the spatial
proximity of methyl group protons with protons at the α position
of the adjacent cycle. Table 5 shows a comparison of the 1H−1H
distances inferred from the NMR data with those obtained from
the X-ray structure. Overall, pairwise comparisons show a very
good match between X-ray and NMR distances.
In order to generate 3D NMR structures, we carried out
molecular dynamic simulations with the NMR constraints.
436
dx.doi.org/10.1021/ci200424a | J. Chem. Inf. Model. 2012, 52, 429−439

Journal of Chemical Information and Modeling
Article
Table 5. NMR Data
quaterpyridine 4
quaterpyridine 5
NOE distance X-ray
distance
NMR
NOE
distance
X-ray
distance
NMR
A5/B6
B3/B6
A3/B6
D5/C3
B3/C2
B5/C2
4.83
3.00
2.91
2.30
4.29
4.29
3.5
3.0
3.8
3.1
2.8
2.5
A5/B6
C6/B5
C3/C6
A3/B6
B3/B6
B3/C6
C5/D2
D5/E3
E5/D3
C3/D2
5.03
4.84
2.97
2.75
2.93
2.98
4.43
2.32
2.27
2.07
3.4
3.0
3.0
3.7
3.0
3.5
2.5
2.8
2.9
2.7
The 10 lowest energy structures for each quaterpyridine were
retained for subsequent analysis. This analysis showed that for
both quaterpyridines the 10 lowest energy structures represent
the same conformation. For 4 and 5, the observed NMR twist
angles lie at the limits of the theoretical prediction, just as the
angles from the X-ray structures do (Figure 3). This result is
not surprising; we expected that the NMR structures would
reproduce the twist angles’ values delimited by the theoretical
predictions. Moreover, the absolute values of the pyridine twist
angles of the NMR and X-ray structures are very close, but the
sense of rotation is not necessary the same (Figure 3). Along
our theoretical calculations, the negative and positive rotations
of pyridines have the same potential energy, and energy barriers
of the rotations in 4 and 5 are not high (ΔE from 3 to 5.5 kcal/mol),
which allows easy rotations around these pyridine junctions.
TheNMR helicity sense for 4, as for 5, is the same as the helicity
sense of the first conformation from corresponding X-ray structures;
this result seems to be a random one to us. For example, the
theoretical simulation carried out on the quaterpyridines showed that
for 4, when ring2 is fixed at a positive twist angle with respect to
ring1 (R1−R2−P), the probability to obtain a negative (R2−R3−M)
or a positive (R2−R3−P) twist of ring 3 with respect to ring 2 is
almost the same; the potential energy difference of both confor-
mations is about 0.16 kcal/mol. When R2−R3 is fixed positive,
the probability to obtain a negative (R3−R4−M) or a positive
(R3−R4−P) was also the same.
Our first study on oligopyridyl scaffold was carried out
with small substituent, but the increase of the substituent
size with the introduction of a charged atom in the sub-
stituent should influence the helicity rotation sense without
necessarily introducing a chiral substituent. For example, our
preliminary theoretical study with tryptophan showed that
only an R1−R2−M rotation of 50° is possible in this
bipyridine.
All of these results, in the solid and the liquid state, show that
the predicted twist angles from the bipyridine models are
reproduced in the oligopyridines. Thus, the preferential angles
library computed using the bipyridine model will be a useful
base to design longer chain, oligopyridines, with a desired
orientation of pyridine rings, and consequently the desired
orientation of the substituents on this oligopyridyl scaffold.
A superposition of conformer 1 and conformer 2 of the X-ray
structure of 5 on an ideal alanine α-helix was carried out with
the aim of verifying whether the positions of methyl
substituents can coincide with the positions of the Cβ atom
of the alanine side chains (see Figure 4). The oligopyridyl
437
dx.doi.org/10.1021/ci200424a | J. Chem. Inf. Model. 2012, 52, 429−439

Journal of Chemical Information and Modeling
Article
scaffolds were aligned successfully along the axis of the helix.
For both conformations of compound 5, the two methyl
substituents are positioned in close proximity to the side chains
of i and i + 3 residues (see Figure 4). Thus, compound 5 will be
ACKNOWLEDGMENTS
■
We thank the CRIHAN (Centre de Ressources Informatiques
de Haute Normandie) and the European Community
(FEDER) for the molecular modelling software.
REFERENCES
■
(1) Stockwin, L. H.; Holmes, S. Antibodies as therapeutic agents: vive
la renaissance! Expert Opin. Biol. Ther. 2003, 3, 1133−1152.
(2) Arkin, M. R.; Wells, J. A. Small-molecule inhibitors of protein-
protein interactions: progressing towards the dream. Nat. Rev. Drug
Discovery 2004, 3, 301−317.
̀
(3) Fauchere, J. L.; Thurieau, C. Evaluation of the stability of
peptides and pseudopeptides as a tool in peptide drug design. Adv.
Drug. Res. 1992, 23, 128−159.
(4) Loffet, A. Peptides as drugs: is there a market ? J. Peptide Sci.
2002, 8, 1−7.
(5) Toogood, P. L. Inhibition of protein-protein association by small
molecules: approaches and progress. J. Med. Chem. 2002, 45, 1543−1558.
(6) Clackson, T.; Wells, J. A. A Hot Spot of Binding Energy in a
Hormone-Receptor Interface. Science 1995, 267, 38−386.
(7) Jacoby, E. Biphenyls as Potential Mimetics of Protein a-Helix.
Bioorg. Med. Chem. Lett. 2002, 12, 891−893.
(8) Ruan, F. Q.; Chen, Y. Q.; Hopkins, P. B. Metal ion-enhanced
helicity in synthetic peptides containing unnatural, metal-ligating
residues. J. Am. Chem. Soc. 1990, 112, 9403−9404.
(9) Horwell, D. C.; Howson, W.; Ratcliffe, G. S.; Willems, H. M. G.
The Design of Dipeptide Helical Mimetics: The Synthesis, Tachykinin
Receptor Affinity and Conformational Analysis of 1,1,6-Trisubstituted
Indanes. Bioorg. Med. Chem. 1996, 4, 33−42.
(10) Ahn, J.-M.; Han, S.-Y. Dacile synthesis of benzamides to mimic
an α-helix. Tetrahedron Lett. 2007, 48, 3543−3547.
Figure 4. The superposition of the X-ray structure of conformer 1 of
the compound 5 (A) and conformer 2 of the compound 5 (B) on an α
helix.
(11) Oguri, H.; Oomura, A.; Tanabe, S.; Hirama, M. Design and
synthesis of a trans-fused polycyclic ether skeleton as an α-helix
mimetic scaffold. Tetrahedron Lett. 2005, 46, 2179−2183.
(12) Oltersdorf, T.; Elmore, S. W.; Shoemaker, A. R.; Armstrong,
R. C.; Augeri, D. J.; Belli, B. A.; Bruncko, M.; Deckwerth, T. L.; Dinges, J.;
Hajduk, P. J.; Joseph, M. K.; Kitada, S.; Korsmeyer, S. J.; Kunzer, A. R.;
Letai, A.; Li, C.; Mitten, M. J.; Nettesheim, D. G.; Ng, S.; Nimmer, P. M.;
O’Connor, J. M.; Oleksijew, A.; Petros, A. M.; Reed, J. C.; Shen, W.;
Tahir, S. K.; Thompson, C. B.; Tomaselli, K. J.; Wang, B.; Wendt, M. D.;
Zhang, H.; Fesik, S. W.; Rosenberg, S. H. An inhibitor of Bcl-2
family proteins induces regression of solid tumours. Nature 2005, 435,
677−681.
able to mimic i and i + 3 side chains of an α helix with its
substituent groups.
CONCLUSION
■
In our study, we assessed the ability of oligopyridyl scaffolds to
mimic the α-helical twist. The first objective was to define all
possible conformations associated with oligopyridines. The
detailed theoretical analysis about possible conformations of
bipyridines as a function of substituent and pyridine nitrogen
position was carried out. Thus, the experimental structural
studies confirmed the theoretical predictions, and we observed
that the bipyridine conformations are maintained within a
longer oligopyridyl scaffold. Furthermore, the use of an
oligopyridyl scaffold gives a wider variation of consecutive
ring twist (five different orientations) with respect to the
oligophenyl one (two different orientations). Thus, oligopyr-
idines allow one to play more subtly with the mutual position of
the rings and so with the distribution of substituents. The
present detailed study of bipyridines will be helpful in
anticipating the conformational space of the designed pyridyl
foldamer and orienting future syntheses. Moreover, this can
be easily envisaged due to the simplicity of our synthetic
methodology.
(13) Yin, H.; Hamilton, A. D. Strategies for targeting protein-protein
interactions with synthetic agents. Angew. Chem., Int. Ed. 2005, 44,
4130−4163.
(14) Saraogi, I.; Hamilton, A. D. α-Helix mimetics as inhibitors of
protein-protein interactions. Biochem. Soc. Trans. 2008, 36, 1414−
1417.
(15) Orner, B. P.; Ernst, J. T.; Hamilton, A. D. Toward
Proteomimetics: Terphenyl Derivatives as Structural and Functional
Mimics of Extended Regions of an α-Helix. J. Am. Chem. Soc. 2001,
123, 5382−5383.
(16) Che, Y.; Brooks, B. R.; Marshall, G. R. Development of small
molecules designed to modulate protein-protein interactions.
J. Comput.-Aided Mol. Des 2006, 20, 109−130.
(17) Discovery Studio Modeling Environment, Release 2.0; Accelrys
Software Inc.: San Diego, CA, 2005.
(18) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.,
Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels,
A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.;
Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford,
S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;
Salvador, P.; Dannenberg, J. J.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul,
A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi,
I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures and details. This material is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: (33)2-31-56-68-21. E-mail: jana.sopkova@unicaen.fr.
438
dx.doi.org/10.1021/ci200424a | J. Chem. Inf. Model. 2012, 52, 429−439

Journal of Chemical Information and Modeling
Article
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.;
Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, Revision
A.11.1; Gaussian, Inc.: Pittsburgh PA, 2001.
(19) Ishikura, M.; Ohta, T.; Terashima, M. A novel synthesis of
4-aryl- and 4-heteroarylpyridines via diethyl(4-pyridyl)borane. Chem.
Pharm. Bull. 1985, 33, 4755−4763.
(20) Cosford, N. D.; Kamenecka, T.; Roppe, J. R. A preparation of
imidazolyl- and pyrazolyl-ethyne derivatives, useful as modulator of
receptors in the nervous system. US 2005085523, 2005.
(21) Bonnet, V.; Mongin, F.; Trecourt, F.; Breton, G.; Marsais, F.;
Knochel, P.; Queguiner, G. Cross-coupling between 3-pyridylmagne-
sium chlorides and heteroatomic halides. Synlett 2002, 6, 1008−1010.
(22) Burzicki, G.; Voisin-Chiret, A. S.; Sopkova-de Oliveira Santos, J.;
Rault, S. Synthesis of dihalo bi- and terpyridines by regioselective
Suzuki-Miyaura cross-coupling reactions. Tetrahedron 2009, 65, 5413−
5417.
(23) Sheldrick, G. M. Phase Annealing in SHELX-90: Direct
Methods for Larger Structures. Acta Crystallogr. 1990, A46, 467−473.
(24) Sheldrick, G. M. A short history of SHELX. Acta Crystallogr.
2008, A64, 112−122.
(25) Davis, J. M.; Truong, A.; Hamilton, A. D. Synthesis of a
2,3′;6′,3″-Terpyridine Scaffold as an α-Helix Mimetic. Org. Lett. 2005,
7, 5405−5408.
(26) Bouillon, A.; Lancelot, J.-C.; Collot, V.; Bovy, P.-R.; Rault, S.
Synthesis of novel halopyridinylboronic acids and esters. Part 1:
6-Halopyridin-3-yl-boronic acids and esters. Tetrahedron 2002, 58,
2885−2890.
(27) Bouillon, A.; Lancelot, J.-C.; Collot, V.; Bovy, P.-R.; Rault, S.
Synthesis of novel halopyridinylboronic acids and esters. Part 2: 2,4, or
5-Halopyridin-3-yl-boronic acids and esters. Tetrahedron 2002, 58,
3323−3328.
(28) Bouillon, A.; Lancelot, J.-C.; Collot, V.; Bovy, P.-R.; Rault, S.
Synthesis of novel halopyridinylboronic acids and esters. Part 3: 2, or
3-Halopyridin-4-yl-boronic acids and esters. Tetrahedron 2002, 58,
4369−4373.
(29) Bouillon, A.; Lancelot, J.-C.; Sopkova-de Oliveira Santos, J.;
Collot, V.; Bovy, P.-R.; Rault, S. Synthesis of novel halopyridinylbor-
onic acids and esters. Part 4: Halopyridin-2-yl-boronic acids and esters
are stable, crystalline partners for classical Suzuki cross-coupling.
Tetrahedron 2003, 59, 10043−10049.
(30) Voisin, A. S.; Bouillon, A.; Berenguer, I.; Lancelot, J.-C.;
Lesnard, A.; Rault, S. Novel oxybispyridylboronic acids: synthesis and
study of their reactivity in Suzuki-type cross-coupling reactions.
Tetrahedron 2006, 62, 11734−11739.
(31) Cailly, T.; Fabis, F.; Bouillon, A.; Lemaitre, S.; Sopkova-de
Oliveira Santos, J.; Rault, S. Synthesis of orthocyanopyridylboronic
acids and esters toward obtention of cyanofunctionnalized bipyridines.
Synlett 2006, 1, 53−56.
(32) De Giorgi, M.; Voisin-Chiret, A. S.; Sopkova-de Oliveira Santos,
J.; Corbo, F.; Franchini, C.; Rault, S. Design and Synthesis of
thienylpyridyl garlands as non-peptidic alpha helix mimetics and
potential protein-protein interactions disruptors. Tetrahedron 2011, 67,
6145−6154.
́
(33) Voisin-Chiret, A. S.; Bouillon, A.; Burzicki, G.; Celant, M.; Legay, R.;
El-Kashef, H.; Rault, S. A general synthesis of halo-oligopyridines.
The Garlanding concept. Tetrahedron 2009, 65, 607−612.
́
(34) Burzicki, G.; Voisin-Chiret, A. S.; Sopkova-de Oliveira Santos, J.;
Rault, S. Synthesis of New [2,3′:6′,3″]Terpyridines Using Iterative
Cross-Coupling Reactions. Synthesis 2010, 16, 2804−2810.
439
dx.doi.org/10.1021/ci200424a | J. Chem. Inf. Model. 2012, 52, 429−439