Design, synthesis, and characterization of peptide nanostructures having ion channel activity

Article abstract of DOI:10.1016/j.bmc.2003.08.037

We report the synthesis and the functional studies of multiple crown α-helical peptides designed to form artificial ion channels. The approach combines the versatility of solid phase peptide synthesis, the conformational predictability of peptidic molecul

Full text of DOI:10.1016/j.bmc.2003.08.037

Bioorganic & Medicinal Chemistry 12 (2004) 1279–1290
Design, synthesis, and characterization of peptide nanostructures
having ion channel activity
Eric Biron, Francois Otis, Jean-Christophe Meillon, Martin Robitaille, Julie Lamothe,
Patrick Van Hove, Marie-Eve Cormier and Normand Voyer*
´ ´ ´ ´ ´
Departement de chimie and Centre de recherche sur la fonction, la structure et l’ingenierie des proteines, Faculte des sciences et de genie,
´ ´
Universite Laval, Quebec, Canada G1K 7P4
Received 17 July 2003; revised 31 July 2003; accepted 1 August 2003
Abstract—We report the synthesis and the functional studies of multiple crown a-helical peptides designed to form artificial ion
channels. The approach combines the versatility of solid phase peptide synthesis, the conformational predictability of peptidic
molecules, and the solution synthesis of crown ethers with engineerable ion-binding abilities. Several biophysical methods were
employed to characterize the activity and the mode of action of these crown peptide nanostructures. The 21 residue peptides bearing
six 21-EC-7 turned out to facilitate the translocation of ions in a similar fashion to natural ion channels.
# 2003 Elsevier Ltd. All rights reserved.
1. Introduction
However, there are several problems associated with
natural ion channel proteins that severely limit their
Nanotechnology and the development of analytical
devices for the detection of important analytes at the
molecular level constitute one of the most active area of
research in bioanalytical chemistry and a formidable
challenge to scientists. However, before getting full
benefits from nanotechnology, scientists need to devise
novel approaches to the preparation of well defined
molecular architectures in the nanometre range that will
serve as discrete components in the construction of
functional and practical devices, such as for example
diagnostic kits. As in many areas of research, Nature,
with an infinite number of functional nanostructures,
serves as a source of inspiration. For instance, channel
proteins exemplify beautifully nanopore technology.
These proteins are large and complex membrane pro-
teins that regulate the flow of ions across cell mem-
branes. Due to their involvement in several diseases and
their enormous potential in nanotechnology and single
molecule detection,1À6 ion channel proteins have been
the subject of intense investigation in the past 15 years.
Truly remarkable results have been obtained by using
natural channel proteins, in particular a-hemolysin.
utility as both component of biosensors and potential
therapeutics. These transmembrane proteins are difficult
to isolate in their pure and functional form and they
tend to denature easily. Also, their chemical modifi-
cation into more elaborated molecular systems is almost
impossible owing to their complex multidomain struc-
ture. To overcome these difficulties, numerous studies
have focused on the development of artificial molecules
designed to mimic natural ion channel proteins.7À19
Important achievements have been made by many
groups around the world and some successful approa-
ches gave compounds that showed single-channel activ-
7À19
ity.
However, there are still several issues that need
to be further addressed before artificial channels and
nanopores can be used in practical applications: these
include ion selectivity, preparation on larger scales,
complete characterization of the mode of action, selec-
tive modifications into more elaborated systems for
recognition of relevant analytes, and implementation
into workable devices.
Our group has devised a unique strategy for the pre-
paration of artificial ion channels.20 The general concept
is to use a-helical peptidic structures as scaffold to ori-
ent multiple macrocyclic ligands on a top of each other.
Thus, to create membrane active channels, we have
designed and prepared 21-residue peptides bearing six
Keywords: Ion channel; Peptide nanostructures; Crown compounds;
Peptide framework; Membrane active; Membrane transport.
* Corresponding author. Tel.: +1-418-656-3613; fax: +1-418-656-
7916; e-mail: normand.voyer@chm.ulaval.ca
0968-0896/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2003.08.037

1280
E. Biron et al. / Bioorg. Med. Chem. 12 (2004) 1279–1290
crown ether modified amino acids at position 2, 6, 9, 13,
16, and 20. When the peptidic framework adopts an a-
helical conformation, the crown rings are proposed to
align to form a polar pore long enough to allow the
passage of ions across a lipid bilayer membrane (Fig. 1).
Because the peptides contain 15 l-leucine residues, they
are sufficiently hydrophobic to incorporate in a lipid
bilayer environnement.21 In addition, the strong helix
induction propensity of leucine suggests that the pepti-
dic backbone should adopt preferentially an a-helical
conformation necessary to bring the crown ring in the
proposed supramolecular array.22,23
acids at positions 2, 6, 9, 13, 16, and 20 of the sequence
allows the crown ligands to be on the same side of the
peptidic framework when it adopts an a-helical struc-
ture. As mentioned in the introduction, we aimed at
exploiting the unique ion binding ability of crown
ethers. The binding selectivity of such neutral macro-
cyclic ligands depends on their size and on the number
of oxygen atoms. On that basis, we devised a general
synthetic strategy toward enantiopure crown ether ami-
noacids. The strategy starts with 3,4-dihydroxy-l-phe-
nylalanine (l-DOPA) commercially available and is
illustrated in Figure 2. Many analogues have been pre-
pared using the synthetic scheme which turned out to be
quite general using Cs2CO3 in DMF as base. The best
yield was obtained with the 21-crown-7 analogue known
to bind Cs+ very tightly, hence the templating effect
being best in this case.
There are several advantages to this strategy. First, ion
selectivity can be easily modified by changing the size of
the crown ether rings. Second, since they are designed to
operate as monomer and not through aggregation, their
detailed mechanism and functional structure should be
more easily assessed. Third, post synthesis modifications
and end group engineering can be easily and efficiently
done selectively. Finally, solid-phase peptide synthesis is
well developed and allows rapid, parallel preparation of
numerous analogues on a relatively large scale. Herein,
we report synthesis and characterization of hexacrown
peptide 1 and its analogues 2–5 as well as studies on
their activity, selectivity, mode of action, and cytotoxi-
city.
Having access to a series of crown ether amino acids, we
devised an efficient synthetic strategy for the peptide
nanostructure. To simplify the synthesis of nanos-
tructures like 1, we envisioned that the 21 amino acid
peptide could come from the trimerization of a hepta-
peptide, where crown ether amino acids are in positions
2 and 6. The choice of a repeating heptad minimizes the
effort in the synthesis if the desired peptide is to be pre-
pared by segment condensation.24,25 Indeed, to avoid
the tedious purification steps associated with a stepwise
synthesis of a 21-residue peptide, we adopted a strategy
involving selective coupling of peptide segments. The
basic idea was to prepare first a seven-residue peptide 28
on oxime resin, and then proceed to its self-condensa-
tion, also on solid support, to afford a 14-mer, then the
desired 21-mer (Fig. 3).
2. Results and discussion
2.1. Synthesis
Molecule 1 represents a prototype of our approach
towards artificial ion channels. It is a 21 amino acid
peptide composed of 15 l-leucines and six 21-crown-7
l-phenylalanines. Incorporation of the modified amino
Coupling reactions to prepare heptapeptide 28 were
performed using N-BOC amino acids and DIC/HOBT
Figure 1. Crown peptide nanostructures synthesized and used in the present studies.

E. Biron et al. / Bioorg. Med. Chem. 12 (2004) 1279–1290
1281
Figure 2. Synthesis of N-t-BOC protected crown ether amino acids: (a) PPh3, Br2, CH3CN; (b) Et3N, TrCl, DMAP, CH2Cl2; (c) NaH, HO–CH2–
(CH2)k–OH (k=1, 2), DMF; (d) 6 N HCl/dioxane, CH2Cl2; (e) TsCl, Py; (f) BOC-l-DOPA-OMe, Cs2CO3, DMF; (g) 1 N NaOH/MeOH (CE,
crown ether ring).
2.2. Conformational studies
as reagents in DCM/DMF (1:1). The N-BOC group was
removed between each coupling by using 50% tri-
fluoroacetic acid (TFA) solution in DCM. After pre-
paration of the key heptapeptide 28 on solid support, a
fraction (3/4) of the total amount of resin was cleaved
with water and DBU in THF to generate the N-BOC
heptapeptide acid 29. On the remaining resin, the N-
BOC group was removed by treatment with TFA.
Coupling of N-BOC heptapeptide acid 29 was per-
formed using DIC/HOBT in DMF to yield the 14-resi-
due intermediate. It is noteworthy that these coupling
conditions are the ones that gave the lowest racemiza-
tion (<10%), as determined by RP-HPLC after cleav-
ing an analytical amount of peptides from the resin.
Indeed, any other conditions attempted (HBTU,
HATU, BOP) led to more epimerization. TFA depro-
tection of the 14-mer followed by another coupling of
segment 29 generated efficiently the desired 21-residue
peptide still attached to the resin. Finally, the peptide
was cleaved from the resin using MeOH in the presence
of 2 equivalent of DBU to yield the 21 amino acid pep-
tide 1. The use of DBU in MeOH for cleavage reactions
was first reported for the cleavage of peptides from
Wang resin and turned out to be very useful to the
construction of peptide ester from oxime resin.26 The
purification of 1 and its analogues (2–5, 31) was easily
achieved by reversed-phase HPLC, and they were char-
Circular dichroism (CD) studies have previously shown
that peptides 1–5 adopt an a-helical conformation in
different media, protic (TFE, MeOH) as well as hydro-
phobic (1,2-dichloroethane), and at different concentra-
tions in TFE (Fig. 4 left).27,28 We have also shown that
these peptides keep their a-helical conformation when
incorporated in lipid bilayers (Fig. 4 left).27,28 These
studies were supported by FTIR spectroscopy studies
using DPPC vesicles and egg yolk lecithin (Fig. 4
right).28 These results suggest that the poly-l-leucine
framework dictates the conformation of these nano-
structures and also that the crown rings are on the same
side of the peptidic framework as predicted initially.
They also point out that compound 1 and its analogues
does not tend to aggregate in low polarity environment
as demonstrated by the lack of change in the ellipticity
at 222 nm at different concentrations (0.1 mM–0.1
mM).27 Likewise, it is noteworthy that CD studies
showed that the conformation of 2 is not affected by the
presence of metal ions or water (data not shown). These
observations indicate clearly the possibility of describ-
ing accurately peptide nanostructures from basic fun-
damental principles.
2.3. Ion transport studies
1
acterized successfully by H NMR and mass spectro-
metry. Thus peptide nanostructures bearing six crown
ethers were efficiently prepared in pure overall yields
higher than 10% generally.
The ion transport ability of hexacrown peptides were
investigated using different techniques. Our first attempt
used the pH stat method.29 In this assay, a sample of

1282
E. Biron et al. / Bioorg. Med. Chem. 12 (2004) 1279–1290
Figure 3. General synthetic strategy for the preparation of crown peptide nanostructures on oxime resin illustrated by the synthesis of the
hexacrown peptide 1.
Figure 4. Left: CD spectra of (A) 21-mer 1, (B) 14-mer 31, (C) 7-mer 30 in TFE, as well as (D) 1 in egg yolk lecithin (80:1 w/w lipid/peptide) (5Â10À5
M) in MOPS 10 mM, 20% TFE pH 7.4. Right: FTIR spectra of 1 in TFE (A), and in lecithin vesicles (B).
unilamellar vesicles with an internal pH of 6.6 is diluted
with an external solution, and the pH is raised to 7.6 to
troneutrality across the vesicle wall (Fig. 5a). The pro-
tons are neutralized continuously to maintain
constant solution pH of 7.6. The graph of the volume of
base added versus time allows the determination of the
transport mode.30,32 A typical result obtained is illu-
strated in Figure 5b for the case of Cs+. As it can be
a
create a proton gradient.30 Then, ions and FCCP,31
a
proton carrier, are added. Upon addition of a func-
tional transporter, its transport ability can be monitored
by the release of protons required to maintain the elec-

E. Biron et al. / Bioorg. Med. Chem. 12 (2004) 1279–1290
1283
lacking the crown ring, did not result in any sig-
nificant transport. Even though, this is in agreement
with the channel-like mechanism of 1 and ions tra-
velling the crown channel, the absence of ion transport
with 6Phe could also be the result of its insolubility in
the media. Interestingly, addition of Triton X-100 at the
end of every experiment led to complete lysis of vesicles
and release of entrapped H+. This phenomenon
demonstrates that vesicles were still intact at the end of
experiments, which strongly suggests that 21-mer pep-
tide 1 is not acting as a sophisticated surfactant. This is
further supported by the physical observation that
addition of 1 to a vesicle suspension does not altered the
opalescent look of the solution. By contrast, addition of
5 mL of a 0.5 mM Triton X-100 solution immediately
clarifies the solution.
It is noteworthy that in the case of Gramicidin A and 1,
proton release is not complete. Indeed, in every case
with these functional compounds transport reaches a
plateau at roughly 80% of maximum theoretical trans-
port. This can be explained by the fact that some of the
vesicles are multilamellar and hence addition of trans-
porter can effect the release of protons only from the
outer layer and not from the inner ones.
In order to characterize further the transport ability of
1, we carried out single-channel measurements33 using a
modified patch clamp technique.34 To demonstrate the
ion channel activity, 2 mL of a 0.1 mM solution of 1 in
methanol was introduced in a 2-mL cell containing an
unbuffered 1 M KCl solution (pH 5.8); the bilayer was
constructed on a pipette tip from zwitterionic DPPC (5
mg/mL in decane). As initial experiments, only K+
transport was studied as this ion is weakly bound by 21-
C-7 rings and therefore should traverse with great facil-
ity channel formed by the crown ether array. The con-
ductance recordings shown in Figure 5c are
characteristic of the results obtained with 1 and typical
of ion channel activity observed with natural pro-
teins.27,35 Typical transitions (‘steps’) between closed
and open states were evident throughout experiments.
The discrete conductance depend on the number of
channels in the open state.33 In DPPC bilayers, com-
pound 1 has a current level of 3.5Æ0.2 pA with lifetimes
for the open states of around 1 s. These results corro-
borate findings from the pH stat studies. However,
more experiments are required with different ions and
analogues 2–5 to quantify accurately their ion channel
activity.
Figure 5. (a) Schematic generalization of the pH-stat method. The
transport ability of test compounds is monitored by the release of
protons from the vesicles as a function of time. (b) Illustrative example
of results obtained using the pH stat technique. Cs+ transport ability
of monomeric crown ether 17 (~), heptapeptide 30 (&), hexacrown
peptide 1 (*), and Gramicidin A (*) (reprinted with permission of
ACS).29 (c) Single-channel conductance recordings of peptide 1 in 1 M
KCl (at À100 mV, filtered at 200 Hz) in a DPPC bilayer (current level
3.5Æ0.2 pA, average lifetimes of open states >1 s (reprinted with
permission of Wiley-VCH).27
seen, the monomeric crown ether (17) and the hepta-
peptide analogue (30), too short to span the membrane,
act as typical carriers by slowly and constantly trans-
porting Cs+ at roughly similar rate. However, addition
of the hexa-crown peptide 1 led to a very rapid release
of protons that reached saturation after less than 2 min.
For comparison purposes, we studied Gramicidin A, a
natural compound known to form exceptionally effi-
cient channels. As it can be seen, hexacrown peptide 1 is
as efficient as Gramicidin A for the transport of Cs+.
This result suggests that 1 functions as an artificial ion
channel in a manner analogous to Gramicidin A. Fur-
thermore, the hexa-crown peptide 1 exhibited the same
transport activity with Li+, Na+, K+ and Rb+.
Although these results support the channel mechanism
of 1 and that ions with diameter smaller than Cs+ are
expected to travel the channel formed by the large 21-C-
7 ring array, it clearly indicates the deficiency of the pH
stat technique to assess the ion selectivity of artificial,
membrane active compounds. Several control experi-
ments were performed to confirm the channel-like
mechanism of 1. Changing the order of addition of the
components (i.e., salts, FCCP, transporter) gave the
same results. Studies using 6Phe, the analogue of 1
In search of a practical method to screen many analo-
gues for channel activity and to further confirm the
activity of peptide nanostructure 1, we used a dynamic
NMR method. Developed by Riddell and Hayer36 and
devoided of any driving force, this method permit the
assessment of the efficiency of sodium cation transport.
In this method, sodium ions trapped inside vesicles,
composed of phosphatidylcholine and phosphatidylgly-
cerol, are distinguished from external sodium cations by
adding a shift reagent (usually Dy3+). Addition of a
functional transporter leads to an alteration of the sig-
nal width caused by an exchange of sodium cations

1284
E. Biron et al. / Bioorg. Med. Chem. 12 (2004) 1279–1290
across the bilayer. These variations can be related to the
transport rate. This methodology, successfully applied
to Gramicidin D,37 was also exploited to evaluate the
efficiency of artificial hydraphile channels.38 One benefit
of this method is the obtention of reliable relative rates.
The NMR experiments reported here are normalized to
a relative rate of 100 representing activity of Gramicidin
D, our standard in this study. The results of 23Na NMR
transport experiments are shown in Table 1.
30, 14-mer 31 and 21-mer 1 to unilamellar vesicles pre-
pared from egg phosphatidylcholine and egg phospha-
tidylglycerol (1:2). Clearly the 21-mer 1 and 7-mer 30
induce very little leakage. Beside, the 14-mer 31 induces
rapid and significant lysis. This unique behaviour of 31
may be the result of an hydrophobic mismatch39 that
introduces weaknesses in the bilayer and lead to lysis.
Contrarily, compound 1 has a definite transport activity
but does not induce lysis confirming that the observa-
tion made in the 23Na NMR experiments is indeed due
to ion transport activity. These results show that cau-
tion should be exerted when using 23Na NMR to probe
ion channel activity, as it can not discriminate between
transport and lysis. Finally, the Na+ transport activity
of 1, even though much weaker than Gramicidin D, is
encouraging as this compound correspond to our start-
ing point in channel development and that it can be
easily optimized, particularly with termini modifications
with polar heads.
We can observe that activity of peptide nanostructure 1
having 21-crown-7 macrocycles (3% of Gramicidine D)
is 30 times higher than its analogue 2, with smaller
macrocycles, and more than 600 times higher than 4 and
5. This strongly supports the hypothesis that cations
travel through the channel formed by the array of mac-
rocycles in order to cross the bilayer membrane. In fact,
cavity of 14-crown-4 and 13-crown-4 is too small to let
sodium cations passing through, while 18-crown-6
seems to have enough binding affinity for Na+ to
diminish partially the channel activity of nanostructure
2. The two most active compounds are 1 and 31,
respectively the 21-mer and the 14-mer built with 21-C-7
amino acids. Surprisingly, shorter analogue 31 is not
long enough to span entirely the lipid bilayer. We rea-
soned that this unexpected result could be explained by
the ability of 31 to induce vesicle lysis, which indeed can
present in 23Na NMR experiments, results similar to
those obtained for cation transport. To verify this pos-
sibility, control experiments were performed using a
fluorescent vesicle lysis assay.39 Shown in Figure 6 are
the calcein release profiles induced by addition of 7-mer
2.4. Mechanism of ion transport
In order to characterize in more details the ion transport
mechanism of 1, we undertook a thorough biophysical
investigation using ATR spectroscopy. Polarized infra-
red ATR spectra of oriented films of DMPC and DPPC
in the absence and presence of crown peptide 1 and
analogues, and at different concentrations, were recor-
ded.28 These experiments led us to conclude that hex-
acrown peptides are in a two-state incorporation
equilibrium, one active parallel to the lipids chains and
the other inactive at 90ꢀ of the same chains. These data
combined with the ones previously obtained by CD and
FTIR spectroscopy allow us to propose a model for the
formation of active ion channels by crown peptide 1
(Fig. 7).28 First, when added to the bilayer system from
MeOH solution, crown peptide 1 and its analogues
precipitate and form solid aggregates. Then, one crown
peptide from the aggregates adsorbs at the surface of
the membrane, most probably through electrostatic
interactions between the crown ether units and the lipid
polar groups. It is also possible that hydrophobic inter-
actions between leucine side chains and the lipid chains
Figure 6. Calcein leakage induced by addition of 7-mer 30,14-mer 31,
and 21-mer 1 (5 mM) to phosphatidylglycerol/phosphatidylcholine
(1:1) vesicles in HEPES 100 mM, NaCl 170 mM, EDTA 5 mM at pH
7.4. Vesicles were lysed with Triton X-100 at 400 s.
Table 1. Rate constants (k) and relative transport rates for Na+
across a phospholipid bilayer membrane by 23Na NMR
Compd
k (sÀ1 mMÀ1
)
Relative rate (%)
Gramicidin D
240
6.6
9.2
0.2
<0.1
<0.1
100
3
4
0.1
<0.05
<0.05
Figure 7. Proposed mode of action for the channel activity of peptide
1. From the aggregates formed in the bulk water solution, one crown
peptide adsorbs at the bilayer surface (A) and is in the adsorbed inac-
tive form. From this partially incorporated form, peptide reorients
itself into the active form B (reprinted with permission of Wiley-
Interscience).28
Boc-21mer-Ome (21-C-7) 1
Boc-14mer-Ome (21-C-7) 31
Boc-21mer-Ome (18-C-6) 2
Boc-21mer-Ome (14-C-4) 4
Boc-21mer-Ome (13-C-4) 5

E. Biron et al. / Bioorg. Med. Chem. 12 (2004) 1279–1290
1285
act as driving force for the initial adsorption. Finally,
under the influence of an electrical potential or an ionic
gradient, one of the adsorbed peptide nanostructure
flips over parallel to the lipid chains and channel activity
can be observed. This mechanism is very close to the
one described by DeGrado et al. for self assembly heli-
cal peptides40,41 and to the barrel-stave mechanism of
some antimicrobial peptides.42
tion of molecular devices.43 The studied hexacrown
peptides 1–5 keep an helical conformation indepen-
dently of the crown ether size or the environment
(MeOH, TFE, DCE or bilayer membranes). Further-
more, they displayed variable ion transport abilities
across membranes in different experiments (pH-stat,
patch clamp and 23Na NMR). Also some ion selectivity
was observed in the 23Na NMR experiments, strongly
suggesting that ions pass through the channel formed by
crown array and not through defects in bilayers caused
by undefined peptide aggregates. Vesicles lysis assays
have demonstrated that the activity observed does not
come from membrane disruption. Conformational and
orientational studies led us to propose a gating
mechanism for the channel activity of compound 1 and
its analogues. On the other hand, peptides 1 and 3
showed very interesting cytotoxic activities towards
cancer cells. Taking advantage of the simplicity and
versatility of our synthetic strategy, we are currently
synthesizing improved analogues of 1 with different
polar heads and reporter groups at C- and N-termini.
Finally, in addition to our ongoing fundamental inves-
tigation of these peptide nanostructures, we are also
working on the engineering of 1 into workable devices
for the detection of important analytes at the molecular
level and continuing their bioactivity investigation.
2.5. Biological activity
As hexacrown compounds mimic properties of natural
membrane active peptides and proteins, we sought to
investigate their bioactivity. Specifically, we were inter-
ested in verifying their antimicrobial activity and cyto-
toxicity. Hence, biological activity of peptide
nanostructures 1 and 3 was tested against bacteria and
mammalian cells. No antimicrobial activity was found
for peptide 1 with several bacteria, both gram+ and
gramÀ bacteria. It appears that presence of positive
charges along the peptide backbone is necessary for
antimicrobial activity.
On the other hand, cytotoxicity of different crown pep-
tides was measured against breast cancer cells (MDA)
and mouse leukemia cells (P388). Results are reported
in Table 2. Interestingly, 21-mer 1 and 3 peptides
showed cytotoxic activities against MDA and P388
cells, while 30, the 7-residue analogue of 1, and 17, the
monomeric crown ether modified amino acid, were
totally inactive. These results point out the necessity of
having nanostructures of appropriate length (3–4 nm) in
order to efficiently form a membrane pore and become
cytotoxic agents. Noteworthy of mention, the depro-
tected analogue of 1 bearing charged head groups is
much less active than 1 itself (last two entries of Table
2). These charged compounds were also shown not to
incorporate easily bilayer membrane (unpublished
results).
4. Experimental
4.1. General
Oxime resin was prepared according to a reported pro-
cedure44 using polystyrene beads (100–200 mesh, 1%
DVB, Advanced ChemTech, Louisville, KY, USA).
Resins with substitution levels of around 0.5 mmol per
gram of oxime group were used. Boc-protected amino
acids were purchased from Advanced ChemTech. All
solvents were Reagent, Spectro, or HPLC grade quality
purchased commercially and used without any further
purification except for DMF (degassed with N2), di-
chloromethane (distilled), and diethyl ether (distilled
from sodium and benzophenone). Water used through-
out the studies was distilled and deionized using a
Barnstead NANOpurII system (Boston, MA, USA)
with four purification columns. Phosphatidylglycerol
(20 mg/mL solution in CHCl3), phosphatidylcholine (20
mg/mL solution in CHCl3) were purchased from Avanti
Polar-Lipids and used without further purification. D2O
was purchased from CDN Isotopes (Pointe-Claire, QC,
Canada) and used without further purification. All
other reagents were purchased from Sigma Aldrich Co.
(Milwauke, WI, USA). Solid phase peptide synthesis
was performed manually using solid-phase reaction
vessels equipped with a coarse glass frit (ChemGlass,
Vineland, NJ, USA). Purification of peptides was per-
formed by reversed phase HPLC with a C4 semi-
preparative column (Phenomenex, Torrance, CA,
USA). All solvents were degassed and gradients of A
(H2O/0.1% TFA) and B (49.9% CH3CN/49.9% iso-
propanol/0.1% TFA) were used. 1H NMR spectra were
recorded on a Bruker AC-F-300 spectrometer. Sonica-
tion was done using a Branson water bath model 3510.
3. Conclusion
We have described a simple and efficient strategy to
prepare new nanoscale devices with well defined molec-
ular architectures that serve as functional synthetic ion
channels. The results reported illustrate well the utility
of peptide nanostructures as scaffolds for the construc-
Table 2. Cytotoxicity of different peptide nanostructures on cancer
cells
Compd
MDA
LD50 (mM)
P388
LD50 (mM)
Gramicidin D
0.2
NDa
15.0
4.0
35.0
8.5
2.5
ND
ND
BOC-7mer-OMe(21-C-7) 30
BOC-21mer-Ome(21-C-7) 1
BOC-21mer-Ome(15-C-5) 3
BOC-21-C-7-OMe 17
H-21mer-EDA(21-C-7)b
Pyr-21mer-OH(21-C-7)c
10.0
ND
15.0
>30.0
>40.0
a ND, not determined due to low toxicity.
b Pyr, pyromellitic acid.
c EDA, ethylene diamine.

1286
E. Biron et al. / Bioorg. Med. Chem. 12 (2004) 1279–1290
Mass spectra were obtained from the Mass Spectro-
metry Laboratory of the Faculty of Medecine at the
University of Toronto, Toronto, Canada.
4.2.5. N-t-BOC-18-C-6-OMe 18. Prepared as described
for compound 17 using the dibromide 7 (44.5 g, 56%).
1H NMR (CDCl3), d 6.80 (m, 1H, H Ar (H2)), 6.69–6.61
(m, 2H, H Ar (H5 +H6)), 4.98 (d, J=7.9 Hz, 1H, NH),
4.52 (m, 1H, aCH), 4.10 (m, 4H, CH2-OAr), 3.88 (m,
4H, CH2–CH2–OAr), 3.78-3.63 (m, 12H, O–CH2–CH2–
O), 3.68 (s, 3H, OCH3), 3.01 (m, 2H, bCH2), 1.40 (s,
9H, H t-Bu). ESMS m/z (M+Na+) 536.
4.2. Crown ether synthesis
4.2.1. 1,17-Dibromo-3,6,9,12,15-pentaoxaheptadecane 6.
Triphenylphosphine (92.9 g, 354 mmol) was suspended
in acetonitrile (400 mL) under N2 at 0 ꢀC. Bromine (56.6
g, 354 mmol) was added during 45 min. Hexaethylene
glycol (50.0 g, 177 mmol) was dissolved in acetonitrile
(60 mL) and added dropwise. Reaction was stirred dur-
ing 48 h at room temperature. The white residue was
eliminated by filtration and the solvent was evaporated.
The resulting pasty orange-coloured residue was
extracted several times with hexane. After evaporation
of the combined extracts, the dibromide was obtained as
4.2.6. N-t-BOC-18-C-6-OH 23. Prepared as described
for compound 22 from crown ether amino ester 18 (38.9
g, 90%). 1H NMR (CDCl3), d 6.77–6.70 (m, 3H, H Ar),
5.42 (m, 1H, OH), 5.20 (d, J=7.9 Hz, 1H, NH), 4.48
(m, 1H, aCH), 4.22–3.92 (m, 4H, CH2–OAr), 3.91–3.78
(m, 4H, CH2–CH2–OAr), 3.77–3.68 (m, 12H, O–CH2–
CH2–O), 3.01 (m, 2H, bCH2), 1.40 (s, 9H, H t-Bu).
ESMS m/z (M+) 499.
1
colourless oil (60.5 g, 84%). H NMR (CDCl3), d 3.71–
3.66 (t, J=6.5 Hz, 4H, CH2–CH2–Br), 3.53 (m, 16H, O–
CH2–CH2–O), 3.37–3.33 (t, J=6.5 Hz, 4H, CH2–Br).
4.2.7. 1,11-Dibromo-3,6,9-trioxaundecane 8. Prepared as
described for compound 6 from tetraetylene glycol (51.5
1
g, 88%). H NMR (CDCl3), d: 3.72–3.68 (t, J=6.5 Hz,
4H, CH2–CH2–Br), 3.56 (m, 8H, O–CH2–CH2–O),
3.39–3.34 (t, J=6.5 Hz, 4H, CH2–Br).
4.2.2. N-t-BOC-21-C-7-OMe 17. BOC-l-DOPA-OMe45
(25.0 g, 80.3 mmol) was dissolved in DMF (800 mL)
under N2. Anhydrous cesium carbonate (26.2 g, 80.4
mmol) was added and the solution was brought to
60 ꢀC. Dibromide 6 (32.8 g, 80.4 mmol) was added dur-
ing 2 h and the suspension was stirred for 24 h at 60 ꢀC.
DMF was evaporated in vacuo and brown residue was
dissolved in CH2Cl2 (200 mL) and washed with 1 N
NaOH and NaCl satd. The organic phase was dried
over MgSO4 and filtered. The oil obtained was purified
by flash column chromatography (silica gel; 2–6%
MeOH in CH2Cl2). The compound was precipitated as
a beige powder (20.5 g, 46%) by triturating in petro-
4.2.8. N-t-BOC-15-C-5-OMe 19. Prepared as described
for compound 17 using the dibromide 8 (55.0 g, 78%).
1H NMR (CDCl3), d 6.81 (m, 1H, H Ar (H2)), 6.64–6.61
(m, 2H, H Ar (H5 +H6)), 4.95 (d, J=7.9 Hz, 1H, NH),
4.53 (m, 1H, aCH), 4.11–4.08 (m, 4H, CH2–OAr), 3.90–
3.88 (m, 4H, CH2–CH2–OAr), 3.75–3.69 (m, 8H, O–
CH2–CH2–O), 3.68 (s, 3H, OCH3), 2.98 (m, 2H, bCH2),
1.41 (s, 9H, H t-Bu). ESMS m/z (M+) 469.
4.2.9. N-t-BOC-15-C-5-OH 24. Prepared as described
for compound 22 from crown ether amino ester 19 (43.7
g, 82%). 1H NMR (CDCl3), d 6.78–6.69 (m, 3H, H Ar),
5.37 (m, 1H, OH), 5.03 (d, J=7.7 Hz, 1H, NH), 4.51
(m, 1H, aCH), 4.09 (m, 4H, CH2–OAr), 3.88 (m, 4H,
CH2–CH2–OAr), 3.75–3.73 (m, 8H, O–CH2–CH2–O),
3.02 (m, 2H, bCH2), 1.41 (s, 9H, H t-Bu). ESMS m/z
(M+) 455.
1
leum ether. H NMR (CDCl3), d 6.80 (m, 1H, H Ar
(H2)), 6.75–6.65 (m, 2H, H Ar (H5+H6)), 5.02 (d,
J=7.9 Hz, 1H, NH), 4.52 (m, 1H, aCH), 4.08–4.03 (m,
4H, CH2–OAr), 3.86–3.82 (m, 4H, CH2–CH2–OAr),
3.78–3.55 (m, 16H, O–CH2–CH2–O), 3.45 (s, 3H,
OCH3), 3.05–3.01 (m, 2H, bCH2), 1.40 (s, 9H, H t-Bu).
ESMS m/z (M+) 558.
4.2.3. N-t-BOC-21-C-7-OH 22. The crown ether amino
ester 17 (20.4 g, 36.6 mmol) was dissolved in MeOH (50
mL) and cooled to 0 ꢀC. 1 N NaOH (60 mL) was added
and the mixture was stirred at room temperature for 3
h. The alkaline solution was washed with diethyl ether
and acidified to pH 2–3 with 1 N HCl. The product was
extracted with CH2Cl2, dried over MgSO4 and filtered.
The compound was precipitated as a beige powder (17.6
g, 88%) by triturating in petroleum ether. 1H NMR
(CDCl3), d 6.78 (m, 1H, H Ar (H2)), 6.65–6.60 (m, 2H,
H Ar (H5+H6)), 4.95 (d, J=7.9 Hz, 1H, NH), 4.48 (m,
1H, aCH), 4.07 (m, 4H, CH2–OAr), 3.85 (m, 4H, CH2–
CH2–OAr), 3.78–3.60 (m, 16H, O–CH2–CH2–O), 2.99
(m, 2H, bCH2), 1.40 (s, 9H, H t-Bu). ESMS m/z (M+)
543.
4.2.10. 2-Triphenylmethoxy-1-chloroethane 9. Chloro-
ethanol (21.8 mL, 326 mmol) was added to a flask con-
taining CH2Cl2 (500 mL). Triethylamine (50 mL, 358
mmol), chlorotriphenylmethane (100.0 g, 358 mmol),
and 4-dimethylaminopyridine (catalytic amount) were
added and this resulting mixture was stirred 24 h at
room temperature. The organic phase was washed with
NaHCO3 5% and H2O, then was dried over MgSO4 and
filtered. CH2Cl2 was evaporated and the solid obtained
was recrystallized in EtOH to give white crystals (103.3
1
g, 98%). H NMR (CDCl3), d 7.49–7.45 (d, J=7.2 Hz,
6H,
H Ar (H2,H6)), 7.33–7.22 (m, 9H, H Ar
(H3,H4,H5)), 3.60–3.57 (t, J=5.9 Hz, 2H, CH2–Cl),
3.40–3.36 (t, J=5.9 Hz, 2H, CH2–OTr). ESMS m/z
(M+) 322.
4.2.4. 1,14-Dibromo-3,6,9,12-tetraoxatetradecane 7. Pre-
pared as described for compound 6 from pentaetylene
glycol (66.2 g, 88%). H NMR (CDCl3), d 3.82–3.77 (t,
J=6.5 Hz, 4H, CH2–CH2–Br), 3.65 (m, 12H, O–CH2–
CH2–O), 3.48–3.44 (t, J=6.5 Hz, 4H, CH2–Br).
4.2.11. 1,9-Di(triphenylmethoxy)-3,7-dioxanonane 11.
Sodium hydride (8.3 g, 346.3 mmol) was added to 1,3-
propanediol (10 mL, 138.5 mmol) previously dissolved
in DMF (250 mL). The mixture was stirred during 30
min at 50 ꢀC. The temperature was lowered to 45 ꢀC and
1

E. Biron et al. / Bioorg. Med. Chem. 12 (2004) 1279–1290
1287
the protected chloroethanol 9 (98.4 g, 304.7 mmol), dis-
solved in DMF by sonication, was added to the reaction
along with sodium iodide (catalytic amount). The mix-
ture was stirred 48 h at 45 ꢀC, then the reaction was
quenched with water. The crude product was extracted
three times with a solution of hexanes/diethyl ether 1:1.
The organic phase was washed with satd NaCl and
H2O, dried over MgSO4, filtered, and evaporated in
CH2–CH2–CH2–O), 3.77–3.75 (m, 8H, 2 CH2–OAr+2
CH2–CH2–OAr), 3.07 (m, 2H, bCH2), 1.82–1.78 (qn,
2H, O–CH2–CH2–CH2–O), 1.35 (s, 9H, H t-Bu). ESMS
m/z (M+) 425.
4.2.16. 3-Triphenylmethoxy-1-chloropropane 10. Pre-
pared as described for compound 9 from 3-chloro-
1
propanol (104.3 g, 95%). H NMR (CDCl3), d: 7.45–
1
vacuo to obtain a white solid (76.4 g, 85%). H NMR
7.41 (d, J=7.5 Hz, 6H, H Ar (H2,H6)), 7.33–7.21 (m,
9H, H Ar (H3,H4,H5)), 3.73–3.68 (t, J=6.7 Hz, 2H,
CH2–Cl), 3.24–3.21 (t, J=5.9 Hz, 2H, CH2–OTr), 2.07–
1.99 (qn, J=6.2 Hz, 2H, Cl–CH2–CH2–CH2–O).
(CDCl3), d 7.49–7.46 (d, J=7.0 Hz, 12H, H Ar
(H2,H6)), 7.30–7.18 (m, 18H, H Ar (H3,H4,H5)), 3.64-
3.59 (m, 8H, O–CH2–CH2–O), 3.23–3.20 (t, J=5.3 Hz,
4H, O–CH2–CH2–CH2–O), 1.95–1.87 (qn, J=6.4 Hz,
2H, O–CH2–CH2–CH2–O). ESMS m/z (M+) 648.
4.2.17. 1,10-Di(triphenylmethoxy)-4,7-dioxadecane 12.
Prepared as described for compound 11 from ethylene
glycol and the protected chloropropanol 10 (59.1 g,
4.2.12. 3,7-Dioxanonane-1,9-diol 13. Diprotected poly-
ether 11 (75 g, 115 mmol) was dissolved in CH2Cl2. A
solution of 6 N HCl/dioxane was added and the solution
was stirred for 30 min. MeOH was added and the mix-
ture was stirred until appearance of a white residue,
which was filtered off. Evaporation led to yellow oil
which was first purified by flash chromatography (silica
gel; CH2Cl2/acetone). The resulting beige oil was dis-
tilled in vacuo (bp 118 ꢀC at 1 mm Hg) to give 13 as a
1
60%). H NMR (CDCl3), d 7.49–7.46 (d, 12H, H Ar
(H2,H6)), 7.30–7.18 (m, 18H, H Ar (H3,H4,H5)), 3.64–
3.59 (t, 4H, O–CH2–CH2–O), 3.35–3.30 (t, 4H, O–CH2–
CH2–CH2–OTr), 3.23–3.20 (t, 4H, O–CH2–CH2–CH2–
OTr), 1.95–1.90 (qn, 4H, O–CH2–CH2–CH2–O).
4.2.18. 4,7-Dioxadecane-1,10-diol 14. Prepared as
described for compound 13 using the diprotected poly-
1
1
colourless oil (17.2 g, 90%). H NMR (CDCl3), d 3.67–
ether 12 (12.7 g, 80%). H NMR (CDCl3), d 3.67–3.64
3.64 (t, J=4.5 Hz, 4H, O–CH2–CH2–OH), 3.56–3.52 (t,
J=6.1 Hz, 4H, O–CH2–CH2–OH), 3.50–3.47 (t, J=4.6
Hz, 4H, O–CH2–CH2–CH2–O), 3.06 (s, 2H, OH), 1.86–
1.77 (qn, J=6.1 Hz, 2H, O–CH2–CH2–CH2–O). ESMS
m/z (M+H+) 165.
(t, 4H, O–CH2–CH2–O), 3.56–3.52 (t, 4H, O–CH2–
CH2–CH2–OH), 3.50–3.47 (t, 4H, O–CH2–CH2–CH2–
OH), 3.06 (s, 2H, OH), 1.95–1.90 (qn, 4H, O–CH2–
CH2–CH2–O).
4.2.19. 1,10-DitosyL-4,7-dioxadecane 16. Prepared as
described for compound 15 from diol 14 (31.0 g, 95%).
1H NMR (CDCl3), d 7.79–7.76 (d, 4H, H Ar (H2,H6)),
7.33–7.30 (d, 4H, H Ar (H3,H5)), 4.14–4.11 (t, 4H, O–
CH2–CH2–O), 3.60–3.56 (t, 4H, O–CH2–CH2–CH2–
OTs), 3.44–3.40 (t, 4H, O–CH2–CH2–CH2–OTs), 2.43
(s, 6H, CH3–Ar), 1.72–1.64 (qn, 4H, O–CH2–CH2–
CH2-O).
4.2.13. 1,9-DitosyL-3,7-dioxanonane 15. Diol 13 (17 g,
103 mmol) was dissolved in pyridine at 0 ꢀC and tosyl
chroride (43.2 g, 227 mmol) in pyridine was added. The
mixture was stirred 4 h at 4 ꢀC, then was transferred in a
beaker filled with ice. 6 N HCl (500 mL) was added. The
resulting mix was extracted with CH2Cl2 and the
organic phase was washed with 2 N HCl and H2O, dried
over MgSO4, and filtered. After evaporation of the sol-
1
vent, 15 was obtained as a beige oil (48.1 g, 99%). H
4.2.20. N-t-BOC-14-C-4-OMe 20. Prepared as descri-
bed for compound 17 using the ditosylate 16 (14.2 g,
NMR (CDCl3), d 7.78–7.77 (d, J=8.4 Hz, 4H, H Ar
(H2,H6)), 7.34–7.32 (d, J=8.1 Hz, 4H, H Ar (H3,H5)),
4.14–4.11 (t, J=4.7 Hz, 4H, O–CH2–CH2–OTs), 3.60–
3.56 (t, J=4.7 Hz, 4H, O–CH2–CH2–OTs), 3.44–3.40 (t,
J=6.3 Hz, 4H, O–CH2–CH2–CH2–O), 2.43 (s, 6H,
CH3–Ar), 1.74–1.66 (qn, J=6.1 Hz, 2H, O–CH2–CH2–
CH2–O). ESMS m/z (M+) 472.
1
49%). H NMR (CDCl3), d 6.83 (m, 1H, H Ar (H2)),
6.70–6.63 m, 2H, H Ar (H5+H6)), 4.98 (d, J=7.7 Hz,
1H, NH), 4.58 (m, 1H, aCH), 4.26 (t, 4H, O–CH2–
CH2–O), 4.05 (m, 4H, CH2–OAr), 3.75 (m, 4H, CH2–
CH2–CH2–OAr), 3.62 (s, 3H, OCH3), 3.00 (m, 2H,
bCH2), 1.84–1.78 (qn, 4H, O–CH2–CH2–CH2–O), 1.35
(s, 9H, H t-Bu).
4.2.14. N-t-BOC-13-C-4-OMe 21. Prepared as descri-
bed for compound 17 using the ditosylate 15 (24.6 g,
56%). 1H NMR (CDCl3), 6.85 (m, 1H, H Ar (H2)),
6.70–6.65 (m, 2H, H Ar (H5 +H6)), 4.95 (d, J=7.7 Hz,
1H, NH), 4.55 (m, 1H, aCH), 4.16 (t, 4H, O–CH2–
CH2–CH2–O), 4.05 (m, 4H, CH2–OAr), 3.75 (m, 4H,
CH2–CH2–OAr), 3.62 (s, 3H, OCH3), 3.00 (m, 2H,
bCH2), 1.85–1.78 (qn, 2H, O–CH2–CH2–CH2–O), 1.35
(s, 9H, H t-Bu). ESMS m/z (M+) 439.
4.2.21. N-t-BOC-14-C-4-OH 25. Prepared as described
for compound 22 using the crown ether amino ester 20
(11.5 g, 85%). H NMR (CDCl3), d 6.83 (m, 1H, H Ar
(H2)), 6.70–6.63 (m, 2H, H Ar (H5+H6)), 5.00 (d,
J=7.7 Hz, 1H, NH), 4.59 (m, 1H, aCH), 4.27 (t, 4H, O–
CH2–CH2–O), 4.09 (m, 4H, CH2–OAr), 3.76 (m, 4H,
CH2–CH2–CH2–OAr), 3.00 (m, 2H, bCH2), 1.84–1.76
(qn, 4H, O–CH2–CH2–CH2–O), 1.35 (s, 9H, H t-Bu).
ESMS m/z (M+NH+4 ) 457.
1
4.2.15. N-t-BOC-13-C-4-OH 26. Prepared as described
for compound 22 using the crown ether amino ester 21
(21.1 g, 90%). H NMR (CDCl3), d 6.88 (m, 1H, H Ar
(H2)), 6.79–6.75 (m, 2H, H Ar (H5+H6)), 4.93 (d,
4.3. Solid-phase synthesis
1
4.3.1. Typical procedure for amino acid coupling on solid
support. The amino acid (5 equiv) was activated with
J=7.7 Hz, 1H, NH), 4.53 (m, 1H, aCH), 4.11 (t, 4H, O–

1288
E. Biron et al. / Bioorg. Med. Chem. 12 (2004) 1279–1290
DIC/HOBt during 30 min at 0C in CH2Cl2/DMF (1:1),
then added to the resin swollen in DMF, followed by
addition of 1.5 equivof DIEA. The mixture was shaken
mechanically for 2 h at room temperature. The resin
was filtered and washed thoroughly with DMF (3Â50
mL), MeOH (3Â50 mL), DMF (3Â50 mL), MeOH
(3ÂmL) and then dried in vacuo. The N-BOC group
was deprotected by a 30 min treatment with a 50%
CF3COOH solution in DCM. The completion of cou-
pling reactions was monitored by the ninhydrin test.
When necessary, a second coupling was performed
under the same conditions.
HOBt during 45 min at 0 ꢀC in CH2Cl2/DMF (1:1), and
then added with 1.5 equivof DIEA to the deprotected
heptapeptide on resin 28 (0.8 g, 0.32 mmol) swollen with
DMF. The mixture was shaken mechanically for 4 h at
room temperature. The resin was drained and washed as
described above.
4.3.7. N-t-BOC-(L-27-L-L-L-27-L)2-OMe 31. Cleavage of
the 14mer peptide from 300 mg of resin was done in
MeOH using 2 equivof DBU. After recoevring the
solution and evaporation, purification was effected by
reversed-phase HPLC (0–100% B in 45 min). Solvents
were removed in vacuo to give a colorless oil that was
dissolved in glacial acetic acid, and lyophilized to yield
4.3.2. Crown peptide nanostructures. All hexacrown
peptides have been synthesized, purified, and character-
ized in the same manner using appropriate crown ether
amino acids. The following synthesis of nanostructure 1
is illustrative of all synthesis.
1
52 mg of a fluffy white solid. H NMR (DMSO-d6), d
8.20–7.30 (m, 13H, NH), 6.85–6.50 (m, 12H, H Ar), 4.50
(m, 4H, aCH 21-C-7), 4.25–3.95 (m, 26H, 10 Leu
aCH+8 CH2–OAr), 3.95–3.50 (m, 83H, 40 CH2–
O+CH3–O), 3.35–3.00 (m, 8H, bCH2 21-C-7), 2.05–1.35
(m, 39H, 10 Leu bCH2+10 Leu gCH+9H t-Bu), 1.05–
0.70 (m, 60H, Leu CH3). ESMS m/z (M+Na)+ 2988.
4.3.3. N-t-BOC-L-27-L-L-L-27-L-Resin 28. The key hep-
tapeptide was prepared using 5 g of oxime resin. Cou-
pling reactions were performed as described above with
N-t-BOC-Leu-OH and N-t-BOC-6-OH.20
4.3.8. N-t-BOC-(L-27-L-L-L-27-L)3-OMe 1. The key
heptapeptide 29 (0.8 g, 0.5 mmol) was activated and
added with 1.5 equivof DIEA to the deprotected 14mer
on resin (0.5 g, 0.2 mmol) swollen in DMF. The mixture
was shaken mechanically for 4 h at room temperature.
After washing and drying, cleavage of the 21mer peptide
from 500 mg of resin was performed in MeOH using 2
equiv of DBU. The cleavage solution was recovered and
evaporated to give crude 1. Purification was done by
reversed-phase HPLC (0–100% B in 45 min). Solvents
were removed in vacuo to give a colorless oil that was
dissolved in acetic acid, and lyophilized to yield 31 mg
of 1 as a fluffy white solid. 1H RMN (300 MHz, CDCl3),
d 8.30–7.40 (m, 21H, NH), 6.85–6.60 (m, 18H, H Ar),
4.35–4.25 (m, 6H, aCH 21-C-7), 4.10–3.85 (m, 24H, 8
CH2–OAr), 3.75–3.60 (m, 24H, 40 CH2–CH2–OAr),
3.60–3.40 (m, 114H, 48 CH2–O+15 Leu aCH+CH3–
O), 3.10–2.90 (m, 12H, bCH2 21-C-7), 1.65–1.35 (m,
54H, 15 Leu bCH2+15 Leu gCH+9H t-Bu), 0.95–0.65
(m, 90H, Leu CH3). ESMS m/z (M+Na)+ 4406,
(M+Na)+ 4406.
4.3.4. N-t-BOC-L-27-L-L-L-27-L-OH 29. Cleavage of the
key heptapeptide from 4 g of resin was realized using a
10% H2O/THF solution with 2 equivof DBU at room
temperature for 2 h. The solution was recovered and
after evaporation of solvent in vacuo, the crude peptide
was dissolved in DCM and washed twice with 0.5 N
HCl and water. After drying with MgSO4, solvent was
evaporated and residue triturated with anhydrous ether
and petroleum ether to give 2.1 g of a beige powder pure
1
enough by HPLC to be used as is in the next step. H
NMR (DMSO-d6), d 8.15 (d, J=8.2 Hz, 1H, NH), 8.05
(d, J=8.3 Hz, 1H, NH), 7.98 (d, J=7.8 Hz, 1H, NH),
7.86 (d, J=8.0 Hz, 1H, NH), 7.72 (d, J=7.9 Hz, 1H,
NH), 7.65 (d, J=7.8 Hz, 1H, NH), 6.97 (d, J=8.1 Hz,
1H, NH-BOC), 6.77–6.63 (m, 6H, H Ar), 4.50–4.48
(m, 2H, aCH 21-C-7), 4.33–4.20 (m, 4H, Leu aCH),
4.10–3.90 (m, 8H, CH2–OAr), 3.84 (m, 1H, Leu-BOC
aCH), 3.79-3.65 (m, 8H, CH2–CH2–OAr), 3.64–3.44 (m,
32H, CH2–O), 3.00–2.59 (m, 4H, bCH2 21-C-7), 1.63–
1.22 (m, 24H, 5 Leu bCH2+5 Leu gCH+9H t-Bu),
0.88–0.75 (m, 30H, Leu CH3). ESMS m/z (M+H+)
1535.
4.4. Ion transport assays
4.4.1. pH stat experiments. The pH stat method and the
preparation of vesicles have been described in details by
Fyles et al.25,30 Briefly, vesicles were prepared by the
sonication of an ethereal solution of an 8:1:1 mixture of
egg phosphatidyl choline/egg phosphatidic acid/choles-
terol in a pH 6.6 buffer solution (0.2 M choline sulphate
and 0.054 M d-mannitol in water). After evaporation of
ether, vesicles were suspended in an unbuffered solution
and purified by filtration through a 0.45-mm syringe fil-
ter to remove large aggregates and by gel filtration on
Sephadex G-25. Vesicles are predominantly unilamellar
with a diameter of around 150 nm and contain a small
portion of smaller (50 nm) unilamellar vesicles and
multilamellar vesicles. In a typical experiment, 0.2 mL
of a fresh vesicle dispersion was added to 3.8 mL of the
external solution (0.11 M choline sulphate and 0.093 M
d-mannitol in water) and the pH was raised to 7.6 by
4.3.5. N-t-BOC-L-27-L-L-L-27-L-OMe 30. Cleavage
from 0.2 g of resin 28 was performed using the same
protocol as above but using MeOH as solvent. Cleavage
yielded 55 mg of 30 as a beige powder. 1H NMR
(DMSO-d6), d 8.31–7.62 (m, 6H, NH), 6.97 (m, 1H,
NH-BOC), 6.80–6.63 (m, 6H, H Ar), 4.55–4.47 (m, 2H,
aCH 21-C-7), 4.33–4.20 (m, 4H, Leu aCH), 4.10–3.90
(m, 8H, CH2–OAr), 3.84 (m, 1H, Leu-BOC aCH), 3.79–
3.65 (m, 8H, CH2–CH2–OAr), 3.64–3.44 (m, 35H, 16
CH2-O+CH3–O), 3.00–2.59 (m, 4H, bCH2 21-C-7),
1.63–1.22 (m, 24H, 5 Leu bCH2+5 Leu gCH+9H t-
Bu), 0.88–0.75 (m, 30H, Leu CH3). ESMS m/z
(M+H)+ 1549.
4.3.6. N-t-BOC-(L-27-L-L-L-27-L)2-resin. The key hepta-
peptide 29 (1.1 g, 0.72 mmol) was activated with DIC/

E. Biron et al. / Bioorg. Med. Chem. 12 (2004) 1279–1290
1289
the addition of a choline hydroxide solution. At 5-min
intervals, the following solutions were added: 10 mL of a
1-mM MeOH solution of FCCP,31 0.5 mL of Cs2SO4
(0.5 M), and 10 mL of a 0.275 mM MeOH solution (2.75
mmol) of a transporter. Release of protons starts at this
point and pH is maintained at 7.5 by addition of 4.75
mM choline hydroxide solution using an automated
titrator. After 1 h, the remaining entrapped protons are
released by lysis of the vesicles by addition of Triton X-
100. The experiments were performed under a stream of
N2.
(Poretics, 0.4 mm). Calcein-containing vesicles were
separated from the free calcein by exclusion chromato-
graphy using a column filled with Sephadex G-50 super-
fine gel swollen in the external buffer (100 mM HEPES,
170 mM NaCl and 5 mM EDTA at pH 7.4). Fluores-
cence measurements were performed on a Varian Cary
Eclipse spectrometer. Fluorescence intensity of calcein
was monitored at emission wavelength of 513 nm
(response time of 0.5 s) after excitation at 490 nm. The
temperature was fixed to 25 ꢀC. In a quartz cuve of 1
cm, 4 mL of external buffer solution and 50 mL of vesi-
cle solution were introduced. Stirring and acquisition
were started. After 50 s, an appropriate amount (1–10
mL) of a TFE solution of transporter (5–40 mM) was
added. Then, after 400 s, a 10% Triton X-100 solution
was added and experiment was terminated after 500 s.
4.4.2. Na+ transport experiments using 23Na NMR. To
prepare vesicles, the procedure described by Gokel38
was followed with some modifications. At room tem-
perature, 13 mg of phosphatidylglycerol and 5.2 mg of
phosphatidylcholine were diluted in 25 mL of CHCl3.
The solvent was evaporated in vacuo and lipids were
dried for 12 h. Under nitrogen, the lipid film was dis-
solved in 25 mL of diethyl ether. 10 mL of buffer solu-
tion was added. The buffer solution has a total Na+
concentration of 100 mM and is composed of NaCl
(98.8 mM), KCl (0.26 mM), Na2HPO4 (0.64 mM) and
KH2PO4 (0.14 mM) in water at pH 7.3. The two-phase
suspension was placed in water bath sonicator at 2–5 ꢀC
and sonicated during 40 min. The organic phase was
evaporated under reduced pressure. The resulting vesi-
cle solution was filtrated through a polycarbonate
membrane (Poretics, 0.4 mm). In a Eppendorf tube, 0.8
mL of vesicle solution was mixed with an appropriate
amount (1–10 mL) of a TFE solution of transporter (5–
40 mM). Sample was warmed at 60 ꢀC during 60 min.
Then, 0.1 mL of D2O and 0.1 mL of the shift reagent
solution (sodium tripolyphosphate (0.133 M) and DyCl3
(0.033 M) in water) were added. After stirring, the sam-
ple was transferred in a NMR tube for study.
4.5. Biological activity assays
4.5.1. Antimicrobial assays. Bacterial strains used were
Escherichia coli ATCC 25922, Staphylococcus aureus
ATCC 25923, Pseudomonas aeruginosa ATCC 27855
and Streptococcus pneumoniae ATCC 6303. After
determination of cell tolerance to DMSO or MeOH,
solutions of crown peptides were prepared in these sol-
vents. Water was added to obtain an amount of 2% of
DMSO, or 4% of MeOH. Compounds were added to
Petri dish infected with bacterial strains and incubated
for 24 h at 37 ꢀC. Visual inspection of dish showed no
significant antimicrobial activity.
4.5.2. Cytotoxicity experiments. Using the same com-
pounds solutions, serial dilutions of stock solutions
were realized. Cancer cells used were MDA MB1231
(breast cancer) and P388 D1 (mouse leukemia). Cells
were disposed into wells of microtiter plates that con-
tained RPMI 1640 culture medium. Then, the different
compound solutions were added. After incubation dur-
ing 72 h at 37 ꢀC, compound efficiency was determined
by treatment with 2-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide (MTT) and absorbance
measurement at 570 nm.
4.4.3. NMR measurements and rate constant. 23Na
NMR spectra were recorded on a Varian Xl-200 NMR
spectrometer operating at 52.9 MHz. Typically, 1800
FID transients were accumulated per data set at a probe
temperature of 25 ꢀC. For each potential transporter
incorporated into phospholipid membranes, any line
broadening was determined. The rate constant k=1/t, is
directly proportional to the observed line broad-
ening,=p(ꢀnÀꢀn0), and is determined from the slope
of a plot of 1/t versus transporter concentration.
Acknowledgements
This work was supported by the NSERC of Canada, the
Fonds FCAR of Quebec, and the Universite Laval. E.B.
and F.O. thanks the NSERC of Canada and the Fonds
FCAR of Quebec for Postgraduate Scholarships. M.-
E.C. is grateful to the NSERC of Canada for a Summer
Research Fellowship. We thank R. C. Levesque and F.
Sanschagrin for their help in performing the anti-
microbial assays, and R. C. Gaudreault for help in the
cytotoxicity assays.
4.4.4. Fluorescence vesicle lysis experiments. For fluor-
escence measurements used to study vesicle lysis, the
procedure described by Lafleur et al.46 was followed
with some modifications. A calcein solution was pre-
pared by dilution calcein in internal buffer (100 mM
HEPES, 30 mM NaCl and 5 mM EDTA) to obtain a
calcein concentration of 80 mM. The pH of the solution
was adjusted to 7.4 with a 1 N NaOH solution. At room
temperature, 13 mg of phosphatidylglycerol and 5.2 mg
of phosphatidylcholine were diluted in 25 mL of CHCl3.
Solvent was evaporated in vacuo and lipids were dried
for 12 h. Calcein solution (5 mL) was added to the lipid
film. The resulting mixture was placed in water bath
sonicator at 2–5 ꢀC and sonicated during 40 min. Vesi-
cles were filtrated through a polycarbonate membrane
References and notes
1. (a) Cheley, S.; Gu, L. Q.; Bayley, H. Chem. Biol. 2002, 9,
829. (b) Howorka, S.; Movileanu, L.; Braha, O.; Bayley,
H. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 12996. (c)
Bayley, H.; Cremer, P. S. Nature 2001, 413, 226. (d)

1290
E. Biron et al. / Bioorg. Med. Chem. 12 (2004) 1279–1290
Howorka, S.; Cheley, S.; Bayley, H. Nat. Biotechnol.
2001, 19, 636.
19. Spencer, D. M.; Wandless, T. J.; Schreiber, S. L.; Crab-
tree, G. R. Science 1993, 262, 1019.
2. Hellinga, H. W.; Marvin, J. S. TIBTECH 1998, 16, 183.
3. Weiss-Wichert, C; Smetazko, M.; Valina-Saba, M.;
Schalkhammer, T. J. Biomol. Screen. 1997, 2, 11.
4. (a) Cornell, B. A.; Krishna, G; Osman, P. D.; Pace, R. D.;
Wieczorek, L. Biochem. Soc. Trans. 2001, 29, 613. (b)
Cornell, B. A.; Braach-Maksvytis, V. L. B.; King, L. G.;
Osman, P. D. J.; Raguse, B.; Wieczorek, L.; Pace, R. J.
Nature 1997, 387, 580.
5. Sugawara, M.; Hirano, A.; Buhlmann, P.; Umezawa, Y.
Bull. Chem. Soc. Jpn. 2002, 75, 187.
6. Borizenko, V.; Zhang, Z. H.; Woolley, G. A. Biochim.
Biophys. Acta 2002, 26, 1558.
7. (a) Arndt, H. D.; Bockelmann; Knoll, A.; Lamberth, S.;
Griesinger, C.; Koert, U. Angew. Chem., Int. Ed. 2002, 41,
21. (b) Arndt, H. D.; Knoll, A.; Koert, U. Angew. Chem.,
Int. Ed. 2001, 40, 2076. (c) Schrey, A.; Vescovi, A.; Knoll,
A.; Rickert, C.; Koert, U. Angew. Chem., Int. Ed. 2000,
39, 5.
8. Julien, L.; Lazrak, T.; Canceill, J.; Lacombe, L.; Lehn, J.-
M. J. Chem. Soc., Perkin Trans. 2 1993, 1011.
9. (a) Waser, P.; Rueping, M.; Seebach, D.; Duchardt, E.;
Schwalbe, H. Helv. Chim. Acta 2001, 84, 1821. (b) See-
bach, D.; Fritz, M. G. Int. J. Biol. Macromol. 1999, 25,
217.
20. Voyer, N. J. Am. Chem. Soc. 1991, 113, 1818.
21. Engelman, D. M.; Steitz, T. A. Cell 1981, 23, 411.
22. Schultz, G. E.; Schrimer, R. H. Principles of Proteins
Structure; Springer: New York, 1996; Chapter 6.
23. Yamamoto, H.; Hayakama, T. Polymer 1979, 18, 979.
24. Benz, H. Synthesis 1994, 4, 337.
25. (a) Osapray, G.; Taylor, J. W. J. Am. Chem. Soc. 1990,
112, 6046. (b) Mihara, H.; Chmielewski, J. A.; Kaiser,
E. T. J. Org. Chem. 1993, 58, 2209.
26. Pichette, A.; Voyer, N.; Larouche, R.; Meillon, J.-C.
Tetrahedron Lett. 1997, 38, 1279.
27. Meillon, J.-C.; Voyer, N. Angew. Chem., Int. Ed. Engl.
1997, 36, 967.
28. Biron, E.; Voyer, N.; Meillon, J.-C.; Cormier, M.-E.;
Auger, M. Biopolymers 2000, 55, 364.
29. Voyer, N.; Robitaille, M. J. Am. Chem. Soc. 1995, 117,
6599.
30. Fyles, T. M.; James, T. D.; Kaye, K. C. J. Am. Chem.
Soc. 1993, 115, 12315.
31. Herve, M.; Cybulska, B.; Gary-Bobo, C. M. Eur. J. Bio-
phys. 1985, 12, 121.
¨
32. Fyles, T. M.; James, T. D.; Kaye, K. C. Can. J. Chem.
1990, 68, 976.
33. (a) Hille, B. Ionic Channels of Excitable Membranes, 2nd
10. (a) Leevy, W. M.; Donato, G. M.; Ferdani, R.; Goldman,
W. E.; Schlesinger, P. E.; Gokel, G. W. J. Am. Chem. Soc.
2002, 124, 9022. (b) Gokel, G. W. Cell. Biochem. Biophys.
2001, 35, 211. (c) Gokel, G. W. Chem. Commun. 2000, 1.
(d) Gokel, G. W.; Murillo, O. Acc. Chem. Res. 1996, 29,
425.
ed.; Sinauer: Sunderland, MA, USA, 1992. (b) Lauger, P.
¨
Angew. Chem., Int. Ed. Engl. 1985, 24, 905. (c) Suarez-
Isla, B. A.; Wan, K.; Lindstrom, J.; Montal, M. Bio-
chemistry 1983, 22, 2319.
34. Wooley, G. A.; Jaikaran, A. S. I.; Zhang, Z.; Peng, S. J.
Am. Chem. Soc. 1995, 117, 4448.
11. (a) Bandyopadhyay, P.; Janout, V.; Zhang, L.-H.; Regen,
S. L. J. Am. Chem. Soc. 2001, 123, 7691. (b) Otto, S.;
Janout, V.; DiGiorgio, A. F.; Young, M. C.; Regen, S. L.
J. Am. Chem. Soc. 2000, 122, 1200.
35. Voyer, N.; Potvin, L.; Rousseau, E. J. Chem. Soc., Perkin
Trans. 2 1997, 1469.
36. Riddell, F. G.; Hayer, M. K. Biochim. Biophys. Acta 1985,
817, 313.
12. Dieckmann, G. R.; Lear, J. D.; Zhong, Q.; Klein, M. L.;
Degrado, W. F.; Sharp, K. A. Biophys. J. 1999, 76, 618.
13. (a) Frenandez-Lopez, S.; Kim, H. S.; Choi, E. C.; Del-
gado, M.; Granja, J. R.; Khasanov, A.; Kraehenbuehl,
K.; Long, G.; Weinberger, D. A.; Wilcoxen, K. M.; Gha-
diri, M. R. Nature 2001, 412, 452. (b) Bong, D. T.; Clark,
T. D.; Granja, J. R.; Ghadiri, M. R. Angew. Chem., Int.
Ed. 2001, 40, 988.
14. (a) Baumeister, B.; Sakai, N.; Matile, S. Org. Lett. 2001,
3, 4229. (b) Sakai, N.; Gerard, D.; Matile, S. J. Am.
Chem. Soc. 2001, 123, 2517. (c) Sakai, N.; Matile, S.
Chem. Eur. J. 2000, 6, 1731.
15. (a) Cameron, L. M.; Fyles, T. M.; Hu, C. W. J. Org.
Chem. 2002, 67, 1548. (b) Fyles, T. M.; Hu, C. W.; Knoy,
R. Org. Lett. 2001, 3, 1335. (c) Fyles, T. M.; Loock, D.;
Zhou, X. J. Am. Chem. Soc. 1998, 120, 2997.
16. (a) Goto, C.; Yanamura, M.; Stake, A.; Kobuke, Y. J. Am.
Chem. Soc. 2001, 123, 12152. (b) Toshino, N.; Satake, A.;
Kobuke, Y. Angew. Chem., Int. Ed. 2001, 40, 457.
37. Buster, D.; Hinton, J.; Millett, F.; Shungu, D. Biophys. J.
1988, 53, 145.
38. Abel, E.; Maguire, G. E. M.; Murillo, O.; Suzuki, I.; De
Wall, S. L.; Gokel, G. W. J. Am. Chem. Soc. 1999, 121,
9043.
39. Vandenburg, Y. R.; Smith, B. D.; Biron, E.; Voyer, N.
Chem. Commun. 2002, 17, 1694.
40. DeGrado, W. F.; Lear, J. D. Biopolymers 1990, 29, 205.
41. Akerfeldt, K. S.; Kienker, P. K.; Lear, J. D.; DeGrado,
W. F. In Comprehensive Supramolecular Chemistry;
Atwood, J. L., Davies, J. E. D., MacNicol, J. D., Vogtle,
¨
F., Reinhoudt, D. N., Eds.; Pergamon: New York, 1996;
Vol. 10, p 659–686.
42. Oren, Z.; Shai, Y. Biopolymers 1998, 47, 451.
43. (a) Voyer, N.; Lamothe, J. Tetrahedron 1995, 31, 9241. (b)
Voyer, N. Top. Curr. Chem. 1997, 184, 1. (c) Voyer, N. In
Method of Organic Chemistry Houben-Weyl; Goodman,
M.; Felix, A.; Moroder, L.; Toniolo, C., Eds.; Thieme,
2002; Vol. E22c, p 153.
17. Perez, C.; Espınola, C. G.; Foces-Foces, C.; Nun
˜
ez-
44. DeGrado, W. F.; Kaiser, E. T. J. Org. Chem. 1980, 45,
1295.
45. Banerjee, S.; Ressler, C. J. Org. Chem. 1976, 41, 3056.
46. Benachir, T.; Lafleur, M. Biophys. J. 1996, 70, 831.
Coella, P.; Carrasco, H.; Martin, J. D. Org. Lett. 2000, 2,
1185.
18. Nolte, R. J. M. Chem. Soc. Rev. 1994, 11.