Expanding sapphyrin: Towards selective phosphate binding

Article abstract of DOI:10.1002/chem.200800860

The anion-templated syntheses and binding properties of novel macrocyclic oligopyrrole receptors in which pyrrole rings are linked through amide or imine bonds are described. The efficient synthesis was accomplished by anion-templated [1 + 1] Schiff-base

Full text of DOI:10.1002/chem.200800860

FULL PAPER
DOI: 10.1002/chem.200800860
Expanding Sapphyrin: Towards Selective Phosphate Binding
Evgeny A. Katayev,*[a, b] Nikolay V. Boev,[b] Ekaterina Myshkovskaya,[a]
Victor N. Khrustalev,[a] and Yu. A. Ustynyuk[b, c]
Abstract: The anion-templated synthe-
ses and binding properties of novel
macrocyclic oligopyrrole receptors in
which pyrrole rings are linked through
amide or imine bonds are described.
The efficient synthesis was accom-
plished by anion-templated [1+1]
Schiff-base condensation and acylation
macrocyclization reactions. Free recep-
tors and their host–guest complexes
with hydrochloric acid, acetic acid, tet-
rabutylammonium chloride, and hydro-
gen sulfate were analyzed by single-
crystal X-ray diffraction analysis. Sta-
bility constants with different tetrabu-
tylammonium salts of inorganic acids
12) exhibit high affinity (logKa =5–7)
for bifluoride, acetate, and dihydrogen
phosphate, and interact weakly with
chloride and hydrogen sulfate. The
amido-bipyrrole receptors 11 and 13
with four pyrrole rings exhibit 104- and
102-fold selectivity for dihydrogen
phosphate, respectively, as inferred
from competitive titrations in the pres-
ence of tetrabutylammonium acetate.
1
were determined by standard H NMR
and UV/Vis titration techniques in
[D6]DMSO/0.5% water solution. Ac-
cording to the titration data, receptors
containing three pyrrole rings (10 and
Keywords: host–guest systems
phosphates receptor pyrroles
receptors · template synthesis
·
·
·
Introduction
that contain a phosphate residue. The binding of phosphate
is essential to a myriad of biological processes ranging from
biosynthesis and metabolism to gene regulation, signal
transaction, and antibiotic resistance. In this regard, the de-
velopment of synthetic receptors for different phosphates
should benefit drug discovery.[7]
Molecular recognition is an emerging area in modern
chemistry that is starting to play a significant role in analyti-
cal, organic, inorganic, and medical chemistry.[1–3] The range
of targets for molecular recognition has dramatically in-
creased in recent years from simple atomic species to more
complex structures such as oligopeptides[4,5] and nucleic
acids.[6] The binding of phosphate species is of particular in-
terest due to its ubiquitous role in living organisms.[7] In
fact, nearly half of all known proteins interact with partners
The general principles of tetrahedral oxoanion recognition
have recently been reviewed by us[8] and others.[9,10] At the
present time a number of different approaches have been
realized for the design of phosphate receptors, among which
strategic, high-yielding anion-templated synthesis has at-
tracted particular attention.[11–15] However, many known arti-
ficial receptors are unable to distinguish between structural-
ly related anions bearing the same overall charge, for exam-
ple, acetate and hydrogen sulfate. Some known efficient re-
ceptors are based on a combination of coulombic, metal co-
ordination, or hydrogen-bonding interactions, for example,
tripodal copper complexes developed by Anslyn[16] and Fab-
[a] Dr. E. A. Katayev, E. Myshkovskaya, Dr. V. N. Khrustalev
Institute of Organoelement Compounds, Russian Academy of Scien-
ces
Vavilov St., 28, Moscow, 119991 (Russia)
Fax : (+7)499-135-5085
E-mail: katayev@ineos.ac.ru
[b] Dr. E. A. Katayev, Dr. N. V. Boev, Prof. Y. A. Ustynyuk
Department of Chemistry
brizzi[17] and their co-workers with KHPO 2À/KOAcÀ, zinc com-
4
plexes by Gunning[18] and Kim and co-workers[19] with
M. V. Lomonosov Moscow State University
Leninskie Gory, 1/3, Moscow, 119991 (Russia)
KHPO 2À/KOAcÀ >28>10 in water, and protonated sapphyrins
4
described by Sessler and co-workers[20] with a binding con-
stant K (relative to 2 equiv of OAcÀ)=32mÀ1 in CD3OD,
and some other charged organic receptors specifically de-
signed for phosphate binding.[21–23] Neutral receptors are
generally less selective mostly due to lower binding affini-
[c] Prof. Y. A. Ustynyuk
Research Center for Magnetic Tomography and Spectroscopy
M. V. Lomonosov Moscow State University
Leninskie Gory, 1, Bldg 73, Moscow, 119991 (Russia)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200800860.
Chem. Eur. J. 2008, 14, 9065 – 9073
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9065

ties, for example, urea-based (KH PO
À
4
/KOAcÀ =28[24] and
C6H5OPO(OH),[20] and cAMP).[34] Thus, an expansion of the
inner cavity of the receptor with additional coordination
sites in appropriate conjunctures is one of the possible ways
to introduce geometrical host–guest complementarity and
hence improved selectivity. In accord with the above-men-
2
KH PO
À
/KOAcÀ =232[25] in [D6]DMSO), amide-based (KH PO
/
À
4
2
4
2
KOAcÀ =5.7 in [D6]DMSO),[26] and amido-pyrrole-based re-
ceptors (KH PO À/KOAcÀ =5–30 also in [D6]DMSO).[27–29] Our
2
4
recent investigations of neutral receptors have shown that a
combination of hydrogen-bond
acceptor (imine or amine) and
donor (amide or pyrrole)
groups in one receptor with ap-
propriate spatial locations and
orientations leads to efficient
synthetic receptors for the
À
H2PO4 anion in CH3CN and
DMSO solutions with selectivi-
ties in the range of 102–103,
values approaching those of
charged receptors (Scheme 1;
values of KH PO /KOAcÀ are
À
2
4
equal to 9, 7.4, and 77 for 1, 2,
and 4, and values of KH PO
À
/
4
2
KHSO
À
are equal to 5.3, 3, and
4
0.56, respectively). The intro-
duction of amino groups (re-
ceptor 3) increased the selectiv-
ity up to 102–103 (KH PO
À
4
/
2
KOAcÀ =250,
KH PO
À
/KHSO =
À
2
4
4
25).[30–32] In our laboratory,
solid-phase analysis of oxoan-
ions with receptors containing
pyrrolic hydrogen bonds re-
vealed that incorporation of di-
pyrromethanes into the recep-
tor structure in most cases leads
to a selectivity for acetate,[33]
whereas the incorporation of
2,2’-bipyrroles leads to a selec-
tivity for phosphate.[32]
With a high phosphate selec-
tivity, sapphyrin has particularly
attracted our attention and
served as a point of origin for
the synthesis of novel receptors.
Sapphyrin is a versatile anion-
binding agent with remarkable
selectivity towards fluoride and
phosphate anions in polar
media. Analysis of the crystal
structures of host–guest com-
plexes leads to the suggestion
that the geometry of the coordi-
nation sites is complementary
only for the fluoride anion and
that phosphate selectivity is a
result of a high affinity for the
phosphate oxygen (see, for ex-
ample, complexes with H3PO4,
(C6H5O)2PO(OH),
Scheme 1. Structures of neutral receptors that have proven selective for phosphate anions.
9066
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9065 – 9073

Selective Phosphate Binding
FULL PAPER
tioned facts, building blocks that contain fragments structur-
ally equivalent to sapphyrin, such as 6–9 (Scheme 2), were
designed. Note that the proposed receptors, 10, 12 and 11,
13, have similar macrocyclic skeletons to sapphyrin and cy-
clo[8]pyrrole, respectively (see the bold structures in
Scheme 2).
Herein, we describe the anion-templated synthesis and
binding properties of novel macrocyclic oligopyrrole recep-
tors in which the pyrrole rings are linked through amide or
imine bonds. The receptor synthesis was accomplished by
covalent template self-assembly (CTS), successfully used in
our laboratory for the synthesis of receptors using template
imine[31] and amide[35] bond formation reactions. To show
the nature of the host–guest interactions and the influence
of geometry, size, and nature of the binding motifs on the
overall selectivity of the receptors, we investigated four dif-
ferent macrocyclic ligands. They can be divided into the fol-
lowing two categories: Oligopyrrole amide receptors (10
and 11) and oligopyrrole amino-imine receptors (12 and 13).
Where possible, the host–guest complexes were investigated
in the solid state. The anion-binding properties of the recep-
tors were determined by 1H NMR and UV/Vis titration
techniques.
Results and Discussion
Synthesis and structure: The receptors 10–13 were synthe-
sized from dicarbaldehyde 6, dicarboxylic acid 7, and the di-
amine building blocks 8 and 9 (Scheme 2). These com-
pounds were chosen to expand the cavity of sapphyrin with
additional amide groups (cf. structure 5 and 10) and enhance
the overall selectivity for oxoanions. Amido-imine receptors
12 and 13 were prepared by acid-catalyzed Schiff-base mac-
rocyclization; a range of inorganic acids, including HCl,
HNO3, H3PO4, H2SO4, and CH3COOH, were tested to opti-
mize the yield of the reaction. As expected, the smallest
macrocycle, 12, was obtained in similar yields using any of
the acids. In accord with our previous investigations, the
yield of the imine macrocyclization reaction was in agree-
ment with the affinity of the resulting macrocycles for the
anion, which corresponds to the acid used in the reaction.[31]
This suggests that 12 does not have a remarkable selectivity
towards any particular anion. The condensation of 6 with 9
gave 13 in a high yield only with H2SO4, H3PO4, and HNO3.
The receptors 10 and 11 were prepared by acylation of the
diamines 8 and 9 with a diacyl dichloride derivative of 7, re-
spectively. In the case of the synthesis of 10 (with a smaller
cavity), the chloride anion eliminated during the acylation
reaction acts a template in the macrocyclization reaction,
which occurs in good yield (46%). However, in the prepara-
tion of 11, tetrabutylammonium dihydrogen phosphate was
added as the anion template, which increased the yield from
15 (no template) to 57% (dihydrogen phosphate template).
A similar enhancement in yield was observed previously
for the acylation with pyridine-2,6-dicarbonyl dichloride in
which the preorganization of 9 upon coordination of dihy-
Scheme 2. Structures of building blocks 6–9 and schemes for the synthesis
of receptors 10–13.
Chem. Eur. J. 2008, 14, 9065 – 9073
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9067

E. A. Katayev et al.
drogen phosphate is crucial to the formation of the [1+1]
acylation product.[35] Interestingly, the benefit of using a
template in this reaction was observed only if the concentra-
tion of the starting materials was not less than around 2 mm.
For all the receptors, we succeeded in obtaining single crys-
tals suitable for X-ray analysis of the receptor in the free
form and in some cases as host–guest complexes (see Fig-
ures 1–4). Receptors 10 and 12 with three pyrrole rings were
crystallized as the tetrabutylammonium chloride and hydro-
gen chloride salts, respectively. Interestingly, crystals of 10
were grown by slow diffusion of hexane into a mixture of
the free receptor and tetrabutylammonium dihydrogen
phosphate in chloroform. However, according to the X-ray
crystal structure analysis, the receptor contained a chloride
anion inside the cavity instead of dihydrogen phosphate.
Similarly, crystallization of the free receptor 12 in the same
solvent mixture gave the hydrogen chloride salt 12·HCl. Pre-
sumably, both receptors facilitate the hydrolysis of chloro-
form in the presence of traces of water and chloride com-
plexes are formed as they are more stable in this system.
Analysis of the structures of the receptors revealed that
amido-imine receptors are more planar than amido recep-
tors, probably due to conjugation in the imino-bipyrrole
fragment. As can be clearly seen from Figure 1, the amide
ligand 10 is more flexible than the amido-imine ligand 12
and can adjust its structure for chloride complexation. The
average ClÀ···N distance is smaller for complex [10·ClÀ]-
ACHTREUNG
[TBA+] (3.28 Š) than for complex 12·HCl (3.43 Š), so a
higher affinity for chloride by 10 can be expected. The
former distance is very close to the Cl···N distance found in
a sapphyrin complex, that is, 3.24 Š. The nonplanarity of the
receptors can also be evaluated from a comparison of dihe-
dral angles between the N4 and N5 atoms of two pyrrole
rings. For 10 this angle is 388 and for 12 it is 288. Unusual
acetate coordination to receptor 12 was found in the com-
Figure 1. ORTEP-POVray rendered view of the structures of the com-
plex a) [10·ClÀ][TBA+]·1.25CHCl3 and b) 12·HCl·CHCl3·Et2O as inferred
ACHTREUNG
from the single-crystal X-ray structure analysis. The tetrabutylammonium
cation for [10·ClÀ][TBA+]·1.25CHCl3, exterior hydrogen atoms, and all b
AHCTREUNG
substituents on the pyrrole rings, as well as solvent molecules have been
removed for clarity.
ACHTREUNG
plex [12·OAcÀ][Et3N+], which crystallized upon neutraliza-
tion with triethylamine of the reaction performed with
acetic acid as template (Figure 2). Instead of the coordina-
tion of one oxygen atom to the receptor, as in sapphyrins,[36]
the 2,5-diamidopyrrole moiety provides NH donor sites for
coordination of the second acetate oxygen atom. The ligand
has a bent structure to provide strong hydrogen-bonding
À
À
À
with the angles O4···H N4, O3···H N7, and O3···H N1
equal to 169, 141, and 175 Š, respectively.
According to the X-ray crystal structure analysis, receptor
11 is a dimer in which two molecules are stabilized by three
intermolecular hydrogen bonds between amide oxygen
À
atoms and pyrrole NHs: O4···H N12 (2.11, 2.903(3) Š,
À
À
1508), O5···H N4 (2.40, 2.961(3) Š, 1228), O7···H N1 (2.10,
2.877(3) Š, 1478). Molecule 11 is observed to adopt a bowl-
like conformation (Figure 3), which is determined by two in-
À
tramolecular hydrogen bonds: O2···H N2 (1.96, 2.722(2) Š,
o
o
À
À
144 ), O4···H N6(1.99, 2.765(2) Š, 146 ) and O5···H N11
o
o
Figure 2. ORTEP-POVray rendered view of the structure of the complex
À
(1.93, 2.697(3) Š, 145 ), O7···H N15 (1.95, 2.729(3) Š, 146 )
for the two independent molecules, respectively, in which
the amide groups are in trans positions. Intramolecular hy-
drogen-bonding introduces high structural rigidity into the
[12·OAcÀ][Et3NH+]MeOH as inferred from the single- crystal X-ray
ACHTREUNG
structure analysis. The triethylammonium cation, exterior hydrogen
atoms, and all b substituents on the pyrrole rings have been removed for
clarity.
9068
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9065 – 9073

Selective Phosphate Binding
FULL PAPER
Figure 4. ORTEP-POVray rendered view of the structures of a) 13·-
AHCTRE(GNU DMSO)4 and b) 13·H2SO4·2CH2Cl2·0.25C6H14 as inferred from a single-
crystal X-ray structure analysis. The exterior hydrogen atoms, all b sub-
stituents on the pyrrole rings, and solvent molecules have been removed
for clarity.
Figure 3. ORTEP-POVray rendered view of a) the structure of the dimer
of 11 (without b substituents on the pyrrole rings and exterior hydrogen
atoms) and b) the molecule 11 alone (without exterior hydrogen atoms)
as inferred from a single-crystal X-ray structure analysis.
the pyrrole planes and the cavity size upon complexation
with tetrabutylammonium hydrogen sulfate. In 13·ACHTRE(UNG DMSO)4,
the dihedral angles between the N4 and N5 pyrroles and the
N1 and N8 pyrroles are 61.2 and 55.28, respectively, and the
receptor and resembles the g turns often found in cyclic nat-
ural peptides. In contrast to the solid phase, in solution no
dimerization was observed according to UV/Vis and
1H NMR dilution measurements.
distances between N1···N4 and N2···N7 are 5.454 and
À
8.423 Š respectively. In the complex [13·HSO4 ]
A
corresponding measurements are 37.3 and 44.68 and 5.901
and 7.290 Š. Thus, upon complexation, the receptor dramat-
ically changes its conformation. The proton coming from the
hydrogen sulfate is statistically disordered over two sites at
the two nitrogen atoms with equal occupancies. Thus, de-
spite the fact that the sulfate dianion forms only seven hy-
drogen bonds, its superimposed hydrogen-bonding network
The amido-imine macrocycle 13 has a completely differ-
ent structure. It was crystallized from DMSO (free ligand)
and chloroform (with TBA+HSO4 ), which resulted in the
À
À
formation of complexes 13·
A
ACHTREUNG
[TBA+], respectively (Figure 4). Knowledge of the struc-
2À
tures of complexes with geometrically different guests pro-
vides an opportunity to assess the conformational flexibility
of the receptor. Thus, 13 coordinates to four DMSO mole-
cules, one on each pyrrolic NH donor site, with additional
stabilization provided by the amide groups (N2 and N7).
The pyrrole rings are rotated in opposite directions so two
resembles that of cyclo[8]pyrroleH22+·SO4 with eight hy-
drogen bonds reported by Sessler and co-workers[37] Nota-
bly, 13 (four pyrrole NH donors, two amide NH donors, and
two imine N acceptors) can be considered an analogue of
cyclo[8]pyrrole (six pyrrole NH donors and two imine-like
N acceptors) in terms of the number and nature of the coor-
dination sites.
À
DMSO molecules are coordinated from one side of the N5
À
À
N4 N1 N8 plane and the other two from the other side. It
is interesting to follow the changes in the angles between
Chem. Eur. J. 2008, 14, 9065 – 9073
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9069

E. A. Katayev et al.
Anion-binding studies: To investigate the anion-binding
properties of these molecules, we performed H NMR titra-
tions with the tetrabutylammonium salts of inorganic anions
in a [D6]DMSO/0.5% water solution (Table 1). The stability
equivalent of hydrogen sulfate. It may be assumed that in
DMSO 12 abstracts a proton from hydrogen sulfate concur-
rent with the binding event. To sum up, the three-pyrrole re-
ceptors show an affinity and selectivity for bifluoride, ace-
tate, and dihydrogen phosphate anions. The data were also
supported by UV/Vis titration experiments (see the Sup-
porting Information).
1
constants were determined by using the HYPERNMR and
[38]
HYPERQUAD 2006programs.
Macrocycle 10 binds fluo-
In the initial titration experiments with the larger recep-
tors 11 and 13, we were only able to calculate the binding
constants with tetrabutylammonium chloride, acetate, and
hydrogen sulfate (Table 1). Recall that the template synthe-
sis of 13 was also efficient in the presence of nitric acid.
However, according to the 1H NMR and UV/Vis titration
experiments, no apparent binding was observed with tetra-
butylammonium nitrate. The binding stoichiometry of ace-
tate complexation showed that the presence of more NH
donor binding sites in 11 resulted in a 1:2 host–guest interac-
tion, whereas for 13 a 1:1 stoichiometry was observed
(Table 1). Complexation of dihydrogen phosphate by 11 and
Table 1. Anion affinities (logK11) for the receptors 10–13 determined
from 1H NMR titrations carried out at 298 K in [D6]DMSO/0.5 water so-
lution.
Anion/Re- 10
12
11
13
ceptor
H2FÀ
logK11 =5.70(5) logK21 =5.50(3) n.d.[a]
logK12 =5.23(4) logK11 =2.22(4)
n.d.[a]
ClÀ
2.38(2)
1.97(1)
1.63(8)
0.89(1)
CH3COOÀ c.e.[a] logK21 >7 c.e.[a] logK21 >7 logK11 =3.54(6) 3.07(4)
logK12 =2.09(5)
À
NO3
n.d.[a]
1.64(6)
c.e.[a]
n.d.[a]
<1
n.d.[a]
n.d.[a]
À
HSO4
H2PO4
1.84(7)
2.18(1)
À
logK21 =7.44(5) >4[b]
logK11 =2.21(4)
2.44(4)[b]
1
13 was very strong and was observed in the H NMR titra-
tion as changes in both the integral intensities and the shifts
of the peaks. However, the binding curves calculated on the
basis of these data were found to be phase-transition-like
curves that did not allow us to calculate the binding con-
stants. The same difficulties were encountered in the UV/
Vis titration experiments. The phase-transition curve is char-
acteristic of a host with multiple binding sites or with highly
cooperative binding.[40] Thus, we suggest that upon addition
of the phosphate, the affinity of the receptor causes aggrega-
tion processes both before and after the saturation point of
1:1 ligand/phosphate. Interesting binding trends were also
[a] n.d. = no apparent binding seen, as reflected in the lack of observa-
ble change in the 1H NMR spectrum upon anion addition; c.e. = com-
plex equilibrium, that could not be fitted by any binding model or the
errors was very large. [b] Binding constants determined from competitive
experiments in the presence of two equivalents of tetrabutylammonium
acetate.
ride with a 1:2 stoichiometry, whereas 12 binds with a 1:1
stoichiometry according to a Job plot analysis. Fitting of the
experimental data reveals multiple equilibria corresponding
À
1
to HF2 binding rather than FÀ. The unusually high affinity
noted through the changes of the H NMR titration spectra
for fluoride can be explained in terms of bifluoride genera-
of 11 and 13 upon addition of tetrabutylammonium acetate
tion in situ upon titration with TBA+FÀ
ACHTRE(UGN H2O)3. Fluoride is
or dihydrogen phosphate. In the 1H NMR spectrum of 11
known to be a strong base in organic solvents, particularly
in the presence of water.[39] As receptor 10 does not have a
proton acceptor group, the first binding event will be the
there were only two signals at d=11.69
AHCTREUNG
ACHTRE(UNG 4H) and 9.17-
the addition of 2 equiv of acetate, four additional peaks ap-
peared, six in total, at d=13.74, 12.95, 12.45, 11.38, 10.51,
and 10.20 ppm. This means that the coordination of two ace-
tates stabilizes the unsymmetrical conformation of the free
receptor. Similarly, after addition of 1 equiv of dihydrogen
phosphate, six signals were observed at d=13.84, 11.69,
11.52, 10.75, 10.28, and 9.22 ppm. However, the addition of
more TBAH2PO4 reduced the number of signals to two (at
d=13.84 and 11.52 ppm). We thus suggest that 11 has two
binding sites for XO2 (e.g., acetate) complexation, but can
adjust its binding sites around XO4 species with high geo-
metrical complementarity. In contrast, upon titration of re-
ceptor 13 with acetate only four (pyrrole and amide) signals
were observed to be shifted downfield. When TBAH2PO4
was added to 13, the number of signals decreased to two,
probably due to fast proton exchange on the NMR time-
scale caused by imine protonation. To determine the desired
binding constants for TBAH2PO4 and TBAOAc we used a
competitive titration method[41] that we had successfully ap-
plied in recent work on acetate-selective receptors.[33] The
macrocycles were mixed with 2 equiv of tetrabutylammoni-
À
binding of two hosts (S) to the HF2 anion: 2S+
À
À
À
HF2 QS2HF2
A
(1:1 stoichiometry); S2HF2À +HF2 QS2-
À
(HF2 )2 (1:2 stoichiometry overall). In contrast, 12 has a
proton acceptor group and the first binding event should be
À
À
protonation of the receptor: S+HF2 QSHF2 (2:1 stoichi-
ometry); SHF2À +SQS2HF2 (1:1 stoichiometry overall). In-
À
terestingly, both receptors bind acetate and dihydrogen
phosphate with a 2:1 stoichiometry with affinities of 107 mÀ1
and have a rather low affinity for the hydrogen sulfate anion
(Ka =43mÀ1 for 10). We were unable to determine the
À
H2PO4 binding constant for 10 due to very strong binding
and a rather complex equilibrium that could not be de-
scribed by the proposed models. The equilibrium is also ac-
companied by a broadening of NMR signals and a down-
field shift with Dd up to 2 ppm. According to the NMR ti-
tration data, fluoride, chloride, sulfate, and acetate did not
induce changes in the NH donor signals, only a shift of Dd=
0.4–2 ppm. An exception was receptor 12 with hydrogen sul-
fate: in addition to the four signals of the NH donor groups,
a fifth broad peak appeared after addition of more than one
9070
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9065 – 9073

Selective Phosphate Binding
FULL PAPER
um acetate to saturate all the binding sites and subsequently
titrated with tetrabutylammonium dihydrogen phosphate. In
all cases (for 10, 11, and 13) changes in integral intensities
were observed rather than shifts of the peaks. In other
words, the integral intensities of the peaks of the acetate
complex gradually decreased as the peaks of the phosphate
complex increased (Figure 5). We were not able to fit the
binding curve for 10 with any appropriate model, however,
11 and 13 were fitted with a 1:1 model and relative binding
constants of Ka >104 and 275mÀ1, respectively, were deter-
mined (Figure 5). Thus, in our studies, receptor 11 with eight
NH donor binding sites (amide and pyrrolic NHs) was the
most selective for the dihydrogen phosphate anion in
[D6]DMSO/0.5% water solution.
Conclusion
In this work we have developed novel receptors with bind-
ing cavities analogous to the expanded porphyrins sapphyrin
and cyclo[8]pyrrole but with the pyrrole rings linked by
amide and amine bonds rather than through direct linkages.
Their easy and efficient synthesis was accomplished by using
anion-templated [1+1] Schiff condensation and acylation
macrocyclization reactions. Their stability constants with dif-
ferent tetrabutylammonium salts of inorganic acids were de-
1
termined by standard H NMR and UV/Vis titration tech-
niques in DMSO. According to the titration data, the recep-
tors that contain three pyrrole rings exhibit high affinities
(logKa =5–7) for bifluoride, acetate, and dihydrogen phos-
phate, but interact weakly with chloride and hydrogen sul-
fate. Receptors 11 and 13, which contain four pyrrole rings,
exhibit unprecedented affinity and selectivity for dihydrogen
phosphate. The determined binding constants were even
higher than those for known charged or metal-containing re-
ceptors in DMSO solution.[8] The shapes of the binding
curves produced from peak shifts or integral changes upon
titration with tetrabutylammonium dihydrogen phosphate
suggest that the binding event is accompanied by a host–
guest aggregation process. This is in agreement with sap-
phyrin behavior and apparently is an integral feature of NH
donor receptors possessing high affinity. Competitive experi-
ments in which 11 and two equivalents of acetate anion
were titrated against dihydrogen phosphate allowed us to
determine a relative binding constant of logKa =>4. More-
over, the selectivity of receptor 11 is higher by at least three
orders of magnitude in comparison with analogue 13, which
bears two imine bonds. Analysis of the X-ray data of the
free receptor 11 revealed that, in the solid state, it forms
inter- and intramolecular hydrogen bonds. All these facts
support the proposition that the geometry of 11 is highly
complementary to the dihydrogen phosphate anion. The re-
sults of this work have shown that a combination of suitable
coordination sites (or binding motifs) in a complementary
geometry to the target anion and the rigidity of the structure
stabilized by rigid ether linkers (e.g., in 13) or hydrogen-
bonding (e.g., in 11) can lead to highly selective receptors
for dihydrogen phosphate. Such receptors possess a unique
property to encapsulate the phosphate and form a hydro-
phobic shell, and hence are of great interest as phosphate
carriers across membranes. They also offer good prospects
for the investigation and development of, among others, se-
lective nucleotides and phosphorylated amino acids as re-
ceptors and biomimetic catalysts, for example, in phospho-
diester hydrolysis. Work in this area is underway.
Experimental Section
Figure 5. a) Integral changes for 11+2TBAOAc upon addition of (from
the bottom) 0.28, 0.41, 0.53, 0.64, 0.75, 0.86, 0.95, 1.16, and 1.37 equiv-
General methods: All solvents and starting materials were of reagent
grade quality and purchased commercially unless noted otherwise. The
tetrabutylammonium (TBA) salts were dried under high vacuum prior to
use. NMR spectra were recorded on a Bruker Avance 400 MHz spec-
*
alents of TBAH2PO4. In the graph b, denote experimental data and the
curves theoretical data generated by the fitting of experimental data
[38]
points using the HYPERNMR 2006computer program.
Chem. Eur. J. 2008, 14, 9065 – 9073
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9071

E. A. Katayev et al.
trometer and referenced to the solvent peak. Column chromatography
was performed on 60 Š (230–400 mesh) silica gel. 3,3’-Dimethyl-4,4’-di-
phenyl-2,2’-bipyrrole-5,5’-dicarbonyl dichloride was prepared by a similar
method to that described by Sessler and co-workers[29] and was used im-
mediately. Compounds 6[31] and 7–9[42] were prepared as described earlier.
served in the DF maps and refined with fixed positions and isotropic dis-
placement parameters. In some cases, where noted, the contribution to
the scattering by solvent molecules was removed by the use of the utility
SQUEEZE in PLATON98.[44] All calculations were carried out by use of
the SHELXTL PLUS program (PC Versions 5.10 and 6.12).[45]
CCDC-685341, CCDC-685342, CCDC-685343, CCDC-685344, CCDC-
685345, and CCDC-685346 contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
Titration conditions: For NMR titrations, stock solutions of the host mol-
ecule to be studied were prepared in [D6]DMSO/0.5% water solution,
with the final concentrations being in the range of (1.3–1.7)”10À3 molLÀ1
for 10–13. Stock solutions of the guest were prepared by dissolving 4–
8 equiv of the TBA salts of the anions in question in 1 mL of the host
stock solution (1”10À2 molLÀ1). The general procedure for the UV/Vis
binding studies involved the preparation of stock solutions with concen-
trations 100-fold more dilute than those used for the NMR experiments,
sequential addition of the titrant (anionic guest) to a 1 mL sample of the
host stock solution in the spectrometric cell, and monitoring of the
changes in the spectral features. The total number of data points in both
NMR and UV/Vis experiments was between 10 and 40, depending on the
stoichiometry of the complexation; for a presumed 1:1 complex 10–15
data points were usually measured. The data points were then collated
and combined to produce plots that, in turn, were processed by the HY-
PERNMR and HYPERQUAD computer programs.[38]
Receptor 10: A freshly prepared solution of 3,3’-dimethyl-4,4’-diphenyl-
2,2’-bipyrrole-5,5’-dicarbonyl dichloride (144 mg in 20 mL THF) was
added dropwise over 10 min to a THF solution (50 mL) containing
N2,N5-bis(2-aminophenyl)-3,4-dimethyl-1H-pyrrole-2,5-dicarboxamide (8)
(100 mg, 0.275 mmol) and pyridine (410 mg). The solution was stirred for
24 h at RT and evaporated under reduced pressure. The residue was dis-
solved in a MeOH/CH2Cl2 mixture (5:95, v/v) and filtered through a
small plug of silica. The filtrate was evaporated and recrystallized from
MeOH to yield 92 mg of 10 as
a
white powder (46%). 1H NMR
(600 MHz, [D6]DMSO): d=11.87 (s, 1H), 11.78 (s, 2H), 9.42 (s, 2H),
9.30 (s, 2H), 7.93 (d, J=7.2 Hz, 2H), 7.55 (d, J=7.2 Hz, 2H), 7.30 (m,
8H), 7.27 (m, 2H), 7.17 (m, 4H), 2.35 (s, 6H), 1.91 ppm (s, 6H);
13C NMR (600 MHz, [D6]DMSO): d=160.1, 158.7, 135.3, 132.54, 129.9,
129.3, 127.4, 126.3, 125.9, 125.6, 124.4, 124.1, 124.0, 123.5, 123.0, 122.7,
122.4, 119.0, 10.3, 10.0 ppm; ESI-MS(+): m/z: 728 [M]+; elemental analy-
sis calcd (%) for C44H37N7O4: C 72.61, H 5.12, N 13.47; found: C 72.54, H
4.89, N 13.42.
X-ray diffraction analysis: Data were collected on Bruker APEX II and
SMART 1K diffractometers equipped with CCD area detectors using
graphite-monochromated MoKa radiation (l=0.71073 Š). Data reduction
was performed by using SAINTPlus (v. 6.01 and 6.2).[43] Crystallographic
data are presented in Table 2. The structures were solved by direct meth-
ods and refined by full-matrix least-squares on F2 with anisotropic dis-
placement parameters for the non-hydrogen atoms. The hydrogen atoms
on carbon were refined in ideal positions with isotropic displacement pa-
rameters set to 1.2Ueq of the attached atom (1.5Ueq for the methyl hydro-
gen atoms). The hydrogen atoms bound to the nitrogen atoms were ob-
Receptor 11: A freshly prepared solution of 3,3’-dimethyl-4,4’-diphenyl-
2,2’-bipyrrole-5,5’-dicarbonyl dichloride (144 mg, 0.33 mmol in 20 mL
THF) was added dropwise over 10 min to a THF solution (50 mL) con-
taining N2,N5-bis(2-aminophenyl)-3,3’-dimethyl-4,4’-diphenyl-2,2’-bipyr-
Table 2. Crystallographic data for the investigated complexes.
10·
A
[TBA+]·
11·1.875CH2Cl2·
G
N
12·
G
[OAcÀ]· 13·4
U
[13H+][TBA+][SO42À]·
ACHTREUNG ACHTREUNG ACHTREUNG
1.25CHCl3
0.5C5H12
CH3OH
2CH2Cl2·0.25C6H14
empirical formula
formula mass
crystal size [mm]
diffractometer
temperature [K]
crystal system
space group
a [Š]
b [Š]
c [Š]
a [8]
b [8]
C61.25H74.25Cl4.75N8O4 C64.375H57.75Cl3.75N8O4 C43H53Cl4N7O3 C47H64N8O5
C62H76N8O6S4
1375.96
0.24”0.21”0.15 0.30”0.25”0.20
SMART 1K
120
monoclinic
P21/n
16.1783(12)
21.7496(15)
18.3369(13)
90
104.211(1)
90
6254.8(8)
4
1.173
C73.5H96.5Cl4N9O6S
1154.92
0.30”0.25”0.25
SMART 1K
120
1140.37
857.72
821.061157.55
0.30”0.25”0.20
SMART 1K
100
0.18”0.15”0.12 0.21”0.18”0.15
SMART 1K
120
APEX II
100
APEX II
100
monoclinic
P21/n
15.4448(12)
21.4657(17)
24.1556(19)
90
103.401(2)
90
monoclinic
C2/c
triclinic
P1
monoclinic
P21/c
monoclinic
P21/c
¯
34.288(8)
18.950(5)
26.151(6)
90
130.299(3)
90
16.8233(14)
18.3589(16)
22.7384(19)
67.581(5)
71.641(5)
72.564(5)
6031.3(9)
4
13.699(5)
17.068(5)
20.866(6)
90
104.593(5)
90
15.456(2)
17.708(3)
18.047(3)
90
114.096(3)
90
g [8]
V [Š3]
12960(6)
8
4721(3)
4
4508.9(12)
4
7790.3(11)
4
Z
1calcd [gcmÀ3
]
1.184
4884
1.2561.207
2383
1.210
1808
1.229
F
A
1768
2464
2930
]
0.263
0.239
0.294
0.080
0.207
0.232
q range [8]
index range
2.04–23.01
À37ꢀhꢀ37
À20ꢀkꢀ20
À28ꢀlꢀ28
42445
1.30–26.23
À20ꢀhꢀ20
À22ꢀkꢀ22
À28ꢀlꢀ28
58090
1.95–25.03
À16ꢀhꢀ16
À20ꢀkꢀ20
À24ꢀlꢀ24
32459
1.85–25.02
À18ꢀhꢀ18
À21ꢀkꢀ21
À21ꢀlꢀ21
38414
1.87–26.06
À19ꢀhꢀ19
À26ꢀkꢀ26
À22ꢀlꢀ22
44508
1.65–27.00
À19ꢀhꢀ19
À27ꢀkꢀ27
À30ꢀlꢀ30
105078
refl. collected
independent
8925
23940
8071
7818
12252
16750
refl.
[Rint=0.092]
2850
0.111
[Rint=0.049]
1487643189
0.062
[Rint=0.073]
336 8389
0.074
[Rint=0.069]
13226
0.092
[Rint=0.058]
[Rint=0.038]
no. of rflns with I>2s(I)
R1 [I>2s(I)]
0.056
0.056
wR2 (all data)
0.277
0.158
0.189
0.215
0.160
0.143
data/restraints/parameters
8925/57/713
1.041
0.783/À0.398
23940/18/1287
1.018
0.490/À0.604
8071/11/431
1.004
0.568/À0.663
7818/32/538
1.009
0.536/À0.581
12252/0/735
1.002
0.370/À0.390
16750/12/783
1.001
0.899/À0.732
GOF on F2
À3
largest diff. peak/hole [eŠ
]
9072
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 9065 – 9073

Selective Phosphate Binding
FULL PAPER
role-5,5’-dicarboxamide (9) (160 mg, 0.275 mmol), pyridine (410 mg), and
TBAH2PO4 (112 mg, 0.33 mmol). The solution was stirred for 24 h at RT
and evaporated under reduced pressure. The residue was dissolved in a
MeOH/CH2Cl2 mixture (5:95 v/v) and filtered through a small plug of
silica. The filtrate was evaporated and recrystallized from MeOH/CH2Cl2
to yield 148 mg of 11 as a white powder (57%). 1H NMR (600 MHz,
[D6]DMSO): d=11.69 (s, 4H), 9.19 (s, 4H), 7.37 (m, 24H), 7.00 (m, 4H),
1.91 ppm (s, 12H); 13C NMR (600 MHz, [D6]DMSO): d=160.5, 142.1,
135.5, 130.7, 128.61, 127.2, 126.0, 125.3, 123.8, 123.0, 118.0, 116.7,
11.24 ppm; ESI-MS (+): m/z: 945 [M]+; elemental analysis calcd (%) for
C60H48N8O4: C 76.25, H 5.12, N 11.86; found C, 76.20, H 5.10, N 11.83.
[10] S. Kubik, C. Reyheller, S. Stꢁwe, J. Inclusion Phenom. Macrocyclic
Chem. 2005, 52, 137–187.
[11] M. D. Lankshear, P. D. Beer, Acc. Chem. Res. 2007, 40, 657–668.
[12] M. D. Lankshear, P. D. Beer, Coord. Chem. Rev. 2006, 250, 3142–
3160.
[13] P. A. Gale, R. Quesada, Coord. Chem. Rev. 2006, 250, 3219–3244.
[14] T. J. Wedge, M. F. Hawthorne, Coord. Chem. Rev. 2003, 240, 111–
128.
[15] M. Bru, I. Alfonso, M. I. Burguete, S. V. Luis, Angew. Chem. 2006,
118, 6301–6305; Angew. Chem. Int. Ed. 2006, 45, 6155–6159.
[16] S. L. Tobey, B. D. Jones, E. V. Anslyn, J. Am. Chem. Soc. 2003, 125,
4026–4027.
[17] M. Allevi, M. Bonizzoni, L. Fabbrizzi, Chem. Eur. J. 2007, 13, 3787–
3795.
[18] P. T. Gunning, Org. Biomol. Chem. 2005, 3, 3877–3879.
[19] M. S. Han, D. H. Kim, Angew. Chem. 2002, 114, 3963–3965; Angew.
Chem. Int. Ed. 2002, 41, 3809–3811.
[20] V. Krꢂl, H. Furuta, K. Shreder, V. Lynch, J. L. Sessler, J. Am. Chem.
Soc. 1996, 118, 1595–1607.
[21] S. K. Kim, N. J. Singh, S. J. Kim, H. G. Kim, J. K. Kim, J. W. Lee,
K. S. Kim, J. Yoon, Org. Lett. 2003, 5, 2083–2086.
Receptor 12: HCl (38%, 0.1 mL) was added to a mixture of diamine 8
(363 mg, 1 mmol) and dialdehyde 6 (300 mg, 1 mmol) in MeOH (50 mL).
On heating the solution at reflux for 20 min, its color changed to orange-
red. An excess of triethylamine was added to this boiling solution and
the mixture was cooled to 08C. The resulting precipitate was filtered off
and recrystallized from MeOH to yield 463 mg (74%) of a yellow
powder. 1H NMR (600 MHz, [D6]DMSO): d=11.84 (s, 1H), 10.88 (s,
2H), 9.48 (s, 2H), 8.55 (s, 2H), 7.88 (m, 2H), 7.42 (m, 2H), 7.15 (m, 4H),
3.32 (m, 4H), 2.69ACHTREUNG(m, 4H), 2.20 (s, 6H), 2.06 (s, 6H), 1.56 ACHTREU(GN m, 4H),
0.97 ppm (t, J=12 Hz, 6H); 13C NMR (600 MHz, [D6]DMSO): d=10.0,
11.1, 13.8, 24.4, 25.4, 45.7, 116.6, 117.6, 123.7, 124.5, 125.0, 125.9, 126.4,
128.7, 131.7, 132.8, 141.8, 145.2, 160.1 ppm; ESI-MS(+): m/z: 627 [M]+;
elemental analysis calcd (%) for C38H41N7O2: C 72.70, H 6.58, N 15.62;
found: C 72.46, H 6.67, N 15.53.
[22] J. Y. Kwon, N. J. Singh, H. N. Kim, S. K. Kim, K. S. Kim, J. Yoon, J.
Am. Chem. Soc. 2004, 126, 8892–8893.
[23] P. Schiessl, F. P. Schmidtchen, J. Org. Chem. 1994, 59, 509–511.
[24] P. S. Lakshminarayanan, I. Ravikumar, E. Suresh, P. Ghosh, Chem.
Commun. 2007, 5214–5216.
Receptor 13: Concentrated nitric acid (0.2 mL) was added to a solution
of diamine 9 (581 mg, 1 mmol) and dialdehyde 6 (300 mg, 1 mmol) in
MeOH (70 mL) at reflux. The solution was heated at reflux for a further
20 min. An excess triethylamine was added to this boiling solution and
the mixture was cooled to 08C. The resulting precipitate was filtered off
and recrystallized from MeOH to yield 515 mg (61%) of a yellow
powder. 1H NMR (600 MHz, [D6]DMSO): d=11.57 (s, 2H), 11.35 (s,
2H), 9.12 (s, 2H), 8.47 (s, 2H), 8.47 (s, 2H), 7.08–7.40 (m, 18H), 2.67 (m,
4H), 1.91 (s, 6H), 1.85 (s, 6H), 1.15 (m, 4H), 0.93 ppm (t, J=12 Hz,
6H); 13C NMR (600 MHz, [D6]DMSO): d=10.3, 18.4, 55.9, 118.5, 122.1,
123.3, 124.3, 125.0, 126.2, 127.4, 129.4, 130.2, 130.4, 135.0, 158.7 ppm;
ESI-MS (+): m/z: 845 [M]+; elemental analysis calcd (%) for
C54H52N8O2: C 76.75, H 6.20, N 13.26, found: C 76.80, H 6.37, N 13.21.
[25] A. N. Leung, D. A. Degenhardt, P. Bꢁhlmann, Tetrahedron 2008, 64,
2530–2536.
[26] J.-H. Liao, C.-T. Chen, J.-M. Fang, Org. Lett. 2002, 4, 561–564.
[27] I. E. D. Vega, S. Camiolo, P. A. Gale, M. B. Hursthouse, M. E. Light,
Chem. Commun. 2003, 1686–1687.
[28] I. E. D. Vega, P. A. Gale, M. B. Hursthouse, M. E. Light, Org.
Biomol. Chem. 2004, 2, 2935–2941.
[29] A. Andrievsky, F. Ahuis, J. L. Sessler, F. Vçgtle, D. Gudat, M.
Moini, J. Am. Chem. Soc. 1998, 120, 9712–9713.
[30] J. L. Sessler, E. Katayev, G. D. Pantos, Y. A. Ustynyuk, Chem.
Commun. 2004, 1276–1277.
[31] E. A. Katayev, G. D. Pantos, M. D. Reshetova, V. N. Khrustalev,
V. M. Lynch, Y. A. Ustynyuk, J. L. Sessler, Angew. Chem. 2005, 117,
7552–7556; Angew. Chem. Int. Ed. 2005, 44, 7386–7390.
[32] E. A. Katayev, J. L. Sessler, V. N. Khrustalev, Y. A. Ustynyuk, J.
Org. Chem. 2007, 72, 7244–7252.
[33] E. A. Katayev, N. V. Boev, V. N. Khrustalev, Y. A. Ustynyuk, I. G.
Tananaev, J. L. Sessler, J. Org. Chem. 2007, 72, 2886–2896.
[34] B. L. Iverson, K. Shreder, V. Krꢂl, P. Sansom, V. Lynch, J. L. Sessler,
J. Am. Chem. Soc. 1996, 118, 1608–1616.
[35] E. A. Katayev, E. N. Myshkovskaya, N. V. Boev, V. N. Khrustalev,
Supramol Chem. 2008, DOI: 10.1080/10610270701561342.
[36] J. L. Sessler, J. M. Davis, Acc. Chem. Res. 2001, 34, 989–997.
[37] D. Seidel, V. Lynch, J. L. Sessler, Angew. Chem. 2002, 114, 1480–
1483; Angew. Chem. Int. Ed. 2002, 41, 1422–1425.
Acknowledgements
This work was supported by the Russian Academy of Sciences, program
“Chemistry and Physico-Chemistry of Supramolecular Systems and
Atomic Clusters”, and the Russian Foundation for Basic Research. We
would like to thank Yu. A. Pirogov (Research Center for Magnetic To-
mography and Spectroscopy) for helpful discussions.
[1] J.-M. Lehn, Chem. Soc. Rev. 2007, 36, 151–160.
[2] J. L. Sessler, P. A. Gale, W.-S. Cho, Anion Receptor Chemistry,
Royal Society of Chemistry, Cambridge, 2006.
[3] K. Severin, Top. Organomet. Chem. 2006, 17, 123–142.
[4] S. Stadlbauer, A. Riechers, A. Späth, B. Kçnig, Chem. Eur. J. 2007,
13, 2536–2541.
[38] P. Gans, A. Sabatini, A. Vacca, Talanta 1996, 43, 1739–1753.
[39] S. O. Kang, R. A. Begum, K. Bowman-James, Angew. Chem. 2006,
118, 8048–8061; Angew. Chem. Int. Ed. 2006, 45, 7882–7894.
[40] K. A. Connors, Binding constants, Wiley, New York, 1987.
[41] B. J. Whitlock, H. W. Whitlock, J. Am. Chem. Soc. 1990, 112, 3910–
3915.
[5] C. Schmuck, Angew. Chem. 2007, 119, 5932–5935; Angew. Chem.
Int. Ed. 2007, 46, 5830–5833;
[42] N. V. Boev, Y. A. Ustynyuk, Russ. J. Org. Chem. 2007, 43, 297–304.
[43] Bruker SAINTPlus, Data Reduction and Correction Program, v.
6.01, Bruker AXS, Madison, WI, 1998.
[6] M. J. Hannon, Chem. Soc. Rev. 2007, 36, 280–295.
[7] A. K. H. Hirsch, F. R. Fischer, F. Diederich, Angew. Chem. 2007,
119, 342–357; Angew. Chem. Int. Ed. 2007, 46, 338–352.
[8] E. A. Katayev, Y. A. Ustynyuk, J. L. Sessler, Coord. Chem. Rev.
2006, 250, 3004–3037.
[44] PLATON,
A Multipurpose Crystallographic Tool, A. L. Spek,
Utrecht University, The Netherlands, 1998.
[45] SHELXTL, v. 5.10, Structure Determination Software Suite, G. M.
Sheldrick, Bruker AXS, Madison, WI, 1998.
Received: May 6, 2008
[9] S. O. Kang, M. A. Hossain, K. Bowman-James, Coord. Chem. Rev.
2006, 250, 3038–3052.
Published online: September 9, 2008
Chem. Eur. J. 2008, 14, 9065 – 9073
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9073