Multiple conformational states in crystals and in solution in αγ hybrid peptides. Fragility of the C12 helix in short sequences

Article abstract of DOI:10.1021/jo8009819

(Chemical Equation Presented) The conformational properties of foldamers generated from αγ hybrid peptide sequences have been probed in the model sequence Boc-Aib-Gpn-Aib-Gpn-NHMe. The choice of α-aminoisobutyryl (Aib) and gabapentin (Gpn) residues greatl

Full text of DOI:10.1021/jo8009819

Multiple Conformational States in Crystals and in Solution in rγ
Hybrid Peptides. Fragility of the C12 Helix in Short Sequences
Sunanda Chatterjee,† Prema G. Vasudev,‡ Kuppanna Ananda,† Srinivasarao Raghothama,*,§
Narayanaswamy Shamala,*,‡ and Padmanabhan Balaram*,†
Molecular Biophysics Unit, Department of Physics, and NMR Research Centre, Indian Institute of Science,
Bangalore-560012, India
pb@mbu.iisc.ernet.in; shamala@physics.iisc.ernet.in; sr@nrc.iisc.ernet.in
ReceiVed May 6, 2008
The conformational properties of foldamers generated from Rγ hybrid peptide sequences have been probed
in the model sequence Boc-Aib-Gpn-Aib-Gpn-NHMe. The choice of R-aminoisobutyryl (Aib) and
gabapentin (Gpn) residues greatly restricts sterically accessible conformational space. This model sequence
was anticipated to be a short segment of the Rγ C12 helix, stabilized by three successive 4f1 hydrogen
bonds, corresponding to a backbone-expanded analogue of the R polypeptide 310-helix. Unexpectedly,
three distinct crystalline polymorphs were characterized in the solid state by X-ray diffraction. In one
form, two successive C12 hydrogen bonds were obtained at the N-terminus, while a novel C17 hydrogen-
bonded γRγ turn was observed at the C-terminus. In the other two polymorphs, isolated C9 and C7
hydrogen-bonded turns were observed at Gpn (2) and Gpn (4). Isolated C12 and C9 turns were also
crystallographically established in the peptides Boc-Aib-Gpn-Aib-OMe and Boc-Gpn-Aib-NHMe,
respectively. Selective line broadening of NH resonances and the observation of medium range NH(i) T
NH(i+2) NOEs established the presence of conformational heterogeneity for the tetrapeptide in CDCl3
solution. The NMR results are consistent with the limited population of the continuous C12 helix
conformation. Lengthening of the (Rγ)n sequences in the nonapeptides Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-
Aib-Gpn-Xxx (Xxx ) Aib, Leu) resulted in the observation of all of the sequential NOEs characteristic
of an Rγ C12 helix. These results establish that conformational fragility is manifested in short hybrid Rγ
sequences despite the choice of conformationally constrained residues, while stable helices are formed
on chain extension.
Introduction
structures like helices and hairpins, thereby providing a route
to generate synthetic mimics of protein structures.1c,2 Several
unique structural features of polypeptide backbones containing
the higher homologues of R amino acids have been established
by extensive studies of oligo-ꢀ- and -γ-peptides.3 Hybrid peptide
design is facilitated by the use of stereochemically constrained
amino acids in which the range of accessible conformations is
limited by backbone substitution. For R residues, the most
widely used constrained residue is R-aminoisobutyric acid
(Aib).4 We have recently described the introduction of the
achiral, ꢀ,ꢀ-disubstituted γ amino acid residue gabapentin (Gpn,
Hybrid polypeptides which incorporate the higher homologues
of the R amino acids provide an entry to a new class of
hydrogen-bonded peptide structures.1 The area of foldamer
design has received great impetus from the realization that ꢀ-
and γ-amino acid residues can be accommodated into folded
* To whom correspondence should be addressed. Fax: (P.B.) 91-80-23600683/
91-80-23600535; (N.S.) 91-80-23602602/91-80-23600683.
† Molecular Biophysics Unit.
‡ Department of Physics.
§ NMR Research Centre.
10.1021/jo8009819 CCC: $40.75
Published on Web 07/29/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 6595–6606 6595

Chatterjee et al.
SCHEME 1. Definition of the Backbone Torsion Angles of
the Gpn Residue
1-(aminomethyl)cyclohexaneacetic acid) as a model residue in
the design of hybrid peptides.5 The geminal ꢀ substituents in
Gpn restrict the backbone torsion angles Cꢀ-Cγ (θ1) and CR-Cꢀ
(θ2) to gauche (θ ) (60°) conformations, thereby restricting
the range of energetically feasible conformations (Scheme 1.)5
Despite the limitations placed on θ1 and θ2, hybrid Rγ sequences
containing Gpn have a much wider range of conformational
choices than their RR counterparts. Indeed, several intramo-
lecular hydrogen bond stabilized conformations like the C7
(NH(i)fCO(i)) and C9 (CO(i-1)rNH(i+1), where Gpn is the
ith residue) structures have been characterized at the Gpn
residues in di- and tripeptides.5a One regular structure has been
observed in crystals for Rγ hybrid sequences in the tetrapeptide
Boc-Aib-Gpn-Aib-Gpn-OMe (1).1d This is the Rγ C12 helix
which may be viewed as the expanded hybrid peptide analogue
of the classical 310-helix observed in R peptides.4b,6 The
observation of a consecutive C12 hydrogen-bonded structure,
corresponding to one turn of a regular Rγ C12 helix, in peptide
1 (Figure 1a) prompted us to explore the possibility of an
extension of such a helical structure by introduction of an
additional hydrogen bond donor at the C-terminus (Figure 1b).
FIGURE 1. (a) Molecular conformation of Boc-Aib-Gpn-Aib-Gpn-
OMe (1) determined in crystals.1d (b) Anticipated conformation of Boc-
Aib-Gpn-Aib-Gpn-NHMe (2), stabilized by three consecutive C12
hydrogen bonds. Model is based on the conformation in Figure 1(a)
by fixing Gpn (4) torsion angles to the same value as Gpn (2).
This apparently minor structural modification resulted in the
observation of unexpected structures, providing new insights
into the accessible conformations for Rγ sequences containing
the conformationally constrained Gpn residue. We describe in
this report the conformational characterization of the peptide
Boc-Aib-Gpn-Aib-Gpn-NHMe (2) and extension of the Rγ
segment in the nonapeptides Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-
Aib-Gpn-Xxx-OMe (Xxx ) Aib (3) and Xxx ) Leu (4)).
Interestingly, 2 crystallized in three different polymorphic forms
permitting characterization of distinctly different backbone
conformations stabilized by intramolecular C7, C9, and C12
hydrogen bonds. An unanticipated C17 hydrogen bond stabilizing
a three-residue γRγ turn has also been observed. Solution NMR
studies support the presence of multiple conformational states
and averaging by exchange effects as revealed by the presence
of distinctive medium range NOEs between backbone NH
protons and selective line broadening. Lengthening of the
polypeptide chain stabilizes the C12 helix as revealed by NMR
studies of the nonapeptides 3 and 4.
(1) (a) Chatterjee, S.; Roy, R. S.; Balaram, P. J. R. Soc. Interface 2006, 4,
587–606. (b) Roy, R. S.; Balaram, P. J. Peptide Res. 2004, 63, 279–289. (c)
Karle, I. L.; Pramanik, A.; Banerjee, A.; Battacharjya, S.; Balaram, P. J. Am.
Chem. Soc. 1997, 119, 9087–9095. (d) Ananda, K.; Vasudev, P. G.; Sengupta,
A.; Raja, K. M. P.; Shamala, N.; Balaram, P. J. Am. Chem. Soc. 2005, 127,
16668–16674. (e) Baldauf, C.; Gunther, R.; Hofmann, H. J. Biopolymers 2006,
84, 408–413. (f) Baldauf, C.; Gunther, R.; Hofmann, H. J. J. Org. Chem. 2006,
71, 1200–1208. (g) Schmitt, M. A.; Choi, H. S.; Guzei, I. A.; Gellman, S. H.
J. Am. Chem. Soc. 2005, 127, 13130–13131. (h) Schmitt, M. A.; Choi, H. S.;
Guzei, I. A.; Gellman, S. H. J. Am. Chem. Soc. 2006, 128, 4538–4539.
(2) (a) Rai, R.; Balaram, P. In Foldamers: Structure, Properties and
Applications; Hecht, S., HucI. Eds.; Wiley-VCH Verlag GmbH & Co.: New
York, 2007; pp 147-171. (b) Goodman, C. M.; Choi, S.; Shandler, S.; DeGrado,
W. F. Nature Chem. Biol. 2007, 3, 252–262. (c) Karle, I. L.; Gopi, H. N.;
Balaram, P. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 5160–5164. (d) Roy, R. S.;
Gopi, H. N.; Raghothama, S.; Karle, I. L.; Balaram, P. Chem. Eur. J. 2006, 12,
3295–3302.
Results
Crystalline Polymorphs of Peptide 2. Figure 2 illustrates
the observed conformations of peptide 2 in distinct poly-
morphic forms. The relevant backbone torsional angles and
intramolecular hydrogen bond parameters in polymorphs
2a-c are listed in Tables 1 and 2, respectively. The two
molecules in the asymmetric unit of 2a in the space group
P21/c adopt almost identical backbone (mirror image)
conformations, with the Aib-Gpn-Aib segment forming two
consecutive Rγ/γR turns resulting in an incipient C12 helix,
which can be formally considered as a backbone-expanded
analogue of the 310 (C10)-helix.6a,b The C-terminus adopts a
folded conformation resulting in a C17 hydrogen bond
between the Aib (1) CO and methylamide NH groups. The
Aib (1) CO group participates in a bifurcated hydrogen bond
interaction with two donor groups, NHMe and Gpn (4) NH.
The two independent molecules in the asymmetric unit differ
in the orientation of the geminal substituents on the cyclo-
hexane ring of the Gpn (4) residue. Polymorphs 2b and 2c
reveal a completely different backbone conformation. In both
(3) (a) Gellman, S. H. Acc. Chem. Res. 1998, 31, 173–180. (b) Seebach, D.;
Beck, A. K.; Bierbaum, D. J. Chem. BiodiVersity 2004, 1, 1111–1239. (c)
Seebach, D.; Hook, D. F.; Glattli, A. Biopolymers 2006, 84, 23–37. (d) Cheng,
R. P.; Gellman, S. H.; DeGrado, W. F. Chem. ReV. 2001, 101, 3219–3232. (e)
Hanessian, S.; Luo, X.; Schaum, R.; Michnick, S. J. Am. Chem. Soc. 1998, 120,
8569–8570. (f) Hanessian, S.; Luo, X.; Schaum, R. Tetrahedron Lett. 1999, 40,
4925–4929. (g) Grotenberg, G. M.; Timmer, M. S. M.; Llamas-Siaz, A. L.;
Verdoes, M.; van der Marel, G. A.; van Raaij, M. J.; Overkleeft, H. S.; Overhand,
M. S. J. Am. Chem. Soc. 2004, 126, 3444–3446.
(4) (a) Prasad, B. V. V.; Balaram, P. CRC Crit. ReV. Biochem. 1984, 16,
307–347. (b) Karle, I. L.; Balaram, P. Biochemistry 1990, 29, 6747–6756. (c)
Venkatraman, J.; Shankaramma, S. C.; Balaram, P. Chem. ReV. 2001, 101, 3131–
3152. (d) Aravinda, S.; Shamala, N.; Roy, R. S.; Balaram, P. Proc. Indian Acad.
Sci. (Chem. Sci.) 2003, 115, 373–400. (e) Toniolo, C.; Benedetti, E. Macro-
molecules 1991, 24, 4004–4009. (f) Crisma, M.; Formaggio, F.; Toniolo, C.;
Moretto, A. Biopolymers (Peptide Sci.) 2006, 84, 3–12.
(5) (a) Vasudev, P. G.; Ananda, K.; Chatterjee, S.; Aravinda, S.; Shamala,
N.; Balaram, P. J. Am. Chem. Soc. 2007, 129, 4039–4048. (b) Ananda, K.;
Aravinda, S.; Vasudev, P. G.; Raja, K. M. P.; Sivaramakrishnan, H.; Nagarajan,
K.; Shamala, N.; Balaram, P. Curr. Sci. 2003, 85, 1002–1011. (c) Banerjee, A.;
Balaram, P. Curr. Sci. 1997, 73, 1067–1077, and ref 3d.
(6) (a) Donohue, J. Proc. Natl. Acad. Sci. U.S.A. 1953, 39, 470–478. (b)
Toniolo, C.; Benedetti, E. Trends Biochem. Sci. 1991, 16, 350–353. (c) For a
discussion on the relationship between the hydrogen-bonded structures observed
in RR and Rγ segments, see ref 1a.
6596 J. Org. Chem. Vol. 73, No. 17, 2008

Fragility of the C12 Helix in Short Sequences
FIGURE 2. Molecular conformation of Boc-Aib-Gpn-Aib-Gpn-NHMe (2) in three polymorphic crystals: (a) polymorph 2a, (b) polymorph 2b, and
(c) polymorph 2c. rmsd for backbone superposition of 2b and 2c is 0.12 Å.
TABLE 1. Backbone Torsion Angles (deg) for Peptides 2, 5, and 6a
2a (mol. 1)
2a (mol. 2)
residue
φ
θ1
θ2
ψ
ω
φ
θ1
θ2
ψ
ω
Aib (1)
Gpn (2)
Aib (3)
Gpn (4)
59.3
127.9
53.1
43.0
112.9
48.7
174.4
176.2
174.7
178.0
-57.8
-129.2
-50.8
108.9
-38.4
-108.9
-51.4
-174.5
-176.5
179.8
-55.3
-63.4
-62.1
-64.8
53.0
64.2
63.3
62.3
-105.7
108.4
-104.9
-177.1
2b
2c
residue
φ
θ1
θ2
ψ
ω
φ
θ1
θ2
ψ
ω
Aib (1)
Gpn (2)
Aib (3)
Gpn (4)
-53.1
-106.9
58.4
-46.2
-77.4
-148.3
-121.1
-174.8
-174.7
-173.4
-174.7
-59.0
-99.7
57.1
-45.9
-76.9
-146.3
-123.6
-177.3
-173.0
-175.3
-174.2
63.9
75.7
68.4
73.8
-103.8
-46.8
-50.1
-104.4
-45.1
-47.5
5 (mol. 1)
5 (mol. 2)
residue
φ
θ1
θ2
ψ
ω
φ
θ1
θ2
ψ
ω
Aib (1)
Gpn (2)
Aib (3)
60.6
128.0
-52.7
33.8
116.7
-47.9
176.9
-176.7
-173.3
-57.5
-131.9
49.2
-38.7
-110.4
49.8
-174.9
179.0
175.9
-56.3
-59.1
55.3
61.5
6
residue
φ
θ1
66.1
θ2
75.1
ψ
ω
Gpn (1)
Aib (2)
-102.3
-60.5
-81.4
-49.4
-173.5
-173.3
a The choice of sign is arbitrary and represents one enantiomeric form present in the centrosymmetric crystals. For convenience, the “-” signs of the
φ, ψ values have been chosen, since this corresponds to the observed handedness in the case of peptides formed by L-amino acids. In the case of 2a
and 5, the two molecules in the asymmetric unit have opposite handedness. Estimated standard deviations ∼1° (2a), 0.2° (2b), 0.6° (2c), and 0.5° (5, 6).
cases, the Gpn (2) adopts a C9 conformation, while Gpn (4)
adopts a C7 conformation. Notably, in both the cases, Aib
(3) residue adopts an unusual polyproline II (PII) conforma-
tion (φ ≈ -60° and ψ ≈ +120°). The Aib residue has been
shown to adopt helical conformations (310/R, φ ) (60°, ψ
) (30°), in a very large number of peptides characterized
so far by X-ray diffraction.4 A lone water molecule bridges
the Boc CO and Aib (3) CO groups in 2b. In 2c, the water
bridge is between Aib (1) CO and Aib (3) CO groups (Table
2). The water molecule forms two strong hydrogen bonds to
the acceptor carbonyl groups in 2c, whereas in 2b, one of
the hydrogen bonds appears distinctly weaker (O · · · H ) 2.51
Å). Polymorphs 2b and 2c have very similar backbone
conformations and differ only in the presence of a methanol
molecule (occupancy 0.25) in the crystals of the former. 2a
and 2b/2c are conformational polymorphs, whereas 2b and
2c differ only in solvation.7 The three-residue γRγ turn (2a)
stabilized by a 17-atom (C17) hydrogen bond results in the
formation of a reverse turn structure, bringing CR(i) and
CR(i+4) into proximity (5.2 Å, Figure 3a.) Inspection of the
torsion angles in Table 1 reveals that the formation of the
C17 structure requires the Gpn (2) and Gpn (4) residues to
J. Org. Chem. Vol. 73, No. 17, 2008 6597

Chatterjee et al.
TABLE 2. Intramolecular Hydrogen-Bond Parameters in Peptides
2, 5, and 6
type donor (D) acceptor (A) D · · ·A (Å) H · · ·A (Å)
2a (mol. 1)
D-H· · ·A (°)
C12
C12
C17
N3
N4
N5
O0
O1
O1
2.93
2.93
2.88
2.08
2.08
2.02
167.8
171.3
174.0
(mol. 2)
O0
O1
C12
C12
C17
N3
N4
N5
2.91
3.07
2.97
2.09
2.21
2.11
159.9
179.0
175.2
O1
2b
O1
O4
O3
C9
C7
N3
N4
O1w
O1w
2.86
2.87
2.81
3.10
1.95
2.22
2.01
2.51
177.5
126.8
165.9
135.9
O0
2c
O1
O4
O1
C9
C7
N3
N4
O1w
O1w
2.85
2.77
2.93
2.80
1.96
2.08
2.07
2.02
168.9
133.6
163.9
167.8
FIGURE 3. (a) The γRγ C17 turn observed in the crystal structure of
Boc-Aib-Gpn-Aib-Gpn-NHMe (2a, -Gpn(2)-Aib(3)-Gpn(4)- segment
of mol. 2). Backbone torsion angles determining the turn are indicated.
(b) An alternate conformation of the tetrapeptide 2, generated by
selecting the allowed/crystallographically observed backbone torsion
angles for Aib and Gpn residues, which explains the medium range
NOEs shown in Figure 6.
O3
5 (mol. 1)
O0
(mol. 2)
O0
6
O0
C12
C12
C9
N3
N3
N2
3.04
2.97
2.83
2.19
2.12
2.02
171.7
172.9
175.9
at 300 K. A notable feature of the spectrum is the selective
broadening of the amide NH resonances of Gpn (2) and methyl
amide NH groups. The inset to Figure 5 illustrates the effect of
raising and lowering of temperature. Broadening is distinctly
more pronounced for the methylamide group upon heating and
cooling, in addition to large chemical shift changes. This
behavior suggests that exchange between multiple conformations
separated by widely different activation barriers is present in
solution. Figure 6 shows a partial ROESY spectrum of peptide
2 in CDCl3. Inspection of the NOEs observed between backbone
NH protons reveals both sequential and medium range dNN
NOEs. The dNN 1T2, 2T3, and 3T4 NOEs are indeed observed
in Figure 6, supporting formation of C12 helical turns over the
Aib(1)-Gpn(2)-Aib(3) segment. Indeed, expected NH(i) T
NH(i+1) distance in a C12 helix are dNN(Rγ) ) 2.7 Å and
dNN(γR) ) 3.4 Å. This is further elaborated in the discussion
on the nonapeptide 3. No detectable NOE was observed between
Gpn (4) NH and the methylamide NH protons, suggesting the
absence of the third helical C12 turn. In the observed crystal
structure of polymorph 2a this interproton distance is 2.7 Å,
suggesting that in solution the novel C17-turn conformation may
be populated only to a limited extent.
The following medium-range dNN NOEs are also observed:
Aib (1) NH T Aib (3) NH and Gpn (2) NH T Gpn (4) NH.
The corresponding interproton distances in the C12 helix are
4.7 and 4.6 Å, respectively, which are beyond the distance limits
for efficient interproton dipolar interactions. Consequently, these
NOEs must arise from a distinctly different set of peptide
conformations. The medium-range NOEs may be rationalized
by considering an alternative “non-helical” conformation sta-
bilized by two C12 hydrogen bonds (Figure 3b). In this structure,
the Aib(1)-Gpn(2) segment forms a “non-helical” C12 turn which
has been built using allowed, crystallographically characterized
conformations of Aib and Gpn residues, resulting in an
interproton distance of 3.8 Å between Aib (1) NH and Aib (3)
NH and an interproton distance of 2.9 Å between Gpn (2) NH
and Gpn (4) NH. This structure is stabilized by two intramo-
lecular hydrogen bonds. Indeed, the local residue conformation
used in this model for Gpn (1) is crystallographically observed
for Gpn (4) in polymorph 2a.
adopt unique local conformations with distinct φ values (Gpn
(2), φ ) 127.9° (mol. 1) and -129.2° (mol. 2); Gpn (4), φ
) -105.7° (mol. 1) and 108.9° (mol. 2)).
Conformations in Crystals of Isolated rγ and γr Units.
In order to evaluate the conformational propensities and
stabilities of Aib-Gpn and Gpn-Aib segments, the crystal
structures of Boc-Aib-Gpn-Aib-OMe (5) and Boc-Gpn-Aib-
NHMe (6) were determined. The molecular conformations are
shown in Figure 4, and the backbone conformational angles and
hydrogen bond parameters are summarized in Tables 1 and 2,
respectively. In 5, the Aib-Gpn segment adopts a helical C12
turn in both molecules in the asymmetric unit. In contrast, the
Gpn-Aib segment in 6 does not reveal the formation of a C12
hydrogen bond. Rather, the Gpn residue adopts a C9 hydrogen
bonded conformation, which can be formally viewed as an
expanded version of the C7 (γ-turn) feature observed in R
peptide structures.8 The N · · ·O distance between the groups Boc
CO and NHMe, which were expected to form a C12 hydrogen
bond in this structure, is 5.0 Å. Notably, the backbone torsion
angles for both the Gpn and Aib residues are not very far from
the values which will result in a C12 hydrogen bond. From the
data presented in Table 1, it is observed that relatively small
changes of approximately 25° of Gpn φ and 30° of Gpn ψ result
in a transformation of the C12 turn into the C9 hydrogen bond,
which is a localized interaction determined only by the torsion
angles at Gpn.
Solution Conformations of Boc-Aib-Gpn-Aib-Gpn-NHMe
(2). Figure 5 shows the 500 MHz NMR spectrum of 2 in CDCl3
(7) (a) The terms conformational polymorphs, pseudopolymorphs, and
solvates have attracted considerable debate in the literature of organic solid-
state chemistry. When considering molecular arrangements in crystals of relatively
large molecules like oligopeptides, a precise application of these terms may be
inappropriate. For a discussion on polymorphism in organic crystals, see ref ,
and for a recent discussion on conformational polymorphism in small organic
molecules, see ref 8c. (b) Bernstein, J. Polymorphism in Molecular Crystals;
Oxford University Press: Oxford, 2002. (c) Nangia, A. Acc. Chem. Res. 2008,
41, 595–604.
(8) (a) Matthews, B. W. Macromolecules 1972, 5, 818–819. (b) Ne´methy,
G.; Printz, M. P. Macromolecules 1972, 5, 755–758. (c) Milner-White, E. J. J.
Mol. Biol. 1990, 216, 385–397.
6598 J. Org. Chem. Vol. 73, No. 17, 2008

Fragility of the C12 Helix in Short Sequences
FIGURE 4. Molecular conformation determined in crystals of (a) Boc-Aib-Gpn-Aib-OMe (5) and (b) Boc-Gpn-Aib-NHMe (6).
1
1
FIGURE 5. 500 MHz H spectrum of Boc-Aib-Gpn-Aib-Gpn-NHMe (2) in CDCl3 at 300 K. Inset shows the H spectrum of 2 at 233, 273, and
323 K.
The delineation of the intramolecularly hydrogen-bonded
(solvent shielded) NH groups in peptides can be conveniently
accomplished by measuring the solvent dependence of the NH
chemical shifts upon addition of a strongly hydrogen-bonding
solvent like DMSO-d6 or CD3OH to a solution of apolar peptides
in a poorly interacting solvent like CDCl3.9 Solvent-exposed
NH groups show large downfield shifts upon addition of DMSO-
d6 to CDCl3 solution of peptides. Buried and strongly hydrogen-
bonded NH groups show little or no solvent dependence. The
interpretation of solvent titration experiments in the case of
systems which exhibit multiple conformational states must be
approached cautiously. Figure 7a shows the amide NH reso-
nances of peptide 2 in varying CDCl3/DMSO-d6 compositions.
While the Gpn (2) and methylamide NH resonances are
appreciably broadened in CDCl3, addition of DMSO-d6 results
in a distinct broadening of the other three NH resonances.
However, over the range of solvent compositions studied, only
five distinct resonances are observed suggesting that the
observed chemical shifts may only represent dynamically
averaged values. The problem of unequal populations of the
exchanging species is another factor that needs to be considered.
Despite these caveats, the solvent titration curves shown in
Figure 7b reveal that the NH groups of Aib (3), Gpn (4), and
methylamide group show very little solvent dependence, sug-
gestive of their involvement in intramolecular hydrogen bonding
in the major species that are populated. This observation is
consistent with both the continuous C12 helix (three hydrogen
bonds) and also the conformation observed in crystals in the
polymorph 2a. Aib (1) NH behaves in the manner expected for
a completely solvent exposed NH group. The diminished solvent
sensitivity observed for the Gpn (2) NH resonance may be
rationalized by contributions from conformations in which this
residue adopts a C7 hydrogen bonded structure (Gpn (2)
NH· · ·Gpn (2) CO), a feature that is observed in the crystal
structures of model Gpn peptides5 (see also Gpn (4) of
polymorphs 2b and 2c in Figure 2).
Solution Conformations of Nonapeptides Boc-Aib-Gpn-Aib-
Gpn-Aib-Gpn-Aib-Gpn-Xxx-OMe. The conformational hetero-
geneity observed in the short (Rγ)n sequences prompted us to
examine the effects of chain extension in the nonapeptides 3
(9) (a) Pitner, T. P.; Urry, D. W. J. Am. Chem. Soc. 1972, 94, 1399–1400.
(b) Iqbal, M.; Balaram, P. J. Am. Chem. Soc. 1981, 103, 5548–5552.
J. Org. Chem. Vol. 73, No. 17, 2008 6599

Chatterjee et al.
groups. The ROESY spectrum illustrated in Figure 9 shows
sequential connectivities between NH protons along the peptide
backbone. Two dNN NOEs, Aib (3) T Gpn (4) and Gpn (4) T
Aib (5), are significantly weaker in intensities than the other
NOEs. In an ideal helical C12 structure, the distances dNN across
the R and the γ residues are 2.7 and 3.4 Å, respectively (Figure
9). This should lead to a pattern of alternating dNN NOE
intensities. Inspection of Figure 9 reveals that the 1T2, 5T6,
and 7T8 NOEs (dNN(Rγ), across the R residues) are appreciably
more intense than the 2T3, 4T5, 6T7, and 8T9 NOEs
(dNN(γR), across the γ residues). In a robust, continuous C12
helix, a strong dNN NOE between Aib (3) NH and Gpn (4) NH
might have been anticipated. However, this NOE is very weak
in the spectrum shown in Figure 9. Together with the observa-
tion of line broadening of Aib (3) NH and Gpn (4) NH
resonances, the experimental findings suggest the possibility of
conformational exchange processes involving the Aib (3)
residue. It is pertinent to note that the Aib (3) residue in the
tetrapeptide Boc-Aib-Gpn-Aib-Gpn-NHMe (2) adopts a polypro-
line II (PII) conformation in both the crystalline polymorphs 2b
and 2c. In a PII conformation, the dNN distance across the Aib
residue is ∼4.5 Å. It should be noted that unlike in the case of
tetrapeptide 2, no dNN[iT(i+2)] NOEs are observed in the
nonapeptide 3. This suggests that lengthening of the Rγ
sequence favors population of the continuous C12 helical
conformation. Further support for the continuous helical con-
formation in peptide 3 is obtained from a measurement of
solvent dependence of NH chemical shifts in CDCl3/DMSO-d6
mixtures. The ∆δ (δ(CDCl3+41.5%DMSO-d6) - δ(CDCl3), ppm) are as
follows: Aib (1) NH 1.59, Gpn (2) NH 0.71, Aib (3) 0.19, Gpn
(4) 0.25, Aib (5) -0.01, Gpn (6) 0.19, Aib (7) -0.03, Gpn (8)
0.05, Aib (9) -0.03. The higher ∆δ values exhibited by Aib
(1) NH and Gpn (2) NH indicate a significantly higher degree
of solvent exposure compared to the other NH resonances,
supporting a conformation in which the NH groups of residues
3-9 are involved in intramolecular hydrogen bonds. Hetero-
geneity of conformations in solution appears to be more
pronounced in the case of short Rγ sequences despite the
backbone constraints inherent in the Aib and Gpn residues.
A nonapeptide sequence Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-
Aib-Gpn-Leu-OMe (4) in which the terminal Aib has been
replaced by the Leu was also synthesized in order to examine
the effects of introduction of chirality on the C12 helix
handedness facilitating future circular dichroism studies. The
500 MHz 1H NMR spectra in CDCl3 revealed that the
distribution of backbone NH chemical shifts for residues 1-8
are almost identical to that observed for peptide 3. The pattern
of dNN NOEs and their intensities and the solvent dependence
of the NH chemical shifts are also almost identical to that
described for peptide 3. Interestingly, in this case also Aib (3)
and Gpn (4) NH resonances showed evidence of selective
broadening (see the Supporting Information).
FIGURE 6. 500 MHz partial ROESY spectrum of Boc-Aib-Gpn-Aib-
Gpn-NHMe (2) in CDCl3 at 300K. Medium-range NOEs (dNN[(i)T(i+2)]
are highlighted (mixing time ) 250 ms).
)
FIGURE 7. (a) Stacked 1H spectrum of 2 at different solvent
compositions (CDCl3 + DMSO-d6). (b) Plot of dependence of the NH
chemical shifts of peptide 2 as a function of solvent composition
(CDCl3+ DMSO-d6) (U ) Aib).
Discussion
Hybrid polypeptide sequences incorporating backbone-ho-
mologated amino acid residues afford an opportunity to mimic
the regular structures found in proteins or biologically active
peptides, using non-R-amino acid residues. They provide new
opportunities for exploring the relationships between the folded
polymeric backbones and the nature of the intramolecular
hydrogen bonds. Historically, the discovery of the R-helix and
the ꢀ-sheet structures for the polypeptides by Pauling and
(Xxx ) Aib) and 4 (Xxx ) LLeu). 500 MHz 1H NMR spectrum
of peptide 3 reveals well-dispersed NH resonances. Sequence-
specific assignments were achieved using ROESY spectra, using
the upfield position of the Aib (1) NH (urethane) as the starting
point for establishing connectivity. The spectrum in Figure 8
reveals the selective broadening of Aib (3) and Gpn (4) NH
6600 J. Org. Chem. Vol. 73, No. 17, 2008

Fragility of the C12 Helix in Short Sequences
FIGURE 8. 500 MHz 1H spectrum of nonapeptide Boc-Aib-Gpn-Aib-Gpn- Aib-Gpn-Aib-Gpn-Aib-NHMe (3) in CDCl3 at 300 K. Inset shows the
1
enlarged view of the amide region of the H spectrum of 3.
FIGURE 9. 500 MHz partial ROESY spectrum of peptide 3 in CDCl3 at 300 K. On the left side are shown the dNN distances across the R and γ
residue (mixing time ) 250 ms).
Corey10 was based on an appreciation of the critical role of
internal hydrogen bonds. The resurgence of interest in polypep-
tide structures follows the establishment of novel hydrogen
bonding possibilities in oligomeric ꢀ-peptides.3 The interest in
foldamer design also stems from the fact that hybrid structures
may be anticipated to be proteolytically more stable than their
all-R counterparts.3b,11 Our studies have been designed to
examine conformational possibilities in regular (Rγ)n sequences
using two conformationally constrained residues, Aib and Gpn.
In the case of all-R amino acid backbones most of our
understanding of hydrogen-bonded conformations has been
derived from the vast body of structural information obtained
from X-ray diffraction studies of natural proteins.12 In contrast,
information on hybrid sequences must necessarily be eked out
by crystallographic and NMR studies of synthetic sequences.
The interpretations of solution NMR data for short peptides are
often uncertain, because of dynamic averaging over multiple
conformational species. The use of backbone-constrained amino
acids limits the range of conformational excursions permitting
definitive characterization of specific folded structures.
(10) (a) Pauling, L.; Corey, R. B.; Branson, H. R. Proc. Natl. Acad. Sci.
U.S.A. 1951, 37, 205–211. (b) Pauling, L.; Corey, R. B. Proc. Natl. Acad. Sci.
U.S.A. 1951, 37, 729–740.
(12) (a) Baker, E. N.; Hubbard, R. E. Prog. Biophys. Mol. Biol. 1984, 44,
97–179. (b) Rose, G. D.; Gierasch, L. M.; Smith, J. A. AdV. Protein Chem.
1985, 37, 1–109. (c) Sibanda, B. L.; Thornton, J. M. Nature 1985, 316, 170–
174. (d) Wilmot, C. M.; Thornton, J. M. J. Mol. Biol. 1988, 203, 221–232. (e)
Richardson, J. S. AdV. Protein Chem. 1981, 34, 167–339.
¨
ˇ
(11) Hook, D. F.; Bindsch; adler, P.; Mahajan, Y. R.; Sebesta, R.; Kast, P.;
Seebach, D. Chem. BiodiVersity 2005, 2, 591–632.
J. Org. Chem. Vol. 73, No. 17, 2008 6601

Chatterjee et al.
FIGURE 10. Representation of the backbone conformations of Aib and Gpn residues in the tetrapeptide 2 in φ,ψ space. The anticipated C12 helix
(pink), the C12/C17 structure in polymorph 2a (blue), and the C9/C7 structure in polymorphs 2b and 2c (red) are shown. The experimentally observed
conformational space for Aib and Gpn residues are marked with circles. The plot for Gpn (at θ1 ≈ θ2 ≈ +60°) includes 46 gabapentin residues
from 25 peptides. The C13 and C12 circles in the quadrant C corresponds to the hydrogen bonded turns in C13/C10 and C12/C10 mixed helices with
alternate hydrogen bond polarity. Lines connect linked residues in the peptide. Conformational transitions correspond to changes in the location of
a given residue in the φ,ψ space. For all three structures, residues 1 and 2 (shown in black color) are in the same region of the conformational
space. The lines and residue numbers follow the same color code as given for the structures.
For the achiral residue Aib, four distinct regions of the
conformational space may be identified as potential energy
minima. These are the helical RR (φ ) -60°, ψ ) -30°), the
helical RL (φ ) +60°, ψ ) +30°), the polyproline PIIL (φ )
-60°, ψ ) +120°), and the polyproline PIIR (φ ) +60°, ψ )
-120°). In the case of this achiral residue, the pairs RR/RL and
tion. Interestingly, the C9 ribbon observed in the (γ)n sequence,
the C12 helix observed in (Rγ)n sequences and the C13 helix
anticipated for the (ꢀγ)n sequence5a,16 lie in the same quadrant
clustered in a limited region of the φ-ψ space. All three regular
structures have the same hydrogen-bond directionality
(CO(i)· · ·NH(i+n)) as in the regular R peptide helices (for all-R
sequences, the corresponding helices are n ) 1, 2.27 ribbon/
helix (C7); n ) 2, 310-helix (C10); n ) 3, 3.613-helix (C13). The
Rγ C12 turn and the ꢀγ C13 turn5a clustered in the quadrant C
are expanded analogues of the “non-helical” ꢀ-turns observed
in isolated two-residue all-R-segments. These are “non-helical”
turns which provide a means of reversing polypeptide chain
directionality, commonly facilitating ꢀ-hairpin structures.4c,17
In the three-residue γRγ C17 hydrogen-bonded structure ob-
served in the tetrapeptide Boc-Aib-Gpn-Aib-Gpn-NHMe in the
polymorph 2a, Gpn (4) lies in quadrant C, while Gpn (2) is in
quadrant D. This conformational feature brings the CR of the
two flanking residues to a distance of 5.2 Å, which is strikingly
similar to that observed for the hairpin nucleating type II′
ꢀ-turn.18 Thus, both helix and hairpin promoting turns may be
readily generated with Rγ hybrid segments. Quadrant B contains
a cluster of C7 hydrogen-bonded structures of Gpn residues and
two examples of non-hydrogen-bonded Gpn residues which are
very close to the C7 conformation. The C7 structure is an
expanded version of the C5 structure proposed for the fully
extended R polypeptides.4f,13 While the C5 hydrogen bond in
the R residues is characterized by an extremely unfavorable
NH· · ·O angle of 90°, the parameters for the C7 hydrogen bond
are more favorable (N · · ·O ) 2.97 Å, H · · ·O ) 2.35 Å,
N-H· · ·O ) 130.5°).5a The C7 conformation corresponds to
a change in hydrogen bond directionality (NH(i)· · ·CO(i+n);
PIIL/PIIR are related by inversion about the origin of the φ-ψ
map. The fully extended conformation (φ ≈ ψ ≈ 180°) is not
considered here as it does not lead to foldamers analogous to
that in R amino acid structures.13 For Aib, the helical conforma-
tions are overwhelmingly favored with only a small number of
examples of PII conformations being observed in peptide crystal
structures.14
The γ residue, Gpn, provides two additional degrees of
backbone torsional freedom. The positioning of the cyclohexyl
substituent at the central Cꢀ atom restricts the torsional degrees
of freedom θ1 (Cꢀ-Cγ) and θ2 (CR-Cꢀ) to predominantly
gauche conformations (θ1 ≈ θ2 ≈ (60°), which are necessary
for nucleating local folding. A body of crystal structures of short
peptides determined in this laboratory has established the
population of three distinct regions of φ-ψ space for fixed
values of θ (θ1 ≈ θ2 ≈ 60°).5,15 Figure 10 illustrates the
observed Gpn conformations as cluster plots in φ-ψ space. In
achiral peptides crystallizing in centrosymmetric space groups,
symmetry-related molecules have inverted signs for all dihedral
angles. The experimental values may then be represented by
arbitrarily choosing one handedness. In the case of sequences
containing chiral residues, the appropriate Gpn torsion angle
signs can be chosen by considering an enantiomeric conforma-
(13) (a) Toniolo, C.; Benedetti, E. In Molecular Conformation and Biological
Interactions; Balaram, P., Ramaseshan, S. Eds.; Indian Academy of Sciences:
Bangalore, India, 1991; pp 511-521. (b) Toniolo, C. CRC Crit. ReV. Biochem.
1980, 9, 1–44.
(14) (a) Aravinda, S.; Shamala, N.; Balaram, P. Chem. BiodiVersity 2008, .
in press. (b) In a total of 1385 Aib residues characterized in peptide crystals,
1315 residues are helical (RL or RR) and 61 residues are in the PII region of the
φ,ψ space.
(16) The ꢀγ helical C13 turn was observed in the tripeptide Boc-ꢀLeu-Gpn-
Val-OMe (refs 1a and 5a).
(17) (a) Hughes, R. M.; Waters, M. L. Curr. Opin. Struct. Biol. 2006, 16,
514–524. (b) Rai, R.; Raghothama, S.; Balaram, P. J. Am. Chem. Soc. 2006,
128, 2675–2681, and references therein.
(15) (a) Aravinda, S.; Ananda, K.; Shamala, N.; Balaram, P. Chem. Eur. J.
2003, 9, 4789–4795. (b) Vasudev, P. G.; Shamala, N.; Ananda, K.; Balaram, P.
Angew. Chem., Int. Ed. 2005, 44, 4972–4975.
(18) The average CR-CR distance for residues flanking hairpin nucleating
type II′ ꢀ-turns (CR(i) · · · R(i+3), where i+1 and i+2 are the turn residues)
determined from analysis of eight peptide hairpin crystal structures is 5.1 Å.
6602 J. Org. Chem. Vol. 73, No. 17, 2008

Fragility of the C12 Helix in Short Sequences
FIGURE 11. dNN distances in different conformations of peptide 2: (a) the anticipated C12 helix; (b) C12 helix terminated with C17 turn determined
in the crystals of polymorph 2a; (c) C9/C7 conformation observed in polymorphs 2b and 2c (in this conformation, the only distance corresponding
to an observable NOE is dH1TH2); (d) model conformation which explains the 2T4 and 1T3 NOEs in the tetrapeptide 2. Residue conformations are
indicated. C12 (nh) and C12 (h) correspond to the helical and nonhelical C12 conformations of Gpn in quadrants C and D, respectively, in Figure 10.
(The 1T3 and 2T4 distances in parts a-c are as follows: (a)/(b) ≈ 4.6 Å; (c) 1T3 ≈ 5.7, 2T4 ≈ 8.5 Å.)
for the C7 structure, n ) 0) which is a reversal of that observed
in R polypeptide structures. In (Rγ)n sequences, reversal of
hydrogen bond polarity has been observed resulting in C10 turns
where n ) 1. Indeed, a novel regular structure consisting of
successive C12 (CO(i)· · ·HN(i+3), normal direction) and C10
(NH(i)· · ·OC(i+1), opposite direction) hydrogen bonds can be
generated in a repetitive (Rγ)n sequence resulting in a C12-C10
helical structure. Such a structure has been postulated on the
basis of theoretical calculations1f and observed in the structure
of the tetrapeptide Boc-Leu-Gpn-Leu-Aib-OMe, in which the
Leu residue adopts a PII conformation and the Gpn residue is
in the nonhelical C12 conformation, falling in quadrant C
(unpublished results). From Figure 10, it is seen that the
polymorphs of Boc-Aib-Gpn-Aib-Gpn-NHMe 2a, 2b, and 2c
and the continuous Rγ C12 helix are all generated by local
residue conformations, which lie in a limited region of φ-ψ
space for both Aib and Gpn residues. Despite the high degree
of local conformational selection in these constrained residues,
a high degree of structural heterogeneity is possible in these
Rγ hybrid sequences. It is evident that the transitions of Gpn
(4) between the quadrants B, C and D may indeed require the
crossing of appreciable activation barriers.
The multiplicity of accessible conformations is most evident
in the shorter sequences, for example, in peptide 2. While some
conformations may be captured in crystals, evidence for other
states begins to appear in NOE studies. In particular, the
interproton NOE between NH(i) and NH(i+2) is a clear
diagnostic of the presence of conformations other than the
continuous C12 helix. The measured short dNN distances in the
two distinct polymorphic forms observed in crystals are shown
in Figure 11. In short peptide sequences, NOEs up to 3.5-4.0
Å are detectable, although there is a sharp fall in intensities for
interproton distances greater than 3.5 Å. It is evident from Figure
11 that in the conformations 2a-c no dNN[iT(i+2)] NOEs are
likely to be observed. Figure 11d illustrates a model conforma-
tion constructed by choosing local residue conformations within
the experimentally observed regions of φ-ψ space depicted in
Figure 10. For an Rγ sequence forming a nonhelical C12 turn,
the dNN[(i)T(i+2)] distance is approximately 3.8 Å. Notably, the
H(1)TH(3) NOE in Figure 6 is extremely weak. Interestingly,
when the γ residue in a C12 nonhelical conformation is followed
by an R residue in a helical conformation, the NH(4) is left
free, resulting in the absence of a C12 hydrogen bond in the
segment, but yields a short H(2)TH(4) distance of 2.9 Å and
brings H(2) and H(3) as close as 2.2 Å. Interestingly, H(2)TH(4)
NOEs observed in Figure 6 are of significantly greater intensities
than the H(1)TH(3) NOE. In the case of peptide 2, the NOE
data strongly suggests the presence of multiple conformational
states, which are distinct from the conformations characterized
in crystals. In such a situation, the interpretations of the NOE
intensities are fraught with uncertainty. Small populations of
conformations with short interproton distances can contribute
significantly to the observed NOEs. The absence or weakness
of a particular NOE cannot be interpreted in terms of a specific
solution conformation.19
Lengthening of the polypeptide chain in the nonapeptides 3
and 4 appears to result in nucleation and stabilization of the
regular Rγ C12 helix. In these cases, the alternating pattern of
intensities of the sequential dNN NOEs across the R and γ
residues provides a clear signature of the C12 helical conforma-
tion. The chain length dependence of helix formation in all-R
peptides is a well-established phenomena.20 The present studies
indicate that continuous helix formation in hybrid Rγ sequences
can also be facilitated by the lengthening of the peptide chain.
The design of repetitive hybrid (Rγ)n peptides using residues
with limited conformational choices, Aib and Gpn, was under-
taken to establish the diversity of intramolecularly hydrogen-
bonded structures in these sequences. The characterization of
the C12 helix and the observation of the γRγ C17 hydrogen
bonded turn illustrate examples of hydrogen-bonded hybrid
structures, which are potentially capable of mimicking secondary
structural features in all-R peptides. The area of hybrid peptide
design using γ-amino acid residues will undoubtedly receive
further impetus from the introduction of novel backbone
modified residues using newly developed synthetic methodolo-
gies.21
(19) Neuhaus, D.; Williamson, M. P. The Nuclear OVerhauser Effect in
Structural and Conformational Analysis; Wiley-VCH: New York, 2000.
(20) (a) Goodman, M.; Verdini, A. S.; Toniolo, C.; Phillips, W. D.; Bovey,
F. A. Proc. Natl. Acad. Sci. U.S.A. 1969, 64, 444–450. (b) Scholtz, J. M.;
Baldwin, R. L. Ann. ReV. Biophys. Biomol. Struct. 1992, 95–118. (c) Fiori, W. R.;
Miick, S. M.; Millhauser, G. L. Biochemistry 1993, 32, 11957–11962.
(21) Chi, Y.; Guo, L.; Kopf, N. A.; Gellman, S. H. J. Am. Chem. Soc. 2008,
130, 5608–5609.
J. Org. Chem. Vol. 73, No. 17, 2008 6603

Chatterjee et al.
added to a precooled solution of 2.5 g (7 mmol) of Boc-Aib-Gpn-
OH in 5 mL of dry THF followed by addition of 1.4 g (7 mmol)
of DCC and 1.0 g (7 mmol) of HOBT. The reaction mixture was
allowed to attain room temperature and was stirred for 2 days. The
reaction was worked up as detailed in (i). Boc-Aib-Gpn-Aib-Gpn-
NHMe 2 (3.0 g, 71%) was obtained as a white solid.
Experimental Procedures
Peptide Synthesis and Characterization. Peptide synthesis was
carried out using conventional solution-phase procedures.22 The
details of the synthesis are discussed below.
(A) Boc-Aib-Gpn-Aib-Gpn-NHMe (2). (i) Boc-Gpn-NHMe.5b
Boc-Gpn-OH (2.7 g, 10 mmol) was dissolved in 10 mL of dry
THF (tetrahydrofuran) and cooled in an ice bath while stirring. Et3N
(1.3 mL, 10 mmol) and EtOCOCl (1.3 mL, 10 mmol) were added
into the reaction mixture. Thirty milliliters of dry THF saturated
with methylamine gas was added. The reaction mixture was left
undisturbed overnight. The solvent was evaporated and the residue
dissolved in ethyl acetate. The organic layer was washed succes-
sively with brine (3 × 30 mL), 2 N HCl (3 × 30 mL), 1 M sodium
carbonate (3 × 30 mL), and brine. This layer was dried over
anhydrous sodium sulfate and evaporated in vacuo to yield Boc-
Gpn-NHMe (2.3 g, 80%) as a white solid.
Peptide 2 was purified by reversed-phase, medium-pressure liquid
chromatography (C18, 40-60 µm) and high-performance liquid
chromatography (HPLC) on a reversed-phase C18 column (5-10
µm, 7.8-250 mm) using methanol/water gradients. ESI-MS,
MH+
(MH+calcd) Da: 608.5 (608). Peptide identity was also
obsd
confirmed by complete assignment of 500 MHz 1H NMR spectra.
(B) Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-OMe (3). (i)
Boc-Aib-Gpn-Aib-OMe. Boc-Aib-Gpn-Aib-OMe was synthesized
by coupling 3.5 g (10 mmol) of Boc-Aib-Gpn-OH obtained as
discussed in section A(iii) to H-Aib-OMe obtained from 3.0 g (20
mmol) of H-Aib-OMe hydrochloride, upon neutralization, extrac-
tion, and concentration using DCC/HOBT as coupling agents. The
reaction was worked up as in section A(iv) to yield the tripeptide
(ii) Boc-Aib-Gpn-NHMe. Boc-Gpn-NHMe (2.3 g, 8 mmol) was
deprotected with 40 mL of 98% formic acid, and the deprotection
was monitored by TLC. After 2 h, formic acid was evaporated;
the residue was taken in 10 mL of water. The pH of the aqueous
layer was adjusted to ∼8 by addition of sodium carbonate and
extracted with ethyl acetate (3 × 30 mL). The combined organic
layer was washed with brine solution (30 mL) and concentrated in
vacuo to ∼2 mL. The peptide-free base was added to a precooled
solution of 1.6 g (8 mmol) of Boc-Aib-OH in 5 mL of dry THF
followed by addition of 1.6 g (8 mmol) of DCC (N,N′-dicyclo-
hexylcarbodiimide) and 1.2 g of HOBT (1-hydroxybenzotriazole).
The reaction mixture was allowed to attain room temperature and
was stirred for 2 days. The reaction was worked up as described in
(i). Boc-Aib-Gpn-NHMe (2.6 g, 87.5%) was obtained as a white
solid and used for synthesis without further purification. ESI-MS,
(3.6 g, 80%) as a white solid. ESI-MS, MH+
(MH+calcd) Da:
obsd
455.5 (455). The peptide was used for synthesis without purification.
(ii) Boc-Aib-Gpn-Aib-Gpn-Aib-OMe. Boc-Aib-Gpn-Aib-OMe
(3.6 g, 8 mmol) was deprotected with 40 mL of 98% formic acid
and was worked up as in section A(iv). The free base was added
to a precooled solution of 2.8 g (8 mmol) of Boc-Aib-Gpn-OH,
obtained as in section A(iii), in 5 mL of dry THF followed by
addition of 1.6 g (8 mmol) of DCC and 1.2 g (8 mmol) of HOBT.
The reaction mixture was allowed to attain room temperature and
was stirred for 2 days. The reaction was worked up as described in
section A(iv). Boc-Aib-Gpn-Aib-Gpn-Aib-OMe was obtained as a
white solid (4.8 g, 87.5%), which was used for synthesis without
further purification. ESI-MS, MH+obsd (MH+calcd) Da: 693.5 (693).
(iii) Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-OMe. Boc-Aib-Gpn-
Aib-Gpn-Aib-OMe (4.8 g, 7 mmol) was deprotected with 35 mL
of 98% formic acid, and the reaction was worked up as discussed
in section A(iv). The free base was added to a precooled solution
of 2.5 g (7 mmol) of Boc-Aib-Gpn-OH, obtained as in section A(iii),
in 5 mL of dry THF followed by addition of 1.4 g (7 mmol) of
DCC and 1.0 g (7 mmol) of HOBT. The reaction mixture was
allowed to attain room temperature and was stirred for 5 days. The
reaction was worked up as in section A(iv) above. Boc-Aib-Gpn-
Aib-Gpn-Aib-Gpn-Aib-OMe (4.8 g, 71%) was obtained as a white
solid, which was used for synthesis without further purification.
MH+
(MH+calcd) Da: 369.5 (369).
obsd
(iii) Boc-Aib-Gpn-OH.23 Boc-Aib-OH (2.0 g, 10 mmol) was
dissolved in 10 mL of dry THF and cooled in an ice bath. HOSu
(2.0 g, 17.4 mmol) followed by 2.0 g (10 mmol) of DCC was added.
The reaction mixture was allowed to attain room temperature. After
12 h, the DCU (N,N′-dicyclohexylurea) was filtered off, and the
filtrate was washed with brine (3 × 30 mL), 1 M sodium carbonate
(3 × 30 mL), and brine (1 × 30 mL). The organic layer was dried
over sodium sulfate and evaporated in vacuo. Boc-Aib-OSu (2.7
g, 90%) was obtained as a white solid. Boc-Aib-OSu (2.7 g, 9
mmol) was dissolved in 10 mL of THF followed by addition of
20% sodium carbonate solution and 2.4 g (9 mmol) of H-Gpn-
OH. The reaction was stirred for 24 h at room temperature. THF
was evaporated, and the residue was taken in 20 mL of water and
was washed with 30 mL of ethyl acetate. The aqueous layer was
acidified with 2 N HCl and was extracted with ethyl acetate (3 ×
30 mL). The pooled organic layer was washed with brine (30 mL)
and was dried over sodium sulfate and was evaporated in vacuo.
Boc-Aib-Gpn-OH (2.8 g, 88.9%) was obtained as a white solid
ESI-MS, MH+
(MH+calcd) Da: 931.5 (931).
obsd
(iii) Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-OMe (3). Boc-
Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-OMe (4.6 g, 5 mmol) was depro-
tected with 25 mL of 98% formic acid and reaction was worked
up as discussed in section A(iv). The free base was added to a
precooled solution of 1.8 g (5 mmol) of Boc-Aib-Gpn-OH, obtained
as in section A(iii), in 5 mL of dry THF followed by addition of
1.0 g (5 mmol) of DCC and 0.8 g (5 mmol) of HOBT. The reaction
mixture was allowed to attain room temperature and was stirred
for 5 days. The reaction was worked up as in section A(iv). Boc-
Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-OMe 3 (3.9 g, 60%) was
obtained as a white solid.
and used without further purification. ESI-MS, MH+obsd (MH+
Da: 356.5 (356).
)
calcd
(iv) Boc-Aib-Gpn-Aib-Gpn-NHMe (2). Boc-Gpn-Aib-NHMe (2.6
g, 7 mmol) was deprotected with 35 mL of 98% formic acid, and
the deprotection was monitored by TLC. After 3 h, formic acid
was evaporated, and the residue was taken in 10 mL of water. The
pH of the aqueous layer was adjusted to ∼8 by addition of sodium
carbonate and extracted with ethyl acetate (3 × 30 mL). The
combined organic layer was washed with brine solution (30 mL)
and concentrated in vacuo to ∼2 mL. The dipeptide-free base was
Peptide 3 was purified by reversed-phase, medium-pressure liquid
chromatography (C18, 40 - 60 µm) and high-performance liquid
chromatography (HPLC) on a reversed-phase C18 column (5-10
µm, 7.8-250 mm) using methanol/water gradients. ESI-MS,
MH+
(MH+calcd) Da: 1169.5 (1169). Peptide identity was also
obsd
confirmed by complete assignment of 500 MHz 1H NMR spectra.
(22) Bodanszky, M.; Bodanszky, A. The Practice of Peptide Synthesis;
Springer-Verlag: Berlin, 1984.
(23) Aravinda, S.; Ananda, K.; Shamala, N.; Balaram, P. Chem. Eur. J. 2003,
9, 4789–4795.
(24) (a) Schneider, T. R.; Sheldrick, G. M. Acta Crystallogr. 2002, D58,
1772–1779. (b) Sheldrick, G. M. SHELXS-97, A program for automatic solution
of crystal structures; University of Go¨ttingen: Go¨ttingen, 1997. (c) Sheldrick,
G. M. SHELXL-97, A program for crystal structure refinement; University of
Go¨ttingen: Go¨ttingen, 1997.
(C) Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-Gpn-Leu-OMe (4).
(i) Boc-Aib-Gpn-Leu-OMe. Boc-Aib-Gpn-Leu-OMe was synthe-
sized by coupling 3.5 g (10 mmol) of Boc-Aib-Gpn-OH obtained
as discussed in section A(iii), to H-Leu-OMe obtained from 3.6 g
(20 mmol) H-Leu-OMe hydrochloride upon neutralization, extrac-
tion, and concentration using DCC/HOBT as coupling agents. The
reaction was worked up as in section A(iv). The tripeptide was
6604 J. Org. Chem. Vol. 73, No. 17, 2008

Fragility of the C12 Helix in Short Sequences
TABLE 3. Crystal and Diffraction Parameters for Boc-Aib-Gpn-Aib-Gpn-NHMe (2), Boc-Aib-Gpn-Aib-OMe (5), and Boc-Gpn-Aib-NHMe (6)
2a
C32 H57 N5 O6
small needle
ethyl acetate/hexane/methanol methanol/water/dioxane
2b
C32 H57 N5 O6. H2O. (0.25) CH3OH C32 H57 N5 O6. H2O
rhombus-shaped plates
2c
5
6
empirical formula
crystal habit
crystallizing solvent
crystal size (mm)
space group
C23 H41 N3 O6
C19 H35 N3 O4
plates
methanol/water
rhombus-shaped plates plates
methanol
0.2 × 0.09 × 0.07
P21/c
methanol/water
0.25 × 0.04 × 0.02 0.5 × 0.04 × 0.02
0.20 × 0.03 × 0.01
0.6 × 0.5 × 0.14
P21/c
P21/c
P21/c
P21/c
a (Å)
b (Å)
c (Å)
17.447(4)
11.722(3)
35.799(9)
96.50(1)
7274(3)
8
14.462(1)
13.179(1)
20.322(1)
104.416(1)
3734.0(5)
4
10.276(2)
12.307(2)
30.071(2)
105.896(3)
3657.4(11)
4
16.892(1)
16.436(1)
18.990(1)
92.128(1)
5268.8(3)
8
9.978(2)
20.950(4)
11.043(2)
110.464(3)
2162.9(6)
4
ꢀ (deg)
volume (Å3)
Z
molecules/asym unit
co-crystallized solvent
formula wt
2
none
607
1.11
2656
46.5
44479
0.34
10424
2397
1
1
2
none
455.5
1.15
1984
1
none
369.5
1.13
808
1 water + 0.25 methanol
632.8
1.13
1382
53.6
27608
0.022
7362
5348
water
625.8
1.14
1368
50.7
25443
0.10
6634
density (g/cm3) (cal)
F (000)
2θ Max. (°)
measured reflns
Rint
unique reflns
obsd reflns
53.1
49.4
38312
0.067
10048
4605
14763
0.035
3676
3348
3048
[|F| >4σ(F)]
final R (%)/wR2 (%)
goodness-of-fit (S)
∆Fmax (e Å3)/∆Fmin (e Å-3
restraints/parameters
data-to-parameter ratio
8.92/10.62
0.884
0.33/-0.19
9/755
4.97/13.85
0.974
0.27/-0.14
0/443
8.08/15.01
1.03
0.25/-0.20
0/425
7.96/16.28
1.008
0.21/-0.14
0/577
8.95/25.26
1.165
0.47/-0.23
0/247
)
4:1
12:1
8:1
8:1
12:1
obtained as a white solid (3.9 g, 80%), which was used for further
MH+
(MH+calcd) Da: 1197.5 (1197). Peptide identity was also
obsd
synthesis without purification. ESI-MS, MH+
483.5(483).
(MH+calcd) Da:
confirmed by complete assignment of 500 MHz 1H NMR spectra.
(D) Boc-Aib-Gpn-Aib-OMe (5) and Boc-Gpn-Aib-NHMe (6).
Peptide 5 was synthesized as discussed in section B(i). Peptide 6
was synthesized by coupling Boc-Gpn-OH (2.71 g, 10 mmol) to
Aib-NHMe, obtained by deprotecting Boc-Aib-NHMe, (2.2 g, 10
mmol) using DCC (2 g,10 mmol)/HOBT (1.5 g,10 mmol) as the
coupling agents. The final peptides were purified by reversed phase,
medium-pressure liquid chromatography (C18, 40-60 µm) using
methanol/water gradients. ESI-MS, MH+obsd (MH+calcd) Da: 5, 455.7
(455); 6, 370.5 (370).
obsd
(ii) Boc-Aib-Gpn-Aib-Gpn-Leu-OMe. Boc-Aib-Gpn-Leu-OMe
(3.9 g, 8 mmol) was deprotected with 40 mL of 98% formic acid
and was worked up as in section A(iv). The free base was added
to a precooled solution of 2.8 g (8 mmol) of Boc-Aib-Gpn-OH,
which was obtained as in section A(iii), in 5 mL of dry THF
followed by addition of 1.6 g (8 mmol) of DCC and 1.2 g (8 mmol)
of HOBT. The reaction mixture was allowed to attain room
temperature and was stirred for 3 days. The reaction was worked
up as in section A(iv). Boc-Aib-Gpn- Aib-Gpn-Leu-OMe (4.3 g,
75%) was obtained as a white solid, which was used for synthesis
without further purification. ESI-MS, MH+obsd (MH+calcd) Da: 721.5
(721).
NMR Spectroscopy. All proton NMR spectra were recorded at
500 MHz. Chemical shifts are reported with respect to internal
tetramethylsilane (TMS). Peptide concentrations used are as follows:
peptide 2, 3.7 mM; peptides 3 and 4, 1.9 mM. Aggregation effects
for peptides at these concentrations are not significant in the solvents
used. Sequence specific assignments were readily achieved, since
the Aib (1) NH (urethane NH) can be unambiguously identified by
its high-field position. The singlet and triplet appearances of Aib
and Gpn NH, respectively, together with sequential NOEs, permit
unambiguous assignment. The nuclear Overhauser effects (NOE)
were observed using ROESY experiments employing standard pulse
sequences from a manufacturer’s library. A single mixing time (CW
spin lock time) of 250 ms was used.
Crystal Structure Determination. Three polymorphic forms of
peptide 2, Boc-Aib-Gpn-Aib-Gpn-NHMe were obtained by slow
evaporation from different solvent mixtures. The peptide, when
crystallized independently from an ethyl acetate/hexane/methanol
mixture and from a methyl acetate/hexane mixture, yielded small
needles (polymorph 2a). Crystallization of a fresh sample of peptide
2 from a methanol/water/dioxane mixture yielded rhombus-shaped
single crystals (polymorph 2b). The crystals of the third polymorph
(2c), obtained by slow evaporation from methanol, resembled the
2b crystals, but were thinner. All three polymorphs crystallized in
the monoclinic space group P21/c. The asymmetric unit of poly-
morph 2a is composed of two independent peptide molecules.
Polymorph 2b crystallized along with a molecule of water and a
partially occupied (0.25) methanol molecule in the asymmetric unit,
while peptide 2c crystallized along with a lone water molecule in
the asymmetric unit. Peptides 5 and 6 were obtained from methanol/
water mixture by slow evaporation, in monoclinic space group P21/
c. Peptide 5 has two molecules in the asymmetric unit, while 6
(iii) Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-Leu-OMe. Boc-Aib-Gpn-
Aib-Gpn-Leu-OMe (4.3 g, 6 mmol) was deprotected with 30 mL
of 98% formic acid, and the reaction was worked up as discussed
in section A(iv). The free base was added to a precooled solution
of 2.1 g (6 mmol) of Boc-Aib-Gpn-OH, obtained as in section A(iii),
in 5 mL of dry THF followed by addition of 1.2 g (6 mmol) of
DCC and 0.9 g (6 mmol) of HOBT. The reaction mixture was
allowed to attain room temperature and was stirred for 5 days. The
reaction was worked up as in section A(iv). Boc-Aib-Gpn-Aib-
Gpn-Aib-Gpn-Leu-OMe (4.7 g, 83%) was obtained as a white solid,
which was used for synthesis without further purification. ESI-MS,
MH+
(MH+calcd) Da: 959.5 (959).
obsd
(iv) Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-Gpn-Leu-OMe (4).
Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-Leu-OMe (4.7 g, 5 mmol) was
deprotected with 25 mL of 98% formic acid, and the reaction was
worked up as discussed in section A(iv). The free base was added
to a precooled solution of 1.8 g (5 mmol) of Boc-Aib-Gpn-OH,
obtained as in section A(iii), in 5 mL of dry THF followed by
addition of 1.0 g (5 mmol) of DCC and 0.8 g (5 mmol) of HOBT.
The reaction mixture was allowed to attain room temperature and
was stirred for 7 days. The reaction was worked up as in section
A(iv), to yield Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-Gpn-Leu-OMe
4 (2.4 g, 40%) as a white solid.
Peptide 4 was purified by reversed-phase, medium-pressure liquid
chromatography (C18, 40 - 60 µm) and high-performance liquid
chromatography (HPLC) on a reversed-phase C18 column (5-10
µm, 7.8-250 mm) using methanol/water gradients. ESI-MS,
J. Org. Chem. Vol. 73, No. 17, 2008 6605

Chatterjee et al.
crystallized with one molecule in the asymmetric unit. X-ray
intensity data for all crystals were collected at room temperature
on a CCD diffractometer (Mo KR radiation, λ ) 0.71073 Å, ω-
scan). The structures were solved by direct methods using
SHELXD24a (2a) or SHELXS-9724b (2b, 2c, 5, 6) and refined
against F,2 with full-matrix least-squares methods using SHELXL-
97.24c The rest of the hydrogen atoms in 2b, 2c, and 6 and all of
the hydrogen atoms for 2a and 5 were fixed geometrically in
idealized positions and were refined as riding over the atoms to
which they were bonded. During the refinement, restraints were
applied on bond lengths and bond angles of Gpn (2) side chain of
Mol. Two in polymorph 2a, in order to maintain an acceptable
geometry of the cyclohexane ring. Poor crystal quality limits the
diffraction data to 0.9 Å resolution, resulting in large thermal
ellipsoids for the carbon atoms of the Gpn (2) side chain. Table 3
summarizes the crystal and diffraction parameters for peptides 2,
5, and 6.
and Program Support in the area of Molecular Diversity and
Design, Department of Biotechnology, India. We thank Dr. K.
Nagarajan of Hikal R&D Centre, Bangalore (India), for the
sample of gabapentin. S.C. and P.G.V. thank the CSIR for
Senior Research Fellowships. X-ray diffraction data were
collected at the CCD facility funded under the IRHPA program
of the Department of Science and Technology, India.
Supporting Information Available: X-ray crystallographic
information files for peptides 2, 5, and 6 (CIF), chemical shifts
1
of amide NH resonances for peptides 3 and 4 (Table S1), H
NMR and ROESY spectra of peptide 4, solvent titration plots
for peptides 3 and 4 (Figures S1-S4). This material is available
free of charge via the Internet at http://pubs.acs.org. The X-ray
crystallographic files (CIF) have also been deposited with the
Cambridge Structural Database with accession numbers CCDC
663392-663394 (2a-2c), 684733 (5), and 684734 (6).
Acknowledgment. This research was supported by a grant
from the Council of Scientific and Industrial Research (CSIR)
JO8009819
6606 J. Org. Chem. Vol. 73, No. 17, 2008