Helix-forming propensity of aliphatic urea oligomers incorporating noncanonical residue substitution patterns

Article abstract of DOI:10.1021/ja401151v

Aliphatic N,N′-linked oligoureas are peptidomimetic foldamers that adopt a well-defined helical secondary structure stabilized by a collection of remote three-center H-bonds closing 12- and 14-membered pseudorings. Delineating the rules that govern helix formation depending on the nature of constituent units is of practical utility if one aims to utilize this helical fold to place side chains in a given arrangement and elaborate functional helices. In this work, we tested whether the helix geometry is compatible with alternative substitution patterns. The central -NH-CH(R)-CH₂-NH-CO- residue in a model oligourea pentamer sequence was replaced by guest units bearing various substitution patterns [e.g., -NH-CH₂-CH₂-NH-CO-, -NH-CH₂-CH(R)-NH-CO-, and -NH-CH(R¹)-CH(R ²)-NH-CO-], levels of preorganization (cyclic vs acyclic residues), and stereochemistries, and the helix formation was systematically assessed. The extent of helix perturbation or stabilization was primarily monitored in solution by Fourier transform IR, NMR, and electronic circular dichroism spectroscopies. Our results indicate that although three new substitution patterns were accommodated in the 2.5-helix, the helical urea backbone in short oligomers is particularly sensitive to variations in the residue substitution pattern (position and stereochemistry). For example, the trans-1,2- diaminocyclohexane unit was experimentally found to break the helix nucleation, but the corresponding cis unit did not. Theoretical calculations helped to rationalize these results. The conformational preferences in this series of oligoureas were also studied at high resolution by X-ray structure analyses of a representative set of modified oligomers.

Full text of DOI:10.1021/ja401151v

Article
pubs.acs.org/JACS
Helix-Forming Propensity of Aliphatic Urea Oligomers Incorporating
Noncanonical Residue Substitution Patterns
Nagendar Pendem,†,⊥ Celine Douat,† Paul Claudon,† Michel Laguerre,† Sabine Castano,†
́
Bernard Desbat,† Dominique Cavagnat,‡ Eric Ennifar,§ Brice Kauffmann,∥ and Gilles Guichard*,†
†Universite
France
́ ́
de Bordeaux-CNRS UMR 5248, CBMN, Institut Europeen de Chimie et Biologie, 2 rue Robert Escarpit, 33607 Pessac,
‡ISM, UMR CNRS 5255, Universite
́
de Bordeaux, 351 cours de la Liber
de Strasbourg-CNRS UPR 9002, Architecture et Reactivite de l’ARN, IBMC, CNRS, 15 rue Rene
Strasbourg, France
∥Universite
́
de Bordeaux-CNRS UMS 3033, INSERM US001, Institut Europeen de Chimie et Biologie, 2 rue Robert Escarpit, 33607
́
ation, 33405 Talence, France
§Universite
́
́
́
́
Descartes, 67084
́
Pessac, France
S
* Supporting Information
ABSTRACT: Aliphatic N,N′-linked oligoureas are peptidomimetic foldamers
that adopt a well-defined helical secondary structure stabilized by a collection of
remote three-center H-bonds closing 12- and 14-membered pseudorings.
Delineating the rules that govern helix formation depending on the nature of
constituent units is of practical utility if one aims to utilize this helical fold to
place side chains in a given arrangement and elaborate functional helices. In this
work, we tested whether the helix geometry is compatible with alternative
substitution patterns. The central −NH−CH(R)−CH2−NH−CO− residue in a
model oligourea pentamer sequence was replaced by guest units bearing various
substitution patterns [e.g., −NH−CH2−CH2−NH−CO−, −NH−CH2−CH-
(R)−NH−CO−, and −NH−CH(R1)−CH(R2)−NH−CO−], levels of preorganization (cyclic vs acyclic residues), and
stereochemistries, and the helix formation was systematically assessed. The extent of helix perturbation or stabilization was
primarily monitored in solution by Fourier transform IR, NMR, and electronic circular dichroism spectroscopies. Our results
indicate that although three new substitution patterns were accommodated in the 2.5-helix, the helical urea backbone in short
oligomers is particularly sensitive to variations in the residue substitution pattern (position and stereochemistry). For example,
the trans-1,2-diaminocyclohexane unit was experimentally found to break the helix nucleation, but the corresponding cis unit did
not. Theoretical calculations helped to rationalize these results. The conformational preferences in this series of oligoureas were
also studied at high resolution by X-ray structure analyses of a representative set of modified oligomers.
acyclic β2,3-amino acids of like configuration13,14 were found to
INTRODUCTION
■
be more effective than their β3 or β2 counterparts in promoting
the requisite synclinal arrangement around Cα−Cβ bonds.
Similarly, adding substituents at the 2-position (like config-
uration) or at both the 2- and 3-positions in γ4-amino acid
monomers was found to reduce significantly the number of
conformations devoid of syn-pentane interactions and to
reinforce optimal preorganization for 14-helix formation.15−17
It is noteworthy that the exploration of substitution patterns
can lead to the discovery of alternative folding patterns. β-
Peptides consisting of the trans-2-aminocyclopentyl carboxylic
acid (trans-ACPC)18,19 and trans-2-aminocyclobutyl carboxylic
acid (trans-ACBC)20 adopt a stable 12-helix with a 1←4 H-
bonding pattern that differs markedly from the 14-helix and
associated 1→3 H-bonding scheme. In addition, a unique β-
peptide 12,10-helical structure remote from the canonical 12-
The formation of stable and regular secondary structures
maintained by remote intramolecular H-bonds (e.g., helices)
requires preorganization of the main chain to position
sequentially remote H-bond donors and acceptors in close
spatial vicinity to enable optimal H-bonding to occur without
significant conformational alteration. In natural α-polypeptides,
preferred backbone conformations derive in part from the
minimization of eclipsing or Pitzer strain and pseudoallylic
A(1,3) strain.1 By analogy, restricted rotation about main-chain
C−C bonds is a key feature that drives helical folding of higher
β- and γ-peptide homologues.2−9 For example, both canonical
14-helical structures of β3- and γ4-peptides are characterized by
a (+)-synclinal arrangement (gauche conformation) around
main-chain ethane bonds. Substantial stabilization of these
helical folds has been achieved by increasing the level of
preorganization of β- and γ-amino acid constituents toward the
gauche conformation. In the β-peptide series, the trans-2-
aminocyclohexyl carboxylic acid (trans-ACHC) unit10−12 and
Received: February 1, 2013
Published: February 27, 2013
© 2013 American Chemical Society
4884
dx.doi.org/10.1021/ja401151v | J. Am. Chem. Soc. 2013, 135, 4884−4892

Journal of the American Chemical Society
Article
Figure 1. (left) Illustration of the canonical (P)-2.5-helical structure formed by N,N′-linked aliphatic oligoureas of type A together with the
nomenclature used for the main-chain torsion angles: OC−N−Cβ−Cα is ϕ, N−Cβ−Cα−N is θ1, and Cβ−Cα−N−CO is θ2. (right) Sequences of
model urea pentamer 1 (canonical units with the S configuration) and the alternative substitution patterns evaluated in this study.
and 14-helices has been reported for “mixed” β-peptides
consisting of alternating β3- and β2-amino acids.14,21,22
oligomers (2−6) were investigated using different experimental
approaches, including Fourier transform IR (FTIR), electronic
circular dichroism (ECD), and 1H NMR spectroscopies and X-
ray crystallography, and finally compared with that of 1.
Restriction of the conformational freedom of the C−C bonds
toward the gauche conformer is also a prerequisite for helical
folding in a family of aliphatic N,N′-linked oligoureas23 having
the general formula [−NHCH(R)CH2NHCO−]n. Homochiral
oligomers as short as four urea linkages have been shown to
fold into a well-defined helical structure with 2.5 residues per
helical turn and a pitch (rise per turn) of 5.1 Å (Figure 1
left).24−26 X-ray diffraction (XRD) analyses of oligoureas A
revealed a (+)-synclinal arrangement (g+ conformation)
around the ethane bond with an average backbone N−Cβ−
Cα−N torsion angle (θ1) of +57.8°.27 The helix is stabilized by
a collection of three-center H-bonds closing 12- and 14-atom
pseudorings. The chain-length dependence, the effect of
capping, and the influence of the solvent on helical folding
have been investigated in previous studies.28,29 However,
relatively little information is available about the folding
propensity of units with alternative substitution patterns and
their compatibility with the 2.5-helix geometry of oligour-
eas.30,31 We recently reported a fragment condensation
approach to long oligourea sequences and found that N-
(pyrrolidine-2-ylmethyl)ureido units can be inserted at discrete
positions without compromising 2.5-helical folding despite the
loss of one H-bond donor site.30 The conformational restriction
around the N−CH(R) bond imposed by the pyrrolidine ring
fixes the CO−N−Cβ−Cα angle (ϕ) to a value of approximately
−95.9°, which matches the one found in the canonical structure
of 2.5-helical oligoureas (i.e., ϕ ≈ −101°). In the present work,
we explored how the helix propensity is influenced by
alternative ethylenediamine monomers with noncanonical
substitution patterns and various conformational constraints
(Figure 1 right). To assess the extent to which these monomers
disrupt or stabilize the canonical 2.5-helical structure of
oligoureas, we prepared singly substituted analogues of the
model pentamer 1. The folding behaviors of these short
RESULTS AND DISCUSSION
■
Design and Synthesis. The singly substituted analogues of
oligourea pentamer 1 considered in this work are shown in
Figure 1. The central −NH−CH(R)−CH2−NH−CO− unit
was replaced by residues with deleted side chains as in 2,
residues with shifted side chains (Cβ → Cα) as in 3 and 4, or
rotationally restricted trans- and cis-1,2-diaminocyclohexane
units as in 5 and 6. Examination of the high-resolution
structures of the 2.5-helices of oligoureas27,30,32 reveals a
number of interesting features. The mean value of ∼60° for θ1
in the helical backbone matches well the corresponding values
in crystal structures of acylated 1,2-diaminocyclohexane
derivatives (mainly trans) reported in the literature,33−38 thus
suggesting that such monomers could fit in the 2.5-helix (e.g., 5
and 6). Whereas the side chains (R) of the canonical
monomeric units in A adopt an equatorial (lateral) orientation
with respect to the helix axis, the two hydrogens on the
adjacent methylene group both point in a more axial direction
(Figure 1 left). Therefore, it is more difficult to predict which
substitution on this carbon would be tolerated (i.e., 3 vs 4; 5 vs
6). To get a first hint about the folding propensities of
monosubstituted pentamers 2−6, we performed a Monte Carlo
(MC) conformational analysis with the GB/SA continuum
solvation model (see the Supporting Information). The
conformational search was optimized (force field and solvent)
with pentaurea 1, and main clusters of conformations were
compared to the previously reported crystal structure of 1.39
The best results were obtained with AMBER* as the force field
and water as the solvent. The lowest-energy cluster was a helical
conformation with a root-mean-square deviation (RMSD) of
0.6 Å relative to the crystal structure of 1. The second- and
4885
dx.doi.org/10.1021/ja401151v | J. Am. Chem. Soc. 2013, 135, 4884−4892

Journal of the American Chemical Society
Article
third-lowest-energy clusters contained partially helical struc-
tures. In such short-chain helical oligomers, the flexibility at
both termini reduces the stability of the helical states, and the
terminal segments tend to unravel. In the case of the
monosubstituted analogues 4, 5, and 6, low-energy clusters
were nevertheless populated by helical conformations akin to
the canonical 2.5-helix. In contrast, helical conformations were
hardly present in the clusters of conformations from
simulations of 2 and 3. This conformational search thus
suggests that in short-chain oligoureas, (i) the 2.5-helix
geometry might accommodate both cis- and trans-1,2-
diaminocyclohexane units, (ii) unsubstituted ethylenediamine
units destabilize helix formation, and (iii) the introduction of a
side chain on Cα is tolerated when combined with inversion of
configuration. To address these possibilities experimentally,
oligoureas 1−6 were synthesized in solution according to
Scheme 1 using (S)-succinimidyl-{2-[(tert-butoxycarbonyl)-
Chart 1. Activated Monomers 7−12 Used for the Synthesis
of Oligoureas 1−6
a
Scheme 1. Synthesis of Oligoureas 1−6
unambiguously determined by XRD analysis (see the
Supporting Information). After coupling of 11 and 12,
reduction of the cis-2-azidocyclohexylurea intermediate to the
corresponding amino derivative proved to be much more
difficult than that of the corresponding trans-2-azidocyclohex-
ylurea oligomer.
FTIR Signatures of 1 and Analogues 2−6. We first used
FTIR spectroscopy as a rapid method to compare the different
oligomers in both solution and the solid state. We previously
showed that the FTIR spectra of helical oligoureas of type A are
dominated by two strong bands centered near 1635 and 1570
cm−1. These two bands, which are essentially due to carbonyl
stretching and coupled C−N stretching/N−H deformation are
termed the “urea I” and “urea II” bands, respectively, by analogy
to the amide I and amide II bands in peptides and proteins.44
In methanol solution (1−2 mg mL−1), the spectrum of
model oligourea 1 (Figure 2a) displays the two intense urea I
and urea II bands at 1633 cm−1 (sharp) and 1583 cm−1
(broader), respectively. As shown previously, the urea II region
is more complex. It is split in two components, a strong one at
1583 cm−1 and a shoulder at 1554 cm−1, corresponding to the
coupled C−N stretching/N−H deformation with the two
vibrational components in phase and out of phase, respectively.
The additional band at 1602 cm−1 is probably a contribution
from the phenyl rings. The solution spectra of oligomer
analogues 2, 4, and 6 (Figure 2c,d,f) display a common
signature with an intense, sharp urea I band around 1630 cm−1
that suggests folding. Very similar spectra were obtained in the
solid state, suggesting that there is no major conformational
change in going from solution to the solid state (see Table 1
and the spectra in the Supporting Information). Conversely,
the solution spectra obtained for oligoureas 3b and 5 (Figure
2b,e) show extremely broad urea I and urea II bands with a shift
of the urea I band toward higher wavenumber. The band
broadening, which is even more pronounced in the solid state,
does not support helical folding but rather reflects strong
intermolecular interactions and aggregation behavior of these
two oligomers. Overall, these FTIR results support the idea that
the noncanonical units in 2, 4, and 6 but not those in 3 and 5
would be compatible with a 2.5-helix geometry.
a
Reagents and conditions: (i) TFA, 30 min; (ii) 7, DIPEA, MeCN, 2
h; (iii) 8−12, DIPEA, MeCN, 2 h; (iv) H2, Pd/C, 20% AcOH in
MeOH, 2 h; (v) ArNCO, DIPEA, DMF or MeCN, 2 h. Abbreviations:
DIPEA, diisopropylethylamine; TFA, trifluoroacetic acid; DMF, N,N-
dimethylformamide.
amino]-3-phenylpropyl}carbamate (7) to introduce the canon-
ical units. The overall yields of oligomers 2−6 ranged from 10
to 45% for the 10 steps (see the Supporting Information).
The specific building blocks 8−12 (whose preparation is
described in the Supporting Information) were used to
introduce residues with alternative substitution patterns
(Chart 1). Briefly, activated carbamates 7−10 were prepared
starting from N-Boc-protected amino acids using procedures
previously described in the literature.40−42 The trans- and cis-2-
azidocyclohexyl carbamate monomers 11 and 12, respectively,
were prepared from trans-2-aminocyclohexanol, which was
obtained by catalytic asymmetric ring opening of cyclohexene
oxide using the Jacobsen procedure.43 The structure of 11 was
1
Folding Propensity in Solution As Determined by H
NMR Spectroscopy. Additional information about the extent
of helix perturbation or stabilization induced by the non-
4886
dx.doi.org/10.1021/ja401151v | J. Am. Chem. Soc. 2013, 135, 4884−4892

Journal of the American Chemical Society
Article
Figure 2. FTIR spectra (1400−1800 cm−1) of oligoureas 1−6 recorded at room temperature in MeOH solution after subtraction of the spectrum of
pure MeOH.
Table 1. CO Stretching Frequencies for Oligomers 1−6 in Solution and the Solid State
a
solution
solid state
compound
urea I
urea II
urea I
urea II
1
1633(s)
1583(m), 1554(sh)
1580(m), 1555(m)
1591(sh)
1631(s)
1571(m), 1549(sh)
1578(s), 1556(sh)
1565(m, sh)
2
1638(m)
1647(m, br)
1633(s)
1635(s)
3b
4
1642(m, br)
1634(s)
1578(s), 1558(s)
1577(m, br)
1576(m), 1555(sh)
1562(m, br)
5
1655(m, br)
1631(s)
1630(m)
1630(s)
6
1580(s), 1554(m, sh)
1572(m), 1555(m)
a
Stretching frequencies in MeOH (1 or 2 mg mL−1). Abbreviations: s, strong absorption; m, medium absorption; br, broad; sh, shoulder.
1
canonical units in oligomers 2−6 was gained by H NMR
spectroscopy in CD3OH at a concentration of 1 or 2 mM. We
have shown that when placed in a helical environment, the
main-chain methylene protons of canonical units exhibit a high
degree of anisochronicity (Δδ ≥ 0.9−1 ppm for central
residues). Thus, the Δδ values can be used to sense the extent
of helix perturbation. Herein, the Δδ values for unmodified
1
residues (i.e., residues 1, 2, 4 and 5) were extracted from H
NMR spectra of oligomers 2−8 and compared to the
corresponding values in 1 (Figure 3). The insertion of an
achiral −NH−CH2−CH2−NH−CO− unit in 2 results in a
lower degree of anisochronicity of the methylene protons (by
∼0.2 ppm relative to 1) in each of the four canonical units. The
Δδ value for the central unsubstituted residue is 0.45 ppm
lower than that of the corresponding canonical chiral unit in 1.
Overall, these lower Δδ values may reflect either destabilization
of the helical structure or some local adjustment of the helix
geometry. Shifting the side chain on the adjacent backbone
carbon atom without configurational reversal as in 3a (in which
all of the residues have the S configuration) resulted in a
Figure 3. Anisochronicities (Δδ in ppm) of the main-chain methylene
protons of each residue in oligourea 1 (white bars) and the singly
■
●
△
substituted analogues 2 (white ), 4 (blue ), and 6 (red
)
measured in CD3OH (1−2 mM; 400, 500, or 700 MHz; 298 K).
Because the sequence assignment could not be inferred from the NMR
spectra of compounds 3a and 5 (see the text), only the range of
chemical shifts is indicated (blue arrow for 3a and red arrow for 5).
The maximum Δδ values measured for 3a and 5 are indicated by blue
and red dashed lines, respectively.
1
completely different outcome. The H NMR spectra of 3a in
CD3OH were poorly resolved, and an unambiguous sequence
assignment was precluded by significant overlap of the
resonances in the NH region. Nevertheless, all of the spin
systems could be resolved, and very limited diastereotopicity of
main-chain methylene protons was found (Δδ = 0.09−0.33
ppm). Substitution of a methyl group for the bulky benzyl side
chain in 3a did not yield any improvement, and 3b was not fully
soluble in CD3OH at 1 mM. In agreement with the MC
simulations and FTIR data, coupling of a shift of the
stereogenic center with inversion of configuration as in 4
(where the central unit has the R configuration) was found to
4887
dx.doi.org/10.1021/ja401151v | J. Am. Chem. Soc. 2013, 135, 4884−4892

Journal of the American Chemical Society
Article
restore a high degree of anisochronicity in the flanking
canonical units (Δδ = 0.66−1.10 ppm), possibly suggesting
2.5-helical folding. The finding that the (S,S)-1,2-diaminocy-
clohexane unit is not accommodated in the 2.5-helix was not
anticipated on the basis of the MC search but is consistent with
the FTIR results. This substitution pattern in 5 led to major
conformational changes, as shown by the significantly reduced
anisochronicity along the backbone (Δδ = 0.23−0.39 ppm). As
in 3b, the NH resonances in 5 were poorly dispersed,
precluding identification of NH(i)/N′H(i+1) nuclear Over-
hauser effect (NOE) cross-peaks and subsequent sequence
assignment. Interestingly, the (S,R)-1,2-diaminocyclohexane
unit was able to maintain high Δδ values in 6, suggesting
that the cis-1,2-diaminocyclohexane unit is compatible with the
2.5-helix geometry while the trans isomer is not.
The presence of nonsequential NOEs between backbone
NH(i)/N′H(i) and CβH(i+2) (see the Supporting Informa-
tion) provided further qualitative evidence that oligomers 2, 4,
and 6, like 1, can adopt a helical conformation. However, the
extraction of full NOE data sets for detailed NMR character-
ization and head-to-head comparisons of these short oligomers
was hampered by the redundancy of benzyl side chains and
numerous resonance overlaps in their NH/CH fingerprint
regions. Measurement of amide proton exchange in proteins
and α-peptides is another established method to obtain useful
structural information (stability, dynamics) at the residue
level.45,46 In the field of foldamers, this approach was recently
used to compare the relative stabilities of short α,β-peptide
helices (nucleated or not) using the H-bond surrogate (HBS)
technique.47 By analogy to oligoamides, we investigated
hydrogen/deuterium (H/D) exchange of backbone ureas in
oligourea helices 2, 4, and 6 in comparison with 1 (Figure 4;
also see the Supporting Information). Although a precise
determination of the H/D exchange rates for all of the
individual ureas was not possible because of overlaps in the 1D
Figure 4. H/D exchange of backbone urea protons in (a) oligourea 1
and (b) oligourea 4. The H spectra (800 MHz) of the oligoureas (1
mM in CD3OD) were recorded at 288 K.
1
1
and 2D H spectra, useful information was gained from the
exchange data for few representative residues with NH protons
putatively involved in intramolecular H-bonding. For example,
the NH urea protons of residues 1 and 3 in 1 and 6 have
exchange rates on the same order of magnitude (kex ≤ 10 ×
10−3 min−1). These protons exchange significantly more slowly
than those in 2 and 4 (25 × 10−3 min−1 ≤ kex ≤ 44 × 10−3
min−1), thus suggesting a higher folding propensity for this
region in 1 and 6 than in 2 and 4.
ECD Spectroscopy. It has been shown that N,N′-linked
oligoureas of type A display a characteristic ECD signature with
an intense maximum at ∼203 nm upon (P)-2.5-helix
formation.28 Additional spectral characteristics of helix
formation include a zero crossing at ∼193 nm and a mimimum
at 188 nm, which can be observed in trifluoroethanol (TFE) in
the absence of aromatic side chains.30 The ECD spectra of
oligomers 1−6 (10−4 M in TFE) are shown in Figure 5. The
poor ECD signal exhibited by oligoureas 3b and 5 reflects the
absence of a well-defined folded conformation, in line with the
NMR and FTIR data. In contrast, oligomers 1, 2, 4, and 6 all
display an ECD signature with a maximum in the positive molar
ellipticity at ∼203 nm, indicative of 2.5-helical folding. This is
also consistent with the FTIR and NMR results. The negative
band at 188 nm and the zero crossing at 193 nm are not visible
in these spectra because they are masked by the positive
contribution of the benzyl side chains.48 Nevertheless, the
intensities of the signals at 203 nm were tentatively used to
compare the helix-folding propensities of the different
Figure 5. ECD spectra of oligoureas 1−6 (10−4 M in TFE) at 298 K.
oligomers qualitatively. The ECD signals exhibited by 2 and
4 are significantly weaker than that of 1, indicating that the
absence of a substituent or the shift of the side chain from Cβ to
Cα in the central unit is destabilizing to some extent. In
contrast, the intensity of the positive band at 203 nm for 6 is
higher than that for 1, suggesting that the insertion of a
constrained cis-1,2-diaminocyclohexane unit is well-tolerated
and stabilizes 2.5-helix formation to some extent in comparison
with canonical acyclic units.
X-ray Crystallographic Analysis. To gain detailed
structural insight into the consequences of incorporating
noncanonical residues into oligoureas, we attempted to grow
single crystals of oligomers 2−6 suitable for XRD analysis. We
4888
dx.doi.org/10.1021/ja401151v | J. Am. Chem. Soc. 2013, 135, 4884−4892

Journal of the American Chemical Society
Article
Figure 6. Comparison of the structures of oligoureas 1, 2, 4, and 6 in the crystalline state: (top) side views; (bottom) top views. C atoms of the
noncanonical units in 2, 4, and 6 are depicted in orange.
succeeded only in those cases where the new substitution
pattern was shown by the FTIR, NMR, and ECD data to be
compatible with 2.5-helical folding (i.e., for 2, 4, and 6). Crystal
structures of 2, 4, and 6 were solved in the P212121 (2) and P21
(4 and 6) space groups (see the Supporting Information) and
compared to the previously solved structure of 1.39 As shown in
Figure 6, all three oligomers adopt a helical conformation akin
to the canonical (P)-2.5-helical structure of 1, thus confirming
the results obtained in solution. Overlays of the structures of 2,
4, and 6 with that of 1 were performed by fitting the five pairs
of β-carbons [CH(R) in the canonical units]. The low RMSDs
of 0.342, 0.380, and 0.241 Å for 2, 4(I), and 6(I), respectively,
indicate a close match between the structures of the oligoureas
with noncanonical residues and that of 1. The intramolecular
H-bond distances are also largely conserved among the four
structures. As shown in Table 2, the modified residues
accommodate the (P)-helix geometry with backbone torsion
angles nearly identical to that of the central canonical unit in 1.
An overlay of the cis-bis(ureido)cyclohexane unit at position
3 in 6 (C atoms in orange) with the corresponding canonical
acyclic unit in 1 is shown in Figure 7. The backbone carbon
Figure 7. Overlay of the cis-bis(ureido)cyclohexane unit at position 3
in the crystal structure of 6 (C atoms in orange) with the
corresponding canonical unit in the crystal structure of 1 (C atoms
in light gray).
Table 2. Main Backbone Torsion Angles (deg) of the
Central Canonical Unit in 1 and the Modified Units in
Oligomers 2, 4, and 6 (from Crystal Structures)
atoms and the first carbon of the side chain (Cγ) show very
little deviation between the two molecules, thus further
highlighting the ability of a six-membered ring to match the
helical backbone conformation of canonical acyclic units. It is
also noteworthy that the introduction of discrete noncanonical
units with substitution on Cα leads to a different spacing and
orientation of the side chains at the surface of the 2.5-helix. For
example, the distances between the i, i + 2, and i + 4 side chains
vary from 5.5 and 6.4 Å in 1 to 8.0 and 4.5 Å in 4 (see the
Supporting Information).
oligomer
ϕ
θ1
θ2
1
2
−101.4
−90.0
−98.6
−95.3
+54.4
+53.6
+51.6
+52.9
+77.6
+79.2
+83.7
+82.9
a
4
b
6
a
Means of the angles in the four independent molecules I−IV in the
asymmetric unit (ASU). Means of the angles in the two independent
molecules I and II in the ASU.
b
Substitution Pattern Requirements for 2.5-Helix
Formation: Why cis- and Not trans-1,2-Diaminocyclo-
4889
dx.doi.org/10.1021/ja401151v | J. Am. Chem. Soc. 2013, 135, 4884−4892

Journal of the American Chemical Society
Article
hexane Monomers? The opposite nature of the trans- and
cis-1,2-bis(ureido)cyclohexane units in 5 and 6, respectively, as
well as the central (S)- and (R)-1,2-bis(ureido)propane units in
3 and 4, respectively, is not easily inferred from examination of
the model shown in Figure 1 but can eventually be rationalized
by considering the torsion angle preferences in these units.
Examination of the torsion angles of (S)-2-substituted-1,2-
diaminoethane units in the crystal structures of model mono-
and diureas [e.g., (S)-Bn−NHCONH−CH(Bn)−CH2−
NHCONH−Me]49 and helical oligoureas27,30 revealed that,
as in L-peptides, the OC−N−Cβ−Cα angle ϕ adopts negative
values. This preference for negative ϕ values in canonical units
having the S configuration may thus suggest that shifting the
substitution position to the second carbon while keeping the
configuration constant, as in 3, may locally impose negative
values of the Cβ−Cα−N−CO angle θ2 and thus an antiparallel
orientation of the ureas, precluding P-helical folding despite the
presence of flanking canonical units. Combining a shift of the
substitution position with inversion of configuration, as in 4,
would then restore helical folding by allowing the central unit
to adopt a positive θ2 value. This may also hold true for the
(S,S)-trans- and (S,R)-cis-1,2-diaminocyclohexane units in 5 and
6, respectively.
energies of the two conformers was smaller (ΔHsyn−anti = −1.2
kcal mol−1).
Geometry optimizations were performed using density
functional theory with the wB97XD functional52 and the 6-
31G** basis set starting from the low-energy syn and anti
conformers identified in the semiempirical calculations. The
results (Figure 8) are consistent with those of semiempirical
To test this hypothesis and gain additional information about
conformational preferences of (1S,2S)-trans- and (1S,2R)-cis-
1,2-bis(ureido)cyclohexane units, we performed theoretical
calculations on the model compounds trans- and cis-1,1′-
(cyclohexane-1,2-diyl)bis(3-methylurea) (CHBU) (Chart 2).
Figure 8. Optimized geometries of (top) the anti conformers
(negative values of ϕ and θ2) and (bottom) the syn conformers
(negative ϕ and positive θ2 values) of (left) trans- and (right) cis-
CHBU.
Chart 2. Model trans- and cis-1,2-Bis(ureido)cyclohexane
Derivatives Used in the Calculations
annealing. For both trans- and cis-CHBU, the anti con-
formation is the more energetically stable. However, the energy
difference between the anti and syn conformers is significantly
smaller for cis-CHBU (0.8 kcal mol−1) than for trans-CHBU
(6.8 kcal mol−1). Altogether, these calculations suggest that a
syn conformation is accessible at a lower energetic cost for the
cis-1,2-diaminocyclohexane derivatives. Furthermore, the values
of the backbone torsion angles in the optimized syn
conformation of cis-CHBU (ϕ = −91°; θ1 = 60°; θ2 = 85°)
match well the backbone torsion angles of the 2.5-helix
geometry.
Finally, we screened the ϕ versus θ2 energy surfaces of trans-
and cis-CHBU by scanning ϕ and θ2 in steps of 18°. The ϕ
versus θ2 energy contours (Ramachandran-type plots) calcu-
lated at the wB97XD/6-31G** level are provided in the
Supporting Information. We observed that (i) both cis- and
trans-CHBU have several minima in the region corresponding
to negative values of ϕ and θ2 [e.g., (ϕ, θ2) = (−100°, −80°),
(−150°, −80°), and (−90°, −160°)] and (ii) only cis-CHBU
exhibits minima with positive values of θ2 [e.g., (−90°, 80°)].53
Overall, these results are in good agreement with the finding
that cis- but not trans-1,2-diaminocyclohexane-type units are
accommodated in the short 2.5-helices reported here.
The entire conformational energy surface of the two molecules
was first sampled by carrying out semiempirical calculations
using the RM1 method50 with the simulated annealing
technique implemented in Ampac 9.51 The search for the
various local minima on this surface was performed in two
stages: (i) a nonlocal search focused on 18 dihedral angles
corresponding to the torsions defining the different conformers
and (ii) a local energetic relaxation of all of the degrees of
freedom for each of the minima collected at each stage. In this
way, 30−50 conformers with energies within 8−10 kcal/mol of
the lowest-energy conformer were determined (see the
Supporting Information). The trans diastereomer (−116.7
kcal mol−1) was calculated to be more stable than the cis one
(−114.1 kcal mol−1). For trans-CBHU, the anti conformation
(i.e., negative values of ϕ and θ2; ureas pointing in opposite
directions) was found to be more stable than the syn
conformation (i.e., negative ϕ and positive θ2 values) by 5.8
kcal/mol. A syn conformation was found to be more stable for
cis-CBHU (−113.8 kcal mol−1), but the difference in the
CONCLUSION
■
We have investigated the extent to which the helical
conformation of short aliphatic peptidomimetic oligoureas is
affected by the substitution pattern and the relative
configuration of ethylenediamine units. Similar to β-amino
acids, the chiral ethylenediamine constituents of oligoureas are
chemically and structurally diverse building blocks with six
substitution positions (vs three for α-amino acids) and eight
possible configurational isomers (vs two for α-amino acids).
4890
dx.doi.org/10.1021/ja401151v | J. Am. Chem. Soc. 2013, 135, 4884−4892

Journal of the American Chemical Society
Article
The possibility of employing monomeric units with different
substitution patterns and various degrees of preorganization
may be useful for fine-tuning the arrangement of functional
groups at the helix surface, modulating the helix stability, and
ultimately enabling the design of more effective peptide mimics
and/or sophisticated folded architectures (e.g., helical bun-
dles54,55).
configuration tolerates the introduction in the reverse direction
of units having the opposite configuration [i.e., intrinsically
programmed to nucleate an (M)-2.5 helix] is intriguing and
may be used to modulate the distribution of side chains at the
surface of this helical fold. However, the maximum number of R
units that can be inserted without causing global unfolding of
the (P)-helix as well as authorized combinations (e.g., 1:1
pattern) still need to be determined precisely. It is worth
mentioning that hybrid β-peptides composed of β2/β3
dipeptide repeats (Cβ → Cα shift)14,21 or heterochiral
sequences [e.g., alternating (S)-β3 and (R)-β3 amino acid
residues]61,62 form mixed-helical structures (i.e., with alternat-
ing orientation of the backbone amides) that differ from the
canonical β-peptide 14-helix. It remains to be seen whether
oligoureas composed of heterochiral sequences would show a
similar propensity to adopt noncanonical secondary structures.
Our solution and solid-state data demonstrate that the
constrained cis-1,2-diaminocyclohexane-type unit is readily
accommodated in the 2.5-helix environment. The backbone
torsion angles ϕ, θ1, and θ2 of the cycloalkyl unit in 6 closely
match those of the acyclic residues in the canonical 2.5-helix
formed by homo-oligomers (e.g., 1). The enhancement of
helical folding by cyclic residues is well-documented among β-
peptide foldamers (e.g., stabilization of the 14-helix by trans-
ACHC residues).56−58 Our spectroscopic measurements
indicate that oligomers 1 and 6 both adopt a stable helical
structure, with the ECD spectra providing some support for an
increase in helical content upon insertion of a central cis-1,2-
diaminocyclohexane residue. However, the presence of benzyl
side chains and their contribution to the far-UV ECD spectra of
oligoureas 1−6 may render a quantitative comparison of the
ECD spectra more difficult. Interestingly, while this article was
in preparation, short homo-oligoureas made of even more
constrained bicyclic units [i.e., (S)-bicyclo[2.2.2]octane-1,2-
diamine) were reported.31 The finding that these oligomers
adopt a 2.5-helical conformation (though partially distorted in
the crystal state of a tetramer) along with the results described
here suggest that homo-oligoureas composed (S,R)-cis-1,2-
diaminocyclohexane units may also adopt a helical structure
akin to that of 1 and 6.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental methods, spectroscopic data, crystallographic
information files (CIF), and ϕ vs θ2 energy contour plots.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
g.guichard@iecb.u-bordeaux.fr
■
Present Address
⊥N.P.: Departments of Chemistry and Biochemistry, University
of Washington, Seattle, WA 98195, United States.
Notes
Calculations and experimental data also reveal that backbone
torsion angle preferences in trans-1,2-diaminocyclohexane units
favor an anti conformation (negative values of ϕ and θ2) with
flanking ureas pointing in opposite directions, which is not
compatible with the 2.5-helix geometry (at least in short-chain
oligomers). One can speculate that repeats of this motif would
rather generate extended conformations prone to the formation
of β-sheet-type arrangements, akin to that of β-peptides of cis-
aminocycloalkyl carboxylic acids, which adopt extended
conformations.59 In this work, we did not investigate cyclic
units with smaller ring sizes such as 1,2-diaminocyclopentane
and 1,2-diaminocyclobutane derivatives. It not clear whether
the constraints imposed by smaller rings would support torsion
angles associated with 2.5-helical folding. The anticipated
preference for larger θ1 values in four- and five-membered rings
may instead promote alternative folding patterns in the
corresponding homo-oligomers similar to what is observed
for β-peptides. trans-ACPC and trans-ACBC residues in the β-
peptide 12-helix environment are characterized by average N−
Cβ−Cα−CO angles of 94 and 96°, respectively.20,60
Previous work on modified oligourea sequences incorporat-
ing noncanonical N-pyrrolidine units (proline-like residues)30
or isosteric amide and carbamate units39 has revealed the
capacity of a few canonical urea monomers to impose their
conformational preference on units with otherwise limited
folding character. Our data indicate that units having no
substituent (glycine-like) or combining a shift of the
substitution position (Cβ → Cα) with inversion of configuration
(S → R) can be substituted for the canonical units at discrete
positions in the sequence, albeit with some destabilization of
the 2.5-helical fold in such short oligomers. Nevertheless, the
finding that the helix formed by units having the S
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported by the Centre National de la
■
Recherche Scientifique (CNRS), the Universite
́
de Bordeaux,
ional d’Aquitaine and by a grant from the
Agence Nationale de la Recherche (ANR-07-PCVI-0018). A
fellowship from the Universite de Strasbourg (N.P.) is
and the Conseil Reg
́
́
gratefully acknowledged. We also thank Immupharma France
and ANRT for a CIFRE predoctoral fellowship to P.C. The
authors also acknowledge computational facilities provided by
the Po
The crystallographic data were collected at the IECB X-ray
facility (UMS 3033, CNRS and Universite de Bordeaux, US
̂ ́ ́
le Modelisation of the Institut des Sciences Moleculaires.
́
001 INSERM), the ESRF (Grenoble, France), and the Swiss
Light Source (SLS) (Villigen, Switzerland). Beamline scientists
from ESRF ID29 and SLS X06SA are warmly acknowledged for
beamtime and help during data collection. We thank Vincent
Olieric for his support at the SLS synchrotron and Axelle
Grel
́
ard for her assistance with NMR measurements.
REFERENCES
■
(1) Quinkert, G.; Egert, E.; Griesinger, C. Aspects of Organic
Chemistry: Structure; Verlag Helvetica Chimica Acta: Basel, Switzer-
land, 1996.
(2) Seebach, D.; Matthews, J. L. Chem. Commun. 1997, 2015−2022.
(3) Banerjee, A.; Balaram, P. Curr. Sci. 1997, 73, 1067−1077.
(4) Gellman, S. H. Acc. Chem. Res. 1998, 31, 173−180.
(5) Cheng, R. P.; Gellman, S. H.; DeGrado, W. F. Chem. Rev. 2001,
101, 3219−3232.
(6) Seebach, D.; Beck, A. K.; Bierbaum, D. J. Chem. Biodiversity 2004,
1, 1111−1239.
4891
dx.doi.org/10.1021/ja401151v | J. Am. Chem. Soc. 2013, 135, 4884−4892

Journal of the American Chemical Society
Article
(7) Le Grel, P.; Guichard, G. In Foldamers: Structure, Properties, and
Applications; Hecht, S., Huc, I., Eds.; Wiley-VCH: Weinheim,
Germany, 2007; pp 35−74.
(37) Hanessian, S.; Vinci, V.; Fettis, K.; Maris, T.; Viet, M. T. P. J.
Org. Chem. 2007, 73, 1181−1191.
(38) Amendola, V.; Boiocchi, M.; Esteban-Gomez, D.; Fabbrizzi, L.;
Monzani, E. Org. Biomol. Chem. 2005, 3, 2632−2639.
(39) Pendem, N.; Nelli, Y. R.; Douat, C.; Fischer, L.; Laguerre, M.;
Ennifar, E.; Kauffmann, B.; Guichard, G. Angew. Chem., Int. Ed. 2013,
DOI: 10.1002/anie.201209838.
(8) Martinek, T. A.; Fulop, F. Chem. Soc. Rev. 2012, 41, 687−702.
̈
̈
(9) Bouiller
Amino Acids 2011, 41, 687−707.
̀
e, F.; Thet
́
iot-Laurent, S.; Kouklovsky, C.; Alezra, V.
(10) Appella, D. H.; Christianson, L. A.; Karle, I. L.; Powell, D. R.;
Gellman, S. H. J. Am. Chem. Soc. 1996, 118, 13071−13072.
(11) Appella, D. H.; Christianson, L. A.; Karle, I. L.; Powell, D. R.;
Gellman, S. H. J. Am. Chem. Soc. 1999, 121, 6206−6212.
(12) Barchi, J. J.; Huang, X.; Appella, D. H.; Christianson, L. A.;
Durell, S. R.; Gellman, S. H. J. Am. Chem. Soc. 2000, 122, 2711−2718.
(13) Seebach, D.; Ciceri, P. E.; Overhand, M.; Jaun, B.; Rigo, D. Helv.
Chim. Acta 1996, 79, 2043−2066.
(40) Kim, J.-M.; Bi, Y.; Paikoff, S. J.; Schultz, P. G. Tetrahedron Lett.
1996, 37, 5305−5308.
(41) Guichard, G.; Semetey, V.; Didierjean, C.; Aubry, A.; Briand, J.-
P.; Rodriguez, M. J. Org. Chem. 1999, 64, 8702−8705.
(42) Aisenbrey, C.; Nagendar, P.; Guichard, G.; Bechinger, B. Org.
Biomol. Chem. 2012, 10, 1440−1447.
(43) Schaus, S. E.; Larrow, J. F.; Jacobsen, E. N. J. Org. Chem. 1997,
62, 4197−4199.
(14) Seebach, D.; Abele, S.; Gademann, K.; Guichard, G.;
Hintermann, T.; Jaun, B.; Matthews, J. L.; Schreiber, J. V.; Oberer,
L.; Hommel, U.; Widmer, H. Helv. Chim. Acta 1998, 81, 932−982.
(15) Hanessian, S.; Luo, X.; Schaum, R.; Michnick, S. J. Am. Chem.
Soc. 1998, 120, 8569−8570.
(44) Cavagnat, D.; Claudon, P.; Fischer, L.; Guichard, G.; Desbat, B.
J. Phys. Chem. B 2011, 115, 4446−4452.
(45) Englander, S. W.; Mayne, L. Annu. Rev. Biophys. Biomol. Struct.
1992, 21, 243−265.
(46) Goodman, E. M.; Kim, P. S. Biochemistry 1991, 30, 11615−
(16) Seebach, D.; Brenner, M.; Rueping, M.; Schweizer, B.; Jaun, B.
Chem. Commun. 2001, 207−208.
11620.
(47) Patgiri, A.; Joy, S. T.; Arora, P. S. J. Am. Chem. Soc. 2012, 134,
11495−11502.
(48) Sreerama, N.; Manning, M. C.; Powers, M. E.; Zhang, J.-X.;
Goldenberg, D. P.; Woody, R. W. Biochemistry 1999, 38, 10814−
10822.
(49) Fischer, L.; Didierjean, C.; Jolibois, F.; Semetey, V.; Lozano, J.
M.; Briand, J. P.; Marraud, M.; Poteau, R.; Guichard, G. Org. Biomol.
Chem. 2008, 6, 2596−2610.
(50) Stewart, J. J. Mol. Model. 2007, 13, 1173−1213.
(51) AMPAC 9; Semichem, Inc.: Shawnee, KS, 2008.
(52) Chai, J.-D.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2008, 10,
6615−6620.
(17) Seebach, D.; Brenner, M.; Rueping, M.; Jaun, B. Chem.Eur. J.
2002, 8, 573−584.
(18) Appella, D. H.; Christianson, L. A.; Klein, D. A.; Powell, D. R.;
Huang, X.; Barchi, J. J.; Gellman, S. H. Nature 1997, 387, 381−384.
(19) Appella, D. H.; Christianson, L. A.; Klein, D. A.; Richards, M.
R.; Powell, D. R.; Gellman, S. H. J. Am. Chem. Soc. 1999, 121, 7574−
7581.
́
(20) Fernandes, C.; Faure, S.; Pereira, E.; Thery, V.; Declerck, V.;
Guillot, R.; Aitken, D. J. Org. Lett. 2010, 12, 3606−3609.
(21) Seebach, D.; Gademann, K.; Schreiber, J. V.; Matthews, J. L.;
Hintermann, T.; Jaun, B.; Oberer, L.; Hommel, U.; Widmer, H. Helv.
Chim. Acta 1997, 80, 2033−2038.
(53) The D-pro-(S,R)-cis-1,2-diaminocyclohexane unit has been used
previously as a reverse-turn scaffold to connect short urea segments. In
this case, the backbone torsion angles for the cis-1,2-diaminocyclohex-
ane unit are ϕ = −105°, θ1 = 56°, and θ2 = 162°. See: Medda, A. K.;
Park, C. M.; Jeon, A.; Kim, H.; Sohn, J.-H.; Lee, H.-S. Org. Lett. 2011,
13, 3486−3489.
(22) Rueping, M.; Schreiber, J. V.; Lelais, G.; Jaun, B.; Seebach, D.
Helv. Chim. Acta 2002, 85, 2577−2593.
(23) Burgess, K.; Shin, H.; Linthicum, D. S. Angew. Chem., Int. Ed.
Engl. 1995, 34, 907−909.
(24) Fischer, L.; Guichard, G. Org. Biomol. Chem. 2010, 8, 3101−
3117.
(54) Daniels, D. S.; Petersson, E. J.; Qiu, J. X.; Schepartz, A. J. Am.
Chem. Soc. 2007, 129, 1532−1533.
(25) Semetey, V.; Rognan, D.; Hemmerlin, C.; Graff, R.; Briand, J.-P.;
Marraud, M.; Guichard, G. Angew. Chem., Int. Ed. 2002, 41, 1893−
1895.
(55) Giuliano, M. W.; Horne, W. S.; Gellman, S. H. J. Am. Chem. Soc.
2009, 131, 9860−9861.
(26) Hemmerlin, C.; Marraud, M.; Rognan, D.; Graff, R.; Semetey,
V.; Briand, J.-P.; Guichard, G. Helv. Chim. Acta 2002, 85, 3692−3711.
(27) Fischer, L.; Claudon, P.; Pendem, N.; Miclet, E.; Didierjean, C.;
Ennifar, E.; Guichard, G. Angew. Chem., Int. Ed. 2010, 49, 1067−1070.
(28) Violette, A.; Averlant-Petit, M. C.; Semetey, V.; Hemmerlin, C.;
Casimir, R.; Graff, R.; Marraud, M.; Briand, J.-P.; Rognan, D.;
Guichard, G. J. Am. Chem. Soc. 2005, 127, 2156−2164.
(29) Violette, A.; Lancelot, N.; Poschalko, A.; Piotto, M.; Briand, J.
P.; Raya, J.; Elbayed, K.; Bianco, A.; Guichard, G. Chem.Eur. J. 2008,
14, 3874−3882.
(56) Raguse, T. L.; Porter, E. A.; Weisblum, B.; Gellman, S. H. J. Am.
Chem. Soc. 2002, 124, 12774−12785.
(57) Raguse, T. L.; Lai, J. R.; Gellman, S. H. J. Am. Chem. Soc. 2003,
125, 5592−5593.
(58) Seebach, D.; Jacobi, A.; Rueping, M.; Gademann, K.; Ernst, M.;
Jaun, B. Helv. Chim. Acta 2000, 83, 2115−2140.
(59) Martinek, T. A.; Tot
Angew. Chem., Int. Ed. 2002, 41, 1718−1721.
́ ́
h, G. K.; Vass, E.; Hollosi, M.; Fulop, F.
̈
̈
(60) Choi, S. H.; Guzei, I. A.; Spencer, L. C.; Gellman, S. H. J. Am.
Chem. Soc. 2010, 132, 13879−13885.
(30) Fremaux, J.; Fischer, L.; Arbogast, T.; Kauffmann, B.; Guichard,
G. Angew. Chem., Int. Ed. 2011, 50, 11382−11385.
(61) Sharma, G. V.; Reddy, K. R.; Krishna, P. R.; Sankar, A. R.;
Narsimulu, K.; Kumar, S. K.; Jayaprakash, P.; Jagannadh, B.; Kunwar,
A. C. J. Am. Chem. Soc. 2003, 125, 13670−13671.
́
(31) Legrand, B.; Andre, C.; Wenger, E.; Didierjean, C.; Averlant-
Petit, M. C.; Martinez, J.; Calmes, M.; Amblard, M. Angew. Chem., Int.
Ed. 2012, 51, 11267−11270.
(62) Man
́
dity, I. M.; Web
́
er, E.; Martinek, T. A.; Olajos, G.; Tot
́
h, G.
K.; Vass, E.; Fulop, F. Angew. Chem., Int. Ed. Engl. 2009, 48, 2171−
2175.
̈
̈
(32) Guichard, G.; Violette, A.; Chassaing, G.; Miclet, E. Magn. Reson.
Chem. 2008, 46, 918−924.
(33) Andre, C.; Luger, P.; Nehmzow, D.; Fuhrhop, J. H. Carbohydr.
Res. 1994, 261, 1−11.
(34) Anthony, S. P.; Basavaiah, K.; Radhakrishnan, T. P. Cryst.
Growth Des. 2005, 5, 1663−1666.
(35) Qiao, J. X.; Chang, C.-H.; Cheney, D. L.; Morin, P. E.; Wang, G.
Z.; King, S. R.; Wang, T. C.; Rendina, A. R.; Luettgen, J. M.; Knabb, R.
M.; Wexler, R. R.; Lam, P. Y. S. Bioorg. Med. Chem. Lett. 2007, 17,
4419−4427.
(36) Alfonso, I.; Bolte, M.; Bru, M.; Burguete, M. I.; Luis, S. V.
Chem.Eur. J. 2008, 14, 8879−8891.
4892
dx.doi.org/10.1021/ja401151v | J. Am. Chem. Soc. 2013, 135, 4884−4892