Conformation and chiral effects in α,β,α-tripeptides

Article abstract of DOI:10.1021/jo300892u

Short α,β,α-tripeptides comprising a central chiral trisubstituted β^2,2,3*-amino acid residue form unusual γ-turns and δ-turns in CDCl₃ and DMSO-d₆ solutions but do not form β-turns. Thermal coefficients of backbone amide protons, 2D-NMR spectra, and molecular modeling revealed that these motifs were strongly dependent on the configuration (chiral effect) of the central β-amino acid residue within the triad. Accordingly, SSS tripeptides adopted an intraresidual γ-turn like (C6) arrangement in the central β-amino acid, whereas SRS diastereomers preferred an extended δ-turn (C9) conformation. A different SRS-stabilizing bias was observed in the crystal structures of the same compounds, which shared the extended δ-turn (C9) found in solution, but incorporated an additional extended β-turn (C11) to form an overlapped double turn motif.

Full text of DOI:10.1021/jo300892u

Featured Article
pubs.acs.org/joc
Conformation and Chiral Effects in α,β,α-Tripeptides
Carlos J. Saavedra,† Alicia Boto,*,† Rosendo Hernandez,† Jose Ignacio Miranda,‡ and Jesus M. Aizpurua*,‡
́
́
†Instituto de Productos Naturales y Agrobiología del CSIC, Avda. Astrofísico Francisco San
́
chez 3, 38206 La Laguna, Tenerife, Spain
‡Departamento de Química Organ
́
ica I, Jose Mari Korta R&D Center, Universidad del País Vasco UPV/EHU, Avda. Tolosa-72,
20018 San Sebastian, Spain
S
* Supporting Information
ABSTRACT: Short α,β,α-tripeptides comprising a central
chiral trisubstituted β2,2,3*-amino acid residue form unusual γ-
turns and δ-turns in CDCl3 and DMSO-d6 solutions but do
not form β-turns. Thermal coefficients of backbone amide
protons, 2D-NMR spectra, and molecular modeling revealed
that these motifs were strongly dependent on the config-
uration (chiral effect) of the central β-amino acid residue
within the triad. Accordingly, SSS tripeptides adopted an
intraresidual γ-turn like (C6) arrangement in the central β-amino acid, whereas SRS diastereomers preferred an extended δ-turn
(C9) conformation. A different SRS-stabilizing bias was observed in the crystal structures of the same compounds, which shared
the extended δ-turn (C9) found in solution, but incorporated an additional extended β-turn (C11) to form an overlapped double
turn motif.
the types C6−C11 (Figure 1) and to compare them to the C9 +
C11 double hydrogen-bond pattern found in the solid state.10
According to crystal data (Figure 2), the N-(4-iodobenzoyl)-
protected tripeptides 2 and 7 with a SRS configuration of the
central triad feature identical double-turned motifs comprising
an extended δ-turn (C9) and an extended β-turn (C11). The two
overlapped turns share all the peptide chain atoms of the central
INTRODUCTION
■
β-Amino acids exhibit a stricking ability to attain secondary
structures far more efficiently than α-amino acids, and the term
“foldamer”1 has been coined for them and for some α,β-hybrids
containing them.2 As a result of their predictable folding patterns
and their improved stability to proteolytic cleavage,3 β-amino
acid containing peptides are the peptidomimetics of choice in
many biomedical applications.4
β
2,2,3-amino acid residue and are further stabilized by
In the late 1990s, Gellman5 and Seebach6 established the
general folding trends for β-peptides containing monosubsti-
tuted β2- and β3-amino acids or disubstituted β2,3-amino acid
units bearing proteinogenic side chains. Seebach also reported
the conformational behavior of β-tripeptides constituted by
geminally disubstituted β2,2-amino acid residues. In the solid
state, and depending on the nature of the β2,2-substituents, such
3-mer peptides adopt either extended γ-turns (C8 hydrogen-
bonded) or doubly extended δ-turns (C10), which are
reminiscent of the canonical β-turns formed by α-amino acid
tetrapeptides.7 Introduction of a DPro-β3,3-amino acid α,β-
dipeptide segment at the (i − 1) − (i + 2) positions of a longer α-
peptide stabilizes β-hairpins through extended β-turns (C11),
either in the crystal or in methanol, as demonstrated by Balaram.8
Far less attention has been drawn to trisubstituted β2,2,3-amino
acid containing hybrid α,β-peptides, despite the valuable
protease inhibition properties shown by some of these
compounds.9
intermolecular NH···OC bonding. In the case of the C11
turn, a hydrogen bond forms between the 4-iodobenzoyl
carbonyl oxygen and the Phe or Leu N7H amides, whereas for
the C9 turn the hydrogen bond is between the N4H amide and
the terminal ester carbonyl oxygen. The structures are further
stabilized inside the crystal cell by additional intermolecular
hydrogen bonds between the N2H amide proton and the CO
oxygen of the β-amino acid residue of a contiguous molecule (not
shown in Figure 2).
RESULTS AND DISCUSSION
■
The model tripeptides 1−7 were prepared as shown in Scheme 1
for compounds 3 and 4. Thus, the α-dipeptide 8 underwent a
one-pot radical decarboxylation−oxidation−alkylation process10
to give the α,β-dipeptide 9 in good yield as a 2:1 diastereomer
mixture. The saponification of the methyl ester, followed by
coupling to H-Phe-OMe, afforded a separable mixture of α,β,α-
tripeptides 3 (51%) and 4 (31%), which were used in the
conformational studies. The other tripeptides 1, 2, and 5−7 have
been previously reported.10
Herein we report the first conformational study conducted in
solution for model hybrid α,β,α-tripeptides 1−7 (Figure 1)
containing a trisubstituted central β2,2,3-amino acid residue with a
single stereocenter at position β3. The objective of this study was
to establish the chiral effect exerted by the configuration of such
stereocenter on the stabilization of several turned structures of
Received: May 7, 2012
Published: July 5, 2012
© 2012 American Chemical Society
5907
dx.doi.org/10.1021/jo300892u | J. Org. Chem. 2012, 77, 5907−5913

The Journal of Organic Chemistry
Featured Article
Scheme 1. Preparation of Model Tripeptides 3 and 4
diastereomers (1/2, 3/4, and 5/6) according to the (S) or (R)
configuration of the central β-amino acid residue, but in all
instances, the configuration of the flanking α-amino acids was S
(L) to represent natural peptides. Finally, the tripeptides were
protected at the N-termini by 4-iodobenzoyl, benzoyl, or Cbz
groups and were capped at the C-termini as methyl esters. As
mentioned above, the 4-iodobenzoylamido tripeptides of SRS
configurations provided crystals (2 and 7) which were used to
compare their C9 + C11 structures with the solution structures of
2 and 6 (the p-I-benzoyl analogue of compound 7), respectively.
Although water was considered first as the most biologically
meaningful medium for the study, finally less polar solvents were
chosen for solubility reasons. Thus, each peptide was dissolved
into DMSO-d6 and CDCl3 (5 × 10−3 M), and all protons were
unambiguously assigned from COSY, TOCSY, HSQC, and
HMBC spectra. Compounds 1−7 yielded sharp, well-resolved
NMR spectra consisting of single sets of signals in both solvents.
Figure 1. Model α,β,α-tripeptides 1−7 selected for conformational
analysis and turn motifs in short peptides, comprising α-amino acids
(top) and combinations of α- and β2,2,3*-amino acids (bottom). All
structures are depicted as N→C sequences.
These model peptides 1−7 (Table 1) were designed to have
different combinations of side chains (Me, i-Pr, i-Bu, Bn) with
variable steric demand and hydrophobicity in order to study their
interference with the intramolecular hydrogen bonding pattern
in solution. Peptides 1−6 could be grouped into three pairs of
1
Experiments were then performed to determine the H NMR
temperature coefficients11 for the exchangeable amide protons of
the tripeptides in DMSO-d6 solutions over a temperature range
of 300−330 K at 5 K intervals (see Table 1). Vicinal coupling
3
constants J(HN−CαH) were also measured, and the corre-
sponding ϕ dihedral angles were calculated using the Karplus
equation.12 Dihedral angles in the solid state for crystalline
compounds 2 and 7 are also included in Table 1.
Relevant noncontiguous interproton distances for peptides 1−
6 (Table 2; see also Supporting Information Table S1) were
calculated following the ISPA (isolated spin-pair approximation)
method13 from the integration of key NOESY crosspeaks
recorded in CDCl3 solvent at 300 K (500 MHz) with mixing
times of 400 ms. Interproton distances calculated in DMSO-d6
solutions delivered values essentially identical to those obtained
in CDCl3 (see Table 2). Diagnostic interproton distances around
the central β-amino acid residue are collected for peptides 1, 2
and 5, 6 in Figure 4.
1
Conformational Analysis. An inspection of the H NMR
spectra recorded from solutions of the tripeptides 1−7 in CDCl3
or DMSO-d6 soon revealed the absence of some key deshielded
amide NH required to support the overlapped double turn
conformations of 2 and 7 in the solid state (Figure 2), suggesting
that the large extended β-turn (C11) of the crystals could break
down when dissolved. Indeed, only the N4H amide protons of
the β-amino acid residues participated in the intramolecular
Figure 2. Crystal structures of α,β,α-tripeptides 2 and 7 containing a
chiral trisubstituted central β2,2,3*-amino acid residue. A (C9 + C11)
hydrogen bonding pattern with two overlapped turns is formed in each
case.
5908
dx.doi.org/10.1021/jo300892u | J. Org. Chem. 2012, 77, 5907−5913

The Journal of Organic Chemistry
Featured Article
Table 1. Chemical shifts (δ), Amide NH Thermal Coefficients (Δδ/ΔT), and HN−CαH Dihedral Angles (ϕ) Measured for α,β,α-
Tripeptides 1−7
peptide
1 (SSS)
2 (SRS)
3 (SSS)
4 (SRS)
5 (SSS)
6 (SRS)
7 (SRS)
a
N2H(δ)
7.42 (+1.11)
7.12 (+0.46)
6.15 (+1.66)
7.83 (+0.80)
7.67 (−0.14)
7.22 (+0.55)
7.08 (+1.34)
6.95 (+0.40)
6.17 (+1.60)
7.26 (+1.24)
7.19 (+0.17)
7.41 (+0.30)
5.00 (+2.00)
7.17 (+0.26)
6.17 (+1.84)
5.24 (+1.99)
7.56 (+0.00)
7.16 (+0.65)
6.71 (+1.25)
7.72 (+0.11)
7.48 (+0.92)
a
a
N4H(δ)
N7H(δ)
b
b
b
N2H(Δδ/ΔT)
N4H(Δδ/ΔT)
N7H(Δδ/ΔT)
−4.5
−3.7
−5.9
6.8 (−140)
9.2 (+164)
−4.9
−3.8
−5.6
−4.0
−4.0
−6.4
5.9 (−133)
10.5 (+180)
−4.8
−4.6
−5.4
7.4 (−142)
9.8 (+180)
−5.9
−3.7
−6.9
8.1 (−152)
9.8 (+180)
7.5 (+146)
−6.5
−4.9
−5.4
−5.2
−4.7
−6.0
c
c
c
d
d
3J(ϕ HN2−C3H)
3J(ϕ HN4−C5H)
3J(ϕ HN7−C8H)
6.3 (−136)[−149]
10.2 (+180)[+137]
7.9 (+147)
7.9 (+147)[−158]
10.0 (+170)[+166]
d
d
10.2 (+174)
7.8 (+147)
d
d
7.5 (−146)
7.9 (−150)[−167]
7.8 (−151)
7.7 (−148)
7.9 (+149)[−148]
a
Chemical shifts (ppm) measured in CDCl3. Values in parentheses represent variations of the chemical shift when the CDCl3 solvent was changed to
b
c
DMSO-d6. Thermal coefficients in ppb/K measured in DMSO-d6. Coupling constants (J) in Hz. The ϕ dihedral angles, in parentheses, were
calculated from the Karplus equation. Dihedral angles, in brackets, from X-ray data, ref 10a.
d
a
Table 2. Key Interproton Distances (Å) Calculated from NOESY Experiments for α,β,α-Tripeptides 1 and 2 in CDCl3 and
DMSO-d6 Solutions
b
c
NOE
Ha-Hb
1 (CDCl3)
1 (DMSO-d6)
2 (CDCl3)
2 (DMSO-d6)
2 (X-ray)
1
2
Ar1ortho−N2H
N2H−C3H
N2H−C3βMe
C3βMe−N4H
N4H−C5H
N4H−C5βMe
N4H−C6MeR
N4H−C6MeS
N4H−N7H
2.16ref
2.16ref
2.16ref
2.16ref
4.0
2.5
3.8
4.3
3.1
2.8
4.1
2.9
2.2
2.8
2.8
2.8
3.4
4.1
4.3
[2.2]
[2.7]
[2.6]
[3.1]
[2.7]
[2.6]
[3.0]
[4.3]
[2.3]
[2.5]
[2.5]
[2.6]
[4.0]
[2.5]
[2.5]
2.9
3.0
3.8
3
2.4
2.4
2.4
4
4.1
4.1
5
3.3
3.1
3.9
6
2.6
3.0
7
4.2
2.7
4.5
2.7
2.6
2.6
2.2
2.9
4.5
2.6
8
2.7
2.9
9
10
11
12
13
14
15
16
C5H−C6MeR
C5H−C6MeS
N7H−C6MeR
N7H−C6MeS
N7H−C8H
2.5
2.5
2.5
3.0
2.6
3.7
3.6
2.7
3.1
2.5
2.7
N7H−C8βH
Ar8H−C9(OMe)
a
b
c
Experiments (500 MHz) carried out at 300 K, mixing time 400 ms. Reference distance. Crystal interproton distances from X-ray analysis of 2 (C9
+ C11 conformer).
hydrogen bond network of the peptides in solution, while the
N7H amide protons were exposed to the solvent. This was
evident from the analysis of thermal coefficients in DMSO-d6
and SRS conformers (2, 4, 6, and 7), we noticed that the chemical
shift of amide N7H protons experienced different downfield
displacement upon solvent changing from CDCl3 to DMSO-d6,
depending on the β-amino acid β-carbon configuration. Thus,
N7H amide peak shifted downfield by 1.7−1.8 ppm in peptides 1,
3 and 5 (SSS configuration), whereas peptides 2, 4, 6, and 7 (SRS
configuration) shifted only 0.3−0.6 ppm. This lower exposition
to coordinating solvent suggests a conformation with the N7H
group oriented toward the inner side of the peptide turn
backbone in SRS diastereomers.
11
(Table 1, rows 4−6) showing absolute values in the range −3.7 to
−4.9 ppb/K for N4H protons, whereas amides N2H and N7H
gave larger thermal coefficients (up to −6.9 ppb/K). Solvent
change from nonacceptor CDCl3 to acceptor DMSO-d6 resulted
in very small chemical shift for β-amino acid N4H amides in all
peptides (0.0−0.4 ppm), but in a significantly larger downfield
shift (0.4−1.9 ppm) for α-amino acid amides N2H and N7H
(Table 1, rows 1−3). Seeking for more insight into the origin of
the chiral effect differentiating the SSS conformers (1, 3 and 5)
All of the ϕ dihedral angles measured in solution for the vicinal
(HN−CαH) atoms in the three amino acids of peptides 1−7
5909
dx.doi.org/10.1021/jo300892u | J. Org. Chem. 2012, 77, 5907−5913

The Journal of Organic Chemistry
Featured Article
showed large values, in the range 133−180°, and were
compatible with quasi-staggered dispositions of the HN−CH
bonds (Table 1, rows 7−9). They were also very close to the
dihedral angles of the crystalline peptides 2 and 7 (Table 1, values
in brackets) suggesting that, after the solution of tripeptides 1−7
in CDCl3 or DMSO-d6, the relative spatial arrangement of the α-
amino acid residues around the β-amino acid should resemble an
open angular shape in SSS diastereomers and a closed turned
shape in SRS diastereomers (Figure 3). These diastereomeric
Figure 3. Main folding pattern (chiral effect) for peptides 1−6 in
solution.
conformers, arising from rotations of the (Me2C−CO) bond in
the central β-amino acid residue, would be spatially directed by
the arrangement of the R2 substituent attached to the β3 chiral
stereocenter *C5H and further stabilized with the formation of
two different intramolecular hydrogen bonds (C6) and (C9).
An examination of the noncontiguous interproton distances
collected in Figure 4 (see also Table 2 and Supporting
Information Table S1) allowed the identification of some
diagnostic NOESY crosspeaks around the central β-amino acid
residue to discriminate the prevalent conformations of SSS and
SRS diastereomers in tripeptides 1, 2 and 5, 6. Unfortunately,
some key peaks of tripeptides 3 and 4 (e.g., the β-amino acid
Figure 4. NMR−NOESY interproton distances (in Å) and main
conformation structures for peptides 1, 2 and 5, 6 in CDCl3 solution and
in the solid state (for 2 and 7). Diagnostic distances are highlighted in
red.
in SSS tripeptides 1 and 5 but much closer in SRS diastereomers 2
and 6, thus being suitable for diagnostic NOE analysis. As shown
in Figure 5 for the NOE signals of peptides 1 and 2, the
crosspeaks between N4H/N7H amide protons (NOE f) were
actually found in the NOESY spectrum of 2, integrating for a 2.5
Å distance, whereas these crosspeaks were absent in the NOESY
spectra of the SSS tripeptide 1 (for analogous spectra of 5 and 6,
see the Supporting Information, Figure S13). Consistent with the
1
geminal methyl groups) were overlapped in the H NMR
spectra, and these compounds were discarded from NOESY-
based interproton distance analysis.
In the β-amino acid residue of peptides 1, 2 and 5, 6, distances
between the C5H proton and the geminal diastereotopic methyl
groups (Me6pro‑R/Me6pro‑S) were very similar to one another in all
instances, either in solution or in the solid state. This confirmed
the antiperiplanar disposition of the C5−H proton and the β-
amino acid’s carbonyl group as the most stable and largely
preferred conformation of the central residue in these tripeptides.
Likewise, the distance between the N7H amide proton and the
geminal methyl groups at position C6 showed a similar pattern in
all instances, with the N7H proton significantly closer to the
proposed conformers, strong NOE crosspeaks for N4H amide
pro‑R
protons with Me6
group were also found only in the SRS
tripeptides 2 and 6, but not in the SSS diastereomer counterparts
1 and 5. In the later tripeptides, the N4H amide protons gave
pro‑S
strong NOE interactions only with the diastereotopic Me6
protons (see NOE d in Figure 5).
No significant additional long-range inter-residual NOE
crosspeaks were found in the NOESY spectra of peptides 1−6.
Furthermore, the Molecular Mechanics minimization of 1−6
including the NMR interproton distance restrictions and ϕ
dihedral angles measured in solution, yielded essentially single-
conformer structures in each case. These observations
collectively suggested that the chiral conformational bias
observed for tripeptides 1−6 in solution is roughly independent
of the size and nature of the R1, R2 and R3 substituents and, most
important, that incorporation of α,β,α-peptide segments with a
β2,2,3*-amino acid in longer peptides could be envisaged to
prepare novel peptidomimetics with predictable shapes in
solution (Figure 6).14
Me6pro‑R group than to the Me6pro‑S group. In the crystal, peptides
pro‑S
2 and 7 had a large N7H − Me6
distance of ∼4.0 Å, which
permitted the involvement of the N7H amide proton in an
extended β-turn (C11). Upon solution in CDCl3 such hydrogen
bond vanished in SRS isomers, the N7H amide proton
pro‑R
approached the Me6
group and only the more stable (C9)
extended δ-turn remained (Figure 4). Conversely, the
conformation of SSS tripeptides 1 and 5 was stabilized by an
unusual intrarresidual (C6) H-bond.
Assuming (C6) and (C9) preferred conformations for α,β,α-
pro‑R
tripeptides 1−6 in solution, the N4H/N7H and N4H/Me6
pairs of protons were anticipated to be separated from each other
5910
dx.doi.org/10.1021/jo300892u | J. Org. Chem. 2012, 77, 5907−5913

The Journal of Organic Chemistry
Featured Article
H-bonded between the N−H and CO groups of the central β-
amino acid residue. Conversely, tripeptides with heterochiral SRS
relative configuration stabilize an extended δ-turn with a (C9)
hydrogen bond between the N−H of the β-amino acid residue
and the CO of the next α-amino acid. These results, based on
amide NH thermal coefficient measurements, diagnostic
NOESY crosspeak analysis and computational modeling, also
evidenced significant differences with the solid state conforma-
tions previously reported for similar heterochiral SRS tripeptides,
which were characterized by the additional stabilization of the
extended δ-turn (C9) with an extended β-turn (C11) spanning
from the CO of the protecting group and the N−H of the C-
terminal α-amino acid.
EXPERIMENTAL SECTION
■
General Experimental Methods. Commercially available reagents
and solvents were analytical grade or were purified by standard
procedures prior to use. All reactions involving air- or moisture-sensitive
materials were carried out under nitrogen atmosphere. The spray
reagents for TLC analysis were 0.25% ninhydrin in ethanol and/or
Fleet’s reagent [Ce(SO4)2 (0.5 g) and ammonium phosphomolybdate
hydrate (2.5 g) in H2SO4 (5 mL) and water (65 mL)]. Once sprayed, the
TLC was heated until development of color. Merck silica gel 60 PF254
and 60 (0.063−0.2 mm) were used for rotatory chromatography and
column chromatography, respectively. Melting points were determined
with a hot-stage apparatus and are uncorrected. Optical rotations were
measured at the sodium line at ambient temperature (26 °C). NMR
spectra were determined at 500 MHz for 1H and 125.7 MHz for 13C, and
Mass spectra (EI) were determined at 70 eV using an ion trap mass
1
analyzer. H NMR references: CDCl3 (δ 7.26), DMSO (δ 2.49). 13C
NMR references: CDCl3 (δ 77.0), DMSO (δ 39.5).
Preparation of Compounds 1−7. The preparation and
spectroscopic data of compounds 1, 2, and 5−7 was reported
previously.10a The syntheses of compounds 3 and 4, and their precursor
9, are reported below.
N-Benzoyl-L-alanyl-α,α-dimethyl-(S)-β-homoleucine Methyl Ester
(9). To a solution of Bz-Ala-Leu-OH (8) (61 mg, 0.2 mmol) in dry
dichloromethane (6 mL) were added iodine (15 mg, 0.06 mmol) and
diacetoxyiodobenzene (DIB) (97 mg, 0.3 mmol). The reaction mixture
was stirred at 26 °C for 4 h under irradiation with visible light. Then the
solution was cooled to 0 °C, and methyl (trimethylsilyl)dimethylketene
acetal (203 μL, 174 mg, 1.0 mmol) was injected, followed by dropwise
addition of BF3·OEt2 (51 μL, 57 mg, 0.4 mmol). The mixture was
allowed to reach room temperature and stirred for 3 h; it was then
poured into 10% aqueous Na2S2O3/saturated aqueous NaHCO3 (1:1,
10 mL) and extracted with CH2Cl2. The organic layer was dried on
sodium sulfate, filtered, and evaporated under vacuum. The residue was
purified by rotatory chromatography 85:15 (hexanes/ethyl acetate
mixtures) to give the product 8 (56 mg, 77%) as a 2:1 diastereomer
mixture: amorphous solid; IR (CHCl3) νmax 3424, 3323, 1721, 1676,
1652, 1511, 1484 cm−1; 1H NMR (500 MHz, CDCl3) δH major
diastereomer: 0.75 (3H, d, J = 6.6 Hz, 5γ-Mea), 0.84 (3H, d, J = 6.3 Hz,
5γ-Meb), 1.19 (3H, s, 6-Mea), 1.19−1.35 (2H, m, 5β-H2), 1.20 (3H, s, 6-
Meb), 1.49 (1H, m, 5γ-H), 1.53 (3H, d, J = 6.9 Hz, 3-Me), 3.66 (3H, s,
OMe), 4.15 (1H, m, 5-H), 4.78 (1H, m, 3-H), 6.83 (1H, br d, J = 9.8 Hz,
NHLeu), 7.21 (1H, br d, J = 7.3 Hz, NHAla), 7.40 (2H, dd, J = 7.3, 7.9 Hz,
Ar), 7.48 (1H, dd, J = 7.6, 7.8 Hz, Ar), 7.78 (2H, d, J = 8.5 Hz, Ar); minor
isomer: 0.88 (3H, d, J = 6.6 Hz, 5γ-Mea), 0.90 (3H, d, J = 6.3 Hz, 5γ-
Meb), 1.11 (3H, s, 6-Mea), 1.15 (3H, s, 6-Meb), 1.11−1.20 (2H, m, 5β-
H2), 1.51 (3H, d, J = 6.9 Hz, 3-Me), 1.57 (1H, m, 5γ-H), 3.63 (3H, s,
OMe), 4.15 (1H, m, 5-H), 4.78 (1H, m, 3-H), 6.81 (1H, br d, J = 8.5 Hz,
NHLeu), 7.23 (1H, br d, J = 9.1 Hz, NHAla), 7.39 (2H, dd, J = 7.9, 8.0 Hz,
Ar), 7.47 (1H, dd, J = 7.6, 7.8 Hz, Ar), 7.79 (2H, d, J = 8.7 Hz, Ar); 13C
NMR (125.7 MHz, CDCl3) δC major diastereomer: 19.2 (CH3), 21.32
(CH3), 22.1 (CH3), 23.1 (CH3), 23.7 (CH3), 25.0 (CH), 40.1 (CH2),
46.7 (C), 49.5 (CH), 51.8 (CH3), 53.3 (CH), 127.0 (2 × CH), 128.5 (2
× CH), 131.7 (CH), 133.9 (C), 167.1 (C), 172.2 (C), 177.0 (C); minor
diastereomer: 18.8 (CH3), 21.27 (CH3), 21.9 (CH3), 23.3 (CH3), 23.8
Figure 5. Diagnostic NOE interactions for peptides 1 and 2: Spatially
close protons in 1 (A) and 2 (D). Molecular Mechanics (SYBIL force
field) minimum energy conformers of 1 (B) and 2 (E) restrained with
NMR interproton distances and ϕ dihedral angles (only amide protons
are shown for clarity). Expansions of key NOESY crosspeaks for 1 (C)
and 2 (F). For position numbering, see structures in Table 2.
Figure 6. General intramolecular hydrogen-bond patterns populated in
solution for α,β-hybrid peptides containing a central β2,2,3* chiral
residue.
CONCLUSIONS
■
A uniform chiral effect was found to dictate the conformational
behavior of α,β,α-tripeptides containing a central β2,2,3*-amino
acid residue with a single stereocenter at β3 position when such
peptides were dissolved in CDCl3 and DMSO-d6 solvents.
Tripeptides with homochiral SSS relative configuration populate
in solution a conformation characterized by an unusual (C6) turn
5911
dx.doi.org/10.1021/jo300892u | J. Org. Chem. 2012, 77, 5907−5913

The Journal of Organic Chemistry
Featured Article
(CH3), 25.2 (CH), 40.3 (CH2), 46.5 (C), 49.5 (CH), 51.8 (CH3), 53.3
(CH), 127.0 (2 × CH), 128.5 (2 × CH), 131.7 (CH), 133.8 (C), 167.1
(C), 172.4 (C), 177.1 (C); MS (EI) m/z (rel intensity) 363 (M+ + H, 2),
176 ([PhCONHCH(Me)CO]+, 25), 149 ([PhCONHCH(Me) + H]+,
53), 148 ([PhCONHCH(Me)]+, 62), 105 ([PhCO]+, 100), 86
([NH2CHCH2CHMe2], 58); HRMS (EI) calcd for C20H31N2O4
363.2284, found 363.2270; calcd for C10H10NO2 176.0712, found
176.0707; calcd for C9H11NO 149.0841, found 149.0841; calcd for
C9H10NO 148.0762, found 148.0758; calcd for C7H5O 105.0340, found
105.0337; calcd for C5H12N, 86.0970, found 86.0967. Anal. Calcd for
C20H30N2O4: C, 66.27; H, 8.34; N, 7.73. Found: C, 66.26; H, 8.35; N,
7.37.
(2 × CH), 131.6 (CH), 133.6 (C), 137.0 (C), 167.6 (C), 172.8 (C),
174.5 (C), 175.6 (C); MS (EI) m/z (rel intensity) 509 (M+, 1), 361 (M+
− PhCONHCHMe, 20), 249 ([Me2CCONHCH(CH2Ph)CO2Me +
H]+, 60), 148 ([PhCONHCH(Me)]+, 36), 105 ([PhCO]+, 100), 86
([NH2CHCH2CHMe2], 70); HRMS (EI) calcd for C29H39N3O5
509.2890, found 509.2882; calcd for C20H29N2O4 361.2127, found
361.2126; calcd for C14H19NO3 249.1365, found 249.1358; calcd for
C9H10NO 148.0762, found 148.0767; calcd for C7H5O 105.0340, found
105.0336; calcd for C5H12N 86.0970, found 86.0970. Anal. Calcd for
C29H39N3O5: C, 68.34; H, 7.71; N, 8.25. Found: C, 68.64; H, 7.83; N,
8.20.
N-Benzoyl-L-alanyl-[α,α-dimethyl-(S)-β-homoleucyl]-L-phenylala-
nine Methyl Eester (3) and N-Benzoyl-L-alanyl-[α,α-dimethyl-(R)-β-
homoleucyl]-L-phenylalanine Methyl Ester (4). To a solution of the
dipeptide mixture 8 (120 mg, 0.33 mmol) in methanol (7 mL) at 0 °C
was slowly added 2 N aqueous NaOH (3 mL). The reaction mixture was
allowed to reach 26 °C and stirred for 64 h, and then it was cooled to 0
°C, diluted with water, poured into 5% HCl, and extracted with EtOAc.
The organic layer was dried and evaporated, and the residue was
dissolved in dry CH2Cl2 (4 mL) and treated with L-phenylalanine
methyl ester hydrochloride (70 mg, 0.33 mmol). The solution was
cooled to 0 °C, and Et3N (45 μL, 33 mg, 0.33 mmol), EDC (69 mg, 0.36
mmol), and HOBt (49 mg, 0.36 mmol) were added. The reaction
mixture was stirred at 0 °C for 2 h, and then it was allowed to reach room
temperature, stirred for 18 h, and finally poured into a saturated aqueous
NaHCO3 solution and extracted with CH2Cl2. After usual drying and
solvent removal, the residue was purified by rotatory chromatography
(hexanes/EtOAc, 65:35), affording compounds 3 (86 mg, 51%) and 4
(52 mg, 31%).
Compound (3): amorphous solid; [α]D +26 (0.43, CHCl3); IR
(CHCl3) νmax 3439, 3420, 1741, 1652, 1505 cm−1; 1H NMR (500 MHz,
CDCl3) δH 0.81 (3H, d, J = 6.6 Hz, 5γ-Mea), 0.85 (3H, d, J = 6.6 Hz, 5γ-
Meb), 1.13 (3H, s, 6-Mea), 1.15 (3H, s, 6-Meb), 1.15−1.27 (2H, m, 5β-
H2), 1.48 (1H, m, 5γ-H), 1.50 (3H, d, J = 6.9 Hz, 3-Me), 3.07 (1H, dd, J
= 6.3, 13.9 Hz, 8β-Ha), 3.15 (1H, dd, J = 5.7, 13.9 Hz, 8β-Hb), 3.75 (3H,
s, OMe), 3.90 (1H, ddd, J = 2.8, 11.1, 11.3 Hz, 5-H), 4.63 (1H, dddd, J =
6.9, 6.9, 6.9, 6.9 Hz, 3-H), 4.78 (1H, ddd, J = 6.0, 6.3, 7.6 Hz, 8-H), 6.11
(1H, br d, J = 7.6 Hz, N7H [NHPhe]), 6.94 (1H, br d, J = 9.8 Hz, N4H
[NHLeu]), 7.06 (1H, br d, J = 6.6 Hz, N2H [NHAla]), 7.08 (2H, d, J = 6.8
Hz, Ar), 7.25−7.31 (3H, m, Ar), 7.41 (2H, dd, J = 6.6, 7.2 Hz, Ar), 7.49
(1H, dd, J = 7.3, 7.6 Hz, Ar), 7.80 (2H, d, J = 7.1 Hz, Ar); 13C NMR
(100.6 MHz, CDCl3) δC 19.2 (CH3), 21.4 (CH3), 23.2 (CH3), 23.8
(CH3), 24.4 (CH3), 25.0 (CH), 37.5 (CH2), 40.1 (CH2), 45.6 (C), 49.5
(CH), 52.4 (CH3), 52.8 (CH), 54.9 (CH), 127.1 (2 × CH), 127.3
(CH), 128.5 (2 × CH), 128.7 (2 × CH), 129.2 (2 × CH), 131.5 (CH),
134.2 (C), 135.8 (C), 166.9 (C), 171.9 (2 × C), 176.4 (C); MS (EI) m/
z (rel intensity) 509 (M+, 1), 361 (M+ − PhCONHCHMe, 15), 249
([Me2CCONHCH(CH2Ph)CO2Me + H]+, 39), 148 ([PhCONHCH-
(Me)]+, 32), 105 ([PhCO]+, 100), 86 ([NH2CHCH2CHMe2], 95);
HRMS (EI) calcd for C29H39N3O5 509.2890, found 509.2888; calcd for
C20H29N2O4 361.2127, found 361.2120; calcd for C14H19NO3 249.1365,
found 249.1356; calcd for C9H10NO 148.0762, found 148.0764; calcd
for C7H5O 105.0340, found 105.0342; calcd for C5H12N 86.0970, found
86.0967. Anal. Calcd for C29H39N3O5: C, 68.34; H, 7.71; N, 8.25.
Found: C, 68.65; H, 7.86; N, 8.22.
ASSOCIATED CONTENT
■
S
* Supporting Information
1H NMR, 13C NMR, and NOESY spectroscopic data of
compounds 1−6 and 9. This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: (J.M.A.) jesusmaria.aizpurua@ehu.es, (A.B.) alicia@
ipna.csic.es.
■
ACKNOWLEDGMENTS
We thank the Gobierno Vasco (Project SAIOTEK S-PR10BF02)
and Ministerio de Ciencia e Innovacion
07109) for financial support and SGIker UPV/EHU for NMR
facilities. C.J.S. thanks Gobierno de Canarias for a predoctoral
fellowship.
■
́
(Project CTQ2009-
REFERENCES
■
(1) (a) Gellman, S. H. Acc. Chem. Res. 1998, 31, 173−180.
(b) Gademann, K.; Hintermann, T.; Schreiber, J. V. Curr. Med. Chem.
1999, 6, 905−925. (c) Licini, G.; Prins, L. J.; Scrimini, P. Eur. J. Org.
Chem. 2005, 969−977. (d) Horne, W. S.; Gellman, S. H. Acc. Chem. Res.
2008, 41, 1399−1408. (e) Seebach, D.; Gardiner, J. Acc. Chem. Res.
2008, 41, 1366−1375. (f) Yashima, E.; Maeda, K.; Iida, H.; Furusho, Y.;
Nagai, K. Chem. Rev. 2009, 109, 6102−6211.
(2) (a) Choi, S. H.; Guzei, I. A.; Spencer, L. C.; Gellman, S. H. J. Am.
Chem. Soc. 2009, 131, 2917−2924. (b) Choi, S. H.; Guzei, I. A.;
Gellman, S. H. J. Am. Chem. Soc. 2007, 129, 13780−13781. (c) Price, J.
L.; Horne, W. S.; Gellman, S. H. J. Am. Chem. Soc. 2007, 129, 6376−
6377. (d) Horne, W. S.; Price, J. L.; Keck, J. L.; Gellman, S. H. J. Am.
Chem. Soc. 2007, 129, 4178−4180.
(3) Aguilar, M.-I.; Purcell, A. W.; Devi, R.; Lew, R.; Rossjohn, J.; Smith,
A. I.; Perlmutter, P. Org. Biomol. Chem. 2007, 5, 2884−2890.
(4) For recent papers, see: (a) Brans, L.; García-Garayoa, E.;
́
Schweinsberg, C.; Maes, V.; Struthers, H.; Schibli, R.; Tourwe, D.
ChemMedChem 2010, 5, 1717−1725. (b) Castelletto, V.; Hamley, I. W.;
Hule, R. H.; Pochan, D. Angew. Chem., Int. Ed. 2009, 48, 2317−2320.
(c) Lee, E. F.; Sadowsky, J. D.; Smith, B. J.; Czabotar, P. E.; Peterson-
Kaufman, K. J.; Colman, P. M.; Gellman, S. H.; Fairlie, W. D. Angew.
Chem., Int. Ed. 2009, 48, 4318−4322. (d) Keresztes, A.; Szucs, M.;
Borics, A.; Kover
́
, K. E.; Forro,
́
E.; Fulop, F.; Tomboly, C.; Pet
́
er, A.;
Compound (4): amorphous solid; [α]D +38 (0.17, CHCl3); IR
̈
̈
̈
̈
̈
1
(CHCl3) νmax 3442, 3362, 1731, 1652 cm−1; H NMR (500 MHz,
Pah
́
i, A.; Fab
́
ian
́
, A.; Muranyi, M.; Tot
́
́
h, G. J. Med. Chem. 2008, 51,
4270−4279. (e) Boersma, M. D.; Haase, H. S.; Peterson-Kaufman, K. J.;
Lee, E. F.; Clarke, O. B.; Colman, P. M.; Smith, B. J.; Horne, W. S.;
Fairlie, W. D.; Gellman, S. H. J. Am. Chem. Soc. 2012, 134, 315−323.
(5) Cheng, R.; Gellman, S. H.; DeGrado, W. F. Chem. Rev. 2001, 101,
3219−3232.
CDCl3) δH 0.88 (3H, d, J = 6.6 Hz, 5γ-Mea), 0.90 (3H, d, J = 6.6 Hz, 5γ-
Meb), 1.11 (3H, s, 6-Mea), 1.12 (3H, s, 6-Meb), 1.23 (1H, ddd, J = 3.8,
12.0, 15.4 Hz, 5β-Ha), 1.34 (1H, ddd, J = 2.8, 10.1, 14 Hz, 5β-Hb), 1.52
(3H, d, J = 6.9 Hz, 3-Me), 1.56 (1H, m, 5γ-H), 3.23 (2H, d, J = 6.8 Hz,
8β-H2), 3.78 (3H, s, OMe), 3.89 (1H, ddd, J = 2.8, 11.0, 11.7 Hz, 5-H),
4.46 (1H, dddd, J = 6.9, 6.9, 6.9, 6.9 Hz, 3-H), 4.78 (1H, ddd, J = 6.3, 6.6,
8.0 Hz, 8-H), 7.15 (1H, d, J = 8.2 Hz, N7H [NHPhe]), 7.24 (1H, br b,
N2H [NHAla]), 7.24−7.37 (8H, m, Ar + N4H [NHLeu]), 7.47 (1H, dd, J
= 7.3, 7.5 Hz, Ar), 7.76 (2H, d, J = 6.9 Hz, Ar); 13C NMR (125.7 MHz,
CDCl3) δC 17.6 (CH3), 21.3 (CH3), 22.6 (CH3), 23.9 (2 × CH3), 25.3
(CH), 36.6 (CH2), 38.8 (CH2), 46.5 (C), 50.8 (CH), 52.7 (CH3), 54.3
(CH), 54.5 (CH), 126.9 (CH), 127.1 (2 × CH), 128.5 (4 × CH), 129.2
(6) Seebach, D.; Abele, S.; Gademann, K.; Jaun, B. Angew. Chem., Int.
Ed. 1999, 38, 1595−1597.
(7) (a) Seebach, D.; Abele, S.; Sifferlen, T.; Hanggi, M.; Gruner, S.;
̈
Seiler, P. Helv. Chim. Acta 1998, 81, 2218−2243. (b) Abele, S.; Seiler, P.;
Seebach, D. Helv. Chim. Acta 1999, 82, 1559−1571.
(8) Rai, R.; Vasudev, P. G.; Ananda, K.; Raghothama, S.; Shamala, N.;
Karle, I. L.; Balaram, P. Chem.Eur. J. 2007, 11, 5917−5926.
5912
dx.doi.org/10.1021/jo300892u | J. Org. Chem. 2012, 77, 5907−5913

The Journal of Organic Chemistry
Featured Article
(9) Lijima, K.; Katada, J.; Yasuda, E.; Uno, I.; Hayashi, Y. J. Med. Chem.
1999, 42, 312−323.
(10) Peptides 1, 2 and 5−7 were previously described: (a) Saavedra,
́
C.; Hernandez, R.; Boto, A.; Alvarez, E. J. Org. Chem. 2009, 74, 4655−
4665. (b) For another recent contribution on structurally close hybrid
α,β,α-peptides, see: Cordero, F. M.; Salvati, M.; Vurchio, C.; de Meijere,
A.; Brandi, A. J. Org. Chem. 2009, 74, 4225−4231.
(11) Kessler, H. Angew. Chem., Int. Ed. Engl. 1982, 21, 512−523.
(12) Ludvigsen, S.; Andersen, K. V.; Poulsen, F. M. J. Mol. Biol. 1991,
217, 731−736.
(13) (a) Neuhaus, D.; Williamson, M. The Nuclear Overhauser Effect in
Structural and Conformational Analysis; VCH Publishers: New York,
1989. (b) Thomas, P. D.; Basus, V. J.; James, T. L. Proc. Nat. Acad. Sci.
U.S.A. 1991, 88, 1237−1241.
(14) NMR structural analyses conducted on α,β,α-tripeptides of the
type Cbz-Pro-(HN-CH(Bn)CMe2CO)-Leu-OMe were also attempted,
but the additional conformation constraint imposed by the N,N-
disubstituted proline residue resulted in multiple sets of signal in the 1H
NMR spectra and coalescence temperatures above 50 °C which
prevented the proper determination of the NH amide thermal
coefficients.
5913
dx.doi.org/10.1021/jo300892u | J. Org. Chem. 2012, 77, 5907−5913