Exploring Selectivity of 22 Acyclic Urea-, Carbazole- and Indolocarbazole-Based Receptors towards 11 Monocarboxylates

Article abstract of DOI:10.1002/ejoc.201700931

Carboxylates are attractive target analytes in supramolecular analytical chemistry. 22 acyclic synthetic receptors with different numbers and geometric arrangements of hydrogen-bond donors (HBD) and hydrophobic moieties have been applied to experimentally study selective binding of 11 carboxylate anions of widely differing basicity, hydrophobicity and steric demand, which resulted in 242 accurately determined binding constants. It was found that besides the basicity of the anions, structural and steric factors of anions and receptors influence the binding. Several interesting cases are pinpointed and analysed. The ability of selected receptors to discriminate between anions according to structural features (hydrophilicity, substitution at α-carbon, etc.) is demonstrated. The present results give insight into carboxylate anion binding and make an important step towards systematic development of receptors with useful selectivity patterns and thereby to the practical use of receptor series in sensor arrays for carboxylate fingerprinting in mixtures.

Full text of DOI:10.1002/ejoc.201700931

A Journal of
Accepted Article
Title: Exploring selectivity of 22 acyclic urea-, carbazole- and
indolocarbazole-based receptors towards 11 monocarboxylates
Authors: Kerli Martin, Juuli Nõges, Kristjan Haav, Sandip A. Kadam,
Astrid Pung, and Ivo Leito
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201700931
Link to VoR: http://dx.doi.org/10.1002/ejoc.201700931
Supported by

10.1002/ejoc.201700931
European Journal of Organic Chemistry
FULL PAPER
Exploring selectivity of 22 acyclic urea-, carbazole- and
indolocarbazole-based receptors towards 11 monocarboxylates
Kerli Martin, Juuli Nõges, Kristjan Haav, Sandip A. Kadam, Astrid Pung and Ivo Leito*
Abstract: Carboxylates are attractive target analytes in
supramolecular analytical chemistry. 22 acyclic synthetic receptors
having a different number and geometric arrangement of hydrogen
bond donor (HBD) fragments and hydrophobic moieties have been
applied to study experimentally selective binding of 11 carboxylate
anions of widely differing basicity, hydrophobicity and steric demand,
resulting in 242 accurately determined binding constants. It was
found that besides the basicity of the anions structural and steric
factors of anions and receptors influence the binding. Several
interesting cases are pinpointed and analysed. The ability of
selected receptors to discriminate between anions according to
structural features (hydrophilicity, substitution at α-carbon, etc.) is
demonstrated. The present results give insight into carboxylate
anion binding and make an important step towards systematic
development of receptors with useful selectivity patterns and thereby
to the practical use of receptor series in sensor arrays for
carboxylate fingerprinting in mixtures.
comprehensive analysis of structural features of carboxylates
related to their binding to synthetic receptors is lacking. Binding
characterization is a major part in receptor design and synthesis
and gives guidelines for designing receptors with better
sensitivity and selectivity. Small studies with 1-2 receptors and
anions in differing solvent media provide little possibility for
making generalizations and make it difficult to do any meaningful
comparison of binding data. Therefore, comparable binding
measurements of a larger set of receptors and anions in one
solvent medium are necessary.
The aim of this work is to quantitatively (via logKass values)
characterize the binding of a series of monocarboxylate anions
to a diverse series of synthetic multidentate HBD receptors. In
total 11 carboxylates of different size, geometry, basicity and
hydrophilicity/hydrophobicity were selected: formate, acetate,
anions of propionic acid derivatives (ibuprofen, ketoprofen,
naproxen, pivalic acid and lactic acid), hexanoate, sorbate,
benzoate, and glucuronate (see Scheme 1).
Formate
Acetate
Pivalate
O
-
O
O
Introduction
-
-
H
O
O
O
pKa(H2O) = 3.74[7]
pKa_exp(DMSO) = 10.3[8]
logPo-w = -0.54[9]
ER = 0
pKa(H2O) = 4.76[7]
pKa_exp(DMSO) = 12.3[10]
logPo-w = -0.17[9]
ER = 17
pKa(H2O) = 5.03[11]
pKa_exp(DMSO) =
12.9[12]
Carboxylates are perhaps the most diverse group of anionic
compounds ranging from some of the smallest anions like
formate and acetate to large peptides. Variations in the carbon
chain length or addition of functional groups can greatly alter
their chemical properties and functionalities.[1],[2] In terms of the
molecular recognition process, binding of a receptor only to the
carboxyl group via hydrogen bonding[3] or ion-ion interactions is
insufficient for achieving good selectivity. In the case of simple
carboxylates, there is a strong relation between binding affinity
and anion basicity.[4] This relation alone is by far insufficient for
useful discrimination between different carboxylates. In order to
improve selectivity, additional interactions are necessary, which
can be achieved by host shape manipulation so that guests of
certain shape/geometry would be preferred.[5] The receptor
should not only be able to differentiate between similar
carboxylates but also other anionic species and be capable of
overcoming solvent and counter-ion competition. Large variety in
logPo-w = 1.47[9]
ER = 59
Lactate
HO
Naproxen
Ibuprofen
O
-
O
O
-
O
O
pKa(H2O) = 3.86[7]
pKa_calc(DMSO) = 9.0
logPo-w = -0.72[9]
ER = 48
pKa(H2O) = 4.15[13]
pKa_calc(DMSO) = 11.7
logPo-w = 3.34[9]
ER = 66
pKa(H2O) = 4.40[14]
pKa_calc(DMSO) = 11.6
logPo-w = 4.5[9]
ER = 66
Ketoprofen
Glucuronate
Hexanoate
O
-
O
pKa(H2O) = 4.30[15]
pKa_calc(DMSO) = 11.1
logPo-w = 3.12[9]
ER = 66
pKa(H2O) = 3.28[16]
pKa_calc(DMSO) = 8.3
logPo-w = -2.57[9]
ER = 57
pKa(H2O) = 4.88[11]
pKa_calc(DMSO) = 12.5
logPo-w = 1.92[9]
ER = 36
the
properties
of
carboxylates
(basicity,
chirality,
hydrophobicity/hydrophilicity, polarizability, etc.) creates diverse
ways to design carboxylate receptors. A number of works
demonstrate good affinity and moderate selectivity of synthetic
receptors towards different carboxylates.[5,6] However, to the
Sorbate
Benzoate
O
-
O
-
O
O
pKa(H2O) = 4.50[17]
pKa_calc(DMSO) = 12.4
logPo-w = 1.33[9]
ER = 36
pKa(H2O) = 4.20[7]
pKa_exp(DMSO) = 11.1[10]
logPo-w = 1.87[9]
ER = 45
best of the knowledge of the authors of this paper,
a
Scheme 1. Investigated anions X-COO– together with pKa values of X-COOH
in water (experimental) and DMSO (experimental, if available, otherwise
computational, see the SI for details), logPo-w values (given for the respective
neutral conjugate acids) and substituent repulsive energies ER (kcal mol-1) as
estimators of the steric demand[18] of X. ER values printed in italic have been
estimated on the basis of structural analogy.
[*]
K. Martin, J. Nõges, K. Haav, S. A. Kadam, A. Pung, Prof. I. Leito
Institute of Chemistry
University of Tartu
Ravila 14a, 50411 Tartu, Estonia
E-mail: ivo.leito@ut.ee
WWW: http://analytical.chem.ut.ee/
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201700931
European Journal of Organic Chemistry
FULL PAPER
As Scheme 1 displays, this selection includes anions X-COO–
with different basicities (ranging from highly basic pivalate and
hexanoate to glucuronate and lactate having low basicity),
[b]
Table 1. Experimentally[a] determined binding constants (logKass
selected carboxylates in DMSO-d6:H2O (99.5%:0.5% m/m).
)
towards
A
hydrophilicities/hydrophobicities
(ranging
from
highly
hydrophobic anion of ibuprofen to highly hydrophilic glucuronate
and formate), size (ranging from the smallest carboxylates
formate and acetate to large anions of naproxen and glucuronic
acid) and steric demand of the substituent X (ranging from non-
existent in the case of formate to serious in the case of pivalate
and the propionic acid derivatives). The counterion for all anions
is tetrabutylammonium (TBA). Binding of these anions to 22
urea-, carbazole- and indolocarbazole-based receptors (see
Scheme 2) was measured in DMSO-d6 with 0.5 % of water.
Polydentate systems where several NH centers can form
hydrogen bonds with carboxylate center are among the most
suitable building blocks for selecting receptors for carboxylate
anions.[19,20] For this reason tetra- to hexadentate receptors
dominate the selection. Some bi- and tridentate receptors are
included for comparison. While these receptors are designed to
recognize monocarboxylates, the results described herein are
also relevant to research related to synthesis of ditopic receptors
Rec
G
L
F
B
K
I
N
S
H
P
1
2
3
4
5
6
7
8
9
1.59 2.45[c] 2.75 2.97[c] 3.22 3.46 3.48 3.55 3.70 3.61[c] 3.84[c]
1.76 2.46[c] 3.07 3.25[c] 3.47 3.73 3.68 3.82 3.96 3.86[c] 4.06[c]
1.81 2.59[c] 3.14 3.35[c] 3.63 3.98 3.84 3.98 4.08 3.88[c] 4.16[c]
1.67 2.31[c] 2.67 3.33[c] 3.47 3.78 3.73 3.84 4.00 3.75[c] 4.29[c]
0.84 1.70[c] 1.44 2.34[c] 2.36 2.55 2.56 2.44 2.63 2.41[c] 3.10[c]
1.48 2.07[c] 2.44 3.03[c] 3.33 3.55 3.59 3.48 3.62 3.38[c] 3.82[c]
2.85 3.63 3.50 3.99 4.26 4.67 4.63 4.62 4.83 4.63
5.07
2.63 3.42[c] 3.65 3.95[c] 4.28 4.68 4.61 4.78 5.08 4.68[c] 4.96[c]
2.76 3.77 3.95 4.32 4.15 4.80 4.74 4.89 5.20 4.86
5.28
5.39
10 2.69 3.83 3.63 4.20 4.13 4.93 4.78 5.02 5.22 4.98
11 2.05 3.19[c] 3.27 4.03[c] 4.06 4.42 4.46 4.43 4.92 4.95[c] 5.02[c]
(e.g. for dicarboxylates)[21,22]
sensors[23], and the like.
,
design of macrocyclic anion
12 2.01 2.72 3.13 3.34 3.84 3.65 3.97 4.05 4.27 4.06
13 2.03 2.56 2.91 3.48 3.77 3.97 3.99 3.95 4.12 3.90
4.28
4.48
14 2.88 3.32[c] 3.63 4.08[c] 4.43 4.76 4.69 4.63 4.81 4.64[c] 5.14[c]
15 2.74 3.30[c] 3.46 4.15[c] 4.48 4.86 4.79 4.62 4.83 4.57[c] 5.40[c]
16 1.50 1.93[c] 2.21 2.56[c] 2.93 3.18 3.19 3.07 3.23 3.09[c] 3.47[c]
17 1.82 2.32[c] 2.62 2.89[c] 3.08 3.24 3.25 3.28 3.31 3.33[c] 3.45[c]
18 2.37 2.90[c] 3.27 3.60[c] 3.78 4.02 3.98 4.06 4.15 4.13[c] 4.23[c]
19 1.69 2.15[c] 2.58 2.77[c] 2.97 3.13 3.13 3.20 3.24 3.27[c] 3.34[c]
20 2.02 2.54[c] 3.00 3.25[c] 3.45 3.67 3.60 3.74 3.81 3.78[c] 3.90[c]
21 1.90 2.43[c] 2.83 3.05[c] 3.23 3.43 3.43 3.51 3.57 3.59[c] 3.67[c]
22 2.07 2.60[c] 3.04 3.29[c] 3.50 3.69 3.67 3.77 3.83 3.87[c] 3.95[c]
Results and Discussion
1
Binding affinities were determined using HNMR-based relative
binding affinity measurement method.[24] 11 binding affinity
scales – one for every anion – were constructed (SI pages S162
to S172). Absolute logKass values were obtained by anchoring
scales to absolute logKass values of 15, 19, 21 and 22 according
to the procedure described in ref. [25]. In total 242 logKass values
were determined (see Table 1). The summary of experimental
results is presented in Scheme 2. The anion-receptor complex
structures of receptors 8, 10, 11 and 15 have been computed for
all 11 anions (complexes between 8, 11, 15 and lactate,
benzoate, acetate, and pivalate are available from ref. [4]). From
practical reasoning receptors showing high binding affinity
towards anions were selected for computations.
For simple receptor structures – e.g. indolocarbazole (19) and
diphenylurea (17) – the binding affinity (logKass) of selected
carboxylates approximately follows the basicity order (see
Scheme 1). This is because HB is the primary binding interaction
in these complexes and the higher the pKa of the acid the higher
is generally the hydrogen bond acceptor ability (HBA) of the
anion.
[a] logKass values were measured in DMSO-d6:H2O (99.5%:0.5% m/m) at
25 °C. [b] See the Scales (in the SI) for uncertainty ranges. Rec = Receptor
(Host); G = Glucuronate; L = Lactate; F = Formate; B = Benzoate; K =
Ketoprofen; I = Ibuprofen; N = Naproxen; S = Sorbate; H = Hexanoate; A =
Acetate; P = Pivalate. [c] logKass reported in ref. [4], some of them slightly
changed due to additional measurements. As there are two anomers of
glucuronate, - and β-form, binding constants for glucuronate are apparent
binding constants. Bold type is used to indicate the strongest binder for the
particular anion. Underline indicates (unexpected) binding affinities that are
discussed in the text.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201700931
European Journal of Organic Chemistry
FULL PAPER
Scheme 2. Binding-structure relationship plot comparing the binding affinity changes of all anions under study. Receptor 5 with
lowest logKass values has been omitted for clarity and depicted in the SI.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201700931
European Journal of Organic Chemistry
FULL PAPER
Principal component analysis (PCA) was performed (using
∆logKass values relative to indolocarbazole, 19) to elucidate the
four anions are distributed in the graph in the order of their
hydrophobicities, not basicities.
trends of binding selectivity of carboxylates to different receptors. The differences between receptors in terms of anion binding are
The results (Figure 1) show that most of the investigated
carboxylates can be discriminated by the used receptors to a
statistically significant extent.
visualized in Scheme 2. In general, the stronger binders are
strong (in relative terms) towards all anions and vice versa.
Closer examination reveals, however, numerous cases where in
a group of two receptors (R1 and R2) and two anions (A1 and
A2) R1 binds A1 stronger than A2, while R2 binds A2 stronger
than R1 (so-called “selectivity reversals”). Let us examine some
of them in the "upper group" of the scheme (receptors 7-11 and
14-15), as there receptors, being the best binders, are the most
interesting in practical terms. An interesting case is the behavior
of the three carboxylates with aliphatic chain: acetate,
hexanoate, pivalate. The three anions have similar basicity and
very similar binding constants with receptors operating solely on
the basis of HB (e.g. 17-19). Thus the differences in binding with
the same receptors are caused by other effects. As a big picture,
acetate is bound weaker by most of the receptors in the upper
group. This is most probably caused by the solvophobic effect
contribution: all these receptors have some hydrocarbon
moieties that according to the structure of the complexes can
interact with the alkyl chains of anions. Among the three anions
this effect is the weakest in the case of acetate, which
possesses by far the smallest alkyl chain. If one examines
specific receptors then striking contrast is observed with receptor
15 binding the three carboxylates: 15 is the weakest binder of
acetate in the upper group and at the same the strongest binder
of pivalate of all studied receptors. Examining the geometries of
the complexes reveals the reason: the "half-pocket" of 15 exactly
accommodates pivalate, allowing solvophobic effects between
its t-Bu group and the outer aromatic rings of the carbazole
fragments. At the same time, for acetate the half-pocket is too
big and the anion will have no interactions with the aromatic
rings. The situation with hexanoate is intermediate – it has
interaction with one of the aromatic rings. Another interesting
receptor is 11, which binds the three carboxylates with almost
equal affinity, meaning that it is among the best binders of
acetate and second worst in the upper group for pivalate.
Examining the geometry of the complex reveals that pivalate
with its t-Bu group will not fit into the pocket, while acetate has
just the right size. It is interesting that out of the upper group 11
is by far the weakest binder of formate. The structure reveals
that – similarly to acetate and 15 – formate is too small for the
binding pocket of 11.
Figure 1. PCA plot of binding constant data (relative to indolocarbazole (19)).
The dot size and shape take into account the uncertainties of the experimental
logKass values.
Distribution of anions across the PCA graph indicates
differentiation of the anions by basicity, hydrophobicity, and
steric hindrance. PC1 mainly describes anion basicity. A large
spread of anion basicity (even though damped to some extent
by the use of relative binding affinities) causes this effect to
dominate the binding. To some extent hydrophobicity also
contributes to PC1. The hydrophobic anions are a group in the
upper right corner and hydrophilic anions (glucuronate, formate,
acetate, and lactate) position themselves in the lower and left
sides of the plot. As anion basicity strongly dominates binding,
hydrophilic lactate is positioned not far from hydrophobic
benzoate and ketoprofen. PC2 expresses mainly the steric
hindrance of the anionic center. Formate is the least sterically
hindered anion, followed by acetate, sorbate, and hexanoate
where the steric hindrance is low. On the other hand, in anions
where substituents are close to carboxylate group they hinder
binding (e.g. in glucuronate or pivalate). Although glucuronate is
significantly larger than lactate or pivalate, the effect of steric
hindrance is of similar magnitude. When anions have similar
basicity then these secondary effects might determine the
binding order. Ketoprofen, naproxen, and ibuprofen form a group
of structurally related (all can be considered derivatives of 2-
phenylpropanoate) large hydrophobic anions and are expectedly
close to each other. Also, benzoate is linked to this group. These
Receptors 9 and 10 are for the majority of anions the most
potent binders in this work. Noticeable differences (0.65 and
0.80 logKass units, respectively) in binding is between ibuprofen
and ketoprofen are evident with receptors 9 and 10: in the upper
group 9 and 10 are among the strongest binders of ibuprofen
and among weakest with ketoprofen. At the same time,
indolocarbazole (19) that has only 2 HBD centers, and where
binding is expected to be dependent only on anion basicity,
binds ibuprofen only by 0.16 logKass units stronger. Both anions
are 2-phenylpropionate derivatives, the steric surroundings of
the carboxylate group are almost identical and the anions differ
only by the substituents in the aromatic ring: 4-isobutyl vs 3-
benzoyl, respectively. The aqueous pKa values of the conjugate
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201700931
European Journal of Organic Chemistry
FULL PAPER
acids of the anions differ by only 0.1 pKa units. The
computational geometries (SI S192-S193) suggest that
carboxylate groups of ibuprofen and ketoprofen are bound by
the same number of NH groups and the substituents have small
effect on the structure of the complexes. In this situation the
large binding differences of 9 and 10 with these anions are
surprising.
Receptors 9 and 10 are the most sophisticated ones where
besides HBs with the carboxylate group an appreciable
solvophobic effect can be observed between the hydrophobic
residues of the anions and the receptors. The difference
between these interactions between the highly hydrophobic
isobutyl group and markedly more polar benzoyl group might be
the main reason for the large difference in binding affinities. This
finding demonstrates that minor (and distant from the anionic
center) structural effects in anions can have large impact on
binding affinity.
three with another oxygen atom. The lowest N-H···O angle is
148°.
Expectedly their binding affinities towards those anions differ on
an average only by 0.15 logKass units from the next receptors (by
0.3 logKass units, as maximum). Thus, their binding affinity is well
comparable to the remaining urea-carbazole combined receptors.
Receptors 7–11 and 15 are the strongest binders with different
carboxylates. In spite of their similar core structures, receptors 9-
11 show different binding patterns. Pivalate, hexanoate, and
acetate are bound by 11 with similar affinity, while 9 and 10
prefer pivalate and hexanoate against acetate. As for the
remaining anions, they are bound stronger by 9 and 10 than by
11. Some anions have additional HBA centers, most notably
lactate and glucuronate. Out of the computationally investigated
receptors 7, 8, 10 and 11 form additional hydrogen bond with
lactate’s 2-OH group as evidenced by computations (the
additional HB is present in the dominant conformers). Such
additional HBs are probably the reason that lactate binds
stronger with 7, 8, 9 and 10 than could be expected from its low
basicity. The strong binding of lactate with 9 and 10 is
noteworthy. The computational geometry of the complex with 10
reveals that even two HBs are formed between the urea
fragment attached to naphthyl and lactate’s OH. Ibuprofen and
naproxen (as well as acetate and sorbate) are bound to 8, 9 and
10 with similar affinity patterns. However, the affinity pattern of
these receptors towards ketoprofen is different, despite very
similar vicinity of the carboxylate group in the three anions.
Glucuronate has, similarly to lactate, OH groups that can behave
as additional HBA (or HBD) centers to have additional
interactions with a receptor. Still, lactate is bound considerably
stronger than glucuronate. Lactate is bound ~ 0.75 logKass units
stronger to receptors 1-8 and 12, ~ 1 logKass units stronger to 9,
10 and 11 and around 0.5 logKass units stronger to receptors 13-
22. Glucuronate is more hydrophilic and a weaker base. The first
factor causes stronger solvation by solvent molecules and the
second factor causes weaker interaction between receptor’s NH
groups and glucuronate’s carboxylate group. 1,3-bis(carbazolyl)
urea based receptors 9, 10 and 11 form a binding ‘pocket’ of
suitable size for lactate but the spatial fit for glucuronate is not
optimal. Lactate’s OH group might be less accessible to solvent
molecules in the complexed state than glucuronate’s OH groups.
It is also interesting that ibuprofen is bound stronger by all
selected receptors except for 12 that binds ketoprofen by 0.18
logKass units stronger, which in our interpretation might be
caused by an additional HB between the ketoprofen’s benzoyl
group and one of the urea fragments of 12.
As a rule, receptors with 4 or more HBD groups dominate in
terms of binding affinity. Out of the bidentate HBD molecules
3,4,4’-Cl3-diphenylurea (18) is on an average the strongest
binder, because of the inductive effect of the three chloro
substituents, enhancing both the acidity and HBD ability of 18. It
has been found, however, that contrary to intuitive expectation,
the correlation between HBD ability and acidity of a molecule is
not strong.[26] Combinations of carbazole (or indole) and urea (7-
11, 14 and 15) have the highest binding affinities. There is an
optimal number of HBD groups that should be implemented into
receptor framework. Computational modeling suggests that as a
maximum, 5-6 NH groups can come to close proximity of a
carboxylate group to form HBs in a co-operative manner.
Furthermore, it is important that the atoms N-H···O are aligned
as closely on one line as possible. 9 and 10 possess 8 NH
groups in total and indeed, together they are the strongest
binders for 7 anions out of 11. For instance, according to the
computational criteria specified in the SI (Computational
structures section), receptor 10 forms five hydrogen bonds with
ibuprofen, two HBs with one carboxylate group’s oxygen and
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201700931
European Journal of Organic Chemistry
FULL PAPER
samples were spiked with the in-house prepared internal calibration
Conclusions
mixture containing perfluorinated Brønsted superacids. Ions used for
–
calibration were: C12F10NO4S2 (m/z = 475.91145), C8F17NO2SH− (m/z =
Differential binding of 11 carboxylate anions of widely differing
basicity, hydrophobicity and steric demand with 22 neutral
synthetic receptors was demonstrated. The results of PCA
reveal that carboxylates with dissimilar structures can be
distinguished using the differential sensing paradigm.
Additionally, this work provides experimentally determined 242
binding constants that can be directly compared with each other.
Several cases of interesting relations between structure and
binding are pinpointed and analysed. Lengthening of the carbon
chain (acetate vs hexanoate) appears to play modest role in
differentiating between the anions. Anion basicity is a key factor
in determining binding and additional influence is exerted by
other groups in the anions, by steric factors and by subtle
structural effects. It is demonstrated that the relative binding
affinity of acetate and pivalate is to a large extent governed by
the suitability of the size of the binding site dimensions to
support solvophobic effect (stronger with pivalate than acetate).
Large binding differences were found between the very similar
anions of naproxen and ibuprofen – anions with similar structure
and basicity and identical surroundings of the carboxylate group.
The presented results give insight into carboxylate anion binding
and contribute towards systematic development of receptors
with useful selectivity patterns and thereby to the practical use of
receptor series in sensor arrays for carboxylate fingerprinting in
mixtures.
–
–
497.94620), C8F18NO4S2 (m/z
= 579.89868), C12F26NO4S2 (m/z =
779.88591)[27]
.
The concentration of the calibrants in the infused
solutions remained within 0.5–1.0 µM.
Solvents and Chemicals. The solvent for binding measurements,
DMSO with 0.5% of water (m/m), was prepared using anhydrous DMSO
99.9% (for UV-Vis and fluorescence measurements) or DMSO-d6 99.8%
(for NMR measurements) and water from MilliQ Advantage A10 system.
The water content of the DMSO solvent was checked with Mettler Toledo
DL 32 titrator. Titrant solutions for binding measurements were prepared
from respective tetrabutylammonium salts.
Receptor Molecules. Receptors 1-6, 8, 11, 14-16 are the same as used
in ref. [4], 17-19, 22 are from ref. [25], and 20, 21 from ref. [24]. Synthesis
procedures of compounds 9 and 10 are described in ref [28]. Literature
approaches were used for the synthesis of 12[29] and 13[30]
.
Preparation of Compound 7. Compound 7a (0.10 g, 0.32 mmol),
prepared as in ref. [4], was dissolved in acetonitrile (20 mL) then phenyl
isocyanate was added drop-wise (0.081 g, 0.67 mmol). The reaction
mixture was stirred at reflux temperature under N2 atmosphere for 17 h.
After the disappearance of the starting material (monitored by TLC)
formed white precipitate was cooled and filtered. The white solid washed
with diethyl ether to obtain the pure compound 7 (0.15 g, 0.27 mmol,
84.7%) as a white solid.
Experimental Section
(1)
Instruments. For compound characterization of 1H and 13C NMR spectra
were recorded at 700.1 MHz (1H), 176.1 MHz (13C) and NMR-based
relative binding affinity measurements were carried out at 700.1 MHz on
Data for 7: Mp: 349.5˗352.9 °C. Rf = 0.57 (10% methanol in DCM). 1H
NMR (700.1 MHz, DMSO-d6, +20 °C) δ: 10.01 (bs, 1H, NH); 8.83 (bs, 2H,
NH); 8.77 (bs, 2H, NH); 7.91 (d, J=2.1 Hz, 2H); 7.54-7.51 (m, 4H); 7.42
(d, J=2.1 Hz, 2H); 7.29-7.26 (m, 4H); 6.99-6.96 (m, 2H); 1.41 (s, 18H,
CH3). 13C NMR (176.0 MHz, DMSO-d6, +20 °C) δ: 153.5 (CO); 142.3;
140.3; 131.6; 129.2; 125.2; 123.5; 122.3; 118.9; 116.4; 112.3; 34.9 (C-
CH3); 32.3 (C-CH3). IR (A R- -IRS) : 329 , 3042, 2950, 2899, 164 ,
1597, 1557, 1497, 1226, 859, 746, 692, 648, 503 cm-1. MALDI FT-ICR
a
Bruker AVANCE III 700 instrument. UV-Vis spectrophotometric
measurements were carried out using Thermo Nicolet Evolution 300
spectrophotometer and fluorescence spectrofluorometric measurements
were carried out using Horiba FluoroMax-4 spectrofluorometer. All the
NMR, UV-Vis, and fluorescence measurements carried out at 25 °C.
ATR-IR spectra were recorded on a Thermo Electron Nicolet 6700 FT-IR
device using a Smart Orbit micro-ATR accessory with diamond crystal.
The spectrometer had a DTGS detector and CsI beamsplitter. 128 scans
were recorded over the range of 400–4000 cm-1. For HRMS-analysis the
samples were first dissolved in MeCN/H2O (80/20) with a concentration
of ~1 mg/ml and then diluted in MeCN/H2O (80/20) for ESI-
measurements. The concentration was chosen so that an appropriate
signal was achieved in LRMS part of the machine. The final analyte
concentration of infused solutions remained in range 0.5-2 µg/ml. High-
resolution ESI-ICR spectra were obtained on a hybrid Varian 910-FT-
ICR-MS system, which is coupled to Varian-J320 triple-quadrupole MS.
Ionization parameters were as follows: spray chamber temperature,
40 °C; spray needle voltage, -4500 V; nebulizing gas (N2) pressure, 30
psi; API-drying gas (N2), 15 psi at 250 °C; shield voltage, 600 V; and
capillary voltage 32 V. Mass range of ions, selected by quadrupole, were
guided into the FT-ICR analyzer cell. Ion guide parameters and FT-ICR
ion guide and excitation parameters were optimized for mass range (m/z
= 100–1000). Ion collection time varied from 300 to 1000 ms; FT-ICR
analyzer cell parameters were: DAC rate: 8000 kHz for m/z range of
100–800 direct (broadband); ADC rate: 4 MHz; transient length: 1024 K;
262.144 ms or 2048 K; 524.288 ms. For the calibration of the mass axis,
(m/z): solvent
~
0.01% DMSO:isopropanol, [M+Na]+ calcd for
[C42H41N5O2+Na]+ 570.28449, found 570.28461.
Preparation of tetrabutylammonium carboxylate salts. Commercially
available TBA salts of acetate and benzoate were used. TBA salts of
[4]
pivalate (trimethylacetate) and lactate are previously described in ref.
.
TBA salts of formate, (S)-(+)-naproxen, ibuprofen (as a racemate), (S)-
(+)-ketoprofen, D-glucuronate, hexanoate and sorbate were prepared by
adding 1 equiv. of tetrabutylammonium hydroxide in methanol to a
solution of the corresponding acid (1 equiv.) in methanol. The mixture
was stirred at room temperature for 24 h, evaporated to dryness under
reduced pressure and then dried under high vacuum at room
temperature overnight. The salts are stored in a glove box under argon
atmosphere.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201700931
European Journal of Organic Chemistry
FULL PAPER
Data for TBA hexanoate: 1H NMR (700.1 MHz, DMSO-d6, +20 °C) δ:
3.19-3.16 (m, 8H); 1.72-1.70 (m, 2H); 1.59-1.55 (m, 8H); 1.37-1.32 (m,
2H); 1.30-1.28 (m, 8H); 1.24-1.28 (m, 2H); 1.16-1.15 (m, 2H); 0.92 (t, J =
7.0 Hz, 12H); 0.83 (t, J = 7.0 Hz, 3H). 13C NMR (176.0 MHz, DMSO-d6,
+20 °C) δ: 1 4.6; 5 .5; 48.5; 32.0; 26.6; 23.1; 22.3; 19.2; 14.1; 13.5. IR
(A R- -IRS) : 2958, 2932, 28 3, 15 6, 1463, 13 3, 883, 38 cm-1. ESI
FT-ICR MS (m/z): solvent MeCN/H2O (80/20), [M]- calcd for [C6H11O2]-
115.07645, found 115.07647.
Data for TBA sorbate: 1H NMR (700.1 MHz, DMSO-d6, +20 °C) δ: 6.58-
6.54 (m, 1H); 6.07-6.03 (m, 1H); 5.74-5.70 (m, 1H); 5.61-5.59 (m, 1H);
3.19-3.15 (m, 8H); 1.72-1.71 (m, 3H); 1.59-1.54 (m, 8H); 1.33-1.28 (m,
8H); 0.92 (t, J = 7.0 Hz, 12H). 13C NMR (176.0 MHz, DMSO-d6, +20 °C)
δ: 169.6; 134.9; 133.8; 132.1; 130.3; 58.0; 23.6; 19. ; 18.5; 13.9. IR
(A R- -IRS) : 3169, 2959, 2936, 28 4, 1568, 1354, 996, 882, 40,
705, 570 cm-1. ESI FT-ICR MS (m/z): solvent MeCN/H2O (80/20), [M]-
calcd for [C6H7O2]- 111.04515, found 111.04522.
(2)
Data for TBA formate: 1H NMR (700.1 MHz, DMSO-d6, +20 °C) δ: 8.5 (s,
1H); 3.20-3.17 (m, s8H); 1.59-1.54 (m, 8H); 1.32-1.28 (m, 8H); 0.92 (t, J
= 7.0 Hz, 12H). 13C NMR (176.0 MHz, DMSO-d6, +20 °C) δ: 165.2; 5 .9;
23.6; 19. ; 14.0. IR (A R- -IRS) : 2959, 28 3, 2590, 159 , 1491,
1456, 1332, 882, 587 cm-1.
Measurement of the relative and absolute binding constants. The
binding constant measurements were carried out in DMSO-d6 or in
DMSO with 0.5 % water (m/m) using the previously described[24,25] NMR
and UV-Vis methodologies. For NMR relative binding measurements, the
concentrations of TBA salt in the concentrated solution were
approximately 0.63-2.20 M and in diluted solutions approximately 0.25-
0.75 M, depending on the degree of solubility and anion basicity. The
initial concentrations of receptors were around 0.006-0.015 M. For UV-
Vis absolute binding measurements, the concentrations of receptors
were in the following ranges: receptor 15 (3.0·10-5 M), receptor 19
(3.0·10-5 M), receptor 21 (8. ·10-5 M), and receptor 22 (8.3·10-5 M). For
fluorescence absolute binding measurements, the concentration of
receptor 15 was around 1.2·10-6 M, and the concentration of receptor 19
was approximately 8.1·10-6 M for formate and around 1.4·10-5 M for
ibuprofen. 1H NMR measurements for absolute logKass determination
were performed only with receptor 22 which concentration was around
.0·10-6 M.
Data for TBA naproxen: 1H NMR (700.1 MHz, DMSO-d6, +20 °C) δ: .69
(d, J = 8.4 Hz, 1H); 7.62 (d, J = 8.4 Hz, 1H); 7.58 (s, 1H); 7.45 (dd, J =
7.0, 2.1 Hz, 1H); 7.21 (d, J = 2.1 Hz, 1H); 7.07 (dd, J = 7.0, 2.1 Hz, 1H);
3.85 (s, 3H); 3.30-3.29 (m, 1H); 3.16-3.14 (m, 8H); 1.57-1.52 (m, 8H);
1.32-1.27 (m, 11H); 0.93 (t, J = 7.0 Hz, 12H). 13C NMR (176.0 MHz,
DMSO-d6, +20 °C) δ: 1 4.3; 155. ; 141. ; 131.8; 128.2; 128.0; 12 .4;
124.9; 124.1; 11 .3; 105.0; 56.9; 54.5; 49.1; 48.0; 22.5; 19.8; 18. ; 13.0.
IR (A R- -IRS) : 318 , 2959, 28 4, 1589, 1485, 13 5, 1339, 1212,
1032, 810, 746, 476 cm-1. ESI FT-ICR MS (m/z): solvent MeCN/H2O
(80/20), [M]- calcd for [C14H13O3]- 229.08702, found 229.08717.
Data for TBA ibuprofen: 1H NMR (700.1 MHz, DMSO-d6, +20 °C) δ: .14
(d, J = 7.7 Hz, 2H); 6.94 (d, J = 7.7 Hz, 2H); 3.19-3.14 (m, 8H); 3.14-3.12
(m, 1H); 2.36 (d, J = 7.0 Hz, 2H); 1.81-1.75 (m, 1H); 1.59-1.54 (m, 8H);
1.33-1.28 (m, 8H); 1.18 (d, J = 7.0 Hz, 3H); 0.93 (t, J = 7.0 Hz, 12H); 0.86
(d, J = 7.7 Hz, 6H). 13C NMR (176.0 MHz, DMSO-d6, +20 °C) δ: 1 5.6;
144.5; 137.5; 128.3; 127.6; 57.9; 49.7; 48.9; 44.9; 30.2; 23.6; 22. ; 20.9;
19. ; 14.0. IR (A R- -IRS) : 318 , 2959, 28 4, 1589, 1485, 13 5,
1339, 1212, 1032, 810, 746, 476 cm-1. ESI FT-ICR MS (m/z): solvent
MeCN/H2O (80/20), [M]- calcd for [C13H17O2]- 205.12340, found
205.12340.
Acknowledgements
This work was supported by the grant No 9105 from the
Estonian Science Foundation, by the Estonian National R & D
infrastructure development program of Measure 2.3 ‘Promotion
of development activities and innovation’ (Regulation No. 34)
funded by the Enterprise Estonia Foundation, by the EU through
the European Regional Development und ( K141 “Advanced
materials and high-technology devices for energy recuperation
systems”) and by the Institutional funding project IU 20-14 from
the Estonian Ministry of Education and Research. We thank the
group of Philip Gale for supplying compounds 14 and 15.
Data for TBA ketoprofen: 1H NMR (700.1 MHz, DMSO-d6, +20 °C) δ:
7.73-7.72 (m, 1H); 7.68-7.66 (m, 1H); 7.56-7.53 (m, 5H); 7.46-7.44 (m,
1H); 7.38-7.36 (m, 1H); 3.26-3.23 (m, 1H); 3.32-3.16 (m, 8H); 1.58-1.54
(m, 8H); 1.33-1.27 (m, 8H); 1.24 (d, J = 3.5 Hz, 3H); 0.92 (t, J = 7.0 Hz,
12H). 13C NMR (176.0 MHz, DMSO-d6, +20 °C) δ: 196.2; 1 4.1; 14 .3;
13 .5; 136.1; 132.4; 132.1; 129.6; 128.9; 128.5; 12 .5; 126.4; 5 .5; 49.6;
23.1; 20.2; 19.2; 13.4. IR (A R- -IRS) : 2959, 2934, 28 3, 2804, 1652,
1593, 1372, 1280, 1048, 864, 722, 704, 642 cm-1. ESI FT-ICR MS (m/z):
solvent MeCN/H2O (80/20), [M]- calcd for [C16H13O3]- 253.08702, found
253.08722.
Keywords: Anion Receptors • Carboxylate • Binding Affinity
Data for TBA glucuronate: 1H NMR (700.1 MHz, DMSO-d6, +20 °C) δ:
3.72-3.71 (m, 1H); 3.57-3.53 (m, 2H); 3.49-3.46 (m, 1H); 3.40-3.38 (m,
1H); 3.32-3.31(m, 1H); 3.17-3.15 (m, 8H); 1.59-1.54 (m, 8H); 1.33-1.28
(m, 8H); 0.93 (t, J = 7.7 Hz, 12H). 13C NMR (176.0 MHz, DMSO-d6,
+20 °C) δ: 1 5.4; 2.8; 2.4; 2.1; 1.5; 64.2; 58.0; 23.5; 19. ; 14.0. IR
(A R- -IRS) : 3413, 3263, 2961, 2935, 28 5, 1612, 1495, 1356, 1093,
1032, 881, 545. ESI FT-ICR MS (m/z): solvent MeCN/H2O (80/20), [M]-
calcd for [C6H9O7]- 193.03538, found 193.03545.
• Hydrogen Bond • Discrimination
References
[1]
[2]
[3]
S. Pagola, P. W. Stephens, D. S. Bohle, A. D. Kosar, S. K. Madsen,
Nature 2000, 404, 307–310.
C. N. R. Rao, S. Natarajan, R. Vaidhyanathan, Angew. Chem. Int. Ed.
2004, 43, 1466–1496.
P. D. Beer, P. A. Gale, Angew. Chem. Int. Ed. 2001, 40, 486–516.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201700931
European Journal of Organic Chemistry
FULL PAPER
[4]
S. A. Kadam, K. Martin, K. Haav, L. Toom, C. Mayeux, A. Pung, P. A.
Gale, J. R. Hiscock, S. J. Brooks, I. L. Kirby, et al., Chem. – Eur. J.
2015, 21, 5145–5160.
[5]
[6]
[7]
[8]
[9]
F. Ulatowski, J. Jurczak, Asian J. Org. Chem. 2016, DOI
10.1002/ajoc.201600093.
Y. Liu, T. Minami, R. Nishiyabu, Z. Wang, P. Anzenbacher, J. Am.
Chem. Soc. 2013, 135, 7705–7712.
R. J. Prankerd, Profiles of Drug Substances, Excipients and Related
Methodology, Elsevier, London, 2007.
S. M. Petrov, Y. I. Umanskii, I. F. Mullin, A. N. Tertichnyi, Z. F. Khim,
Russ. J. Phys. Chem. 1973, 47, 363, 647.
C. Hansch, A. Leo, D. Hoekman, Exploring QSAR Hydrophobic,
Electronic, and Steric Constants, American Chemical Society,
Washington, DC, 1995.
[10]
[11]
F. G. Bordwell, Acc. Chem. Res. 1988, 21, 456–463.
D. D. Perrin, B. Dempsey, E. P. Serjeant, pKa Prediction for Organic
Acids and Bases, Chapman And Hall, 1981.
[12]
[13]
[14]
I. A. Koppel, J. B. Koppel, V. O. Pihl, Org. React. 1988, 25 (1), 77–101.
X. Li, M. A. Cooper, Anal. Chem. 2012, 84, 2609–2613.
M. Meloun, S. Bordovská, L. Galla, J. Pharm. Biomed. Anal. 2007, 45,
552–564.
[15]
A. Fini, P. De Maria, A. Guarnieri, L. Varoli, J. Pharm. Sci. 1987, 76,
48–52.
[16]
[17]
R. Kohn, P. Kovac, Chem Zvesti 1978, 32, 478–485.
E. Lück, M. Jager, N. Raczek, in Ullmanns Encycl. Ind. Chem. (Ed.:
Wiley-VCH Verlag GmbH & Co. KGaA), Wiley-VCH Verlag GmbH &
Co. KGaA, Weinheim, Germany, 2000.
[18]
[19]
[20]
[21]
D. P. White, J. C. Anthony, A. O. Oyefeso, J. Org. Chem. 1999, 64,
7707–7716.
N. H. Evans, P. D. Beer, Angew. Chem. Int. Ed. 2014, 53, 11716–
11754.
V. Blažek Bregović, N. Basarić, K. Mlinarić-Majerski, Coord. Chem.
Rev. 2015, 295, 80–124.
S.-Y. Liu, Y.-B. He, J.-L. Wu, L.-H. Wei, H.-J. Qin, L.-Z. Meng, L. Hu,
Org. Biomol. Chem. 2004, 2, 1582–1586.
[22]
[23]
[24]
C. L. Beck, A. H. Winter, J. Org. Chem. 2014, 79, 3152–3158.
P. Piątek, J. Jurczak, Chem Commun 2002, 2450–2451.
S. A. Kadam, K. Haav, L. Toom, T. Haljasorg, I. Leito, J. Org. Chem.
2014, 79, 2501–2513.
[25]
K. Haav, S. A. Kadam, L. Toom, P. A. Gale, N. Busschaert, M. Wenzel,
J. R. Hiscock, I. L. Kirby, . Haljasorg, M. Lõkov, et al., J. Org. Chem.
2013, 78, 7796–7808.
[26]
[27]
S. Tshepelevitsh, A. Trummal, K. Haav, K. Martin, I. Leito, J. Phys.
Chem. A 2017, 121, 357–369.
A. Teearu, S. Vahur, T. Rodima, K. Herodes, W. Bonrath, T. Netscher,
S. shepelevitsh, A. rummal, M. Lõkov, I. Leito, J. Mass Spectrom.
2017, DOI 10.1002/jms.3943.
[28]
[29]
S. A. Kadam, K. Haav, L. Toom, A. Pung, C. Mayeux, I. Leito, Eur. J.
Org. Chem. 2017, DOI 10.1002/ejoc.201601583.
N. Busschaert, P. A. Gale, C. J. E. Haynes, M. E. Light, S. J. Moore, C.
C. Tong, J. T. Davis, W. A. Harrell, Jr., Chem. Commun. 2010, 46,
6252.
[30]
J. Y. Lee, E. J. Cho, S. Mukamel, K. C. Nam, J. Org. Chem. 2004, 69,
943–950.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201700931
European Journal of Organic Chemistry
FULL PAPER
Entry for the Table of Contents
FULL PAPER
242 accurately determined
Carboxylates Binding
binding constants between 11
carboxylates and 22 synthetic
receptors are used to reveal the
main structural factors of anions
and receptors governing the
binding and demonstrate that the
receptors selected for this study
enable discriminating between
anions according to structural
features (hydrophilicity,
K. Martin, J. Nõges, K.
Haav, S. A. Kadam, A.
Pung, I. Leito*
Page No. – Page No.
Exploring selectivity of 22
acyclic urea-, carbazole-
and indolocarbazole-
based receptors towards
11 monocarboxylates
substitution at α-carbon, etc).
This article is protected by copyright. All rights reserved.