Substrate modulation of the activity of an artificial nanoesterase made of peptide-functionalized gold nanoparticles

Article abstract of DOI:10.1002/anie.200602581

Nanozymes with a heart of gold: A functional artificial protein has been prepared by grafting a dodecapeptide onto the surface of gold nanoparticles (see picture). The system catalyzes the hydrolysis of carboxylate esters and features enzyme-like properti

Full text of DOI:10.1002/anie.200602581

Communications
DOI: 10.1002/anie.200602581
Artificial Proteins
Substrate Modulation of the Activity of an Artificial Nanoesterase
Made of Peptide-Functionalized Gold Nanoparticles**
Paolo Pengo, Lars Baltzer, Lucia Pasquato,* and Paolo Scrimin*
Dedicated to Professor Richard W. Franck on the occasion of his 70th birthday
Enzymes have evolved to work at optimal efficiency within a
precise interval of substrate concentration and, in many cases,
the substrate itself controls the activity of the enzyme in order
to modulate the metabolic cascade.[1] It would be desirable for
an artificial enzyme to present not only the rate accelerations
shown by natural enzymes but also these peculiar features. We
have shown in the last six years[2–4] how gold nanoclusters
passivated by a monolayer of functional thiols (Au-MPCs,
MPC = monolayer-protected cluster)[5] may manifest excep-
tional catalytic properties connected with the characteristics
of the functional groups of the monolayer and the multivalent
nature of the system.[6] The behavior of Au-MPCs led us to
coin the term “nanozymes” for these systems.[3] We report
herein the outstanding case of peptide-functionalized Au-
MPCs[7–13] that, like real enzymes, present unique catalytic
pathways not present in analogous monomeric systems,
including substrate modulation of activity during an ester-
olytic process.
well as stabilization of the negatively charged transition state
that arises along the pathway of ester hydrolysis. Further-
more, charged residues are known to favor the stabilization of
a peptide-capped nanoparticle.[11] In order to bring about
substrate binding and proximity effects, hydrophobic residues
were incorporated in positions buried in the core of the
peptide assembly, with one Phe and one Tyr close to the key
catalytic His residue. A Leu residue was also incorporated
into the sequence, in the position closest to the nanoparticle,
to further enhance the hydrophobic nature of the reactive site.
The positions close to the spherical nanoparticle are severely
constrained, and therefore an aliphatic side chain was favored
over an aromatic one in the design. While there may be
sequences that are capable of more efficient catalysis and a
thorough search of chemical space might help to identify
them, the number of combinations to screen becomes
enormous even for short peptides and the process would
require combinatorial strategies. Herein we have taken a
rational approach in order to investigate whether a simple
construct based on fundamental principles of organic reac-
tivity would be enough to generate efficient catalysis.
Following these principles, we have prepared, by using
standard solid-phase synthesis, five dodecapeptide-function-
alized thiols[14] and used them for the passivation of Au
nanoparticles by exploiting the site-exchange protocol and
starting from nanoparticles covered with the water-soluble
thiol N-(3,6,9-trioxadecyl)-8-sulfanyloctanamide (HS-C8-
TEG).[15] In these five peptides the His catalytic site was
moved along the sequence, with the aim of addressing the role
of an increasingly hydrophobic environment in the catalytic
process. Only the MPC functionalized with thiol 1 possessed
the solubility and stability required for the kinetic investiga-
tion. This functional MPC, Au-PEP, contains HS-C8-TEG
and peptide 1 in a 3:1 ratio, as determined by 1H NMR
spectroscopy; that is, there are approximately 62peptides per
nanoparticle on the basis of the average size (3.5 nm) of the
gold core. Structurally this nanosystem resembles proteins
both in size and in functionality.[16] The molecular weight of
the organic monolayer is approximately 150 kDa.
The peptide sequence was designed according to the
following principles. Ionizable residues are invariably
involved in catalysis in nature. A combination of one His
with two Arg and one Lys residue was expected to enable
nucleophilic, general-acid, and/or general-base catalysis as
[*] Prof. Dr. L. Pasquato
Dipartimento di Scienze Chimiche
Università degli Studi di Trieste
via L. Giorgieri, 1-34127 Trieste (Italy)
Fax: (+39)040-558-3903
E-mail: lpasquato@units.it
Homepage: http://www.dsch.units.it/pasquato
Dr. P. Pengo, Prof. Dr. P. Scrimin
Dipartimento di Scienze Chimiche
Università degli Studi di Padova
and CNR ITM, Padova Section
via Marzolo, 1-35131 Padova (Italy)
Fax: (+39)049-827-5239
E-mail: paolo.scrimin@unipd.it
Homepage: http://www.chimica.unipd.it/paolo.scrimin/pubblica/
Prof. Dr. L. Baltzer
Department of Biochemistry and Organic Chemistry
Uppsala University
Functional groups that may be involved in esterolytic
activity of Au-PEP are the imidazole N atom of His (pKa =
6.2), the phenol O atom of Tyr (pKa = 9.7), the amine N atom
of Lys (pKa = 10.4), and the guanidinium N atoms of Arg
(pKa = 12).[17,18] Furthermore, the free carboxylate group of
the C-terminal Arg residue may also be involved in a catalytic
process. At physiological pH values (around 7) the most
relevant functional groups are the imidazole unit of His
which, because of its pKa value, may act either as a general
Uppsala (Sweden)
[**] This work was funded by the Ministry of University and Research of
Italy (grant no. PRIN2003).
Supporting information for this article (including 1H NMR spectra
of the nanoparticles, high-resolution transmittance electron micro-
graphs, and bimodal kinetic traces in the cleavage of Z-Leu-PNP) is
available on the WWW under http://www.angewandte.org or from
the author.
400
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 400 –404

Angewandte
Chemie
catalyzed process.[20] The
profile is rather complex
and evidences the contribu-
tion to catalysis of three
nucleophilic species with
apparent pKa values of 4.2,
7.2, and 9.9. We attribute
the first pKa value to the
terminal carboxylate anion,
the second to the imidazole
moiety of His, and the third
to the amino group of Lys
or the phenolic hydroxy
group of Tyr. For compar-
ison, we have also reported
the profiles obtained with
nanoparticles functional-
ized with peptide 2 (Au-2)
and with S-acetylated pep-
tide 1 (see Figure 1).
At low pH values the
two nanoparticle-based cat-
alysts behave in a similar
way and catalysis depends
on moieties with compara-
ble pKa values, most likely
the terminal carboxylate
anion. Up to approximately
pH 5 Au-PEP is almost one
order of magnitude more
efficient than Au-2. This
base–general acid or as a nucleophile and the C-terminal
carboxylate which may act as a general-base catalyst. At
variance with our previous dipeptide-functionalized nano-
particles containing thiol 2[2] (functionalized with the dipep-
tide His-Phe-OH), the His residue in the present system is
buried inside the monolayer and is probably in a less solvated
region,[19] by analogy with what is typically the case for
functional groups in the catalytic site of enzymes. From the
structural point of view, the peptide sequence of 1 is not
expected to fold into a particular conformation.
amounts to a 3000-fold rate acceleration over that exerted
by the simple dipeptide AcHis-Phe-OH and a 500-fold
acceleration over that of S-acetylated 1. The most likely
explanation for this behavior is that a carboxylate anion acts
as a general base to activate a water molecule and a
In order to test the activity of Au-PEP we have run
kinetics studies with the activated substrate 2,4-dinitrophen-
ylbutyrate (DNPB) and the p-nitrophenyl esters of benzyl-
oxycarbonyl N-protected leucine and glycine (Z-Leu-PNP
and Z-Gly-PNP, respectively; PNP = p-nitrophenol). DNPB
is quite useful because the very low pKa value of 2,4-
dinitrophenol allows one to run kinetics studies down to pH
values of 3–4 by following the absorbance of the 2,4-
dinitrophenoxide ion at 400 nm (at lower wavelengths the
molar absorbtivity of the nanoparticles is very high). Z-Leu-
PNP and Z-Gly-PNP represent amino acid derivatives with
high (the first one) and low (the second one) lipophilicities.
They allow one to compare the role of their binding (and
hence of their insertion into the putative catalytic site of the
monolayer) in the catalytic process.
Figure 1. Logarithm of k2,app against the pH value for the hydrolysis of
*
*
DNPB catalyzed by Au-PEP nanoparticles ( ), Au-2 nanoparticles ( ),
^
and S-acetylated peptide 1 ( ). The solid lines represent the best fits
of functions describing the dissociation of residues involved in
catalysis with pKa values of 4.2, 7.2, and 9.9 for Au-PEP, 4.2 and 8.1
for Au-2, and 6.1 and 9.2 for S-acetylated 1. The dotted lines represent
the calculated contribution of each species to the solid curve for Au-
PEP. Conditions: [catalyst]=4.0”10À5 m, [buffer]=10–20 mm, 258C.
Figure 1 reports the activity, against the pH value, of the
functional Au-PEP nanoparticles in the hydrolysis of DNPB,
as k2,app, the apparent second-order rate constant of the
Angew. Chem. Int. Ed. 2007, 46, 400 –404
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
401

Communications
protonated imidazole group acts as a general acid to form a
residue of 1 anchored to the gold nanocluster, this intermedi-
ate is very likely to be the acylated imidazole group. Strikingly
enough, the formation of this intermediate is not observed
either with DNPB or Z-Gly-PNP, substrates that are less
lipophilic (Z-Gly-PNP) or more reactive (DNPB) than Z-
Leu-PNP. Furthermore, we had previously established that no
intermediate was formed in the use of Au-2 nanoparticles
with Z-Leu-PNP as a substrate. Thus, the presence of both a
lipophilic substrate (Z-Leu-PNP) and a catalytic site (His)
buried inside the monolayer covering the nanoparticle
apparently leads to a change in the rate-determining step of
the hydrolytic process.
In order to get further insight into this intriguing behavior
we have carefully analyzed the rate of formation of the
intermediate and the rate of its hydrolysis working under pre-
steady-state conditions (that is, with a catalyst concentration
comparable to that of the substrate)[23] and the results
obtained are reported in Figure 2. For Z-Leu-PNP the
analysis reveals the following: a) the first step does not
strong hydrogen bond to the developing negative charge in
the tetrahedral intermediate.[2] The larger catalytic efficiency
of Au-PEP in comparison with Au-2 arises from the stronger
acidity of the protonated imidazole group (the pKa value of
the His of Au-PEP is 0.9 units lower than that of Au-2), as
described by the Brønsted equation, lgk2,app = AÀapKa,
where A is a constant and a is the Brønsted coefficient for
general-acid catalysis. The observed difference in pKa values
between the His residues in the two systems is explained by
differential interactions with other charged residues. While in
Au-2 the His residue is close to the carboxylate anion which
increases the pKa value of the imidazolium group, the three
cationic Lys and Arg residues of Au-PEP give rise to a
decrease in the His pKa value due to electrostatic repulsion, in
spite of the presence of the C-terminal carboxylate anion.
The His residue and the C-terminal carboxylate group are
far apart in the sequence but the suggested cooperativity
requires proximity in space. It is unlikely that the folding of
peptide 1 in the monolayer covering the nanoparticle is such
as to bring the carboxylate group of Arg into close proximity
with the His residue. Accordingly, the most probable situation
is that in which the carboxylate and imidazolium ions belong
to two different, although spatially close, peptides. The
suggestion that a carboxylate and an imidazolium ion are
very close in the monolayer is also supported by the fact that
the apparent pKa value of the imidazolium ion is 1.1 units
higher than that observed for S-acetylated 1. In this peptide
the protonated side chain of Lys completely shields the
carboxylate effect on the pKa value of the imidazolium ion
which is, consequently, very close to the standard value
reported for His. Thus, the experimental evidence indicates
that the confinement of peptide 1 on the monolayer covering
the surface of the gold nanoparticles significantly alters the
properties of key functional groups involved in the catalytic
process and triggers cooperative processes that are totally
absent in the monomeric system.
*
Figure 2. Dependence of the initial rates of intermediate formation ( )
&
*
and of its hydrolysis ( ) with Z-Leu-PNP and of hydrolysis ( ) with Z-
Gly-PNP upon substrate concentration. Conditions: [1]=1.3”10À5
(bound to Au-PEP), pH 7, 258C.
m
As expected, the presence of the catalytic residue with an
apparent pKa value of 9.9 increases also the efficiency of Au-
PEP in comparison to Au-2 at the highest pH values studied.
Thus, Au-PEP at pH 10 is almost 40-fold more active than Au-
2. At this high pH regime either the amine group of Lys or the
phenoxide moiety of Tyr may be responsible for catalysis.
When the cleavage of DNPB was performed at pH 9.5 ESI-
MS experiments indicated the formation of a butanoyl
derivative of 1. When treated with butylamine this derivative
was completely deacylated, which suggests that Tyr and not
Lys acted as a nucleophile.[21]
DNPB is an excellent substrate for studying the reactivity
profile in the complete pH interval of 3–10 but it binds only
weakly to the monolayer of Au-PEP. For this reason we have
also studied the lipophilic substrate Z-Leu-PNP. However, in
running kinetics studies with excess substrate to determine
the Michaelis–Menten-like profile we observed bimodal
kinetics: a first, fast burst of absorbance was followed by a
slower process.[22] This behavior is typical of systems in which
the catalyst is first transformed into an intermediate that is
slowly cleaved to turn over the catalyst again. Since at pH 7
the relevant catalytic species is the imidazole group of the His
follow the typical Michaelis–Menten saturation kinetics but
becomes faster as the substrate concentration increases, with
a clear discontinuity at [substrate] ꢀ 3 ” 10À5 m; b) the second
step, that is, the breakdown of the intermediate, which should
be independent of the substrate concentration, is also
substrate-dependent and its rate monotonically decreases as
the substrate concentration increases. Figure 2also shows the
behavior observed for Z-Gly-PNP, where there is no build-up
of an intermediate: the initial rate of the cleavage process
increases steadily as expected since, because of the poor
lipophilicity of the substrate, we are far from saturation of the
catalyst.
The data indicate that Au-PEP catalyzes with equal
efficiency the cleavage of Z-Leu-PNP and Z-Gly-PNP up to a
concentration of substrate of 3 ” 10À5 m. Above this concen-
tration the nucleophilic attack of the His imidazole group
becomes much more efficient for Z-Leu-PNP than for Z-Gly-
PNP. Paradoxically this does not result in an overall better
performance of the catalyst because the cleavage of the
intermediate becomes the rate-determining step and, under
steady-state conditions, the hydrolysis is faster for Z-Gly-PNP
than for Z-Leu-PNP. A possible explanation for this behavior
402
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 400 –404

Angewandte
Chemie
phenolates at a given pH value. Catalyst concentration when the
is that the binding of lipophilic Z-Leu-PNP alters the
hydration of the catalytic site placed on the interior of the
monolayer covering the gold core of Au-PEP.[19] A decrease in
the water content may increase the nucleophilicity of the
imidazole group by decreasing its solvation (thus increasing
the rate of its acylation) and, on the other hand, may decrease
the rate of hydrolysis (which depends on the concentration of
water at the catalytic site). The binding of the substrate may
also affect the structure of the catalytic site. Out of the three
substrates we have investigated, selectively only one, Z-Leu-
PNP, regulates the activity of the catalyst.
The present results indicate that several copies of a
peptide bound to the surface of a gold nanocluster may lead to
the formation of a functional nanoparticle with enzyme-like
structure and properties. This is the first example of a protein-
like system with considerable complexity yet self-assembled.
Not only it is a good esterolytic catalyst but it is also capable
of regulation of its activity. Thus, the anchoring of a functional
peptide to the surface of a gold nanocluster results in a) the
modulation of the properties of functional groups present on
the side arms of the constituent amino acids, b) the induction
of cooperativity between different groups at the catalytic site
(an imidazolium and a carboxylate ion), and c) the creation of
an environment different from the bulk solution and more
similar to that found in the catalytic site of an enzyme (by
depleting water molecules).[24] None of these properties is
present in the monomeric peptide 1, although all of the
functional groups are obviously present in the oligomer.
These are novel and striking features of these systems that
compound with their multivalent nature[25,26] which has been
shown already to lead to exceptionally high binding constants
with selected substrates.[27]
nanoparticles are used as catalysts refers to the peptide component
present on the monolayer. Typical catalyst concentrations were 1.3 ”
10À5 m for reactions run with excess substrate and 4.0 ” 10À5 m for
reactions run with excess catalyst. Under pre-steady-state conditions
the substrate concentration was varied between 1.0 ” 10À5 and 4.0 ”
10À5 m. The buffer concentration was 10–20 mm. The buffers used
were N-cyclohexyl-3-aminopropanesulfonic acid (CAPS; pH 10.00),
2-(cyclohexylamino)ethanesulfonic acid (CHES; pH 9.00), 4-(2-
hydroxyethyl)piperazine-4-(3-propanesulfonic
acid)
(EPPS;
pH 8.00), 2-[4-(2-hydroxyethyl)-1-piperazinyl]ethanesulfonic acid
(HEPES; pH 7.00 and 6.50), b-morpholinoethanesulfonic acid
(MES; pH 6.00 and 5.5), acetic acid/sodium acetate (pH 4.00), and
citric acid/sodium citrate (pH 3.00).
Received: June 28, 2006
Revised: October 4, 2006
Published online: November 28, 2006
Keywords: artificial proteins · gold · hydrolysis · nanoparticles ·
.
peptides
[1] R. Davydov, R. Perera, S. X. Jin, T. C. Yang, T. A. Bryson, M.
Sono, J. H. Dawson, B. M. Hoffman, J. Am. Chem. Soc. 2005,
127, 1403 – 1413.
[2] P. Pengo, S. Polizzi, L. Pasquato, P. Scrimin, J. Am. Chem. Soc.
2005, 127, 1616 – 1617.
[3] F. Manea, F. B. Houillon, L. Pasquato, P. Scrimin, Angew. Chem.
2004, 116, 6291 – 6295; Angew. Chem. Int. Ed. 2004, 43, 6165 –
6169.
[4] L. Pasquato, F. Rancan, P. Scrimin, F. Mancin, C. Frigeri, Chem.
Commun. 2000, 2253 – 2254.
[5] Reviews: a) M.-C. Daniel, D. Astruc, Chem. Rev. 2004, 104, 293 –
346; b) R. Shenhar, V. M. Rotello, Acc. Chem. Res. 2003, 36,
549 – 561; c) A. C. Templeton, W. P. Wuelfing, R. W. Murray,
Acc. Chem. Res. 2000, 33, 27 – 36; d) A. Badia, R. B. Lennox, L.
Reven, Acc. Chem. Res. 2000, 33, 475 – 481.
[6] M. Mammen, S. K. Choi, G. M. Whitesides, Angew. Chem. 1998,
110, 2908 – 2953; Angew. Chem. Int. Ed. 1998, 37, 2755 – 2794.
[7] Z. Wang, R. Levy, D. G. Fernig, M. Brust, Bioconjugate Chem.
2005, 16, 497 – 500.
[8] N. Higashi, J. Kawahara, M. Niwa, J. Colloid Interface Sci. 2005,
288, 83 – 87.
[9] M. L. Becker, E. E. Remsen, D. Pan, K. L. Wooley, Bioconjugate
Chem. 2004, 15, 699 – 709.
[10] J. Fan, S. W. Chen, Y. Gao, Colloids Surf. B 2003, 28, 199 – 207.
[11] R. Levy, N. T. K. Thanh, R. C. Doty, I. Hussain, R. J. Nichols,
D. J. Schiffrin, M. Brust, D. G. Fernig, J. Am. Chem. Soc. 2004,
126, 10076 – 10084.
Experimental Section
Synthesis of 1: The peptide sequence of 1 was synthesized in the solid
phase by using standard 9-fluorenylmethoxycarbonyl (Fmoc), tert-
butoxycarbonyl (Boc), and tBu protecting groups. Activation of the
carboxylate was performed with 2-(1H-benzotriazol-1-yl)-1,1,3,3-
tetramethyluronium tetrafluoroborate (TBTU). The hydrocarbon
spacer terminating with the S-acetylated thiolic function was intro-
duced in the last step just before removal from the resin with acidic
treatment (95% trifluoroacetic acid (TFA)). The crude S-acetylated 1
was purified by HPLC (HICHROM 10C8 column), and the proper
fractions were analyzed by MALDI-TOF mass spectrometry (m/z
calcd for [M+]: 1588.8781; found: 1588.996). Deacetylation was neatly
performed by treatment with NH2NH2 in methanol.
[12] P. Pengo, Q. B. Broxterman, B. Kaptein, L. Pasquato, P. Scrimin,
Langmuir 2003, 19, 2521 – 2524.
Synthesis of Au-PEP: HS-C8-TEG-functionalized nanoparti-
cles[15] (15 mg) were dissolved in N2-flushed methanol (15 mL) in a
jacketed reactor kept at 288C. Peptide 1 (9 mg) was added to the
solution, and the reaction mixture was kept under an argon
atmosphere with stirring for 72h. The methanol was then partly
evaporated, and the mixture was passed through a Sephadex LH-60
column with elution with methanol. Evaporation of the proper
fractions afforded Au-PEP (22 mg). The exchange did not signifi-
cantly alter the size of the original HS-C8-TEG nanoparticles ((3.4 Æ
0.5) nm).
Kinetic experiments: Kinetics were studied under first-order
conditions at (25 Æ 0.2)8C and followed by monitoring the change of
absorbance at 400 nm due to the release of 2,4-dintrophenolate or 4-
nitrophenolate. Whenever necessary, absorbance was converted into
concentration by using the molar extinction coefficients of the
[13] L. Fabris, S. Antonello, L. Armelao, R. L. Donkers, F. Polo, C.
Toniolo, F. Maran, J. Am. Chem. Soc. 2006, 128, 326 – 336.
[14] The sequences of the four peptides not discussed in the text were
as follows: Leu-Gly-Tyr-Ala-Ala-Lys-Phe-Arg-Gly-His-Gly-
Arg-OH, Ala-Gly-Ala-Gly-Gly-Lys-Phe-His-Ala-Arg-Ala-Arg-
OH, Leu-Gly-Tyr-His-Ala-Lys-Phe-Arg-Ala-Gly-Gly-Arg-OH,
and Leu-His-Tyr-Lys-Ala-Arg-Phe-Arg-Ala-Gly-Gly-Gly-OH.
[15] P. Pengo, S. Polizzi, M. Battagliarin, L. Pasquato, P. Scrimin, J.
Mater. Chem. 2003, 13, 2471 – 2478.
[16] R. Levy, Z. X. Wang, L. Duchesne, R. C. Doty, A. I. Cooper, M.
Brust, D. G. Fernig, ChemBioChem 2006, 7, 592– 594.
[17] L. K. Andersson, G. T. Dolphin, L. Baltzer, ChemBioChem
2002, 3, 741 – 751.
[18] L. K. Andersson, M. Caspersson, L. Baltzer, Chem. Eur. J. 2002,
8, 3687 – 3697.
Angew. Chem. Int. Ed. 2007, 46, 400 –404
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
403

Communications
[19] M. Lucarini, P. Franchi, G. F. Pedulli, P. Pengo, P. Scrimin, L.
[23] A. Fersht, Structure and Mechanism in Protein Science, Freeman,
New York, 1999.
Pasquato, J. Am. Chem. Soc. 2004, 126, 9326 – 9329.
[20] k2,app = (kobsÀk0)/[catalyst], where kobs is the experimentally
observed first-order rate constant, k0 is the rate constant for
the spontaneous hydrolysis of the substrate, and [catalyst] is the
nominal concentration of peptide 1 on the monolayer. Since we
are far from full binding of the substrate to the monolayer, we
may consider the catalytic process like a bimolecular one.
[21] Under these experimental conditions, there is no evidence of
deacylation of Tyr, either for the nanoparticle or the peptide.
[22] The DA value (and hence the p-nitrophenolate released) during
the burst increases along with the substrate concentration, but it
is less than the catalyst concentration because full saturation of
the catalyst was not obtained, even at the highest substrate
concentration used.
[24] Recent data from our laboratory indicate that not only the
substrate but also the size of the nanoparticles may alter the
hydrophobicity of the monolayer covering a gold nanoparticle;
see: V. Lucarini, P. Franchi, G. F. Pedulli, C. Gentilini, S. Polizzi,
P. Pengo, P. Scrimin, L. Pasquato, J. Am. Chem. Soc. 2005, 127,
16384 – 16385.
[25] L. Pasquato, P. Pengo, P. Scrimin, J. Mater. Chem. 2004, 14, 3481 –
3487.
[26] L. Pasquato, P. Pengo, P. Scrimin, Supramol. Chem. 2005, 17,
163 – 171.
[27] G. Fantuzzi, P. Pengo, R. Gomila, P. Ballester, C. A. Hunter, L.
Pasquato, P. Scrimin, Chem. Commun. 2003, 1004 – 1005.
404
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 400 –404