Flat peptides

Article abstract of DOI:10.1021/ja9842114

We have synthesized by solution methods the first homopeptide series, pBrBz-(ΔAla)(n)-OMe (n = 1-6), based on a C(α,β)-didehydro-α-amino acid, to determine the preferred conformation of this residue, characterized by an sp² α-carbon atom and th

Full text of DOI:10.1021/ja9842114

3272
J. Am. Chem. Soc. 1999, 121, 3272-3278
Flat Peptides
Marco Crisma,† Fernando Formaggio,† Claudio Toniolo,*,† Taichi Yoshikawa,‡ and
Tateaki Wakamiya*,‡
Contribution from the Biopolymer Research Centre, CNR, Department of Organic Chemistry,
UniVersity of PadoVa, 35131 PadoVa, Italy, and Department of Chemistry, Faculty of Science and
Technology, Kinki UniVersity, Osaka 577, 8502 Japan
ReceiVed December 7, 1998
Abstract: We have synthesized by solution methods the first homopeptide series, pBrBz-(∆Ala)n-OMe (n )
1-6), based on a CR,â-didehydro-R-amino acid, to determine the preferred conformation of this residue,
characterized by an sp2 R-carbon atom and the smallest side chain. To this aim, we have exploited FTIR
1
absorption and H NMR techniques in solution and X-ray diffraction in the crystal state. Our investigation
shows that a multiple, consecutive, fully extended conformation (2.05-helix) largely predominates for all
oligomers in deuteriochloroform solution and occurs in the crystal state for the monomer, dimer, and trimer
as well. These peptide molecules are completely flat, including the amino acid side chains, and form planar
sheets. This novel peptide structure is stabilized by two types of intramolecular H-bonds, Ni-H‚‚‚OidC′i
(typical of the 2.05-helix) and Câi+1-H‚‚‚OidC′i (characteristic of ∆Ala peptides). The results obtained are
compared with those of the oligopeptides based on the related Câ-substituted, CR,â-didehydro-R-amino acid
residues.
Introduction
solution2,7 experimental investigations have been devoted to the
conformational preference of the simplest residue of this class,
∆Ala. In addition, they have been restricted to small compounds
such as linear derivatives (“monopeptides”) and dipeptides, the
latter, however, containing only a single ∆Ala residue [the only
exception is represented by the X-ray diffraction analysis of
the cyclic homodipeptide c(∆Ala)2,6a the conformation of which,
however, is strongly forced to be folded by the constraints
imposed by the small ring size]. All of these studies provide
evidence that the fully extended (C5) conform-
ation3b,8 is preferred by a single ∆Ala residue. A number of
theoretical analyses confirmed this conclusion for short (n <
6) ∆Ala homooligomers.2,9 However, recent conformational
energy calculations predicted that the 310-helix is the most stable
structure in longer homooligopeptides of this family.9a-c
In an attempt to contribute to solving this issue, we have
synthesized a terminally blocked, complete, monodispersed
∆Ala homooligomeric series, namely pBrBz-(∆Ala)n-OMe
(where pBrBz is p-bromobenzoyl and OMe is methoxy), from
monomer to hexamer (n ) 1-6) (Figure 1). We have carried
out a conformational analysis in solution using FTIR absorption
Identification of peptide backbones with new, well-defined,
regular secondary structural elements (helices, sheets, and turns)
is of outmost importance in the design of predetermined, simple
structural and functional motifs with potential applications in
biochemistry and materials science.1
Among noncoded R-amino acids, the class of CR,â-didehydro-
R-amino acids (∆AAs) is of particular interest. Both electronic
and steric factors play important roles in directing the confor-
mational properties of didehydropeptides.2 More specifically,
a variety of recent studies has unambiguously recognized the
strong tendency of the conformationally constrained, Câ-
substituted, γ-branched residues of this class ∆ZPhe and ∆Z-
Leu to stabilize â-turns3 in short model compounds and to
nucleate the 310-helical structure4 in longer peptides.2,5 On the
other hand, only few and nonsystematic crystal-state2,6 and
* Address correspondence to Prof. Claudio Toniolo, Department of
Organic Chemistry, University of Padova, Via Marzolo 1, 35131 Padova,
Italy. Tel.: +39-049-827-5247. Fax: +39-049-827-5239. E-mail: biop02@
chor.unipd.it.
†University of Padova.
‡Kinki University.
1
and H NMR on the complete series and a X-ray diffraction
(1) (a) Gellman, S. H. Acc. Chem. Res. 1998, 31, 173-180. (b) Toniolo,
C.; Yoshikawa, S. Biopolymers (Pept. Sci.) 1998, 47, 3-4.
(2) (a) Singh, T. P.; Narula, P.; Patel, H. C. Acta Crystallogr. 1998, B46,
539-545. (b) Singh, T. P.; Kaur, P. Prog. Biophys. Mol. Biol. 1996, 66,
141-165. (c) Jain, R.; Chauhan, V. S. Biopolymers (Pept. Sci.) 1996, 40,
105-119.
(3) (a) Venkatachalam, C. M. Biopolymers 1968, 6, 1425-1436. (b)
Toniolo, C. CRC Crit. ReV. Biochem. 1980, 9, 1-44. (c) Rose, G. D.;
Gierasch, L. M.; Smith, J. A. AdV. Protein Chem. 1985, 37, 1-109.
(4) Toniolo, C.; Benedetti, E. Trends Biochem. Sci. 1991, 16, 350-353.
(5) Ramagopal, U. A.; Ramakumar, S.; Joshi, R. M.; Chauhan, V. S. J.
Pept. Res. 1998, 52, 208-215.
(6) (a) Ongania, K. H.; Granozzi, G.; Busetti, V.; Casarin, M.; Ajo`, D.
Tetrahedron 1985, 41, 2015-2018. (b) Pietrzynski, G.; Rzeszotarska, B.;
Ciszak, E.; Lisowski, M.; Kubica, Z.; Boussard, G. Int. J. Pept. Protein
Res. 1996, 48, 347-356. (c) Lombardi, A.; D’Agostino, B.; Nastri, F.;
D’Andrea, L. D.; Filippelli, A.; Falciani, M.; Rossi, F.; Pavone, V. Bioorg.
Med. Chem. Lett. 1998, 8, 1153-1156.
(7) Broda, M. A.; Rzeszotarska, B.; Smelka, L.; Rospenk, M. J. Pept.
Res. 1997, 50, 342-351.
(8) Toniolo, C.; Benedetti, E. In Molecular Conformation and Biological
Interactions: G. N. Ramachandran Festschrift; Balaram, P., Ramaseshan,
S., Eds.; Indian Institute of Science: Bangalore, India, 1991; pp 511-521.
(9) (a) Aleman, C. Biopolymers 1993, 33, 1811-1817. (b) Aleman, C.
Biopolymers 1994, 34, 841-847. (c) Fabian, P.; Chauhan, V. S.; Pongor,
S. Biochim. Biophys. Acta 1994, 1208, 89-93. (d) Jezowska-Bojczuk, M.;
Varnagy, K.; Sovago, I.; Pietrzynski, G.; Dyba, M.; Kubica, Z.; Rzeszo-
tarska, B.; Smelka, L.; Kozlowski, H. J. Chem. Soc., Dalton Trans. 1996,
3265-3268. (e) Cativiela, C.; Garcia, J. I.; Mayoral, J. A.; Salvatella, L. J.
Mol. Struct. (THEOCHEM) 1996, 368, 57-66. (f) Rao, G. S.; Siddiqui,
M. I.; Arora, S. J. Biomol. Struct. Dyn. 1996, 14, 21-24. (g) Rao, G. S.;
Arora, S.; Kataria, S. J. Biomol. Struct. Dyn. 1998, 15, 1053-1058. (h)
Thormann, M.; Hofmann, H. J. J. Mol. Struct. (THEOCHEM) 1998, 431,
79-96. (i) Broda, M. A.; Rzeszotarska, B. Lett. Pept. Sci. 1998, 5, 441-
443.
10.1021/ja9842114 CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/30/1999

Flat Peptides
J. Am. Chem. Soc., Vol. 121, No. 14, 1999 3273
Table 1. Physical and Analytical Properties of the A2pr(Z) and ∆Ala Homopeptides
compd
yield (%)
melting point (°C)
recryst solvent b
[R]25D (deg)c
FAB-MS (m/z) e
Boc-[A2pr(Z)]2-OMe
Boc-[A2pr(Z)]3-OMe
Boc-[A2pr(Z)]4-OMe
Boc-[A2pr(Z)]5-OMe
Boc-[A2pr(Z)]6-OMe
pBrBz-A2pr(Z)-OMe
pBrBz-[A2pr(Z)]2-OMe
pBrBz-[A2pr(Z)]3-OMe
pBrBz-[A2pr(Z)]4-OMe
pBrBz-[A2pr(Z)]5-OMe
pBrBz-[A2pr(Z)]6-OMe
pBrBz-∆Ala-OMe
pBrBz-(∆Ala)2-OMe
pBrBz-(∆Ala)3-OMe
pBrBz-(∆Ala)4-OMe
pBrBz-(∆Ala)5-OMe
pBrBz-(∆Ala)6-OMe
83
84
86
99
99
92
81
93
88
97
90
74
65
62
58
30
22
119-122
153-155
193-196
218-220a
236-239a
113-116
195-197
213-215
234-242a
250-259a
255-264a
94-96
EtOAc/hexane
CHCl3/hexane
CHCl3/hexane
DMF/H2O
DMF/H2O
EtOAc/Et2O/hexane
DMF/H2O
DMF/H2O
DMF/H2O
DMF/H2O
DMF/H2O
EtOAc/hexane
EtOAc/hexane
CHCl3/hexane
CHCl3/hexane
CHCl3/hexane
DMF/H2O
-0.6
-18.0
-23.1d
-25.8d
-20.6d
+13.4
-24.3d
-34.9d
-38.2d
-38.1d
-35.6d
-
573f/573.3g
793/793.3
1013/1013.4
1234/1233.5
1454/1453.6
435/435.1
655/655.1
875/875.1
1095/1095.2
1315/1315.3
1535/1535.4
284/284.2
353/353.2
422/422.2
491/491.1
560/560.1
629/629.1
119-122
-
>210a
-
>210a
-
>210a
-
>210a
-
a With decomposition. b EtOAc, ethyl acetate; DMF, N,N′-dimethylformamide; Et2O, diethyl ether. c c 1.0, CHCl3. d c 1.0, DMF. e (M + H)+.
f Found. g Calculated.
The physical and analytical properties of the newly synthesized A2-
pr(Z) (A2pr, R,â-diaminopropionic acid; Z, benzyloxycarbonyl) and
∆Ala homopeptides are listed in Table 1. All peptides were also
chemically homogeneous by 1H NMR spectrometry. Typical procedures
were the following.
pBrBz-A2pr(Z)-OMe. To a solution of HCl‚H-A2pr(Z)-OMe (1.20
g, 4.15 mmol) in N,N’-dimethylformamide (DMF) (5 mL) were added
pBrBz-Cl (0.913 g, 4.15 mmol) and triethylamine (TEA) (1.27 mL,
9.13 mmol) in 10 equal and alternate portions over 100 min at 0 °C.
The mixture was stirred for 2 h at 0 °C and then overnight at room
Figure 1. (A) ∆ZPhe and ∆ZLeu residues. (B) pBrBz-(∆Ala)n-OMe
(n ) 1-6) homopeptides investigated in this work.
temperature. The precipitated TEA‚HCl was filtered off, and the filtrate
was concentrated in vacuo. The residue was triturated with water, and
the resulting crystalline product was collected by filtration, followed
by recrystallization from ethyl acetate (EtOAc)-diethyl ether (Et2O)-
hexane.
investigation on the crystalline monomer, dimer, and trimer.
The data obtained strongly support the view that the fully
extended, 2.05-helix is the conformation preferred by the ∆Ala
homopeptides to the hexamer level. A preliminary account of
the results of this work has been reported.10
It is also worth mentioning that these findings may have
additional implications as ∆AA residues are frequently found
in naturally occurring peptides of microbial and fungal metabo-
lite origins,11,12 and they are also constituents of a few proteins.
In particular, it has been shown that ∆Ala is posttranslationally
formed from Ser in the active site of the enzyme histidine
ammonia-lyase,13 and segments of consecutive 1-4 ∆Ala
residues are found in a number of thiopeptide antibiotics.12
pBrBz-∆Ala-OMe. To a solution of pBrBz-A2pr(Z)-OMe (0.750
g, 1.72 mmol) in acetic acid (AcOH) (2 mL) was added 25% HBr/
AcOH (5 mL) at room temperature. After being stirred for 2 h, the
reaction mixture was concentrated in vacuo. The residue was crystallized
by trituration with Et2O, and the pBrBz-A2pr-OMe hydrobromide was
used for the subsequent reaction without further purification. To a
solution of the above-mentioned hydrobromide (0.640 g, 1.67 mmol)
in a methanol (MeOH)/DMF (1:3) mixture (64 mL) were added CH3I
(4.0 mL, 64.2 mmol) and KHCO3 (3.21 g, 32.1 mmol). After the
reaction mixture was stirred for 6 h at room temperature, EtOAc (200
mL) was added. The precipitated insoluble material was filtered off,
and the filtrate was concentrated in vacuo. EtOAc (100 mL) and water
(20 mL) were added to the residue. The EtOAc layer was washed with
water (20 mL × 3), 10% aqueous citric acid (20 mL × 3), saturated
aqueous NaHCO3 (20 mL × 3), and brine (20 mL × 3). The organic
layer was dried over anhydrous MgSO4 and concentrated in vacuo
below 25 °C. The residue was crystallized by trituration with hexane.
The crystalline product was collected by filtration and recrystallized
from EtOAc/hexane.
Boc-[A2pr(Z)]2-OMe. To a stirred solution of Boc-A2pr(Z)-OSu
(Boc, tert-butyloxycarbonyl; OSu, N-hydroxysuccinimido)14 (1.50 g,
3.44 mmol) and HCl‚H-A2pr(Z)-OMe (0.994 g, 3.27 mmol) in DMF
(15 mL) was added TEA (0.474 mL, 3.44 mmol) at 0 °C. After the
solution was stirred for 1 h at 0 °C and overnight at room temperature,
the precipitated TEA‚HCl was filtered off, and the filtrate was
concentrated in vacuo. The residue was dissolved in CHCl3 (10 mL)
and washed with 10% aqueous citric acid (5 mL × 3), brine (5 mL ×
3), saturated aqueous NaHCO3 (5 mL × 3), and brine (5 mL × 3).
The organic layer was dried over anhydrous Na2SO4 and concentrated
in vacuo. The residue was crystallized by trituration with hexane. The
crystalline product was collected by filtration and recrystallized from
EtOAc/hexane.
Materials and Methods
Synthesis and Characterization of Peptides. Melting points were
determined with a Yanaco model MP-J3 (Kyoto, Japan) apparatus and
are uncorrected. Silica gel column chromatography was carried out with
Merck (Darmstadt, Germany) silica gel 60 (70-230 mesh). Analytical
1H NMR spectra were recorded on a Jeol model JNM-EX (Tokyo,
Japan) 270-MHz or a Bruker model AM 400 (Karlsruhe, Germany)
spectrometer. Peptide mass numbers were determined by fast-atom
bombardment mass spectrometry (FAB-MS) using a Jeol model JMS-
HX 100 spectrometer. Specific optical rotations were measured with a
Horiba model SEPA-200 polarimeter (Kyoto, Japan).
(10) Toniolo, C.; Formaggio, F.; Crisma, M.; Yoshikawa, T.; Wakamiya,
T. In Proceedings of the First International Peptide Symposium; Shimonishi,
Y., Ed.; Kluwer: Dordrecht, The Netherlands, 1999 (in press).
(11) (a) Stammer, C. H. In Chemistry and Biochemistry of Amino Acids,
Peptides and Proteins, Vol. 6; Weinstein, B., Ed.; Dekker: New York,
1982; pp 33-74. (b) Schmidt, U.; Lieberknecht, A.; Wild, J. Synthesis 1988,
159-172.
(12) Fate, G. D.; Benner, C. P.; Grode, S. H.; Gilbertson, T. J. J. Am.
Chem. Soc. 1996, 118, 11363-11368.
(13) Langer, M.; Lieber, A.; Re´tey, J. Biochemistry 1994, 33, 14034-
14038.
(14) Teshima, T.; Nomoto, S.; Wakamiya, T.; Shiba, T. Bull. Chem. Soc.
Jpn. 1977, 50, 3372-3380.

3274 J. Am. Chem. Soc., Vol. 121, No. 14, 1999
Crisma et al.
Table 2. Crystallographic Data and Diffraction Parameters for the pBrBz-(∆Ala)n-OMe (n ) 1-3) Homopeptides
parameter
monopeptide
dipeptide
tripeptide
empirical formula
formula weight (amu)
color, habit
crystal system
space group
a (Å)
C11H10NO3Br
284.1
colorless, plates
triclinic
P1h
8.899(1)
13.709(2)
4.895(1)
96.2(1)
89.9(1)
107.2(1)
566.8(2)
C14H13N2O4Br
353.2
colorless, plates
triclinic
P1h
9.937(2)
11.187(2)
6.743(1)
102.3(1)
94.2(1)
93.7(1)
727.9(2)
C17H16N3O5Br
422.2
colorless, plates
monoclinic
P21/c
9.885(1)
6.779(1)
26.339(2)
90.0
98.3(1)
90.0
1746.5(3)
4
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
Z (molecules/unit cell)
density (calcd) (g/cm3)
absorption coefficient (mm-1
F(000)
2
2
1.665
3.616
284
1.611
2.840
356
1.606
2.388
856
)
collected reflections
independent reflections
observed reflections [I g 2σ(I)]
solved by
2752
3525
4316
2749 [R(int) ) 0.0164]
1170
SHELXS 86
3513 [R(int) ) 0.0188]
1190
SHELXS 86
4207 [R(int) ) 0.0291]
2071
SHELXS 86
refined by
SHELXL 93
SHELXL 93
SHELXL 93
final R indices [I g 2σ(I)]
final R indices (all data)
temperature (K)
radiation (λ)
scan method
θ range (deg)
index ranges
R1 ) 0.0562, wR2 ) 0.1383
R1 ) 0.1596, wR2 ) 0.1838
293(2)
R1 ) 0.0668, wR2 ) 0.1620
R1 ) 0.2254, wR2 ) 0.2386
293(2)
R1 ) 0.0432, wR2 ) 0.1069
R1 ) 0.1214, wR2 ) 0.1390
293(2)
Mo KR (0.71073 Å)
θ-2θ
2-28
Mo KR (0.71073 Å)
θ-2θ
2-28
Mo KR (0.71073 Å)
θ-2θ
2-28
-11 e h e 11, -18 e k e 17,
0 e l e 6
-13 e h e 13, -14 e k e 14,
-1 e l e 8
-13 e h e 12, 0 e k e 8,
0 e l e 34
refinement method
full-matrix least-squares on F2
2749/0/133
full-matrix least-squares on F2
3510/0/179
full-matrix least-squares on F2
4207/0/223
data/restraints/parameters
goodness of fit on F2
crystallization solvent
crystal size (mm)
0.940
acetone
0.854
acetone
0.869
acetone
0.2 × 0.2 × 0.1
0.4 × 0.4 × 0.2
0.4 × 0.2 × 0.1
∆Fmax and ∆Fmin (e‚Å)
0.478/-0.760
0.673/-0.698
0.530/-0.416
HCl‚H-[A2pr(Z)]2-OMe. To a solution of Boc-[A2pr(Z)]2-OMe (1.53
g, 2.67 mmol) in EtOAc (25 mL) was added 4 M HCl/EtOAc (20 mL)
at room temperature. After being stirred for 1 h, the mixture was
concentrated in vacuo, and the residue was crystallized by trituration
with Et2O. The resulting ester hydrochloride was used for the subsequent
reaction without further purification.
times the Ueq of the attached atoms. Crystal data and diffraction
parameters are listed in Table 2.
Results and Discussion
Peptide Synthesis. Since ∆AAs are relatively unstable
compounds, their incorporation at an early stage in peptide
synthesis may limit later reactions.11 Therefore, the methodology
we report here allows conversion to the ∆AAs on completion
of peptide synthesis. Saturated R-amino acids such as Ser, Cys,
and A2pr are commonly exploited as precursors of ∆Ala
residues.11
In the synthesis of our (∆Ala)n homopeptides, we took
advantage of Shiba’s procedure16a based on the Hofmann
degradation of A2pr as shown below:
FTIR Absorption. The FTIR absorption spectra were recorded using
a Perkin-Elmer model 1720 X (Norwalk, CT) FTIR spectrophotometer,
nitrogen flushed, equipped with a sample-shuttle device, at 2 cm-1
nominal resolution, averaging 100 scans. Solvent (baseline) spectra were
obtained under the same conditions.Cells with path lengths of 0.1, 1.0,
and 10 mm (with CaF2 windows) were used. Spectrograde deuterio-
chloroform (99.8% D) was purchased from Fluka (Buchs, Switzerland).
1
NMR Spectroscopy. The H NMR spectra were recorded with a
Bruker model AM 400 spectrometer. Measurements were carried out
in deuteriochloroform (99.96% D; Acros, Geel, Belgium) and deuterated
dimethyl sulfoxide (99.96% D6; Stohler, Waltham, MA) with tetra-
methylsilane as the internal standard. The free radical 2,2,6,6-
tetramethylpiperidinyl-1-oxy (TEMPO) was purchased from Sigma (St.
Louis, MO).
X-ray Diffraction. Data collection was performed using a Philips
PW 1100 (Eindhoven, The Netherlands) four-circle diffractometer. Unit
cell determination was carried out by least-squares refinement of the
setting angles of 25 high-angle reflections accurately centered using
Mo KR radiation. All of the structures were solved by direct methods,
using the SHELXS 8615a program. Refinement was carried out on F2,
with all non-H atoms anisotropic, by application of the SHELXL 9315b
program. The H atoms of the three peptides were calculated at idealized
positions, and during the refinement they were allowed to ride on their
carrying atom, with Uiso set equal to 1.2 (or 1.5 for the methyl groups)
The Z-protecting group for the â-amino function of A2pr was
removed from the intermediate A2pr(Z) homopeptide with 25%
HBr/AcOH16b prior to the Hofmann degradation. The resulting
free side-chain amino group was then quaternized according to
the Chen-Benoiton method (CH3I/KHCO3).16c The intermediate
Nâ-trimethyl-A2pr peptides were not isolated during the synthetic
procedure.
(16) (a) Nomoto, S.; Sano, A.; Shiba, T. Tetrahedron Lett. 1979, 521-
522. (b) Ben-Ishai, D.; Berger, A. J. Org. Chem. 1982, 17, 1564-1570.
(c) Chen, F. C. M.; Benoiton, N. L. Can. J. Chem. 1976, 54, 3310-3311.
(d) Shin, C.; Okumura, K.; Ito, A.; Nakamura, Y. Chem. Lett. 1994, 1301-
1304. (e) Okumura, K.; Ito, A.; Nakamura, Y.; Shin, C. Bull. Chem. Soc.
Jpn. 1996, 69, 2309-2316.
(15) (a) Sheldrick, G. M. In SHELXS 86. Program for the Solution of
Crystal Structures; University of Go¨ttingen: Go¨ttingen, Germany, 1990.
(b) Sheldrick, G. M. In SHELXL 93. Program for Crystal Structure
Refinement; University of Go¨ttingen: Go¨ttingen, Germany, 1993.

Flat Peptides
J. Am. Chem. Soc., Vol. 121, No. 14, 1999 3275
Chart 1
Figure 2. FTIR absorption spectra in the N-H stretching region for
the pBrBz-(∆Ala)n-OMe (n ) 1-6) homopeptides in CDCl3 solution.
Peptide concentration, 1 mM.
Table 3. Chemical Shifts Values for the Amide/Peptide NH Proton
Resonances of the pBrBz-(∆Ala)n-OMe (n ) 1-6) Homopeptides
(1 mM Concentration in CDCl3 Solution)
NH protons
N-terminal
residue
C-terminal
residue
n
internal residues
1
2
3
4
5
6
8.48
8.58
8.56
8.56
8.56
8.56
8.85
8.85
8.86
8.86
8.86
8.96
8.95
8.95
8.95
8.97
8.95
8.95
8.97
8.95
8.97
demonstrate that, at 1 mM concentration, intermolecular H-
bonding is negligible for all oligomers (not shown). At 1 mM
concentration or below, a very weak band is seen at about 3450
cm-1 (free, solvated NH groups), followed for all oligomers
by a broad and intense absorption centered in the 3405-3380-
cm-1 region (weakly intramolecularly H-bonded NH groups).
This latter spectral region is characteristic of the fully extended
(C5) peptide conformation.17b,c As the peptide main-chain length
is enhanced, (i) the frequency of the absorption maximum of
the 3400-cm-1 band tends to decrease slightly and (ii) the
intensity of this band, relative to that of the 3450-cm-1 band,
tends to increase regularly. The effect of heating to 55 °C does
not alter the overall pattern (not shown).
The present FTIR absorption investigation has provided
convincing evidence that intramolecular H-bonding of the C5-
type is the most important factor influencing the thermally stable
conformation of (∆Ala)n homooligomers in CDCl3 solution.
To get more detailed information on the preferred conforma-
tion of the (∆Ala)n homooligomers in CDCl3 solution, we
carried out a 400-MHz 1H NMR investigation. The delineation
of inaccessible (or intramolecularly H-bonded) amide/peptide
NH groups was performed by using (i) solvent dependence of
NH chemical shifts by adding increasing amounts of the
H-bonding acceptor solvent DMSO18a to the CDCl3 solution,
(ii) free-radical TEMPO-induced line broadening of NH
resonances,18b and (iii) temperature dependence of NH chemical
shifts in CDCl3 solution.18c
The preparation of the A2pr(Z) homopeptides was achieved
by using the C-activated Boc-A2pr(Z)-OSu derivative.14 Re-
moval of the NR-protecting Boc group was obtained by mild
acidolysis. Blocking of the NR-amino function with the pBrBz
moiety was achieved with pBrBz-Cl in the presence of a base
(TEA). The pBrBz group is useful in that it contains a heavy
atom (Br) which helps crystallographers in solving the phase
problem during the X-ray diffraction analysis. The final products
were characterized by melting point determination, polarimetry,
1H NMR, and mass spectrometry. The synthetic procedures are
schematically outlined in Chart 1. An alternative synthesis of a
stretch of four consecutive ∆Ala residues by repetition of
stepwise elongation of a Ser derivative and subsequent â-elim-
ination has been reported by Shin and co-workers in connection
with the synthesis of thiopeptide antibiotics.16d,e
Solution Conformational Analysis. The preferred conforma-
tion adopted by the terminally blocked (∆Ala)n (n ) 1-6)
homopeptide series was determined in a solvent of low polarity
1
(CDCl3) by FTIR absorption and H NMR techniques.
Figure 2 illustrates the FTIR absorption spectra in the
conformationally informative N-H stretching region. Using
Mizushima’s dilution technique,17a we have been able to
A complete assignment has been achieved for the proton
resonances of the monomer, dimer, and trimer by means of
homonuclear decoupling and NOE experiments. A partial
(17) (a) Mizushima, S.; Shimanouchi, T.; Tsuboi, M.; Souda, P. J. Am.
Chem. Soc. 1952, 74, 270-271. (b) Cung, M. T.; Marraud, M.; Ne´el, J.
Ann. Chim. 1972, 7, 183-209. (c) Toniolo, C.; Pantano, M.; Formaggio,
F.; Crisma, M.; Bonora, G. M.; Aubry, A.; Bayeul, D.; Dautant, A.; Boesten,
W. H. J.; Schoemaker, H. E.; Kamphuis, J. Int. J. Biol. Macromol. 1994,
16, 7-14.
(18) (a) Kopple, K. D.; Ohnishi, M. Biochemistry 1969, 8, 4087-4095.
(b) Kopple, K. D.; Schamper, T. J. J. Am. Chem. Soc. 1972, 94, 3644-
3646. (c) Bonora, G. M.; Mapelli, C.; Toniolo, C.; Wilkening, R. R.; Stevens,
E. S. Int. J. Biol. Macromol. 1984, 6, 179-188.

3276 J. Am. Chem. Soc., Vol. 121, No. 14, 1999
Crisma et al.
Table 4. Chemical Shifts Values for the Vinyl CâH Proton Resonances of the pBrBz-(∆Ala)n-OMe (n ) 1-6) Homopeptides (1 mM
Concentration in CDCl3 Solution)
H-bonded vinyl CH protons
other residues
free vinyl CH protons
other residues
n
N-terminal residue
C-terminal residue
1
2
3
4
5
6
6.80
6.79
6.80
6.80
6.80
6.80
6.01
6.03
6.04
6.04
6.04
6.04
6.65
6.65
6.65
6.65
6.65
5.50
5.51
5.52
5.52
5.53
6.66
6.65
6.65
6.66
5.55
5.56
5.56
5.55
6.66
6.66
6.66
5.64
5.56
5.56
6.67
6.67
5.57
5.57
6.67
5.57
fields (8.95-8.97 ppm). On the other hand, there are two groups
(50% each) of vinyl CâH protons (Table 4): (i) deshielded
resonances (6.65-6.80 ppm), which we assign to intramolecu-
larly H-bonded CâH protons, and (ii) shielded resonances (5.50-
6.04 ppm), which we assign to free CâH protons. Within the
former group, the resonance corresponding to the N-terminal
residue is that found at lowest fields, whereas within the latter
group, the resonance corresponding to the C-terminal residue
is that found at lowest fields.
Furthermore, from the data illustrated in Figure 3, it is evident
that none of the proton NH chemical shifts is markedly sensitive
to the addition of DMSO and heating, nor do their resonances
broaden significantly upon addition of TEMPO. This behavior
is characteristic of intramolecularly H-bonded NH protons.
Taken together, these 1H NMR results fit nicely with the FTIR
absorption data discussed above, in that a multiple, consecutive
C5 conformation seems to largely prevail for all of the (∆Ala)n
homooligomers in CDCl3 solution. In this conformation, all of
the NH groups are intramolecularly H-bonded, whereas inter-
molecular H-bonds do not play a significant role, even at high
concentrations. These fully extended structures appear to be
further stabilized by intramolecular H-bonds involving one vinyl
CâH proton from each residue as donor.
Crystal-State Conformational Analysis. The molecular
structures of the pBrBz-(∆Ala)n-OMe (n ) 1-3) homopeptides
with the atomic numbering schemes are illustrated in Figure 4.
Average bond distances and bond angles characterizing the ∆Ala
residue are shown in Figure 5A,B. Relevant backbone torsion
angles19 are presented in Table 5. In Table 6, the inter- and
intramolecular H-bond parameters are listed.
Interestingly, in peptides from protein amino acids, where
the R-carbon is sp3 hybridized, the CR-N (1.45 Å) and CR-C′
(1.52 Å) bond lengths20 are significantly longer than those in
∆Ala peptides, as expected, since in the latter compounds a
resonance effect with the CRdCâ bond is operative. Furthermore,
the bond angle at the nitrogen is much wider than the
corresponding parameter in normal peptides,20 and the confor-
mationally informative τ (N-CR-C′) bond angle20 is greatly
narrowed compared to that typical of an sp2-hybridized atom
(120°). This latter finding may be considered a preliminary
indication of the onset of the fully extended (C5) structure3b,8
for the ∆Ala homooligomers in the crystal state.
1
Figure 3. (a) Plot of NH chemical shifts in the H NMR spectra of
pBrBz-(∆Ala)6-OMe as a function of increasing percentages of DMSO
added to the CDCl3 solution (v/v). (b) Plot of the bandwidths of the
NH protons of the same homopeptide as a function of increasing
percentages of the free radical TEMPO (w/v) in CDCl3 solution. (c)
Plot of temperature coefficients for the amide NH protons of the same
homopeptide in CDCl3 solution. Peptide concentration, 1 mM.
tentative assignment of the NH and vinyl CâH proton resonances
of the N- and C-terminal ∆Ala residues of the longer oligomers
has been performed by analogy with the corresponding values
for the shorter oligomers. From an analysis of the spectra of
the more soluble oligomers (from monomer to tetramer) in the
10-1 mM concentration range (results not shown), we have
been able to conclude that dilution does not induce any
significant shift of any of the proton resonances.
Tables 3 and 4 list the chemical shifts of the amide/peptide
NH and vinyl CâH protons, respectively, for all oligomers in
CDCl3 solution, while Figure 3 shows the effects of the added
perturbing agents DMSO and TEMPO and heating on the NH
resonances of the hexamer, selected as a typical example. An
inspection of the values in Table 3 clearly indicates that all of
the NH protons are at low fields compared to usual peptide
NH protons and are of three different types: (i) the least
deshielded peptide NH proton (8.48-8.58 ppm) of the C-
terminal residue at highest fields, (ii) the amide NH proton of
the N-terminal residue at intermediate fields (8.85-8.86 ppm),
The three peptides examined form single (monomer) or
multiple and consecutive (dimer and trimer) fully extended (C5)
conformations. The resulting helical structure (2.05-helix) has
been experimentally verified so far only in the terminally
blocked (Deg)2-5 and (Dpg)2 (Deg, CR,R-diethylglycine; Dpg,
C
R,R-di-n-propylglycine) homooligomers.8 Thus, our (∆Ala)n
dimer and trimer are the first 2.05-helical peptides not based on
(19) IUPAC-IUB Commission on Biochemical Nomenclature. J. Mol.
Biol. 1970, 52, 1-17.
(20) Benedetti, E. In Chemistry and Biochemistry of Amino Acids,
Peptides and Proteins, Vol. 6; Weinstein, B., Ed.; Dekker: New York,
1983; pp 105-184.
and (iii) a clustering of the most deshielded peptide NH protons
of all of the internal residues in a very narrow range at lowest

Flat Peptides
J. Am. Chem. Soc., Vol. 121, No. 14, 1999 3277
Table 5. Selected Backbone Torsion Anglesa (deg) for the
pBrBz-(∆Ala)n-OMe (n ) 1-3) Homopeptides
torsion angle
monopeptide
dipeptide
tripeptide
ω0
φ1
ψ1
ω1
φ2
ψ2
ω2
φ3
ψ3
ω3
-178.0(5)
156.2(5)
175.7(6)
179.5(7)
170.3(3)
-178.2(3)
-172.3(3)
179.9(3)
-178.5(3)
176.8(3)
173.1(5)b
-179.5(5)c
-170.8(6)
-169.7(7)
170.3(7)
178.5(6)b
177.9(6)c
172.6(3)
-175.8(3)
-179.9(3)b
-178.8(3)c
a The torsion angles for rotation about bonds of the peptide backbone
(φ, ψ, ω) are described in ref 19. b N-CR-C′-OT torsion angle. c CR-
C′-OT-CT torsion angle.
in ∆Ala peptides, sp3 in Deg and Dpg peptides). The parameters
characterizing the Ni-H‚‚‚OidCi′ intramolecular H-bond shown
in Figure 5C should be compared to the corresponding ones in
Deg and Dpg peptides: N‚‚‚O and H‚‚‚O distances 2.54 and
2.00 Å, respectively, and N-H‚‚‚O angle 110.6°.8 Overall, it
seems that the pentagonal ring of the C5 structure of ∆Ala
peptides would be somewhat more expanded than the analogous
C5 ring structure of Deg and Dpg peptides.
Figure 4. X-ray diffraction structures of (A) pBrBz-∆Ala-OMe, (B)
pBrBz-(∆Ala)2-OMe, and (C) pBrBz-(∆Ala)3-OMe with numbering of
the atoms. The N-H‚‚‚OdC′, C′dO‚‚‚H-Câ, and Câ-H‚‚‚OT in-
tramolecular H-bonds are represented by dashed lines.
The æ,ψ,ω backbone torsion angles are all very close to the
trans disposition (180° ( 10°), as expected for a fully extended
(C5) conformation.3b,8 The only significant exception is repre-
sented by the æ1 torsion angle of the monomer, where the C5
structure seems to be slightly distorted. In the fully extended
structure of the ∆Ala dimer and trimer, the CR ‚‚‚CRi+1 distance
i
is 3.77 Å, while in the antiparallel pleated-sheet â-structure,
310-helix, and R-helix, this same distance is 3.47, 1.94, and 1.56
Å, respectively.4
The structure of these peptides is further stabilized by Câ
-
i+1
H‚‚‚OidC′i intramolecular H-bonds.21a The average parameters
characterizing this hexagonal (C6) ring structure are given in
Figure 5D. An additional H-bond is observed between the Câ-
H2 group of the C-terminal ∆Ala residue and the OT oxygen
atom of the methyl ester group. These findings closely parallel
the 1H NMR observations about the chemical shifts of the CâH
protons in solution.
The most striking feature in the packing mode of these ∆Ala
homopeptides is their arrangement in planar, parallel layers, with
an approximate interlayer separation of 3.30 Å (for the trimer,
see Figure 6A). There are no N-H‚‚‚OdC′ intermolecular
H-bonds, with the single exception of the monomer, in which
the slight warping of the molecule allows the N1-H group to
approach the C′0dO0 group of a symmetry-related molecule
in an adjacent layer, to form a H-bond of normal strength.21b
To our knowledge, this unusual lack of N-H‚‚‚OdC′ inter-
molecular H-bonds is shared only by the 2.05-helix forming Deg
and Dpg homopeptides.8 A significant contribution to the
stability of the planar arrangement of the molecules might be
ascribed to the intermolecular (aryl) C-H‚‚‚OdC′ short contacts
listed in Table 6 and illustrated for the tripeptide in Figure 6B.
It has also to be mentioned that, in all of the three structures,
the distances between the Br atoms of centrosymmetric pairs
of molecules (within the same layer) are in the range 3.556-
(3)-3.681(1) Å, in which dispersion effects between the
bromines may be operative.
Figure 5. Average bond distances (A) and bond angles (B) for the
∆Ala residue and average parameters for the Ni-H‚‚‚OidC′i intramo-
lecularly H-bonded C5 conformation (C), further stabilized by intramo-
lecular Câi+1-H‚‚‚OidC′i H-bonds (D), of the ∆Ala homopeptides.
CR-tetrasubstituted R-amino acids. An additional interesting
observation is that the ∆Ala homooligomers are flat molecules,
including the amino acid side chains. The dihedral angle between
adjacent C5 rings is in the range 3.5-8.0°. This property is
obviously not shared by the Deg and Dpg homooligomers. The
average parameters for the C5 conformation in ∆Ala peptides
are given in Figure 5C. Interestingly, the τ bond angle is greatly
expanded compared to the corresponding bond angle for the
C5 conformation of Deg and Dpg homooligomers (102-103°).8
This difference may be explained, at least in part, on the basis
of the different type of hybridization for the R-carbon atom (sp2
(21) (a) Fabiola, G. F.; Krishnaswamy, S.; Nagarajan, V.; Pattabhi, V.
Acta Crystallogr. 1997, D53, 316-320. (b) Go¨rbitz, C. H. Acta Crystallogr.
1989, B45, 390-395.

3278 J. Am. Chem. Soc., Vol. 121, No. 14, 1999
Crisma et al.
Table 6. Intra- and Intermolecular H-Bond Parameters for the pBrBz-(∆Ala)n-OMe (n ) 1-3) Homopeptides
peptide
donor D-H acceptor A
symmetry operation
distance (Å) D ‚‚‚A distance (Å) H‚‚‚A angle (deg) D-H‚‚‚A
pBrBz-∆Ala-OMe
N1-H
O1
O0
OT
O0
O1
O0
O1
O2
O0
O1
OT
O0
O1
O2
O3
O0
O1
O2
OT
O0
x, y, z
x, y, z
x, y, z
x, y, 1 + z
-x, 2 - y, 1 - z
1 - x, 2 - y, -1 - z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, 1 + z
x, y, z
x, y, z
x, y, z
x, y, z
x, y, z
2.687(7)
2.869(7)
2.730(8)
3.005(6)
3.386(6)
3.386(7)
2.599(8)
2.624(8)
2.855(11)
2.910(10)
2.737(10)
3.147(8)
2.596(4)
2.583(4)
2.624(4)
2.850(5)
2.870(4)
2.895(5)
2.757(5)
3.075(3)
2.38
2.35
2.41
2.31
2.47
2.61
2.17
2.22
2.25
2.32
2.43
2.50
2.16
2.14
2.21
2.26
2.27
2.30
2.45
2.40
101
115
100
139
170
141
110
109
122
121
99
127
111
111
109
121
122
121
99
Câ1-H1
Câ1-H2
N1-H
C2-H
C5-H
pBrBz-(∆Ala)2-OMe
pBrBz-(∆Ala)3-OMe
N1-H
N2-H
Câ1-H1
Câ2-H1
Câ2-H2
C2-H
N1-H
N2-H
N3-H
Câ1-H1
Câ2-H1
Câ3-H1
Câ3-H2
C2-H
x, y, z
x, y, z
x, -1 + y, z
129
Exciting, new features of ∆Ala fully extended peptides are
the following. (i) These molecules are flat, including the amino
acid side chains. (ii) The molecules pack in layers, without any
significant contribution from intermolecular N-H‚‚‚OdC H-
bonds. (iii) The average intramolecular CR ‚‚‚CR
distance
i
i+1
(3.77 Å) is the largest known separation of this type for any
peptide structure experimentally found to date. (iv) This novel
peptide structure is stabilized by two types of intramolecular
H-bonds, Ni-H‚‚‚OidC′i, involving all of the NH groups of
the molecule and typical of the five-membered ring C5 form,
and Câi+1-H‚‚‚OidC′i, characteristic of ∆Ala peptides and
giving rise to a six-membered ring system. In this connection,
it is gratifying to note that the structure found in the crystal
state corresponds to the most populated conformation in a poorly
interacting solvent such as deuteriochloroform.
In summary, in the present conformational study of (∆Ala)1-6
homopeptides, we have identified for the first time a completely
flat peptide structure. For these striking planar peptide sheets,
we foresee a bright future in biochemical and materials science
applications.
Figure 6. Packing mode of the pBrBz-(∆Ala)3-OMe molecules in the
crystal as viewed (A) along the b direction and (B) along the a direction.
For clarity, in the latter view, only the molecules belonging to two
adjacent layers are shown with different bond thickness. Intermolecular
C-H‚‚‚OdC′ contacts are represented by dashed lines.
Acknowledgment. M.C., F.F., and C.T. are grateful to the
Ministry of University and Technological Research (M.U.R.S.T.)
and the National Research Council (C.N.R.) of Italy for their
continuous and generous support of this research.
Conclusions
The experimental results reported by other groups2,5-7 and
in the present work strongly support the view that γ-branched
∆AAs (∆ZPhe and ∆ZLeu) are very effective â-turn/310-helix
formers, whereas ∆Ala, with no Câ-substituents, behaves quite
differently, overwhelmingly preferring the fully extended (2.05-
helical) conformation. Steric effects that might cause warping
of a Câ-substituted ∆AA peptide main chain are absent in ∆Ala
peptides. Thus, these findings emphasize the need for carefully
taking into account Câ-substitution in ∆AAs for a correct peptide
design. It still remains to be seen whether longer ∆Ala stretches
would fold into the 310-helical structure, as suggested by recent
theoretical studies.9a-c
Supporting Information Available: Experimental details
for the synthesis and characterization of peptides, and tables of
crystal data and structure refinement, atomic coordinates and
equivalent isotropic displacement parameters, bond lengths and
angles, anisotropic displacement parameters, hydrogen coordi-
nates and isotropic displacement parameters, and selected torsion
angles for pBrBz-∆Ala-OMe, pBrBz-(∆Ala)2-OMe, and pBrBz-
(∆Ala)3-OMe (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
JA9842114