A flexible solution to anion transport: Powerful anionophores based on a cyclohexane scaffold

Article abstract of DOI:10.1002/anie.201311071

Transmembrane anion carriers (anionophores) have potential for biological activity, including the treatment of channelopathies such as cystic fibrosis. A new family of anionophores has been synthesized, in which three thiourea groups are mounted on a cyclohexane-based scaffold. Though conceptually related to earlier polycyclic systems, these molecules are simpler and far more accessible. Preorganization is somewhat reduced compared to earlier systems, and anion affinities are correspondingly lower. However, transport activities set new records. This surprising performance suggests a role for controlled flexibility in the design of transmembrane anion carriers. Speedy shuttling: New anionophores were developed which are simpler and far more accessible than conceptually related earlier systems. They are also less preorganized and anion affinities are lower (as expected), but transport activities set new records. This surprising performance suggests a role for controlled flexibility in the design of transmembrane anion carriers.

Full text of DOI:10.1002/anie.201311071

Angewandte
Chemie
DOI: 10.1002/anie.201311071
Anion Transport
A Flexible Solution to Anion Transport: Powerful Anionophores Based
on a Cyclohexane Scaffold**
James A. Cooper, Steven T. G. Street, and Anthony P. Davis*
Abstract: Transmembrane anion carriers (anionophores) have
potential for biological activity, including the treatment of
channelopathies such as cystic fibrosis. A new family of
anionophores has been synthesized, in which three thiourea
groups are mounted on a cyclohexane-based scaffold. Though
conceptually related to earlier polycyclic systems, these mole-
cules are simpler and far more accessible. Preorganization is
somewhat reduced compared to earlier systems, and anion
affinities are correspondingly lower. However, transport activ-
ities set new records. This surprising performance suggests
a role for controlled flexibility in the design of transmembrane
anion carriers.
materials (depending on the choice of side-chain R). Herein
we describe two new families of transporters, 3 and 4
(Figure 1), which are accessible in 3 to 4 steps from
T
here is widespread interest in the design of anionophores,
commercially available starting materials. Transporters 4, in
particular, are remarkably active and include the most
powerful anionophore yet reported from our laboratory.
molecules capable of transporting anions across bilayer
[
1]
membranes. Such agents would complement the well-
known cationophore natural products (valinomycin, monen-
sin, etc.), and may result in the discovery of new biological
activities. In particular, there is hope that they could be used
in “channel replacement therapy”, for illnesses such as Best
disease, Bartterꢀs syndrome and (especially) cystic fibrosis.
These genetic conditions result from defective anion chan-
nels, and could perhaps be treated by providing alternative
[1a]
[
2]
[
3]
pathways for transmembrane anion transport. If anion
transporters are to serve practical purposes, it is clearly
desirable to achieve high activities so that only small amounts
are required. However, despite extensive work, there are still
few systems which show proven effectiveness at low trans-
[
4,5]
porter loadings.
1
Among the most active are the cholapods
[
4a,b]
[4c]
and the closely related diaxial diureidodecalins 2.
With optimal substitution, both series are capable of promot-
ing chloride/nitrate exchange in synthetic vesicles at levels of
transporter:lipid ꢀ 1:250000.
Figure 1. Cyclohexane-based tris-(thio)ureas discussed in this paper.
The molecules are labeled according to scaffold (3/4), X (O/S) and
aromatic substituent; for example 3OP corresponds to the tris-N-
phenylurea based on scaffold 3. Side-chains are shown in the axial/
NH-in conformation required for cooperative binding.
The success of 1 and 2 is obtained at a cost, as these rigid,
polycyclic structures require lengthy syntheses. The cholapods
1
are prepared from the natural steroid cholic acid in 7 to 12
steps, depending on the variant. The diureidodecalins 2
require at least 7 steps from commercially available starting
The new designs extend our theme of exploiting 1,5-
diaxial arrangements of H-bond donors to bind and transport
anions. In the case of 1 and 2, this positioning creates
preorganized binding sites which are complementary to
inorganic anions such as chloride, while also preventing
intramolecular hydrogen bonding (see Figure 2a). We rea-
soned that if two binding groups with this arrangement were
effective, then three should be better, and that in this case
some flexibility might be allowable. In particular, the
trisubstituted cyclohexanes 3 seemed interesting. Ab initio
calculations suggested that the tris-(axial/NH-in) conforma-
[
*] J. A. Cooper, S. T. G. Street, Prof. A. P. Davis
School of Chemistry, University of Bristol
Cantock’s Close, Bristol, BS8 1TS (UK)
E-mail: Anthony.Davis@bristol.ac.uk
[
**] This work was supported by the Engineering and Physical Sciences
Research Council through grant number EP/J00961X/1, a Ph.D.
studentship to J.A.C., and a studentship funded via the Bristol
Chemical Synthesis Doctoral Training Centre (EP/G036764/1) to
S.T.G.S.
ꢁ
tion shown can form six hydrogen bonds to Cl with lengths in
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201311071.
[6]
[7]
the range 2.6–2.8 ꢁ. Although slightly longer than ideal,
Angew. Chem. Int. Ed. 2014, 53, 5609 –5613
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5609

.
Angewandte
Communications
Figure 2. a) Conformational restrictions in diaxial diureidodecalins 2.
Rotation about the axial CꢁN bonds is strongly disfavored by steric
interactions with axial CH. Intramolecular H-bonding is thus disal-
lowed. b) Similar restrictions apply in 3 and 4. Attempts to form
intramolecular H-bonds are prevented by axial CH.
[
6]
Figure 3. Calculated C3 symmetrical structure for chloride bound to
4
ꢁ
SP. NH···Cl distances are shown in green. A transparent surface
highlights the Me···S interactions which compress the binding site,
thus shortening and strengthening the H-bonds.
these should provide substantial binding energy. Rotations
would be possible about bonds a and b (see Figure 1), and the
[
6]
binding conformation would not be favored. However, it
could presumably be adopted in the presence of substrate.
Intramolecular H-bonding would be prevented as for other
1
,5-diaxial systems (Figure 2b). The tris-urea 3OP was
known, having been synthesized from commercial triacid 5
[
8]
in just four steps. We also realized that the hexamethyl
analogues 4 could be still more promising. A short synthesis
from 5 seemed realistic (see later). The lipophilicity of the
added methyl groups would favor applications in membranes,
and binding studies in organic solvents. Moreover, introduc-
tion of the methyl groups would promote binding in two ways.
Firstly, they would alter the balance between side-chain
conformations, favoring the axial/NH-in arrangement. This
was demonstrated by calculations on the single-armed
analogues 6 and 7. Although not the ground state, the axial/
Scheme 1. Synthesis of triamines 9 and 13: a) SOCl , DMF, 908C, then
2
NH , DCM, 97%; b) SOCl , DMF, 458C, 60%; c) H (1 atm), Raney
3
2
2
Ni, 10% NH OH/MeOH, 75%; d) SOCl , DMF, 908C, then MeOH,
4
2
Pyridine, 708C, 99%; e) MeMgBr, THF, 81%; f) TMSN , BF ·Et O,
3
3
2
CHCl , 49%; g) H (1 atm), Pd/C, EtOH, 93%.
3
2
overall yield. For the present work we developed a new route
[6]
via hydrogenation of trinitrile 8, as shown in Scheme 1.
Starting material 8 is commercially available, but can also
be prepared from 5 by amidation and dehydration. The novel
triamine 13 was synthesized in three steps from triester 10
(Scheme 1). Treatment of 10 with methylmagnesium bromide
gave triol 11, which was subjected to azide displacement using
ꢁ
1
NH-in conformation for 7 was 3.6 kJmol closer to baseline
[
6]
than for 6. Secondly the methyl groups would strengthen the
H-bonding by compressing the binding site, pushing the O/S
downwards and thus moving the Ar-NH groups inwards.
Calculations suggested that this effect could shorten some
NH···Cl distances to 2.54 ꢁ, within the optimum range
predicted from crystallography (see Figure 3).
[9]
Me SiN /BF .Et O as described by Koziara and Zwierzak.
3
3
3
2
[
10]
The product triazide 12 was then converted to triamine 13
by catalytic hydrogenation. The overall yield from 10 to 13
was 37%. Triester 10 is commercially available, but may also
be prepared economically from 5 by simple esterification
ꢁ
[7]
The syntheses of 3 and 4 required practical routes to tris-
aminoalkyl)cyclohexanes 9 and 13. The literature prepara-
[
8]
(
(SOCl then MeOH).
2
tion of 9 involved conversion of 5 to the corresponding tris-
Triamines 9 and 13 were treated with aryl isocyanates and
amide followed by reduction by LiAlH , proceeding in 10%
isothiocyanates to give the tris-(thio)ureas listed in Table 1.
4
5
610
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 5609 –5613

Angewandte
Chemie
ꢁ
1
Table 1: Chloride binding affinities (K , m ) and transport data for the
compounds discussed in this paper.
rable to 14, was found to be 18-times weaker than the
cholapod. Where solubility permitted, binding constants to
a
+
ꢁ
ꢁ
1 [a]
ꢁ
ꢁ
Compound
Ka [m
]
Cl /NO3 exchange
Et N Cl in chloroform were measured using a variant of
4
[
b]
[13]
(
[s
transporter:lipid=1:2500)
Cramꢀs extraction method, as used previously for chola-
pods
[
c]
ꢁ1
[d]
[14]
[4c]
I
]
t
1
[s]
=
2
and other powerful receptors.
The values were
[
[
e]
f]
[e]
[f]
considerably higher than those in wet DMSO (by ca. 5 orders
of magnitude) but the trends were the same. For example
3
3
3
3
3
4
4
4
4
4
4
1
OP
OF2
SF
SN
SF2
OP
OF2
SP
27
57
68
79
[
15]
6
7
0.00091
0.0014
0.0046
440
340
3SF2, 4SF2 and 14 gave apparent K = 2.4 ꢂ 10 , 3 ꢂ 10 and
a
9
ꢁ1
4
.3 ꢂ 10 m , respectively. Once again, it seemed that the
99
120
methyl groups in 4 improved preorganization, but not enough
to compete with the rigid steroidal scaffold.
[
e]
[e]
180
390
340
400
480
670
0.036
0.0017
0.029
0.01
0.065
0.093
16
190
22
66
9
As we planned to test for transport using chloride/nitrate
exchange, the binding properties towards nitrate were also
SF
ꢁ
ꢁ
SN
SF2
4
studied. Calculations on 3SP·NO3 and 4SP·NO3 revealed C3
symmetrical structures with NH···O distances of 2.07–
12000
6
[6]
1
2
.34 ꢁ.
H NMR titrations of 4SN and 4SF2 with
+
ꢁ
1
+
ꢁ
3
[
a] To Bu N Cl in [D ]DMSO/H O, 200:1, measured by H NMR
Bu N NO in [D ]DMSO/H O, 200:1 produced evidence of
4 6 2
4
6
2
titration. T=298 K. [b] In 200 nm vesicles formed from POPC/choles-
terol (7:3) measured by following decay of lucigenin fluorescence F.
binding (downfield shifts of NH signals), but with affinities
too small for meaningful analysis. This mirrors previous work
which suggests that nitrate is especially difficult to bind in this
[
c] Initial rate of change of relative fluorescence, F/F . [d] Decay half-life
0
of F/F . [e] Not measured due to low solubility. [f] Fluorescence decay
0
[5g]
+
ꢁ
medium. The affinities of 3SF2, 4SF2 and 14 to Et N NO
4
3
indistinguishable from background.
in chloroform were measured by extraction, giving values of
5
6
8
ꢁ1
5
.2 ꢂ 10 , 2.5 ꢂ 10 and 1.9 ꢂ 10 m , respectively. Thus all three
All could be dissolved in [D ]DMSO for binding studies (see
receptors bound nitrate somewhat more weakly than chlo-
ride, in line with previous observations.
6
[
4b,14b]
below), but solubilities in less polar solvents varied consid-
erably. In general solubility was favored by X = S (rather than
O), methylated scaffold 4 (rather than 3) and trifluoromethy-
lated aryl groups, especially F2. Compound 4SF2 could be
dissolved in chloroform to the level of ca. 1 mm. Cholapod 14
was synthesized as a comparison using previously reported
Anion transport by the new compounds was investigated
in large unilamellar vesicles (200 nm mean diameter) using
[
4,6]
the previously reported “lucigenin method”.
Briefly,
vesicles containing aqueous NaNO3 (225 mm) and the
halide-sensitive fluorescent dye lucigenin were prepared
from a 7:3 ratio of 1-palmitoyl-2-oleoylphosphatidylcholine
[11]
procedures.
(
POPC) and cholesterol (7:3), with a small amount of added
transporter. Suspensions of these vesicles in aqueous NaNO3
225 mm) were placed in a fluorescence spectrometer. An
(
external pulse of sodium chloride (25 mm) was added and the
ꢁ
influx of Cl was followed through the decay in lucigenin
fluorescence. The transport rates were quantified in two ways.
Firstly the initial rate of change of relative fluorescence I was
determined by empirical curve fitting, and secondly the
approximate half-life t = was estimated through fitting to
1
2
a single exponential decay function.
An early series of experiments employed the (thio)ureas
at loadings of transporter:lipid = 1:2500. Typical decay curves
The tris-(thio)ureas were assessed as receptors for chlo-
are shown in Figure 4, and the values of I and t = are listed in
1
2
1
ride in [D ]DMSO/H O, 200:1 through H NMR titrations
Table 1. Only one tris-urea, 4OF2, gave positive results; 3OP
and 4OP were not tested because of their very low solubilities,
and 3OF2 gave no transport (probably again due to poor
solubility). However all the tris-thioureas showed useful,
measurable activity. For both series 3S and 4S, we observed
the expected correlation between anion affinity and transport
rate, with one exception in that 4SN was less active than
expected. We were pleased to find that the advantage of
methylated scaffold 4 was retained. For example 4SF2 was ca.
14-times more active than 3SF2. Most importantly the
absolute levels of activity were surprisingly high, considering
the modest chloride affinities. The most powerful, 4SF2,
6
2
+
ꢁ
[6]
with Bu N Cl as substrate. In all experiments the most
4
significant movement was observed for the two NH protons,
which could be followed throughout. The titration data could
be fitted to a 1:1 model using the HypNMR2008 computer
[12]
program,
giving the results summarized in Table 1. As
expected, the use of more electron-withdrawing aryl groups
increases the chloride affinities of both series 3 and 4, while
the thioureas are nearly twice as powerful as the ureas. We
were pleased to find that the affinities for 4 were higher than
those for 3 by factors of 6–7, consistent with the conforma-
tional arguments discussed earlier (Figure 3). On the other
hand the affinities were somewhat disappointing when
compared to cholapod 14. Tris-thiourea 4SF2, which is the
most powerful of the new compounds and the most compa-
showed t = < 10 s, quite similar to cholapod 14.
1
2
To obtain more meaningful comparisons, we performed
a second series of measurements at lower transporter load-
Angew. Chem. Int. Ed. 2014, 53, 5609 –5613
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5611

.
Angewandte
Communications
Figure 4. Chloride transport by 3 and 4 at receptor:lipid=1:2500,
detected by the lucigenin method (see text). These traces were used to
Figure 6. Chloride transport by 4SF2 at different receptor:lipid ratios
(lucigenin method). A trace for 14 at receptor:lipid=1:250000 is
included for comparison.
obtain the I and t = values listed in Table 1.
1
2
ings, focusing on the most active systems from each series
3SF2 and 4SF2) as well as control cholapod 14. Results from
experiments at transporter:lipid = 1:25000 are shown in
Figure 5 and Table 2. 4SF2 and 14 were also compared at
Figure 5). More significantly, 4SF2 was found to overtake 14
at these concentrations (possibly reflecting some self-associ-
ation at higher loadings). At transporter:lipid = 1:25000 the
cyclohexane-based system was roughly 3-times more power-
ful than the cholapod (Table 2, Figure 5). Indeed, 4SF2 was
exceptional by historic standards. Transport was clearly
detectable at transporter:lipid = 1:500000 (Figure 6), and
comparison with archived traces implied higher activity than
all previous systems reported from this laboratory (see
Figure S48, Supporting Information).
(
The surprisingly high activities observed for 3 and 4
suggested that they might be acting as self-assembled
channels rather than mobile carriers. To test this possibility,
4
SF2 was tested for chloride/nitrate exchange in vesicles
composed of dipalmitoylphosphatidylcholine (DPPC), which
undergoes a gel–liquid phase transition at 418C. At 458C
rapid chloride transport was observed, but at 258C 4SF2
[4b]
showed no activity (see Figure S49). Like the cholapods it
seems that 4SF2 is sensitive to bilayer viscosity, suggesting
[
16,17]
a mobile carrier mechanism.
Finally we were interested to know whether, like some
Figure 5. Chloride transport by 3SF2, 4SF2 and cholapod 14 at
receptor:lipid=1:25000 (lucigenin method). These traces were used
[
4c,5c,g,11]
other systems,
the new transporters might also be
to obtain the I and t values listed in Table 2.
1
=
2
effective for the biologically important anions bicarbonate
and sulfate. As these hydrophilic substrates are resistant to
transport, they can be studied by pairing with chloride in
Table 2: Anion transport data for thioureas 3SF2, 4SF2, and 14 at low
ꢁ
[
a]
antiport experiments and following the change in Cl
transporter loadings.
concentration (bicarbonate/sulfate transport being the rate-
limiting process). Transporter 4SF2 was tested in vesicles
prepared using 1) aqueous Na SO and 2) aqueous NaHCO .
In the former case the rate of fluorescence decay was
negligible after an initial small drop (see Figure S51). It
therefore seems that 4SF2 does not transport sulfate at an
appreciable rate. The initial drop may be due to unbalanced
chloride influx, creating a potential across the membrane.
Compound Transporter:lipid=1:25000 Transporter:lipid=1:250000
ꢁ1
ꢁ1
I [s
]
t
1
[s]
I [s
]
t [s]
=
1
=
2
2
2
4
3
3SF2
4SF2
14
0.00097
0.013
0.0037
408
54
147
0.0023
0.00095
214
309
[a] Details as for Table 1.
transporter:lipid = 1:250000 (Table 2), and 4SF2 was tested at
the very low loading of transporter:lipid = 1:500000. The
traces for 4SF2 at all loadings, with that of 14 at 1:250000, are
shown in Figure 6. As expected, the advantage of 4 over 3 was
confirmed, the methylated 4SF2 proving to be ca. 10-times
more active than 3SF2 at transporter:lipid = 1:25000 (Table 2,
With the vesicles formed using NaHCO an initial drop in
3
fluorescence was followed by slow long-term decay, suggest-
ing that bicarbonate can be transported but only very poorly.
The remarkable effectiveness of these cyclohexane-based
transporters may provide new insight into anionophore
design. Studies thus far have suggested that binding affini-
5
612
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 5609 –5613

Angewandte
Chemie
[
1h,4]
[18]
ty
and lipophilicity
are major factors determining
C. J. E. Haynes, J. Gonzalez, J. L. Sutton, S. J. Brooks, M. E.
Light, J. Herniman, G. J. Langley, V. Soto-Cerrato, R. Perez-
Tomas, I. Marques, P. J. Costa, V. Felix, P. A. Gale, Chem. Sci.
2013, 4, 103; f) L. Adriaenssens, C. Estarellas, A. V. Jentzsch,
M. M. Belmonte, S. Matile, P. Ballester, J. Am. Chem. Soc. 2013,
activity, and trends within the 3/4 series are consistent with
both effects. Lipophilic balance (an even distribution of
[
19]
[18a]
lipophilic groups) and molecular size
also seem to be
relevant. However, the difference between 4SF2 and 14 is
135, 8324; g) N. Busschaert, M. Wenzel, M. E. Light, P. Iglesias-
difficult to rationalize on the basis of these factors. The two
Hernandez, R. Perez-Tomas, P. A. Gale, J. Am. Chem. Soc. 2011,
133, 14136.
[6]
molecules possess similar sizes and clogP values and are not
obviously different in terms of lipophilic balance. Nonethe-
less, 4SF2 is 3-times more active than 14 as a transporter, but
[6] For details of calculations, synthetic procedures, binding experi-
ments and transport studies, see the Supporting Information.
[7] M. Mascal, J. Chem. Soc. Perkin Trans. 2 1997, 1999.
8] C. Raposo, M. Almaraz, M. Martin, V. Weinrich, L. Mussons, V.
Alcazar, C. Caballero, J. R. Moran, Chem. Lett. 1995, 759. This
paper also describes a brief study of 3OP as a phosphate receptor
[
20]
1
8-times less powerful as a receptor. A possible explanation
[
may be the difference in flexibility. The cholapods 1 (and
decalins 2) possess preorganized NH groups which are well-
suited for binding but perhaps less so for transport, where on–
off rates could be critical. The conformational freedom of 3/4
may lower affinities but, to compensate, accelerates binding
kinetics. The methyl groups in 4 promote transport by
improving binding without substantially affecting flexibility.
In conclusion, we report a new series of cyclohexane-
based anion carriers which are less preorganized but more
accessible than earlier systems. While their flexibility reduces
anion affinities, transport rates remain high. Especially
notable is the hexamethyl system 4 in which acyclic conforma-
tional control improves binding and transport, and which sets
new records for carrier activity. The results suggest that
control of flexibility could be an important parameter for the
design of future anionophores.
in [D ]DMSO.
6
[9] A. Koziara, A. Zwierzak, Tetrahedron Lett. 1987, 28, 6513.
10] Tris-azide 12 failed to explode on heating or impact. Never-
theless, for preparative scale work it was handled as ꢀ 5%
solutions in organic solvents.
11] A. V. Koulov, T. N. Lambert, R. Shukla, M. Jain, J. M. Boon,
B. D. Smith, H. Li, D. N. Sheppard, J.-B. Joos, J. P. Clare, A. P.
Davis, Angew. Chem. 2003, 115, 5081; Angew. Chem. Int. Ed.
2003, 42, 4931.
[12] a) C. Frassineti, S. Ghelli, P. Gans, A. Sabatini, M. S. Moruzzi, A.
Vacca, Anal. Biochem. 1995, 231, 374; b) C. Frassineti, L.
Alderighi, P. Gans, A. Sabatini, A. Vacca, S. Ghelli, Anal.
Bioanal. Chem. 2003, 376, 1041.
13] J. M. Timko, S. S. Moore, D. M. Walba, P. C. Hiberty, D. J. Cram,
J. Am. Chem. Soc. 1977, 99, 4207.
14] a) A. J. Ayling, S. Broderick, J. P. Clare, A. P. Davis, M. N. Pꢃrez-
Payꢄn, M. Lahtinen, M. J. Nissinen, K. Rissanen, Chem. Eur. J.
2002, 8, 2197; b) J. P. Clare, A. J. Ayling, J. B. Joos, A. L. Sisson,
G. Magro, M. N. Pꢃrez-Payꢄn, T. N. Lambert, R. Shukla, B. D.
Smith, A. P. Davis, J. Am. Chem. Soc. 2005, 127, 10739.
15] Cramꢀs extraction protocol makes various assumptions (e.g. 1:1
binding stoichiometry) which may be difficult to verify. For this
[
[
[
[
Received: December 20, 2013
Revised: February 24, 2014
Published online: April 7, 2014
[
Keywords: anion binding · anion transport ·
.
reason the derived K values should be considered “apparent”.
a
molecular recognition · receptors · supramolecular chemistry
Despite the uncertainties, the method provides a direct indica-
tion of a receptorꢀs ability to extract an anion into non-polar
media as required for anion transport.
[
1] a) A. P. Davis, D. N. Sheppard, B. D. Smith, Chem. Soc. Rev.
[
16] G. Deng, T. Dewa, S. L. Regen, J. Am. Chem. Soc. 1996, 118,
2007, 36, 348; b) G. W. Gokel, N. Barkey, New J. Chem. 2009, 33,
8975.
947; c) J. Mareda, S. Matile, Chem. Eur. J. 2009, 15, 28; d) J. T.
[
17] The interpretation of this experiment is complicated by the
possibility that the fluid!gel transition might expel the trans-
porter from the membranes: see S. Otto, M. Osifchin, S. L.
Regen, J. Am. Chem. Soc. 1999, 121, 10440. However, a control
experiment suggested that this does not occur for 4SF2 (see
Supporting Information).
18] a) N. Busschaert, S. J. Bradberry, M. Wenzel, C. J. E. Haynes,
J. R. Hiscock, I. L. Kirby, L. E. Karagiannidis, S. J. Moore, N. J.
Wells, J. Herniman, G. J. Langley, P. N. Horton, M. E. Light, I.
Marques, P. J. Costa, V. Felix, J. G. Frey, P. A. Gale, Chem. Sci.
Davis, O. Okunola, R. Quesada, Chem. Soc. Rev. 2010, 39, 3843;
e) P. A. Gale, Acc. Chem. Res. 2011, 44, 216; f) S. Matile, A. V.
Jentzsch, J. Montenegro, A. Fin, Chem. Soc. Rev. 2011, 40, 2453;
g) N. Busschaert, P. A. Gale, Angew. Chem. 2013, 125, 1414;
Angew. Chem. Int. Ed. 2013, 52, 1374; h) H. Valkenier, A. P.
Davis, Acc. Chem. Res. 2013, 46, 2898.
[
[
[
2] F. M. Ashcroft, Ion Channels and Disease, Academic Press,
London, 2000.
3] J. R. Broughman, L. P. Shank, W. Takeguchi, B. D. Schultz, T.
Iwamoto, K. E. Mitchell, J. M. Tomich, Biochemistry 2002, 41,
2013, 4, 3036; b) V. Saggiomo, S. Otto, I. Marques, V. Felix, T.
7350.
Torroba, R. Quesada, Chem. Commun. 2012, 48, 5274.
19] H. Valkenier, C. J. E. Haynes, J. Herniman, P. A. Gale, A. P.
Davis, Chem. Sci. 2014, 5, 1128.
20] As suggested by a referee, the possibility that the earlier
transporters bind too strongly should also be considered (see:
A. V. Jentzsch, D. Emery, J. Mareda, P. Metrangolo, G. Resnati,
S. Matile, Angew. Chem. 2011, 123, 11879; Angew. Chem. Int. Ed.
[
4] a) B. A. McNally, A. V. Koulov, B. D. Smith, J. B. Joos, A. P.
Davis, Chem. Commun. 2005, 1087; b) B. A. McNally, A. V.
Koulov, T. N. Lambert, B. D. Smith, J. B. Joos, A. L. Sisson, J. P.
Clare, V. Sgarlata, L. W. Judd, G. Magro, A. P. Davis, Chem. Eur.
J. 2008, 14, 9599; c) S. Hussain, P. R. Brotherhood, L. W. Judd,
A. P. Davis, J. Am. Chem. Soc. 2011, 133, 1614.
[
[
[
5] a) J. L. Seganish, J. T. Davis, Chem. Commun. 2005, 5781; b) J. L.
Sessler, L. R. Eller, W. S. Cho, S. Nicolaou, A. Aguilar, J. T. Lee,
V. M. Lynch, D. J. Magda, Angew. Chem. 2005, 117, 6143;
Angew. Chem. Int. Ed. 2005, 44, 5989; c) J. T. Davis, P. A. Gale,
O. A. Okunola, P. Prados, J. C. Iglesias-Sanchez, T. Torroba, R.
Quesada, Nat. Chem. 2009, 1, 138; d) S. Bahmanjah, N. Zhang,
J. T. Davis, Chem. Commun. 2012, 48, 4432; e) S. J. Moore,
2011, 50, 11675; J. P. Behr, M. Kirch, J. M. Lehn, J. Am. Chem.
Soc. 1985, 107, 241). However, within closely related series we
have generally found that affinities correlate with transport
activity, even for very powerful receptors. It thus seems that we
have not yet reached the zone where improved binding inhibits
transport.
Angew. Chem. Int. Ed. 2014, 53, 5609 –5613
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5613