Colorful methods to detect ion channels and pores: Intravesicular chromogenic probes that respond to pH, pM and covalent capture

Article abstract of DOI:10.1039/b900130a

The activity of synthetic pores, ion channels, transporters and carriers is usually determined with fluorescent probes in vesicles or by conductance measurements in planar lipid bilayers. Elaborating on more colorful alternatives, we here introduce 5(6)-c

Full text of DOI:10.1039/b900130a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Colorful methods to detect ion channels and pores: intravesicular chromogenic
probes that respond to pH, pM and covalent capture†
Sara M. Butterfield, Andreas Hennig and Stefan Matile*
Received 6th January 2009, Accepted 18th February 2009
First published as an Advance Article on the web 17th March 2009
DOI: 10.1039/b900130a
The activity of synthetic pores, ion channels, transporters and carriers is usually determined with
fluorescent probes in vesicles or by conductance measurements in planar lipid bilayers. Elaborating on
more colorful alternatives, we here introduce 5(6)-carboxynaphthofluorescein (CNF) as an
intravesicular pH probe for the colorimetric detection of activity, selectivity and cooperativity of ion
channels such as gramicidin A. We further report that intravesicular pyrocatechol violet (PV), together
with extravesicular Cu2+, extravesicular 4-carboxyphenylboronic acid (CBA) or intravesicular
4-(benzyl-N-glutamate)boronic acid (BGBA) can detect the activity of synthetic pores or
cell-penetrating peptide (CPP) sensors. Their response to analytes such as dodecylphosphate,
hyaluronan or IP6 are reported as high-contrast color changes from yellow to blue, from yellow to red,
or from red to green.
characterization of selected probes that respond to changes in pH
and pM as well as to covalent capture.
Introduction
In routine assays,1 the activity of synthetic pores, ion channels,
transporters and carriers2–16 is determined with a broad variety
Results and discussion
of fluorescent probes in vesicles or by conductance measure-
ments in planar (black) or supported lipid bilayer membranes.
These two complementary methods are most popular because
they are informative, sensitive and adaptable to probe specific
characteristics such as ion selectivity, voltage gating, ligand gating,
blockage, and so on.1 Several alternative methods exist as well. For
example, ion selective electrodes have been used in vesicle assays
despite limitations in sensitivity and adaptability. The use of NMR
spectroscopy has been explored in several variations. Particular
attention has been given to sodium NMR spectroscopy with shift
reagent additives to discriminate between intra- and extravesicular
sodium. Although appealing from a conceptual point of view,
this method is rather slow, inflexible, insensitive and inadaptable
to probe specific characteristics. Recently, G-quartets have been
introduced as ECCD (exciton-coupled circular dichroism) probes
for the detection of the activity of ion channels and pores by
circular dichroism spectroscopy.17,18
With the recently emerging sensing and screening
applications19–27 in diagnostics and drug discovery, there is
growing interest in the development of new methods in general
and colorimetric approaches in particular to detect the activity
of ion channels, transporters and pores with the naked eye.
This is particularly true for the recent introduction of synthetic
pores23 and cell-penetrating peptides27 as multianalyte sensors in
complex matrices. Colorimetric detection is also of interest for the
development of screening assays in drug discovery efforts to better
target membrane proteins. Here, we report the identification and
pH Indicators
The detection of the activity of ion channels with pH probes is
particularly attractive because they can report on transmembrane
H+/M+- and OH-/A--antiport in a very general manner.1 Loaded
into vesicles, the velocity of the decay of an applied pH gradient
in the presence and absence of an ion channel can be followed
easily and continuously. High selectivities do not hinder detection
because activity is not only observed for H+/M+- and OH-/A--
antiport but also for symport or simple probe export.1 Moreover,
the detection of important characteristics such as anion/cation
selectivity, ion selectivity sequences, voltage gating, ligand gating,
blockage, catalysis, Hill plots and the dependence on surface
potential, membrane composition, membrane fluidity, and so on is
unproblematic with pH probes.1 Only the detection of pH profiles
is not straightforward. The most popular fluorescent pH probe is
HPTS 1 (Fig. 1). With pKa ~7.3 and two complementary excitation
maxima, HPTS is perfect for ratiometric detection near neutral
water (Fig. 1). Other common fluorescent pH probes include CF
2, although alternative modes of operation are more common for
this fluorophore (pKa ~6.5).
From the rich collection of commercially available colori-
metric pH indicators, the usefulness of Congo red and 5(6)-
carboxynaphthofluorescein28–30 (CNF, 3, Fig. 1) to detect the
activity of ion channels was explored first. Congo red was dropped
quickly because loading into egg yolk phosphatidycholine large
unilamellar vesicles (EYPC LUVs) was problematic.
Consistent with the literature, CNF changed the color from
red to blue between pH 6 and pH 10 (Fig. 1). The emission
around 668 nm increased with increasing pH. The pH profile
confirmed a global pKa = 7.3 0.1 in water. CNF was entrapped
in EYPC LUVs using standard freeze–thaw–extrusion methods.
Department of Organic Chemistry, University of Geneva, Geneva,
Switzerland. E-mail: stefan.matile@unige.ch; Fax: +41 22 379 5123;
Tel: +41 22 379 6523; Web: www.unige.ch/sciences/chiorg/matile/
† This paper is dedicated to Professor Seiji Shinkai in recognition of his
contributions to supramolecular chemistry.
1784 | Org. Biomol. Chem., 2009, 7, 1784–1792
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Fig. 1 The absorption spectrum and the color (inset) of CNF in water
(24 mM) at pH 5.8 (dotted), 6.7, 7.1 8.4 and 9.1 (solid).
Extravesicular chromphores were removed by gel filtration. The
obtained EYPC-LUVs…CNF were bluish.
The absorption of CNF at 598 nm was monitored continuously
during the addition of first HCl, then gramicidin A (gA), and
finally triton X-100 (tX) to EYPC-LUVs…CNF (Fig. 2A). Insen-
sitivity toward the initial acid pulse confirmed that intravesicular
CNF is inaccessible for extravesicular protons (Fig. 2A and 2E).
Addition of gA, a classical model ion channel,18,31,32 caused a
decrease in blue absorption of CNF. This change was consistent
with intravesicular acidification by proton/cation antiport. Final
vesicle lysis with tX gave a pale pink color independent of sample
history (Fig. 2E). This final step was essential to calibrate different
experiments (Fig. 2A). The absorption at 598 nm after lysis was
higher that the minimal values obtained with gA. This apparent
influence of tX on optical probes is neither unique nor problematic.
Namely, different experiments were first calibrated to identical
maximal fractional absorption at the beginning of the experiment
and identical minimal fractional absorption at 598 nm after vesicle
lysis. In a second calibration step, all normalized curves were
then adjusted to A = 0 at the beginning and A = 1 for the
highest activity observed before lysis. This second recalibration
reduced the uniformized fractional absorption after lysis to A ~
0.45 (Fig. 2A).
Fig. 2 (A) Fractional absorption A (l 598 nm) during the addition of HCl
(9 ml, 0.3 M aq, 50 s), gA (from 3 mM DMSO, 7.5 nM (dotted), 40 (dashed),
and 80 nM (solid) final concentration) and tX (40 ml, 1.2% aq, 350 s) to
EYPC-LUVs…CNF (12 mM CNF, 10 mM Tris, 80 mM NaCl, pH 9.1) in
buffer (10 mM Tris, 107 mM NaCl, pH 9.1). (B) Same for the addition of
gA (60 nM, 50 s) before HCl (100 s) with extravesicular MCl, M = Cs, Rb
(dashed), K, Na, or Li (dotted). (C) Dependence of the fractional activity
Y (initial velocity of change in absorption or final absorption before lysis)
on the concentration of gA, with fit to Hill equation (from A). (D) Same
for the reciprocal radius of the cation M in the extravesicular buffer (from
B). (E) EYPC-LUVs…CNF after addition of HCl (left) and gA (right).
The colorimetric CNF assay was first used to record the
dependence of the activity on the concentration of gA (Fig. 2A).
The obtained dose response curve exhibited weak cooperativity
(Fig. 2C). A Hill coefficient n ~ 2 was consistent with the
endergonic self-assembly of gA into thermodynamically unstable
dimers.18,31,32
Interference from vesicle light scattering is intrinsically irrel-
evant for qualitative detection of color changes with the naked
eye. Eventual contributions to quantitative kinetics from light
scattering require consideration for single-wavelength kinetics,
whereas ratiometric double-channel kinetics are not affected.
LUVs with a diameter of 100 nm produce significant light
scattering at high energy only. Chromophores with red-shifted
absorption were selected for this study to avoid any possible
interference from light scattering.
The dependence of colorimetric CNF assay on the sequence
of addition was explored next (Fig. 2A vs. 2B). Addition of gA
before the acid pulse produced active channels that were naturally
invisible (Fig. 2B). Application of an acid pulse to these preformed
gA channels caused an instantaneous change in color (Fig. 2B).
This fast response to an acid pulse applied to preformed channels
differed clearly to the slow response that was observed when
channel formation was initiated after the acid pulse. This difference
confirmed that velocity of H+/M+-exchange through preformed
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1784–1792 | 1785
©

View Article Online
gA observed in the fast kinetics is far beyond the timescale of
this experiment (Fig. 2B). The slow kinetics observed for channel
formation after acid pulse reflects a slowly increasing number
of gA-containing vesicles that is affected by many parameters
including intravesicular transfer, self-assembly and ion selectivity
of the final channel (Fig. 2A). Identical observations concerning
sequence of addition have been made with the fluorometric pH
probe HPTS.33
The detectability of the ion selectivity of gA channels with
the colorimetric CNF assay was explored by extravesicular ion
exchange. This method has been used before with the fluoro-
metric HPTS assay.1,34 Moreover, qualitative compatibility of
the obtained values with permeability ratios from conductance
measurements in planar bilayers has been confirmed.34 With the
colorimetric CNF assay, extravesicular cation exchange affected
the apparent activity of gA channels in a meaningful manner
(Fig. 2B). The obtained Eisenman selectivity topology I with
a strong Li-anomaly was consistent with results from conduc-
tance experiments as well as fluorometric and chiroptical probes
(Fig. 2D).18,32
to the situation with protons, the detection of Cu2+ influx
appeared too invasive to study ion channels. However, PV export
and chromogenic binding to extravesicular Cu2+ appeared fully
appropriate to detect activity, activation or inactivation of anion
selective pores, anion transporters or simple vesicle destruction.
The colorimetric detection of pore activity was explored with
synthetic pores 5. These pores are formed by one of the best un-
derstood artificial b-barrels (Fig. 4).21,23,24,39 Their cylindrical self-
assembly is achieved by the interdigitation of peptide strands to
form short amphiphilic b-sheets that position the non-planar rigid-
rod p-octiphenyl scaffolds in close proximity.40,41 The sequence of
the short peptides is chosen to produce a hydrophobic barrel exte-
rior and a cationic, functionalized interior. The former is thought
to maximize interactions with the surrounding membrane, the
latter to stabilize internal space and to recognize and translocate
anionic analytes and probes such as PV.
pM Indicators
The use of cation complexing dyes to detect the activity of ion
channels and pores is generally underexplored.1 Initial screening
revealed pyrocatechol violet (PV, 4)35–38 as preferable compared
to the similarly functional pyrogallol red and arsenazo (III) for
various reasons (Fig. 3). PV has been used previously to determine
permeability35 and composition36 of lipid bilayer vesicles. More-
over, PV was of vital importance for the discovery of indicator
displacement assays.37,38 Cation screening for PV in neutral water
revealed the highest contrast for the binding of Cu2+, which
occurred with a drastic change in color from yellow to blue.
Fig. 4 Structure of pore 5 and pore inactivator 6. In the rigid-rod b-barrel
5, b-sheets are given as solid (backbone) and dotted lines (hydrogen bonds,
top) or as arrows (N→C, bottom); external amino-acid residues are dark
on white (L, leucine), internal residues are white on dark (K, lysine; H,
histidine).
To explore the detectability of synthetic pore 5 at work with
colorimetric pM indicators, PV was loaded into EYPC LUVs
using routine freeze–thaw–extrusion methods, and extravesicular
PV was removed by gel filtration. The obtained EYPC-LUVs…PV
were yellow (Fig. 5D). To detect pore activity with EYPC-
LUVs…PV, the change of the absorption at 612 nm was followed
continuously during the addition of first CuCl2, then pore 5, and
finally tX (Fig. 5A). Whereas the presence of CuCl2 alone passed
unnoticed by EYPC-LUVs…PV, the additional presence of pore 5
and tX was reported as a high-contrast color change from yellow
to blue (Fig. 5D).
As with pH probes, the sequence of addition mattered with
pM probes. The color change caused by Cu2+ addition after pore
addition was much faster than that caused by pore addition after
Cu2+ addition (Fig. 5B vs. 5A). The reasons for this drastic and
informative dependence of kinetics in pM assays on the sequence
of addition were as discussed above for the CNF assay. The strong
dependence of assay kinetics on the sequence of addition did,
however, not influence the observed final activities (Fig. 5A vs.
Fig. 3 The absorption spectrum of PV in water (24 mM) in the presence
of 0 (solid), 10 (broken), 50 (dashed) and 100 mM Cu2+ (dotted, 10 mM
Tris, 10 mM NaCl, pH 7.5). Inset: Color change of PV (100 mM) upon
addition of cations (300 mM, 10 mM Tris, 10 mM NaCl, pH 7.5).
The pM indicator assay was developed in analogy to pH
indicator assays, with intravesicular dyes meeting extravesicular
cations instead of protons to cause a change in color. Different
1786 | Org. Biomol. Chem., 2009, 7, 1784–1792
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
presence of increasing phytate concentrations gave an IC50 = 3.7
0.5 mM (Fig. 5C, ᭺). This finding suggested that pore sensors
that operate with the colorimetric PV/Cu assay would be about
two orders of magnitude less sensitive than pore sensors that
operate with the fluorometric CF assay. Similar losses in sensitivity
with the PV/Cu assay were observed for pore inactivation by
polyglutamate (IC50 = 92 30 mM) or hyaluronan (IC50 = 10
1 mg/ml).39 However, the poor responsiveness of the PV/Cu assay
was observed only when Cu2+ was added before the pore (Fig. 5A
and 5C, ᭺). This observation suggested that binding of Cu2+ to
the analytes reduces their affinity for the pore sensors. The poor
responsiveness of the PV/Cu assay should thus be improvable by
adding the pore before the Cu2+, that is, reversal of the sequence
of addition. This was found to be true. The efficiency of phytate 6
as inactivator of pore 5 improved almost 10-times to IC50 = 385
30 nM when Cu2+ was added after the pore (Fig. 5B and C, ᭹).
The compatibility of colorimetric PV/Cu assay with CPP-
counteranion transporters was explored next. When activated
by amphiphilic anions such as dodecylphosphate (DP), cell-
penetrating peptides (CPPs)42–51 such as polyarginine (pR) can
efficiently mediate the export of fluorogenic anions such as CF
from vesicles.52–56 This counteranion-mediated activity is of use
not only for cytosolic CPP delivery57 but also for the development
of cost-efficient multianalyte sensors, particularly for the otherwise
problematic hydrophobic analytes.22,27
Used as described in the previous section with synthetic pores
and their inactivators, the PV/Cu assay reported increasing pR
activity with increasing DP concentrations (Fig. 6A). Hill analysis
of the obtained dose response curve gave an EC50 = 6.4 1.3 mM
for pR activation by DP (Fig. 6B). This value was in the range of
the one determined for pR-DP complexes with the fluorogenic
CF assay (19.0
1.0 mM).22 This independence of activator
efficiencies on the assay system indicated that Cu2+ does not affect
the interaction between pR and DP significantly.
Fig. 5 (A) Fractional absorption A (l 612 nm) during the addition of
CuCl2 (1.25 mM final, 50 s), 5 (180, 375 and 750 nM monomer, 100 s) and
tX (40 ml, 1.2% aq, 350 s) to EYPC-LUVs…PV (2 mM PV, 10 mM Tris,
100 mM NaCl, pH 7.4) in buffer (10 mM Tris, 107 mM NaCl, pH 7.5).
(B) Same for addition of 5 (750 nM monomer, 0 s) before CuCl2 (313 mM,
240 s) in the presence of IP6 (varied, see C). (C) Dose response curve for
IP6 for sequence of addition IP6–5-Cu2+ (᭹, from B) and IP6-Cu2+-5 (᭺, as
in A). (D) EYPC-LUVs…PV after addition of Cu2+ (left) and tX (right).
5B, solid lines before 5 min). Identical observations concerning
the independence of thermodynamics but not kinetics on the
sequence of addition have been made and understood previously
with conventional assays.33
The dependence of pore activity on the concentration of pore
5 was correctly reported by the PV/Cu assay (Fig. 5A). An EC50
~400 nM was found (EC50 is the monomer concentration needed
to obtain 50% activity for the tetrameric pore 5). This value was
similar to values obtained previously with conventional assays.21,39
Because of a central importance for sensing applications,19–27 the
colorimetric detection of the activation and inactivation of stimuli-
responsive pores and transporters was of particular interest. In
practice, inactivator efficiency is best described as reciprocal IC50,
the inactivator concentration needed to reduce pore activity to
50%. Phytate 6 was selected as a representative analyte of broad
scientific importance.24 With the conventional fluorogenic CF
assay, phytate 6 inactivated pore 5 with an IC50 = 45 nM.24
Colorimetric detection of the changes in activity of pore 5 in the
Fig. 6 (A) Fractional change in absorption A (l 612 nm) during the
addition of DP (0–200 mM final, <0 sec), CuCl2 (313 mM, 50 s), pR
(350 nM, 100 s) and tX (40 ml, 1.2% aq, 350 s) to EYPC-LUVs…PV (2 mM
PV, 10 mM Tris, 100 mM NaCl, pH 7.5) in buffer (10 mM Tris, 107 mM
NaCl, pH 7.5). (B) Dose response curve for DP.
The colorimetric PV/Cu assay is necessarily incompatible with
the detection of pH profiles. Copper complexation by catechols
is highly pH dependent, and insoluble hydroxides form under
basic conditions. Both pH and copper concentration were kept
unchanged in all studies for these reasons. The fraction of the
released PV that is complexed by the excess of extravesicular
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1784–1792 | 1787
©

View Article Online
copper (1.25 mM) under these conditions was sufficient to cause
the desired change in color and was not further quantified.
Covalent capture
Pioneering work by the Shinkai group contributed substantially
to sensing applications of the covalent capture of catechols,
a-hydroxy acids and vicinal diols by boronic acids.25,26,58–69 Success-
ful studies in several groups focused particularly on carbohydrate60
and polyphenol25 binding, applications in biology (aptamers,
artificial amino acids)67,68 and materials sciences (porous solids)69
are currently emerging. Covalent capture by boronic acids has been
used to inactivate synthetic pores.25 However, reversible covalent
capture by boronic acids has so far not been explored for the
colorimetric detection of the activity of synthetic pores.
The color change of pyrocatechol violet (PV, 4) in response
to the reaction with boronic acids has been exploited in the
Anslyn group to introduce indicator displacement assays to
chemosensing.37,38,63,64 4-Carboxyphenylboronic acid (CBA, 7) was
selected as one of the simplest possible boronic acids with the
negative charge needed to minimize passive diffusion through lipid
bilayer membranes (Fig. 7). Addition of CBA to PV caused a high-
contrast change in color from yellow to red (Fig. 7C). The dose
response curve revealed an EC50 = 333 17 mM for dynamic
covalent capture of PV by CBA at pH 7.5 (Fig. 7B). This value
was as expected for the formation of boronate esters such as 8 with
a KA ª 3000 M-1 under these conditions.25,60,62
To use the chromogenic reaction between catechol 4 and boronic
acid 7 to detect pore activity, activation and inactivation with
the naked eye, different approaches were conceivable. However,
slow spontaneous release of intravesicular CBA was observed
during the preparation of vesicles in the presence or the absence
of co-entrapped PV. This finding established that in any practical
assay, the still relatively hydrophobic CBA would have to be added
extravesicularly to already made vesicles.
To use the chromogenic reaction between catechol 4 and boronic
acid 7 to detect the activity of pores, vesicles were thus loaded with
the unproblematic PV. The presence of extravesicular CBA above
the EC50 = 333 17 mM did not change the yellow color of EYPC-
LUVs…PV (Fig. 7C). The addition of pore 5 or tX, however, caused
a high-contrast color change from yellow to red (Fig. 7C). This
color change demonstrated the formation of boronic ester 8 after
either efflux of intravesicular PV or influx of extravesicular CBA
became possible. The change in color thus reflected the formation
of pore 5 (or more complex events such as vesicle destruction).
To validate the PV/CBA assay, changes in absorption at 525 nm
were detected continuously during the addition of synthetic pore 5
at different concentrations to EYPC-LUVs…PV in the presence of
extravesicular CBA (Fig. 8A). Hill analysis of the obtained dose
response curve gave an EC50 = 350 30 nM for the monomer
concentration needed to obtain 50% activity of the tetrameric
pore 5 (Fig. 8B, ᭹). Similarity of this value from the PV/CBA
assay with values from the PV/Cu assay (above) and conventional
assays21,39 suggested that PV, boronic acid 7 and boronate ester 8
do not affect formation and activity of pore 5.
Fig. 7 (A) The absorption spectrum of PV (100 mM) in buffer (10 mM
Tris, 100 mM NaCl, pH 7.5) in the presence of increasing concentrations of
CBA. (B) Absorption (l 525 nm) of PV as a function of the concentration
of CBA. (C) EYPC-LUVs…PV after addition of CBA (1 mM, left) and tX
(excess, right). Possible products of PV and CBA other than boronic ester
8 are not shown for clarity (e.g., conjugate bases and dimers).
᭺, IC50 = 3.7 0.5 mM). The interaction between pore 5 and
phytate 6 was thus less disturbed by extravesicular CBA than by
extravesicular Cu2+. In this example, the PV/CBA assay, based
on covalent capture, is clearly preferable compared to the more
interfering inorganic PV/Cu assay.
The colorimetric PV/CBA assay was not compatible with the
colorimetric detection of CPP-counteranion transporters. In the
presence of extravesicular CBA, pR transporters were active
already without DP counterion activators (Fig. 9B, ᭺). Appar-
ently, CBA can function as a pR activator. This interpretation
appeared reasonable considering the negative charge and relative
hydrophobicity of CBA. The slight increase in pR activity around
10 mM DP was consistent with activator exchange from CBA to
DP (Fig. 9B, ᭺). The effective DP concentration roughly matched
the EC50 = 6.4 1.3 mM found in the PV/Cu assay (Fig. 6B).
Decreasing pR activity at excess activator has been observed before
According to the PV/CBA assay, inactivation of pore 5 by
extravesicular phytate 6 occurred with an IC50 = 600 100 nM
(Fig. 8C). This inactivator efficiency was better than the one
obtained from the PV/Cu assay under similar conditions (Fig. 6C,
1788 | Org. Biomol. Chem., 2009, 7, 1784–1792
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Fig. 8 (A) Fractional absorption A (l 525 nm) during the addition of
pore 5 (c in B) and tX (excess 40 ml) to EYPC-LUVs…PV (20 mM PV,
10 mM Tris, 70 mM NaCl, pH 7.5) in buffer (1 mM CBA, 10 mM Tris,
100 mM NaCl, pH 7.5). (B) Hill analysis of (A), reported as Y as a function
of the monomer concentration c (>4-times c of the active tetramer, ᭹;
compared to results in EYPC-LUVs…PV/BGBA, ᭺; see Fig. 9). (C)
Dose response curve for IP6 at constant concentration of pore 5 (800 nM
monomer). (D) EYPC-LUVs…PV after addition of CBA and pore 5 in the
presence (left) and absence (right) of IP6 (200 mM).
(Fig. 9B, ᭺).27,52–54 It occurs in response to the self-assembly of the
amphiphilic DP monomers into hydrophilic DP micelles that bind
pR strongly but avoid the membrane. The cmc of DP should thus
be around ~200 mM.
pR-activation by the extravesicular boronic acid could be
avoided if the latter could be loaded together with PV as red
boronate ester 8 into the LUV. Interference-free extravesicular
formation of pR-DP transporters should then mediate the export
of intravesicular boronate ester 8, which would hydrolyze upon
dilution and afford the yellow PV together with the colorless
CBA. To replace the slightly membrane permeable CBA with
a more hydrophilic boronic acid, 4-(benzyl-N-glutamate)boronic
acid (BGBA) 9 was prepared by reductive amination of 4-
formylphenylboronic acid with glutamate. The preparation of
EYPC-LUVs…PV/BGBA was unproblematic, the obtained vesi-
cles were stable and could be used without any extra precaution.
Addition of either DP or pR did not change the rusty color of
intravesicular boronate ester 10 (Fig. 9A, solid; 9C, left). Addition
of both DP and pR (Fig. 9A, dashed, 9C, middle) and the addition
of tX (Fig. 9A, solid, 9C, left) caused a color change from rusty to
olive. The dose response curve for pR activation with DP revealed
the normal sigmoidal behavior centered at EC50 = 6.8 0.4 mM
(Fig. 9B, ᭹). This demonstrated that, different to extravesicular
CBA probes, intravesicular BGBA probes do not interfere with
the activity of pR-DP transporters.
Fig.
9
(A) The absorption spectrum of EYPC-LUVs…PV/BGBA
(200 mM) in buffer (1.8 ml, 10 mM Tris, 100 mM NaCl, pH 7.5) after the
addition of DP (solid; 10 mM final), pR (dashed; 350 nM mM final) and
excess tX (dotted). (B) Hill analysis of spectral changes as in A, measured at
l 600 nm and reported as Y as a function of DP concentration c at constant
pR concentration (350 nM, ᭹; compared to results in EYPC-LUVs…PV
and external CBA, ᭺; see Fig. 8). (C) EYPC-LUVs…PV/BGBA in the
presence of pR (left), DP and pR (middle), and tX (right).
pore formation in EYPC-LUVs…PV/BGBA exhibited, with a Hill
coefficient n = 4.1 0.2, higher apparent cooperativity than with
extravesicular CBA (n = 1.7 0.1). Much experimental evidence
from other studies support that pore 5 is a tetramer.39–41 The
difference found in cooperativity thus suggested that intravesicular
BGBA interferes less with the self-assembly of pore 5 than
extravesicular CBA. Many effects can naturally be imagined to
account for the more subtle influence of extravesicular CBA on
pore 5.70
Conclusions
For completion, the Hill plot of pore 5 was also recorded
EYPC-LUVs…PV/BGBA (Fig. 8B, ᭺). The colorimetric re-
sponse obtained with intravesicular BGBA was similar to that
with extravesicular CBA (Fig. 8B, ᭹) and other probes.39 However,
Four new methods to detect activity, activation, inactivation and
selectivity of ion channels, transporters and pores with the naked
eye are introduced. The four introduced methods use colorimetric
probes that respond to pH, pM and covalent capture.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1784–1792 | 1789
©

View Article Online
The CNF assay exhibits a color change from bluish to light
pink. In the CNF assay, carboxynaphthofluorescein is used as an
intravesicular pH probe to report on facilitated anion or cation
exchange in response to an applied pH gradient. The CNF assay
is the colorimetric version of the HPTS assay, that is arguably the
most popular fluorometric assay to study activity and selectivity
of ion channels.
The PV/Cu assay exhibits a color change from yellow to blue.
This color change occurs when intravesicular pyrocatechol violet
and extravesicular Cu2+ meet. Presumably less suitable for ion
channels, the PV/Cu assay is of use to characterize pores and
transporters and their response to chemical stimulation.
The PV/CBA assay exhibits a color change from yellow
to red. It occurs when intravesicular PV and extravesicular 4-
carboxyphenylboronic acid meet. Like the PV/Cu assay, the
PV/CBA assay is of use to characterize stimuli-responsive pores
and transporters.
precipitation. Addition of NaBH3CN (256 mg, 4.08 mmol) in
MeOH (2 ml) gave a clear solution. The reaction mixture was
stirred at room temperature. After ~1 h, precipitation of a colorless
solid could be observed. The reaction mixture was stirred for an
additional 48 h, while the pH was carefully maintained at pH
7 by stepwise addition of 1 M HCl (1.2 ml). Subsequently, the
precipitate is filtered through a Bu¨chner funnel and washed with
acetone. Drying under high vacuum yielded 412 mg (46%) of pure
9 as a colorless solid. Mp: 231–232 ◦C; IR (neat): 3190 (br), 2952
(w), 2923 (w), 1645 (m), 1631 (m), 1572 (s), 1471 (m), 1388 (s),
1210 (m); 1H NMR (400 MHz, D2O): 7.78 (d, 3J = 7.8 Hz, 2H),
7.46 (d, 3J = 8.1 Hz, 2H), 4.15–4.26 (m, 2H), 3.57–3.59 (m, 1H),
2.25–2.39 (m, 2H), 2.00–2.05 (m, 2H); 13C NMR (100 MHz, D2O):
181.3 (s), 173.5 (s), 134.2 (d), 133.3 (s), 129.2 (d), 62.0 (d), 50.1 (t),
33.9 (t), 26.0 (t); MS (ESI, MeOH–H2O 9 : 1): 280 (100, [M - H]-).
EYPC-LUVs…CNF
The related PV/BGBA assay exhibits a color change from red to
green. It occurs when the boronate ester formed between intraves-
icular PV and intravesicular BGBA hydrolyzes. This happens upon
probe dilution when leaving the vesicles. The PV/BGBA assay is
of use to characterize stimuli-responsive pores and transporters
and is less interfering than the PV/CBA assay.
The difference between PV/Cu and PV/BGBA (or PV/CBA)
assays is that the former operates with coordination chemistry
between Cu2+ and catechols, whereas the latter use covalent
capture between boronic acids and catechols. The PV/BGBA
assay appears more robust than the PV/Cu and PV/CBA assays
with regard to the use of pores as sensors because the intravesicular
boronic acid interacts less than extravesicular Cu2+ or boronic
acids with analytes such as phytate, activators such as dodecyl
phosphate, synthetic pores or CPP transporters.
Solutions of EYPC in CHCl3–MeOH 1 : 1 (25 ml, 1.0 g/ml)
were dried under vacuum (>2 h) to give a transparent thin
film. Hydration with “inside” buffer (12 mM CNF, 10 mM Tris,
80 mM NaCl, pH 9.1) for 1 h was followed by several freeze–
thaw cycles (5¥) and extrusions (>10¥, mini-extruder with a
stacked polycarbonate membrane of pore size 100 nm, Avanti).
External CNF was removed by gel filtration (Sephadex G-50)
with “outside” buffer (10 mM Tris, 107 mM MCl, pH 9.1; M =
Cs, Rb, K, Na, or Li). The LUV fractions were combined and
diluted to 6 ml. Final conditions: ~2.5 mM EYPC; inside: 12 mM
CNF, 10 mM Tris, 80 mM NaCl, pH 9.1; outside: 10 mM Tris,
107 mM MCl, pH 9.1; M = Cs, Rb, K, Na, or Li.
CNF assay
Taken together, the four introduced colorimetric assays operate
in a complementary manner to satisfy complementary needs. They
demonstrate that detection of activity, activation, inactivation
and selectivity of ion channels, transporters and pores with the
naked eye is possible and straightforward, identify a marvelous
topic to enjoy creative supramolecular chemistry and illustrate
much potential for future refinements toward the development of
advanced screening and sensing devices.
EYPC-LUVs…CF (750 ml) were added to gently stirred buffer in
a thermostated cuvette (1250 ml; 10 mM Tris, 107 mM MCl, pH
9.1; M = Cs, Rb, K, Na, or Li). Absorption At (l = 598 nm) was
monitored as a function of time (t) during the addition of HCl
(9 ml, 0.5 M aq, t = 50 or 100 s, external pH ~6.5), gA (c variable,
3 mM DMSO, t = 50 or 100 s) and tX (40 ml, 1.2% aq, t = 350 s).
CV/Cu assay
EYPC-LUVs…CV were prepared following above procedure for
EYPC-LUVs…CNF. Final conditions: ~2.5 mM EYPC; inside:
2 mM PV, 10 mM Tris, 100 mM NaCl, pH 7.5; outside: 10 mM
Tris, 107 mM NaCl, pH 7.5. EYPC-LUVs…CV (200 ml) were
added to gently stirred buffer in a thermostated cuvette (1800 ml;
10 mM Tris, 107 mM NaCl, pH 7.5, activator (DP; c variable)
or inactivator (IP6, pE, hyaluronan; c variable)). Absorption At
(l = 612 nm) was monitored as a function of time (t) during the
addition of CuCl2 (10 ml, 250 mM DMSO, t = 50 or 240 s), pore
5 (c variable, 37.5 mM DMSO, t = 100 or 0 s) or pR (350 nM aq,
t = 100 s) and tX (40 ml, 1.2% aq, t = 350 s).
Experimental part
Materials and methods
All compounds, probes, channels, transporters, analytes, buffers,
salts, detergents, solvents, etc, were commercially available from
standard suppliers except for pore 5, which was synthesized
following previously reported procedures.39 EYPC was from
Avanti Polar Lipids. UV-vis spectra were measured on a Varian
Cary 1 Bio spectrophotometer equipped with a magnetic stirrer
and a temperature controller (25 ◦C).
4-(Benzyl-N-glutamate)boronic acid (BGBA, 9)
CV/CBA assay
Glutamic acid (466 mg, 3.17 mmol) was suspended in H2O (8 ml),
and 10 M aqueous NaOH (0.32 ml, 3.20 mmol) was added to
yield a clear solution. Addition of 4-formylphenylboronic acid
(475 mg, 3.17 mmol) in MeOH (8 ml) at rt caused immediate
EYPC-LUVs…CV were prepared following above procedure for
EYPC-LUVs…CNF. Final conditions: ~2.5 mM EYPC; inside:
20 mM PV, 10 mM Tris, 70 mM NaCl, pH 7.5; outside: 10 mM
Tris, 100 mM NaCl, pH 7.5. EYPC-LUVs…CV (200 ml) were
1790 | Org. Biomol. Chem., 2009, 7, 1784–1792
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
added to gently stirred buffer in a thermostated cuvette (1800 ml;
1 mM CBA, 10 mM Tris, 100 mM NaCl, pH 7.5, activator (DP; c
variable) or inactivator (IP6; c variable)). Absorption At (l = 525
nm) was monitored as a function of time (t) during the addition
of pore 5 (c variable, t ~ 50 s) or pR (350 nM aq, t ~ 50 s) and tX
(40 ml, 1.2% aq, t ~ 160 s).
Acknowledgements
We thank D.-H. Tran for contributions to synthesis, D. Jeannerat,
A. Pinto and S. Grass for NMR measurements, and P. Perrottet,
N. Oudry and G. Hopfgartner for MS measurements. This work
was supported by the University of Geneva and the Swiss NSF.
References
CV/BGBA assay
1 S. Matile and N. Sakai, The Characterization of Synthetic Ion Channels
and Pores, in Analytical Methods in Supramolecular Chemistry, C. A.
Schalley, Ed., Wiley, Weinheim, 2007, 391–418.
2 W. H. Binder, Angew. Chem. Int. Ed., 2008, 47, 3092–3095.
3 P. Davis, D. N. Sheppard and B. D. Smith, Chem. Soc. Rev., 2007, 36,
348–357.
4 T. M. Fyles, Chem. Soc. Rev., 2007, 36, 335–347.
5 P. A. Gale, Acc. Chem. Res., 2006, 39, 465–475.
6 A. L. Sisson, M. R. Shah, S. Bhosale and S. Matile, Chem. Soc. Rev.,
2006, 35, 1269–1286.
7 U. Koert, L. Al-Momani and J. R. Pfeifer, Synthesis, 2004, 8, 1129–
1146.
8 S. Matile, A. Som and N. Sorde´, Tetrahedron, 2004, 60, 6405–6435.
9 G. W. Gokel and A. Mukhopadhyay, Chem. Soc. Rev., 2001, 30, 274–
286.
EYPC-LUVs…CV/BGBA were prepared following above proce-
dure for EYPC-LUVs…CNF. Final conditions: ~2.5 mM EYPC;
inside: 20 mM PV, 20 mM BGBA, 10 mM Tris, 40 mM NaCl,
pH 7.5; outside: 10 mM Tris, 100 mM NaCl, pH 7.5. EYPC-
LUVs…CV/BGBA (200 ml) were added to gently stirred buffer
in a thermostated cuvette (1800 ml; 10 mM Tris, 100 mM NaCl,
pH 7.5). Absorption spectra were monitored before and after the
addition of first activator (DP in DMSO; c variable) or inactivator
(IP6; c variable), then pR (350 nM aq) or pore 5 (c variable), and
finally tX (40 ml, 1.2% aq, t ~ 160 s).
Data analysis
10 A. Satake, M. Yamamura, M. Oda and Y. Kobuke, J. Am. Chem. Soc.,
2008, 130, 6314–6315.
Absorption kinetics were normalized to fractional absorption A
applying eqn (1)
11 L. Ma, M. Melegari, M. Colombini and J. T. Davis, J. Am. Chem. Soc.,
2008, 130, 2938–2939.
12 C. P. Wilson and S. J. Webb, Chem. Commun., 2008, 4007–4009.
13 I. Izzo, S. Licen, N. Maulucci, G. Autore, S. Marzocco, P. Tecilla and
F. De Riccardis, Chem. Commun., 2008, 2986–2988.
14 M. Jung, H. Kim, K. Baek and K. Kim, Angew. Chem. Int. Ed., 2008,
47, 5755–5757.
A = {(At - A0)/(A• - A0)}/{(AtMAX - A0)/(A• - A0)}
(1)
where A0 = At at channel addition, A• = At after final lysis, and
AtMAX = maximal At before final lysis. From the obtained curves,
fractional activities Y were determined with eqn (2)
15 X. Li, B. Shen, X. Q. Yao and D. Yang, J. Am. Chem. Soc., 2007, 129,
7264–7265.
16 P. L. Boudreault and N. Voyer, Org. Biomol. Chem., 2007, 5, 1459–1465.
17 N. Sakai and S. Matile, Chirality, 2003, 15, 766–771.
18 A. Hennig and S. Matile, Chirality, 2008, 20, 932–937.
19 S. Matile, H. Tanaka and S. Litvinchuk, Top. Curr. Chem., 2007, 277,
219–250.
20 G. Das, P. Talukdar and S. Matile, Science, 2002, 298, 1600–1602.
21 S. Litvinchuk, N. Sorde´ and S. Matile, J. Am. Chem. Soc., 2005, 127,
9316–9317.
Y = (YMAX - YMAX(0))/(YMAX(•) - YMAX(0)
)
(2)
where YMAX is A after a constant period of time after channel
addition (typically ~2 min), YMAX(0) is YMAX for lowest activity
and YMAX(•) is YMAX for the highest activity (Y = 1) for a given
experiment (e.g., YMAX(•) = YMAX with Cs+, YMAX(0) = YMAX without
gA in Fig. 3B). For Hill analysis of channels/pores, Y was plotted
against the monomer concentration cMONOMER and fitted to the
Hill eqn (3) to give the effective concentration EC50 and the Hill
coefficient n.
22 T. Miyatake, M. Nishihara and S. Matile, J. Am. Chem. Soc., 2006, 128,
12420–12421.
23 S. Litvinchuk, H. Tanaka, T. Miyatake, D. Pasini, T. Tanaka, G. Bollot,
J. Mareda and S. Matile, Nat. Mater, 2007, 6, 576–580.
24 S. M. Butterfield, D.-H. Tran, H. Zhang, G. D. Prestwich and S. Matile,
J. Am. Chem. Soc., 2008, 130, 3270–3271.
25 S. Hagihara, H. Tanaka and S. Matile, J. Am. Chem. Soc., 2008, 130,
Y = Y• + (Y0 - Y•)/{1 + (cMONOMER/EC50)n}
(3)
5656–5657.
26 S. Hagihara, H. Tanaka and S. Matile, Org. Biomol. Chem., 2008, 6,
2259–2262.
where Y0 is Y in the absence of channels/pores, Y• is Y with
excess channels/pores (e.g., Fig. 2B).
For Hill analysis of activators, Y was plotted against the
activator concentration cACTIVATOR and fitted to the Hill eqn (4)
to give the effective concentration EC50 and the Hill coefficient n.
27 S. M. Butterfield, T. Miyatake and S. Matile, Angew. Chem. Int. Ed.,
2009, 48, 325–328.
28 O. S. Wolfbeis, Mikrochim. Acta, 1992, 108, 133–141.
29 A. Song, S. Parus and R. Kopelman, Anal. Chem., 1997, 69, 863–
867.
30 M. Ramanathan and A. L. Simonian, Biosens. Bioelectr., 2007, 22,
Y = Y• + (Y0 - Y•)/{1 + (cACTIVATOR/EC50)n}
(4)
3001–3007.
31 A. M. O’Connell, R. E. Koeppe, II and O. S. Andersen, Science, 1990,
250, 1256–1259.
where Y0 is Y in the absence of activators, Y• is Y with excess
activators (e.g., Fig. 7B).
For Hill analysis of inactivators, Y was plotted against the
inactivator concentration cINACTIVATOR and fitted to the Hill eqn (5)
to give the inhibitory concentration IC50 and the Hill coefficient n.
32 A. Vescovi, A. Knoll and U. Koert, Org. Biomol. Chem., 2003, 1, 2259–
2262.
33 V. Gorteau, G. Bollot, J. Mareda and S. Matile, Org. Biomol. Chem.,
2007, 5, 3000–3012.
34 N. Sakai, D. Houdebert and S. Matile, Chem. Eur. J, 2003, 9, 223–232.
35 Y. Jin, M. Li and X. Hou, J. Pharm. Biomed. Anal., 2005, 37, 379–382.
36 R. G. Hanshaw, E. J. O’Neil, M. Foley, R. T. Carpenter and B. D.
Smith, J. Mater. Chem., 2005, 15, 2707–2713.
37 L. Zhu, Z. Zhong and E. V. Anslyn, J. Am. Chem. Soc., 2005, 127,
4260–4269.
38 L. Zhu and E. V. Anslyn, J. Am. Chem. Soc., 2004, 126, 3676–3677.
Y = Y• + (Y0 - Y•)/{1 + (cINACTIVATOR/IC50)n}
(5)
where Y0 is Y in the absence of inactivators, Y• is Y with excess
inactivators (e.g., Fig. 6C).
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 1784–1792 | 1791
©

View Article Online
39 S. Litvinchuk, G. Bollot, J. Mareda, A. Som, D. Ronan, M. R. Shah,
P. Perrottet, N. Sakai and S. Matile, J. Am. Chem. Soc., 2004, 126,
10067–10075.
54 M. Nishihara, F. Perret, T. Takeuchi, S. Futaki, A. N. Lazar, A. W.
Coleman, N. Sakai and S. Matile, Org. Biomol. Chem., 2005, 3, 1659–
1669.
40 N. Sakai, J. Mareda and S. Matile, Acc. Chem. Res, 2005, 38, 79–
55 N. Sakai, T. Takeuchi, S. Futaki and S. Matile, ChemBioChem, 2005,
87.
6, 114–122.
41 N. Sakai, J. Mareda and S. Matile, Acc. Chem. Res, 2008, 41, 1354–
1365.
56 A. Hennig, G. J. Gabriel, G. N. Tew and S. Matile, J. Am. Chem. Soc.,
2008, 130, 10338–10344.
57 T. Takeuchi, M. Kosuge, A. Tadokoro, Y. Sugiura, M. Nishi, M.
Kawata, N. Sakai, S. Matile and S. Futaki, ACS Chem. Biol., 2006,
1, 299–303.
58 T. D. James, K. R. A. S. Sandanayake and S. Shinkai, Angew. Chem.
Int. Ed., 1994, 33, 2207–2209.
59 T. D. James, K. R. A. S. Sandanayake and S. Shinkai, Nature, 1995,
374, 345–347.
60 T. D. James, Top. Curr. Chem., 2007, 277, 107–152.
61 J. Yoon and A. W. Czarnik, Bioorg. Med. Chem., 1993, 1, 267–271.
62 G. Springsteen and B. Wang, Tetrahedron, 2002, 58, 5291–5300.
63 N. Y. Edwards, T. W. Sager, J. T. McDevitt and E. V. Anslyn, J. Am.
Chem. Soc., 2007, 129, 13575–13583.
42 I. Nakase, T. Takeuchi, G. Tanaka and S. Futaki, Adv. Drug Del. Rev.,
2008, 60, 598–607.
43 K. M. Stewart, K. L. Horton and S. O. Kelley, Org. Biomol. Chem.,
2008, 6, 2242–2255.
44 A. Verma, O. Uzun, Y. H. Hu, Y. Hu, H. S. Han, N. Watson, S. L.
Chen, D. J. Irvine and F. Stellacci, Nat. Mater, 2008, 7, 588–595.
45 B. A. Smith, D. S. Daniels, A. E. Coplin, G. E. Jordan, L. M. McGregor
and A. Schepartz, J. Am. Chem. Soc., 2008, 130, 2948–2949.
46 E. K. Esbjorner, P. Lincoln and B. Norden, Biochem. Biophys. Acta,
2007, 1768, 1550–1558.
47 S. M. Fuchs and R. T. Raines, ACS Chem. Biol., 2007, 2, 167–
170.
48 T. Shimanouchi, P. Walde, J. Gardiner, Y. R. Mahajan, D. Seebach, A.
Thomae, S. D. Kraemer, M. Voser and R. Kuboi, Biochem. Biophys.
Acta, 2007, 1768, 2726–2736.
49 M. Tang, A. J. Waring and M. Hong, J. Am. Chem. Soc., 2007, 129,
11438–11446.
50 E. A. Goun, T. H. Pillow, L. R. Jones, J. B. Rothbard and P. A. Wender,
ChemBioChem, 2006, 7, 1497–1515.
51 S. Pujals, J. Fernandez-Carneado, C. Lopez-Iglesias, M. J. Kogan and
E. Giralt, Biochem. Biophys. Acta, 2006, 1758, 264–279.
52 N. Sakai and S. Matile, J. Am. Chem. Soc., 2003, 125, 14348–14356.
53 F. Perret, M. Nishihara, T. Takeuchi, S. Futaki, A. N. Lazar, A. W.
Coleman, N. Sakai and S. Matile, J. Am. Chem. Soc., 2005, 127, 1114–
1115.
64 S. L. Wiskur, J. J. Lavigne, A. Metzger, S. L. Tobey, V. Lynch and E. V.
Anslyn, Chem. Eur. J., 2008, 10, 3792–3804.
65 M. Maue and T. Schrader, Angew. Chem. Int. Ed., 2005, 44, 2265–
22670.
66 P. J. Duggan, T. A. Houston, M. J. Kiefel, S. M. Levonis, B. D. Smith
and M. L. Szydzik, Tetrahedron, 2008, 64, 7122–7126.
67 M. Li, N. Lin, Z. Huang, L. Du, C. Altier, H. Fang and B. Wang, J.
Am. Chem. Soc., 2008, 130, 12636–12638.
68 E. Brustad, M. L. Bushey, J. W. Lee, D. Groff, W. Liu and P. G. Schultz,
Angew. Chem. Int. Ed., 2008, 47, 8220–8223.
69 H. M. El-Kaderi, J. R. Hunt, J. L. Mendoza-Corte´s, A. P. Coˆte´, R. E.
Taylor, M. O’Keeffe and O. M. Yaghi, Science, 2007, 316, 268–272.
70 S. Bhosale and S. Matile, Chirality, 2006, 18, 849–856.
1792 | Org. Biomol. Chem., 2009, 7, 1784–1792
This journal is
The Royal Society of Chemistry 2009
©