Highly efficient and directional homo-and heterodimeric energy transfer materials based on fluorescently derivatized α,γ(γ)-cyclic octapeptides

Article abstract of DOI:10.1002/asia.201000545

Cyclic octapeptides composed of α-amino acids alternated with cis-3-aminocycloalkanecarboxylic acids, self-assemble as drumlike dimers through β-sheet-like, backbone-to-backbone hydrogen bonding. Heterodimerization appears to be significantly more favored

Full text of DOI:10.1002/asia.201000545

FULL PAPERS
DOI: 10.1002/asia.201000545
Highly Efficient and Directional Homo- and Heterodimeric Energy Transfer
Materials Based on Fluorescently Derivatized a,g-Cyclic Octapeptides
Roberto J. Brea,[a] Marꢀa Jesffls PØrez-Alvite,[a] Michele Panciera,[a] Manuel Mosquera,[b]
Luis Castedo,[a] and Juan R. Granja*[a]
Abstract: Cyclic octapeptides com-
posed of a-amino acids alternated with
cis-3-aminocycloalkanecarboxylic acids,
self-assemble as drumlike dimers
through b-sheet-like, backbone-to-
backbone hydrogen bonding. Heterodi-
merization appears to be significantly
more favored than homodimerization,
and this represents a novel approach
for the design and fabrication of highly
multicomponent equilibrium network
based on fluorescently derivatized self-
assembling a,g-cyclic octapeptides has
been successfully used to form light-
harvesting/light-converting ensembles
with a distinctive organization of donor
and acceptor units able to act as effi-
cient artificial photosystems.
stable heterodimeric assemblies.
A
Keywords: amino acids · cyclic pep-
tides · FRET · heterodimers · self-
assembly
Introduction
one of the most powerful “bottom-up” techniques to mimic
such orderliness for the preparation of well-organized opto-
electronic materials.[3] However, a suitable structural matrix
that allows both orientation and proximity between donor
and acceptor units is required to obtain the appropriate spa-
tial organization for efficient energy transfer. Indeed, with-
out such an arrangement, fine control of energy transfer is
not possible. Porous materials built of a solid-state array of
self-assembled peptide nanotubes (SPNs), for which the
homo- and/or heterodimerization can be easily controlled,
would be of great interest.[4] Self-assembly of biomolecules
that are capable of selectively forming either homo- and/or
heterodimeric entities represents one of the great challenges
today because, in general, current procedures do not provide
sufficient recognition power for the selective assembly in
either homo- or heteromeric fashion.[5] In nature, protein-
subunit interactions (either homodimer or heterodimer) are
an important phenomena in regulation and catalysis and
appear to constitute an economical means of achieving
structural diversity and functional versatility.[6] One would
expect that the same goals may be achieved by the develop-
ment of synthetic assemblers that are capable of selectively
forming homo- and/or heterodimeric species. Herein, we
report the controlled fabrication of homo- and heterodi
meric light-harvesting nanobiomaterials through the self-or-
ganization of cyclic peptides (CPs). These systems can act as
platforms for efficient energy transfer by providing an op
timized spatial distribution.
Natural photosynthesis provides an efficient example of the
hierarchical organization in space of a multitude of chromo-
phoral molecules in space for energy transfer.[1,2] Self-assem-
bly processes, in which individual components associate
spontaneously in a predetermined fashion, have emerged as
[a] R. J. Brea, M. J. Pꢀrez-Alvite, M. Panciera, Prof. Dr. L. Castedo,
Prof. Dr. J. R. Granja
Departamento de Quꢁmica Orgꢂnica y
Unidad Asociada al C.S.I.C. Centro Singular de
Investigaciꢃn en Quꢁmica Biolꢃgica y Materiales Moleculares
Campus Vida
Universidad de Santiago
15782 Santiago de Compostela (Spain)
Fax : (+34)981-595012
E-mail: juanr.granja@usc.es
[b] Prof. Dr. M. Mosquera
Departamento de Quꢁmica Fꢁsica y
Centro Singular de Investigaciꢃn en
Quꢁmica Biolꢃgica y Materiales Moleculares
Campus Vida
Universidad de Santiago
15782 Santiago de Compostela (Spain)
The Supporting Information for this article contains Figures SI-1–12,
the H and 13C NMR, NOESY and/or ROESY, and FTIR spectra of
1
peptides 9, 10a, 10b, 10c, 11a, 11b, 11c, 12a, 12c, 12d, and 12 f, to-
1
gether with the H NMR spectra of compounds 13 and 14, the
1H NMR, NOESY, and/or ROESY spectra of heterodimers D9a/10a
and D11a/12a, and the UV spectra of peptides 11c, 12c, and 12 f and is
available on the WWW under http://dx.doi.org/10.1002/
asia.201000545.
As part of an extensive program directed towards the
design, synthesis, and structural and functional evaluation of
110
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 110 – 121

peptide nanotubes,[4] we recently developed a novel class of
CPs in which a-amino acids are alternated with cyclic g-
amino acids [such as g-Ach (3-aminocyclohexanecarboxylic
acid, Scheme 1, 1–3) or g-Acp (3-aminocyclopentanecarbox-
tive heterodimer formation between cyclic a,g-octapeptides
based on g-Ach residues with those containing g-Acp. In
these systems the preferential heterodimer formation is,
once again, driven by the backbone–backbone interactions,
as well as by the entropy of mixing (DSmix).[11,14] Although
the DSmix is expected to increase, the heterodimer formation
should not alter significantly the overall trend in association
constants.
Results and Discussion
Cyclic octapeptides cyclo[(d-Leu-l-MeN-g-Acp)4-] (9),[7b]
cyclo[(d-Phe-l-MeN-g-Ach)4-]
(10a),[7c]
and
cycloACHTUNGTRENNUNG[(d-
Ser(Bn)-l-MeN-g-Ach)4-] (10b) were synthesized first
(Scheme 2). The 1H NMR spectra of these compounds in
nonpolar solvents reflect the typical features of homodimer
formation (D9, D10a, and D10b, respectively), for example,
coupling constants (JNH,Ha =9.5, 8.6, and 8.4 Hz for 9, 10a,
and 10b, respectively) and the downfield shift of the NH
signal of a-Aa (d=8.33, 8.57, and 8.41 ppm for 9, 10a, and
10b, respectively). The FTIR spectra in chloroform show
the characteristic bands of b-sheet-like hydrogen bond-
ing.[7,15] In all of these samples, the association constant for
homodimer formation must be greater than 105 mꢀ1, based
Scheme 1. a,g-Cyclopeptides (1–8) and their corresponding dimer models
(D1–D8). For clarity, amino acid side chains have been omitted from the
representations of the nanotube and dimers.
ꢀ
on the fact that the location of the N H signals remains
constant at concentrations as low as 1ꢅ10ꢀ4 m, regardless of
whether the solvent contained 30% methanol or the sam-
ples were heated at 323 K.
ylic acid, Scheme 1, 4–8)], to give systems with large homo-
dimerization association constants.[7–9] Of particular interest
is the preferential heterodimer formation between g-Ach-
based and g-Acp-based cyclic a,g-hexapeptides, in which the
selective formation of the heterodimer is driven by back-
bone-to-backbone hydrogen-bonding interactions.[10–12] This
is a very opportune property, because it allows side chain
modifications to be exploited for other purposes without af-
fecting the self-assembly properties, as exemplified by our
recent findings on electron and/or energetic transfer pro-
cesses.[13] Herein, we report a further extension of the selec-
The possibility of heterodimer formation was followed
and confirmed by the successive additions of 10a (from zero
to 1.7 equivalents) to a chloroform solution of 9 (1.7 mm).
This experiment resulted in the emergence of a new set of
1
signals in the H NMR spectrum that did not correspond to
either of the possible homodimers (Figure 1 and Figure SI-1
in the Supporting Information). The dimeric nature of the
new species, D9/10a, was confirmed by nuclear Overhauser
effect (nOe) cross-peaks between the signals of HgAcp (d=
4.96 ppm) and HaAch (d=2.67 ppm), and also HgAch (d=
4.33 ppm) and HaAcp (d=3.17 ppm).
To determine the interactions (backbone versus side-chain
interactions) responsible for the preferential formation of
Abstract in Spanish: Los ciclooctapꢀptidos compuestos de
a-aminoꢂcidos alternados con g-aminoꢂcidos cꢁclicos (ꢂcidos
cis-3-aminocicloalcanocarboxꢁlicos) dan lugar a dꢁmeros ci-
lꢁndricos mediante un proceso de autoensamblaje molecular
mediante la formaciꢃn de una hoja plegada b. La formaciꢃn
de los heterodꢁmeros es mꢂs favorable que la formaciꢃn de
los correspondientes homodꢁmeros y esta discriminaciꢃn se
debe a las interacciones entre los esqueletos peptꢁdicos y no
al tipo de a-aminoꢂcido que componen los cꢁclopꢀptide.
Esto permite la preparaciꢃn de una red de equilibrios de au-
toensablaje molecular entre distintos componentes ciclopetꢁ-
dicos marcados con fluorꢃforos que se han utilizado en el
diseÇo de complejos de captura y conversiꢃn de luz median-
te la adecuada disposiciꢃn de las unidades dadoras y acepto-
ras que podrꢁa tener utilidad en el diseÇo de fotosistemas ar-
tificiales eficientes.
heterodimeric assemblies, cycloACHTUNGTRNEUNG
{[d-Ser(Bn)-l-MeN-g-Ach]4-}
(10b) was added to CP 9 to study the formation of the cor-
responding heterodimer D9/10b (Scheme 2). Any significant
change in the heterodimer formation with respect to previ-
ous experiments could be attributed to the differences be-
tween the side-chain cross-strand interactions. Analysis of
1
the H NMR spectrum of the 1:1 mixture of the two pep-
tides in chloroform again reveals the preferential formation
of heterodimer D9/10b (see Figure SI-2 in the Supporting In-
formation). This result suggests that the selective formation
of the heterodimer is mainly driven by the backbone–back-
bone hydrogen-bonding interactions between the Acp- and
the Ach-containing CPs in which the better complementari-
ty between the donors and acceptors of the two CPs induces
the preferred formation of the heterodimeric form.[10,14]
Chem. Asian J. 2011, 6, 110 – 121
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
111

J. R. Granja et al.
FULL PAPERS
corresponding homodimer D10c
with a large association con-
stant. Addition of one equiva-
lent of
a
solution of 10c
(8.4 mm) in methanol to
a
chloroform solution of
9
(1.7 mm) resulted, once again,
in the predominant formation
of the heterodimeric species
(D9/10c) (Figure SI-3 in the Sup-
porting Information) confirm-
ing the predominant effect of
backbone–backbone
interac-
tions over side chain-side chain
interactions.
To investigate whether self-
assembly might be significantly
affected by interactions be-
tween amino acid side chains
on adjacent CPs in more com-
plex systems, new octapeptides
cyclo{[d-Phe-l-MeN-g-Ach-d-
Ser(Bn)-l-MeN-g-Ach]2-} (11a),
cyclo{[d-Phe-l-MeN-g-Acp-d-
Ser(Bn)-l-MeN-g-Acp]2-} (12a),
and cyclo{[d-Phe-l-MeN-g-Acp-
d-Glu(Bn)-l-MeN-g-Acp]2-}
Scheme 2. Structures of Acp-based (9) and Ach-based (10a–c) fourfold symmetric cyclic octapeptides and
their corresponding homodimers (D9 and D10aꢀc) and heterodimers (D9/10a, D9/10b and D9/10c).
(12d)
were
synthesized
(Scheme 3). These CPs all pres-
A new peptide, cyclo[(d-Ser-l-MeN-g-Ach)4-] (10c), ob-
tained by hydrogenation of 10b (10% Pd/C, balloon pres-
sure), was prepared with the aim of evaluating whether the
potential side-chain/side-chain interactions between serine
hydroxy groups in its homodimer form (D10c) play an impor-
tant role in stabilizing the system or if formation of the hete-
rodimer is still more favorable.[16] Although CP 10c is not
very soluble in nonpolar solvents, analysis of its 1H NMR
spectrum in chloroform again reveals the formation of the
ent twofold symmetry instead of the fourfold symmetry of
previously described CPs (9 and 10a–c). For this reason,
these systems can form two non-equivalent homodimers
that differ in the cross-strand interactions between mono-
mers; i) the eclipsed homodimer, in which identical amino
acids face each other (such as Phe to Phe, Ser to Ser, or Glu
to Glu) and ii) the alternated dimer in which Phe faces the
other amino acids (Ser for 12a or 12b, and Glu for 12d and
12e).[17] In addition, the functional group of Ser and Glu
Figure 1. a) 1H NMR spectra corresponding to the formation of heterodimer D9/10a by addition of 1.7 equivalents of 10a to a 1.7 mm CDCl3 solution of 9
(298 K). Signals in purple correspond to the heterodimer D9/10a. b) ROESY spectrum (CaH and CgH of the g-amino acids) of a 1:1.7 mixture of CPs 9
and 10a, showing HgAcp(9)-HaAchACTHNUGRTNEN(GU 10a) and HaAcp(9)-HgAchAHCUTNTGRNE(NNUG 10a) nOe cross-peaks in heterodimer D9/10a.
112
www.chemasianj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 110 – 121

Heterodimeric Energy Transfer Materials
gidity derived from the cyclo-
hexanecarboxylic acid and to
the difference in the bulk of the
side chains of Phe and Ser,
which might slightly distort the
flat conformation allowing the
Ser residues to form stronger
hydrogen bonds with Phe.[7c]
The preferential formation of
the corresponding heterodimers
was confirmed by the addition
of 11a to a solution of 12a
(2.8 mm) in chloroform, which
once again resulted in the ap-
pearance of a new set of signals
in the 1H NMR spectrum that
did not correspond to either of
the possible homodimers (Fig-
ure SI-4 in the Supporting In-
formation). Interestingly, the
ratio between the eclipsed and
alternated heterodimeric as-
Scheme 3. Structures of Ach-based (11a–c) and Acp-based (12a–f) twofold symmetric cyclic octapeptides and
models of their corresponding two non-equivalent (eclipsed [E] and alternated [A]) homodimers (D11aꢀc and
D
12aꢀf) and heterodimers (D11aꢀc/12aꢀf).
semblies
(D11a/12a[E]
and
D11a/12a[A]) was 1:1 (Table 1,
side chains can be used to introduce additional groups for
studying other properties. As expected, the H NMR spectra
entry 10). This finding again corroborated that the selective
formation of the heterodimeric ensembles is driven mainly
by the b-sheet hydrogen bonding interactions and is inde-
pendent of the side-chains of the a-Aa employed in the CPs.
In addition, the flexibility of the cyclopentyl rings allows the
Acp-based CPs, such as 12a, to fit their conformation to the
buckled form of the Ach-based CP regardless of the cross-
strand interaction.
1
of these CPs in chloroform showed two sets of signals that
correspond to the two dimers both of which reflect b-sheet-
like hydrogen bonding between monomers in the all-trans
conformation. Only CP 11a showed the preferential forma-
tion of one dimer (alternated) over the eclipsed one
(D11a[A]/D12a[E]ꢁ14:1), while the other two peptides (12a
and 12d) showed a similar ratio of both dimers (eclipsed,
D
12a[E] and D12d[E], versus alternated, D12a[A] and
Steady-State and Time-Resolved Fluorescence Studies
D12d[A]; Table 1, entries 1, 4, and 7). The dimer ratios were
not dependent on peptide concentration or temperature. In
the case of Ach-based CP 11a, the predominant formation
of the alternated dimer (D11a[A]) was attributed to the ri-
Fluorescence methods have been extensively used in the
study of self-assembly processes in supramolecular chemis-
try,[18] but their application to the study of SPNs is still
recent.[13,19] We proceeded to use the spectral properties of
pyrene (i.e., excimer emission) to evaluate different systems
involving homo- and heterodimeric entities in equilibrium.
The information derived from the CPs containing pyrene
moieties was completed by using a second chromophore
such as perylene, which could act as a donor for fluores-
cence resonance energy transfer (FRET, Figure 2b). Pery-
lene was chosen because it has spectroscopic characteristics
such as high quantum yield and spectral absorption overlap
with the emission spectrum of pyrene, thus making these
two moieties an efficient FRET pair for studying heterodi-
meric systems.[20] In addition perylene can also form excimer
species that could be used in evaluating homodimer proper-
ties. Pyrene and perylene excimers formation and FRET ef-
fects can be combined to analyze complex intermolecular
association systems.
Table 1. Ratio between eclipsed and staggered dimers of Acp and Ach-
based CPs.
Entry Dimer Subunit 1 [R]
Subunit 2 [R]
D[E]/
D[A][a]
1
2
3
4
5
6
7
D11a
D11b
D11c
D12a
D12b
D12c
D12d
11a, R=CH2OBn
11b, R=CH2OH
11c, R=CH2OPyr
12a, R=CH2OBn
12b, R=CH2OH
12c, R=CH2OPyr
12d,
11a, R=CH2OBn
11b, R=CH2OH
11c, R=CH2OPyr
12a, R=CH2OBn
12b, R=CH2OBn
12c, R=CH2OPyr
12d,
1:14
1:6
1:10
1:1
1:1
1:1
1:1.2
R=(CH2)2CO2Bn
12e,
R=(CH2)2CO2H
12f,
R=(CH2)2CO2Per
11a, R=CH2OBn
R=(CH2)2CO2Bn
12e,
R=(CH2)2CO2H
12f,
R=(CH2)2CO2Per
12a, R=CH2OBn
8
9
D12e
D12f
1:1.2
1:0.6
1:1
The fluorescence network developed in our study was
formed by different equilibria involving three fluorescently
labeled a,g-CPs (Scheme 3): cyclo{[d-Phe-l-MeN-g-Ach-d-
10 D11a/
12a
1
[a] Ratio determined by using H NMR spectroscopy.
Chem. Asian J. 2011, 6, 110 – 121
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
113

J. R. Granja et al.
FULL PAPERS
Figure 2. a) Pyrene fluorescence emission of 12c (l=340 nm excitation wavelength) in CHCl3 from 3.35 nm to 0.23 mm, denoting homodimer formation.
Inset shows the titration data for KA calculation. b) Absorption spectrum of perylene (solid line) showing the overlap with pyrene emission (dashed
line). All spectra were normalized for the sake of comparison.
Ser
d-Ser
Acp-d-Glu
G
assemblies (from 1:1.2 for D12d[E]/D12d[A] or D12e[E]/
12e[A] to 1:0.6 for D12f[E]/D12f[A]; Table 1, entry 9), with
the eclipsed dimer now the main form. Preferential forma-
tion of the eclipsed dimer (D12f[E]) instead of the alternated
one (D12f[A]) can perhaps be attributed to an edge-face in-
teraction between perylene moieties, vide infra.
R
D
G
based and Ach-based CPs were synthesized as follows. Pre-
viously obtained peptides 11a, 12a, and 12d were deprotect-
ed (H2, Pd/C, or HBr/TFA) to provide the corresponding
CPs 11b, 12b, and 12e, respectively. The unprotected Acp-
based CPs showed almost the same ratio between the iso-
meric dimers as the protected ones (1:1 for 12b and 1:1.2
for 12e; Table 1, entries 5 and 8), whereas unprotected Ach-
based CP 11b showed a slight increase in the eclipsed/alter-
nated ratio (from 1:14 for 11a to 1:6 for 11b; Table 1, en-
tries 1 and 2). Peptides 11b and 12b were subsequently cou-
pled with 1-pyreneacetic acid in the presence of N,N’-di
The emission of CPs containing pyrene moieties presents
two characteristic well-defined emission bands. The pyrene
band, in the ultraviolet region, is identified by its typical vi-
bronic structure, and the emission at longer wavelength cor-
responds to the short-lived dimeric species—one member of
which is in an electronically excited state, called an exci-
mer.[22] The relative intensities of the two bands reflect the
ratio between monomer and dimer in solution. The first
equilibrium studied was the homodimerization process of
the Acp-based cyclic octapeptide 12c because in nonpolar
solvents the dimers formed (D12c[E] and D12c[A], Scheme 3)
exist in an almost equimolecular ratio. Dilute solutions of
12c (0.15 nm in chloroform) showed only the characteristic
emission spectrum of the pyrene moiety, with two maxima
at l=377 and 397 nm. Addition of successive aliquots of a
concentrated stock solution of this CP (1.5 mm) resulted in
the appearance of the excimer-emission band at l=470 nm
(Figure 2a).[23] Considering that the formation of excimers is
geometrically only possible in the eclipsed dimer (D12c[E]),
we used the observed excimer band to calculate its homodi-
merization binding constant, with a value of KA (CHCl3)=
7.3ꢅ108 mꢀ1 obtained (Figure 2a, inset).[24] In this way, the
formation of two additional hydrogen-bonding interactions,
in comparison to cyclic hexapeptides,[13b] leads to an increase
in the association constant by almost three orders of magni-
tude. This is by far the largest association constant measured
for a nanotube-forming CP and is five orders of magnitude
higher than that of the dimer-forming CPs made of alternat-
ing d- and l-a-amino acids (a-d,l-CP), which has an identi-
cal number of hydrogen-bonding interactions.[25] Analogous
isopropylcarbodiimide/4-(dimethylamino)pyridine
(DIC/
DMAP) to give 11c and 12c, respectively, whereas CP 12 f
was prepared by analogous coupling of 12e with perylen-3-
ylmethanol (14).[21] The NMR spectra of the resulting pep-
tides in chloroform exhibited clear b-sheet signals, character-
istic of homodimeric structures, in which the two non-equiv-
alent dimers are slowly exchanging on the NMR timescale.
As expected for 12c, the ratio between eclipsed and alter-
nated forms did not change with the introduction of a pyre-
neacetyl group into the serine side-chain (1:1 for D12c[E]/
D12c[A]; Table 1, entry 6). However, fluorescent Ach-based
CP 11c presented a different ratio between non-equivalent
dimers (D11c[E]/D11c[A]) than the corresponding unprotect-
ed CP (11b), with an increase observed in the proportion of
the alternated dimer (1:10; Table 1, entry 6). This is close to
the ratio observed for the benzyl ester 11a and suggests that
perhaps the observed stabilization of the eclipsed form of
the CP containing the free hydroxy group (11b) could arise
from the formation of week hydrogen-bonding interactions
between the side chains of the serine residues. Surprisingly,
the introduction of the perylene moiety into CP 12e led to
the inversion of the ratio between eclipsed and alternated
114
www.chemasianj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 110 – 121

Heterodimeric Energy Transfer Materials
results were obtained in a 20% DMSO/CHCl3 solution, al-
though, as expected, the association constant is slightly
lower, KA (1:4 DMSO/CHCl3)=1.5ꢅ107 mꢀ1 (Figure SI-5 in
the Supporting Information). Further experiments were car-
ried out in polar media (CH3CN) and the characteristic exci-
mer band was observed.[26] This band disappeared on addi-
tion of different aliquots of a more strongly hydrogen bond-
ing competing solvent such as milliQ water (Figure SI-6 in
the Supporting Information).[27]
The second equilibrium analyzed was the homodimeriza-
tion process of CP 11c (Scheme 3). As described above, at
low concentration (0.12 mm) in CHCl3 this CP showed the
typical emission spectrum of the pyrene moiety. Successive
addition of aliquots of a concentrated stock solution of 11c
(12.0 mm) resulted in the appearance of the corresponding
pyrene excimer. In this case, as mentioned before, the exci-
mer formation is only geometrically possible for the minor
dimer (D11c[E]), a situation that made the measurement of
its association constant more difficult and less accurate. De-
spite this, the association constant was estimated to be of a
similar order, albeit slightly lower, than that for the Acp-
based CPs, KA (CHCl3)=1.2ꢅ108 mꢀ1 (Figure SI-7 in the
Supporting Information), as predicted by our previous DFT
calculations.[11]
Addition of Ach-based CP 11a to a 0.13 mm chloroform
solution of homodimer D12c resulted in an increase in the
natural emission of pyrene at l=377 and 397 nm, and a re-
duction in the l=472 nm pyrene excimer band (Figure SI-8
in the Supporting Information). This observation is consis-
tent with the formation of heterodimer D11a/12c as homodi-
mer D12c disappears.[28] As expected, addition of one equiva-
lent of 12a, which does not have a fluorophore in its side
chain, to a solution of Ach-based CP 11c resulted in the dis-
appearance of the pyrene excimer-emission band, again con-
sistent with formation of the heterodimeric species D11a/12c
and the disruption of excimer-forming homodimer D12c.
Both results clearly indicate that the homodimeric model is
responsible for the excimer emission and that heterodimer
assemblies are more stable than the corresponding homo-
dimers.
Preferential heterodimer formation was also studied
through FRET processes between pyrene and perylene moi-
eties of corresponding CPs (11c and 12 f, respectively).[29] Ir-
radiation at l=340 nm of an 8.0 mm chloroform solution of
perylene-derivatized CP 12 f resulted in a small emission
characterized by the three feature bands (at l=449, 479,
and 513 nm) of the perylene system. Addition of successive
aliquots of complementary CP (11c) led to a clear increase
in the fluorescence emission intensity of the three bands
(Figure 3b), which must arise from intramolecular FRET
between pyrene and perylene in the resulting heterodimeric
[20,30]
complex D11c/12f
.
As expected, inverse titration by the
addition of successive aliquots of 12 f to a 6.0 mm solution of
11c again led to a substantial increase in the fluorescence
emission intensity of the perylene bands and a significant re-
duction of the pyrene emission, denoting once more the het-
erodimeric FRET process (Figure SI-9 in the Supporting In-
formation). As previously indicated, the efficient energy
transfer is possible, owing to the spectral overlap between
the two fluorophores (Figure 3b).
Disruption of the supramolecular heterodimeric species
D11c/12f was achieved by addition of competitive CP 11a, re-
sulting in a reduction of the three signals of perylene and an
increase in the typical emission of pyrene (Figure SI-10 in
the Supporting Information). This result is a consequence of
the formation of the heterodimers D11a/12f and D11a/11c, which
do not have any energy transfer process.
These results demanded a full characterization of the dy-
namic processes taking place using time-resolved fluores-
cence techniques. Irradiation of homodimer D12c resulted in
Figure 3. a) Fluorescence emission of 12c (5ꢅ10ꢀ6 m), excitation at l=345 nm. Fluorescence emission of a mixture of 11c (5ꢅ10ꢀ6 m) and 12 f (5ꢅ10ꢀ6 m),
excitation at l=345 nm. b) Fluorescence emission of homodimer D12f and heterodimer D11c/12f. Addition of successive aliquots of 11c (from 0 to 1.2 eq)
to a 8.0 mm solution of 12 f increases the fluorescence emission intensity of perylene bands, denoting an efficient Per/Pyr FRET process in the resulting
heterodimeric D11c/12f complex.
Chem. Asian J. 2011, 6, 110 – 121
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
115

J. R. Granja et al.
FULL PAPERS
a fluorescence decay that could be fitted to a three exponen-
tial model with three different times of 25, 6.4, and 0.3 ns
(Figure 4). The longer time (25 ns), with a contribution of
95%, corresponds to the normal pyrene emission decay in a
non-deoxygenated solvent and it must correspond to the
dimer (D12c[A]) for which excimer formation is geometrical-
ly impossible (Scheme 3). The shorter lifetime of 6.4 ns, with
a 5% contribution, and the 0.3 ns time, with a residual con-
tribution of only 0.2%, are assigned to the process of con-
formational change that gives the required geometrical ar-
rangement for excimer formation—that is, the dimer with
pyrene side chains in register (D12c[E], Figure 4c). Fluores-
cence decay measurements at the excimer emission wave-
length (l=480 nm) also showed a three exponential decay
(22.7, 7.4, and 2.1 ns). The two shortest lifetimes are associ-
ated with negative pre-exponential factors, whereas the long
lifetime has a positive pre-exponential factor for which the
numerical value is somewhat higher than the absolute value
of the sum of the previous two. The 22.7 ns time corre-
sponds to the excimer decay that, as can be deduced from
the pre-exponential factors, is mainly formed in the excited
state. This means that in the ground state, the two pyrene
moieties must not be interacting with each other to any
great extent. From the pre-exponential factors it can be in-
ferred that at least 70% of the excimer-emitting moieties
are formed in the excited state. This conformation can come
from two different conformers that evolve to the excimer
formation. The dimer equilibrium between alternated and
eclipsed forms is slower, with both species observed in the
NMR spectrum; as this equilibrium is slower than the NMR
time scale, we conclude that this change in conformation is
related with the side chain that links pyrene and the CP
backbone. These two conformers are in a 7:1 ratio and the
most abundant is the one that evolves in 7.4 ns.
The time-resolved emission of CP 12 f resulted in a fluo-
rescence decay that could be fitted again to a three expo-
nential process, for which the longest lifetime was 4 ns,
which corresponds well with the typical perylene emission,
with an 80% contribution (Figure SI-11 in the Supporting
Information).[31] The other two exponential factors of 1.4
and 0.2 ns have a contribution of 18 and 2%, respectively.
This could represent the existence of conformers that evolve
towards an excimer, although this could not be confirmed
because the perylene excimer could not be observed under
these conditions.[29]
The emission temporal dependence of the heterodimeric
complex D11c/12f at the perylene band (l=480 nm), as the
pyrene is excited (l=333 nm), can again be fitted to a three
Figure 4. a) Time resolved l=395 nm emission (blue) of 5 mm monomeric CP 12c, with excitation at l=333 nm; l=450 nm emission (green) of homodi-
meric D12c. b) Time resolved l=480 nm emission of heterodimeric D11c/12f (5 mm 11c and 3 mm 12 f) with excitation at l=333 nm. The black line repre-
sents the laser pulse and the Red lines are the best fit to the experimental data. c) Proposed conformational change of eclipsed dimer to form the exci-
mer upon pyrene excitation.
116
www.chemasianj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 110 – 121

Heterodimeric Energy Transfer Materials
exponential process for which the two shortest time expo-
nentials, 1.2 and 3.6 ns,[31] have the same amplitude, but op-
posite sign (Figure 4b). Consequently, the emitting species
with a lifetime of 3.6 ns comes from the one with the short-
est lifetime (1.2 ns). The emission time of perylene is 3.6 ns
and this results from the energy transfer from the pyrene.
The 1.2 ns lifetime was assigned to the energy-transfer pro-
cess. The transfer-energy constant is (1/1.2)=0.8 nsꢀ1, which
represents the weighted-average value of the different con-
formations and distances between the donor (pyrene) and
the acceptor (perylene) moieties. The third time, 23.4 ns,
corresponds to the pyrene emission from the excimer-form-
ing homodimer (D11c), which in this case is in excess. The
lifetime of the energy transfer process (1.2 ns), together with
the previously estimated Fçrster radius (R0) of 22 ꢆ,[20b,e] al-
lowed us to determine an average pyrene–perylene distance
of 18 ꢆ, with 96% efficiency in the transfer process by using
Equation (1):
nanotubes, with the Acp-based dimers strongest (>108 mꢀ1).
Despite the strength of the interactions, there is clear selec-
tivity for the heterodimer formation between Acp-based
CPs and Ach-based CPs. The selective formation of the het-
erodimeric ensembles is driven mainly by the backbone-to-
backbone hydrogen bonding interactions and is not a result
of interstrand side chain–side chain interactions. This homo/
hetero equilibrium allows the selective formation of a wide
collection of supramolecular assemblies with potential appli-
cations as sensors, catalysts, and electronic/optical devices.
Introduction of pyrene and perylene functionalities was par-
ticularly relevant for the design and characterization of a
novel multicomponent system based on fluorescently deriv-
atized a,g-CPs. These materials were specifically tailored for
studying networks involving homo- and heterodimerization
processes. Exhaustive fluorescence studies have shown that
such CPs act as suitable platforms for efficient energy trans-
fer by providing an optimized spatial distribution between
fluorophores, a characteristic that could be successfully used
for the development of efficient light-harvesting/light-con-
verting ensembles.
R0
r ¼
ð1Þ
1=6
ðt0D=ktÞ
For which t0D is the excited-state lifetime of pyrene in the
absence of transfer (25 ns) and kt is the transfer-energy con-
stant (0.8 nsꢀ1).[32] This distance corresponds to the dimers
in which pyrene and perylene are not in proximity
(D11c/12f[A]). The other form (D11c/12f[E]), in which both
fluorophores are in close contact, must have an even faster
transfer that is not detectable in the nanosecond timescale.
Time-resolved fluorescence experiments at the perylene
band (l=450 nm) when the pyrene is excited (l=333 nm)
of heterodimer D11c/12f solutions, in which perylene-contain-
ing CP (12 f) was in excess, gave similar results (see
Figure 12-SI in the Supporting Information). The fluores-
cence decay could be fitted again to a three exponential pro-
cess, for which the lifetime of 4.5 ns is the emission time of
perylene, which is formed by the energy transfer from the
pyrene, but also directly by the l=333 nm light, because it
is in excess compared to pyrene CP. This is the explanation
for why, in this case, the 1.2 and 4.5 ns have the same ampli-
tude, but opposite signs, as there is more excited perylene
than pyrene. Finally, the third lifetime (14 ns) corresponds
to the pyrene emission and derives from the small amount
of monomer-emitting pyrenes.
Experimental Section
Synthesis of 3-Formylperylene (13)[21,33]
An ice-cooled suspension of perylene (1.00 g, 3.96 mmol) in 1,2-dichloro-
benzene (60 mL) was treated with 1,1-dichloromethyl methyl ether
(466 mL, 5.15 mmol) [CAUTION: CARCINOGEN] and TiCl4 (1.0m so-
lution in CH2Cl2; 5.94 mL, 5.94 mmol). The resulting mixture was stirred
at 08C for 1 h, allowed to warm to room temperature, and poured onto
ice (200 g) and HCl (conc. 10 mL). The organic layer was diluted with
CHCl3 (200 mL), washed with HCl (5%) (200 mL) and H2O (3ꢅ
200 mL), dried over Na2SO4, and evaporated to dryness. The residue was
purified by flash chromatography (100% CHCl3) to give the desired com-
pound 3-formylperylene as orange crystals [1.02 g, 92%, Rf =0.74 (100%
CHCl3)]. 1H NMR (CDCl3, 250.13 MHz): d=10.25 (s, 1H, CHO), 9.10
(d, J=8.2 Hz, 1H), 8.41–7.98 (m, 4H), 7.93–7.56 (m, 4H), 7.50 ppm (t,
J=7.8 Hz, 2H); MS (ESI): m/z (%): 281 ([MH]+, 100), 245 (78), 126
(72); HRMS (ESI): calcd for C21H13O ([MH]+): 281.0966; found:
281.0961.
Synthesis of Perylen-3-ylmethanol (14)[21]
A solution of 3-formylperylene (400.0 mg, 1.43 mmol) in a mixture of tet-
rahydrofuran/methanol (THF/MeOH, 2:1; 60 mL) was treated with
sodium borohydride (56.7 mg, 1.50 mmol), and the mixture was stirred at
RT for 30 min. The resulting mixture was diluted with H2O (40 mL),
acidified to pH 1–2 by addition of HCl (conc.), and finally extracted with
CHCl3 (2ꢅ200 mL). The combined organic layers were washed with H2O
(2ꢅ100 mL) and saturated aqueous NaHCO3 (1ꢅ100 mL), dried over
Na2SO4, and concentrated under reduced pressure. The crude material
was purified by using flash chromatography (100% CHCl3), to give pery-
len-3-ylmethanol [348 mg, 86%, Rf =0.24 (100% CHCl3), yellow solid].
1H NMR (CDCl3, 250.13 MHz): d=8.30–8.11 (m, 4H), 7.96 (d, J=
8.4 Hz, 1H), 7.70 (d, J=8.0 Hz, 2H), 7.63–7.42 (m, 4H), 5.10 (s, 2H),
1.76 ppm (br s, 1H); MS (EI): m/z (%): 282 ([M]+, 100), 265 (59), 252
(59); HRMS (EI): calcd for C21H14O ([M]+): 282.104465; found:
282.104829.
Conclusions
In summary, analysis by using NMR and fluorescence spec-
troscopies have provided conclusive evidence that cyclic a,g-
octapeptide heterodimers of a,gACTHUNRTGENNUG(Acp)-CPs with a,gACHTUNGTNER(NUGN Ach)-
CPs, which are held together by b-sheet-like hydrogen
bonds, are more stable than the corresponding homodimers.
Pyrene-labeled CPs allowed measurement of the association
constant and showed the strength of the b-sheet interaction
responsible for homodimer formation; this interaction is
stronger than those previously reported for CPs that form
Peptide Synthesis
Linear peptides Boc-[(d-Leu-l-MeN-g-Acp)4-]OFm, Boc-[(d-Phe-l-MeN-g-
Ach)4-]OFm, Boc-{[d-Ser(Bn)-l-MeN-g-Ach]4-}OFm, Boc-{[l-Phe-d-MeN-
g-Acp-l-Ser(Bn)-d-MeN-g-Acp]2-}OFm,
Boc-{[l-Phe-d-MeN-g-Ach-l-
Chem. Asian J. 2011, 6, 110 – 121
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
117

J. R. Granja et al.
FULL PAPERS
Ser(Bn)-d-MeN-g-Ach]2-}OFm, and Boc-{[l-Phe-d-MeN-g-Acp-l-Glu(Bn)-
d-MeN-g-Acp]2-}OFm were prepared following the synthetic strategy de-
scribed previously.[10]
afford
9
as
a
white solid [29.0 mg, 76%, Rt =16 min (Phenomenex
Maxsil-10 semipreparative column, 5–15% MeOH in CH2Cl2, 30 min)].
1H NMR (CDCl3, 500.13 MHz): d=8.33 (d, J=9.5 Hz, 1H, NHLeu), 5.16
(m, 1H, HaLeu), 4.84 (m, 1H, HgAcp), 3.06 (m, 1H, HaAcp), 3.03 (s, 3H,
NCH3), 2.09–1.14 (m, 9H, 3ꢅCH2 g-Acp+CH2Leu +CHLeu), 0.94 ppm
(dd, J1 =17.7 Hz, J2 =6.3 Hz, 6H, CH3Leu); 13C NMR (CDCl3,
125.77 MHz): d=174.9 (CO), 173.4 (CO), 54.6 (CH), 46.8 (CH), 43.3
(CH), 42.2 (CH2), 32.8 (CH2), 29.4 (NCH3), 27.9 (CH2), 26.9 (CH2), 24.7
(CH), 23.4 (CH3), 22.2 ppm (CH3); FT-IR (293 K, CHCl3): n˜ =3309
(amide A), 3003, 2959, 1665, 1626 (amide I), 1536 cmꢀ1 (amide IIII); MS
(MALDI-TOF): m/z (%): 993 ([M+K]+, 7), 975 ([M+Na]+, 100), 953
([MH]+, 34); HRMS (MALDI-TOF): calcd for C52H88N8O8Na ([M+Na]+
): 975.6617; found: 975.6573.
General Method for Amino Acid Coupling: Synthesis of Boc-[d-Leu-
l-MeN-g-Acp-]OFm
A solution of 6b (775 mg, 1.84 mmol) in 18 mL of trifluoroacetic acid/di-
chloromethane (TFA/DCM, 1:1) was stirred at RT for 15 min. The sol-
vent was removed and the residue was dried under high vacuum for 3 h.
The resulting TFA salt was dissolved in dry DCM (18 mL), after which
d-Boc-Leu-OH·H2O (467 mg, 2.02 mmol), 2-(1H-7-azabenzotriazol-1-yl)-
1,1,3,3-tetramethyl uronium hexafluorophosphate methanaminium
(HATU, 769 mg, 0.601 mmol), and N,N-diisopropylethylamine (DIEA,
1.29 mL, 7.36 mmol) were successively added. The mixture was stirred
for 1 h at RT and the solution was poured into a separating funnel and
washed with HCl (5%) and NaHCO3 (sat.). The organic layers were
dried over Na2SO4 and concentrated under reduced pressure to give a
yellow oil, which was purified by flash chromatography (15–25% EtOAc/
Hexanes) to give the target dipeptide as a white foam [897 mg, 91%,
Rf =0.66 (50% EtOAc/hexanes)]. 1H NMR (CDCl3, 250.13 MHz): d=
7.67 (d, J=7.3 Hz, 2H), 7.50 (d, J=7.3 Hz, 2H), 7.27 (td, J1 =7.4 Hz, J2 =
22.2 Hz, 4H), 5.24 (m, 1H), 4.85 (m, 1H), 4.70–4.30 (m, 3H), 4.12 (m,
1H), 2.76 (s, 3H), 2.64 (m, 1H), 2.08–1.05 (m, 18H), 0.88 ppm (ddd, J1 =
5.1 Hz, J2 =7.7 Hz, J3 =10.2 Hz, 6H); 13C NMR (CDCl3, 62.90 MHz): d=
175.5 (CO), 173.2 (CO), 155.5 (CO), 143.5 (C), 141.2 (C), 127.7 (CH),
127.0 (CH), 124.7 (CH), 119.9 (CH), 79.3 (C), 65.8 (CH2), 57.2 (CH),
53.9 (CH), 49.2 (CH), 46.9 (CH), 42.7 (CH2), 41.5 (CH), 30.9 (CH2), 29.0
(CH3), 28.3 (CH3), 27.5 (CH2), 27.0 (CH2), 23.4 (CH3), 21.7 ppm (CH3);
MS (FAB+): m/z (%): 535 ([MH]+, 17), 479 (8), 435 ([MHꢀBoc]+, 7);
HRMS (FAB+): calcd for C32H43N2O5 ([MH]+): 535.31720; found:
535.31845.
Synthesis of cyclo[(d-Phe-l-MeN-g-Ach)4-] (10a)[7c]: Previously obtained
from Boc-[(d-Phe-l-MeN-g-Ach)4-]OFm (201 mg, 140 mmol) in the same
way as 9 to yield, after HPLC purification, CP 10a (112 mg, 70%); white
1
solid. H NMR (CDCl3, 300.13 MHz): d=8.57 (d, J=8.6 Hz, 1H, NHPhe),
7.17 (m, 5H, Ar-HPhe), 5.30 (dd, J=5.7 Hz, 1H, HaPhe), 4.45 (br s, 1H,
HgAch), 3.00–2.71 (m, 3H, CH2bPhe+HaAch), 2.64 (s, 3H, NCH3), 1.82–
1.04 ppm (m, 8H, CH2Ach); 13C NMR (CDCl3, 75.40 MHz): d=175.1
(CO), 171.3 (CO), 136.5 (C), 129.6 (CH), 128.2 (CH), 126.8 (CH), 51.4
(CH), 49.9 (CH), 43.3 (CH), 40.4 (CH2), 30.8 (CH2), 30.1 (CH2), 29.5
(NCH3), 28.5 (CH2), 24.7 ppm (CH2); FT-IR (293 K, CHCl3): n˜ =3312
(amide A), 2935, 2864, 1660, 1623 (amide I), 1525 cmꢀ1 (amide IIII); MS
(FAB+): m/z (%): 1145 ([MH]+, 100); HRMS (FAB+): calcd for
C68H89N8O8 ([MH]+): 1145.68034; found: 1145.68176.
Synthesis of cycloACTHNUTRGNEUNG
{[d-Ser(Bn)-l-MeN-g-Ach]4-} (10b): Prepared from
Boc-{[d-Ser(Bn)-l-MeN-g-Ach]4-}OFm (99 mg, 78 mmol) in the same way
as 9 to yield, after HPLC purification, CP 10b (50 mg, 50%); white solid.
1H NMR (CDCl3, 250.13 MHz): d=8.41 (d, J=8.4 Hz, 1H, NHSer), 7.32–
7.10 (m, 5H, Ar-HBn), 5.39 (dd, J=21.8 Hz, 1H, HaSer), 4.54 (m, 1H,
HgAch), 4.37 (overlapping doublets, J=8.0 Hz, 2H, CH2Bn), 3.61–3.42 (m,
2H, CH2bSer), 3.04 (s, 3H, NCH3), 2.84 (m, 1H, HaAch) 1.92–1.21 ppm (m,
8H, CH2Ach); 13C NMR (CDCl3, 62.90 MHz): d=175.4 (CO), 170.6 (CO),
137.9 (C), 128.2 (CH), 127.5 (CH), 127.1 (CH), 73.1 (CH2), 71.4 (CH2),
51.7 (CH), 47.9 (CH), 43.2 (CH), 31.1 (CH2), 29.9 (NCH3), 29.8 (CH2),
28.7 (CH2), 24.6 ppm (CH2); FT-IR (293 K, CHCl3): n˜ =3309 (amide A),
2936, 2863, 1660, 1628 (amide I), 1524 cmꢀ1 (amide IIII); MS (FAB+):
m/z (%): 1265 ([MH]+, 100); HRMS (FAB+): calcd for C72H96N8O12
([MH]+): 1265.72260; found: 1265.72583.
General Method for Peptide Coupling
Synthesis of Boc-[(d-Leu-l-MeN-g-Acp)2-]OFm: A solution of the dipep-
tide Boc-[d-Leu-l-MeN-g-Acp-]OFm (225 mg, 0.42 mmol) in 20% piperi-
dine/DCM (5 mL) was stirred at RT for 20 min and the solvent was re-
moved in vacuo. The residue was dissolved in DCM (10 mL), washed
with HCl (5%), dried over Na2SO4, filtered, and concentrated to give
Boc-[d-Leu-l-MeN-g-Acp-]OH, which was used without further purifica-
tion.
A solution of Boc-[d-Leu-l-MeN-g-Acp-]OFm (225 mg, 0.42 mmol) in
4 mL of TFA/DCM (1:1) was stirred at RT for 15 min. The solvent was
removed and the residue was dried under high vacuum for 3 h. The re-
sulting dipeptide was dissolved in dry DCM (4 mL) to which Boc-[d-Leu-
l-MeN-g-Acp-]OH, HBTU (176 mg, 0.46 mmol), and DIEA (294 mL,
1.68 mmol) were successively added. The mixture was stirred for 1 h at
RT and the solution was washed with HCl (5%) and NaHCO3 (sat.),
dried over Na2SO4, and concentrated under reduced pressure. The resi-
due was purified by flash chromatography (0–3% MeOH/CH2Cl2) to
give Boc-[(d-Leu-l-MeN-g- Acp)2-]OFm, as a white foam [270 mg, 83%,
Rf =0.28 (3% MeOH in DCM)]. MS (FAB+): m/z (%): 773 ([MH]+,
100), 673 ([MHꢀBoc]+, 14), 560 (22), 452 (48); HRMS: calcd for
C45H65N4O7 ([MH]+): 773.48533; found: 773.48575.
Synthesis of cyclo[(d-Leu-l-MeN-g-Acp)4-] (9)[7b]: A solution of Boc-[(d-
Leu-l-MeN-g-Acp)4-]OFm (50.0 mg, 40.1 mmol) in 20% piperidine in
DCM (400 mL) was stirred at RT for 20 min. The solvent was removed
and the residue was dissolved in DCM (5 mL). The solution was washed
with HCl (5%), dried over Na2SO4, filtered, and concentrated. The re-
sulting residue was dissolved in 400 mL of TFA/DCM (1:1) and stirred at
RT for 15 min. The solvent was removed and the residue was dried under
high vacuum for 3 h and used without further purification. The linear
peptide was dissolved in DCM (40.1 mL) and treated with TBTU
(14.1 mg, 44.1 mmol), followed (dropwise) by DIEA (28 mL, 160.3 mmol)
[an additional 1 equiv of TBTU (12.9 mg, 40.1 mmol), and 4 equiv of
DIEA (28 mL, 160.3 mmol) were added when the starting material was
observed by HPLC, and the resulting mixture was stirred for 3 h at RT to
complete the reaction]. After 12 h, the solvent was removed under re-
duced pressure, and the crude mixture was purified by using HPLC to
Synthesis of cyclo[(d-Ser-l-MeN-g-Ach)4-] (10c): A solution of cyclo
ACHTUNGTRENNUNG{[d-
Ser(Bn)-l-MeN-g-Ach]4-} (10b; 50 mg, 40 mmol) in EtOH (1.5 mL) was
treated with 10% Pd/C (12 mg, 11 mmol) and stirred at RT under hydro-
gen overnight. The resulting mixture was filtered through a Celite pad,
the residue was washed with ethanol, and the combined filtrates and
washings were concentrated under reduced pressure to afford 10c as a
white solid (36 mg, 99%). 1H NMR ([D6]DMSO, 250.13 MHz): d=8.14–
7.87 (m, 1H, NHSer), 4.71 (m, 1H, HaSer), 4.27 (m, 1H, HgAch), 3.40 (m,
2H, CH2bSer), 2.88 (m, 1H, HaAch), 2.86 (s, 3H, NCH3), 1.72–1.22 ppm
(m, 8H, CH2Ach); 13C NMR (CDCl3, 62.90 MHz): d=174.0 (CO), 169.8
(CO), 61.8 (CH2), 51.2 (CH), 42.7 (CH), 39.7 (CH), 31.3 (CH2), 29.1
(NCH3), 29.0 (CH2), 28.5 (CH2), 24.3 ppm (CH2); FT-IR (293 K, CaF2):
n˜ =3310 (amide A), 2933, 2864, 1661, 1621 (amide I), 1541 cmꢀ1 (amide
IIII); MS (FAB+): m/z (%): 905 ([MH]+, 100); HRMS (FAB+): calcd for
C44H73N8O12 ([MH]+): 905.53480; found: 905.53476.
Synthesis
of
cyclo{[d-Phe-l-MeN-g-Ach-d-Ser(Bn)-l-MeN-g-Ach]2-}
(11a)[7c]: Previously obtained from Boc-{[d-Phe-l-MeN-g-Ach-d-Ser(Bn)-
l-MeN-g-Ach]2-}-OFm (100.0 mg, 66.7 mmol) in the same way as 9 to
yield, after HPLC purification, CP 11a (73.4 mg, 92%); white solid.
1H NMR (CDCl3, 750.00 MHz): d=8.59 (d, J=8.2 Hz, 2H, NHSer
D11a[A]), 8.52 (overlapping doublets, J=9.4 Hz, 2.14H, NHPhe
D11a[E]+NHPhe D11a[A]), 8.29 (d, J=8.4 Hz, 0.14H, NHSer D11a[E]), 7.23–
7.06 (m, 21.4H, ArPhe+ArBn), 5.35–5.29 (m, 2.14H, HaSer), 5.22–5.14 (m,
2.14H, HaPhe), 4.52–4.37 (m, 4.28H, HgAch), 4.32 (d, J=3.3 Hz, 4.28H,
CH2Bn), 3.61–3.54 (m, 2.14H, CH2bSer), 3.50–3.45 (m, 2.14H, CH2bSer),
3.19–2.93 (m, 6.42H, HaAch+CH2bPhe), 2.91 (s, 6.42H, NCH3), 2.85–2.77
(m, 2.14H, CH2bPhe), 2.74 (s, 6.42H, NCH3), 2.45–2.35 (m, 4.28H,
CH2Ach), 1.92–0.91 ppm (29.96H, CH2Ach); 13C NMR (CDCl3, 75.40 MHz):
118
www.chemasianj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 110 – 121

Heterodimeric Energy Transfer Materials
d=175.9 (CO), 174.5 (CO), 171.5 (CO), 169.9 (CO), 137.6 (C), 137.2 (C),
127.8 (CH), 127.2 (CH), 126.5 (CH), 125.4 (CH), 73.2 (CH2), 71.6 (CH2),
51.9 (CH), 51.0 (CH), 49.8 (CH), 48.5 (CH), 44.2 (CH), 42.4 (CH), 38.9
(CH2), 31.8 (CH2), 31.0 (CH2), 30.3 (CH3), 29.9 (CH3), 29.6 (CH3), 28.7
(CH2), 28.4 (CH2), 24.6 ppm (CH2); FT-IR (293 K, CHCl3): n˜ =3311
(amide A), 2934, 2861, 1660, 1625 (amide I), 1523 cmꢀ1 (amide IIII); MS
(ESI): m/z (%): 1227 ([M+Na]+, 46) 1205 ([MH]+, 100); HRMS (ESI):
calcd for C70H93N8O10 ([MH]+): 1205.7015; found: 1205.7009.
0.92 mmol) were added when the starting material was observed by
HPLC]. The resulting mixture was stirred for 3 h at RT to complete the
reaction. After 12 h, the solvent was removed under reduced pressure
and the crude product was purified by HPLC to afford 11a as a white
foam [40.7 mg, 31%, Rt =18 min (Phenomenex Maxsil-10 semiprepara-
tive column, 3–12% MeOH in CH2Cl2, 25 min)]. 1H NMR (CDCl3,
500.13 MHz): d=8.66 (d, J=9.0 Hz, 2H, NHPhe D12a[E]), 8.52 (d, J=
8.6 Hz, 2H, NHSer D12a[A]), 8.48 (d, J=9.2 Hz, 2H, NHPhe D12a[A]), 8.27
(d, J=9.1 Hz, 2H, NHSer D12a[E]), 7.25–7.02 (m, 40H, Ar-HPhe+Ar-HBn),
5.41–5.09 (m, 8H, HaSer+HaPhe), 4.94–4.65 (m, 8H, HgAcp), 4.44 (d, J=
12.5 Hz, 2H, CH2Bn), 4.41–4.35 (overlapping doublets, J=12.4 Hz, 4H,
CH2Bn), 4.21 (d, J=12.5 Hz, 2H, CH2Bn), 3.59 (m, 4H, CH2bSer D12a[A]),
3.27 (m, 4H, CH2bSer D12a[E]), 3.17–2.84 (m, 24H, HaAcp+CH2bPhe
D12a[E]+NCH3), 2.77 (m, 4H, CH2bPhe D12a[A]), 2.47 (m, 12H, NCH3),
2.02–0.76 ppm (m, 48H, CH2Acp); 13C NMR (CDCl3, 125.77 MHz): d=
174.8 (CO), 174.4 (CO), 172.3 (CO), 171.5 (CO), 137.9 (C), 136.6 (C),
129.3 (CH), 128.3 (CH), 128.2 (CH), 127.5 (CH), 127.1 (CH), 126.9
(CH), 72.8 (CH2), 71.0 (CH2), 54.6 (CH), 54.2 (CH), 50.0 (CH), 48.1
(CH), 42.6 (CH), 42.3 (CH), 40.6 (CH2), 33.3 (CH2), 29.7 (CH3), 29.1
(CH3), 27.6 (CH2), 27.3 (CH2), 26.8 (CH2), 26.6 (CH2), 26.4 (CH2),
26.0 ppm (CH2); MS (ESI): m/z (%): 1172 ([M+Na]+, 2), 1149 ([MH]+,
6), 594 ([M+K]2+, 100); HRMS (ESI): calcd for C66H85N8O10 ([MH]+):
1149.6383; found: 1149.6398.
Synthesis of cyclo[(d-Phe-l-MeN-g-Acp-d-Ser-l-MeN-g-Acp)2-] (12b):
Anisole (12.4 mL, 0.13 mmol), pentamethylbenzene (12.2 mg, 82.5 mmol),
TFA (1.85 mL), and HBr solution (33% in AcOH, 400 mL) were succes-
sively added to cyclo{[d-Phe-l-MeN-g-Acp-d-Ser(Bn)-l-MeN-g-Acp]2-}
(12a, 12.0 mg, 10.4 mmol). The resulting mixture was vigorously stirred
for 2 h at RT and then evaporated to dryness in vacuo to give 12b as an
orange oil, which was used without further purification [10.1 mg, 100%].
MS (ESI): m/z (%): 992 ([M+Na]+, 16), 969 ([MH]+, 100); HRMS
(ESI): calcd for C52H72N8O10 ([MH]+): 969.5444; found: 969.5459.
Synthesis of cyclo[(d-Phe-l-MeN-g-Ach-d-Ser-l-MeN-g-Ach)2-] (11b)[7c]
:
A solution of cyclo{[d-Phe-l-MeN-g-Ach-d-Ser(Bn)-l-MeN-g-Ach]2-} (11a,
20.0 mg, 16.6 mmol) in EtOH (250 mL) was treated with 10% Pd/C
(3.5 mg, 3.3 mmol) and stirred overnight at RT under a hydrogen atmos-
phere. The resulting mixture was filtered through a Celite pad, the resi-
due was washed with ethanol, and the combined filtrates and washings
were concentrated under reduced pressure to afford 11b as a white solid
(16.2 mg, 95%). 1H NMR (CDCl3, 750.00 MHz): d=8.96 (s, 2H, NHSer
D11b[A]), 8.65 (brs, 0.32H, NHPhe D11b[E]), 8.49 (d, J=9.8 Hz, 2H, NHPhe
D11b[A]), 8.14 (brs, 0.32H, NHSer D11b[E]), 7.24–7.08 (m, 11.6H, Ar-
H
Phe), 5.31–5.24 (m, 2.32H, HaSer), 5.22–5.16 (m, 2.32H, HaPhe), 4.52–4.45
(m, 4.64H, HgAch), 3.89–3.81 (m, 2.32H, CH2bSer), 3.76–3.69 (m, 2.32H,
CH2bSer), 3.32–3.25 (m, 4.64H, HaAch), 3.09–3.01 (m, 9.28H,
NCH3+CH2bPhe), 2.97–2.90 (m, 2.32H, CH2bPhe), 2.80 (s, 6.96H, NCH3),
2.45–2.39 (m, 4.64H, CH2Ach), 1.92–1.03 (m, 34.8H, CH2Ach+OHSer);
13C NMR (CDCl3, 75.40 MHz): d=177.9 (CO), 174.2 (CO), 171.8 and
171.6 (CO), 168.6 and 168.6 (CO), 136.8 and 136.8 (C), 129.8 (CH), 128.2
(CH), 126.7 (CH), 67.0 (CH2), 53.3 (CH), 52.2 (CH), 51.2 (CH), 50.0
(CH), 44.4 (CH), 42.5 (CH), 39.3 (CH2), 32.2 and 32.1 (CH2), 30.8 (CH2),
29.9 and 29.8 (CH3), 29.1, 28.7 and 28.6 (CH2), 24.7 ppm (CH2); FTIR
(293 K, CHCl3): n˜ =3317 (amide A), 2935, 2863, 1660, 1620 (amide I),
1524 cmꢀ1 (amide IIII); MS (FAB+): m/z (%): 1025 ([MH]+, 100);
HRMS (FAB+): calcd for C56H81N8O10 ([MH]+): 1025.60757; found:
1025.60908.
Synthesis
of
cyclo{[d-Phe-l-MeN-g-Ach-d-Ser(Pyr)-l-MeN-g-Ach]2-}
Synthesis
of
ACTHNGUTERNNUG
cyclo{[d-Phe-l-MeN-g-Acp-d-Ser(Pyr)-l-MeN-g-Acp]2-}
(11c): A solution of 1-pyreneacetic acid (3.3 mg, 12.7 mmol) in CDCl3
(500 mL) was stirred and sonicated at RT for 10 min, after which DIC
(3.0 mL, 19.1 mmol), cyclo[(d-Phe-l-MeN-g-Ach-d-Ser-l-MeN-g-Ach)2-]
(11b, 6.5 mg, 6.4 mmol), and DMAP (2.3 mg, 19.1 mmol) were successively
added. The mixture was stirred for 1 h and the solution was washed with
HCl (5%) and NaHCO3 (sat.), dried over Na2SO4, and concentrated
under reduced pressure. The residue was purified by flash chromatogra-
phy (0–7% MeOH in CH2Cl2) to give 11c as a yellow solid [9.1 mg,
(12c): A solution of 1-pyreneacetic acid (5.4 mg, 20.9 mmol) in DMF/
CDCl3 (1:4) (500 mL) was stirred and sonicated at RT for 10 min, and
then DIC (2.5 mL, 15.7 mmol), cyclo[(d-Phe-l-MeN-g-Acp-d-Ser-l-MeN-g-
Acp)2-] (12b, 10.1 mg, 10.4 mmol), and DMAP (1.9 mg, 15.7 mmol) were
successively added. The mixture was stirred for 1 h at RT, the solution
was then washed with HCl (5%) and NaHCO3 (sat.), dried over Na2SO4,
and concentrated under reduced pressure. The residue was purified by
HPLC to give 12c as a yellow solid [3.5 mg, 23%, Rt =23 min (Phenom-
enex Maxsil-10 semipreparative column, 0–15% MeOH in CH2Cl2,
30 min)]. 1H NMR (CDCl3, 500.13 MHz): d=8.55 (d, J=9.4 Hz, 2H,
NHPhe D12c[E]), 8.47 (d, J=9.0 Hz, 2H, NHSer D12c[A]), 8.16 (d, J=
9.4 Hz, 2H, NHPhe D12c[A]), 8.14–7.86 (m, 30H, NHSer D12c[E]+Ar-HPyr),
7.81 (d, J=8.0 Hz, 2H, Ar-HPyr), 7.74 (d, J=7.9 Hz, 2H, Ar-HPyr), 7.64
(m, 2H, Ar-HPyr), 7.46 (m, 2H, Ar-HPyr), 7.17–7.00 (m, 20H, ArPhe), 5.39–
4.97 (m, 16H, HaSer+HaPhe+CH2Pyr), 4.74–4.41 (m, 8H, HgAcp), 4.38–3.76
(m, 8H, CH2bSer), 3.08–2.63 (m, 8H, CH2bPhe), 2.61–0.56 ppm (m, 80H,
HgAcp+NCH3+CH2Acp); MS (ESI): m/z (%): 746 ([M+K]2+, 100), 738
([M+Na]2+, 26); HRMS (ESI): calcd for C88H92N8O12Na ([M+Na]2+):
738.3400; found: 738.3395.
1
95%, Rf =0.37 (3% MeOH in CH2Cl2)]. H NMR (CDCl3, 500.13 MHz):
d=8.55 (d, J=7.8 Hz, 2H, NHSer D11c[A]), 8.36 (d, J=8.0 Hz, 0.20H,
NHPhe D11c[E]), 8.29 (d, J=9.5 Hz, 2H, NHPhe D11c[A]), 8.19–7.88 (m,
17.8H, NHSer D11c[E]+Ar-HPyr), 7.80 (overlapping doublets, J=7.8 Hz,
2.2H, Ar-HPyr), 7.16–7.01 (m, 11H, ArPhe), 5.44–5.33 (m, 2H, HaSer
D11c[A]), 5.25–5.13 (m, 2.4H, HaPhe D11c[E]+HaPhe D11c[A]+HaSer
D11c[E]), 4.42–4.31 (m, 2.2H, HgAch), 4.29–4.12 (m, 8.8H,
CH2bSer+CH2Pyr +HgAch), 3.98–3.89 (m, 2.2H, CH2bSer), 3.10–2.95 (m,
4.4H, HaAch+CH2bPhe), 2.91–2.83 (m, 2.2H, CH2bPhe), 2.75 (overlapping
singlets, 6.6H, NCH3), 2.68 (overlapping singlets, 6.6H, NCH3), 2.58–2.48
(m, 2.2H, HaAch), 2.25–2.12 (m, 2.2H, CH2Ach), 1.79–1.68 (m, 2.2H,
CH2Ach), 1.66–0.68 ppm (30.8H, CH2Ach); MS (MALDI-TOF): m/z (%):
1548 ([M+K]+, 12), 1532 ([M+Na]+, 100), 1509 ([MH]+, 17); HRMS
(MALDI-TOF): calcd for C92H101N8O12 ([MH]+): 1509.75; found:
1509.74.
Synthesis
of
cyclo{[d-Phe-l-MeN-g-Acp-d-Glu(Bn)-l-MeN-g-Acp]2-}
(12d): Prepared from Boc-[d-Phe-l-MeN-g-Acp-d-Glu(Bn)-l-MeN-g-Acp-
]OFm (1000.0 mg, 1.09 mmol) in the same way as 11a to yield, after
HPLC purification, the desired CP (254.1 mg, 38%); white foam.
1H NMR (CDCl3, 500.13 MHz): d=8.65 (d, J=8.9 Hz, 1.7H, NHPhe
D12d[E]), 8.54 (d, J=9.0 Hz, 2H, NHGlu D12d[A]), 8.30 (d, J=9.0 Hz, 2H,
NHPhe D12d[A]), 8.16 (d, J=9.0 Hz, 1.7H, NHGlu D12d[E]), 7.29–7.05 (m,
37H, Ar-HPhe+Ar-HBn), 5.28–5.15 (m, 3.7H, HaPhe D12d[E]+HaPhe
D12d[A]), 5.13–5.06 (m, 2H, HaGlu D12d[A]), 5.05–5.00 (m, 5.7H, HaGlu
D12d[E]+CH2Bn D12d[A]), 4.98–4.96 (overlapping singlets, 3.4H, CH2Bn
D12d[E]), 4.83–4.62 (m, 7.4H, HgAcp), 3.08–2.76 (m, 25.9H,
HaAcp+CH2bPhe+NCH3), 2.53–2.19 (m, 18.5, NCH3+CH2gGlu), 2.08–
1.09 ppm (m, 51.8H, CH2bGlu+CH2Acp); 13C NMR (CDCl3, 125.77 MHz):
d=175.0 (CO), 174.5 (CO), 172.5 (CO), 172.3 (CO), 172.1 (CO), 136.5
(C), 135.8 (C), 129.3 (CH), 128.5 (CH), 128.4 (CH), 128.3 (CH), 128.2
(CH), 126.9 (CH), 66.4 (CH2), 54.7 (CH), 54.4 (CH), 50.2 (CH), 47.7
Synthesis of cyclo{[d-Phe-l-MeN-g-Acp-d-Ser(Bn)-l-MeN-g-Acp]2-} (12a):
A
solution of Boc-[d-Phe-l-MeN-g-Acp-d-Ser(Bn)-l-MeN-g-Acp]-OFm
(200.0 mg, 0.23 mmol) in 20% piperidine in DCM (2.3 mL) was stirred at
RT for 20 min. The solvent was removed and the residue was dissolved
in DCM (10 mL). The solution was washed with HCl (5%), dried over
Na2SO4, filtered, and concentrated. The resulting residue was dissolved in
TFA/DCM (2.3 mL, 1:1) and the solution was stirred at RT for 15 min.
The solvent was removed and the residue was dried under high vacuum
for 3 h and used without further purification. The linear peptide was dis-
solved in DCM (46 mL) and treated with HATU (104.9 mg, 0.28 mmol),
followed (dropwise) by DIEA (160.6 mL, 0.92 mmol) [an additional
1 equiv of HATU (87.4 mg, 0.23 mmol) and 4 equiv of DIEA (160.6 mL,
Chem. Asian J. 2011, 6, 110 – 121
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
119

J. R. Granja et al.
FULL PAPERS
(CH), 42.4 (CH), 42.2 (CH), 40.8 (CH2), 34.0 (CH2), 33.2 (CH2), 33.0
(CH2), 30.0 (CH2), 29.8 (CH3), 29.5 (CH2), 29.3 (CH2), 29.2 (CH2), 29.0
(CH3), 28.8 (CH2), 28.6 (CH2), 28.0 (CH2), 27.7 (CH2), 27.5 (CH2), 27.2
(CH2), 26.9 (CH2), 26.4 (CH2), 26.0 ppm (CH2); MS (MALDI-TOF): m/z
(%): 1272 ([M+K]+, 2), 1256 ([M+Na]+, 57), 1234 ([MH]+, 100); HRMS
(MALDI-TOF): calcd for C70H89N8O12 ([MH]+): 1233.6600; found:
1233.6639.
Synthesis of cyclo[(d-Phe-l-MeN-g-Acp-d-Glu-l-MeN-g-Acp)2-] (12e): A
solution of cyclo{[d-Phe-l-MeN-g-Acp-d-Glu(Bn)-l-MeN-g-Acp]2-} (12d,
50.0 mg, 40.6 mmol) in 5% AcOH in CH2Cl2 (1 mL) was treated with
10% Pd/C (8.6 mg, 8.1 mmol) and stirred at RT under hydrogen over-
night. The resulting mixture was filtered through a Celite pad, the residue
was washed with MeOH/CH2Cl2 (1:1) and the combined filtrates and
washings were concentrated under reduced pressure to afford 12e as a
white solid (40.2 mg, 94%); MS (ESI-TOF): m/z (%): 1075 ([M+Na]+,
24), 1053 ([MH]+, 100), 527 ([MH]2+, 68); HRMS (ESI-TOF): calcd for
C56H77N8O12 ([MH]+): 1053.566091; found: 1053.563722.
Chem. 1989, 101, 849–871; Angew. Chem. Int. Ed. Engl. 1989, 28,
848–869.
[2] For light harvesting energy-transfer system based on synthetic pro-
tein assemblies, see; a) R. A. Miller, A. D. Presley, M. B. Francis, J.
Am. Chem. Soc. 2007, 129, 3104–3109; b) M. Endo, M. Fujitsuka, T.
Majima, Chem. Eur. J. 2007, 13, 8660–8666; c) Y. Z. Hu, S. Tsukiji,
S. Shinkai, S. Oishi, I. Hamachi, J. Am. Chem. Soc. 2000, 122, 241–
253.
[3] a) E. R. Young, J. Rosenthal, J. M. Hodgkiss, D. G. Nocera, J. Am.
Chem. Soc. 2009, 131, 7678–7684; b) G. Fernꢂndez, E. M. Pꢀrez, L.
Sꢂnchez, N. Martꢁn, Angew. Chem. 2008, 120, 1110–1113; Angew.
Chem. Int. Ed. 2008, 47, 1094–1097; c) P. G. A. Janssen, J. Vanden-
bergh, J. L. J. Van Dongen, E. W. Meijer, A. P. H. J. Schenning, J.
Am. Chem. Soc. 2007, 129, 6078–6079; d) A. Del Guerzo, A. G. L.
Olive, J. Reichwagen, H. Hopf, J. P. Desvergne, J. Am. Chem. Soc.
2005, 127, 17984–17985.
[4] a) R. J. Brea, C. Reiriz, J. R. Granja, Chem. Soc. Rev. 2010, 39,
1448–1456; b) R. J. Brea, J. R. Granja in Dekker Encyclopedia of
Nanoscience and Nanotechnology (Eds: J. A. Schwarz, C. I. Contes-
cu, K. Putyera), Marcel Dekker Inc, New York, 2004, pp. 3439–
3457; c) G. R. Patzke, F. Krumeich, R. Nesper, Angew. Chem. 2002,
114, 2554–2571; Angew. Chem. Int. Ed. 2002, 41, 2446–2461;
d) D. T. Bong, T. D. Clark, J. R. Granja, M. R. Ghadiri, Angew.
Chem. 2001, 113, 1016–1041; Angew. Chem. Int. Ed. 2001, 40, 988–
1011.
[5] a) V. Gauba, J. D. Hartgerink, J. Am. Chem. Soc. 2007, 129, 2683–
2690; b) E. S. Barrett, T. J. Dale, J. Rebek, Jr., J. Am. Chem. Soc.
2007, 129, 8818–8824; c) W. Zheng, H. O. Jacobs, Adv. Funct. Mater.
2005, 15, 732–738; d) M. V. Ovchinnikov, A. M. Brown, X. Liu,
C. A. Mirkin, L. N. Zakharov, A. L. Rheingold, Inorg. Chem. 2004,
43, 8233–8235.
[6] a) P. M. Bowers, S. J. Cokus, D. Eisenberg, T. O. Yeates, Science
2004, 306, 2246–2249; b) A. Tsuchisaka, A. Theologis, Proc. Natl.
Acad. Sci. USA 2004, 101, 2275–2280; c) S. K. Nair, S. K. Burley,
Cell 2003, 112, 193–205; d) D. W. Felsher, Nat. Rev. Cancer 2003, 3,
375–380; e) M. R. Young, H.-S. Yang, N. H. Colburn, Trends Mol.
Med. 2003, 9, 36–41; f) R. Eferl, R. Ricci, L. Kenner, R. Zenz, J.-P.
David, M. Rath, E. F. Wagner, Cell 2003, 112, 181–192.
[7] a) C. Reiriz, M. Amorꢁn, R. Garcꢁa-FandiÇo, L. Castedo, J. R.
Granja, Org. Biomol. Chem. 2009, 7, 4358–4361; b) R. J. Brea, L.
Castedo, J. R. Granja, Chem. Commun. 2007, 3267–3269; c) M.
Amorꢁn, L. Castedo, J. R. Granja, Chem. Eur. J. 2005, 11, 6543–
6551; d) M. Amorꢁn, R. J. Brea, L. Castedo, J. R. Granja, Org. Lett.
2005, 7, 4681–4684; e) M. Amorꢁn, L. Castedo, J. R. Granja, J. Am.
Chem. Soc. 2003, 125, 2844–2845.
Synthesis
(12 f):
of
solution of cyclo[(d-Phe-l-MeN-g-Ach-d-Glu-l-MeN-g-Acp)2-]
ACTHNGUTERNNUG
cyclo{[d-Phe-l-MeN-g-Acp-d-Glu(Per)-l-MeN-g-Acp]2-}
A
(12e, 30.0 mg, 28.5 mmol) in DMF/CDCl3 (1:11, 3.0 mL) was stirred and
sonicated at RT for 10 min and then DIC (3.0 mL, 19.1 mmol), perylen-3-
ylmethanol (14, 24.1 mg, 85.6 mmol), and DMAP (15.7 mg, 128.3 mmol)
were successively added. The mixture was stirred for 1 h at RT and the
solution was washed with HCl (5%) and NaHCO3 (sat.), dried over
Na2SO4, and concentrated under reduced pressure. The residue was puri-
fied by HPLC to give 12 f as a yellow solid [26.3 mg, 58%, Rt =17 min
(Phenomenex Maxsil-10 semipreparative column, 0–10% MeOH in
CH2Cl2, 30 min)]. 1H NMR (CDCl3, 500.13 MHz): d=8.62 (d, J=8.9 Hz,
2H, NHPhe D12f[E]), 8.53 (d, J=8.7 Hz, 1.3H, NHGlu D12f[A]), 8.28 (d,
J=8.8 Hz, 1.3H, NHPhe D12f[A]), 8.17 (d, J=8.8 Hz, 2H, NHGlu D12f[E]),
8.14–7.25 (overlapping doublets+doublets, 36.3H, Ar-HPer), 7.18–6.99 (m,
16.5H, Ar-HPhe), 5.38 (d, J=1.6 Hz, 2.6H, CH2Per D12f[A]), 5.29 (d, J=
6.5 Hz, 4H, CH2Per D12f[E]), 5.20 (m, 2H, HaPhe D12f[E]), 5.18–5.08 (m,
2.6H, HaGlu D12f[S]+HaPhe D12f[A]), 5.04 (m, 2H, HaGlu D12f[E]), 4.83–
4.55 (m, 6.6H, HgAcp), 3.07–2.83 (m, 23.1H, HaAcp+CH2bPhe+NCH3),
2.51–2.17 (m, 16.5, NCH3+CH2gGlu), 2.04–1.03 ppm (m, 46.2H,
CH2bGlu+CH2Acp); MS (MALDI-TOF): m/z (%): 1619 ([M+K]+, 52),
1603 ([M+Na]+, 50), 1581 ([MH]+, 2); HRMS (MALDI-TOF): calcd for
C98H100N8O12K ([M+K]+): 1619.7098; found: 1619.7094.
Time-Resolved Fluorescence
Fluorescence lifetimes were determined by time-correlated single-photon
counting by using an Edinburgh Instruments CD-900 spectrometer
equipped with a hydrogen-filled nanosecond flash lamp. The instrumental
response width of the system was 1.0 ns. Measurements were usually car-
ried out until 10000 counts were reached in (2ꢅ103 channels). The emis-
sion band-pass for the lifetime measurements was usually 20 nm. The ex-
periments were performed at room temperature and samples were
purged with argon prior to measurement.
[8] M. Amorꢁn, L. Castedo, J. R. Granja, Chem. Eur. J. 2008, 14, 2100–
2111.
[9] For a,g-peptide nanotube formation, see: a) R. Garcꢁa-FandiÇo, J. R.
Granja, D. A. Marco, M. Orozco, J. Am. Chem. Soc. 2009, 131,
15678–15686; b) C. Reiriz, R. J. Brea, R. Arranz, J. L. Carrascosa,
A. Garibotti, B. Manning, J. M. Valpuesta, R. Eritja, L. Castedo,
J. R. Granja, J. Am. Chem. Soc. 2009, 131, 11335–11337.
[10] R. J. Brea, M. Amorꢁn, L. Castedo, J. R. Granja, Angew. Chem.
2005, 117, 5856–5859; Angew. Chem. Int. Ed. 2005, 44, 5710–5713.
[11] For theoretical studies carried out with homo-and heterodimers, see:
R. Garcꢁa-FandiÇo, L. Castedo, J. R. Granja, S. Vꢂzquez, J. Phys.
Chem. B 2010, 114, 4973–4983.
[12] For heteromeric peptide nanotube formation driven mainly by side-
chain/side-chain interactions, see: a) K. Rosenthal-Aizman, G.
Svensson, A. Undꢀn, J. Am. Chem. Soc. 2004, 126, 3372–3373; b) J.
Sꢂnchez-Quesada, M. P. Isler, M. R. Ghadiri, J. Am. Chem. Soc.
2002, 124, 10004–10005.
Acknowledgements
This work was supported in part by the Spanish Ministry of Science and
Innovation (Micinn), ERDF [under projects SAF2007-61015 and Consol-
ider Ingenio 2010 (CSD2007-00006)], the European project Magnifyco
(NMP4-SL-2009-228622) and the Xunta de Galicia (PGIDIT08C-
SA047209PR and R2006/124). R.J.B. and M.J.P.-A. thank the Spanish
M.E.C. for their FPU and FPI fellowships, respectively. M.P. thanks
Spanish M.A.E.C. for her MAEC-AECI fellowship.
[13] For donor-acceptor systems based on a,g-CP heterodimer formation,
see: a) R. J. Brea, M. A. Herranz, L. Sꢂnchez, L. Castedo, W. Seitz,
D. M. Guldi, N. Martꢁn, J. R. Granja, Proc. Natl. Acad. Sci. USA
2007, 104, 5291–5294; b) R. J. Brea, M. E. Vꢂzquez, M. Mosquera,
L. Castedo, J. R. Granja, J. Am. Chem. Soc. 2007, 129, 1653–1657.
[14] a) J. J. Havranek, P. B. Harbury, Nat. Struct. Biol. 2003, 10, 45–52;
b) A. Apelblat, Can. J. Chem. 1991, 69, 638–647.
[1] a) T. Brixner, J. Stenger, H. M. Vaswani, M. Cho, R. E. Blankenship,
G. R. Fleming, Nature 2005, 434, 625–628; b) L. M. Yoder, A. G.
Cole, R. J. Sension, Photosynth. Res. 2002, 72, 147–158; c) T. Puller-
its, V. Sundstrçm, Acc. Chem. Res. 1996, 29, 381–389; d) M. R. Wa-
sielewski, Chem. Rev. 1992, 92, 435–461; e) R. Huber, Angew.
120
www.chemasianj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 110 – 121

Heterodimeric Energy Transfer Materials
[15] S. Krimm, J. Bandekar in Advances in Protein Chemistry (Eds: C. B.
Anfinsen, J. T. Edsall, F. M. Richards), Academic Press, Orlando,
1986, pp. 181–364.
[16] M. Amorin, V. Villaverde, L. Castedo, J. R. Granja, J. Drug Delivery
Sci. Technol. 2005, 15, 87–92.
[25] a) M. R. Ghadiri, K. Kobayashi, J. R. Granja, R. K. Chadha, D. E.
McRee, Angew. Chem. 1995, 107, 76–78; Angew. Chem. Int. Ed.
Engl. 1995, 34, 93–95; b) T. D. Clark, J. M. Buriak, K. Kobayashi,
M. P. Isler, D. E. McRee, M. R. Ghadiri, J. Am. Chem. Soc. 1998,
120, 8949–8962.
[17] For a recent studies of molecular tweezers based on a,g-CP dimer/
porphyrin hybrids, see: L. P. Hernꢂndez-Eguꢁa, R. J. Brea, L. Caste-
do, P. Ballester, J. R. Granja Chem. Eur. J. 2010, DOI:10.1002/
chem.201002271.
[18] a) V. A. Azov, A. Schlegel, F. Diederich, Angew. Chem. 2005, 117,
4711–4715; Angew. Chem. Int. Ed. 2005, 44, 4635–4638; b) Y. Kane-
kiyo, R. Naganawa, H. Tao, Chem. Commun. 2004, 1006–1007;
c) R. B. Martin, K. Fu, H. Li, D. Cole, Y.-P. Sun, Chem. Commun.
2003, 2368–2369.
[26] Homodimerization binding constant of 12c in CH3CN was deter-
mined to be 1.3ꢅ108 mꢀ1
.
[27] Fluorescence studies of 12c carried out in milliQ water did not show
the appearance of the pyrene excimer-emission band, which suggests
that the formation of the corresponding dimer is not possible in
water or that the solubility of the peptide is much lower.
[28] Heterodimerization binding constant of D11c/12f in chloroform was es-
timated by dilution experiments to be 2.5ꢅ109 mꢀ1
.
[29] Perylene-based CP 12 f present only one band with the typical vi-
bronic structure, so it could not be used for association constant de-
termination through excimer formation, because its characteristic
lower energy peak than the monomer emission could not be detect-
ed even at high concentration. This was attributed to the short life-
time and low quantum yields of the perylene moiety, which pre-
cludes the excimer formation and detection if it is not previously
preformed.
[30] Given that pyrene-based CP 11c does not emit at l=449, 479, and/
or 513 nm, and the concentration of the perylene-based CP 12 f was
kept constant throughout the titration, the increased fluorescence
emission intensity can only be consequence of the FRET process in
the heterodimer D11cꢀ12c. Furthermore, addition of 1-pyreneacetic
acid to 12 f did not induce any change in its emission intensity, a
finding consistent with the absence of the heterodimeric species.
[31] Perylene life times could not be precisely measured in heterodimeric
species, owing to the perturbation caused by the presence of other
supramolecular entities with similar lifetimes.
[19] a) N. Ashkenasy, W. S. Horne, M. R. Ghadiri, Small 2006, 2, 99–102;
b) W. S. Horne, N. Ashkenasy, M. R. Ghadiri, Chem. Eur. J. 2005,
11, 1137–1144.
[20] For pyrene/perylene FRET pair, see: a) D. Lindegaard, A. S.
Madsen, I. V. Astakhova, A. D. Malakhov, B. R. Babu, V. A. Kor-
shunb, J. Wengel, Bioorg. Med. Chem. 2008, 16, 94–99; b) M.
Masuko, S. Ohuchi, K. Sode, H. Ohtani, A. Shimadzu, Nucleic Acids
Res. 2000, 28, 34e; c) J. N. Wilson, J. Gao, E. T. Kool. Tetrahedron
2007, 63, 3427–3433; d) Y. N. Teo, J. N. Wilson, E. T. Kool, J. Am.
Chem. Soc. 2009, 131, 3923–3933; e) H. Kashida, T. Takatsu , K. Se-
kiguchi , H. Asanuma, Chem. Eur. J. 2010, 16, 2479–2486.
[21] For the preparation of perylen-3-ylmethanol (14), see: M. V. Skoro-
bogatyi, A. A. Pchelintseva, A. L. Petrunina, I. A. Stepanova, V. L.
Andronova, G. A. Galegov, A. D. Malakhov, V. A. Korshun, Tetra-
hedron 2006, 62, 1279–1287.
[22] J. B. Birks, Rep. Prog. Phys. 1975, 38, 903–974.
[23] a) C. Garcꢁa-Echeverrꢁa, J. Am. Chem. Soc. 1994, 116, 6031–6032;
b) A. Ueno, I. Suzuki, T. Osa, J. Am. Chem. Soc. 1989, 111, 6391–
6397.
[32] Bernard Valeur, Molecular Fluorescence: Principles and Applica-
tions, Wiley-VCH, Weinheim, 2002.
[24] KA was determined, taking in consideration ratios between alternat-
ed and eclipse forms calculated by NMR experiments, by least-
squares analysis fitting to appropriate equations using Kaleidagraph
3.5 (Synergy Software, Reading, PA). For model and equation used,
see: a) J. W. Park, H. E. Song, S. Y. Lee, J. Org. Chem. 2003, 68,
7071–7076; b) R. B. Martin, Chem. Rev. 1996, 96, 3043–3064.
[33] A. Rieche, H. Gross, E. Hçft, Chem. Ber. 1960, 93, 88–94.
Received: August 8, 2010
Published online: October 26, 2010
Chem. Asian J. 2011, 6, 110 – 121
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
121