Quantification of nitrate-π interactions and selective transport of nitrate using calix[4]pyrroles with two aromatic walls

Article abstract of DOI:10.1021/ja4021793

Herein we disclose the results of our investigations regarding the interactions between the biologically relevant nitrate oxoanion and several "two-wall" aryl-extended calix[4]pyrroles. There exists a clear relationship between the electronic nature of th

Full text of DOI:10.1021/ja4021793

Article
pubs.acs.org/JACS
Quantification of Nitrate−π Interactions and Selective Transport of
Nitrate Using Calix[4]pyrroles with Two Aromatic Walls
Louis Adriaenssens,† Carolina Estarellas,‡ Andreas Vargas Jentzsch,§ Marta Martinez Belmonte,†
Stefan Matile,*,§ and Pablo Ballester*,†,∥
†Institute of Chemical Research of Catalonia (ICIQ), Avda. Països Catalans 16, 43007 Tarragona, Spain
‡Department de Química, Universitat de les Illes Balears, crta. Valldemossa km 7.5, 07122, Palma de Mallorca, Spain
§Department of Organic Chemistry, University of Geneva, Geneva, Switzerland
∥Catalan Institution of Research and Advanced Studies (ICREA), Passeig Lluís Companys 23, 08010 Barcelona, Spain
S
* Supporting Information
ABSTRACT: Herein we disclose the results of our inves-
tigations regarding the interactions between the biologically
relevant nitrate oxoanion and several “two-wall” aryl-extended
calix[4]pyrroles. There exists a clear relationship between the
electronic nature of the aromatic walls of the calix[4]pyrroles
and the stability of the nitrate⊂calix[4]pyrrole complex. This
−
suggests that NO3 −π interactions have an important
electrostatic component. We provide energetic estimates for
the interaction of nitrate with several phenyl derivatives.
Additionally, we report solid-state evidence for a preferred
binding geometry of the nitrate anion included in the
calix[4]pyrroles. Finally, the “two-wall” aryl-extended calix[4]-
pyrroles show excellent activity in ion transport through lipid-
based lamellar membranes. Notably the best anion transporters are highly selective for transport of nitrate over other anions.
INTRODUCTION
■
In recent years, calix[4]pyrroles1 have been studied extensively as
effective receptors for ion pairs,2−4 fundamental components of
molecular switches,5−7 and efficient anion carriers through
lipophilic membranes.8−10 While in the majority of organic
solvents calix[4]pyrroles adopt the relatively nonpolar 1,3
alternate conformation, when introduced to guests capable of
acting as hydrogen-bond acceptors, the calix reorganizes itself
into the cone conformation so that all four pyrrole moieties can
participate as donors in hydrogen-bonding interactions with the
guest (Figure 1).11
When two opposite meso-carbons of the calix[4]pyrrole core
are substituted with aryl groups, “two-wall” aryl-extended
calix[4]pyrroles arise.6,12,13 In the cone conformation, the α,α-
isomer of the “two-wall” aryl-extended calix[4]pyrrole contains a
deep aromatic cleft suitable for the inclusion of mono- and
polyatomic anions.14 Anions bound to “two-wall” aryl-extended
calix[4]pyrroles are sandwiched between the two aromatic walls
rendering these receptors the perfect tools with which to explore
the energetics between anions and various aromatic systems.
Structurally related “four-wall” aryl-extended calix[4]-
pyrroles15,16 have already been used as a model system for the
quantitative evaluation of chloride−π interactions in solution.17
Attractive interactions between anions and π systems, so-
called anion−π interactions, have been extensively studied
computationally.18−23 Increasing experimental evidence for
Figure 1. (a) Molecular structures of octamethyl calix[4]pyrrole 1 and
the “two-wall” aryl-extended derivative α,α-2a. (b) Equilibirum between
the 1,3-alternate conformation of 1 and the cone conformation adopted
for the 1:1 complex with an anion.
these unorthodox interactions is accumulating.23−30 Experimen-
tal evidence based on solution and gas phase studies is especially
important. In the solid state, the close proximity of anions to π
Received: March 1, 2013
Published: May 14, 2013
© 2013 American Chemical Society
8324
dx.doi.org/10.1021/ja4021793 | J. Am. Chem. Soc. 2013, 135, 8324−8330

Journal of the American Chemical Society
Article
systems is more likely to be circumstantial and provoked by other
stronger interactions that dominate in the packing of the lattice.
In our groups, we have previously used “four-wall” aryl-extended
calix[4]pyrroles17 and naphthalenediimides25 to uncover
solution and gas-phase evidence, respectively, of both repulsive
and attractive anion−π interactions. So far, these studies have
focused solely on the spheroid shaped halide anions (Cl−, Br−,
and I−), and thus, we thought it was important to extend our
investigations to the interaction of polyatomic anions with
aromatic systems.27,31
In particular, we were intrigued by the trigonal planar shape of
the nitrate oxoanion and the consequent variety of possible
binding geometries for this anion within “two-wall” aryl-
extended calix[4]pyrroles. Furthermore, the nitrate anion plays
a pivotal role in various biological systems. Nitrate is one of the
most significant nutrients in photosynthesis and growth
representing the source of nitrogen in plant amino acid
production.32 In this context, Taylor et al.31 recently reported
experimentally determined estimates of the attractive interaction
of nitrate and other oxoanions with perfluoroaryl groups, in the
order of 0.3−0.5 kcal/mol. More recently, Wang and Wang have
disclosed an association constant value of 16 950 M−1 (ΔG =
−5.7 kcal/mol) for the 1:1 complex formed between a neutral
tetraoxacalix[2]arene[2]triazine receptor and nitrate in acetoni-
trile solution.27
multiple substitution. Something that is not possible with the
more sterically constrained “four-wall” analogues.17 This allows
for the interrogation of the highly π-acidic pentafluorophenyl and
dinitro-phenyl systems. Furthermore, “two-wall” calix[4]-
pyrroles offer the opportunity to evaluate the preferred
arrangement of the nitrate anion in relation to the aromatic
walls, i.e., nitrate anion oriented parallel or perpendicular to the
planes of the two meso-phenyl substituents.
We probed the interaction of nitrate with the series of “two-
wall” calix[4]pyrroles 2a−e using 1H NMR spectroscopy. Due to
solubility constraints and weak thermodynamic stability of the
complexes, the binding of nitrate to the receptor series 2 could
1
not be investigated accurately using calorimetric methods. H
NMR titrations had the advantage of providing structural
−
information about the binding geometry of the NO3 ⊂2
complexes. Tetrabutylammonium nitrate was used as the source
of nitrate anion, and the binding experiments were performed in
deuterated acetonitrile CD3CN. The selection of a polar solvent
like CD3CN was made for two principal reasons. First, in
acetonitrile, and at the concentrations used, the [Bu4N]+[NO3]−
ion pair can be considered as fully disassociated.35 Second, while
acetonitrile effectively solvates the cation, it is far less able to
solvate the anion. Thus, acetonitrile encourages a binding
process between calix[4]pyrroles and nitrate that is expected to
−
Herein, we report the results of our investigations of the
interactions between various “two-wall” aryl-extended calix[4]-
pyrroles and the biologically important nitrate oxoanion. The
obtained results reveal a clear relationship between the electronic
nature of the calix’s aromatic walls and the stability of the
produce 1:1 NO3 ⊂2 anionic complexes with a minimum of ion
pairing.
As increasing amounts of [Bu4N]+[NO3]− are added to
calix[4]pyrrole 2e, a shifting of the proton signals corresponding
to 2e is observed (Figure 3). The fact that one averaged set of
−
NO3 ⊂calix[4]pyrrole host−guest complex suggesting that
−
signals is seen for the free calix 2e and bound calix NO3 ⊂2e
nitrate−π interactions are dominated by electrostatics. Finally,
we show that “two-wall” aryl-extended calix[4]pyrroles are active
ion transporters. Significantly, “two-wall” calix[4]pyrroles
composed of the most π-acidic aromatic substituents show a
marked selectivity for the transport of nitrate over other anions,
representing very rare examples of small-molecule anion
transporters selective for the transport of nitrate.33
1
indicates that the binding process is fast on the H NMR time
scale. Furthermore, it is necessary to add an excess (4 equiv) of
RESULTS AND DISCUSSION
Nitrate Binding. We analyzed the complexation of nitrate
with a series of “two-wall” calix[4]pyrroles 2 (Figure 2). The
■
Figure 2. Molecular structures of the α,α-isomers of “two-wall” aryl-
extended calix[4]pyrroles 2 used in this study.
synthesis of the receptors was accomplished using two
subsequent acid-mediated condensations. The first using pyrrole
as the solvent and constructing the aryl substituted dipyrro-
methane and the second using acetone as solvent effecting
cyclodimerization to give receptors α,α-2a−e.34 We focused on
“two-wall” calix[4]pyrroles as their aromatic walls permit
Figure 3. (a) Binding equilibrium of calixpyrrole 2e with nitrate. (b)
Selected downfield regions of the 1H NMR spectra (400 MHz, CD3CN,
298 K) obtained during the titration of calix[4]pyrrole 2e (1.1 mM)
with incremental amounts of [Bu4N]+[NO3]−.
8325
dx.doi.org/10.1021/ja4021793 | J. Am. Chem. Soc. 2013, 135, 8324−8330

Journal of the American Chemical Society
Article
[Bu4N]+[NO3]− to induce saturation in the chemical shift
changes experienced by the proton signals of 2e.
Table 1. Association Constants (Ka), Free Energy Values
(ΔG) for the 1:1 Complexes of Calix[4]pyrroles 1, 2 and 3e
−
This observation indicates that the binding affinity of 2e for the
oxoanion is not very high. The direction in which the proton
signals of 2e shift during the titration is significant and gives
information regarding the binding geometry of the final host−
guest complex. After the addition of 4 equiv of [Bu4N]+[NO3]−,
the signal belonging to the pyrrole NH protons (Ha) has shifted
downfield to δ = 9.5 ppm (Δδ = 1.7 ppm), suggesting that
hydrogen-bonding interactions exist between the nitrate anion
and the four pyrrolic NH protons of 2e. The two signals
belonging to the β-pyrrole protons (Hb and Hc) shift in opposite
directions. The dissimilar chemical shift changes observed for
these protons suggest that calix[4]pyrrole 2e has undergone the
above-described conformational change, that is, from 1,3
alternate in the free state to cone in the host−guest complex
with NO3 , Statistically Corrected Free Energy Values (ΔΔG)
−
Calculated for the NO3 −π Interactions, Chemical Shifts
(δNH), and Complexation Induced Changes of Chemical Shift
(ΔδNH) for the Pyrrole NH
a
b
c
d
calix
R
Ka
ΔG
ΔΔG
δNH
ΔδNH
1
Me
Ph
60
30
−2.4
−2.0
−2.2
−2.9
−3.9
−4.3
−2.6
−
0.2
7.38
7.83
7.85
7.88
7.71
7.87
8.11
2.0
1.3
2.0
1.9
2.1
1.9
1.3
α,α-2a
α,α-2b
α,α-2c
α,α-2d
α,α-2e
α,β-3e
4-N3-Ph
40
0.2
4-NO2-Ph
C6F5
140
710
1550
85
−0.3
−0.7
−0.9
−
3,5-(NO2)2-Ph
3,5-(NO2)2-Ph
a
Association constants (M−1) measured in CD3CN at 298 K for 1:1
complexes of calixpyrroles 1, 2 and 3e with the nitrate anion. Values
NO3 ⊂2e.36 This places the anion deep within the aromatic cleft
−
b
1
determined by H NMR titrations. Estimated error is 20% kcal/
of 2e, sandwiched between the two aromatic meso-substituents.
Notably, the two signals belonging to protons in the dinitro-
substituted meso-aromatic walls (Hd and He) both shift upfield.
This observation is important as it suggests that the oxoanion
c
d
−
−⊂2
mol. ΔΔG = (ΔGNO
− ΔGNO ⊂1)/2. ΔδNH = complexation
3
3
induced changes (ppm) of the signal corresponding to the pyrrole NH
protons.
−
does not interact with the aromatic systems by C−H···NO3
(hydrogen bonding) interactions but rather, if any, via
Table 1 can be related to the free binding energies of the
interaction of the nitrate anion with a single aromatic wall. They
are simply calculated by dividing by two (statistical correction)
the difference in free energy between the complex of NO3− with a
interactions with the aromatic surface (anion-π). In short, in
−
the NO3 ⊂2 complex, the nitrate anion sits deep inside the
pocket formed by the two axially oriented aromatic groups and is
sandwiched between them, probably experiencing anion−π
given aryl-extended calix[4]pyrole 2 and that of the complex of
−
interactions. This binding mode suggests that the NO3 ⊂2
−
−⊂2
NO3 with octamethyl calix[4]pyrrole 1 (ΔGNO
−
3
complexes are ideal tools with which to investigate the energetic
relationships between the nitrate anion and π surfaces.
Quantitative assessment of the binding affinity of “two-wall”
−
ΔGNO ⊂1)/2. As stated before, in doing so, we consider receptor
3
1 to be a suitable reference for the quantification of the primary
hydrogen-bonding interaction that is operative in all the 1:1
complexes and that the value of this interaction is constant.
−
calix[4]pyrroles 2 for NO3 was achieved by fitting the data
obtained in the 1H NMR titrations using the HypNMR software
package.37 In all cases, we obtained a good fit to a theoretical 1:1
binding model, and the calculated binding constants ranged
between 101 and 103 M−1. As the calix[4]pyrrole core is
maintained constant in the receptor series 2, the differences in
binding energy can provide a direct measurement of the relative
interaction energy of nitrate with different aromatic π-systems. In
addition, the free energy of binding determined for the complex
−
Based on the calculated values of ΔΔG, complexes NO3 ⊂2a
−
and NO3 ⊂2b experience repulsive nitrate−π interactions. On
the other hand, 2c binds nitrate slightly stronger than octamethyl
calixpyrrole 1, and therefore, it can be concluded that there are
slightly attractive contributions from nitrate−π interactions in
−
the NO3 ⊂2c complex. Receptors 2d and 2e, with highly π-
acidic aromatic walls, show the existence of clearly attractive
−
−
of NO3 with octamethyl calix[4]pyrrole 1 provides an ideal
nitrate−π interactions in their NO3 complexes. A nice linear
reference for the component of binding free energy attributable
to hydrogen-bonding interactions. This reference value can then
be subtracted from the free energy of binding for nitrate with the
“two-wall” model systems to directly quantify the nitrate−π
interaction for the various π systems. The calculated association
constant values and the corresponding free energies of binding
correlation exists between the values determined for ΔΔG and
the calculated electrostatic surface potential (ESP) values
determined at the centroid of the respective aromatic walls
(Figure 4).39 This correlation is consistent with a nitrate-π
interaction mainly dominated by electrostatic factors.
X-ray diffraction analysis of single crystals of the complex
[Me4N]+[NO3]−⊂2e grown from acetonitrile solution revealed a
fascinating asymmetric unit composed of two distinct
nitrate⊂calix[4]pyrrole 2e complexes exhibiting different bind-
ing geometries (Figure 5). The binding geometry proposed
above for the structure of the complex in solution is supported by
the solid-state structures. Interestingly, in both solid-state
complexes, the nitrate anion binds with only one oxygen atom
to the four pyrrole NH protons of the calix[4]pyrrole core.
In one of the binding motifs, the nitrate anion is located almost
perpendicular to the calixpyrrole’s aromatic walls. The
tetramethyl ammonium counteranion is included into the
electron-rich aromatic cup, opposite to the bound nitrate,
provided by the calix[4]pyrrole core in cone conformation. In
short, receptor 2e acts as a separated ion-pair receptor. The
distances between the centroids of the dinitro aromatic walls and
the respective, most proximal nitrate oxygen atoms are 3.2 and
−
for the 1:1 NO3 ⊂2 complexes are summarized in Table 1.
The chemical shift data reported in Table 1 (δNH and ΔδNH
)
serve to illustrate that the strength of the primary interaction,
which is hydrogen bonding between the pyrrolic NHs of the
calix[4]pyrrole core and the nitrate anion, is not significantly
perturbed by the introduction of the two meso-phenyl
substituents or the modification of their substitution. Both the
values for the chemical shift of the NH proton in the free receptor
and the complexation-induced shift experienced by the NH
proton upon binding to nitrate are quite similar for all six
receptors 1 and 2a−e.38
For all complexes, the free energy of binding is attractive,
ranging from −2.0 kcal/mol for the complex of NO3− with the
electron-rich phenyl-substituted calix[4]pyrrole 2a to −4.3 kcal/
mol for the complex of NO3− with calixpyrrole 2e decorated with
3,5-dinitro phenyl aromatic walls. The ΔΔG values shown in
8326
dx.doi.org/10.1021/ja4021793 | J. Am. Chem. Soc. 2013, 135, 8324−8330

Journal of the American Chemical Society
Article
strong electrostatic interaction between the cation and the anion
dictates the arrangement of the nitrate anion in the binding
geometry of this complex. Thus, this structure is likely unreliable
as a judge of the lowest energy geometry for the anionic
−
NO3 ⊂2e complex. We suggest that the perpendicular geometry
−
is likely to be prevalent in a purely anionic NO3 ⊂2e complex in
solution, although the π,π-sandwich geometry cannot be fully
excluded.41
Ion Transport. The nitrate anion is one of the most
important nutrients in plant biology. As such, natural systems
have developed clever ways to sequester nitrate from the
environment. For example, marine blue-green algae display high
levels of the nitrate-selective solute binding protein NrtA on the
surface of their plasma membrane.32 NrtA is integrated with
three other proteins including a permease, and the complex
system of polypeptides selectively passes nitrate from the
extracellular environment through the membrane and into the
bacteria where it acts as the source of nitrogen in protein
synthesis.42 Comparatively simple small molecule anion trans-
porters have recently attracted significant attention.43−45 Among
other examples, calix[4]pyrroles9,10 and electron-poor aromatic
systems25 have proven to be effective at passing ions across lipid
membranes. In the case of electron-poor systems, anion−π
interactions have been postulated as being integral to anion
transport. Given that α,α “two-wall” aryl-extended calix[4]-
pyrroles 2 display aromatic walls of varied electronic nature that
seem to induce a range of repulsive and attractive anion-π
interactions, we were curious to test their activity in anion
transport. We were pleased to discover that α,α “two-wall” aryl-
extended calix[4]pyrroles are not only active in ion transport but
also that their activity is related to the electronic nature of their
aromatic walls. Most notably, the best ion transporters show a
pronounced selectivity for transport of the biologically relevant
nitrate anion.
Figure 4. Experimental values for the nitrate-π interaction derived from
the binding of nitrate with calix[4]pyrroles 2 correlate with the ESP
values calculated at the centroid of the aromatic walls.
The transport activity of calix[4]pyrroles was determined in
large unilamellar vesicles (LUVs) composed of egg yolk
phosphatidylcholine (EYPC) using two different fluorescent
probes.46 To probe ion transport activity, vesicles containing the
pH−sensitive fluorophore 8-hydroxy-1,3,6-pyrenetrisulfonate
(HPTS) were prepared. These vesicles were immersed in an
ionic medium and exposed to an extravesicular pulse of NaOH.
As the subsequent pH gradient between the extra- and
intravesicular media dissipated, a change of fluorescence by
HPTS⊂EYPC-LUVs was seen. This change can be attributed to
various ion transport mechanisms: H+/M+ antiport, OH−/A−
antiport, H+/A− symport, and OH−/M+ symport as well as
vesicle destruction. To show that fluorescence change is not due
to calix[4]pyrrole-induced vesicle rupture and subsequent HPTS
efflux, a carboxyfluorescein (CF) assay was also performed. In the
CF assay, vesicles were loaded with CF at concentrations high
enough to ensure self-quenching. Due to the size of CF, the most
likely mechanism by which CF can escape the vesicle and “turn
on” is by vesicle rupture. Therefore, CF⊂EYPC-LUVs report
vesicle destruction and CF efflux as fluorescence recovery.
Results from HPTS and CF assays were quantified with EC50
values, which were obtained from dose response curves. The
EC50 value is the ‘effective’ calix[4]pyrrole concentration needed
to reach 50% activity. A lower EC50 value for the HPTS assay
than the CF assay demonstrates that ion transport is more
efficient than dye export or vesicle destruction. In all cases the
calix[4]pyrroles performed far better in the HPTS assay than the
CF assay, clearly showing that the movement of ions between the
Figure 5. Side and top views of the two distinct binding geometries seen
in the solid-state structures of [Me4N]+[NO3]−⊂2e complexes. Solvent
molecules and nonpolar hydrogen atoms of calixpyrrole 2e are omitted
for clarity. The Me4N+ cation is omitted from top view of the
perpendicular binding geometry, also for clarity. [Me4N]+[NO3]− is
shown in CPK model and 2e in stick representation.
3.3 Å. Moreover, the nitrate anion’s oxygen atom that is closest to
its respective aromatic wall is located just 3.0 Å from the carbon
atom directly attached to the nitro substituent (Figure 5, top
right). This distance is shorter than the sum of the van der Waal
radii of the atoms (C 1.70 and O 1.52 Å) and suggests the
existence of a weak sigma interaction instead of pure anion-π.40
The analogous distance on the other side of the complex is only
3.25 Å.
In the other binding motif, the cation is positioned to the side
of the nitrate anion and pushes the two aromatic walls apart
destroying their parallel relationship. In this second binding
geometry the nitrate anion is situated pseudoparallel to the
calix[4]pyrrole’s aromatic walls. The distances between the
nitrate oxygen atoms and the closest aromatic carbons, those
holding nitro substituents, in opposite phenyl rings are 3.20 and
3.33 Å, very close to the sum of van der Waals radii. In this case,
2e functions as a close-contact ion-pair receptor. Most likely, the
8327
dx.doi.org/10.1021/ja4021793 | J. Am. Chem. Soc. 2013, 135, 8324−8330

Journal of the American Chemical Society
Article
a
Table 2. Summary of Transport Data
b
b
b
c
c
c
entry
calixpyrrole
HPTS (NaCl) EC50 (nM)
HPTS (CsCl) EC50 (nM)
CF (NaCl) EC50 (nM)
Cs+/Na+
Cl−/Br−
NO3−/Cl−
1
2
3
4
5
2a
2c
2e
1
350 20
8.4 0.3
2.0 0.4
960 300
9.7 0.4
7.3 0.3
0.73 0.06
4.5 0.3
7.0 2.0
1300 100
1600 500
10 000 2000
>50 000
2.8
1.2
1.0
Nd
0.9
0.9
1.1
1.2
Nd
1.5
0.8
1.4
1.7
Nd
0.4
3e
11.0 1.0
>50 000
a
b
c
For original data, see SI. Effective calix[4]pyrrole concentration needed to reach 50% activity in the assay. Relative activity in the HPTS assay with
different extravesicular ions: external cesium (CsCl) compared with external sodium (NaCl), external chloride (NaCl) compared with external
bromide (NaBr), external nitrate (NaNO3) compared with external chloride (NaCl).
extra- and intravesicular environments was due to ion transport
and not vesicle rupture.
In the initial tests a HEPES buffered NaCl solution was used as
the extravesicular medium. Calix[4]pyrroles 1, 2a, 2c, and 2e
were tested in the above-described assay, and it was found that
they all promoted ion transport.47 In particular, calix[4]pyrroles
2c and 2e (Table 2, entries 2 and 3) were the most active,
showing low nM EC50 values. Phenyl-substituted 2a (Table 2,
entry 1) was less active, while octamethyl calix[4]pyrrole 1
(Table 2, entry 4) performed worst of all with an EC50 of 1 μM.
Octamethyl calix[4]pyrrole 1 is reported to transport ions via an
anion/cation symport mechanism that is highly selective for the
transport of Cs+ salts.10 Therefore, to make a fair comparison
between octamethyl calix[4]pyrrole 1 and “two-wall” calix[4]-
pyrroles 2 and to check the “two-wall” systems for cation
selectivity, we performed the same tests using HEPES buffered
CsCl in place of NaCl as the extravesicular medium. As expected,
the activity of 1 increased dramatically (EC50 = 4.5 nM). The
activity of phenyl-substituted 2a also increased significantly,
while the activities of 2c and 2e only increased moderately
(Table 2). This suggests that the cation is pivotal for transport
mediated by calix[4]pyrrole 2a and that, like octamethylcalix[4]-
pyrrole 1, “two-wall” calix[4]pyrrole 2a can transport both
cations and anions via an anion/cation symport mechanism.
Meanwhile, ion transport mediated by calix[4]pyrroles 2c and 2e
seems to be independent of the cation used (Na+, Cs+, Li+, K+,
Rb+), suggesting that cations are not significantly involved in
transport and that an anion/anion antiport mechanism is the
more likely predominant process in these cases (Figure 6, left
side). Notably 3,5-dinitro phenyl substituted 2e is significantly
more active in our assay than octamethyl calix[4]pyrrole 1.
Whether functioning via anion/cation symport (1 and 2a) or
anion/anion antiport (2c and 2e), the manner in which the
calix[4]pyrroles transport the ions is almost certainly a carrier
mechanism. Such a mechanism has been proposed previously for
octamethyl calix[4]pyrrole 110 and seems to be likely for aryl-
extended calix[4]pyrroles 2 as well. Certainly, the span and
nature of the membrane in combination with the dimensions,
electrostatic natures, and shapes of the calix[4]pyrroles mean
that the formation of a channel is unlikely.
Figure 6. Transport selectivity. Dependence of the fractional transport
activity Y of calixpyrroles 2a (panel A), 2c (panel B), and 2e (panel C)
on different external cations ([M]+[Cl]−) and anions ([Na]+[A]−) in
the HPTS ion transport assay. Y has been normalized to 1 for Cl− and
Na+.
electropositive 2e to the electron-rich phenyl substituted 2a, the
activities steadily decrease. These results nicely match with our
previous study, where it was shown that the activity of various
naphthalenediimides in anion transport increases as their π
surface becomes more and more electropositive.25
We tested the calix[4]pyrroles 2 for anion transport activity
with a variety of extravesicular anions: F−, AcO−, Cl−, Br−, I−,
NO3 , and ClO4 (Table 2; Figure 6, right side).48 For all
calix[4]pyrroles tested, iodide and perchlorate were the most
difficult anions to transport. This is likely the result of a reduced
complementarity between the size and shape of these anions and
the size and shape of the binding pocket of the calix[4]pyrroles.
Little preference was seen for transport of bromide or chloride.
Most interesting to us was that the best transport agents 2c and
especially 2e show a pronounced preference for the transport of
nitrate. The emergence of a weak preference for nitrate has been
observed previously for naphthalenediimide transporters with
optimized anion−π interactions.25 In sharp contrast to these
−
−
In the absence of the preferred cesium cation, calix[4]pyrrole
mediated anion/cation symport or cation/cation antiport should
be minimized. Under these circumstances anion/anion antiport
is expected to predominate. In this way we hoped to reduce the
role of the cation, allowing us to better relate activity to the
electronic nature of the calix[4]pyrrole’s aromatic walls and the
interactions of these walls with anions. Indeed, the results
obtained while using a HEPES buffered NaCl solution as the
extravesicular medium show a clear relationship between the ESP
values of the aromatic walls and the activity in ion transport
(entries 1−3, third column in Table 2). Going from the most
8328
dx.doi.org/10.1021/ja4021793 | J. Am. Chem. Soc. 2013, 135, 8324−8330

Journal of the American Chemical Society
Article
weak effects, the nitrate selectivity found for 2e is significant.
Given the stability of host−guest complexes between calix[4]-
pyrroles 2 and anions in acetonitrile follows the trend Cl− ≫ Br−
> NO3− > I−, this result caught us by surprise.49
Despite the fact that differences between binding and
transport selectivity are not unusual,50 the preferred transport
of nitrate by 2c and 2e seemed more of an aberration than the
consequence of the typical bell-shaped dependence of transport
on binding affinity,46 which can be seen in the slight preference
shown by calixpyrrole 2a for the transport of bromide (Figure 6,
panels B and C vs panel A).
It is likely that the complementary geometry that exists
between the nitrate anion and two-wall calixpyrroles 2c and 2e is
the key factor responsible for the observed transport selectivity.
As can be seen in the X-ray structure, the nitrate anion is able to
satisfy its three centers of charge when bound to the two-wall
calix[4]pyrrole host. In cases where the nitrate−π interaction is
favorable (2c and 2e), the walls of the calix[4]pyrrole in
combination with the tetrapyrrole core create a pocket perfectly
functionalized to protect and stabilize the nitrate anion as it
passes through the membrane, in effect diffusing the clash of
polarities that exists between the charged anion and the nonpolar
intermembrane space.51
walls constituting the calix[4]pyrrole’s binding pocket. The more
electron poor the aromatic walls, the stronger the binding. This
points to interactions, both repulsive and attractive, between the
nitrate anion and π systems and shows that this nitrate−π
interaction is highly dependent on electrostatics. We show
experimental evidence for the preferred binding geometry of the
nitrate anion within calix[4]pyrrole 2e. One oxygen atom of the
nitrate anion accepts four hydrogen bonds from the tetrapyrrolic
core of the calix[4]pyrrole, while the other two nitrate-oxygen
atoms interact with the π surface of the calix[4]pyrrole’s aromatic
walls. Finally, we have shown the stereoisomeric aryl-extended
calix[4]pyrroles α,α-2 and α,β-3 are active ion transport agents.
Their activity correlates well to the electronic nature of their
aromatic walls, giving credence to the importance of anion−π
interactions in anion transport. The most active ion transporters,
2c and 2e, exhibit a remarkable selectivity for the transport of
nitrate anions over all other anions tested. This is likely a
consequence of the complementary geometry that exists
between the nitrate and the “two-wall” α,α-calix[4]pyrrole’s
binding site and is especially notable given the importance of the
nitrate anion to a variety of biological processes.
ASSOCIATED CONTENT
* Supporting Information
■
During the synthesis of two-wall α,α calix[4]pyrroles 2 the α,β
calix[4]pyrroles 3 with one aromatic wall above and one aromatic
wall below their tetrapyrrolic core are produced as byproducts
and also isolated (Figure 7). In these cases, only one wall can
S
General experimental methods for the transport experiments,
synthesis and binding studies, spectral data for new compounds
and two X-ray crystallographic files of [Me4N]+[NO3]−⊂2e.
This material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
stefan.matile@unige.ch; pballester@iciq.es
■
Notes
The authors declare no competing financial interest.
Figure 7. Schematic representation of nitrate bound to two-wall α,β
calix[4]pyrrole 3e.
ACKNOWLEDGMENTS
We thank Spanish Ministerio de Economıa y Competitividad
■
́
(CTQ2011-23014), Generalitat de Catalunya
(2009SGR00686), ICIQ Foundation, the University of Geneva,
the European Research Council (ERC Advanced Investigator),
the National Centre of Competence in Research (NCCR)
Chemical Biology, and the Swiss NSF for funding.
protect the anion from the intermembrane space during
transport. The manner in which this configurational distinction
affects the ability of α,β calix[4]pyrroles 3 to act as anion
transporters intrigued us. First, using 1H NMR titration
experiments we determined the stability constant value for the
NO3 ⊂3e complex to be 85 M−1 (ΔG= −2.6 kcal/mol). This
−
REFERENCES
■
value is slightly smaller than would be expected for the sum of the
interaction with the pyrrole core and a single nitrate−π
interaction (ΔGexpected = ΔG(1) + ΔΔG(2e) = −3.3 kcal/
mol). Most likely, the pseudoequatorial orientation that must be
adopted by one of the dinitro-phenyl substituents in the cone
(1) Gale, P. A.; Sessler, J. L.; Kral, V.; Lynch, V. J. Am. Chem. Soc. 1996,
118, 5140−5141.
(2) Park, I. W.; Yoo, J.; Adhikari, S.; Park, J. S.; Sessler, J. L.; Lee, C. H.
Chem.Eur. J. 2012, 18, 15073−15078.
́
(3) Ciardi, M.; Tancini, F.; Gil-Ramírez, G.; Escudero Adan, E. C.;
−
conformation of the NO3 ⊂3e complex provokes an energetic
Massera, C.; Dalcanale, E.; Ballester, P. J. Am. Chem. Soc. 2012, 134,
13121−13132.
penalty. Next, we tested α,β calix[4]pyrrole 3e in various HPTS
and CF assays (Table 2, entry 5). As anticipated, the ion
transport activity is reduced compared to the α,α isomer 2e, but it
is still quite significant. Notably, the selectivity for transport of
nitrate completely disappeared. Most probably, the presence of
only one wall to protect the anion during transport is responsible
for both of these observations.51
(4) Custelcean, R.; Delmau, L. H.; Moyer, B. A.; Sessler, J. L.; Cho, W.
S.; Gross, D.; Bates, G. W.; Brooks, S. J.; Light, M. E.; Gale, P. A. Angew.
Chem., Int. Ed. 2005, 44, 2537−2542.
(5) Park, J. S.; Karnas, E.; Ohkubo, K.; Chen, P.; Kadish, K. M.;
Fukuzumi, S.; Bielawski, C. W.; Hudnall, T. W.; Lynch, V. M.; Sessler, J.
L. Science 2010, 329, 1324−1327.
(6) Park, J. S.; Yoon, K. Y.; Kim, D. S.; Lynch, V. M.; Bielawski, C. W.;
Johnston, K. P.; Sessler, J. L. Proc. Natl. Acad. Sci. U.S.A. 2011, 108,
20913−20917.
(7) Park, J. S.; Bejger, C.; Larsen, K. R.; Nielsen, K. A.; Jana, A.; Lynch,
V. M.; Jeppesen, J. O.; Kim, D.; Sessler, J. L. Chem. Sci. 2012, 3, 2685−
2689.
CONCLUSION
■
In conclusion, we have shown that the strength with which the
nitrate anion binds with two-wall aryl-extended calix[4]pyrroles
2 is highly dependent on the electronic nature of the aromatic
8329
dx.doi.org/10.1021/ja4021793 | J. Am. Chem. Soc. 2013, 135, 8324−8330

Journal of the American Chemical Society
Article
(8) Park, I. W.; Yoo, J.; Kim, B.; Adhikari, S.; Kim, S. K.; Yeon, Y.;
Haynes, C. J. E.; Sutton, J. L.; Tong, C. C.; Lynch, V. M.; Sessler, J. L.;
Gale, P. A.; Lee, C. H. Chem.Eur. J. 2012, 18, 2514−2523.
(9) Gale, P. A.; Tong, C. C.; Haynes, C. J. E.; Adeosun, O.; Gross, D. E.;
Karnas, E.; Sedenberg, E. M.; Quesada, R.; Sessler, J. L. J. Am. Chem. Soc.
2010, 132, 3240−3241.
(38) Calculations of the electrostatic potential values at the nitrogen
nuclei in the pyrrole units of phenyl substituted dipyromethanes hint to
a decrease of two pKa units for the pyrrolic NHs of 2e with respect to 2a.
Liu, S. B.; Pedersen, L. G. J. Phys. Chem. A 2009, 113, 3648−3655.
According to the pKa slide rule introduced by Gilli et al. such change in
the difference of donor−acceptor acidities, ΔpKa ∼ 16−18, should have
a reduced impact in the strength of hydrogen bonding between nitrate
and the calix[4]pyrrole core for the receptor series studied in this paper.
See SI for details and Gilli, P.; Pretto, L.; Bertolasi, V.; Gilli, G. Acc. Chem.
Res. 2009, 42, 33−44 . These findings in combination with the 1H NMR
data discussed in the text provide strong support to the claim that the
primary hydrogen-bonding interaction of the calix[4]pyrrole core is not
significantly perturbed by the introduction of aromatic walls or their
modification by different substituent groups.
(10) Tong, C. C.; Quesada, R.; Sessler, J. L.; Gale, P. A. Chem. Commun.
2008, 6321−6323.
(11) Blas, J. R.; Lopez-Bes, J. M.; Marquez, M.; Sessler, J. L.; Luque, F.
J.; Orozco, M. Chem.Eur. J. 2007, 13, 1108−1116.
(12) Bruno, G.; Cafeo, G.; Kohnke, F. H.; Nicolo, F. Tetrahedron 2007,
63, 10003−10010.
(13) Cafeo, G.; Kohnke, F. H.; Valenti, L.; White, A. J. P. Chem.Eur.
J. 2008, 14, 11593−11600.
(39) ESP values determined at the BP86/def2-TZVP level of theory.
(40) For a previous example of related weak σ interaction between
trinitro-substituted aromatics and halides see: Chiavarino, B.; Crestoni,
M. E.; Fornarini, S.; Lanucara, F.; Lemaire, J.; Maitre, P.; Scuderi, D.
Chem.Eur. J. 2009, 15, 8185−8195.
(14) Sokkalingam, P.; Kee, S.-Y.; Kim, Y.; Kim, S.-J.; Lee, P. H.; Lee, C.-
H. Org. Lett. 2012, 14, 6234−6237.
(15) Anzenbacher, P.; Jursikova, K.; Lynch, V. M.; Gale, P. A.; Sessler, J.
L. J. Am. Chem. Soc. 1999, 121, 11020−11021.
(16) Bonomo, L.; Solari, E.; Toraman, G.; Scopelliti, R.; Latronico, M.;
Floriani, C. Chem. Commun. 1999, 2413−2414.
(41) In addition to the structurally rich X-ray results shown in Figure 5,
other crystallization conditions furnished solid−state structures for the
[Me4N]+[NO3]−⊂2e complex where only the perpendicular binding
geometry is seen. See SI for details. We are not aware of experimental or
computational reports discussing in detail the various interaction
geometries and forces of nitrate−π interactions. For combined
crystallographic and theoretical studies of nitrate−π interactions in
heterocyclic systems see: (a) Maheswari, P. U.; Modec, B.; Pevec, A.;
Kozlevcar, B.; Massera, C.; Gamez, P.; Reedijk, J. Inorg. Chem. 2006, 45,
6637−6645. (b) Garcia-Raso, A.; Alberti, F. M.; Fiol, J. J.; Tasada, A.;
Barcelo-Oliver, M.; Molins, E.; Estarellas, C.; Frontera, A.; Quinonero,
D.; Deya, P. M. Cryst. Growth Des. 2009, 9, 2363−2376.
(17) Gil-Ramirez, G.; Escudero-Adan, E. C.; Benet-Buchholz, J.;
Ballester, P. Angew. Chem., Int. Ed. 2008, 47, 4114−4118.
(18) Mascal, M.; Armstrong, A.; Bartberger, M. D. J. Am. Chem. Soc.
2002, 124, 6274−6276.
(19) Quinonero, D.; Garau, C.; Rotger, C.; Frontera, A.; Ballester, P.;
Costa, A.; Deya, P. M. Angew. Chem., Int. Ed. 2002, 41, 3389−3392.
(20) Alkorta, I.; Rozas, I.; Elguero, J. J. Am. Chem. Soc. 2002, 124,
8593−8598.
(21) Schottel, B. L.; Chifotides, H. T.; Dunbar, K. R. Chem. Soc. Rev.
2008, 37, 68−83.
(22) Gamez, P.; Mooibroek, T. J.; Teat, S. J.; Reedijk, J. Acc. Chem. Res.
2007, 40, 435−444.
(42) Omata, T. Plant Cell Physiol. 1995, 36, 207−213.
(43) Davis, A. P.; Sheppard, D. N.; Smith, B. D. Chem. Soc. Rev. 2007,
36, 348−357.
(23) Salonen, L. M.; Ellermann, M.; Diederich, F. Angew. Chem., Int.
Ed. 2011, 50, 4808−4842.
(44) Davis, J. T.; Okunola, O.; Quesada, R. Chem. Soc. Rev. 2010, 39,
3843−3862.
(24) Rosokha, Y. S.; Lindeman, S. V.; Rosokha, S. V.; Kochi, J. K.
Angew. Chem., Int. Ed. 2004, 43, 4650−4652.
(45) Behr, J. P.; Kirch, M.; Lehn, J. M. J. Am. Chem. Soc. 1985, 107,
241−246.
(25) Dawson, R. E.; Hennig, A.; Weimann, D. P.; Emery, D.;
Ravikumar, V.; Montenegro, J.; Takeuchi, T.; Gabutti, S.; Mayor, M.;
Mareda, J.; Schalley, C. A.; Matile, S. Nat. Chem. 2010, 2, 533−538.
(26) Ballester, P. Acc. Chem. Res. 2012, 46, 874−884.
(27) Wang, D.-X.; Wang, M.-X. J. Am. Chem. Soc. 2013, 135, 892−897.
(28) Wang, D. X.; Zheng, Q. Y.; Wang, Q. Q.; Wang, M. X. Angew.
Chem., Int. Ed. 2008, 47, 7485−7488.
(46) Matile, S.; Sakai, N. In Analytical Methods in Supramolecular
Chemistry, 2nd ed.; Schalley, C. A., Ed., Wiley-VCH: Weinheim, 2012,
pp 711−742.
(47) We also tested calixpyrroles 2b and 2d for ion transport. The
transport data for these receptors follow the same trends discussed for
2a, 2c, and 2e. See the SI for complete ion transport data for all the two-
wall calix[4]pyrroles 2.
(29) Arranz-Mascaros
Godino-Salido, M.-L.; Gutier
́
, P.; Bazzicalupi, C.; Bianchi, A.; Giorgi, C.;
rez-Valero, M.-D.; Lopez-Garzon, R.;
́
́
(48) Transport of nitrite and bicarbonate, two other biologically
relevant oxyanions, was not measured because of their weak basicity. In a
detailed analysis, fluoride has been shown to passively diffuse across
bilayer membranes because of the formation of the neutral HF in minor
quantities, also at pH 7 (pKa = 3.2). See: Gorteau, V.; Bollot, G.;
Mareda, J.; Matile, S. Org. Biomol. Chem. 2007, 5, 3000−3012 . The same
is true for acetate but not for nonbasic anions including nitrate (nitric
acid, pKa −1.3), chloride, bromide, iodide, etc. Nitrous acid (pKa = 3.3),
obtained by protonation of nitrite, is highly volatile and dispropor-
tionates into nitrate and NO. Transport experiments with nitrite will
thus provide results that can presumably not be interpreted in the
context of anion selectivity topologies. Comparative studies of
bicarbonate (pKa 10.3; carbonic acid, pKa 6.4) are obviously not
possible with an assay that operates with a pH gradient.
Savastano, M. J. Am. Chem. Soc. 2013, 135, 102−105.
(30) Wang, D. X.; Wang, M. X. Chimia 2011, 65, 939−943.
(31) Chudzinski, M. G.; McClary, C. A.; Taylor, M. S. J. Am. Chem. Soc.
2011, 133, 10559−10567.
(32) Guerrero, M. G.; Vega, J. M.; Losada, M. Annu. Rev. Plant Physiol.
1981, 32, 169−204.
(33) For a rare example of small molecule-mediated significantly
nitrate-selective ion transport see: Santacroce, P. V.; Okunola, O. A.;
Zavalij, P. Y.; Davis, J. T. Chem. Commun. 2006, 3246−3248.
(34) See Supporting Information for experimental details. Calix[4]
pyrroles 2c (ref 12) and 2e (ref 6) have been prepared previously.
(35) de Namor, A. F. D.; Shehab, M. J. Phys. Chem. B 2003, 107, 6462−
6468.
(49) Gil-Ramírez, G. PhD thesis, Universitat Rovira i Virgili, 2009.
(50) Vargas Jentzsch, A.; Emery, D.; Mareda, J.; Metrangolo, P.;
Resnati, G.; Matile, S. Angew. Chem., Int. Ed. 2011, 50, 11675−11678.
(51) The same nitrate selectivity in ion transport is seen for receptor 2d
but not for the corresponding α,β-3d isomer. See SI.
(36) X-ray structures suggest that when in the 1,3 alternate
conformation Hb is shielded by the meso-aromatic rings. Hence, in the
1H NMR spectrum, the signal belonging to Hb appears upfield with
respect to the signal corresponding to Hc. Upon binding, the
conformational change to cone conformation moves Hb away from
the anisotropic current of the meso-aromatic ring and so the 1H NMR
signal corresponding to Hb is shifted downfield past the signal for Hc.
(37) Gans, P.; Sabatini, A.; Vacca, A. HypNMR 2008, version 4.0.66;
Protonic Software GmbH: Hanau, Germany; peter.gans@hyperquad.
co.uk.
8330
dx.doi.org/10.1021/ja4021793 | J. Am. Chem. Soc. 2013, 135, 8324−8330