Design, synthesis and characterisation of a fluorescently labelled CyPLOS ionophore

Article abstract of DOI:10.1002/chem.201000611

A novel fluorescently labelled synthetic ionophore, based on a cyclic phosphate-linked disaccharide (CyPLOS) backbone and decorated with four tetraethylene glycol tails carrying dansyl units, has been synthesised in 12 steps in 26 % overall yield. The key

Full text of DOI:10.1002/chem.201000611

FULL PAPER
DOI: 10.1002/chem.201000611
Design, Synthesis and Characterisation of a Fluorescently
Labelled CyPLOS Ionophore**
Cinzia Coppola,[a] Antonio Paciello,[a] Gaetano Mangiapia,[b, c] Sabina Licen,[d]
Mariangela Boccalon,[d] Lorenzo De Napoli,[a] Luigi Paduano,[b, c] Paolo Tecilla,[d] and
Daniela Montesarchio*[a]
Abstract:
A
novel fluorescently la-
activated carboxylic acid derivatives, or
by a direct coupling with a terminal
alkyne in a CuI-promoted 1,3-dipolar
cycloaddition. Tagging of the mono-
meric building block with dansyl resi-
dues allowed us to prepare a fluores-
cent, amphiphilic macrocycle, which
was investigated for its propensity to
self-aggregate in CDCl3—studied by
means of concentration-dependent
31P NMR spectroscopy experiments—
and in aqueous solution, in which com-
bined dynamic light scattering (DLS)
and small-angle neutron scattering
(SANS) measurements provided a de-
tailed physico-chemical analysis of the
self-assembled systems, mainly organ-
ised in the form of large vesicles. Its
ion-transport properties through phos-
pholipid bilayers, determined by HPTS
fluorescence assays, showed this com-
pound to be more active than the pre-
viously synthesised CyPLOS congeners.
belled synthetic ionophore, based on a
cyclic phosphate-linked disaccharide
(CyPLOS) backbone and decorated
with four tetraethylene glycol tails car-
rying dansyl units, has been synthesised
in 12 steps in 26% overall yield. The
key intermediate in the synthetic strat-
egy is a novel glucoside building block,
serving through its 2- and 3-hydroxy
groups as the anchor point for flexible
tetraethylene glycol tentacles with re-
active azido moieties at their ends. To
test the versatility of this glucoside
scaffold, it was preliminarily function-
alised with a set of diverse probes—as
fluorescent, redox-active or hydropho-
bic tags—either by reduction of the
azides followed by condensation with
Solvent-dependent
fluorescence
changes for the labelled ionophore in
liposome suspension established that
the dansyl moieties are dispersed in en-
vironments with polarity intermediate
between those of CH2Cl2 and propan-
2-ol, suggesting that the CyPLOS ten-
tacles infiltrate the mid-polar region of
the membranes.
Keywords: amphiphilic
oligosac-
charides · fluorescent probes · ion
transport · macrocyclic compounds ·
self-assembly
Introduction
[a] Dr. C. Coppola, Dr. A. Paciello, Prof. L. De Napoli,
Prof. D. Montesarchio
A wide number of diverse synthetic ionophores, functional
analogues of ion channels or carriers, but typically display-
ing poor or no structural similarity with their natural, poly-
peptide-based counterparts, have been described in the
recent literature.[1] Notwithstanding the large amounts of
data collected on ion transport across cell membranes, pro-
moted either by natural or by artificial agents, the overall
picture of the structure–activity relationships is still not
completely clear, with many different mechanisms of action
being operative and with high ionophore efficiencies being
exhibited both by very complex proteic architectures[2] and
by structurally simple, small organic molecules.[3]
Department of Organic Chemistry and Biochemistry
University “Federico II” of Napoli
Via Cintia, 4, 80126 Napoli (Italy)
Fax : (+39)081-674393
E-mail: daniela.montesarchio@unina.it
[b] Dr. G. Mangiapia, Prof. L. Paduano
Department of Chemistry “Paolo Corradini”
University “Federico II” of Napoli
Via Cintia, 4, 80126 Napoli (Italy)
[c] Dr. G. Mangiapia, Prof. L. Paduano
CSGI, Consorzio univ. per lo sviluppo
dei Sistemi a Grande Interfase (Italy)
[d] Dr. S. Licen, Dr. M. Boccalon, Prof. P. Tecilla
Department of Chemical Sciences
University of Trieste, Via L. Giorgieri
34127 Trieste (Italy)
Suitably derivatised carbohydrates are ideal platforms for
the construction of effective artificial receptors in molecular
or ion recognition, because they are easily available starting
materials, displaying multiple functional groups in well-de-
[**] CyPLOS=Cyclic phosphate-linked oligosaccharide.
Chem. Eur. J. 2010, 16, 13757 – 13772
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
13757

fined stereochemical presentations. These valuable synthetic
scaffolds offer further advantages from the perspective of in
vivo applications, playing pivotal roles in many cellular pro-
cesses and showing favourable pharmacokinetic properties.
In addition, when conjugated to valuable metabolites, they
can confer resistance to enzymatic degradation and cell
uptake enhancement, eliciting specific molecular recognition
processes, mediated by membrane proteins (lectins).[4] For
these and other reasons the development of glycoconjugates
or hybrid backbones incorporating saccharide moieties is a
research field in continuous development.[5]
Aiming at the preparation of novel macrocycles useful for
specific ion recognition, we recently described the synthesis
and conformational properties of a novel class of cyclic oli-
gosaccharide analogues, 4,6-linked through stable phospho-
diester bonds (1–3), that we named CyPLOSs (cyclic phos-
phate-linked oligosaccharide analogues).[6]
Studies on the ionophoric activities of amphiphilic Cy-
PLOSs[8] showed 5 to be the most effective compound in the
investigated series, capable of discharging the pH gradient
across a liposomial membrane completely in less than
20 min at 2% ionophore concentration. This property was
strictly correlated to the presence of TEG chains, with 5
being much more active than 6 and the tetra-alkylated de-
rivative 4 being almost completely inactive. Analysis of the
ionophoric properties of 5 in comparison with those of its
linear and fully protected cyclic congeners led us to the con-
clusion that both the four TEG tentacles and the fairly rigid
anionic macrocycle are structural motifs essential for activi-
ty. When investigated for selectivity in ion transport,
CyPLOS 5 did not show any significantly different behav-
iour with variation in the nature of the cation in solution;
on the other hand, when tested with different anions, high
transport activities were found with halogens (except fluo-
ride), nitrate and perchlorate, whereas on the contrary, fluo-
ride, acetate, glutamate and sulfate were not effectively
transported. The lack of selectivity towards lipophilic anions
was attributed to the formation of a poorly structured active
species with little specificity towards the transported ions.
With the final objective of preparing novel and more effi-
cient analogues of 5, containing reporter groups that should
in principle allow better insight into their mechanism of
action, we first set out to investigate the synthesis of a versa-
tile glucoside intermediate that would be susceptible to fur-
ther derivatisation and widely applicable for the construc-
tion of multi-labelled architectures. For this purpose we de-
signed the glucoside scaffold 8 (Scheme 1, below), bearing
at its 2- and 3-OH positions flexible tetraethylene glycol
chains connected at their ends to further reactive azido
groups. This novel glucoside has been exploited to generate
a small library of model labelled monosaccharides conjugat-
ed with different dyes (10a–
Amphiphilic macrocycles with remarkable propensities to
aggregate were then obtained by attaching long tentacles
with different hydrophobicities, particularly n-undecyl or
benzyl-capped tetraethylene glycol (TEG) chains, to the
CyPLOS backbones (4–6).[7]
10c, Scheme 3, below, and 12,
Scheme 4, below), to demon-
strate its effective applicability.
The building blocks 16 and 17
(Scheme 5, below) were then
selected to generate the multi-
labelled
macrocycle
21
(Scheme 6, below), very similar
to 5 except that the terminal
benzyl groups are replaced by
the fluorescent 5-N,N-dime-
thylaminonaphthalenesulfonyl
(dansyl) residues. The use of
fluorescent chemosensors as ex-
tremely powerful analytical and
diagnostic tools, particularly for
supramolecular applications, is
widely documented in the
recent literature.[9] Of the most
commonly used fluorescent
dyes, the dansyl group is consid-
ered an excellent probe for the
13758
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13757 – 13772

A Fluorescently Labelled CyPLOS Ionophore
FULL PAPER
study of self-assembly or recognition processes, being dra-
matically sensitive to external stimuli or different environ-
ments.[10] Typically, it exhibits large solvent-dependent fluo-
rescence shifts and has been inserted in several synthetic ion
transporters to investigate their mechanism of action, also
allowing easy monitoring of their incorporation in cells.[11]
As a further advantage, it is chemically stable and can be
easily conjugated to systems containing primary amines or
hydroxy functions by direct coupling with its commercially
available chloride derivative. Herein, we describe the syn-
thesis and properties of the novel dansyl-labelled CyPLOS
macrocycle 21; a variety of analytical techniques were also
exploited to study its self-assembly, partition, localisation
within liposomes and its ionophoric activity, also in compari-
son with the previously synthesised congeners.
tune to its compatibility with a broad range of functional
groups and different solvent systems, including water. In ad-
dition, the newly generated triazole ring is stable to hydro-
lytic and enzymatic cleavage and is virtually inert to a varie-
ty of conditions. In all cases, the obtained functionalised glu-
cosides provide novel amphiphilic scaffolds useful in the
construction of carbohydrate-based receptors with peculiar
spectroscopic properties, allowing easy detection of interac-
tion events in molecular recognition.
The synthetic route to 8 first required the elaboration of
tetraethylene glycol (TEG),[14] a flexible bifunctional linker
widely used for the covalent attachment of dyes to biomole-
cules,[15] about 17 ꢁ long in its fully extended conforma-
tion.[16] In our synthetic scheme, tetraethylene glycol was
subjected to the following straightforward and high-yielding
reactions (depicted in Scheme 2): monotosylation, azidation
Results and Discussion
Synthesis of the versatile glucoside 8 and its conversion into
the labelled derivatives 10a–10c and 12: The glucoside scaf-
fold 8 (Scheme 1) is a highly versatile building block, char-
Scheme 2. Synthesis of the activated TEG-azido derivative III.
and activation of the free OH group as a mesyl ester. Once
the activated monoazido-TEG III had been obtained[17] it
was employed in a coupling with the 4,6-benzylidene-pro-
tected sugar 7 (Scheme 1) in a classical Williamson reaction,
carried out in DMF in the presence of an excess of NaH,
which allowed the persubstitution of the secondary 2- and 3-
OH groups, leading to 8 in 90% yields.
Scheme 1. General scheme for the synthesis and applications of the glu-
coside 8.
To demonstrate the feasibility of this synthetic strategy,
glucoside 8 was functionalised with a selected panel of di-
verse labels, either by classical Staudinger ligation reactions
or by the click chemistry approach. In the first synthetic in-
vestigation we prepared the derivatives 10a–10c, each incor-
porating a fluorescent tag in the form of the 5-N,N-dimeth-
acterised by two tetraethylene glycol appendages with azido
residues at their ends. These groups, generally chemically
stable, can be activated under appropriate conditions and
coupled with a range of diverse labels, including fluorescent
and/or electrochemically active probes. The derivative 8 can
undergo high-yielding condensations through the exploita-
tion of different but equally reliable synthetic schemes. As
summarised in Scheme 1, two approaches have been adopt-
ed here. The first involves a two-step, one-pot reaction, re-
quiring reduction of the azido groups to primary amines, fol-
lowed by coupling with suitably activated derivatives of car-
boxylic acids, in a classical Staudinger ligation approach.[12]
In the second strategy, glucoside 8 is exploited in a direct
coupling with a terminal alkyne by means of the CuI-assisted
azide–alkyne 1,3-dipolar cycloaddition (the so called “click-
chemistry” approach).[13] This reaction owes its recent for-
ACHUTNGRENyNUG lACTHUGNRTENaUNG minonaphthalenesulfonyl (dansyl) residue, an aromatic
a-amino acid (i.e., 9-fluorenylmethoxycarbonyl (Fmoc)-pro-
tected tryptophan) and an electrochemically active probe
(ferrocenecarboxylic acid), respectively (Scheme 3).
In a typical experiment, glucoside 8 was first treated with
triphenylphosphine in water, followed by coupling of the
freshly generated primary diamine 9 with the reactive tag in
CH2Cl2 at room temperature. No condensing reagent was re-
quired for the coupling with dansyl chloride; in the cases of
the derivatives 10b and 10c, dicyclohexylcarbodiimide
(DCC) was adopted as the activating agent for the carboxyl-
ic acid functions. The reactions, monitored by TLC and
Chem. Eur. J. 2010, 16, 13757 – 13772
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13759

D. Montesarchio et al.
even with variation of the solvent and temperature condi-
tions, but good yields of the target compound 12 were finally
obtained with addition of CuI in stoichiometric amounts
and use of DMF as the solvent in the presence of diisopro-
pylethylamine (DIPEA).
Successive synthetic elaboration of the labelled glucosides
10a–10c and 12 first involved benzylidene removal, liberat-
ing the 6- and 4-OH groups, which exhibited very different
reactivities. The obtained compounds might be of interest as
such, being amenable to a variety of further manipulations,
including immobilisation on a solid support either through
direct loading on carboxyl-functionalised resins or through
exploitation of derivatisation at C6 through a succinate
linker for coupling with an amino-functionalised matrix.[21]
Studies directed towards exploitation of these glucoside-
functionalised solid supports to generate differently labelled
glycoconjugates are in progress. In view of the widely docu-
mented use of the dansyl group as an effective probe for
mechanistic studies,[11] glucoside 10a was selected here as a
suitable building block to provide the desired environment-
sensitive CyPLOS macrocycle.
Synthesis of the dansyl-labelled macrocycle 21: From the
starting glucoside 10a, two building blocks, functionalised as
a phosphoramidite (16, Scheme 5) and as a 2-chlorophenyl-
phosphate derivative, 17, were prepared, ready for coupling
and cyclisation to finally give macrocycle 21.
Scheme 3. Synthesis of the labelled glucoside derivatives 10a–10c.
quenched after the disappearance of the monofunctionalised
derivatives, in all cases gave the target compounds in good
yields (80–85%).
In the first step, compound 10a was converted into the N-
methyl derivative 13, to mask the NH groups permanently.
Benzylidene removal by standard acidic treatment with di-
luted trifluoroacetic acid (TFA) in CH2Cl2/H2O gave the
4,6-deprotected derivative 14 in high yields. Treatment with
DMTCl (DMT=4,4’-dimethoxytriphenylmethyl) in pyridine
gave almost quantitative yields of the DMT-protected com-
pound 15, which was then phosphitylated to give the phos-
phoramidite derivative 16 in 90% yield. The synthesis of the
derivative 17 was accomplished in 86% yields over two
steps, first involving phosphorylation of the 4-OH group
with 2-chlorophenyldichlorophosphate in the presence of
triethylamine (TEA) and triazole in pyridine, followed by
on-column DMT removal by elution of the silica gel column
with CHCl3 with added TFA (0.01%). The coupling of
building blocks 16 and 17 was achieved with the aid of 4,5-
dicyanoimidazole (DCI) in CH3CN (0.25m) as the condens-
ing agent, followed by oxidation with tert-BuOOH (5.5m in
decane). The linear dimer 18 (Scheme 6) was isolated after
chromatography, involving on-column DMT removal, with
72% yield for the three steps.
To test the effective potential of 8 in click chemistry reac-
tions, it was coupled with a suitable terminal alkyne. Here
we used substrate 11 (Scheme 4), previously prepared ad
hoc by treatment of propargylamine with cholic acid in
CH2Cl2, in the presence of DCC as the condensing agent.
Cholic acid was chosen as a model hydrophobic residue, of
interest for the non-covalent immobilisation of biomolecules
on hydrophobic surfaces[18] and for the development of
novel drug delivery systems,[19] as well as of novel chemosen-
sors for sensitive ion detection.[20] In our hands the reaction
was never successful in catalytic mode either with CuI or
with the CuSO4/ascorbate system (0.01 up to 0.2 equiv.),
The cyclisation of 18 was then carried out by a phospho-
triester methodology, with use of 1-mesitylensulfonyl-3-
nitro-1,2,4-triazole (MSNT) as the condensing agent under
high-dilution conditions (cꢀ10ꢁ3 m), affording the fully pro-
tected 19 in 75% yields. Final deprotection was obtained by
treatment first with piperidine, to promote b-elimination of
the 2-cyanoethyl protecting group, leading to 20, and subse-
quently with a saturated LiOH solution in dioxane/water
(1:5, v/v) at 508C to cleave the 2-chlorophenyl group (95%
Scheme 4. Synthesis of the labelled glucoside derivative 12.
13760
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13757 – 13772

A Fluorescently Labelled CyPLOS Ionophore
FULL PAPER
31P NMR spectroscopy studies on the aggregation behav-
iour of the dansyl-labelled macrocycle 21 in CDCl3: In anal-
ogy with the studies previously carried out on the macrocy-
cles 4–6,[7] compound 21 was analysed by NMR spectroscopy
experiments for its tendency to form self-aggregated species
in organic solvents. CDCl3, roughly mimicking the polarity
of the internal milieu of a lipid bilayer, was the deuterated
solvent of choice, offering preliminary, but still useful indica-
tions about the behaviour of this amphiphilic macrocycle in
1
bulk membranes. Inspection of the H and 31P NMR spectra
of 21 showed dramatic line broadening and chemical shift
anisotropy of the resonances, diagnostic of slow equilibria
on the NMR spectroscopy timescale, which is consistent
with the formation of large aggregates in CDCl3.[22] The
1H NMR spectra of 21 did not show any detectable simplifi-
cation of the spectral lines either with increasing tempera-
ture or with increasing dilution of the sample. Concentra-
tion-dependent 31P NMR spectroscopic analysis was more
informative about the solution behaviour of 21. At 3.0 mm
concentration (CDCl3, 161.98 MHz, 298 K) a very broad
signal, centred at about d=0 ppm and dispersed over a
15 ppm region, was apparent in its 31P NMR spectrum. Two
main bands emerged when the sample was analysed at 1.5
and 0.75 mm, and these then collapsed into a unique large
signal centred at about d=2.0 ppm when the sample was
further diluted to 0.35 mm concentration. At 0.18 mm con-
centration a well-resolved signal was registered, indicative
of the full magnetic equivalence of the two phosphorus
atoms in the macrocycle under these conditions. It can be
deduced that at this concentration 21 predominantly adopts
a
symmetric, presumably monomeric and well-solvated
form, so that different conformations exist in rapid dynamic
equilibria with respect to the NMR spectroscopy timescale.
Taken together, these data allowed us to evaluate the criti-
cal aggregation concentration (cac) value of 21 over the con-
centration range of 0.30–0.20 mm (Figure 1). Variable-tem-
perature (VT) 31P NMR spectra for 21 showed no tendency
towards even partial simplification of the peaks with in-
creasing temperature up to 338 K, thus indicating the pres-
ence of thermally stable aggregates (data not shown). Com-
pound 21 showed a stronger propensity than the previously
described benzyl-capped derivative 5 towards self-aggrega-
tion in CDCl3, as can be inferred by analysis of their cac
values (8–6 mm[7] vs. 0.3–0.2 mm). In part, this can be attrib-
uted to possible interactions of the dansyl groups at the ex-
tremities of the TEG tails, plausibly contributing to provide
extra stabilisation to self-aggregated species through p-
stacking interactions.
Scheme 5. Synthesis of the labelled building blocks 16 and 17; Py=pyri-
dine.
yield for the two steps). Compound 21 was prepared by the
described procedure in four steps, with 51% yield on aver-
age from the building blocks 16 and 17, each in turn ob-
tained from the glucoside 10a in four steps with 81–85%
yield.
All the intermediates and the final compounds were puri-
fied by column chromatography and characterised by 1H
and 13C (and 31P, where present) NMR spectrometry and MS
analysis
Characterisation of the aggregates formed by CyPLOS 21 in
a pseudo-physiological solution: Self-assembly of amphiphil-
ic ion transporters in physiological solutions, that is, in the
aqueous environment surrounding cell membranes, can be
crucial for their interactions with the phospholipid bilayers
and possibly also for their activity. In principle, nano-aggre-
gation allows these systems to be stabilised in the extracellu-
lar media, vehiculated and more easily transferred into the
Chem. Eur. J. 2010, 16, 13757 – 13772
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13761

D. Montesarchio et al.
The slow diffusive mode found,
which corresponds to Rh =
(670ꢂ20) nm, is compatible
with the presence of large ag-
gregates such as vesicles. On
the other hand, the fast diffu-
sive mode is consistent with the
presence of smaller aggregates,
presumably micelles. The latter
have a hydrodynamic radius of
44 nm, which suggests the for-
mation of elongated micelles,
probably rod-like aggregates. In
fact, the linear length of the
molecule, with the assumption
of an all-trans conformation, is
estimated to be around 6 nm,
so a hypothetical spherical mi-
celle would have the same
radius. On the other hand, the
Rh value found for the slow dif-
fusive mode is more than one
order of magnitude above the
usual size of micellar aggre-
gates, and hence comparable
with the size of vesicles. This
value is also much higher than
the hydrodynamic radii found
for the vesicles formed by other
CyPLOS molecules. Vesicles
Scheme 6. Synthesis of the macrocycle 21. i) a) DCI in CH3CN (0.25m), RT, 2 h; b) tert-BuOOH in decane
(5.5m), RT, 30 min (72% for the two steps); ii) DMAP, MSNT, Py, RT, 12 h (75%); iii) piperidine, 708C, 12 h
(96%); iv) saturated aq. LiOH/dioxane (5:1, v/v), 508C, 12 h (99%).
phospholipid bilayers. From this perspective, we analysed
the aggregative behaviour of CyPLOS 21 in a pseudo-phys-
iological solution by means of dynamic light scattering
(DLS) and small-angle neutron scattering (SANS) measure-
ments.
formed by 5, the largest aggregates in the series 4–6, show a
value of Rh =(330ꢂ30) nm.[23c] The replacement of benzyl
groups with the larger dansyl groups at the ends of the TEG
tentacles of CyPLOS backbones therefore seem to affect
their overall aggregation behaviour dramatically.
Figure 2 shows an example of the diffusion coefficient dis-
tribution function obtained at 908 from inverse Laplace
transformation of the intensity correlation function g(2)(t)ꢁ1
for CyPLOS 21 dissolved in a pseudophysiological solu-
The results obtained by DLS were confirmed by small-
angle neutron scattering. Figure 3 reports the scattering
cross sections obtained for CyPLOS 21 as a function of the
scattering vector modulus q. The cross section profile re-
veals the presence of large scattering aggregates, typically
large vesicles. In particular, the power law dS/dW/qꢁ2 is
found in the low q region; this trend is typical of bidimen-
sional objects, such as double layers.[23a] A theoretical ex-
pression for dS/dW for a collection of such scattering objects
exists[23a] and has been fitted to the experimental data to
provide the thickness t of the double layer. Results of the
fitting procedure are reported in Table 1, in which a thick-
ness t of (13ꢂ0.9) nm was obtained. This value is close to
double that of the linear extension of the molecule
(ꢀ6 nm), suggesting a possible conformation not very dis-
similar from the all-trans disposition and not very pro-
nounced interdigitation between molecules belonging to the
two opposite sides of the double layer.
ACHTUNGTRENNUNG
tion.[23a] Inspection of the figure shows a bimodal distribu-
tion with well-separated modes. The slow mode has a great-
er amplitude than the faster mode, as would be expected for
two objects of different sizes diffusing in solution. Provided
that the solution is sufficiently dilute, the Stokes–Einstein
equation, which rigorously holds at infinite dilution for spe-
cies diffusing in a continuum medium, may legitimately be
used to evaluate the hydrodynamic radius (Rh) of the aggre-
gates [Eq. (1)]:[23b]
kT
6phD
ð1Þ
Rh ¼
in which k is the Boltzmann constant, T is the absolute tem-
perature, and h is the medium viscosity.
Diffusion coefficients, together with hydrodynamic radius
values obtained through Eq. (1), are collected in Table 1.
Finally, we would like to highlight the fact that SANS in-
vestigations have indeed revealed the presence of vesicles
but that it was not possible, unlike in the case of the DLS
measurements, to infer the existence of smaller aggregates
13762
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13757 – 13772

A Fluorescently Labelled CyPLOS Ionophore
FULL PAPER
Figure 2. Translational diffusion coefficient distribution function at 258C
obtained by means of DLS measurements for the binary aqueous system
containing CyPLOS 21 at the concentration of 2.00 mmolkgꢁ1
.
Table 1. Structural parameters obtained at 258C for the aggregates
formed by CyPLOS 21 at a concentration of 2.00 mmolkgꢁ1, through
DLS and the fitting of the model described in the Experimental section
(SANS data). The table reports the translational diffusion coefficients D
and hydrodynamic radii Rh for both the diffusive modes, together with
the average lamellar thickness t of vesicular aggregates, determined for
the slow diffusive mode.
D [cm2 sꢁ1
(3.6ꢂ0.1)ꢀ10ꢁ9
(5.5ꢂ0.3)ꢀ10ꢁ8
]
Rh [nm]
t [nm]
slow mode
fast mode
N
670ꢂ20
44ꢂ2
13.1ꢂ0.9
Figure 1. 31P NMR spectra (161.98 MHz, CDCl3, 298 K) of the macrocycle
21 at different concentrations (3.0, 1.5, 0.75, 0.35, 0.18, 0.09 mm).
(micelles). This, of course, does not imply the absolute ab-
sence of smaller aggregates, but can be ascribed to an over-
whelming predominance of vesicles over micelles in solu-
tion, thus rendering the contribution of the smaller aggre-
gates almost negligible. On the other hand, the scattering
depends on the square of the volume and consequently is
more relevant the larger the aggregates under investigation.
Figure 3. Scattering cross sections obtained at 258C by means of SANS
measurements for the binary aqueous system containing CyPLOS 21, at
a concentration of 2.00 mmolkgꢁ1. The curve obtained through the fitting
of the model described in the experimental part is also shown.
dylcholine (EYPC) and egg phosphaACTHNUTRGNEUNGtidylglycerol (EYPG)
Ionophoric activity of the dansyl-labelled macrocycle 21: In
analogy with the previous members of the CyPLOS family,
the ionophoric properties of the dansyl-labelled macrocycle
21 were investigated in large unilamellar vesicles (100 nm
diameter, prepared by extrusion) with a 95:5 egg phosphati-
lipid composition. The presence of fluorescent tags in 21 al-
lowed easier determination of the partition coefficient of
the ionophore in the membrane than in the case of com-
pound 5, in which the TEG chains terminate in benzyl
groups. The unbound ionophore was separated from the lip-
Chem. Eur. J. 2010, 16, 13757 – 13772
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13763

D. Montesarchio et al.
osome suspension with the aid of a gel-permeation column
and the liposome fraction, diluted in methanol to destroy
aggregated forms, was analysed by fluorescence spectrosco-
py. Comparison with a calibration curve obtained by meas-
uring the emission intensities of solutions of the ionophore
at known concentrations in methanol showed that about
40% of the CyPLOS derivative was partitioned into the
membrane.
with higher activity than compound 5, which had been the
most effective CyPLOS derivative so far tested. Fitting of
the data to a first-order rate equation gives the apparent
rate constants (kobs, sꢁ1) for the pH-discharge process, which
are 4.4ꢂ10ꢁ3 and 1.5ꢂ10ꢁ3 sꢁ1 for 21 and 5, respectively.
The replacement of the benzyl moieties with the dansyl
groups, therefore, roughly doubles the activity of the iono-
phore.
The ability of the ionophore to mediate a pH gradient col-
lapse across the membrane was measured with the aid of
the pH-sensitive fluorescent dye HPTS (HPTS=8-hydroxy-
pyrene-1,3,6-trisulfonic acid, pKa =7.2).[24] In this assay the
dye is entrapped in phospholipid vesicles buffered at pH 7.0
and, after addition of the ionophore, a pH gradient of
0.6 pH units is applied by external addition of NaOH. The
collapse of this transmembrane pH-gradient implies basifica-
tion of the liposome inner water pool, which is signalled by
an increase in the HPTS fluorescence emission. Figure 4a
shows the kinetic profiles obtained in the presence of deriv-
ative 21 and, for comparison, of CyPLOS 5 both at concen-
trations of 1%, expressed as the molar ratio of ionophore
with respect to total lipid concentration. These data show
the dansyl derivative to be a reasonably effective ionophore,
For comparison, the naturally occurring ionophore am-
photericin B (AmB), at the same concentration and under
the same experimental conditions, is characterised by a
k
obs =2.3ꢂ10ꢁ3 sꢁ1, which is comparable to the values mea-
sured for the CyPLOS derivatives.[8]
The dependence of the apparent rate constant (kobs, sꢁ1)
for the pH-discharge process from the concentration of ion-
ophore 21 is shown in Figure 4b. The activity increases line-
arly with the concentration, suggesting that the active spe-
cies responsible for the ionophoric process is monomeric.
This observation, which parallels the results obtained for
compound 5, is not in contradiction with the aggregation be-
haviour in low-polarity solvents described above. Indeed,
even if account is taken of the partition of the ionophore in
the small liposome membrane volume,[25] the concentration
used here in the kinetic experiments lies well below the cac
determined in CDCl3 (range 0.30–0.20 mm).
The HPTS assay described above can be modified to pro-
vide information on the cation/anion selectivity of the trans-
port process.[24] Indeed, the basification of the inner water
pool as a response to the pH gradient has to be balanced to
maintain the ionic strength constant. The basification may
originate either from H+ efflux or from OHꢁ influx and so
four overall transmembrane charge translocation transport
processes can be set up to balance the pH increase: H+/Na+
antiport, OHꢁ/Clꢁ antiport, H+/Clꢁ symport and Na+/OHꢁ
symport. In this modified version of the HTPS assay the lip-
osome suspension is prepared in a HEPES buffer (pH 7.0)
containing 100 mm MCl or NaX (in which M and X are the
cation and anion under investigation, respectively) and the
pH gradient is established with NaOH in the cation selectiv-
ity experiments or MOH in the anion selectivity experi-
ments. The observed HPTS fluorescence emission changes
therefore report indirectly on the effect of the cation/anion
present on the overall transport process. The results ob-
tained are reported in Figure 4c and d for the cation and
anion selectivity experiments, respectively. The activity is
not significantly affected with a change in the Group I alkali
metal cation present (Figure 4c). On the other hand, the
anions show a strong effect on the transport process (Fig-
ure 4d) and they group into two distinct families: the more
hydrophilic ones (fluoride, acetate, glutamate), which are
almost not transported, and the less hydrophilic ones (the
other halogens, nitrate and perchlorate) which are efficiently
transported, but all at a similar rate. This behaviour (indif-
ference to the cation present and anion-hydrophilicity-con-
trolled off/on behaviour in the transport of the anions)
strictly parallels that described for compound 5, which has
been interpreted in terms of a transport process dominated
Figure 4. a) Normalised fluorescence change in HPTS emission (FI, lex
460 nm, lem 510 nm) as a function of time after addition of the base
(50 mL of 0.5m NaOH) to 95:5 EYPC/EYPG LUVs (100 nm diameter)
loaded with HPTS (0.1 mm HPTS, 0.17 mm total lipid concentration,
25 mm HEPES, 100 mm NaCl, pH 7.0, total volume 3 mL, 258C), in the
presence of CyPLOS derivatives 21 and 5. The concentration of the iono-
phores (1%) is given in percent with respect to the total concentration of
lipids. The control trace was recorded in the absence of ionophore.
b) Dependence of the apparent rate constants (kobs, sꢁ1) for the pH-dis-
charge process on the % concentration of CyPLOS derivative 21.
c) Cation selectivities for ionophore 21 (0.2% concentration) by HPTS
assay (100 mm MCl, pH 7.0, base pulse by addition of 50 mL of 0.5m
MOH). d) Anion selectivities for ionophore 21 (0.2% concentration) by
HPTS assay (100 mm NaX, pH 7.0, base pulse by addition of 50 mL of
0.5m NaOH, Glu=glutamate).
13764
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13757 – 13772

A Fluorescently Labelled CyPLOS Ionophore
FULL PAPER
by OHꢁ/Xꢁ antiport or H+/Xꢁ symport in which the rate-
limiting factor switches from Xꢁ to H+ (or OHꢁ) transport
with decreasing anion hydration energy.[8] Highly hydrophil-
ic anions are not transported, but when the dehydration cost
decreases below that of acetate, the process is controlled by
H+ (or OHꢁ) transport, and all the anions with lower hydra-
tion energy are therefore transported at a similar rate. In
any case the absence of selectivity among the transported
anions suggests the formation of a poorly structured pore
and little or no interaction between the transported species
and the ionophore. However, control experiments with the
relatively large calcein dye indicate that the pore formed,
whatever its structure, is not large enough to allow the
transport of calcein and that the ionophore is not able to
cause lysis of the liposomes, at least up to the highest con-
centration tested (7%, in terms of ionophore/phospholipids
ratio).
The emission spectra of the dansyl dye exhibit large sol-
vent-dependent fluorescent shifts and this property might
provide valuable information on its position when inserted
in a phospholipid bilayer.[11] The fluorescence spectra of the
dansyl-labelled macrocycle 21 (1.7 mm) were recorded in sol-
vents of different polarities and the observed emissions
(lmax) were plotted against Reichardtꢃs ET polarity parame-
ter (Figure 5).[26] Fitting of the data gives a linear correlation
(y=1.1x+458.87, r2 =0.9807) over the range of the solvents
four polarity regimes[11d], 1) the surface (water polarity),
2) the phosphoryl headgroup (high polarity), 3) the glyceryl
esters (mid-polar regime, medium polarity) and 4) the insu-
lator regime (hydrocarbon-like polarity), the observed value
suggests that the dansyl groups of 21 are located at or near
the mid-polar (glyceryl) region of the bilayer. This observa-
tion is consistent with the mode of insertion in the mem-
brane postulated for the previously investigated CyPLOS
derivatives.[8] The polar, negatively charged, macrocycle lies
on the surface of the membrane with the four amphiphilic
TEG chains inserted in the phospholipid bilayer. The TEG
chains are apparently not deeply inserted in the phospholip-
id bilayer but are located in the midpolar region in-between
the polar surface and the hydrocarbon core of the mem-
brane. This, probably, results in a destabilisation of the struc-
ture of the membrane and, therefore, in alteration of its per-
meability. Taken together, the data obtained for compound
21 confirm the general picture on the mechanism of action
that was drawn for CyPLOS 5, experimentally supporting its
mode of insertion in the membranes. The higher activity ob-
served for 21 could probably be related either to greater
partition of the more lipophilic derivative in the membrane,
or to the larger size of the dansyl residues relative to the
benzyl groups, which may result in a more effective pertur-
bation of the phospholipid bilayer order.
Interestingly, the fluorescence maximum (lmax) observed
for 21 (3.4 mm) in HEPES buffer was 496 nm. Interpolation
results in an ET value of 34, lower that that obtained when
the ionophore is inserted in the membrane. This result indi-
cates that the dansyl dye is experiencing a less polar envi-
ronment in an aqueous buffer, suggesting the formation of
pre-aggregates in water even at this very low concentration.
Apparently, these pre-aggregates are destroyed when the
CyPLOS derivative partitions in the membrane, as suggest-
ed by the kinetic profile of Figure 4b and by the change in
the emission spectra of the dansyl group.
Conclusion
Figure 5. Plot of fluorescence emission maxima for 21 ([21]=1.7 mm) in
solvents of different polarity. The dotted lines correspond to emission
maxima of 21 ([21]=3.4 mm, lex =340 nm) measured in an EYPC/EYPG
suspension (95:5, [total lipid]=0.17 mm, 25 mm HEPES, 100 mm NaCl,
pH 7.0).
A novel, dansyl-labelled CyPLOS analogue, 21, is described
and synthesised in 12 steps and 26% overall yield from the
starting benzylidene-protected phenyl b-d-glucopyranoside
7. The overall strategy is based on the synthesis of a versa-
tile glucoside containing azido functionalities, 8, which—
through straightforward and high-fidelity reactions—gave
easy access to differently labelled monosaccharides (10a–
10c and 12), prepared here as model synthetic scaffolds.
These multifunctional entities are of potential interest them-
selves in a diversity-oriented synthetic approach for the
preparation of various glycoconjugates or glyco-architec-
tures, tailored for the development of sophisticated sensors
in biomolecule detection and in supramolecular chemistry.
In this context, these glucosides might be exploitable as
useful precursors to provide diversely functionalised
CyPLOS analogues.
investigated. Similar linear correlations were also obtained
with different polarity scales, such as Z, or by plotting the
lmax values against the dielectric constants (e) of the sol-
vents. However, the solvent polarity parameter ET is deter-
mined directly by spectral measurements in solution and so
its use appears more appropriate in the present case.[11a]
The fluorescence maximum for 21 in EYPC/EYPG lipo-
somes (95:5, [21]=3.4 mm, [total lipid]=0.17 mm) was
509 nm. This corresponds to an ET value of 45.5, intermedi-
ate between the alcohols and the less polar solvents. Be-
cause the phospholipid bilayer can be thought of as having
Chem. Eur. J. 2010, 16, 13757 – 13772
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13765

D. Montesarchio et al.
Synthetic elaboration of the dansyl-labelled glucoside 10a
gave the phosphorylated building blocks 16 and 17, which,
after coupling, cyclisation and final deprotection, furnished
the target macrocycle 21. The self-assembly properties of
this molecule in CDCl3 and in a pseudo-physiological aque-
ous solution were studied, and it was shown to form stable
aggregated systems in both media. Concentration-dependent
31P NMR spectroscopic analysis allowed a critical aggrega-
tion concentration in the 0.3–0.2 mm range to be estimated
for this amphiphilic macrocycle in CDCl3, which is an envi-
ronment roughly mimicking the polarity of the mid-polar
region of the phospholipid bilayers. Characterisation of the
aggregative behaviour of 21 in aqueous solution, providing
useful information on its organisation prior to membrane in-
sertion, was carried out by a combined experimental ap-
proach, using DLS measurements to reveal the formation
and size distribution of the aggregates and SANS measure-
ments to give an estimation of the thickness of the aggre-
gates. The data obtained from DLS and SANS studies
showed that in an aqueous environment 21 is mainly organ-
ised in the form of large vesicles, much greater in size than
all previously investigated CyPLOS molecules. DLS data
also revealed the presence in a pseudo-physiological buffer
of smaller aggregates, presumably micelles, and the diffusion
coefficients and the hydrodynamic radii for both systems
were determined.
The HPTS assay showed 21 to be able to discharge a pH
gradient across a liposomal membrane completely in 15 min
at 1% ionophore concentration, therefore displaying a
higher activity than all the CyPLOS derivatives described
previously, although the determination of the cation and
anion selectivities and of other kinetic parameters suggests a
common mechanism of action. The presence of the dansyl
dye gave valuable information on the mode of insertion of
the ionophore in the membrane, essentially confirming the
overall mechanism of action previously hypothesised for its
analogues. In particular, in EYPC/EYPG liposomes at the
concentration used in the kinetic experiments, the iono-
phore appears to be active in a monomeric form with the
polar macrocyclic head lying on the surface of the mem-
brane and the four amphiphilic TEG chains inserted in the
phospholipid bilayer. The TEG chains are apparently not
deeply inserted in the phospholipid bilayer but are located
in the mid-polar region in-between the polar surface and the
hydrocarbon core of the membrane. This probably results in
a destabilisation of the structure of the membrane and
therefore in the alteration of its permeability.
gregation, allowing these systems to be vehiculated and sta-
bilised in the extracellular media, and eventually transferred
into the phospholipid bilayers, can be crucial for their dy-
namic interactions with cell membranes and their activity.
Future studies will be addressed at the specific investigation
of these aspects, so as to produce ionophores with increased
activity and selectivity. For this purpose, novel differently
end-functionalised CyPLOS derivatives will be produced,
with advantage taken of the ready synthetic access to di-
verse analogues in this class of molecules and also with ex-
ploitation of the major role played by the reporter groups
attached to their tentacles both in aggregation and in ion-
transport activity.
Experimental Section
General methods: All reagents were of the highest commercially avail-
able quality and were used as received. TLC analyses were carried out
with silica gel plates (Merck, 60, F254). Reaction products on TLC plates
were visualised with UV light and then by treatment with a CeACHTUNGTRENNUNG(SO4)2/
H2SO4 aqueous solution (10%). For column chromatography, silica gel
(Merck Kieselgel 40, 0.063–0.200 mm) was used. The following abbrevia-
tions are used throughout the text: AcOEt=ethyl acetate; Dans=dansyl
(that
is,
5-N,N-dimethylaminonaphthalenesulfonyl);
DBU=1,8-
diazabicycloAHCTUNRTGEG[NNUN 5,4,0]undec-7-ene; DCC=dicyclohexylcarbodiimide; DCI=
4,5-dicyanoimidazole; DIPEA=N,N-diisopropylethylamine; DMAP=4-
(N,N-dimethylamino)pyridine; DMF=N,N-dimethylformamide; DMT=
4,4’-dimethoxytriphenylmethyl; EtOH=ethanol; AcOEt=ethyl acetate;
Fmoc=9-fluorenylmethoxycarbonyl;
HPTS=8-hydroxypyrene-1,3,6-trisulfonic acid, trisodium salt; MSNT=1-
mesitylensulfonyl-3-nitro-1,2,4-triazole; Ph3P=triphenylphosphine;
TEG=tetraethylene glycol; THF=tetrahydrofuran; TEA=triethyl-
amine; TFA=trifluoroacetic acid.
HOBt=1-hydroxybenzotriazole;
AHCTUNGTRENNUNG
NMR spectra were recorded with Bruker WM 400, Varian Gemini 300,
Varian Gemini 200 and Varian Inova 500 spectrometers as specified, in
all cases at 258C. All the chemical shifts are expressed in ppm with re-
spect to the residual solvent signal. Peak assignments were carried out
through standard H–H COSY and HSQC experiments. For the ESI MS
analyses, a Waters Micromass ZQ instrument—fitted with an Electro-
spray source—was used in the positive and/or negative mode. MALDI
TOF mass spectrometric analyses were performed with a PerSeptive Bio-
systems Voyager-De Pro MALDI mass spectrometer in the Linear mode
with 3,4-dihydroxybenzoic acid as the matrix.
Preparation of the investigated systems: All the samples used for DLS
and SANS investigations were prepared by direct weighing of all the
components. In particular, for each solution an appropriate amount of
the solute was dissolved in an appropriate buffer solution at pH 7.4, in
order to provide a final concentration of 2.00 mmolkgꢁ1. Because of the
intrinsically slow kinetics, dissolution was promoted by slight warming
(ꢀ408C) and a short sonication treatment (ꢀ15 min).
The buffer solution was obtained by starting from an aqueous solution of
NaCl (0.140 moldmꢁ3) and KH2PO4 (0.010 moldmꢁ3), prepared in dis-
tilled and previously degassed water, and by adding an appropriate
amount of KOH up to pH 7.4. This value was checked to be within
0.1 pH units with a Radiometer pHM220 pH-meter, with a saturated cal-
omel electrode and a glass electrode previously calibrated with the
IUPAC standard buffer solutions.[24] For SANS measurements, heavy
water (D2O, isotopic enrichment>99.8%, molar mass 20.03 gmolꢁ1), pur-
chased from Aldrich and used as such, was used as the solvent in order
to minimise the incoherent contribution to the scattering cross sections
arising from the system.
At 1% ionophore concentration, aggregation processes
for 21 may be considered—at least to a first approxima-
tion—negligible within the phospholipid bilayers, and the
ionophoric activity may therefore be reasonably accounted
for by simple monomeric forms, which can generate defects
capable of facilitating ion trafficking. On the other hand, in
an aqueous milieu, this amphiphilic ion transporter has a
strong tendency to form stable self-assembled systems even
at mm concentrations, as also confirmed by the fluorescence
shift of the appended dansyl dye. Such high levels of pre-ag-
SANS measurements: Small-angle neutron scattering measurements
were performed at 258C with the KWS2 instrument located at the Heinz
13766
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13757 – 13772

A Fluorescently Labelled CyPLOS Ionophore
FULL PAPER
Maier-Leibnitz neutron source of the Jꢄlich Centre for Neutron Science
(Garching, Germany). Neutrons with an average wavelength of 7 ꢁ and
19 ꢁ and a wavelength spread Dl/lꢃ0.2 were used. A two-dimensional
128ꢂ128 array scintillation detector at two different collimation (C)/
sample-to-detector(D) distances (C8D8 and C8D2, with all the lengths ex-
press in m) measured neutrons scattered from the samples. These config-
urations allowed data collection in a range of the scattering vector modu-
LUV suspension was separated from extravesicular dye by size exclusion
chromatography (SEC) (stationary phase: pre-packed column Sepha-
dex G-25, mobile phase: HEPES buffer) and diluted with HEPES buffer
to give a stock solution with a lipid concentration of 5 mm (with the as-
sumption that 100% of lipids were incorporated into liposomes). The
lipid suspension (104 mL) was placed in a fluorimetric cell, diluted to
3040 mL with the same buffer solution as used for the liposome prepara-
tion and kept with gentle stirring. The total lipid concentration in the flu-
orimetric cell was 0.17 mm. An aliquot of solution of the ionophore (5–
30 mL of the appropriate mother solution in order to obtain the desired
molcompound/mollipid ratio) was then added to the lipid suspension and the
cell was incubated at 258C for 30 min. After incubation the time course
of fluorescence was recorded for 200 s (lex 460 nm, lem 510 nm), NaOH
(0.5m, 50 mL) was then rapidly added through an injector port, and the
fluorescence emission was recorded for 1200 s. Maximum changes in dye
emission were obtained by final lysis of the liposomes with detergent
(40 mL of 5% aqueous Triton X-100). Fluorescence time courses were
normalised by use of the following equation [Eq. (3)]:
lus q=4p/lsinACHTUNGTRENNUNG
(q/2) located between 0.002 and 0.33 ꢁꢁ1, with q scattering
angle. The obtained raw data were then corrected for background and
empty cell scattering. Detector efficiency corrections, radial average and
transformation to absolute scattering cross sections dS/dW were made
with a secondary plexiglass standard.[28,29]
DLS measurements: Dynamic light scattering investigations were per-
formed with a setup composed of a Photocor compact goniometer, a
SMD 6000 Laser Quantum 50 mW light source operating at 5325 ꢁ and a
PMT and correlator obtained from Correlator.com. All the measure-
ments were performed at (25.00ꢂ0.05)8C with use of a thermostat bath.
In DLS, the intensity autocorrelation function g(2)(t) is measured and re-
lated to the electric field autocorrelation g(1)(t) by the Siegert relation.[30]
g(1)(t) can in turn be related, through a Laplace transform, to the distribu-
tion function of the relaxation rates G from which the z-average of the
diffusion coefficient D may be obtained [Eq. (2)]:[31]
À
Á
Ft ꢁ F0
ð3Þ
FI ¼
ꢄ 100
ðF1 ꢁ F0Þ
where Ft is the fluorescence intensity measured at time t, F0 is the fluo-
rescence intensity at ionophore addition, and F1 is the fluorescence in-
tensity at saturation after lysis with Triton. The apparent first-order rate
constants for the transport process were obtained by nonlinear regression
analysis of the fluorescence data vs. time.
G
q!0 q2
ð2Þ
(q/2) is the modulus of the scattering vector, n0 is
D ¼ lim
in which q=4pn0/lsin
U
the refractive index of the solution, l is the incident wavelength, and q
represents the scattering angle. D is thus obtained from the limit slope of
G as a function of q2, in which G is measured at different scattering
angles.
Determination of cation and anion selectivity by HPTS assay: The vesi-
cle suspension (104 mL stock solution, prepared as described above) was
placed in a fluorimetric cell and diluted to 3040 mL with the appropriate
buffer solution (HEPES (pH 7, 25 mm), MCl (100 mm) with M=Li+,
ꢁ
Na+, K+, Rb+, Cs+, or NaX (100 mm) with X=Fꢁ, Clꢁ, Brꢁ, Iꢁ, NO3
,
Ionophoric activity
ClO4ꢁ, Acꢁ, Glu). The total lipid concentration in the fluorimetric cell
was 0.17 mm. An aliquot of DMSO solution of the ionophore (5–30 mL of
the appropriate mother solution in order to obtain the desired
molcompound/mollipid ratio) was then added to the lipid suspension and the
cell was incubated at 258C for 30 min. After incubation the time course
of fluorescence was recorded for 200 s (lex 460 nm, lem 510 nm) and then
MOH (0.5m, with M=Li+, Na+, K+, Rb+, Cs+ depending on the cation
present in the extravesicular buffer solution, 50 mL) was rapidly added
through an injector port and the fluorescence emission was recorded for
1200 s. Maximum changes in dye emission were obtained by final lysis of
the liposomes with detergent (aqueous solution of Triton X-100, 5%,
40 mL). Fluorescence time courses were normalised as previously de-
scribed.
General procedures: l-a-Phosphatidyl-dl-glycerol sodium salt (EYPG,
20 mgmLꢁ1 chloroform solution) was purchased from Avanti Polar
Lipids; egg yolk phosphatidylcholine (EYPC, 100 mgmLꢁ1 chloroform
solution) and 8-hydroxypyrene-1,3,6-trisulfonic acid trisodium salt
(HPTS) were from Sigma. Triton X-100 and HEPES buffer were from
Fluka; all salts were of the best grade available from Aldrich and were
used without further purification.
Ultrafiltration was performed with Microcon YM-10 filters from Milli-
pore. Size exclusion chromatography (SEC) was performed with Sepha-
dex G-75 or pre-packed columns Sephadex G-25M (PD-10) from Amer-
sham Biosciences.
Liposomes were prepared by extrusion with use of a Lipex Thermobarrel
EXTRUDER (10 mL, Northern Lipids. Inc.) connected to a thermostatic
bath (258C if not otherwise indicated). The 100 nm polycarbonate mem-
branes are Nucleopore Track-Etch Membranes from Whatman.
Calcein release assay: The LUV suspension was prepared as previously
described with a mixture of EYPC chloroform solution (100 mgmLꢁ1
,
9.5 mmol, 72 mL) and EYPG chloroform solution (20 mgmLꢁ1, 0.5 mmol,
19 mL). The lipid cake was hydrated in calcein solution (HEPES (pH 7.4,
1 mm) 50 mm, 2.0 mL). The LUV suspension was separated from extrave-
sicular dye by size exclusion chromatography (SEC, stationary phase:
column Ø1ꢂ25 cm Sephadex G-75, mobile phase: buffer HEPES (1 mm),
NaCl (150 mm), pH 7.4) and diluted with the same HEPES buffer to give
a stock solution with a final lipid concentration of 0.4 mm (with the as-
sumption that 100% of lipid was incorporated into liposomes). The vesi-
cle suspension (550 mL stock solution) was placed in a fluorimetric cell
and diluted to 2200 mL with the buffer solution used for preparation. The
total lipid concentration in the fluorimetric cell was 0.1 mm. Calcein emis-
sion was monitored at 520 nm with excitation at 490 nm. During the ex-
periment, aliquots of DMSO solution of the ionophore were then added
up to the maximum concentration investigated, of 7% ionophore. Maxi-
mum changes in dye emission were obtained by lysis of liposomes with
detergent (aqueous Triton X-100 solution, 5%, 40 mL).
Fluorescence spectra were recorded with a Perkin–Elmer LS-50B fluo-
rimeter and a Varian Cary Eclipse Fluorescence spectrophotometer. UV
spectra were recorded with a Unicam helios b UV/Vis spectrophotome-
ter. For solvent-dependence experiments, freshly distilled solvent (2 mL)
was used instead of buffer.
The concentrations of ionophores are given as percentages with respect
to the total concentrations of lipids. Mother solutions of ionophores were
prepared in DMSO. Control experiments showed that the amounts of
DMSO added to the vesicular suspensions in the different experiments
(maximum amount 1.0% in volume) did not affect the permeabilities of
the membranes.
HPTS assay: A mixture of EYPC chloroform solution (100 mgmLꢁ1
,
225 mL, 30 mmol) and EYPG chloroform solution (20 mgmLꢁ1, 60 mL,
1.5 mmol) was first concentrated under an Ar flow to form a thin film and
then dried under high vacuum for 3 h. The lipid cake was hydrated in
HPTS solution (HEPES (25 mm), NaCl (100 mm), pH 7, 1.5 mL, 0.1 mm)
for 30 min at 408C. The lipid suspension was subjected to five freeze-
thaw cycles (ꢁ1968C/408C) with use of liquid nitrogen and a thermostat-
ic bath, and then extruded under nitrogen pressure (15 bar) at room tem-
perature (10 extrusions through a 0.1 mm polycarbonate membrane). The
Ionophore partition in liposome: The vesicle suspension was prepared as
described above in HEPES buffer (25 mm), NaCl (100 mm), pH 7, with a
final total concentration of lipids of 3.7ꢂ10ꢁ3 m. An aliquot of mother so-
lution of compound 21 was added in order to obtain a 1% concentration
of ionophore. After 30 min of incubation at 258C, the LUV suspension
Chem. Eur. J. 2010, 16, 13757 – 13772
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13767

D. Montesarchio et al.
was separated from non-membrane-bound 21 by size exclusion chroma-
tography (SEC, stationary phase: pre-packed column Sephadex G-25,
mobile phase: HEPES buffer). The liposome fraction eluted from the
column was analysed for the ionophore content: an aliquot of the suspen-
sion was dissolved in methanol and the fluorescence emission of the
dansyl dye was measured (lex =340 nm). The maximum of emission in-
tensity (lem =520 nm) was correlated to the ionophore concentration with
the aid of a calibration curve obtained by measuring the emission intensi-
ties of solutions of ionophore at known concentrations in methanol.
(75 MHz, CDCl3): d=156.9, 137.0, 129.5, 129.3, 128.8, 128.0, 125.9, 122.8,
ꢁ
116.8 and 116.3 (aromatic carbons), 101.6 (C-1), 101.0 (Ph CH), 82.3 (C-
2), 81.5 and 80.4 (2ꢂ-CH2-CH2-O-sugar), 77.1 (2ꢂ-CH2-CH2-N3), 72.4
and 72.2 (2ꢂ-CH2-O-sugar), 70.4 (12ꢂO-CH2 TEG), 70.2 (C-6), 69.8 (C-
3), 68.5 (C-4), 66.0 (C-5), 50.4 ppm (2ꢂ-CH2-N3); IR (KBr): n˜max =2104,
1636, 1558, 1490, 1456, 1225, 1085, 1036, 771, 706 cmꢁ1; MS (MALDI+):
m/z: calcd for C35H50N6O12: 746.3487; found: 747.49 [M+H]+, 769.59
[M+Na]+, 785.62 [M+K]+; HRMS (MALDI-TOF): m/z: calcd for
C35H50N6O12Na: 769.3384; found: 769.3398 [M+Na]+.
Synthesis of TEG-azido III
Synthesis of 9: Compound 8 (320 mg, 0.43 mmol) was dissolved in anhy-
drous THF (2 mL), Ph3P (281 mg, 2.5 mmol) was then added, and the so-
lution was stirred at room temperature for 8 h. Next, water was added
(5 mL) and the resulting mixture was left stirring for 48 h. The reaction
mixture was treated with a dilute aqueous HCl solution and then washed
several times with diethyl ether to remove the impurities. The aqueous
phase was then neutralised and extracted with CHCl3/CH3OH (9:1, v/v),
and the organic phases were collected and taken to dryness, giving pure 9
(300 mg, 0.43 mmol, almost quantitative yields), which was used as such
without further purification. Oil, Rf =0.4 (n-butanol/acetic acid/water
60:15:25, v/v/v); 1H NMR (500 MHz, CDCl3): d=7.48–7.02 (complex sig-
ꢁ
Synthesis of I: Ag2O (1.1 g, 4.7 mmol) and p-toluenesulfonyl chloride
(1.0 g, 5.2 mmol) were sequentially added to a solution of tetraethylene
glycol (0.89 g, 4.6 mmol) in anhydrous CH2Cl2 (10 mL). The reaction mix-
ture was stirred at room temperature for 48 h. The reaction mixture was
filtered on celite and washed twice with AcOEt and then with a saturated
NaCl solution. The organic phase, dried over anhydrous Na2SO4, was fil-
tered, concentrated under reduced pressure and then chromatographed
on a silica gel column. Elution with CHCl3 containing increasing amounts
of CH3OH (from 0 to 5%) gave the product I (1.3 g, 3.7 mmol) in 80%
yield. Oil, Rf =0.35 (CHCl3/CH3OH 95:5, v/v); 1H NMR (500 MHz,
CDCl3): d=7.81–7.33 (m, 4H; aromatic protons), 4.16 (t, 3J
ACTHUNRTGNE(GNU H,H)=4.8,
nals, 10H; aromatic protons), 5.53 (s, 1H; Ph CH), 5.02 (d, 3J
ACHTUNGTRENNUNG(H,H)=
8.0 Hz, 1H; H-1), 4.34 (m, 1H; H-4), 4.07–3.89 (overlapped signals, 4H;
-CH2-O-C-2 and -CH2-O-C-3), 3.77 (t, 3J
(H,H)=10.0 and 10.5 Hz, 1H;
4.8 Hz, 2H; -CH2-SO2), 3.72–3.60 (overlapped signals, 14H; 7ꢂ-O-CH2-
TEG), 2.44 ppm (s, 3H; CH3); 13C NMR (125 MHz, CDCl3): d=144.7,
ꢁ
AHCTUNGTRENNUNG
129.7, 127.9 and 104.2 (aromatic carbons), 72.4, 70.7, 70.6 70.4, 70.2, 69.1,
H-3), 3.65–3.52 (overlapped signals, 27H; 12ꢂ(O-CH2-TEG), H-2 and
H2-6), 3.47 (m, 1H; H-5), 2.83 ppm (brs, 4H; 2ꢂ-CH2-NH2); 13C NMR
(125 MHz, CDCl3): d=157.0, 137.1, 132.0, 129.4, 128.9, 128.3, 128.1,
+
ꢁ
68.6 and 61.7 (-O-CH2-TEG), 21.5 ppm ( CH3); MS (ESI ): m/z: calcd
for C15H24O7S: 348.1243; found: 371.23 [M+Na]+, 387.20 [M+K]+.
Synthesis of II: NaN3 (721 mg, 11.1 mmol) was added to a solution of I
(1.3 g, 3.7 mmol), dissolved in anhydrous EtOH (15 mL). The solution
was stirred for 8 h at reflux and then concentrated under reduced pres-
sure. The reaction mixture, diluted in CHCl3, was washed twice with a sa-
turated NaCl solution. The organic phase, dried over anhydrous Na2SO4,
was filtered and concentrated under reduced pressure to give II (0.81 g,
3.7 mmol), which was used as such without further purification. Oil, Rf =
0.30 (CHCl3/CH3OH 95:5, v/v); 1H NMR (500 MHz, CDCl3): d=3.71–
3.59 (overlapped signals, 12H; 6ꢂ-O-CH2-TEG), 3.41 (t, 2H; -CH2-OH),
2.32 ppm (t, 2H; -CH2-N3); 13C NMR (125 MHz, CDCl3): d=72.3, 70.5,
ꢁ
125.9, 122.8 and 116.9 (aromatic carbons), 101.7 (C-1), 101.1 (Ph CH),
82.4 (C-2), 81.5 and 80.5 (2ꢂ-CH2-CH2-O-sugar), 73.3 (2ꢂ-CH2-CH2-
NH2), 72.4 and 72.2 (2ꢂ-CH2-O-sugar), 70.6, 70.5, 70.4 and 70.3 (12ꢂO-
CH2 TEG), 70.2 (C-6), 70.1 (C-3), 68.5 (C-4), 66.0 (C-5), 41.6 ppm (2ꢂ
CH2-NH2); IR (KBr): n˜max =3504 (broad), 1638, 1558, 1486, 1452, 1230,
1089, 1032, 772, 699 cmꢁ1; MS (MALDI+): m/z: calcd for C35H54N2O12
:
694.3677; found: 695.93 [M+H]+, 717.95 [M+Na]+, 733.83 [M+K]+;
HRMS (MALDI-TOF): m/z: calcd for C35H54N2O12Na: 717.3574; found:
717.3583 [M+Na]+.
Synthesis of dansyl-tagged derivative 10a: Compound
9 (130 mg,
ꢁ
70.2, 69.9 (-O-CH2-TEG), 61.6 (-CH2-OH), 50.5 ppm ( CH2N3); MS
(ESI+): m/z: calcd for C8H17N3O4: 219.1219; found: 242.22 [M+Na]+,
258.21 [M+K]+.
0.19 mmol), previously dried by repeated coevaporations with anhydrous
benzene and kept under reduced pressure, was dissolved in anhydrous
CH2Cl2 (3 mL). TEA (100 mL, 0.76 mmol) and dansyl chloride (150 mg,
0.57 mmol) were sequentially added and the resulting mixture was left
stirring overnight at room temperature. The reaction mixture was then
Synthesis of III: DIPEA (1.5 mL, 8.6 mmol) and methanesulfonyl chlo-
ride (345 mL, 4.4 mmol) were sequentially added to compound II (0.81 g,
3.7 mmol), dissolved in anhydrous CH2Cl2 (8 mL), and the resulting mix-
ture was left overnight at room temperature. The reaction mixture was
then concentrated under reduced pressure, transferred into a separating
funnel and washed twice with a saturated NaCl solution. The organic
phase, dried over anhydrous Na2SO4, was filtered, concentrated under re-
duced pressure and then chromatographed on a silica gel column. Elution
with CHCl3 gave product III (0.99 g, 3.4 mmol) in 92% yield. Oil, Rf =
0.6 (CHCl3/CH3OH 95:5, v/v); MS (ESI+): m/z: calcd for C9H19N3O6S:
297.0995; found: 320.19 [M+Na]+, 336.28 [M+K]+.
concentrated under reduced pressure, transferred into
a separating
funnel and washed twice with a saturated NaCl solution. The organic
phase, dried over anhydrous Na2SO4, was filtered, concentrated under re-
duced pressure and then chromatographed on a silica gel column. Elution
with CHCl3 containing increasing amounts of CH3OH (from 0% to 5%)
gave product 10a (185 mg, 0.16 mmol) in 85% yield. Oil, Rf =0.5
(CHCl3/CH3OH 95:5, v/v); 1H NMR (500 MHz, CDCl3): d=8.51 (d, 3J-
A
E
ACHTUNGTREN(NUGN H,H)=8.5 Hz, 2H; 2ꢂH8-
3
(Dans)), 8.22 (d, J
CHTUNGTRENNUNG
Synthesis of 8: Compound 7 (380 mg, 1.1 mmol), prepared as previously
described,[6] was dissolved in anhydrous DMF (7.0 mL) and treated with
sodium hydride (80 mg, 3.4 mmol). The mixture was stirred for 10 min,
and compound III (0.99 g, 3.4 mmol) was then added at room tempera-
ture. After 4 h with stirring, the reaction mixture was quenched by addi-
tion of CH3OH and concentrated under reduced pressure. The crude
product was then diluted with CHCl3, transferred into a separating
funnel and washed three times with water. The organic phase was con-
centrated under reduced pressure and purified by column chromatogra-
phy. Elution of the column with CHCl3 containing increasing amounts of
CH3OH (from 0% to 5%) gave product 8 (665 mg, 0.90 mmol) in 82%
yield. Oil, Rf =0.35 (CHCl3/CH3OH 95:5, v/v); 1H NMR (400 MHz,
CDCl3): d=7.46–7.01 (complex signals, 10H; aromatic protons), 5.52 (s,
ꢁ
U
ꢁ
tons), 5.64–5.59 (brs, exchangeable protons, 2H; 2ꢂNH TEG), 5.53 (s,
1H; Ph CH), 5.06 (d, 3J
(H,H)=7.5 Hz, 1H; H-1), 4.33 (m, 1H; H-4),
ꢁ
4.07–3.91 (overlapped signals, 5H; -CH2-O-C-2, -CH2-O-C-3 and H-6a),
3.76 (t, 3J
ACTHNUTRGNEUNG(H,H)=10.5 and 10.0 Hz, 1H; H-3), 3.66–3.49 (overlapped sig-
nals, 23H; 10ꢂO-CH2-TEG, H-6b, H-2 and H-5), 3.36–3.31 (m, 4H; 2ꢂ
CH2-CH2-NHSO2-), 3.10–3.06 (m, 4H; 2ꢂ-CH2-CH2-NHSO2-), 2.87 ppm
(s, 12H; 4ꢂCH3-N-Dans); 13C NMR (75 MHz, CDCl3): d=132.1, 129.4,
128.9, 128.5, 128.1, 126.0, 122.8 and 116.9 (aromatic carbons), 101.4 (C-1),
ꢁ
101.1 (Ph CH), 82.5 (C-2), 81.5 and 80.4 (2ꢂ-CH2-CH2-O-sugar), 72.5
and 72.3 (2ꢂ-CH2-O-sugar), 70.6 and 70.4 (10ꢂO-CH2 TEG), 70.2 (C-6),
69.2 (C-3), 68.5 (C-4), 66.0 (C-5), 45.8 (4ꢂCH3-N-Dans), 43.0 ppm (2ꢂ
CH2-CH2-NHSO2); IR (KBr): n˜max =3494 (broad), 1593, 1494, 1459, 1325,
1H; Ph CH), 5.01 (d, 3J
ACTHUNRTGNE(GNU H,H)=7.6 Hz, 1H; H-1), 4.33 (m, 1H; H-4),
1223, 1142, 1089, 768, 699 cmꢁ1
;
MS (ESI+): m/z: calcd for
4.05–3.88 (overlapped signals, 5H; -CH2-O-C-2, -CH2-O-C-3 and H-6a],
3.76 (t, J=10.4, 10.0 Hz, 1H; H-3), 3.64–3.57 (overlapped signals, 25H;
12ꢂO-CH2-TEG and H-6b), 3.51–3.43 (overlapped signals, 2H; H-2 and
C59H76N4O16S2: 1160.47; found: 1161.65 [M+H]+, 1183.64 [M+Na]+,
1199.69 HRMS (MALDI-TOF): m/z: calcd for
[M+K]+;
C59H76N4O16S2Na: 1183.4595; found: 1183.4612 [M+Na]+.
ACHTUNGTRENNUNG
H-5), 3.34 ppm (t, 3J(H,H)=4.8, 4.8 Hz, 4H; -CH2-N3); 13C NMR
13768
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13757 – 13772

A Fluorescently Labelled CyPLOS Ionophore
FULL PAPER
Synthesis of tryptophan-tagged derivative 10b: Compound 9 (130 mg,
0.19 mmol), previously dried by repeated coevaporations with anhydrous
benzene and kept under reduced pressure, was dissolved in anhydrous
CH2Cl2 (3 mL). DIPEA (163 mL, 0.76 mmol), DCC (235 mg, 1.14 mmol)
and Fmoc-protected tryptophan (240 mg, 0.57 mmol) were sequentially
added and the resulting solution was left stirring overnight at room tem-
perature. The reaction mixture was then concentrated under reduced
pressure, transferred into a separating funnel and washed twice with a sa-
turated NaCl solution. The organic phase, dried over anhydrous Na2SO4,
was filtered, concentrated under reduced pressure and then chromato-
graphed on a silica gel column. Elution with CHCl3 containing increasing
amounts of CH3OH (from 0% to 10%), allowed product 10b to be iso-
lated in a pure form (235 mg, 0.16 mmol) in 84% yield. Oil, Rf =0.5
(CHCl3/CH3OH 9:1, v/v); 1H NMR (500 MHz, CDCl3): d=7.76–6.95
(complex signals, 38H; aromatic protons), 6.00 (brs, exchangeable pro-
stirred mixture, which was left at room temperature for 4 h. The reaction
mixture was then diluted with CHCl3, transferred into a separating
funnel and washed twice with water. The organic phase, dried over anhy-
drous Na2SO4 and filtered, was then concentrated under reduced pressure
and purified by column chromatography. Elution of the column with
CHCl3 containing increasing amounts of CH3OH (from 1 to 10%) gave
pure 11 (1.0 g, 2.2 mmol) in 92% yield. White amorphous solid, Rf =0.6
(CH2Cl2/CH3OH 85:15, v/v); 1H NMR (400 MHz, CDCl3): d=5.70 (brs,
ꢁ
1H; exchangeable proton, NH), 4.05 (brs, 2H; CH2 NH), 3.98, 3.86 and
3.47 (3ꢂbrs, each 1H; 3ꢂCHOH), 2.23 (1H; H-alkyne), 2.27–1.15 (over-
lapped signals, 24H; CH2 and CH protons of the cholic acid residue),
1.00, 0.92 and 0.71 ppm (3ꢂs, each 3H; CH3 protons of the cholic acid
residue); 13C NMR (50 MHz, CDCl3): d=174.0 (CO), 79.9 (Csp), 71.7
ꢁ
(CHsp), 73.0, 71.0 and 68.3 (CHOH), 46.2 (CH2 NH), 41.4, 39.2, 35.2,
34.6, 32.6, 31.4, 30.2, 28.9, 28.0, 27.5, 26.1 and 23.1 (CH2 and CH carbons
of the cholic acid residue), 22.3, 17.3 and 12.3 ppm (CH3 carbons of the
cholic acid residue); IR (KBr): n˜max =3306 (broad), 1649, 1540, 1445,
1370, 1073, 1037, 754, 661 cmꢁ1; MS (ESI+): m/z: calcd for C27H43NO4:
445.3192; found: 468.71 [M+Na]+, 484.59 [M+K]+.
ꢁ
tons, 2H; 2ꢂNH TEG), 5.84 (brs, exchangeable protons, 2H; 2ꢂNH-
3
ꢁ
Trp), 5.29 (s, 1H; Ph CH), 4.88 (d, J
N
(overlapped signals, 6H; 2ꢂCH2-Fmoc and 2ꢂCH-Trp), 4.26 (m, 1H; H-
4), 4.21 (m, 2H; 2ꢂCH-Fmoc), 4.02–3.79 (overlapped signals, 5H; -CH2-
O-C-2, -CH2-O-C-3 and H-6a), 3.66–3.18 (overlapped signals, 31H; H-3,
12ꢂO-CH2 TEG, 2ꢂCH2-Trp, H-2 and H-6b), 3.11–3.05 ppm (overlapped
signals, 5H; 2ꢂ-CH2-NHCO and H-5); 13C NMR (125 MHz, CDCl3): d=
171.0 (CO), 156.8, 155.7, 143.7, 141.1, 137.1, 136.2, 129.4, 128.9, 128.1,
127.6, 127.3, 127.0, 126.0, 125.0, 123.7, 122.8, 121.8, 119.8, 119.4, 118.7,
CuI-promoted 1,3-dipolar cycloaddition—synthesis of triazole 12: Com-
pound 8 (84 mg, 0.11 mmol) and compound 11 (150 mg, 0.33 mmol) were
dissolved in anhydrous DMF (1.0 mL). DBU (165 mL, 1.1 mmol) and CuI
(63 mg, 0.33 mmol) were added to the stirred mixture, which was left at
room temperature for 48 h. The reaction mixture was concentrated under
reduced pressure, diluted with CHCl3 and then transferred into a separat-
ing funnel and washed twice with water. The organic phase, dried over
anhydrous Na2SO4 and filtered, was then concentrated under reduced
pressure and the resulting residue was purified by column chromatogra-
phy. Elution of the column with CHCl3 containing increasing amounts of
CH3OH (from 1 to 10%) gave pure 12 (165 mg, 0.10 mmol) in 92%
yield. Oil, Rf =0.6 (CH2Cl2/CH3OH 85:15, v/v); 1H NMR (500 MHz,
CDCl3): d=7.44–7.01 (complex signals, 12H; aromatic protons), 5.57 (s,
ꢁ
ꢁ
116.6, 111.4 and 109.9 (aromatic carbons), 101.2 (C-1), 101.0 (Ph CH),
82.3 (C-2), 81.4 and 80.3 (2ꢂ-CH2-CH2-O-sugar), 72.1 and 71.9 (2ꢂ-CH2-
O-sugar), 70.6, 70.4, 70.2 and 70.1 (10ꢂO-CH2 TEG), 69.9 (C-3), 69.2 (C-
6), 68.4 (C-4), 66.8 (CH2-Fmoc), 65.8 (C-5), 55.5 (CH-a Trp), 47.0 (CH-
Fmoc), 39.1 (2ꢂ-CH2-NHCO), 29.6 ppm (CH2-Trp); IR (KBr): n˜max
=
3412 and 3297 (broad), 1703, 1663, 1499, 1455, 1228, 1085, 743 cmꢁ1; MS
(ESI+): m/z: calcd for C87H94N6O18: 1510.6625; found: 1535.96 [M+Na]+,
1552.05 [M+K]+; HRMS (MALDI-TOF): m/z: calcd for C87H94N6O18Na:
1533.6522; found: 1533.6539 [M+Na]+.
1H; Ph CH), 5.02 (d, 3J
ACTHNUGRTENUNG(H,H)=7.6 Hz, 1H; H-1), 4.56 (brs, exchangea-
Synthesis of ferrocene-tagged derivative 10c: Compound
9
(130 mg,
ble protons, 2ꢂNH), 4.37 (m, 1H; H-4), 4.32–3.70 (overlapped signals,
ꢁ
0.19 mmol), previously dried by repeated coevaporations with anhydrous
benzene and kept under reduced pressure, was dissolved in anhydrous
CH2Cl2 (3 mL). DIPEA (130 mL, 0.76 mmol), DCC (235 mg, 1.1 mmol),
and ferrocenecarboxylic acid (132 mg, 0.57 mmol) were sequentially
added and the resulting solution was left stirring overnight at room tem-
perature. The reaction mixture was then concentrated under reduced
pressure, transferred into a separating funnel and washed twice with a sa-
turated NaCl solution. The organic phase, dried over anhydrous Na2SO4,
was filtered, concentrated under reduced pressure and then chromato-
graphed on a silica gel column. Elution with CHCl3 containing increasing
amounts of CH3OH (from 0% to 10%) gave product 10c (150 mg,
0.13 mmol) in 68% yield. Oil, Rf =0.3 (CHCl3/CH3OH 9:1, v/v);
1H NMR (500 MHz, CDCl3): d=7.51–7.03 (complex signals, 10H; aro-
matic protons), 6.12 (brs, exchangeable protons, 2ꢂNH), 5.64 (s, 1H;
8H; -CH2-O-C-2, -CH2-O-C-3 and 2ꢂCH2 NHCO), 3.63–3.40 (over-
lapped signals, 35H; 12ꢂO-CH2 TEG, 6ꢂCHOH of the cholic acid resi-
dues, H-3, H2-6, H-2 and H-5), 2.90 (brs, 4H; 2ꢂ-CH2-N TEG), 2.40–
1.20 (overlapped signals, 48H; CH2 and CH protons of the cholic acid
residues), 0.85, 0.66 and 0.63 ppm (3ꢂs, each 6H; CH3 protons of the
cholic acid residues); 13C NMR (50 MHz, CDCl3): d=166.3 (CO), 157.2,
145.8, 137.3, 129.4, 128.8, 128.0, 126.0, 122.9, 117.1 (aromatic carbons),
ꢁ
101.9 (C-1), 101.2 (Ph CH), 82.5 (C-2), 81.7 and 80.7 (2ꢂ-CH2-CH2-O-
sugar), 72.4 and 72.3 (2ꢂ-CH2-O-sugar), 72.9, 71.8 and 68.2 (CHOH of
the cholic acid residues), 70.5 (10ꢂO CH2 TEG), 68.8 (C-3), 68.6 (C-4),
ꢁ
ꢁ
ꢁ
66.2 (C-5), 62.8 (C-6), 54.4 (N CH2 TEG), 46.5 (CH2 NHCO), 48.7,
41.6, 39.7, 37.9, 35.3, 34.7, 32.2, 31.4, 30.5, 28.9, 27.5, 26.7, 23.9, 23.2 and
22.4 (CH2 and CH carbons of the cholic acid residues), 19.5, 17.4 and
12.4 ppm (CH3 carbons of the cholic acid residues); IR (KBr): n˜max =3365
(broad), 1646, 1540, 1455, 1375, 1222, 1077, 1043, 756, 661 cmꢁ1; MS
3
ꢁ
Ph CH), 5.30 (d, J
N
4 ferrocene rings A), 4.34 (m, 1H; H-4), 4.28 (brs, 4H; H-2 and H-5 fer-
rocene rings A), 4.19 (s, 10H; ferrocene unsubstituted rings B), 4.05–3.95
(overlapped signals, 5H; -CH2-O-C-2, -CH2-O-C-3 and H-6a), 3.84–3.49
(overlapped signals, 26H; H-3, 12ꢂO-CH2 TEG and H-6b), 3.42–3.38 (m,
2H; H-2 and H-5), 3.08–3.03 ppm (m, 4H; 2ꢂ-CH2-NHCO); 13C NMR
(100 MHz, CDCl3): d=170.8 (CO), 156.8, 137.2, 130.3, 129.7, 128.9, 128.7,
128.1, 127.3, 127.0, 125.4, 123.7, 122.0, 117.5 and 115.9 (aromatic car-
(MALDI+): m/z: calcd for C89H136N8O20
: 1636.9871; found: 1637.48
[M+H]+, 1676.63 [M+K]+; HRMS (MALDI-TOF): m/z: calcd for
C89H136N8O20Na: 1659.9769; found: 1659.9806 [M+Na]+.
Removal of benzylidene group—synthesis of 14: Compound 10a
(170 mg, 0.15 mmol), dissolved in anhydrous THF (1.3 mL), was treated
with sodium hydride (18 mg, 0.44 mmol) and the reaction mixture was
left stirring for 10 min. Methyl iodide (20 mL, 0.33 mmol) was then added
and the resulting system was left overnight at room temperature. The re-
action mixture was then concentrated under reduced pressure, diluted
with CHCl3, transferred into a separating funnel and washed twice with
water. The organic phase was then taken to dryness, which gave target
compound 13 in a pure form and almost quantitative yields. Derivative
ꢁ
bons), 102.1 (Ph CH), 100.4 (C-1), 83.3 (C-2), 81.8 and 80.4 (2ꢂCH2-
CH2-O-sugar), 79.4 (ferrocene ring B carbons), 71.5 and 71.1 (2ꢂ-CH2-
O-sugar), 70.5, 70.1, 69.3 and 68.8 (10ꢂO-CH2 TEG and ferrocene
ring A carbons), 67.7 (C-3), 66.7 (C-4), 66.2 (C-5), 65.2 (C-6), 40.7 ppm
ꢁ
(2ꢂCH2 NHCO); IR (KBr): n˜max =3410 and 3290 (broad), 1635, 1540,
1490, 1452, 1226, 1085, 1024, 733, 696 cmꢁ1; MS (ESI+): m/z: calcd for
C57H70N2Fe2O14: 1118.3526; found: 1141.34 [M+Na]+, 1157.36 [M+K]+;
HRMS (MALDI-TOF): m/z: calcd for C57H70N2Fe2O14Na: 1141.3424;
found: 1141.3446 [M+Na]+.
13 (170 mg, 0.15 mmol) was then treated with
a TFA/CH2Cl2/H2O
(1:10:0.5 v/v/v) solution (10 mL) and the resulting mixture was left at
08C. After 4 h, the reaction mixture was diluted with CHCl3. The organic
phase was washed twice with water and concentrated under reduced
pressure. The crude product was purified by column chromatography.
Elution of the column with CHCl3 containing increasing amounts of
CH3OH (from 1 to 5%) gave pure 14 (160 mg, 0.15 mmol) in almost
quantitative yields. Oil, Rf =0.3 (CHCl3/CH3OH 95:5, v/v); 1H NMR
Synthesis of N-propargylamide of cholic acid (11): Cholic acid (1.0 g,
2.4 mmol) was dissolved in anhydrous CH2Cl2 (6 mL). DIPEA (640 mL,
3.7 mmol), DCC (760 mg, 3.7 mmol), HOBt (360 mg, 2.7 mmol) and
propargylamine (185 mL, 2.7 mmol) were sequentially added to the
Chem. Eur. J. 2010, 16, 13757 – 13772
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13769

D. Montesarchio et al.
3
(500 MHz, CDCl3): d=8.52 (d, J
G
2A
4A
(H,H)=7.5 Hz, 2H;
(Dans)), 7.27–6.98 (complex signals, 5H; phenyl protons), 4.95 (d,
(H,H)=7.0 Hz, 1H; H-1), 4.12 (m, 1H; H-4), 4.07–3.72 (overlapped
signals, 6H; -CH2-O-C-2, -CH2-O-C-3 and H2-6), 3.66–3.52 (overlapped
{2ꢂs, each 3H; 24H; 2ꢂN[CHACTHGNUTERNNUG
(CH3)2]2}. 13C NMR (75 MHz, CDCl3): d=
3
(d, J
G
(Dans)), 8.14 (m, 2H; 2ꢂH
R
158.2, 151.6, 145.0, 139.4, 134.4, 130.1, 129.3, 129.0, 128.2, 127.8, 127.7,
126.5, 123.0, 119.6, 116.8, 115.1, 113.1, 112.9 and 112.1 (aromatic car-
bons), 116.8 (CN), 101.0 (C-1), 85.8 (quaternary C of DMT group), 84.0
and 82.4 (C-2 and C-5), 75.3 (C-3), 72.3 and 71.9 (2ꢂCH2-O-sugar), 70.4,
70.3 and 69.8 (O-CH2-CH2-O TEG and, overlapped, C-4), 63.7 and 63.5
(m, 4H; 2ꢂH
2ꢂH6A
3J
G
U
CHTUNGTRENNUNG
ACHTUNGTRENNUNG
ꢁ
ꢁ
ꢁ
signals, 25H; 12ꢂO CH2 TEG and H-3), 3.44 (m, 4H; N CH2 TEG),
3.41 (t, J=5.5 and 5.5 Hz, 1H; H-5), 3.38 (m, 1H; H-2), 2.94 and 2.93
(2ꢂs, each 3H; 2ꢂ-CH2-NCH3SO2-), 2.87 ppm (s, 12H; 4ꢂCH3-N-
Dans); 13C NMR (50 MHz, CDCl3): d=157.1, 151.7, 134.7, 130.1, 129.4,
127.8, 123.0, 119.7, 116.7 and 115.2 (aromatic carbons), 101.4 (C-1), 85.9
(C-2), 82.0 (C-5), 75.6 (C-3), 72.2 and 71.9 (2ꢂCH2-O-sugar), 71.9, 70.5,
(C-6), 58.0 (-O-CH2-CH2-CN), 55.0 (OCH3 of DMT group), 49.1 (2ꢂN
CH2 TEG), 46.7 (4ꢂCH3-N-Dans), 42.9 {N[CH
NCH3SO2-), 24.3 and 22.2 {N[CH
(CH3)2]2}, 35.7 (2ꢂ-CH2-
(CH3)2]2}, 19.9 and 19.1 ppm (-O-CH2-
CH2-CN); 31P NMR (161.98 MHz, CDCl3): d=151.0 and 150.4 ppm; IR
(KBr): n˜max =2252, 1606, 1507, 1457, 1324, 1217, 1138, 1027, 973,
765 cmꢁ1; MS (ESI+): m/z: calcd for C84H111N6O19PS2: 1602.7083; found:
1605.02 [M+H]+, 1705.12 [M+Et3NH]+; HRMS (MALDI-TOF): m/z:
calcd for C84H111N6O19PS2Na: 1625.6981; found: 1625.6997 [M+Na]+.
ꢁ
ꢁ
70.4 and 70.2 (12ꢂO CH2 TEG), 62.8 (C-6), 60.1 (C-4), 49.2 (2ꢂN CH2
TEG), 45.2 (4ꢂCH3-N-Dans), 35.7 ppm (2ꢂ-CH2-NCH3SO2-); IR (KBr):
n˜max =3451 (broad), 1589, 1574, 1452, 1324, 1229, 1142, 1070, 983, 933,
774, 695 cmꢁ1. MS (MALDI+): m/z: calcd for C54H76N4O16S2: 1100.4698;
found: 1101.14 [M+H]+, 1123.27 [M+Na]+, 1139.24 [M+K]+; HRMS
(MALDI-TOF): m/z: calcd for C54H76N4O16S2Na: 1123.4595; found:
1123.4630 [M+Na]+.
Synthesis of 4-O-(2-chlorophenylphosphate) derivative 17: 2-Chlorophen-
yl-dichlorophosphate (228 mL, 1.40 mmol, 4 equiv) was added dropwise
at 08C to
a stirred solution of compound 15 (500 mg, 0.35 mmol,
1 equiv), 1,2,4-triazole (193 mg, 2.80 mmol, 8 equiv) and triethylamine
(390 mL, 2.80 mmol, 8 equiv) in anhydrous pyridine (9.0 mL). The mix-
ture was allowed to warm to room temperature. After 3 h the reaction
mixture was concentrated under reduced pressure. The crude product
Synthesis of 15: Compound 14 (160 mg, 0.15 mmol), dissolved in anhy-
drous pyridine (2.0 mL), was treated with DMTCl (60 mg, 0.18 mmol).
The reaction mixture, left at room temperature overnight with stirring,
was then diluted with CH3OH and concentrated under reduced pressure.
The crude product was next purified on a silica gel column. Elution with
CH2Cl2 containing increasing amounts of CH3OH (from 1 to 5%) in the
presence of a few drops of pyridine gave pure 15 (210 mg, 0.15 mmol) in
almost quantitative yield. Glassy solid, m.p. dec.> 908C. Rf =0.8 (CHCl3/
was then diluted with CHCl3, transferred into
a separating funnel,
washed three times with water, concentrated under reduced pressure and
purified by column chromatography. Elution with CH2Cl2 containing in-
creasing amounts of CH3OH (from 1 to 10%), with the addition of a few
drops of TFA, afforded pure 17 (395 mg, 0.30 mmol) in 86% yield. Oil.
Rf =0.3 (CHCl3/CH3OH, 98:2, v/v). 1H NMR (500 MHz, CDCl3): d=8.51
CH3OH 95:5, v/v). 1H NMR (500 MHz, CDCl3): d=8.52 (d, 3J
8.5 Hz, 2H; 2ꢂH2A (H,H)=8.5 Hz, 2H; 2ꢂH8A(Dans)),
(Dans)), 8.33 (d, 3J
8.14 (m, 2H; 2ꢂH4A(Dans)), 7.52 (m, 4H; 2ꢂH3A(Dans) and 2ꢂH7A(Dans)),
7.49–6.71 (complex signals, 20H; DMT aromatic protons, phenyl protons
and 2ꢂH6A (H,H)=8.0 Hz, 1H; H-1), 4.13–3.71 (over-
(Dans)), 4.95 (d, 3J
A
(d, 3J
G
2A
(H,H)=7.5 Hz, 2H; 2ꢂH
(Dans)), 7.30–6.90 (complex signals, 11H; phenyl
(Dans)), 4.78 (d, 3J
(H,H)=7.0 Hz, 1H; H-1), 4.48 (m,
1H; H-4), 4.00 (m, 2H; CH2-O-sugar), 3.91 (d, 3J
(H,H)=12.0 Hz, 1H;
ACHTUNGTREN(NGNU H,H)=9.0 Hz, 2H;
3
C
E
U
2ꢂH
2ꢂH
8A
3A(Dans) and 2ꢂH
6A
CHTUNGTRENNUNG
C
E
U
protons and 2ꢂH
ACHTUNGTRENNUNG
C
R
ACHTUNGTRENNUNG
lapped signals, 5H; -CH2-O-C-2, -CH2-O-C-3 and H-4), 3.74 (s, 6H;
H-6a), 3.75–3.68 (overlapped signals, 3H; CH2-O-sugar and H-6b), 3.64–
3.38 (overlapped signals, 30H; H-3, H-2 and O-CH2-CH2-O TEG), 3.30
(m, 1H; H-5), 2.90 (s, 6H; 2ꢂ-CH2-NCH3SO2-), 2.86 and 2.81 ppm (s,
12H; 4ꢂCH3-N-Dans); 13C NMR (125 MHz, CDCl3): d=157.3, 151.6,
149.5, 134.3, 130.5, 130.1, 130.0, 129.8, 129.7, 129.4, 129.2, 128.6, 127.9,
127.1, 125.4, 123.4, 123.0, 122.3, 122.1, 119.5, 118.9, 116.8 and 115.0 (aro-
matic carbons), 101.5 (C-1), 85.1 (C-2), 82.0 (C-3), 76.0 (C-5), 72.6 (C-4),
71.5 and 70.9 (2ꢂCH2-O-sugar), 70.3, 70.2, 70.1, 69.7 and 69.2 (O-CH2-
ꢁ
OCH3 of the DMT group), 3.65–3.53 (overlapped signals, 26H; 12ꢂO
ꢁ
CH2 TEG and H2-6), 3.47–3.42 (overlapped signals, 6H; 2ꢂN CH2 TEG,
H-3 and H-5), 3.38 (t, 1H; H-2), 2.94 and 2.91 (2ꢂs, each 3H; 2ꢂ-CH2-
NCH3SO2-), 2.87 ppm (s, 12H; 4ꢂCH3-N-Dans); 13C NMR (CDCl3,
50 MHz): d=158.2, 157.2, 151.6, 144.9, 139.4, 136.1, 134.4, 130.0, 129.3,
129.0, 128.2, 127.7, 126.9, 126.4, 123.0, 122.4, 119.6, 116.9, 115.1, 113.1 and
112.9 (aromatic carbons), 101.2 (C-1), 86.3 (quaternary
C of DMT
ꢁ
group), 85.8 (C-2), 82.1 (C-5), 75.3 (C-3), 72.3 and 71.9 (2ꢂCH2-O-
sugar), 70.5, 70.4 and 69.9 (12ꢂOꢂCH2 TEG and, overlapped, C-4), 63.4
CH2-O TEG), 61.0 (C-6), 49.2 and 49.0 (2ꢂN CH2 TEG), 46.1 and 45.3
(4ꢂCH3-N-Dans),
35.7 ppm
(2ꢂ-CH2-NCH3SO2-)];
31P NMR
ꢁ
(C-6), 55.1 (OCH3 of the DMT group), 49.1 (2ꢂN CH2 TEG), 46.2 and
(161.98 MHz, CDCl3): d=ꢁ6.8 ppm; HRMS (MALDI-TOF): m/z: calcd
45.3 (4ꢂCH3-N-Dans), 35.8 ppm (2ꢂ-CH2-NCH3SO2-); IR (KBr): n˜max
=
for C60H80ClN4O19PS2: 1290.4284; found: 1289.3460 [MꢁH]ꢁ.
1608, 1511, 1453, 1301, 1247, 1142, 1080, 1030, 820, 787, 700 cmꢁ1; MS
(ESI+): m/z: calcd for C75H94N4O18S2: 1402.6004; found: 1425.66
[M+Na]+, 1441.68 [M+K]+; HRMS (MALDI-TOF): m/z: calcd for
C75H94N4O18S2Na: 1425.5902; found: 1425.5980 [M+Na]+.
Synthesis of DMT-free linear dimer 18: Derivative 17 (150 mg,
0.12 mmol, 1 equiv) and compound 16 (228 mg, 0.14 mmol, 1.2 equiv),
previously dried by repeated coevaporations with anhydrous CH3CN and
kept under reduced pressure, were treated with a DCI solution (0.25m)
in anhydrous CH3CN (5.0 mL). The reaction was left stirring and moni-
tored by TLC in the eluent system CHCl3/CH3OH 96:4 v/v. After 2.0 h, a
tBuOOH solution in n-decane (5.5m, 0.3 mL) was added to the mixture,
which was left stirring at room temperature. After an additional 30 min
the reaction mixture was diluted with CHCl3, transferred into a separat-
ing funnel and washed three times with water. The organic phase, con-
centrated under reduced pressure, was then purified by column chroma-
tography. Elution with CH2Cl2 containing increasing amounts of CH3OH
(from 1 to 10%) in the presence of a few drops of TFA afforded pure 18
(218 mg, 0.087 mmol) in 72% yield. Oil. Rf =0.5 (CHCl3/CH3OH 98:2, v/
Synthesis of 16: Compound 15 (200 mg, 0.14 mmol), dissolved in anhy-
drous CH2Cl2 (2 mL) and previously left in contact with activated molec-
ular sieves (3 ꢁ), was treated with DIPEA (145 mL, 0.84 mmol) and 2-
cyanoethyl-N,N-diisopropylchlorophosphoramidite (100 mL, 0.45 mmol),
sequentially added with stirring at room temperature. After 2 h, the reac-
tion mixture was concentrated under reduced pressure. The crude prod-
uct was then chromatographed on a silica gel column, with elution with
n-hexane containing increasing amounts of ethyl acetate (from 30 to
50%) in the presence of a few drops of triethylamine, furnishing the de-
sired compound 16 (200 mg, 0.13 mmol) in 90% yield. Oil, as a mixture
of diastereomers: Rf =0.5 (CH2Cl2/CH3OH, 99:1, v/v); 1H NMR
v). H NMR (400 MHz, CDCl3): d=8.52 (d, J
(Dans)), 8.33 (d, 3J
(H,H)=8.5 Hz, 4H; 4ꢂH
4A(Dans)), 7.50 (m, 8H; 4ꢂH3A(Dans) and 4ꢂH
plex signals, 18H; phenyl protons and 4ꢂH6A(Dans)], 4.89 and 4.80 (2ꢂm,
ACHTUNGTRENNUNG
1
3
3
(300 MHz, CDCl3): d=8.52 (d, J
G
E
G
R
CHTUNGTRENNUNG
(d, 3J
4ꢂH4A
(H,H)=8.5 Hz, 4H; 4ꢂH
8A
G
H
G
G
CHTUNGTRENNUNG
R
G
U
CHTUNGTRENNUNG
each 1H; H-1 and H-1’), 4.23 (t, 2H; O-CH2-CH2-CN), 4.10–3.09 (over-
ꢁ
H
G
lapped signals, 74H; H2-6-O-P, 2ꢂH-4, 2ꢂH-3 and 2ꢂH-5, 32ꢂO CH2
-O-CH2-CH2-CN, 4ꢂ-CH2-O-sugar, 2ꢂH-6a, 2ꢂH-4), 3.79 (s, 12H; 4ꢂ
OCH3 of DMT group), 3.76–3.37 {overlapped signals, 68H; 2ꢂH-6b, 2ꢂ
H-3, O-CH2-CH2-O TEG, 2ꢂH-2, 2ꢂH-5 and 4ꢂN[CHACTHUNGTRNENUG(CH3)2]2}, 2.93 (s,
12H; 4ꢂ-CH2-NCH3SO2-), 2.86 (s, 24H; 8ꢂCH3-N-Dans), 2.73 (t, 4H;
2ꢂ-O-CH2-CH2-CN), 1.28, 1.27, 1.26, 1.25, 1.22, 1.20, 0.98 and 0.96 ppm
TEG and H2-6-OH), 3.40 and 3.38 (obscured, 2H; 2ꢂH-2), 2.91 (s, 12H;
4ꢂ-CH2-NCH3SO2-), 2.85 (s, 24H; 8ꢂCH3-N-Dans), 2.49–2.37 ppm (m,
2H; O-CH2-CH2-CN; 13C NMR (100 MHz, CDCl3): d=165.1, 156.9,
151.6, 144.9, 134.4, 134.2, 130.8, 130.2, 130.0, 129.4, 128.7, 127.9, 123.0,
122.6, 121.6, 119.5, 116.6, 116.4, 115.1 (aromatic carbons), 116.1 (CN),
13770
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13757 – 13772

A Fluorescently Labelled CyPLOS Ionophore
FULL PAPER
100.6 and 100.2 (2ꢂC-1), 84.2 and 84.0 (2ꢂC-2), 81.7 and 81.6 (2ꢂC-5),
75.3 (2ꢂC-3), 74.0 and 73.8 (4ꢂCH2-O-sugar), 71.7, 70.2, 69.5 and 69.2
centrated under reduced pressure and purified by column chromatogra-
phy. Elution of the column with CH2Cl2 containing increasing amounts of
ꢁ
CH3OH (from
0 to 15%) gave pure cyclic dimer 21 (20 mg,
(28ꢂO CH2 TEG and, overlapped, 2ꢂC-4), 68.0 (C-6-O-P), 62.5 (C-6-
ꢁ
0.0086 mmol) in an almost quantitative yield. Oil, Rf =0.5 (CHCl3/
OH), 60.3 (-O-CH2-CH2-CN), 49.1 (4ꢂN CH2 TEG), 45.5 and 45.3 (8ꢂ
CH3OH 9:1 v/v); 1H NMR (500 MHz, CDCl3): d=8.78 (broad, 4H; 4ꢂ
CH3-N-Dans), 35.6 (4ꢂ-CH2-NCH3SO2-), 29.6 ppm (-O-CH2-CH2-CN);
31P NMR (161.98 MHz, CDCl3): broad signals centred at d =ꢁ3.1 and
ꢁ7.4 ppm; HRMS (MALDIꢁ-TOF): m/z: calcd for C117H158ClN9O37P2S4:
2505.8805; found: 2503.2009 [MꢁH]ꢁ.
H
C
(Dans)), 8.04 (broad, 4H; 4ꢂH4-
N
C
CHTUNGTRENNUNG
CTHUNGTRENNUNG
5.36–4.92 (m, 2H; H-1 and H-1’), 4.40–3.37 (overlapped signals, 74H; 2ꢂ
H2-6, 2ꢂH-4, O-CH2-CH2-O TEG, 2ꢂH-2 and 2ꢂH-3), 3.11 (m, 2H; 2ꢂ
H-5), 2.97 (s, 12H; 4ꢂ-CH2-NCH3SO2-), 2.93 ppm (s, 24H; 8ꢂCH3-N-
Dans); 13C NMR (100 MHz, CDCl3): d=165.1, 158.0, 153.9, 142.4, 139.1,
134.9, 130.5, 129.4, 128.2, 124.8, 123.1, 122.6, 119.6, 117.2 and 115.1 (aro-
matic carbons), 101.4 and 101.1 (C-1 and C-1’), 84.7 (C-5 and C-5’), 75.9
(C-4 and C-4’), 71.9, 70.3 and 69.8 (O-CH2-CH2-O TEG), 61.6 (C-6), 49.0
Synthesis of fully protected cyclic dimer 19: Derivative 18 (30 mg,
0.012 mmol, 1 equiv), previously dried by repeated coevaporations with
anhydrous pyridine, DMAP (3.0 mg, 0.024 mmol, 2 equiv) and MSNT
(106 mg, 0.36 mmol, 30 equiv) were dissolved in anhydrous pyridine
(12 mL) and left two days with stirring at room temperature. The reac-
tion mixture was then concentrated under reduced pressure, dissolved in
ethyl acetate, transferred into a separating funnel and washed three times
with water. The organic phase was concentrated under reduced pressure
and purified by column chromatography with elution with CHCl3 con-
taining increasing amounts of CH3OH (from 1 to 15%), affording pure
19 (22 mg, 0.009 mmol) in 75% yield. Oil, Rf =0.5 (CHCl3/CH3OH
ꢁ
(2ꢂN CH2 TEG), 45.3 and 45.2 (4ꢂCH3-N-Dans), 35.8 and 35.6 ppm
(2ꢂ-CH2-NCH3SO2-); 31P NMR (161.98 MHz, CDCl3, 3.0 mm): very
broad signal, centred at d=1.33 ppm; 31P NMR (161.98 MHz, CDCl3,
0.09 mm): sharp signal at d=1.5 ppm; HRMS (MALDIꢁ-TOF): m/z:
calcd for C108H150N8O36P2S4: 2324.8511; found: 2323.1221 [MꢁH]ꢁ;
1
3
95:5, v/v). H NMR (400 MHz, CDCl3): d=8.59 (d, JCAHTUNGTRENNNUG
1160.8435 [Mꢁ2H]2ꢁ
.
3
4ꢂH
4ꢂH
(Dans)), 8.37 (d, J
4A(Dans)), 7.53 (m, 8H; 4ꢂH
R
8ACHTUNGTRENNUNG
U
CHTUNGTRENNUNG
(complex signals, 18H; phenyl protons and 4ꢂH HCTUNGTRENNUNG
(2ꢂm, each 1H: H-1 and H-1’), 4.34 (overlapped signals, 8H; 4ꢂCH2-
CH2-O-sugar), 4.07 (t, 2H; O-CH2-CH2-CN), 3.97–3.39 (overlapped sig-
ꢁ
nals, 72H; H2-6-O-P, 2ꢂH-4, 2ꢂH-3, 2ꢂH-5, 28ꢂO CH2 TEG and H2-6-
Acknowledgements
OH), 3.37 and 3.28 (obscured, 2H; 2ꢂH-2), 2.99 (s, 12H; 4ꢂ-CH2-
NCH3SO2-), 2.92 (s, 24H; 8ꢂCH3-N-Dans), 2.40–2.34 ppm (m, 2H; O-
CH2-CH2-CN); 13C NMR (125 MHz, CDCl3): d=165.0, 158.1, 157.0,
156.9, 151.5, 149.4, 144.6, 135.7, 135.6, 134.2, 130.1, 130.0, 129.9, 129.4,
129.2, 129.0, 127.9, 127.7, 127.6, 127.4, 126.5, 123.1, 122.9, 122.5, 121.5,
119.5, 119.4, 116.7 and 115.0 (aromatic carbons), 116.1 and 116.0 (CN),
100.6 and 100.3 (2ꢂC-1), 85.9 and 85.7 (2ꢂC-2), 81.9 and 81.2 (2ꢂC-5),
73.8 (2ꢂC-3), 72.0 and 71.9 (4ꢂCH2-O-sugar), 70.2 and 69.7 (28ꢂO-CH2
TEG and, overlapped, 2ꢂC-4), 62.6 (C-6), 60.3 (-O-CH2-CH2-CN), 48.9
We thank the MIUR (PRIN) for grants in support of this investigation
and the Centro di Metodologie Chimico-Fisiche (CIMCF), Universitꢅ di
Napoli “Federico II”, for the MS and NMR spectroscopy facilities.
[1] For recent reviews, see for example: a) G. W. Gokel, M. M. Dasch-
bach, Coord. Chem. Rev. 2008, 252, 886–902; b) S. Licen, F. De Ric-
cardis, I. Izzo, P. Tecilla, Curr. Drug Discovery Technol. 2008, 5, 86–
97; c) T. M. Fyles, Chem. Soc. Rev. 2007, 36, 335–347; d) A. P. Davis,
D. N. Sheppard, B. D. Smith, Chem. Soc. Rev. 2007, 36, 348–357;
e) G. W. Gokel, I. A. Carasel, Chem. Soc. Rev. 2007, 36, 378–389;
f) R. S. Hector, M. S. Gin, Supramol. Chem. 2005, 17, 129–134.
[2] a) R. Dutzler, Curr. Opin. Struct. Biol. 2006, 16, 439–446; b) R.
MacKinnon, Angew. Chem. 2004, 116, 4363–4376; Angew. Chem.
Int. Ed. 2004, 43, 4265–4277.
[3] For recent reports, see for example: a) W. Wang, R. Li, G. W.
Gokel, Chem. Eur. J. 2009, 15, 10543–10553; b) N. Bernier, S. Car-
valho, F. Li, R. Delgado, V. Fꢆlix, J. Org. Chem. 2009, 74, 4819–
4827; c) I. Izzo, S. Licen, N. Maulucci, G. Autore, S. Marzocco, P. Te-
cilla, F. De Riccardis, Chem. Commun. 2008, 2986–2988; d) R. Fer-
dani, R. Li, R. Pajewski, J. Pajewska, R. K. Winter, G. W. Gokel,
Org. Biomol. Chem. 2007, 5, 2423–2432; e) V. Gorteau, G. Bollot, J.
Mareda, S. Matile, Org. Biomol. Chem. 2007, 5, 3000–3012.
[4] Essentials of Glycobiology (Eds.: A. Varki, R. Cumming, J. Esko, H.
Freeze, G. Hart, J. Marth), Cold Spring Harbor Laboratory Press,
Cold Spring Harbor, 1999, Chapter 1–21, pp. 1–331.
[5] For recent reports, see for example: a) J. M. Benito, M. Gꢇmez-
Garcꢈa, C. Ortiz Mellet, I. Baussanne, J. Defaye, J. M. Garcia
Fernꢉndez, J. Am. Chem. Soc. 2004, 126, 10355–10363; b) J. Ni, H.
Song, Y. Wang, N. M. Stamatos, L.-X. Wang, Bioconjugate Chem.
2006, 17, 493–500; c) S. G. Gouin, E. Vanquelef, J. M. Garcia
Fernꢉndez, C. Ortiz Mellet, F.-Y. Dupradeau, J. Kovensky, J. Org.
Chem. 2007, 72, 9032–9045; d) E. A. B. Kantchev, C.-C. Chang, S. F.
Cheng, A.-C. Roche, D.-K. Chang, Org. Biomol. Chem. 2008, 6,
1377–1385; e) X. Ning, J. Guo, M. A. Wolfert, G.-J. Boons, Angew.
Chem. 2008, 120, 2285–2287; Angew. Chem. Int. Ed. 2008, 47, 2253–
2255.
ꢁ
(2ꢂN CH2 TEG), 45.8, 45.7, 45.6, 45.4, 45.3 and 45.2 (4ꢂCH3-N-Dans),
35.7 and 35.6 (2ꢂ-CH2-NCH3SO2-), 22.9 ppm (-O-CH2-CH2-CN);
31P NMR (161.98 MHz, CDCl3): d=ꢁ2.1, ꢁ5.0, ꢁ9.5 and ꢁ10.6 ppm;
HRMS (MALDI+-TOF): m/z: calcd for C117H156ClN9O36P2S4: 2487.8700;
found: 2487.1564 [M+H]+, 2509.0653 [M+Na]+.
Synthesis of partially protected cyclic dimer 20: Compound 19 (22 mg,
0.009 mmol) was treated with piperidine (1.0 mL) and the resulting mix-
ture was left overnight with stirring at 708C. The reaction was quenched
by removal of the solvent in vacuo. The crude product was then purified
by column chromatography with elution with CHCl3 containing increas-
ing amounts of CH3OH (from 1 to 15%), affording pure 20 (21 mg,
0.0086 mmol) in 96% yields. Oil, Rf =0.3 (CHCl3/CH3OH 95:5 v/v);
1H NMR (400 MHz, CDCl3): d=8.52 (d, 3J
(Dans)), 8.33 (d, 3J
(H,H)=8.0 Hz, 4H; 4ꢂH
4A(Dans)), 7.49 (m, 8H; 4ꢂH3A(Dans) and 4ꢂH
plex signals, 18H; phenyl protons and 4ꢂH6A(Dans)), 5.16 and 4.90 (2ꢂm,
ACHTUNGTRENNUNG
A
R
CHTUNGTRENNUNG
H
N
N
CHTUNGTRENNUNG
CTHUNGTRENNUNG
each 1H; H-1 and H-1’), 4.40–3.35 (overlapped signals, 76H; 2ꢂH2-6, 2ꢂ
H-4, O-CH2-CH2-O TEG, 2ꢂH-2, 2ꢂH-3 and 2ꢂH-5), 2.98 (s, 12H; 4ꢂ
CH2-NCH3SO2-), 2.86 ppm (s, 24H; 8ꢂCH3-N-Dans); 13C NMR
(100 MHz, CDCl3): d=165.1, 158.0, 153.9, 142.4, 139.1, 134.9, 130.5,
129.4, 128.2, 124.8, 123.1, 122.6, 119.6, 117.2 and 115.1 (aromatic car-
bons), 101.4 and 101.1 (C-1 and C-1’), 84.7 (C-5 and C-5’), 75.9 (C-4 and
C-4’), 71.9, 70.3 and 69.8 (O-CH2-CH2-O TEG), 61.6 (C-6), 49.0 (2ꢂ
NꢁCH2 TEG), 45.3 and 45.2 (4ꢂCH3-N-Dans), 35.8 and 35.6 ppm (2ꢂ
CH2-NCH3SO2-); 31P NMR (161.98 MHz, CDCl3): two broad signals, cen-
tred at d=0.7 and ꢁ8.6 ppm; HRMS (MALDIꢁ-TOF): m/z: calcd for
C
114H153ClN8O36P2S4: 2434.8434; found: 2434.3841 [MꢁH]ꢁ.
Synthesis of fully deprotected cyclic dimer 21: Compound 20 (21 mg,
0.0086 mmol), dissolved in dioxane (200 mL), was treated with saturated
aq. LiOH solution (1.0 mL) and the resulting mixture was left overnight
with stirring at 708C. The reaction mixture was then concentrated under
reduced pressure, redissolved in CHCl3, transferred into a separating
funnel and washed three times with water. The organic phase was con-
[6] a) G. Di Fabio, A. Randazzo, J. DꢃOnofrio, C. Ausꢊn, A. Grandas, E.
Pedroso, L. De Napoli, D. Montesarchio, J. Org. Chem. 2006, 71,
3395–3408; b) J. D’Onofrio, C. Coppola, G. Di Fabio, L. De Napoli,
D. Montesarchio, Eur. J. Org. Chem. 2007, 3849–3858.
Chem. Eur. J. 2010, 16, 13757 – 13772
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
13771

D. Montesarchio et al.
[7] C. Coppola, V. Saggiomo, G. Di Fabio, L. De Napoli, D. Monte-
sarchio, J. Org. Chem. 2007, 72, 9679–9689.
[8] S. Licen, C. Coppola, J. D’Onofrio, D. Montesarchio, P. Tecilla, Org.
Biomol. Chem. 2009, 7, 1060–1063.
[9] For a recent review on fluorescent chemosensors, see: F. Mancin, E.
Rampazzo, P. Tecilla, U. Tonellato, Chem. Eur. J. 2006, 12, 1844–
1854.
overall yields. Its synthesis, achieved in 5 steps with 50% yield, and
use has been also more recently reported, in: C. Hçltke, A. von
Wallbrunn, K. Kopka, O. Schober, W. Heindel, M. Schaefers, C.
Bremer, Bioconjugate Chem. 2007, 18, 685–694; another remarkable
example of desymmetrized tetraethylene glycol linker is in: A. W.
Schwabacher, J. W. Lane, M. W. Schiesher, K. M. Leigh, C. W. John-
son, J. Org. Chem. 1998, 63, 1727–1729.
[10] For applications of dansyl groups in recognition/sensing systems, see
for example: a) B. Branchi, P. Ceroni, V. Balzani, G. Bergamini,
F. G. Klaerner, F. Voegtle, Chem. Eur. J. 2009, 15, 7876–7882; b) M.
Montalti, L. Prodi, N. Zaccheroni, G. Battistini, S. Marcuz, F.
Mancin, E. Rampazzo, U. Tonellato, Langmuir 2006, 22, 5877–5881;
c) S. Pagliari, R. Corradini, G. Galaverna, S. Sforza, A. Dossena, M.
Montalti, L. Prodi, N. Zaccheroni, R. Marchelli, Chem. Eur. J. 2004,
10, 2749–2758; d) A. Ueno, A. Ikeda, I. Ikeda, T. Ikeda, F. Toda, J.
Org. Chem. 1999, 64, 382–387.
[18] Y. Erkan, I. Czolkos, A. Jesorka, L. M. Wilhelmsson, O. Orwar,
Langmuir 2007, 23, 5259–5263.
[19] X. Wu, Y. Wu, L. Hai, Y. Bao, Lett. Org. Chem. 2006, 3, 911–914.
[20] H. Wang, W.-H. Chan, Tetrahedron 2007, 63, 8825–8830.
[21] F. Peri, F. Nicotra, C. P. Leslie, F. Micheli, P. Seneci, C. Marchioro, J.
Carbohydr. Chem. 2003, 22, 57–71.
[22] a) R. E. Stafford, T. Fanni, E. A. Dennis, Biochemistry 1989, 28,
5113–5120; b) P. Babu, D. Chopra, T. N. Guru Row, U. Maitra, Org.
Biomol. Chem. 2005, 3, 3695–3700.
[11] a) L. You, G. W. Gokel, Chem. Eur. J. 2008, 14, 5861–5870; b) L.
You, R. Li, G. W. Gokel, Org. Biomol. Chem. 2008, 6, 2914–2923;
c) W. M. Leevy, G. M. Donato, R. Ferdani, W. E. Goldman, P. H.
Schlesinger, G. W. Gokel, J. Am. Chem. Soc. 2002, 124, 9022–9023;
d) E. Abel, G. E. M. Maguire, O. Murillo, I. Suzuki, S. L. De Wall,
G. W. Gokel, J. Am. Chem. Soc. 1999, 121, 9043–9052; e) M. K.
Mathew, R. Nagaraj, P. Balaram, J. Biol. Chem. 1982, 257, 2170–
2176.
[12] For a recent review, see: M. Kçhn, R. Breinbauer, Angew. Chem.
2004, 116, 3168–3178; Angew. Chem. Int. Ed. 2004, 43, 3106–3116.
[13] For recent reports, see: a) H. C. Kolb, M. G. Finn, K. B. Sharpless,
Angew. Chem. 2001, 113, 2056–2075; Angew. Chem. Int. Ed. 2001,
40, 2004–2021; b) V. D. Bock, H. Hiemstra, J. H. van Maarseveen,
Eur. J. Org. Chem. 2006, 51–68.
[14] For recent synthetic elaborations of TEG-based linkers, see for ex-
ample: a) H. Herzner, H. Kunz, Carbohydr. Res. 2007, 342, 541–557;
b) H. T. Uyeda, I. L. Medintz, J. K. Jaiswal, S. M. Simon, H. Mat-
toussi, J. Am. Chem. Soc. 2005, 127, 3870–3878.
[15] For a recent example see W. Wang, A. D. Q. Li, Bioconjugate
Chem. 2007, 18, 1036–1052.
[23] a) M. Vaccaro, A. Accardo, D. Tesauro, G. Mangiapia, D. Loef, K.
Schillen, O. Soederman, G. Morelli, L. Paduano, Langmuir 2006, 22,
6635–6643; b) A. Vergara, L. Paduano, R. Sartorio, J. Phys. Chem.
B 2001, 105, 328–334; c) G. Mangiapia, C. Coppola, G. Vitiello, G.
DꢃErrico, L. De Napoli, A. Radulescu, D. Montesarchio, L. Padua-
no, unpublished results.
[24] N. Sakai, S. Matile, J. Phys. Org. Chem. 2006, 19, 452–460.
[25] D. D. Lasic, Liposomes from Physics to Applications, Elsevier, Am-
sterdam, 1993.
[26] C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, 3rd
ed., Wiley-VCH, Weinheim, 2003.
[27] A. K. Covington, F. S. Bates, R. A. Durst, Pure Appl. Chem. 1985,
57, 531–542.
[28] T. P. Russell, J. S. Lin, S. Spooner, G. D. Wignall, J. Appl. Crystallogr.
1988, 21, 629–638.
[29] G. D. Wignall, F. S. Bates, J. Appl. Crystallogr. 1987, 20, 28–40.
[30] B. J. Berne, R. Pecora, Dynamic Light Scattering with Applications
to Chemistry, Biology, and Physics, Dover, New York, 1975.
[31] G. A. Brehm, V. A. Bloomfield, Macromolecules 1975, 8, 663–665.
[16] S. Rumney IV, E. T. Kool, J. Am. Chem. Soc. 1995, 117, 5635–5646.
[17] This reagent has been previously described for the synthesis of a
mannose–fluorescein conjugate in: C. R. Bertozzi, M. D. Bednarski,
J. Org. Chem. 1991, 56, 4326–4329; the authors recovered it in 44%
Received: March 9, 2010
Revised: July 26, 2010
Published online: November 4, 2010
13772
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 13757 – 13772