Development of a series of cross-linking agents that effectively stabilize α-helical structures in various short peptides

Article abstract of DOI:10.1002/chem.200700843

A series of cross-linking agents of varying rigidity and length were designed to stabilize helical structures in short peptides and were then synthesized. The sequences of the short peptides employed in this study each include two X residues (X=Dap, Dab, Orn, and Lys) at the i/i+4, i/i+7, or i/i+11 positions to provide the sites for cross-linking. These peptides were subjected to reaction with the synthesized cross-linking agents, and the helical content of the resulting cross-linked peptides were analyzed in detail by circular dichroism. For each of the peptide classes we found combinations with the cross-linking agents suitable for the construction of stable helical structures up to > 95 % helicity at 5°C. Our method could also be applied to biologically related sequences seen in native proteins such as Rev.

Full text of DOI:10.1002/chem.200700843

FULL PAPER
DOI: 10.1002/chem.200700843
Development of a Series of Cross-Linking Agents that Effectively ACHTRESUGN tabilize
a-Helical Structures in Various Short Peptides
Kazuhisa Fujimoto,* Masaoki Kajino, and Masahiko Inouye*[a]
Abstract:
A
series of cross-linking
cross-linking. These peptides were sub-
jected to reaction with the synthesized
cross-linking agents, and the helical
content of the resulting cross-linked
peptides were analyzed in detail by cir-
cular dichroism. For each of the pep-
tide classes we found combinations
with the cross-linking agents suitable
for the construction of stable helical
structures up to >95% helicity at 58C.
Our method could also be applied to
biologically related sequences seen in
native proteins such as Rev.
agents of varying rigidity and length
were designed to stabilize helical struc-
tures in short peptides and were then
synthesized. The sequences of the short
peptides employed in this study each
include two X residues (X=Dap, Dab,
Orn, and Lys) at the i/i+4, i/i+7, or
i/i+11 positions to provide the sites for
Keywords:
cross-linking
alpha
helices
·
·
agents
helical structures · peptides
Introduction
tides probably need to display much higher helical content
and greater thermal stabilities at room temperature for prac-
tical uses. Moreover, there was no reason to believe that the
combination of Lys residues and the acetylenes represented
the optimal linkage. It was thus decided that a variety of
cross-linked patterns should be developed for investigation
of the application of these peptides to life sciences, especial-
ly to peptide inhibitors and peptide drugs.[7] Here we report
the results obtained for a series of new cross-linking agents
Because particular amino acid sequences that exist as heli-
ces in proteins usually adopt random-coiled structures in
their isolated short-peptide forms, effective stabilization of
the helices requires artificial efforts[1] such as the utilization
of hydrogen bond surrogates,[2] the introduction of non-natu-
ral amino acids,[3] the use of specific sequences,[4] or the
cross-linking of two side chains in the residues.[5] In our pre-
vious paper, we were able to
stabilize a-helices of short pep-
tides with acetylenic cross-link-
ing agents (Figure 1).[6] In these
peptides, Lys residues were lo-
cated at the i/i+4and i/i+7 po-
sitions for the cross-linking re-
actions, and the cross-linked
peptides showed moderate a-
Figure 1. A schematic representation of the stabilization of helical structures in short peptides by use of cross-
linking agents.
helical content of up to 35%
(i/i+4 ) or 65% i(/i+7) at 58C.
However, the cross-linked pep-
ACHTREUNG
that effectively stabilize a-helical structures in various short
peptides, as well as further information on the acetylenic
ones.
[a] Dr. K. Fujimoto, M. Kajino, Prof. Dr. M. Inouye
Graduate School of Pharmaceutical Sciences
University of Toyama
Sugitani 2630, Toyama 930-0194(Japan)
Fax : (+81)76-434-5049
E-mail: fujimoto@pha.u-toyama.ac.jp
inouye@pha.u-toyama.ac.jp
Results and Discussion
Short peptides used in this study: One turn in an a-helix
consists of 3.6 amino acid residues, so the two side chains of
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2008, 14, 857 – 863
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
857

the residues at positions i and i+4(one turn: 0.5 nm from i),
i and i+7 (two turns: 1.1 nm from i), or i and i+11 (three
turns: 1.6 nm from i) in the helices are oriented in almost
the same direction. These distances in the helices help us to
identify favorable cross-linking agents for stabilizing helical
structures in short peptides. As shown in Figure 2a, two “X”
tides, which we called a “trigger effect”. If such regulation
could also be achieved by linking shorter distances (i/i+4 ) in
peptides, this strategy should be fruitful for regulating vari-
ous interactions, such as in peptide–protein and peptide–
DNA complexes. Therefore, if the peptide possesses a se-
quence relating to biological recognition events at its C ter-
minus, a cross-linked region at the N terminus can be uti-
lized as a control unit for the interactions.
In the cases of the peptides B–D, the cross-linking regions
are positioned around the centers of the sequences. This sit-
uation causes one side of such a cross-linked peptide to be
broadly masked with the cross-linking agents, leaving the
opposite side untouched. When the sequence of the amino
acid residues protruding at the opposite side is designed to
interact side by side with other biomolecules, this opposite
region could be used as a novel recognition motif with sev-
eral functional groups lined up linearly. Indeed, in the pep-
tides D, the positions of X were decided upon so as not to
hamper the arginine-rich side of the peptide from binding
with RRE RNA, on the basis of X-ray crystallographic re-
sults for the Rev–RRE complex.[11,12]
Type of X residue for the cross-linking sites: In the previous
paper, we chose Lys as a cross-linking residue because of its
natural presence in the peptides of interest. In the cases of
short peptides, however, solid-phase synthesis allows incor-
poration of any type of natural or non-natural amino acid
into the sequences, which can be applied to biological recog-
nition events at will. Moreover, cross-linked Lys residues no
longer maintain their inherent cationic properties for contri-
buting to interactions with other species. These features sug-
gest that the cross-linking site X should be variable with
each individual experiment for stabilizing a-helices. There-
fore, we additionally introduced diaminopropionic acid
(Dap), diaminobutyric acid (Dab), and ornithine (Orn) into
the peptide sequences as cross-linking sites X for extending
our strategy (Figure 2b). These four amino acids have vari-
ous lengths of methylene spacers between the terminal
amino group of the side chain and the a-carbon: Dap, Dab,
Orn, and Lys contain one, two, three, and four carbons, re-
spectively, for the spacers.
Figure 2. a) Helical structures and sequences of short peptides used in
this study: peptide classes A (i/i+4), B (i/i+7), C (i/i+11), and D (i/i+11).
b) Amino acids “X” (Dap, Dab, Orn, and Lys) as cross-linking agents in
short peptides.
residues are placed at the i/i+4, i/i+7, and i/i+11 positions
as cross-linking sites for peptide classes A (13 residues), B
(16 residues), and C (20 residues), respectively. Peptide class
D (20 residues) also has the i/i+11 relationship; its sequence
is extracted from the arginine-rich region seen in natural
Rev, a HIV-1 regulatory protein.[8,9]
Cross-linking agents: Ten cross-linking agents have been
newly tested, in addition to the acetylenic 1 and 2, as shown
in Figure 3.[6] In the new cross-linking agents, the spacers are
made up of short alkane, benzene, naphthalene, biphenyl,
phenanthrene, and fluorene cores without flexible alkoxy
chains, unlike in 1 and 2. The alkyl side chains of the cross-
linked amino acids X thus have a role for varying the overall
length of the resulting cross-linker, meaning that cross-link-
ing agents of slightly shorter length than a targeted pitch
might be advantageous for formation of stable helices. With
this in mind, short alkylene-based 3 and 4 would be expect-
ed to be suitable for stabilizing a-helices in the peptide class
A. Benzene-based 5 and 6, of moderate spacer length,
should be potent candidates for the peptides B, while the
longer ones of the biphenylene-based 7–9 and the highly
Locations of cross-linking sites in the short peptides: The
peptide class A was designed to have its cross-linking site X
at the N terminus, in view of our previous results. In that
study, a photochromic cross-linking agent bearing a spiro-
pyran skeleton was introduced between two Lys residues in
the N-terminal region in the i/i+7 positions.[10] The local
helix thus formed by the cross-linkage at the terminal was
able to provide total structural regulation of the target pep-
858
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 857 – 863

a-Helical Structures
FULL PAPER
spectra of native B (X=Orn)
and its cross-linked B·5. All of
the native peptides A–C adopt
almost random-coiled structures
(ꢀ20% helicities) as judged
from their CD spectra at 58C.
On the other hand, the cross-
linked peptides turned into
stable helices when the combi-
nation was preferable, as in the
case of B·5 (see Figures S2–S22
in the Supporting Information
for the spectra of other cross-
linked peptides). Most of the
cross-linked peptides include
chromophores absorbing at
ꢁ190 nm, the absorption bands
of which overlap with the CD
Figure 3. Cross-linking agents: acetylenic 1 and 2, short alkylene-based 3 and 4, benzene-based 5 and 6, biphe-
nylene-based 7–9, and highly rigid 10–12.
rigid 10–12 should serve for the peptides C. Furthermore, bi-
phenyl, naphthalene, phenanthrene, and fluorene cores are
fluorescent, so that peptides cross-linked with 7–12 could be
useful for analyzing biological recognition events such as
peptide–DNA interactions through their fluorescence.[13]
Preparation and CD spectra of cross-linked peptides: All of
the cross-linking agents were synthesized from the corre-
sponding dicarboxylic acids and N,N’-disuccinimidyl carbon-
ate. The dicarboxylic acids are commercially available or
easily synthesized. Short peptides A–D were prepared with
a peptide synthesizer. The cross-linking reactions and purifi-
cation were performed by the previously reported proce-
dures,[6] and the isolation yields were about 90–30% except
in a few cases. To elucidate the helical content of the cross-
linked peptides accurately, CD measurements must be car-
ried out below this concentration, thus at a concentration
sufficiently low that intermolecular association of the cross-
linked peptides is negligible. Therefore, the intermolecular
association tendencies of the cross-linked peptides were
studied in advance with C·12 with X=Lys, as the fluorene
core of 12 is one of the most hydrophobic of the cross-link-
ing agents tested and can self-associate in water. The UV
and CD spectra of C·12 with X=Lys were measured in
phosphate buffer (100 mm, pH 6.6) at 58C. The shapes of
the spectra are not sensitive to their concentrations, and at
ꢀ5.0”10À4 m the absorbances and ellipticities fit in propor-
tion to the concentration, obeying Beerꢁs law (Figure S1).
Contribution of intermolecular association of the cross-
linked peptides can thus be ruled out below that concentra-
tion. Taking this into account, all of the following measure-
ments were conducted at ca. 1.0”10À4 m concentrations of
the cross-linked peptides in phosphate buffer (100 mm,
pH 6.6) at 5 and 258C.
Figure 4. CD spectra of native and the cross-linked B (X=Orn) with 5
dissolved in phosphate buffer (100 mm, pH 6.6) at 5 and 258C: native B
(c) at 58C, B·5 at 58C (c), and B·5 at 258C (a).
active region arising from peptide backbones. Nevertheless,
isodichroic points were observed at 202 nm in the spectra of
the cross-linked peptides on varying the temperature from 5
to 608C, indicating that the chromophores do not interfere
with the CD region (see Figure S23 in the Supporting Infor-
mation).
Evaluation of helical content of the cross-linked peptides:
We evaluated the helical content of the cross-linked pep-
tides by means of their CD spectra. Helical content in repre-
sentative cross-linked peptides were calculated from the
mean residue ellipticity at 222 nm,[5d] the concentrations of
the solutions being determined by directly weighing the
samples on a microbalance. Table 1 shows the helical con-
tent (%) of the cross-linked peptides that showed outstand-
ing a-helicities (ꢁ60% except for A·4 with X=Orn). Other
combinations were roughly estimated for their a-helicities
by comparison of their CD spectra with those of the strictly
calculated ones and are marked as H (high), M (middle),
and L (low), corresponding to 60–40%, 40–20%, and
<20% helical content, respectively.
To identify effective cross-linking agents in each of the
peptide classes A–C, we measured the CD spectra of the
cross-linked peptides formed by various combinations be-
tween A–C and 1–12. Figure 4, for instance, shows the CD
Chem. Eur. J. 2008, 14, 857 – 863
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
859

K. Fujimoto, M. Inouye, and M. Kajino
Table 1. Helical content of peptides cross-linked with various cross-linking agents at 5 and 258C (in brackets).[a]
Cross-
linking
agents
Helical content
Peptides A (i/i+4: 0.5 nm)
X=
Peptides B (i/i+7: 1.1 nm)
X=
Peptides C (i/i+11: 1.6 nm)
X=
Dap
Dab
Orn
Lys
Dap
Dab
Orn
Lys
[b]
Dap
Dab
Orn
Lys
1
2
3
4
5
6
7
8
9
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
[b]
>95 (70)
90 (60)
–
–
–
–
90 (55)
M (L)
M (M)
L (L)
L (L)
–
–
–
–
–
–
L (L)
L (L)
L (L)
L (L)
–
–
–
–
–
–
60 (40)
50 (35)
L (L)
H (L)
–
–
–
–
–
–
M (L)
M (L)
L (L)
L (L)
–
–
–
–
–
–
L (L)
L (L)
M (L)
90 (60)
H (M)
M (L)
L (L)
L (L)
M (L)
L (L)
H (M)
L (L)
80 (60)
85 (60)
H (M)
M (M)
L (L)
M (L)
M (M)
L (L)
M (L)
M (M)
H (M)
H (M)
H (M)
H (M)
L (L)
M (L)
H (M)
L (L)
–
–
–
–
–
–
–
–
–
–
L (L)
–
–
–
–
–
–
–
–
–
–
M (M)
H (M)
M (L)
M (M)
H (H)
90 (70)
M (M)
85 (70)
H (H)
>95 (75)
H (M)
95 (70)
10
11
12
[a] Standard deviation is <6% for the number noted. “–”: No change or not measured. [b] See ref. [6].
i) Peptide class A (i/i+4—distance of 0.5 nm): The most ef-
fective cross-linking agent for the peptide class A was ethyl-
ene-based 3, with an appropriate spacer length (ca. 0.3–
cross-linked peptide: the peptide skeleton is too close to the
rigid benzene core of Dap.
À
À
0.4nm) between the two COON groups. In particular,
the cross-linked A·3 with X=Orn displayed a higher helical
content—of 60%—than any other combination at 58C and
kept ca. 40% of its helical structure even at 258C. This ob-
servation confirms that the whole helical structure in the
peptide of 13 residues would be influenced by the local
structural change at the N terminus through the trigger
effect described above. While other short cross-linking
agents—methylene-based 4 and m-phenylene-based 6—were
able to stabilize the helices somewhat, the cross-linking
agents with longer spacers never contributed to the stabiliza-
tion of the helices.
iii) Peptide class C (i/i+11—distance of 1.6 nm): Diacetylen-
ic 2 (spacer length: ca. 0.8–1.3 nm), naphthalene-based 10
(spacer length: ca. 0.8 nm), and fluorene-based 12 (spacer
length: ca. 0.9 nm) provided much higher helical content
AHCTRE(UNG >80%) in the cross-linked peptides C at 58C. In particular,
C·2 (X=Dab) and C·10 (X=Lys) were confirmed to exist in
almost fully helical states (>95%), as if the sequences of
the peptides had been incorporated into naturally occurring
proteins. We succeeded in constructing almost complete
helices from the short peptides, which comprise only ꢂ20%
helix in their native forms, with our cross-linking agents.
Even when the temperature was raised to 258C, the cross-
linked C·10 with X=Lys still sustained a high helical con-
tent (75%), offering promise that our method can be used
for biological application in vivo.
As speculated in the section on cross-linking agents, cross-
linkers that were rather short, in relation to the target dis-
tances between the two X residues, were found to be favora-
ble for effective stabilization of a-helices in the three short
peptides. On the basis of the data discussed above, the heli-
cal structures of the cross-linked peptides could roughly be
optimized in the cases in which the lengths of the cross-link-
ers were about 50–60% of the target pitches.
ii) Peptide class B (i/i+7—distance of 1.1 nm): We have re-
ported the use of the acetylenic cross-linking agents 1 and 2
for stabilizing a-helical structures of the peptide B (X=
Lys).[6] To explore the influence of the spacer lengths on the
helical content in detail, we examined all of the cross-linking
agents for peptide class B. Among them, p-phenylene-based
5 (spacer length: ca. 0.6 nm) and m-phenylene-based 6
(spacer length: ca. 0.5 nm) formed extremely stable helices
both at 58C (80–90%) and at 258C (60%). These values for
the helical content are rather higher than those reported
previously with use of the acetylenic 1 and 2. Although bi-
phenylene-based 9 (spacer length: ca. 1.0 nm) and fluorene-
based 12 (spacer length: ca. 0.9 nm) are approximately
fitted for bridging the i/i+7 pitch, the lengths of the alkyl
side chains in the cross-linked amino acids X should be con-
sidered as described above. For 9 and 12, the extra lengths
of the alkyl side chains might cause the cross-linked pep-
tides to be flexible, which should result in the formation of
relatively unstable helices. On the other hand, the phenyl-
ene-based 5 and 6 are likely more suitable in terms of their
slightly shorter and rigid spacers. Even in the case of B·5,
the alkyl side chain of Dap is too short to form a stable
Application to biologically important sequences: Finally, we
applied our strategy to biologically related sequences direct-
ed toward natural recognition events. As a target we select-
ed an arginine-rich sequence existing in the Rev protein that
binds to Rev responsive element (RRE) RNA in HIV repli-
cation. In view of the above observations with the peptides
C, the two arginine-rich sequences of D (X=Orn and Lys)
were cross-linked with 2 and 10. Figure 5 displays the greatly
increased helical content after the cross-linking reaction in
the case of D (X=Orn) with the cross-linking agent 2. Thus,
only ꢂ20% of native D fold up at 58C, whereas the cross-
860
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 857 – 863

a-Helical Structures
FULL PAPER
CDCl3): d=2.93 (s, 8H), 7.64(t, J=7.3 Hz, 2H), 7.93 (m, 2H), 8.18 (m,
2H), 8.38 ppm (t, J=1.6 Hz, 2H); 13C NMR (67.5 MHz, CDCl3): d=
169.1, 161.6, 140.4, 133.6, 130.1, 129.7, 129.2, 126.1, 25.7 ppm; IR (KBr):
n˜ =1774, 1736 cmÀ1
; HRMS (ESI): m/z: calcd for C22H16N2NaO8:
459.0804; found: 459.0803 [M+Na]+.
Biphenyl-2,2’-dicarboxylic acid bis(2,5-dioxopyrrolidin-1-yl) ester (8):
Yield 96% (420 mg); m.p. 225–2288C; 1H NMR (270 MHz, CDCl3): d=
2.77 (s, 8H), 7.33 (dd, J=1.4, 7.6 Hz, 2H), 7.52 (dt, J=1.1, 8.0 Hz, 2H),
7.66 (dt, J=1.4, 7.6 Hz, 2H), 8.21 ppm (dd, J=1.4, 7.8 Hz, 2H);
13C NMR (67.5 MHz, CDCl3): d=168.5, 161.1, 142.7, 133.3, 130.7, 130. 5,
127.8, 123.0, 25.5 ppm; IR (KBr): n˜ =1774, 1732 cmÀ1; HRMS (ESI): m/z:
calcd for C22H16N2NaO8: 459.0804; found: 459.0801 [M+Na]+.
Naphthalene-2,6-dicarboxylic acid bis(2,5-dioxopyrrolidin-1-yl) ester (10):
Yield 19% (38 mg); m.p. >2958C (decomp); 1H NMR (300 MHz,
DMSO-d6): d=2.95 (s, 8H), 8.20 (d, J=8.4Hz, 2H), 8.51 (d, J=8.4Hz,
2H), 9.03 ppm (s, 2H); 13C NMR (75 MHz, DMSO-d6): d=169.8, 161.2,
Figure 5. CD spectra of the native and the cross-linked D (X=Orn) with
2 dissolved in phosphate buffer (100 mm, pH 6.6) at 5 and 258C: native
D (c) at 58C, D·2 at 58C (c), and D·2 at 258C (a).
134.5, 131.8, 131.0, 125.5, 124.5, 25.5 ppm; IR (KBr): n˜ =1771, 1738 cmÀ1
;
HRMS (ESI): m/z: calcd for C20H14N2NaO8: 433.0648; found: 433.0646
[M+Na]+.
linked D·2 was found to exist 95% in the helical state at the
same temperature. The cross-linked D·2 still kept 60% a-
helical structure at 258C, close to the temperature range
covering artificial experiments. This finding implies that the
cross-linked D·2 might be applicable in artificial segments
with affinity for naturally occurring RRE RNA.
Phenanthrene-3,6-dicarboxylic acid bis(2,5-dioxopyrrolidin-1-yl) ester
(11): Yield 26% (12 mg); m.p. 121–1248C; 1H NMR (300 MHz, CDCl3):
d=2.98 (s, 8H), 7.96 (s, 2H), 8.04(d, J=8.4Hz, 2H), 8.31 (dd, J=1.5,
8.4Hz, 2H), 9.49 ppm (d, J=1.5 Hz, 2H); 13C NMR (75 MHz, CDCl3):
d=25.7, 123.5, 126.7, 127.5, 129.3 129.5, 129.6, 135.8, 161.6, 168.3,
168.9 ppm; IR (KBr): n˜ =1770, 1736 cmÀ1; HRMS (ESI): m/z: calcd for
C24H16N2NaO8: 483.0804; found: 483.0803 [M+Na]+.
Fluorene-2,7-dicarboxylic acid bis(2,5-dioxopyrrolidin-1-yl) ester (12):
Yield 98% (47 mg); m.p. >2748C (dec); 1H NMR (270 MHz, CDCl3):
d=2.94(s, 8H), 4.08 (s, 2H), 7.99 (d, J=4.2 Hz, 2H), 8.24 (d, J=4.2 Hz,
2H), 8.38 ppm (s, 2H); 13C NMR (67.5 MHz, CDCl3): d=169.2, 161.9,
148.0, 144.7, 130.1, 127.5, 124.7, 121.4, 36.9, 25.7 ppm; IR (KBr): n˜ =1774,
1736 cmÀ1; HRMS (ESI): m/z: calcd for C23H16N2NaO8: 471.0804; found:
471.0805 [M+Na]+.
Conclusion
We have developed a series of cross-linking agents and have
found combinations of cross-linking agents and short pep-
tides suitable for effective stabilization of helical structures
in the cross-linked states. The best combinations were ethyl-
ene-based 3 for peptide class A (i/i+4 ; X=Orn), m-phenyl-
ene-based 6 for B (i/i+7; X=Dab), and naphthalene-based
10 for C (i/i+11; X=Lys) at both 5 and 258C. In particular,
the cross-linked C·10 showed >95% helical content at 58C
and 75% even at 258C. Furthermore, we were able to
obtain stable helical structures in peptide D (i/i+11; X=
Orn and Lys), an arginine-rich motif from natural Rev. In
future investigations, the cross-linking agents identified in
this study will be widely applied to regulation and inhibition
of biological recognition events in which a-helices partici-
pate.
Solid-phase peptide synthesis (SPPS): All of the peptides were synthe-
sized with an automated peptide synthesizer by standard Fmoc chemistry.
Peptides were constructed on an Fmoc-NH-SAL resin (capacity
0.59 mmolgÀ1). After the automated SPPS, N-terminal amino groups
were acetylated with Ac2O (5%) in NMP (or 2% N,N-diisopropylamine,
9% Ac2O, and 1.9% HOBt·H2O in DMF for peptides D) over 12 min
(30 min for peptides D) at room temperature. Peptide cleavage and side
chain deprotection of amino acids were carried out by treatment under
suitable conditions: TFA/ethanedithiol/thioanisole 18:1:1 for peptides A
and C, TFA/ethanedithiol/thioanisole/2-methylindole 90:5:5:0.1 for pep-
tides B, and TFA/ethanedithiol/thioanisole/thiophenol/DMSO/H2O
83:2.5:5:2:3 for peptides D over 2 h (12 h for peptides D) at room tem-
perature.
Peptide purification: Peptides A–D were purified by reversed-phase
HPLC (column; COSMOSIL 5C18-AR-300 nacalai tesque, 10”150 mm)
and eluted with TFA buffer (0.1%) and the following CH3CN (including
0.1% TFA) linear gradients at a flow rate of 2.0 mLminÀ1; 5–45% (0–
40 min) for peptides A, 10–50% (0–40 min) for peptides B, 15–55% (0–
40 min) for peptides C, and 5–45% (0–40 min) for peptides D. The frac-
tions of the peptides were monitored at 220 nm with a UV detector, and
were identified by ESI-MS and MALDI TOF MS.
Experimental Section
Materials and general procedures: The cross-linking agents 3 and 5 were
commercially available, while 4,[14] 6,[15] and 9[16] had been reported previ-
ously. Other cross-linking agents were prepared from their corresponding
dicarboxylic acids by the same procedure as described for 1 and 2.[6] The
dicarboxylic acid precursors for 7, 8, and 10 were commercially available,
and those for 11 and 12 were synthesized by published procedures.[17,18]
NMR spectra were recorded on a JEOL FX-270 or a Varian Gemini 300
spectrometer. IR spectra were measured on a JASCO-FT/IR-460 plus
spectrometer. MALDI TOF-MS spectra were obtained by use of a
Brüker Autoflex mass spectrometer. ESI-HRMS analyses were carried
out on a JEOL JMS-T100 LC mass spectrometer. Melting points were
determined with a Yanako MP-500D instrument and were not corrected.
Peptides A: X=Dap: m/z: calcd for C49H81N16O22: 1245.6; found: 1245.1
[M+H]+ (ESI); X=Dab: m/z: calcd for C51H85N16O22: 1273.6; found:
1273.5 [M+H]+ (MALDI); X=Orn: m/z: calcd for C53H89N16O22: 1301.6;
found: 1301.5 [M+H]+ (MALDI); X=Lys: m/z: calcd for C55H93N16O22
1329.7; found: 1329.5 [M+H]+ (MALDI).
:
Peptides B: X=Dap: m/z: calcd for C66H102N20O25: 787.4; found: 787.2
[M+2H]2+ (ESI); X=Dab: m/z: calcd for C68H106N20O25: 801.4; found:
801.3 [M+2H]2+ (ESI); X=Orn: m/z: calcd for C70H110N20O25: 815.4;
found: 815.1 [M+2H]2+ (ESI); X=Lys: m/z: calcd for C72H114N20O25
829.4; found: 829.2 [M+2H]2+ (ESI).
:
Physical and spectroscopic data for cross-linking agents
Peptides C: X=Dab: m/z: calcd for C74H123N23O31: 914.9; found: 914.6
[M+2H]2+ (ESI); X=Orn: m/z: calcd for C76H126N23O31: 1856.9; found:
Biphenyl-3,3’-dicarboxylic acid bis(2,5-dioxopyrrolidin-1-yl) ester (7):
Yield 71% (15 mg); m.p. >1658C (decomp); 1H NMR (270 MHz,
Chem. Eur. J. 2008, 14, 857 – 863
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
861

K. Fujimoto, M. Inouye, and M. Kajino
1856.9 [M+H]+ (MALDI); X=Lys, calcd for C78H130N23O31: 1884.9;
found: 1884.9 [M+H]+ (MALDI).
C61H90N16NaO24: 1453.6; found: 1453.1 [M+Na]+ (MALDI); X=Lys:
m/z: calcd for C63H94N16NaO24
:
1481.7; found: 1481.1 [M+Na]+
(MALDI).
Peptides D: X=Orn: m/z: calcd for C101H196N57O22: 2559.6; found: 2559.3
[M+H]+ (MALDI); X=Lys, calcd for C103H201N57O22: 1294.3; found:
1294.0 [M+2H]2+ (ESI).
Cross-linked peptide B·3: X=Dab: m/z: calcd for C72H108N20O27: 842.4;
found: 842.3 [M+2H]2+ (ESI); X=Orn: m/z: calcd for C74H111N20O27:
1711.8; found: 1711.1 [M+H]+ (MALDI); X=Lys: m/z: calcd for
C76H116N20O27: 870.4; found: 870.0 [M+2H]2+ (ESI).
Cross-linking reactions—Peptides A–C: A cross-linking agent solution in
DMSO (0.5 mL, 5.0”10À4 m) was added to a peptide solution (0.5 mL,
1.0”10À4 m) in phosphate buffer (100 mm, pH 6.6). The reaction mixture
was incubated at 258C in a thermo-mixer for 0.5–2 h. The cross-linked
peptides A–C were purified by use of the following CH3CN (including
0.1% TFA) linear gradients (0–40 min) at a flow rate of 1.0 mLminÀ1; 5–
65% for cross-linked peptides A and 10–70% for cross-linked peptides B
and C.
Cross-linked peptide B·4: X=Dab: m/z: calcd for C73H110N20O27: 849.4;
found: 849.1 [M+2H]2+ (ESI); X=Orn: m/z: calcd for C75H113N20O27:
1725.8; found: 1725.7 [M+H]+ (MALDI); X=Lys: m/z: calcd for
C77H116N20NaO27: 1775.8; found: 1775.4[ M+Na]+ (MALDI).
Cross-linked peptide B·5: X=Dap: m/z: calcd for C74H104N20O27: 852.4;
found: 852.0 [M+2H]2+ (ESI); X=Dab: m/z: calcd for C76H108N20O27:
866.4; found: 866.2 [M+2H]2+ (ESI); X=Orn: m/z: calcd for
C78H112N20O27: 880.4; found: 880.6 [M+2H]2+ (ESI); X=Lys: m/z: calcd
for C80H114N20NaO27: 1809.8; found: 1809.6 [M+Na]+ (MALDI).
Peptides D: A cross-linking agent solution in DMSO (0.4mL, 4.17”
10À4 m) was added to an EtOH solution (0.6 mL) of a native peptide D
(1.25”10À4 m). The reaction mixture was stirred at 308C in a thermo-
mixer for 12 h. The cross-linked peptides D were isolated under condi-
tions similar to those described above, except for use of a 0–60%
CH3CN (including 0.1% TFA) linear gradient. All of the cross-linked
peptides were identified by ESI-MS and MALDI TOF-MS.
Cross-linked peptide B·6: X=Dab: m/z: calcd for C76H108N20O27: 866.4;
found: 866.2 [M+2H]2+ (ESI); X=Orn: m/z: calcd for C78H112N20O27:
880.4; found: 880.6 [M+2H]2+ (ESI); X=Lys: m/z: calcd for
C80H116N20O27: 894.4; found: 894.0 [M+2H]2+ (ESI).
Yields of cross-linked peptides: The reaction yields of the cross-linked
peptides were estimated by comparison of the peak areas of the remain-
ing native peptides with those of the cross-linked peptides in the HPLC
charts of the reaction mixtures. The molar ratios for the unreacted and
the cross-linked peptides were calculated by calibration of their molar ex-
tinction coefficients (e) at 220 nm. The e values of the cross-linked pep-
tides were each defined to be the sum of the values for the native peptide
and the cross-linking agent at 220 nm.
Cross-linked peptide B·7: X=Dab: m/z: calcd for C82H112N20O27: 904.4;
found: 904.1 [M+2H]2+ (ESI); X=Orn: m/z: calcd for C84H116N20O27:
918.4; found: 918.0 [M+2H]2+ (ESI); X=Lys: m/z: calcd for
C86H120N20O27: 932.4; found: 932.0 [M+2H]2+ (ESI).
Cross-linked peptide B·8: X=Dab: m/z: calcd for C82H111N20O27: 1807.8;
found: 1807.5 [M+H]+ (MALDI); X=Orn: m/z: calcd for C84H116N20O27:
918.4; found: 918.0 [M+2H]2+ (ESI); X=Lys: m/z: calcd for
C86H120N20O27: 932.4; found: 932.0 [M+2H]2+ (ESI).
Cross-linked peptides: A·3: X=Dap, 85%; X=Dab, 70%; X=Orn,
75%; X=Lys, 85%. A·4: X=Dap, 10%; X=Dab, 70%; X=Orn, 55%;
X=Lys, 55%. A·5: X=Dap, 20%; X=Dab, 45%; X=Orn, 60%; X=
Lys, 75%. A·6: X=Dap, 50%; X=Dab, 75%; X=Orn, 75%; X=Lys,
70%. B·3: X=Dab, 85%; X=Orn, 85%; X=Lys, 90%. B·4: X=Dab,
40%; X=Orn, 45%; X=Lys, 35%. B·5: X=Dap, 15%; X=Dab, 45%;
X=Orn, 50%; X=Lys, 40 %. B·6: X=Dab, 75%; X=Orn, 80%; X=
Lys, 85%. B·7: X=Dab, 40%; X=Orn, 40%; X=Lys, 40 %. B·8: X=
Dab, 30%; X=Orn, 15%; X=Lys, 30%. B·9: X=Dab, 15%; X=Orn,
20%; X=Lys, 15%. B·10: X=Dab, 20%; X=Orn, 30%; X=Lys, 30%.
B·11: X=Dab, 55%; X=Orn, 45%; X=Lys, 55%. B·12: X=Dab, 55%;
X=Orn, 70%; X=Lys, 40 %. C·2: X=Dab, 85%; X=Orn, 65%; X=
Lys, 60%. C·9: X=Dab, 65%; X=Orn, 45%; X=Lys, 30%. C·10: X=
Dab, 60%; X=Orn, 35%; X=Lys, 35%. C·11: X=Dab, 65%; X=Orn,
55%; X=Lys, 65%. C·12: X=Dab, >95%; X=Orn, 90%; X=Lys,
>95%. D·2: X=Orn, 15%; X=Lys, 30%. D·10: X=Orn, <5%; X=
Lys, <5%.
Cross-linked peptide B·9: X=Dab: m/z: calcd for C82H111N20O27: 1807.8;
found: 1807.6 [M+H]+ (MALDI); X=Orn: m/z: calcd for C84H116N20O27:
918.4; found: 918.0 [M+2H]2+ (ESI); X=Lys: m/z: calcd for
C86H120N20O27: 932.4; found: 932.0 [M+2H]2+ (ESI).
Cross-linked peptide B·10: X=Dab: m/z: calcd for C80H110N20O27: 891.4;
found: 891.2 [M+2H]2+ (ESI); X=Orn: m/z: calcd for C82H114N20O27:
905.4; found: 905.2 [M+2H]2+ (ESI); X=Lys: m/z: calcd for
C84H118N20O27: 919.4; found: 919.1 [M+2H]2+ (ESI).
Cross-linked peptide B·11: X=Dab: m/z: calcd for C84H111N20O27: 1831.8;
found: 1831.5 [M+H]+ (MALDI); X=Orn: m/z: calcd for
C86H114N20NaO27: 1881.8; found: 1881.7 [M+Na]+ (MALDI); X=Lys:
m/z: calcd for C88H118N20NaO27: 1909.8; found: 1909.7 [M+Na]+
(MALDI).
Cross-linked peptide B·12: X=Dab: m/z: calcd for C83H111N20O27: 1819.8;
found: 1819.4[ M+H]+ (MALDI); X=Orn: m/z: calcd for C85H115N20O27:
1847.8; found: 1847.5 [M+H]+ (MALDI); X=Lys: m/z: calcd for
C87H118N20NaO27: 1897.8; found: 1897.4[ M+Na]+ (MALDI).
Mass spectral data for cross-linked peptides
Cross-linked peptide A·3: X=Dap: m/z: calcd for C53H83N16O24: 1327.6;
found: 1327.8 [M+H]+ (MALDI); X=Dab: m/z: calcd for
C55H86N16NaO24: 1377.6; found: 1377.8 [M+Na]+ (MALDI); X=Orn:
Cross-linked peptide C·2: X=Dab: m/z: calcd for C84H129N23O35: 1010.0;
found: 1009.7 [M+2H]2+ (ESI); X=Orn: m/z: calcd for C86H132N23O35
:
2046.9; found: 2047.1 [M+H]+ (MALDI); X=Lys: m/z: calcd for
C88H136N23O35: 2075.0; found: 2075.0 [M+H]+ (MALDI).
m/z: calcd for C57H90N16NaO24
:
1405.6; found: 1405.1 [M+Na]+
(MALDI); X=Lys: m/z: calcd for C59H94N16NaO24: 1433.7; found: 1433.1
[M+Na]+ (MALDI).
Cross-linked peptide C·9: X=Dab: m/z: calcd for C88H127N23NaO33
:
2056.9; found: 2057.2 [M+Na]+ (MALDI); X=Orn: m/z: calcd for
C90H132N23O33: 2062.9; found: 2062.5 [M+H]+ (MALDI); X=Lys: m/z:
calcd for C92H135N23NaO35: 2112.9; found: 2112.8 [M+Na]+ (MALDI).
Cross-linked peptide A·4: X=Dap: m/z: calcd for C54H84N16NaO24
:
1363.6; found: 1363.8 [M+Na]+ (MALDI); X=Dab: m/z: calcd for
C56H88N16NaO24: 1391.6; found: 1391.6 [M+Na]+ (MALDI); X=Orn:
:
1419.6; found: 1419.4 [M+Na]+
Cross-linked peptide C·10: X=Dab: m/z: calcd for C86H127N23O33
:
m/z: calcd for C58H92N16NaO24
1004.9; found: 1004.7 [M+2H]2+ (ESI); X=Orn: m/z: calcd for
C88H129N23NaO33: 2058.9; found: 2059.2 [M+Na]+ (MALDI); X=Lys:
m/z: calcd for C90H134N23O35: 2065.0; found: 2064.6 [M+H]+ (MALDI).
(MALDI); X=Lys: m/z: calcd for C60H96N16NaO24: 1447.7; found: 1447.3
[M+Na]+ (MALDI).
Cross-linked peptide A·5: X=Dap: m/z: calcd for C57H83N16O24: 1375.6;
found: 1375.4[ M+H]+ (ESI); X=Dab: m/z: calcd for C59H86N16NaO24
:
Cross-linked peptide C·11: X=Dab: m/z: calcd for C90H128N23O33: 2058.9;
found: 2058.8 [M+H]+ (MALDI); X=Orn: m/z: calcd for
C92H132N23O33: 2086.9; found: 2086.7 [M+H]+ (MALDI); X=Lys: m/z:
calcd for C94H136N23O33: 2115.0; found: 2115.2 [M+H]+ (MALDI).
1425.6; found: 1425.6 [M+Na]+ (MALDI); X=Orn: m/z: calcd for
C61H90N16NaO24: 1453.6; found: 1453.6 [M+Na]+ (MALDI); X=Lys:
m/z: calcd for C63H94N16NaO24
:
1481.7; found: 1481.7 [M+Na]+
(MALDI).
Cross-linked peptide C·12: X=Dab: m/z: calcd for C89H127N23NaO33
:
2068.9; found: 2068.8 [M+Na]+ (MALDI); X=Orn: m/z: calcd for
C91H131N23NaO33: 2096.9; found: 2096.6 [M+Na]+ (MALDI); X=Lys:
m/z: calcd for C93H136N23O33: 2103.0; found: 2102.6 [M+H]+ (MALDI).
Cross-linked peptide A·6: X=Dap: m/z: calcd for C57H83N16O24: 1375.6;
found: 1375.4[ M+H]+ (ESI); X=Dab: m/z: calcd for C59H86N16NaO24
1425.6; found: 1425.4 [M+Na]+ (MALDI); X=Orn: m/z: calcd for
:
862
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 857 – 863

a-Helical Structures
FULL PAPER
Cross-linked peptide D·2: X=Orn: m/z: calcd for C111H202N57O26: 2749.6;
found: 2749.2 [M+H]+ (MALDI); X=Lys: m/z: calcd for
C113H206N57O26: 2777.7; found: 2778.2 [M+H]+ (MALDI).
Res. 2005, 38, 486–493; c) D. J. Cline, C. Thorpe, J. P. Schneider, J.
Am. Chem. Soc. 2003, 125, 2923–2929; d) C. E. Schafmeister, J. Po,
G. L. Verdine, J. Am. Chem. Soc. 2000, 122, 5891–5892.
[6] K. Fujimoto, N. Oimoto, K. Katsuno, M. Inouye, Chem. Commun.
2004, 1280–1281.
[7] a) A. K. Sato, M. Viswanathan, R. B. Kent, C. R. Wood, Curt. Opin.
Biotechnol. 2006, 17, 638–642; b) D. J. Barlow, In Smith and Wil-
liams’ Introduction to the Principles of Drug Design and Action
(Ed.: H. J. Smith), CRC Press, Boca Raton, 4th ed., 2006, pp. 327–
354; c) J. A. Kritzer, O. M. Stephens, D. A. Guarracino, S. K.
Reznik, A. Schepartz, Bioorg. Med. Chem. 2005, 13, 11–16.
[8] a) M. L. Zapp, M. R. Green, Nature 1989, 342, 714–716; b) T. J.
Daly, K. S. Cook, G. S. Gray, T. E. Maione, J. R. Rusche, Nature
1989, 342, 816–819.
Cross-linked peptide D·10: X=Orn: m/z: calcd for
C113H200N57O24:
2739.6; found: 2739.2 [M+H]+ (MALDI); X=Lys: m/z: calcd for
C115H204N57O24: 2767.6; found: 2767.6 [M+H]+ (MALDI).
Acknowledgement
This work was partly supported by the Sasakawa Scientific Research
Grant from The Japan Science Society.
[9] Recent papers for Rev–RRE interactions: See refs. [1c,4a].
[10] K. Fujimoto, M. Amano, Y. Horibe, M. Inouye, Org. Lett. 2006, 8,
285–287.
[1] Recent reviews: a) N. Errington, T. Iqbalsyah, A. J. Doig, In Protein
Design: Methods and Applications (Methods in Molecular Biology)
(Eds.: R. Guerois, M. Lopez De La Paz), Humana Press, Totowa,
2006, pp. 3–26; b) A. J. Doig, N. Errington, T. Iqbalsyah, In Protein
Folding Handbook, Vol. 1 (Eds.: J. Buchner, T. Kiefhaber), Wiley-
VCH, Weinheim, 2005, pp. 247–313; c) G. Licini, L. J. Prins, P. Scri-
min, Eur. J. Org. Chem. 2005, 969–977.
[2] a) D. Wang, K. Chen, J. L. Kulp, III, P. S. Arora, J. Am. Chem. Soc.
2006, 128, 9248–9256; b) D. Wang, W. Liao, P. S. Arora, Angew.
Chem. 2005, 117, 6683–6687; Angew. Chem. Int. Ed. 2005, 44, 6525–
6529; c) R. N. Chapman, G. Dimartino, P. S. Arora, J. Am. Chem.
Soc. 2004, 126, 12252–12253.
[3] a) S. Hyun, H. J. Kim, N. J. Lee, K. H. Lee, Y. Lee, D. R. Ahn, K.
Kim, S. Jeong, J. Yu, J. Am. Chem. Soc. 2007, 129, 4514–4515; b) M.
Bellanda, S. Mammi, S. Geremia, N. Demitri, L. Randaccio, Q. B.
Broxterman, B. Kaptein, P. Pengo, L. Pasquato, P. Scrimin, Chem.
Eur. J. 2007, 13, 407–416; c) A. Barazza, A. Wittelsberger, N. Fiori,
E. Schievano, S. Mammi, C. Toniolo, J. M. Alexander, M. Rose-
nblatt, E. Peggion, M. Chorev, J. Pept. Res. 2005, 65, 23–35.
[4] a) J. C. Lin, B. Barua, N. H. Andersen, J. Am. Chem. Soc. 2004, 126,
13679–13684; b) J. W. Neidigh, R. M. Fesinmeyer, N. H. Andersen,
Nat. Struct. Biol. 2002, 9, 425–430.
[11] J. L. Battiste, H. Mao, N. S. Rao, R. Tan, D. R. Muhandiram, L. E.
Kay, A. D. Frankel, J. R. Williamson, Science 1996, 273, 1547–1551.
[12] A reference paper for the sequence of peptide D: R. Tan, A. D.
Frankel, Proc. Natl. Acad. Sci. USA 1998, 95, 4247–4252.
[13] Since the cross-liking agents 7–12 include the fluorescent cores, we
measured the fluorescence spectra of the peptides C cross-linked
with 7–12. Figure S24in the Supporting Information shows the exci-
tation and emission spectra of cross-linked C·12 with X=Lys as rep-
resentative.
[14] R. J. Bergeron, G. W. Yao, G. W. Erdos, S. Milstein, F. Gao, W. R.
Weimer, O. Phanstiel IV, J. Am. Chem. Soc. 1995, 117, 6658–6665.
[15] O. Safarowsky, E. Vogel, F. Vçgtle, Eur. J. Org. Chem. 2000, 499–
505.
[16] Q. Chen, D. A. Sowa, J. Cai, R. Gabathuler, Synth. Commun. 2003,
33, 2401–2421.
[17] S. M. Langenegger, R. Hꢂner, Helv. Chim. Acta 2002, 85, 3414–
3421.
[18] M. Dotrong, R. C. Evers, Polym. Mater. Sci. Eng. 1989, 60, 507–511.
Received: June 4, 2007
Revised: September 21, 2007
Published online: October 30, 2007
[5] a) N. L. Mills, M. D. Daugherty, A. D. Frankel, R. K. Guy, J. Am.
Chem. Soc. 2006, 128, 3496–3497; b) G. A. Woolley, Acc. Chem.
Chem. Eur. J. 2008, 14, 857 – 863
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
863