Zinc(II) as an allosteric regulator of liposomal membrane permeability induced by synthetic template-assembled tripodal polypeptides.

Article abstract of DOI:10.1002/1521-3765(20020617)8:12<2753::AID-CHEM2753>3.0.CO;2-4

Three copies of peptide sequences from the peptaibol family, known to affect the permeability of the lipid bilayer of membranes, were connected to tris(2-aminoethyl)amine (TREN), a tripodal metal ion ligand, to prepare functional peptides capable of modifying the permeability of liposomal membranes. Some of the resulting tripodal polypeptide derivatives are very effective in promoting carboxyfluorescein (CF) leakage from CF-loaded unilamellar vesicles composed of a 70:30 phosphatidylcholine/cholesterol blend. The activity of these novel compounds was shown to be tunable upon metal ion coordination of the TREN subunit; the tripodal apopeptide was far more effective than its ZnII complex. Leakage experiments showed that a minimum number of five amino acids per peptide chain is required to form active systems. A mechanism is proposed in which the ZnII ion changes the conformation of the template from extended to globular and thus acts as an allosteric regulator of the activity of the systems. Molecular modeling studies indicate that when the three peptide chains are connected to the template in the extended conformation, the resulting tripodal polypeptide is able to span across the membrane, thus allowing the formation of permeable channels made of a cluster of molecules. The same change of conformation induces, to some extent, fusion of the membranes of different liposomes.

Full text of DOI:10.1002/1521-3765(20020617)8:12<2753::AID-CHEM2753>3.0.CO;2-4

FULL PAPER
Zinc(ii) as an Allosteric Regulator of Liposomal Membrane Permeability
Induced by Synthetic Template-Assembled Tripodal Polypeptides
Paolo Scrimin,*[a] Paolo Tecilla,[c] Umberto Tonellato,*[a] Andrea Veronese,[a]
Marco Crisma,[b] Fernando Formaggio,[b] and Claudio Toniolo[b]
Abstract: Three copies of peptide se-
quences from the peptaibol family,
known to affect the permeability of the
lipid bilayer of membranes, were con-
The activity of these novel compounds
was shown to be tunable upon metal ion
coordination of the TREN subunit; the
tripodal apopeptide was far more effec-
tive than its ZnII complex.Leakage
experiments showed that a minimum
number of five amino acids per peptide
chain is required to form active systems.
A mechanism is proposed in which the
ZnII ion changes the conformation of the
template from extended to globular and
thus acts as an allosteric regulator of the
activity of the systems.Molecular mod-
eling studies indicate that when the
three peptide chains are connected to
the template in the extended conforma-
tion, the resulting tripodal polypeptide is
able to span across the membrane, thus
allowing the formation of permeable
channels made of a cluster of molecules.
The same change of conformation in-
duces, to some extent, fusion of the
membranes of different liposomes.
nected
to
tris(2-aminoethyl)amine
(TREN), a tripodal metal ion ligand, to
prepare functional peptides capable of
modifying the permeability of liposomal
membranes.Some of the resulting tri-
podal polypeptide derivatives are very
effective in promoting carboxyfluores-
cein (CF) leakage from CF-loaded uni-
lamellar vesicles composed of a 70:30
phosphatidylcholine/cholesterol blend.
Keywords: allosterism ¥ liposomes
¥ membrane permeability ¥ peptides
¥ zinc
able structural features and properties have been designed
and synthesized.[5] The so-called ™bouchet approach∫ has
been exploited to prepare ion channels based on membrane
spanning monomers attached to a pre-organized macrocyclic
central core.[6] Very efficient transmembrane channels have
recently been prepared by designing macrocyclic peptides
that can adopt flat-ring conformations and stack one upon the
other to form self-assembling nanotubes.[7]
Introduction
Biological membranes are organized assemblies of lipids and
proteins whose primary functions are containing, compart-
mentalizing, and regulating the transfer of metabolites and
macromolecules in a living organism.[1] Their permeability is
highly selective because of the occurrence of specific carriers
or channel-formers for ions or organic molecules.[2] Several
physiologically important channel proteins have been isolated
and sequenced,[3] and the preparation of artificial ion channels
is an ongoing ambitious goal attracting considerable efforts.[4]
To this aim, different strategies have been used.Artificial
proteins, based on specifically tailored peptides with predict-
A common feature of many biological systems is the presence
of regulatory elements that allow modulation of the perme-
ation processes across the membranes.[1] A reversible modu-
lation of any given physicochemical property is the basic
feature of any molecular or supramolecular device.[8] Artificial
ionophores which can reversibly modulate membrane perme-
ability in response to a chemical or electrochemical input are
examples of such artificial devices.[9] Such a regulation may be
achieved by taking advantage of allosteric effects, such as those
resulting from a change of conformation owing to interaction
with an effector in a remote subsite of the molecule which is
not directly involved in the process.Examples of systems
whose functions have been regulated by means of allosteric
interactions have been reported.[10] They include hydrolyti-
cally active catalysts,[11] and enzyme inhibitors.[12]
[a] Prof.Dr.P.Scrimin, Prof.Dr.U.Tonellato, Dr.A.Veronese
Dipartimento di Chimica Organica and CNR-CMRO
¡
Universita di Padova, Via Marzolo, 1 35131 Padova (Italy)
Fax : (‡39)049-827-5239
E-mail: paolo.scrimin@unipd.it, umberto.tonellato@unipd.it
[b] Dr.M.Crisma, Prof.Dr.F.Formaggio, Prof.Dr.C.Toniolo
Dipartimento di Chimica Organica and CNR-CSB
¡
Universita di Padova, Via Marzolo, 1 35131 Padova (Italy)
[c] Prof.Dr.P.Tecilla
Dipartimento di Scienze Chimiche
Universita di Trieste, Via Giorgieri, 1 34126 Trieste (Italy)
¡
Peptaibols[13] constitute an interesting class of natural
antibiotics produced by soil fungi.It is widely accepted that
Supporting information for this article is available on the WWW under
http://www.chemeurj.org or from the author.
Chem. Eur. J. 2002, 8, No. 12
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0812-2753 $ 17.50+.50/0
2753

P.Scrimin, U.Tonellato et al.
FULL PAPER
these linear peptides act on cell membranes through channel
formation involving leakage of the cytoplasmic material and,
eventually, lead to cell death.They are characterized by
closely related sequences of linear peptides, ranging from
19 amino acids (alamethicin)[14] to 11 amino acids (tricho-
gin)[15], and even to as low as six amino acids (trichodece-
nins).[16] Peptaibols are known to contain a large portion of a-
aminoisobutyric acid (Aib), which strongly promotes helix
formation.[17] While the helix formed by the 19-amino acid
sequence is long enough to span across the double layer of
membranes, shorter peptides such as trichogin require the
presence of a fatty acid chain at the N-terminus to elicit
membrane activity (hence the term lipopeptides).[18] A mixed
a/310 helix was proposed for natural trichogin GA IV from
extensive solution[15a,c] and crystal-state studies. [15b]
In 1996, we reported[19] the synthesis and the metal ion
modulation of membrane permeability properties induced by
a polypeptide template.[20] This was designed by connecting
together three copies of a decapeptide analogue of trichogin
GA IV through an appropriate spacer to the known metal ion
ligand tris(2-aminoethyl)amine (TREN).[21] Membrane activ-
ity was tested by monitoring the leakage of trapped carboxy-
fluorescein (CF) from small unilamellar vesicles.[15a,c, 18, 22]
Here, we report a comparative study of a wide class of
compounds obtained by functionalizing the same ligand with
peptides of different main chain length aimed at defining the
minimal requirements for affecting membrane permeability
as well as identifying the source of the activity observed.
bled T(P4)3, T(P6)3 and T(P10)3 reveals the following
features: 1) the presence of free (stretching band at nÅ >
3400 cmÀ1) and hydrogen-bonded (stretching band in the
3356 3319 cmÀ1 range) NH groups, 2) an increase in the
intensity of the hydrogen-bonded NHs as the number of
amino acids in the peptide chain increases.This behavior
strongly suggests formation of a folded structure even with the
short, four-residue peptides.Upon increasing the length of the
peptide, the content of the folded conformation tends to
increase.The crystal structures of the Z-protected tetrapep-
tide P4 and pentapeptide P5 confirm the spectroscopic
evidence, although they show that in the crystals, the
molecules assume conformations not amenable to a regular
310 or a helix.[24] For the purpose of our investigation, which
was carried out in a lipid bilayer, it is conceivable that the
environment experienced by the peptide in the membrane
would be similar to that in the chloroform solution used for
the spectroscopic investigations.Hence, the folded conforma-
tion of the peptides is probably maintained when bound to the
liposomes.This is also supported by our recent observation
that peptides rich in Ca-tetrasubstituted amino acids assume a
prevailing helical conformation even in aqueous solutions.[25]
Synthesis and conformation of the template-assembled pep-
tides: Peptides P3 P6 and P10 were used for the function-
alization of TREN.It is known that coordination of N-func-
tionalized TREN to ZnII involves, typically, formation of
pentacoordinate trigonal bipyramidal complexes with four
amino groups and an extra apical ligand.[21] A 4-carboxy-1-
methylbenzene spacer was introduced between the TREN
molecule and the peptide sequences to avoid undesired
interactions between the vacant coordination site of the metal
and the peptide chains, which, inter alia, could result in the
modification of their secondary structures.
Results and Discussion
Peptide design and conformation: The peptides used for the
construction of our systems are made of sequences ranging
from three to ten amino acids; all of them are portions of the
naturally occurring antibiotic trichogin.The constituent
amino acids and the acronyms used for their identification
are given below.Peptides P4, P5, P6, and P10 correspond to
the C-terminal sequence of trichogin, except for the C-termi-
The series of five related compounds was prepared as
indicated in Scheme 2, while the templating platform was
prepared according to Scheme 1.Thus, using SnCl and HCl,
2
À
O
nal Lol which was replaced by Leu OMe (the non-reduced
O
amino acid ester).According to previous studies, this mod-
ification has little influence on the activity of the antibio-
tic.[15c, 23] Furthermore, all sequences contain at least one
helicogenic Aib residue.The synthesis and characterization of
peptides P3 P6 and P10 have been reported elsewhere.[23]
i)
ii)
OCH3
OCH3
NC
1
O
O
O
OCH3
OCH3
iii)
iv)
H
Trichogin GA IV
n-C7H15CO-Aib-Gly-Leu-Aib-Gly-Gly-Leu-Aib-Gly-Ile-Lol
n-C7H15CO-Aib-Gly-Leu-Aib-Gly-Gly-Leu-Aib-Gly-Ile-LeuOMe
Gly-Leu-Aib-Gly-Gly-Leu-Aib-Gly-Ile-LeuOMe
Gly-Leu-Aib-Gly-Ile-LeuOMe
N
N
N
N
C8P11
P10
3
3
3
2
P6
P5
Leu-Aib-Gly-Ile-LeuOMe
O
P4
Aib-Gly-Ile-LeuOMe
O
P3
Gly-Leu-Aib-NHiPr
OH
Boc
N
OCH3
Boc
N
v)
Ac(P11)
Z(P10)
CH3CO-Aib-Gly-Leu-Aib-Gly-Gly-Leu-Aib-Gly-Ile-LeuOMe
PhCH2OCO-Gly-Leu-Aib-Gly-Gly-Leu-Aib-Gly-Ile-LeuOMe
N
N
3
3
5
4
A conformational investigation carried out by FT-IR
absorption spectroscopy in deuterochloroform on the Z-pro-
tected peptides used for the preparation of template-assem-
Scheme 1.Synthesis of the TREN-based template.i) SnCl 2, HCl, Et2O;
ii) TREN, CH3CN; iii) NaBH4, EtOH; iv) (Boc)2O; v) NaOH, EtOH, then
diluted HCl.
2754
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0812-2754 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 12

Template-Assembled Tripodal Polypeptides
2753 2763
4-cyanomethylbenzoate was converted into the corresponding
aldehyde, which was then treated with TREN, and the
resulting tris-imine was reduced with NaBH4.This TREN
derivative was subsequently protected with Boc and the
methyl ester was hydrolyzed to give the free carboxylic acid.
All peptide sequences were prepared in solution by stepwise
elongation of the peptide chain starting from the C-terminal
leucine methyl ester.[23] Na-Benzyloxycarbonyl amino acid
derivatives were incorporated in the peptide chain either by
the mixed anhydride or by the symmetrical anhydride
coupling method.Docking between the core unit and the
peptide sequences was performed by coupling the free
benzoyl groups of the ligand with the N-terminus of each
different peptide using 1-hydroxybenzotriazole (HOBt) and
N-ethyl-N'-(3-dimethylaminopropyl)carbodiimide (EDC) as
coupling reagents.Removal of the Boc protecting groups and
semi-preparative reversed-phase HPLC purification gave the
final products.
Figure 1.FT-IR absorption spectra of the Z-protected peptides (1), the
tripodal polypeptides (2), and their ZnII complexes (3) recorded in CDCl3
(10% CD3CN added for T(P10)3).a) P5 and T(P5)3; b) P6 and T(P6)3;
c) P10 and T(P10)3.
indicate an increase of the hydrogen-bonded versus free NH
groups when compared to those of the corresponding single,
Z-protected peptides.This behavior suggests that, within the
template-assembled tripodal peptides, the peptide chains tend
to interact reciprocally, and this interaction induces formation
of additional hydrogen bonds.Although it is not possible to
know from this behavior whether this extra hydrogen bonding
occurs within the same peptide chain or between different
chains, it clearly suggests some sort of interaction between the
peptides in the template leading to an overall increase of
hydrogen bonding.We note, however, that the main amide I
band occurs at 1662 cmÀ1, in full accord with a 310 helical
conformation.[27] Induced helicity by the clustering of several
peptide molecules is well documented in the literature.[20a, 28]
Upon formation of the ZnII complexes, the only change
observed is a shift (10 15 cmÀ1) of the absorption maximum
towards lower wavenumbers in accordance with a further
strengthening of the hydrogen bonds, probably owing to a
stronger interaction between the peptide chains.This could
imply a ZnII template conformation that can induce the
alignment of the three peptide chains.
Further support for a change of conformation in the
presence of ZnII is obtained by dynamic light scattering
experiments.Figure 2 shows the amount of scattered intensity
of aqueous solutions of T(P10)3 as a function of its concen-
tration as a free ligand and as a ZnII complex.Since the
intensity of scattered light depends not only on the concen-
tration but also on the size of the system, a steeper gradient in
Figure 2 indicates a larger object in solution.Hence, the data
suggest a more compact (probably globular, see above)
conformation for the ZnII complex.From the relatively high
intensity of the scattered light, it appears that both systems
form aggregates under the conditions employed.The size of
the aggregates does not change in the concentration interval
explored, as there is no change of gradient and both lines show
the same intercept that represents the background intensity in
the absence of added peptide.Hence, aggregates are already
present at the lowest concentration explored.As mentioned
above, the onset of aggregation or a change in the size of the
aggregates would result in a net change of gradient.However,
O
P3
P4
P5
P6
i)
OH
Boc
N
+
N
3
P10
5
T(P3)3
T(P4)3
T(P5)3
T(P6)3
T(P10)3
O
ii)
Peptide
Boc
N
N
3
6-10
Scheme 2.Synthesis of the tripodal polypeptide templatesi.) HOBt, EDC,
CH2Cl2; ii) TFA, CH2Cl2.
The purity and the identity of these compounds were
assessed by analytical HPLC, H NMR and mass spectrom-
1
etry (MALDI).Final compounds where isolated as trifluoro-
acetate salts; in this form they are all soluble in methanol,
insoluble in chloroform and dichloromethane.Template-
assembled derivatives T(P5)3, T(P6)3, and T(P10)3 are soluble
in water up to about 5 Â 10À5 m at pH 7.3, while T(P3)3 and
T(P4)3 exhibit a slightly higher solubility in this solvent.
The uncomplexed ligands have a flexible structure with no
geometrical constraint.Previous work conducted in our
laboratory[10d] has shown that on binding to ZnII ions, a
benzyl-functionalized TREN derivative forms a pentacoordi-
nate complex with the three phenyl groups pointing in the
same direction, defining a basket-like pseudocavity.Thus, in
principle, the tertiary structure of the template-assembled
polypeptides can be controlled by metal ion complexation.
The binding constant of the TREN platform for the 1:1
complex with ZnII ions is estimated from literature data to be
about 1010 mÀ1 at pH 7.[26] This means that a very strong
complex can be formed, and also that it is reasonable to
assume that a negligible amount of free metal ion is present
when a 1:1 solution of ligand and metal ion is prepared and
diluted to a concentration of 10À7 m.
The FT-IR absorption spectra of apopeptides T(P5)3,
T(P6)3, and T(P10)3 in deuterochloroform solution (Figure 1)
Chem. Eur. J. 2002, 8, No. 12
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0812-2755 $ 17.50+.50/0
2755

P.Scrimin, U.Tonellato et al.
FULL PAPER
were prepared by sonication, and unencapsulated carboxy-
fluorescein was separated by passing the resulting dispersion
through a Sephadex column.Light scattering experiments on
the vesicle solutions devoid of the trapped dye gave a
hydrodynamic diameter of 63 Æ 5 nm.This fact and the
method of preparation of the aggregates (sonication) are
consistent with the formation of unilamellar aggregates.The
time course of the fluorescence emission is very similar to that
reported for analogous experiments conducted on antibiotics
of the peptaibol family.[29] A typical profile is shown in
Figure 4 for T(P6)3 and T(P10)3.An initial burst of fluo-
rescence emission, the amount of which was found to be
Figure 2.Dependence of the scattering intensity of aqueous solutions of
II
*
*
T(P10)3 ( ) and its Zn complex ( ) as a function of the concentration of
tripodal peptide at 258C.
we cannot rule out that the difference in gradient is related, at
least in part, to a different number of monomers constituting
the two different aggregates.
These observations lend support to the hypothesis that
complexation with ZnII induces a change of conformation of
the polypeptide template with the formation of a three-helix
bundle, similar to that depicted in Figure 3b, while the
polypeptide template devoid of the metal ion assumes a more
open conformation like, for instance, that illustrated in
Figure 3a.
Figure 4.Time course of the release of trapped CF from CF-loaded
liposomes at 258C upon addition of apopeptides (filled symbols) and ZnII-
peptides (open symbols).Squares: T(P10)3 ; circles: T(P6)3.The arrow
indicates the time of addition of excess EDTA to a cuvette containing the
ZnII peptides while a second cuvette was left unchanged for reference.
Conditions: [tripodal polypeptide] ˆ [ZnII] ˆ 5  10À7 m, pH 7.4.
dependent on the amount of peptide added, is immediately
observed, followed by a slower increase until the final
fluorescence intensity is eventually reached.Because of the
kinetic complexity of the fluorescence versus time profiles we
found it convenient to evaluate the efficiency of the systems
by determining the percentage of fluorescence increase
observed after 20 min, a time that was chosen after a
preliminary screening of the different peptide templates.
The results obtained for the different systems and their ZnII
complexes are shown in Figure 5 as a function of their
concentration.Table 1 and Figure 6 report the percentage of
fluorescence increase after 20 min at a fixed concentration of
5 Â 10À7 m polypeptide.Table 1 also reports the amount of CF
leaked spontaneously by the vesicles, and the effect due to the
single peptide models Ac(P11) and Z(P10) at the same
concentration.
II
Figure 3.Suggested conformational change induced by the addition of Zn
ions to the tripodal polypeptides.
Trapped-dye release from liposomes: As stated in the
Introduction, our basic idea was to verify if the template-
assembled polypeptides would be able to alter the perme-
ability of a liposomal membrane, and if the complexation of a
ZnII ion could modulate this effect.Furthermore, we wanted
to know the influence of the length of each peptide chain on
the efficiency of the modification of membrane permeability.
The CF-leakage technique[15a,c, 18, 22] takes advantage of the
increase in fluorescence of the dye when released into the
surrounding medium from a condition of self-quenching due
to its high concentration when trapped inside the liposome.
Membrane-modifying properties of the new compounds were
thus examined by following their effect on CF leakage from
small unilamellar vesicles prepared from a 70:30 blend of egg
l-a-phosphatidylcholine and cholesterol loaded with a high
concentration (5 Â 10À2 m) of the fluorescent dye.Vesicles
Analysis of Figures 5, 6, and Table 1 allows one to make the
following observations: 1) Some of the systems synthesized
are very effective in promoting CF leakage from the vesicles.
In the best cases, the efficiency is higher than that of the
II
trichogin analogue C8(P11).2) The Zn complexes are con-
sistently much less effective than the apopeptides in promot-
2756
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0812-2756 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 12

Template-Assembled Tripodal Polypeptides
2753 2763
eventually vanishes completely with T(P3)3.5) The presence
of the template is critical for the obtainment of an active
system.Single peptide systems Ac(P11) and Z(P10), although
made of ten amino acids, are essentially inactive.6) The
TREN platform, which is devoid of peptide functionalization,
is also inactive.
For T(P6)3 and T(P10)3, the permeability can, at least in
part, be reversibly controlled (Figure 4).When we start the
experiment with the ZnII complexes, we observe a very slow
increase of fluorescence, in accord with the behavior de-
scribed above.However, the rate of CF release becomes much
faster when an excess of ethylendiaminetetraacetic acid
(EDTA) is added to the solution.EDTA (log Kb ˆ 16.7 for
ZnII)[30] removes the metal ion from the TREN platform,
thereby transforming the tripodal polypeptide into a more
efficient system for promoting CF release from vesicles.The
time course of an experiment which starts with the free ligand
is less amenable to unravel such a reversible control of
permeation, as the very fast initial CF release is followed by a
slower process (see above) which is difficult to differentiate
from that observed in the presence of the ZnII complex.
™Snapshots∫ taken after 20 min for different vesicular
solutions exposed to 2.5 Â 10À7 m T(P6)3 in the presence of
an increasing concentration of metal ion clearly indicate that
one equivalent of ZnII required to elicit the maximum
inhibition of the effect of the free tripodal polypeptide
(Figure 7).This finding rules out any perturbation of perme-
Figure 5.Percentage of CF released after 20 min at 25 8C and pH 7.4 from
CF-loaded liposomes upon addition of increasing amounts of the different
-
tripodal polypeptides and their ZnII complexes.
Figure 6.Percentage of CF released after 20 min at 25 8C from CF-loaded
liposomes as a function of the number of amino acids present in each
peptide chain.[Tripodal polypeptide] ˆ 5  10À7 m; filled symbols refer to
apopeptides while open symbols refer to the ZnII complexes.
Table 1.Percentage increase of fluorescence intensity at 520 nm after
20 min upon addition of 5 Â 10À7 m polypeptide (free or as a 1:1 ZnII
complex) or the same concentration of other additives to CF-loaded
liposomes at 258C.
Additive
F20 [%]
free
83
79
65
12
7
55
3
5
ZnII
18
18
15
8
T(P10)3
T(P6)3
T(P5)3
T(P4)3
T(P3)3
C8P11
Ac(P11)
Z(P10)
TOMe[a]
none
Figure 7.Percentage of CF released after 20 min at 25 8C and pH 7.4 from
CF-loaded liposomes as a function of the [ZnII]/[T(P6)3] ratio.[ T(P6)3] ˆ
2.5 Â 10À7 m.
5
ability associated with an interaction of ZnII with the lipid
rather than with the TREN template.Indeed, it has been
reported that divalent ions (including ZnII) decrease mem-
brane permeability,[31] although this effect shows up at
concentrations far higher than those employed in our experi-
ments.
3
3
[a] The methyl ester of the TREN platform devoid of peptide chains.
ing CF release from the vesicles.3) The six residue per chain
tripodal polypeptide (T(P6)3) is as effective as the longer
homologue T(P10)3.4) As the number of amino acids in each
peptide chain decreases, the efficiency decreases as well, and
Membrane fusion experiments: The kinetic experiments
illustrated in Figure 4 indicate that the process of dye release
is by no means unimodal.What was particularly intriguing was
the very fast increase of fluorescence at the very early stages
Chem. Eur. J. 2002, 8, No. 12
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0812-2757 $ 17.50+.50/0
2757

P.Scrimin, U.Tonellato et al.
FULL PAPER
of the experiments.It occurred to us that the amphiphilic
nature of the active templates could be responsible for a
fusion process between the membranes of different liposomes.
A fast fusion process could be associated with a rapid release
of part of the trapped dye.For this purpose, we have run lipid
mixing experiments using vesicle preparations containing 1%
of hexadecylphosphatidylethanolamine functionalized at the
amine of the head group with the 7-nitrobenzen-2-oxo-1,3-
diazol-4-yl group (PE-NBD) and lissamine-rhodamine (PE-
LRB), respectively.Excitation of PE-NBD (450 nm) leads to
an emission band at 530 nm, which is very close to the
absorption band of PE-LRB.When the two dyes are on the
membrane of the same liposome, their proximity leads, upon
excitation of PE-NBD to the emission of PE-LRB and a
concomitant decrease of the emission of PE-NBD because of
energy transfer between the two systems.[32]
Effects of the polypeptides on the liposomal membranes: The
experiments reported above point to at least two different
effects exerted by the polypeptide templates on the liposomal
membrane: 1) fusion, and 2) altered permeability.
The former is a fast process occurring immediately after the
addition of the templates (devoid of ZnII) to the liposomes.
During fusion some of the trapped dye is released and this
accounts for the instantaneous burst of fluorescence in
Figure 4.Interestingly enough, a similar kinetic behavior has
been observed with other antibiotics of the peptaibol family
and we speculate that in those cases also, induced fusion may
be the explanation.In the present case, it is conceivable that
several polypeptide templates in their extended conformation
(see Figure 3a) bind to two different liposomes, inserting one
peptide chain into the membrane of one liposome and the
other two into the membrane of the second liposome.This
should allow the interaction of the constituent lipids that
eventually leads to fusion.In the presence of Zn II, the system
assumes a conformation that does not allow this interaction
(see Figure 3b) and, consequently, fusion is negligible.
As for the alteration of permeability, this may be attained
by the following mechanisms: 1) formation of pores or
channels that span across the membrane,[4] 2) formation of
lipophilic complexes with the trapped dye (carrier mecha-
nism),[33] or, 3) formation of defects or ™soft∫ patches in the
bilayer with enhanced permeability.[34] This last mechanism
occurs more frequently when large amounts of additives
(typically amphiphilic molecules) are added to the vesicular
preparation.The very low concentration at which the present
tripodal polypeptides exert their effects (ꢀ1% with respect to
the lipid concentration) allows us to dismiss the possibility
that this is actually the mechanism occurring.The channel or
carrier mechanisms appear more likely.With the data
obtained from the present investigation we favor the afore-
mentioned mechanism 1) on the basis of the following
arguments:
1) A concentration lag is observed before the onset of activity
of the tripodal polypeptide.All profiles of Figure 5 are
sigmoidal as typically expected for a cooperative process.
In the present case, this finding could be related to the
obtainment of the minimum concentration required for
the formation of a channel by a cluster of tripodal
polypeptides within the vesicular membrane.In the case
of the carrier mechanism, a lower efficiency and a linear
dependence on tripodal peptide concentration would be
expected.[35]
2) It has been suggested[6a, 35] and experimentally verified[6a]
with naturally occurring ionophores operating with chan-
nel or carrier mechanism, that the carrier mechanism
becomes much less effective in membranes in their gel
state, that is, below their phase transition temperature (Tc).
For this purpose we have tested the tripodal polypeptide
T(P10)3 with CF-loaded liposomes made of diphosphati-
dylcholine (DPPC), a lipid characterized by a Tc of 418C
and, hence, in the gel state at 258C.By contrast, the
membrane of the liposomes of egg phosphatidylcholine
used in the leakage experiments described above are in
their fluid state at 258C (Tc in the range À15 to À78C).As
can be seen in Figure 9, the amount of released CF is
Accordingly, when two liposome preparations containing
PE-NBD and PE-LRB, respectively, are mixed in the
presence of a fusogenic agent, the mixing of their lipids leads
to the phenomenon described above.Figure 8 shows the effect
Figure 8.Fluorescence emission spectra ( lex 450 nm) of an equimolar
mixture of PE-NBD and PE-LRB containing vesicles (1 mol% fluorescent
probe/total lipid; [total lipid] ˆ 1.14  10À4 m) and increasing concentration
of T(P10)3.The different curves are relative to the following percentage of
[T(P10)3] respect to the total lipid concentration ˆ 0, 0.5, 0.9, 1.4, 1.8, 2.3,
and 2.7%.
on the fluorescent emission of a solution of two liposome
preparations, each containing one of the two fluorescent
lipids, upon addition of increasing amounts of T(P10)3.The
fluorescent emission of PE-NBD decreases, while that of PE-
LRB increases; this behavior suggests that the template
induces considerable fusion between the two liposome
preparations.The phenomenon is instantaneous and occurs
immediately after the addition of the peptide template with
no further time-dependent change of fluorescence.When the
same experiment is carried out with T(P10)3-ZnII the effect is
almost negligible (data not shown), in line with the lack of the
burst of fluorescence in the kinetic experiments in the
presence of the complex (see Figure 4).We suggest that the
fast fusion process induced by T(P10)3 is also responsible for
the burst in the release of trapped CF in the experiments
described in the previous paragraph.
2758
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0812-2758 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 12

Template-Assembled Tripodal Polypeptides
2753 2763
extended conformation.[36] When a critical concentration is
reached within a liposome, a cluster of tripodal polypeptides is
formed.They may interact with another liposome to give
membrane fusion or, by inserting more deeply into the
membrane, they get aligned and oriented in such a way as to
expose the hydrophobic surface of each helix to the hydro-
carbon chains of the surrounding lipids, while the more polar
surface defines a channel suitable for the permeation of polar
species.As the number of amino acids in each peptide
decreases below six, the length of the tripodal polypeptide is
too short to span across the membrane and, furthermore, its
helical conformation becomes less stable.Consequently, the
activity decreases.Activity is also much lower if it binds to the
membrane in its globular conformation as in the ZnII complex.
In this case again, the system is too short to allow for the
fusion process or to ensure the formation of a channel
spanning the membrane, and the residual activity observed
could be due to the occasional alignment of two clusters
residing on different leaflets of the membrane.[37]
Figure 9.Percentage of CF released after 20 min at 25 8C from CF-loaded
liposomes (open circles, egg phosphatidylcholine, filled circles DPPC) upon
addition of increasing amounts of T(P10)3.
practically the same irrespective of the fluidity of the
membrane of the liposomes, lending strong support to the
channel mechanism.
3) Molecular models of the ZnII-ligated polypeptide template
(Figure 10) obtained by manually docking three copies of
the decapeptide (in the conformation obtained from the
crystallographic data of trichogin GA IV)[15b] to the TREN
platform[10d] and subsequently minimizing the energy of
the structure for the new bonds introduced, have an
estimated distance between the termini of the helices and
the CH2 groups of TREN close to the pivotal nitrogen of
approximately 18 ä, that is, half the length of the
phospholipid bilayer (ca.36 ä).However, in the absence
of the metal ion and, hence, in the extended conformation,
the polypeptide template is able to span the full length of
the bilayer.
Conclusion
By connecting three copies of a series of peptides, with amino
acid sequences representing portions of the peptaibol tricho-
gin GA IV, to a modified TREN platform, we have obtained
template-assembled polypeptides that are very effective in
promoting CF leakage from unilamellar vesicles.A minimum
number of five amino acids per peptide chain are required to
elicit activity in the polypeptide template.The rate of CF
release can be controlled by formation of a ZnII complex with
the TREN subunit in a reversible process.The apopeptides
are far more effective than the ZnII polypeptides in inducing
CF leakage.This behavior is probably associated with a
change of conformation of the polypeptide template from an
extended one, able to span across the vesicular membrane, to
a globular one, too short for this purpose.We note, finally, that
this conformational change may regulate the activity of a
II
system according to its functions.In the present case, Zn
binding leads to inhibition of activity.In a recent example
from our own laboratory based on the same templating
unit,[12] ZnII complexation leads, in contrast, to enhancement
of the efficiency of a HIV-1 protease inhibitor.In this latter
case, the conformational change induced by metal ion binding
ensures a tighter binding with the dimerization interface of
the protein.
Experimental Section
II
Proton NMR spectra were recorded at 200 MHz on a Bruker AC200F
spectrometer, at 250 MHz on a Bruker AC250F spectrometer, or at
400 MHz on a Bruker AM400 spectrometer.All chemical shifts are
reported in parts per million (ppm) as downfield shifts from Me4Si.Light
scattering experiments were performed with a Spectra Physics2016 argon
laser operating at 488 nm, interfaced with a Nicomp370 Model particle
sizing autocorrelator.The experiments were carried out at 25 8C on
previously centrifuged samples.Positive Maldi-MS measurements were
performed at the CNR Research Area, Camin, Padova.
Figure 10.Computer graphic picture of the Zn
polypeptide T(P10)3.
complex of tripodal
On the basis of the above considerations, we propose the
following mechanism to explain the alteration of liposomal
membrane permeability induced by the tripodal polypeptides.
The tripodal polypeptide binds to a liposome mostly in its
Chem. Eur. J. 2002, 8, No. 12
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0812-2759 $ 17.50+.50/0
2759

P.Scrimin, U.Tonellato et al.
FULL PAPER
HPLC purification: All peptide derivatives were purified by high-perform-
ance liquid chromatography.HPLC was performed using a Pharmacia
LKB-LCC2252 liquid chromatograph equipped with a UV Uvicord SV
detector operating at 226 nm.All separations were accomplished on a
reversed-phase semi-preparative C18 column (Vydac model218TP1010)
using a solvent gradient (solvent A: 0.05% TFA in H2O; solvent B: 0.05%
TFA in 9:1 CH3CN/H2O).Purity was confirmed to be at least 95% by
analytical HPLC performed using a Gilson liquid chromatograph equipped
with a model306 double pump, a manometric device model306, a Rainin
UV detector (DINAMAX operating at 226 nm), and a reversed-phase C18
column (Vydac model218TP54) using the same solvent mixture.
mixture until two liquid layers separated.A solution of methyl-4-
cyanobenzoate (10 g, 62.05 mmol), dissolved in diethyl ether (50 mL),
was added and the mixture was vigorously stirred for two days.A white
precipitate formed, the mixture was decanted overnight, and the solid
collected by filtration.A 10% NaHCO solution was then added to the
3
residue until all the acid was neutralized.The mixture was extracted three
times with dichloromethane, the organic phase was then dried over
Na2SO4, and the solvent was evaporated to dryness.Purification on silica
gel chromatography (CHCl3) afforded the desired product 1 (5.18 g, 51%).
M.p. 628C (lit.61 62 8C)[39]; 1H NMR (250 MHz, CDCl3): d ˆ 3.95 (s, 3H;
OCH3), 7.92, 7.95, 8.15, 8.18, (AA'BB', 4H; Ph), 10.07 (s, 1H; CHO).
Membrane permeability measurements: Leakage from vesicles was
evaluated using the CF-entrapped vesicle technique.[15a,c, 18, 22] CF-encapsu-
lated vesicles were prepared according to the following procedure: egg
phosphatidylcholine (20.5 mg) and cholesterol (4.5 mg) were dissolved in
freshly distilled CH2Cl2 (5 mL).The resulting mixture was stirred for 5 min
under nitrogen and then evaporated under a steady nitrogen stream.The
residue was dried under vacuum and then allowed to swell for one night at
room temperature under nitrogen in a pH 7.4 HEPES solution (0.05m,
5 mL) containing CF (50 mm).The slurry was sonicated (Braunsonic
immersion probe sonicator, 40% duty cycle, output control set to five) at
08C for 45 min, and the solution was filtered through a 0.45 mm Millipore
filter.Unencapsulated CF was removed by passing the solution through a
Sephadex G-75 column (0.1m NaCl, 0.01m HEPES, pH 7.4). The fractions
containing the liposomes were collected, and the total lipid concentration
was determined by phosphate analysis.[38] The liposomes were then diluted
with buffer to a 4 Â 10À5 m for the leakage experiments.Three mL of this
solution were used for each experiment.The phospholipid concentration
was kept constant, and increasing [peptide]/[lipid] molar ratios were
obtained by adding aliquots of the peptide dissolved in a 1:1 methanol/
water mixture (pH 7.4, HEPES 0.01m; the pH refers to the aqueous phase
before mixing).Methanol concentration was kept below 2% in all the
experiments.After peptide addition the mixture was vigorously stirred, and
the time course of the fluorescence change corresponding to CF efflux was
recorded at 258C on a Perkin Elmer LS50B instrument (lem 520 nm, lex
488 nm).The percentage of released CF was determined according to
Equation (1):
N,N,N-Tris{2-[(4-methoxycarbonylphenyl)methylenimino]ethyl}amine
(2): TREN (1.54 g, 10.5 mmol) was dissolved in dry CH3CN (40 mL) and
then added dropwise to a magnetically stirred solution of 1 (5.18 g,
31.5 mmol) in the same solvent (100 mL). After 4 h, a white precipitate
formed.The solid was collected by filtration and then washed with
additional solvent.The organic solution was evaporated to dryness, and the
oily residue was dissolved in toluene (10 mL) and then evaporated to
dryness again.After repeating this procedure three times, a white solid
compound was recovered.This was washed with CH 3CN and combined to
the previous fraction to obtain compound 2 (5.2 g, 85%). The product was
used directly in the following step. 1H NMR (250 MHz, CDCl3): d ˆ 2.95 (t,
J ˆ 6.25 Hz, 6H; NCH2CH2), 3.70 (t, J ˆ 6.25 Hz, 6H; NCH2CH2), 3.94 (s,
9H; OCH3), 7.55, 7.59, 7.97, 8.01 (AA'BB', 12H; Ph), 8.13 (s, 3H; NCHPh).
N,N,N-Tris{2-[(4-methoxycarbonylphenyl)methylamino]ethyl}amine (3):
A solution of 2 (5.1 g, 87.2 mmol) dissolved in the minimum amount of
methanol was carefully added to a suspension of NaBH4 (1.5 g, 39.7 mmol)
in the same solvent (100 mL) which was stirred under nitrogen at 08C in an
ice bath.The ice bath was removed, and the mixture was stirred at room
temperature overnight.Water (3 mL) was added, and the mixture was
stirred for 30 min.The solvent was removed under vacuum, and H 2O
(50 mL) was added to the resulting oily residue.The aqueous phase was
extracted with CH2Cl2 (3 Â 50 mL), the combined organic phases were
dried over Na2SO4, and the solvent was removed under vacuum to afford
compound 3 (5.15 g, 99%) as a yellow oil. 1H NMR (250 MHz, CDCl3): d ˆ
1.78 (brs, 3H; NH), 2.57 2.59 (m, 6H; NCH2CH2), 2.63 2.65 (m, 6H;
NCH2CH2), 3.78 (s, 9H; OCH3), 3.90 (s, 6H; CH2Ph), 7.29, 7.32, 7.91, 7.94
(AA'BB', 12H; Ph).
% CF ˆ (Ft À F0)/(FT À F0)  100
(1)
N,N,N-Tris{2-[(4-methoxycarbonylphenyl)methyl(tert-butoxycarbonyl)-
amino]ethyl}amine (4): A mixture of di-tert-butyldicarbonate (2.97 g,
13.6 mmol) and compound 3 (2.41 g, 4.08 mmol) in freshly distilled CH2Cl2
(100 mL) was stirred overnight at room temperature with protection by a
CaCl2 drying tube.NaHCO 3 (5%, 30 mL) was added and the mixture was
stirred for 30 min.The aqueous phase was extracted with CH 2Cl2 (3 Â
10 mL).The organic layers were combined, dried over Na 2SO4, and the
solvent was evaporated.Purification on silica gel (CHCl 3/ethyl acetate 6:1)
afforded the desired product 4 (2.5 g, 69%) as a yellow oil. 1H NMR
(250 MHz, CDCl3, 328 K): d ˆ 1.40 (s, 27H; Boc), 2.55 (brt, 6H;
NCH2CH2), 3.18 (brt, 6H; NCH2CH2), 3.91 (s, 9H; OCH3), 4.43 (s, 6H;
CH2Ph), 7.23, 7.26, 7.96, 7.99 (AA'BB', 12H; Ph).
where F0 is the fluorescence intensity of the vesicles suspension in the
absence of peptide, Ft is the fluorescence intensity at time t in the presence
of peptide, and FT is the total fluorescence determined by disrupting the
vesicles by addition of 30 mL of a 10% Triton X-100 solution.
Vesicle fusion assay: Peptide-induced vesicle fusion was monitored using
resonance energy transfer (RET) between PE-NBD and PE-LRB
probes.[31] PE-NBD and PE-LRB-loaded vesicles were prepared as
following: egg phosphatidylcholine (11.0 mg) and cholesterol (2.4 mg)
were dissolved in freshly distilled CH2Cl2 (2 mL).This solution was split in
two aliquots (each 1 mL), to which were added PE-NBD (230 mL of a
3.12 Â 10À4 m solution in CH2Cl2) and PE-LRB (100 mL of a 7. 14Â 10À4
m
solution in CH2Cl2) to obtain a lipid/probe ratio of 100:1.The organic
solvent was evaporated under a steady nitrogen stream followed by high
vacuum for two hours.The two lipid preparations were dispersed in 10 mL
of a pH 7.4, 0.01m HEPES/0.1m NaCl buffer solution, sonicated separately
(Braunsonic immersion probe sonicator, 40% duty cycle, output control set
to five) at 08C for 45 min, and the solutions obtained were filtered through
a 0.45 mm Millipore filter.The fluorophore-labeled liposomes (160 mL of
each) were diluted with buffer in the same cuvette to a final volume of
2 mL.Fluorescence emission spectra (475 700 nm, lex 450 nm) were
recorded at 258C on a Perkin Elmer LS-50B instrument before and after
the addition of increasing amounts of the peptide dissolved in a 1:1
methanol/water mixture (pH 7.4, HEPES 0.01m; the pH refers to the
aqueous phase before mixing).
N,N,N-Tris{2-[(4-carboxyphenyl)methyl(tert-butoxycarbonyl)amino]-
ethyl}amine (5): NaOH (1m, 20 mL) was added to a solution of 4 (2.5 g,
2.81 mmol) in EtOH (140 mL). After stirring the mixture at room
temperature for 24 h, no more starting material was detected by TLC
(CHCl3/MeOH 9:1).The solvent was evaporated leaving a white residue
that was collected and dissolved in H2O (30 mL).After adjusting to pH 4
with 1m HCl, a white solid precipitated.This was separated by centrifu-
gation, washed several times with neutral water, and then dried under
vacuum in the presence of anhydrous CaCl2 to obtain compound 5 (1.7 g,
71%).Mp. .153 155 8C; 1H NMR (250 MHz, CD3OD, 328 K): d ˆ 1.44 (s,
27H; Boc), 2.62 (brm, 6H; NCH2CH2), 3.26 (brm, 6H; NCH2CH2), 4.51 (s,
6H; CH2Ph), 7.32, 7.36, 8.00, 8.03 (AA'BB', 12H; Ph).
N,N,N-Tris{2-[(4-(isopropyl-NH-Aib-Leu-Gly)carboxyphenyl)methyl-
(tert-butoxycarbonyl)amino]ethyl}amine (6): EDC (80 mg, 0.42 mmol) was
added to a solution of compound 5 (95 mg, 0.112 mmol) and HOBt (50 mg,
0.37 mmol) in freshly distilled CH2Cl2 (2 mL) which was kept at 08C.The
mixture was stirred for 15 min at the same temperature and then mixed
with a second solution prepared by dissolving tripeptide P3 (200 mg,
0.64 mmol) in the same solvent (30 mL). The mixture was left stirring at
room temperature overnight.Further EDC (20 mg) was added, and the
Syntheses: The synthesis of Z-protected peptides P3, P4, P5, P6, and P10
has already been reported.[23] Deprotection was performed by hydro-
genation on Pd/C following standard procedures.Analytical HPLC profiles
of compounds T(P4)3 T(P6)3 and T(P10)3, their 1H NMR spectra and
MALDI-MS spectra are reported in the Supporting Information.
4-Formylmethylbenzoate (1): Stannous chloride (33.33 g, 175.7 mmol) was
suspended in diethyl ether (100 mL) and HCl gas was bubbled into the
2760
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0812-2760 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 12

Template-Assembled Tripodal Polypeptides
2753 2763
mixture was stirred for an additional 3 h.The organic solution was washed
with H3BO3 (0.5m, 3 Â 10 mL), NaHCO3 (10%, 3 Â 10 mL), and H2O (3 Â
10 mL), dried over Na2SO4, and then concentrated under reduced pressure.
The oily residue was precipitated from CH2Cl2/light petroleum and purified
by flash chromatography (CHCl3/EtOH 15:1 to 7:1) to afford compound 6
(131 mg, 71%). 1H NMR (250 MHz, CDCl3): d ˆ 0.88 (dd, 18H;
LeudCH3), 1.02 (dd, 18H; NHCH(CH3)2), 1.40 (s, 27H; Boc), 1.45 (s,
18H; AibbCH3), 1.65 (m, 9H; LeubCH2, gCH), 2.46 (brm, 6H;
NCH2CH2), 3.10 (brm, 6H; NCH2CH2), 3.85 4.00 (m, 6H; GlyaCH2),
4.00 4.20 (m, 3H; NHCH(CH3)2), 4.20 4.30 (m, 3H; LeuaCH), 4.30 (brs,
6H; CH2Ph), 6.40 (brd, 3H; NHCH(CH3)2), 6.88 (s, 3H; AibaNH), 7.15,
7.17 (AA'BB', 6H; Ph), 7.29 (brs, 3H; LeuaNH), 7.67 (brs, 3H; GlyaNH),
7.72, 7.75 (AA'BB', 6H; Ph).
and g'CH3), 1.30 1.90 (m, 54H; LeugCH, Boc, AibbCH3, LeubCH2),
1.95 2.15 (m, 3H; IlebCH), 2.62 (brm, 6H; NCH2CH2), 3.25 (brm, 6H;
NCH2CH2), 3.70 (s, 9H; OCH3), 3.71 4.00 (m, 6H; GlyaCH2), 4.21 4.41
(m, 6H; IleaCH, LeuaCH), 4.50 (brs, 6H; CH2Ph), 4.55 4.70 (m, 3H;
LeuaCH), 7.25 7.45 (brAA', 6H; Ph), 7.87, 7.9 (BB', 6H; Ph).
N,N,N-Tris{2-[(4-(CH3O-Leu-Ile-Gly-Aib-Leu)carboxyphenyl)methyl-
amino]ethyl}amine (T(P5)3): Compound 8 (85 mg, 0.04 mmol) was dis-
solved in 3:1 CH2Cl2/CF3COOH (16 mL), and the mixture was stirred at
room temperature for 1 h.The solvent was evaporated under vacuum to
give a residue that was purified by semi-preparative reversed-phase HPLC.
(gradient from 50% to 62% solvent B over 25 min, flow rate 2.5 mLminÀ1
,
retention time 16.7 min). The fractions containing the pure product were
combined and lyophilized to give the title compound (44 mg, 51%) as a
white solid; the purity was ascertained by analytical reversed-phase HPLC
(linear gradient from 50% to 68% solvent B over 36 min, flow rate
1 mLminÀ1, retention time 18.81 min). M.p. 155 1588C; [a]2D0 ˆ À35.0
(c ˆ 0.1 in MeOH); 1H NMR (400 MHz, CD3OD): d ˆ 0.75 1.10 (m, 42H;
LeudCH3, IledCH3, gCH2, g'CH3), 1.20 1.90 (m, 36H; LeugCH,
AibbCH3, LeubCH2), 1.90 2.10 (m, 3H; IlebCH), 2.93 (brm, 6H;
NCH2CH2), 3.23 (brm, 6H; NCH2CH2), 3.72 (s, 9H; OCH3), 3.72, 3.78,
3.88, 3.94 (AB, 6H; GlyaCH2), 4.20 4.40 (m, 12H; LeuaCH, IleaCH,
GlyaCH2, CH2Ph), 4.60 4.70 (m, 3H; LeuaCH), 7.62, 7.64, 7.97, 8.00
N,N,N-Tris{2-[(4-(isopropyl-NH-Aib-Leu-Gly)carboxyphenyl)methylami-
no]ethyl}amine (T(P3)3): Compound 6 was dissolved in a 3:1 CH2Cl2/
CF3COOH solution (20 mL) and the mixture was stirred at room temper-
ature for 1 h.The solvent was then evaporated under vacuum to afford the
title compound as a white solid.Mp. .130 132 8C; [a]2D0 ˆ À18.4 (c ˆ 0.5 in
MeOH); 1H NMR (250 MHz, CD3OD): d ˆ 0.99 (dd, 18H; LeudCH3), 1.04
(dd, 18H; NHCH(CH3)2), 1.49 (s, 18H; AibbCH3), 1.60 (m, 6H; LeubCH2),
1.64 (m, 3H; LeugCH), 2.99 (brt, 6H, NCH2CH2), 3.26 (brt, 6H;
NCH2CH2), 3.94 (m, 3H; NHCH(CH3)2), 4.11 (dd, 6H; GlyaCH2), 4.23
(m, 3H; LeuaCH), 4.34 (s, 6H; CH2Ph), 7.62, 7.65, 7.95, 7.99 (AA'BB', 12H;
Ph); elemental analysis calcd (%) for C75H120O12N16 ¥ 8CF3COOH ¥ H2O: C
46.16, H 5.53, N 9.46; found: C 46.03, H 5.49, N 9.00.
‡
‡
(AA'BB', 12H; Ph); MALDI-MS: m/z: 2059 [M‡Na] , 2037 [M] .
N,N,N-Tris{2-[(4-(CH3O-Leu-Ile-Gly-Aib-Leu-Gly)carboxyphenyl)meth-
yl(tert-butoxycarbonyl)amino]ethyl}amine (9): A procedure similar to that
described for the synthesis of compound 7 was followed using EDC
(37.3 mg, 0.195 mmol), compound 5 (47.2 mg, 0.055 mmol), HOBt (24 mg,
0.18 mmol), and hexapeptide P6 (174 mg, 0.305 mmol). The crude oily
material was purified by column chromatography (silica gel, CHCl3/EtOH
from 9:1 to 7:3) to give compound 9 (55 mg, 39%). 1H NMR (250 MHz,
CD3OD): d ˆ 0.80 1.10 (m, 60H; LeudCH3, IledCH3, gCH2, g'CH3),
1.20 1.80 (m, 57H; AibbCH3, Boc, LeubCH2), 1.97 2.20 (m, 3H;
IlebCH), 2.63 (brm, 6H; NCH2CH2), 3.25 (brm, 6H; NCH2CH2), 3.68 (s,
9H; OCH3), 3.90 4.60 (m, 27H; GlyaCH2, LeuaCH, IleaCH, CH2Ph),
7.35 7.45 (brAA', 6H; Ph), 7.83, 7.89, (BB', 6H; Ph).
N,N,N-Tris{2-[(4-(CH3O-Leu-Ile-Gly-Aib)carboxyphenyl)methyl(tert-bu-
toxycarbonyl)amino]ethyl}amine (7): EDC (25.1 mg, 0.13 mmol) was
added to a solution of compound 5 (33.1 mg, 0.04 mmol) and HOBt
(15.2 mg, 0.11 mmol) in freshly distilled CH2Cl2 (1 mL) at 08C.The mixture
was stirred for 15 min at this temperature and then mixed with a second
solution prepared by dissolving tetrapeptide P4 (89.9 mg, 0.22 mmol) in the
same solvent (3 mL).After 2 h, EDC (15 mg) was added, and the mixture
stirred for an additional 24 h.The organic solution was then washed with
H3BO3 (0.5m, 3 Â 10 mL), NaHCO3 (10%, 3 Â 10 mL), and H2O (3 Â
10 mL), dried over Na2SO4, and concentrated under reduced pressure.
The residual crude oily product was purified by silica column chromatog-
raphy (CHCl3/EtOH 9:1 ! 8:2).Concentration of the proper fractions gave
compound 7 (58.2 mg, 74%). 1H NMR (250 MHz, CDCl3): d ˆ 0.80 1.10
(m, 42H; LeudCH3, IlegCH2, d-, g'CH3), 1.10 1.35 (m, 3H; LeugCH),
1.39 1.75 (m, 51H; Boc, AibbCH3, LeubCH2), 1.95 2.13 (m, 3H;
IlebCH), 2.46 (brm, 6H; NCH2CH2), 3.10 (brm, 6H; NCH2CH2), 3.67 (s,
9H; OCH3), 3.80 4.10 (m, 6H; GlyaCH2), 4.35 (d, J ˆ 6.25 Hz, 3H;
IleaCH), 4.49 4.62 (m, 3H; LeuaCH), 4.90 (brs, 6H; CH2Ph), 6.75 (brm,
6H; NH), 7.10 (brm, 3H; NH), 7.25 (brm, 3H; NH), 7.30, 7.35, 7.72, 7.76
(AA'BB', 12H; Ph).
N,N,N-Tris{2-[(4-(CH3O-Leu-Ile-Gly-Aib-Leu-Gly)carboxyphenyl)meth-
ylamino]ethyl}amine (T(P6)3): Compound 9 (55 mg, 0.02 mmol) was dis-
solved in 3:1 CH2Cl2/CF3COOH (14 mL) and the mixture was stirred at
room temperature for 1 h.The solvent was evaporated under vacuum to
give a residue that was purified by semi-preparative reversed-phase HPLC
(linear gradient from 48 to 60% solvent B over 22 min, flow rate
2.5 mLminÀ1).The fractions containing the pure product were combined
and lyophilized to obtain the title compound (18 mg, 32%) as a white solid
whose purity was ascertained by analytical reversed-phase HPLC (linear
gradient from 50 to 68% solvent B over 36 min, flow rate 1 mLminÀ1
,
N,N,N-Tris{2-[(4-(CH3O-Leu-Ile-Gly-Aib)carboxyphenyl)methylamino]-
ethyl}amine (T(P4)3): Compound 7 (58.2 g, 0.03 mmol) was dissolved in a
3:1 CH2Cl2/CF3COOH solution (4 mL) and the mixture was stirred at room
temperature for 1 h.The solvent was then evaporated under vacuum to
leave a white solid that was purified by semi-preparative reversed-phase
retention time 20.96 min). M.p. 183 1858C; [a]2D0 ˆ À43.6 (c ˆ 0.5 in
MeOH); 1H NMR (400 MHz, CD3OD): d ˆ 0.80 1.10 (m, 60H; LeudCH3,
IlegCH2, g'CH3, dCH3), 1.20 1.80 (m, 36H; LeugCH, LeubCH2,
AibbCH3), 2.00 2.20 (m, 3H; IlebCH), 2.93 (brm, 6H; NCH2CH2), 3.23
(brm, 6H; NCH2CH2), 3.51, 3.56, 3.66, 3.90 (AB, 6H; GlyaCH2), 3.69 (s,
9H; OCH3), 4.10 4.50 (m, 21H; GlyaCH2, LeuaCH, IleaCH, CH2Ph),
HPLC (isocratic conditions, 47% solvent B, flow rate 2 mLminÀ1
,
11.57 min retention time). The fractions containing the pure product were
combined and lyophilized to obtain the title compound (12.3 mg, 21%) as a
white solid; the purity was ascertained by analytical reversed-phase HPLC
(linear gradient from 40% to 59% solvent B over 36 min, flow rate
1 mLminÀ1, retention time 12.04 min). M.p. 142 1448C; [a]2D0 ˆ À24.0
(c ˆ 0.1 in MeOH); 1H NMR (400 MHz, CD3OD): d ˆ 0.80 1.10 (m, 42H;
LeudCH3, IlegCH2, g'CH3), 1.20 1.40 (m, 3H; LeugCH), 1.50 1.80 (m,
27H; AibbCH3, LeubCH2), 2.00 2.20 (m, 3H; IlebCH), 2.92 (brm, 6H;
CH2CH2N), 3.23 (brm, 6H; CH2CH2N), 3.73 (s, 9H; OCH3), 3.74 4.00 (m;
GlyaCH2), 4.29 (d, J ˆ 6.25 Hz, 3H; IleaCH), 4.38 (s, 6H; CH2Ph), 4.40
4.50 (m, 3H; LeuaCH), 7.67, 7.69, 8.07, 8.09 (AA'BB', 12H; Ph); MALDI-
‡
7.62, 7.64, 7.95, 7.97 (AA'BB', 12H; Ph); MALDI-MS: m/z: 2225 [M‡Na] .
N,N,N-Tris{2-[(4-(CH3O-Leu-Ile-Gly-Aib-Leu-Gly-Gly-Aib-Leu-Gly)car-
boxyphenyl)methyl(tert-butoxycarbonyl)amino]ethyl}amine (10): A pro-
cedure similar to that described for the synthesis of compound 7 was
employed using EDC (14.7 mg, 0.08 mmol), compound 5 (18.6 mg,
0.02 mmol), HOBt (8.9 mg, 0.065 mmol), and decapeptide P10 (96.2 mg,
0.11 mmol). The resulting oily material was purified by column chroma-
tography (silica gel, CHCl3/MeOH 9:1) to give compound 10 (66.6 mg,
88%). 1H NMR (250 MHz, CD3OD, 398 K): d ˆ 0.85 1.15 (m, 78H;
LeudCH3, IledCH3, gCH2, g'CH3), 1.20 1.85 (m, 90H; AibbCH3, Boc,
LeubCH2, LeugCH), 2.00 2.20 (m, 3H; IlebCH), 2.65 (brt, 6H;
NCH2CH2), 3.30 (brt, 6H, NCH2CH2), 3.67 3.83 (m, 15H; GlyaCH2,
OCH3), 3.89 (d, J ˆ 10.5 Hz, 12H; GlyaCH2), 4.08 (d, J ˆ 8.7 Hz, 6H;
GlyaCH2), 4.33 (t, J ˆ 7.5 Hz, 12H; LeuaCH, IleaCH), 7.34, 7.37, 7.86, 7.89
(AA'BB', 12H; Ph).
‡
‡
‡
MS: m/z: 1732.1 [M‡K] , 1716.0 [M‡Na] , 1694.4 [M] .
N,N,N-Tris{2-[(4-(CH3O-Leu-Ile-Gly-Aib-Leu)carboxyphenyl)methyl-
(tert-butoxycarbonyl)amino]ethyl}amine (8): A procedure similar to that
described for the synthesis of compound 7 was followed using EDC
(34.4 mg, 0.18 mmol), compound 5 (45.4 mg, 0.05 mmol), HOBt (20.8 mg,
0.15 mmol), and pentapeptide P5 (158 mg, 0.31 mmol). The crude oily
material obtained was purified by column chromatography (silica gel,
CHCl3/EtOH 8.5:1.5) to give compound 8 (85 mg, 69%). 1H NMR
(250 MHz, CD3OD): d ˆ 0.80 1.05 (m, 42H; LeudCH3, IledCH3, gCH2
N,N,N-Tris{2-[(4-(CH3O-Leu-Ile-Gly-Aib-Leu-Gly-Gly-Aib-Leu-Gly)car-
boxyphenyl)methylamino]ethyl}amine (T(P10)3): Compound 10 (66.6 mg,
0.02 mmol) was dissolved in 3:1 CH2Cl2/CF3COOH (10 mL) and the
mixture was stirred at room temperature for 1 h.The solvent was
Chem. Eur. J. 2002, 8, No. 12
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0812-2761 $ 17.50+.50/0
2761

P.Scrimin, U.Tonellato et al.
FULL PAPER
evaporated under vacuum to leave a residue that was purified by semi-
preparative reversed-phase HPLC (isocratic conditions 63% solvent B,
flow rate 2 mLminÀ1, retention time 22.2 min). The fractions containing the
pure product were combined and lyophilized to obtain the title compound
(11 mg, 16%) as a white solid; the purity was ascertained by analytical
reversed-phase HPLC (linear gradient from 60% to 88% solvent B over
30 min, retention time 20.96 min). M.p. 197 1998C; [a]2D0 ˆ À31.6 (c ˆ 0.5
in MeOH); 1H NMR (400 MHz, CD3OD): d ˆ 0.85 1.15 (m, 78H;
LeudCH3, IledCH3, gCH2, g'CH3), 1.30 1.90 (m, 65H; AibaCH3,
LeubCH2, LeugCH), 1.90 2.20 (m, 3H; IlebCH), 2.95 (brm, 6H;
NCH2CH2), 3.25 (brm, 6H; NCH2CH2), 3.71 (s, 9H; OCH3), 3.70 4.50
(m, 36H; GlyaCH2, LeuaCH, IleaCH), 7.62, 7.64, 7.95, 7.97 (AA'BB', 12H;
2297; d) M.M.Tedesco, B.Ghebremariam, N.Sakai, S.Matile, Angew.
Chem. 1999, 111, 523; Angew. Chem. Int. Ed. Engl. 1999, 38, 541; e) N.
Sakai, B.Baumeister, S.Matile, ChemBioChem 2000, 2, 123; f) C.D.
Hall, G.J.Kirkovits, A.C.Hall,
Chem. Commun. 1999, 1897; g) D.
Wang, L.Guo, J.Zhang, L.R.Jones, Z.Chen, C.Pritchard, R.W.
Roeske, J. Pept. Res. 2001, 57, 301.
[10] For selected examples see: a) S.Shinkai, M.Ikeda, A.Sugasaki, M.
Takeuchi, Acc. Chem. Res. 2001, 34, 494; b) F.Wang, A.W.Schwa-
bacher, J. Org. Chem. 1999, 64, 8922; c) P.Tecilla, U.Tonellato, A.
Veronese, F.Felluga, P.Scrimin, J. Org. Chem. 1997, 62, 7621; d) P.
Scrimin, P.Tecilla, U.Tonellato, G.Valle, A.Veronese, J. Chem. Soc.
Chem. Commun. 1995, 1163; e) R.Baldes, H-.J.Schneider,
Chem. 1995, 107, 380; Angew. Chem. Int. Ed. Eng. 1995, 34, 321.
[11] a) I.O.Fritsky, R.Ott, R.Kr‰mer, Angew. Chem. 2000, 112, 3403;
Angew.
‡
Ph); MALDI-MS: m/z: 3183 [M‡K] .
Angew. Chem. Int. Ed. 2000, 39, 3255; b) I.O. Fritsky, R. Ott, H.
Pritzkow, R.Kr‰mer, Chem. Eur. J. 2001, 7, 1221.
Acknowledgements
[12] S.Valente, M.Gobbo, G.Licini, A.Scarso, P.Scrimin,
Angew. Chem.
2001, 113, 4017; Angew. Chem. Int. Ed. 2001, 40, 3899.
The authors thank Dr.R.Battistutta for the molecular modeling and Drs.
R.Mantovani, L.Corte, and N.Paggiarin for synthetic work.This work was
funded by MIUR (COFIN2000-MM03194891) and by Regione Friuli
Venezia Giulia ™Fondo 2000∫.
[13] a) E.Benedetti, A.Bavoso, B.Di Blasio, V.Pavone, C.Pedone, C.
Toniolo, G.M. Bonora, Proc. Natl. Acad. Sci USA 1982, 79, 7951;
b) M.S.P.Sansom, Q. Rev. Biophys. 1993, 26, 365.
[14] a) R.C.Pandey, J.C.Cook, Jr,. K.L.Rinehart,
1977, 99, 8469; b) R.O. Fox, F.M. Richards, Nature 1982, 300, 325.
[15] a) C.Auvin-Guette, S.Rebuffat, Y.Prigent, B.Bodo, J. Am. Chem.
Soc. 1992, 114, 2170; b) C.Toniolo, C.Peggion, M.Crisma, F.
Formaggio, X.Shui, D.S.Eggleston, Nat. Struct. Biol. 1994, 1, 908;
J. Am. Chem. Soc.
[1] M.Jain, Introduction to Biological Membranes, 2nd ed., Wiley, New
York, 1988, Chapter 9.
c) V.Monaco, E.Locardi, F.Formaggio, M.Crisma, S.Mammi, E.
Peggion, C.Toniolo, S.Rebuffat, B.Bodo, J. Pept. Res. 1998, 52, 261.
[16] a) T.Fujita, S.Wada, A.Iida, T.Nishimura, M.Kanai, N.Toyama,
Chem. Pharm. Bull. 1994, 42, 489; b) V.Monaco, F.Formaggio, M.
Crisma, C.Toniolo, X.Shui, D.S.Eggleston, Biopolymers 1996, 39, 31;
c) R.Gurunath, P.Balaram, Biopolymers 1995, 35, 21.
[2] D.W.Urry, Top. Curr. Chem. 1985, 128, 175.
[3] a) M.Montal, Curr. Opin. Struct. Biol. 1995, 5, 501; b) F.Hucho, C.
Weise, Angew. Chem. 2001, 113, 3194; Angew. Chem. Int. Ed. 2001, 40,
3101 and references therein.
[4] a) T.M.Fyles, T.D.James, K.C.Kaye,
12315; b) M.F.M.Rocks, R.J.M.Nolte,
J. Am. Chem. Soc. 1993, 115,
Macromolecules 1992, 25,
[17] a) C.Toniolo, E.Benedetti, Trends Biochem. Sci. 1991, 16, 350; b) C.
Toniolo, E.Benedetti, Macromolecules 1991, 24, 4004; c) I.L.Karle, P.
Balaram, Biochemistry 1990, 29, 6747.
[18] T.Le Doan, M.El Hajji, S.Rebuffat, MR. .Rajesvari, B.Bodo,
Biochim. Biophys. Acta 1986, 858, 1.
[19] P.Scrimin, A.Veronese, P.Tecilla, P.Tonellato, V.Monaco, F.
Formaggio, M.Crisma, C.Toniolo, J. Am. Chem. Soc. 1996, 118, 2505.
[20] For template-assembled synthetic peptides see: a) M.Mutter, S.
Vuilleumier, Angew. Chem. 1989, 101, 551; Angew. Chem. Int. Ed.
Engl. 1989, 28, 535; b) M.Mutter, P.Dumy, P.Garrouste, C.Lehmann,
M.Mathieu, C.Peggion, S.Peluso, A.Razaname, G.Tuchscherer,
Angew. Chem. 1996, 108, 1587; Angew. Chem. Int. Ed. Engl. 1996, 35,
1482.For peptide bundles affecting membrane permeability see:
c) K.S.äkerfeldt, R.M.Kim, D.Camac, J.T.Groves, J.D.Lear, W.F.
DeGrado, J. Am. Chem. Soc. 1992, 114, 9656; d) A.Matsubara, K.
Asami, A.Akagi, N.Nishino, J. Chem. Soc. Chem. Commun. 1996,
2069.
5398; c) H.Wagner, K.Harms, U.Koert, S.Meder, G.Boheim, Angew.
Chem. 1996, 108, 2836; Angew. Chem. Int. Ed. Engl. 1996, 35, 2643;
d) Y.Tanaka, Y.Kobuke, M.Sokabe, Angew. Chem. 1995, 107, 717;
Angew. Chem. Int. Ed. Engl. 1995, 34, 693; e) G.W.Gokel, O.Murillo,
Acc. Chem. Res. 1996, 29, 425; f) F.M. Menger, D.S. Davis, R.A.
Persichetti, J.J. Lee, J. Am. Chem. Soc. 1990, 112, 2451; g) F.M.
Menger, P.Aikens, Angew. Chem. 1992, 104, 919; Angew. Chem. Int.
Ed. Engl. 1992, 31, 898; h) U.F. Kragten, M.F.M. Roks, R.J.M.
Nolte, J. Chem. Soc. Chem. Commun. 1985, 1275; i) N.Voyer, M.
Robitaille, J. Am. Chem. Soc. 1995, 117, 6599; j) T.M.Fyles, D.Loock,
X.W.F. van Straaten-Nijenhuis, X. Zhou,
8866; k) O.Murillo, I.Suzuki, E.Abel, G.W.Gokel,
J. Org. Chem. 1996, 61,
J. Am. Chem.
Soc. 1996, 118, 7628; l) N.Sakai, K.C.Brennan, L.A.Weiss, S.Matile,
J. Am. Chem. Soc. 1997, 119, 8726; m) N.Sakai, S.Matile, Chem. Eur.
J. 2000, 6, 1731; n) A.F.DiGiorgio, S.Otto, P.Bandyopadhyay, S.L.
Regen, J. Am. Chem. Soc. 2000, 122, 11029; o) G.W.Gokel, Chem.
Commun. 2000, 1; p) N.Yoshino, A.Satake, Y.Kobuke,
Angew.
[21] R.H. Prince in Comprehensive Coordination Chemistry (Eds.: G.
Wilkinson, R.D.Gillard, A.McCleverty), Pergamon Press, London,
1987, pp.937 938.
Chem. 2001, 113, 471; Angew. Chem. Int. Ed. 2001, 40, 457; q) G.W.
Gokel, A.Mukhopadhyay, Chem. Soc. Rev. 2001, 30, 274.
[5] a) A.Grove, M.Mutter, J.E.Rivier, M.Montal,
J. Am. Chem. Soc.
[22] JN. .Winstein, S.Yoshikami, P.Henkari, R.Blumenthal, WH. .
Hagins, Science 1977, 195, 489.
[23] C.Toniolo, M.Crisma, F.Formaggio, C.Peggion, V.Monaco, C.
Goulard, S.Rebuffat, B.Bodo, J. Am. Chem. Soc. 1996, 118, 4952.
1993, 115, 5919; b) K.S.äkerfeldt, J.DL. ear, Z.R.Wasserman, L.A.
Chung, W.F. DeGrado, Acc. Chem. Res. 1993, 26, 191; M.S.P.
Sansom, Prog. Biophys. Mol. Biol. 1991, 55, 139.
[6] a) M.J.Pregel, L.Jullien, J.Canceill, L.Lacombe, J.M.Lehn,
Chem. Soc. Perkin Trans. 2 1995, 417; b) L.Jullien, T.Lazrak, J.
Canceill, L.Lacombe, J.M.Lehn, J. Chem. Soc. Perkin Trans. 2 1993,
1011.
J.
[24] a) M.Crisma, G.Valle, V.Monaco, F.Formaggio, C.Toniolo,
Acta
Crystallogr. Sect. C 1994, 50, 563; b) M.Crisma, G.Valle, F.
Formaggio, V.Monaco, C.Toniolo, Lett. Pept. Sci. 1996, 3, 121.
[25] F.Formaggio, M.Crisma, P.Rossi, P.Scrimin, B.Kaptein, QB. .
Broxterman, J.Kamphuis, C.Toniolo, Chem. Eur. J. 2000, 6, 4498.
[26] a) G.Anderegg, V.Gramlich, Helv. Chim. Acta 1994, 77, 685; b) J.W.
[7] a) R.M.Ghadiri, J.R.Granja, L.K.Buehler,
Nature 1994, 26, 301;
b) N.Khazanovich, J.R.Granja, D.E.McRee, R.A.Milligan, M.R.
Ghadiri, J. Am. Chem. Soc. 1994, 116, 6011; c) M.R. Ghadiri, J.R.
Canary, J.Xu, J.M.Castagnetto, D.Rentzeperis, L.A.Marky,
J. Am.
Granja, R.A.Milligan, D.E.McRee, N.Khazanovich,
Nature 1993,
Chem. Soc. 1995, 117, 11545.
366, 324; d) H.S.Kim, J.D.Hartgerink, M.R.Ghadiri,
J. Am. Chem.
[27] D.F.Kennedy, M.Crisma, C.Toniolo, D.Chapman,
Biochemistry
Soc. 1998, 120, 4417; e) S.Fernandez-Lopez, H.-S.Kim, E.C.Choi, M.
Delgado, J.R.Granja, A.Khasanov, K.Kraehenbuehl, G.Long, D.A.
Weinberger, K.M. Wilcoxen, M.R. Ghadiri, Nature 2001, 412, 452.
[8] J.M. Lehn, Angew. Chem. 1990, 102, 1347; Angew. Chem. Int. Ed.
Engl. 1990, 29, 1304.
1991, 30, 6541.
[28] For selected examples see: a) M.R. Ghadiri, C. Soares, C. Choi, J.
Am. Chem. Soc. 1992, 114, 4000; b) T.Sasaki, S.T.Kaiser,
Chem. Soc. 1989, 111, 380.
J. Am.
[29] G.Schwarz, C.H.Robert, Biophys. J. 1990, 58, 577.
[9] a) G.A.Woolley, A.S.I.Jaikaran, Z.Zhang, S.Peng,
J. Am. Chem.
[30] D.D.Perrin, B.Dempsey,
Buffers for pH and Metal Ion Control,
Soc. 1995, 117, 4448; b) T.M.Fyles, B.Zeng, Chem. Commun. 1996,
2295; c) T.M.Fyles, D.Loock, X.Zhou, J. Am. Chem. Soc. 1998, 120,
Chapman and Hall, London, 1974.
2762
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0812-2762 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 12

Template-Assembled Tripodal Polypeptides
2753 2763
[31] M.Kaszuba, G.R.Hunt, J. Inorg. Biochem. 1990, 40, 217.
[32] a) C.Puyal, L.Maurin, G.Miquel, A.Bienven¸e, J.Philippot,
Biochim. Biophys. Acta 1994, 1195, 259; b) D.Hoekstra, Biochemistry
1982, 21, 2833; c) D.K. Struck, D. Hoekstra, R.E. Pagano, D.
Hoekstra, Biochemistry 1981, 20, 4093.
[36] A similar extended geometry of complexation to a lipid bilayer by a
tetrasteroid-functionalized porphyrin has been reported: J.Lahiri,
G.D.Fate, S.B.Ungashe, J.T.Groves,
2347.
J. Am. Chem. Soc. 1996, 118,
[37] A reviewer pointed out that the role of ZnII could be to decrease the
partition of the template into the membrane as a consequence of a
lower lipophilicity of the metal complex.However, we note that at
pH 7.4 the TREN platform is triprotonated (pKa ˆ 10.1; 9.5; 8.4 see:
R.M.Smith, A.E.Martell, Critical Stability Constants, Vol. 6, Plenum
[33] P.Lauger, Science 1972, 178, 24.
[34] a) Y.Liu, S.L.Regen, J. Am. Chem. Soc. 1993, 115, 708; b) P.Scrimin,
P.Tecilla, R.A.Moss, K.Bracken, J. Am. Chem. Soc. 1998, 120, 1179.
[35] Recently Regen has reported compelling evidence that ionophores
may act with a dominant carrier mechanism at low concentrations and
with a channel mechanism at high concentrations (see: G.Deng, M.
II
Press, New York, 1989).In our experience the Zn complexes are
slightly less water soluble than the metal free (protonated) templates.
Merritt, K.Yamashita, V.Janout, A.Sadownik, S.L.Regen,
J. Am.
[38] P.S. Chen, T.Y. Toribara, H. Warner,
Anal. Chem. 1956, 28, 1756.
Chem. Soc. 1996, 118, 3307.). We cannot rule out that this is the case
with the present tripodal peptides and that the lower part of each
sigmoidal curve represents the carrier-mechanism contribution to the
transport process.
[39] K.Hass, M.Bender, J. Am. Chem. Soc. 1949, 71, 1767.
Received: December 27, 2001 [F3765]
Chem. Eur. J. 2002, 8, No. 12
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0812-2763 $ 17.50+.50/0
2763