A small synthetic molecule forms selective potassium channels to regulate cell membrane potential and blood vessel tone

Article abstract of DOI:10.1039/c4ob01420k

In living cell membranes, K⁺ permeability is higher than that of other ions such as Na⁺ and Cl^- owing to abundantly expressed K⁺ channels. Polarized membrane potential is mainly established by K⁺ outward flow because the K⁺ concentration in the intracellular side is much higher than that in the extracellular side. We have found that the small synthetic molecule 1 is capable of self-assembling into selective K⁺ channels, enhancing K⁺ permeability and hyperpolarizing liposome membrane potential. Interestingly, molecule 1 also functions as K⁺ channel hyperpolarizing living cell membrane potential and relaxing agonist-induced blood vessel contraction. Therefore, it may have the potential to become a lead compound for the treatment of human diseases associated with K⁺ channel dysfunction. This journal is

Full text of DOI:10.1039/c4ob01420k

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
A small synthetic molecule forms selective
potassium channels to regulate cell membrane
potential and blood vessel tone†
Cite this: Org. Biomol. Chem., 2014,
12, 8174
Hui-Yan Zha,‡a Bing Shen,‡b,c Kwok-Hei Yau,a Shing-To Li,a Xiao-Qiang Yao*b and
Dan Yang*a
In living cell membranes, K+ permeability is higher than that of other ions such as Na+ and Cl− owing to
abundantly expressed K+ channels. Polarized membrane potential is mainly established by K+ outward
flow because the K+ concentration in the intracellular side is much higher than that in the extracellular
side. We have found that the small synthetic molecule 1 is capable of self-assembling into selective
K+ channels, enhancing K+ permeability and hyperpolarizing liposome membrane potential. Interestingly,
molecule 1 also functions as K+ channel hyperpolarizing living cell membrane potential and relaxing
agonist-induced blood vessel contraction. Therefore, it may have the potential to become a lead
compound for the treatment of human diseases associated with K+ channel dysfunction.
Received 8th July 2014,
Accepted 14th August 2014
DOI: 10.1039/c4ob01420k
www.rsc.org/obc
Introduction
Ion channels, membrane proteins that mediate ion flow
between plasma membranes, participate in various pivotal bio-
logical processes. Specifically, the potassium (K+) channel is
expressed in all kinds of cells involved in numerous physio-
logical and pathological processes, such as the excitation of
excitable cells, secretion of gland cells, cytosolic H+/OH−
balance, apoptosis and arrhythmia.1–4 The basic structure of
natural K+ channel is a tetrameric molecule, which is com-
posed of four identical subunits. The subunits form a four-
fold symmetrical protein complex, which is arranged around a
central pore through which ions are conducted.5–16 In the lit-
erature, most studies focus on finding new molecules, such as
nicorandil, pinacidil, flupirtine, to regulate natural K+ chan-
nels. How to create synthetic K+ channels and mimic natural
K+ channel function still remain challenging. Although several
synthetic cation channels have been created, small synthetic
K+ channel candidates possessing good K+ channel behavior
not only in artificial lipid membrane but also in living system
have not been reported yet.17–20 Previously, we discovered that
Scheme 1 Chemical structures of compounds C11 and 1.
a
small C2-symmetric molecule with an isophthalamide
scaffold linking two R-(α)-aminoxy leucine units (C11:
Scheme 1) was capable of self-assembling into selective Cl−
channels and modulating the membrane potential of living
cells.21 Herein we report that compound 1 (Scheme 1), which
has the isophthalamide scaffold linking two Boc-protected (α)-
aminoxy lysine residues, can self-assemble into K+ channels in
liposomes and living cell membranes.
aMorningside Laboratory for Chemical Biology, Department of Chemistry,
The University of Hong Kong, Pokfulam Road, Hong Kong, P. R. China.
E-mail: yangdan@hku.hk
bSchool of Biomedical Sciences, The Chinese University of Hong Kong, Shatin, N.T.,
Hong Kong, China. E-mail: yao2068@cuhk.edu.hk
cDepartment of Physiology, Anhui Medical University, Hefei, China
†Electronic supplementary information (ESI) available: Experimental procedures.
See DOI: 10.1039/c4ob01420k
Results and discussion
We prepared compound 1 in four steps from readily available
(R)-6-(tert-butoxycarbonylamino)-2-(1,3-dioxoisoindolin-2-yloxy)-
hexanoic acid with an overall yield of 30% (see ESI†).
‡These authors contributed equally to this work.
8174 | Org. Biomol. Chem., 2014, 12, 8174–8179
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
We first used the pH-sensitive fluorescent dye 8-hydroxy- were changed to 200 mM NMDG-Cl, then single channel cur-
1,3,6-pyrene-trisulfonate (HPTS, pyranine) to test the ion trans- rents would disappear (Fig. 2b). Because NMDG+ (N-methyl-D-
port ability and selectivity of compound 1. The addition of a glucamine) is a huge organic cation, normally it is unable to
KOH solution created a gradient of approximately one pH unit. pass through ion channels. Therefore, the results suggested
The resulting pH gradient caused H+ efflux (outward flow) or that the ion channel formed by 1 is a K+ channel, not a Cl−
OH− influx (inward flow) and built up an electrostatic poten- channel.
tial, which was expected to be compensated by the exchange of
To explore the selectivity between K+ and Na+ of the ion
intra- and extravesicular electrolytes across the lipid bilayers of channel formed by 1, patch clamp experiments with inside-out
the liposomes, mediated by compound 1.22 In KCl-filled lipo- configuration were performed. With symmetrical 200 mM KCl
somes suspended in an isotonic KCl solution, the application or NaCl solution on both sides, single channel currents were
of increasing concentrations of compound 1 did induce a observed at both positive and negative holding potentials
rapid exchange of intra- and extravesicular electrolytes as indi- (Fig. 3a and b), indicating that besides K+, Na+ can also be per-
cated by the increase in relative fluorescence that accompanied meated through this synthetic ion channel formed by 1. Next,
the rise in intravesicular pH. This indicated that compound 1 the relative permeability of the channel formed by 1 for K+ and
mediated ion transport in a concentration dependent fashion Na+ ions was examined. The concentration gradients of K+ and
(Fig. 1a and b). To clarify the species of ions mediated by com- Na+ ions were created across the membrane using 180 mM
pound 1, potassium-binding benzofuran isophthalate (PBFI, NaCl and 20 mM KCl as the pipette solution and 180 mM KCl
see ESI†), a K+ sensitive fluorescent dye, was employed.23 Lipo- and 20 mM NaCl as the bath solution. The I–V relationship
somes encapsulating 100 mM NMDG-Cl and 500 µM PBFI showed that the reversal potential was shifted to −15 3 mV
were prepared and then suspended in a 100 mM KCl solution. (n = 5, Fig. 3c). Calculated from the Goldman equation, the
The result showed that 25 µM compound
1
markedly relative permeability of K+ to Na+ ion was 2.2, implying that K+
enhanced PBFI fluorescence intensity compared to the DMSO is more permeable than Na+ in this ion channel formed by 1.
control (Fig. 1c and d), which suggests K+ ions moving into Therefore, the data of ion permeability from the patch clamp
liposomes and binding with PBFI. Therefore, the PBFI assay corroborate with the results from the HPTS assay.
directly demonstrates that compound 1 indeed transports K+
The membrane potential fluorescence assay with POPC
ions. In addition, because both K+ and Na+ are very important liposomes was utilized to further characterize the ion transport
monovalent cations in biological systems, the HPTS assay was properties of 1.26 The liposomes were filled with 100 mM KCl
also used to demonstrate the difference in transport activity and placed in 100 mM NaCl solution containing 60 nM safra-
between Na+ and K+ for compound 1. The intravesicular solu- nin O. Safranin O is a membrane potential sensitive fluo-
tion contained 10 mM HEPES, pH 6.8, and 100 mM KCl and rescent dye and its fluorescence intensity increase represents
the extravesicular solution contained 10 mM HEPES, pH 6.8, hyperpolarization. Our results revealed that 1 induced mem-
and 100 mM MCl (M = Na+ (blue line), K+ (red line)). At t = 100 brane potential hyperpolarization in a dose-dependent fashion
s, 20 mL of a DMSO solution of compound 1 at the final con- (Fig. 4a and b). Because the concentration of Cl− is symmetric
centration of 10 μM was added to extravesicular solution. Then at both sides of the lipid membrane, only K+ efflux will cause
20 μL of a 0.5 M KOH solution was added subsequently to membrane potential hyperpolarization. In addition, valino-
create a gradient of approximately one pH unit. At this point, mycin, which is a well-known K+-specific carrier, was also used in
the exterior pH reached 7.8 while the interior pH remained at our experiment (Fig. 4c and d). When 0.6 µM valinomycin was
6.8, which results in the electrostatic potential across the mem- added to liposome suspension (inside vesicles: 100 mM KCl,
brane. Since compound 1 could function as a cation transpor- pH 6.8; outside vesicles: 100 mM NaCl, 60 nM safranin O, pH
ter, the resulting electrostatic potential caused by H+ efflux or 6.8), K+ ions inside liposomes were transported out and the
OH− influx from the liposomes can be compensated by K+ or membrane potential was hyperpolarized after several minutes.
Na+ influx. Since the relative fluorescence intensity increased Then the addition of compound 1 further enhanced the hyper-
faster in extravesicular KCl solution than in extravesicular polarization of the membrane potential (Fig. 4d), indicating
NaCl solution (Fig. 1e and f), we concluded that compound 1 that 1 facilitates K+ ion movement through the lipid mem-
could transport K+ faster than Na+.
brane and induces membrane potential hyperpolarization.
As a patch clamp is a powerful tool for characterizing the Moreover, the data also suggest that the permeability of 1 for
ion channel, single channel recording was applied to identify K+ is much higher than that for Na+. Since the outside solution
whether compound 1 can form ion channels in giant lipo- of liposomes was NaCl and the inside solution was KCl, the
somes.24,25 In POPC giant liposomes (POPC–PS = 4 : 1) with asymmetric positive charge distribution and polarized poten-
symmetric extracellular and intracellular solutions (200 mM tial can only be established when the permeability of 1 for K+
KCl), typical single channel currents were observed in ramp is much higher than that for Na+.
recording from −100 mV to +100 mV when 5 µM compound 1
Above results are all based on cell-free studies. The big chal-
was applied to the extracellular solution compared to the lenge is whether 1 can self-assemble into functional K+ chan-
DMSO control (Fig. 2a, c and d). The results demonstrate that nels in living cell membranes. To address this issue, we used a
compound 1 is capable of self-assembling into ion channels to current clamp, one type of patch clamp technique, to measure
transport ions. Furthermore, if the solutions in both sides the resting membrane potential and investigate the effect of
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 8174–8179 | 8175

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 1 (a and b) HPTS assays with compound 1 at different concentrations from 1 to 100 μM. (c) Graphical representations of the PBFI fluorescence
assay. (d) Representative traces showing that 25 µM compound 1 markedly enhanced the PBFI fluorescence ratio. Inside vesicles: 100 mM NMDG-Cl,
500 µM PBFI, pH 7.0; outside vesicles: 100 mM KCl, pH 7.0. Compound 1 (25 µM final concentration) was added at the time indicated at the arrow.
(e and f) Comparison of K+ and Na+ permeability of compound 1 by the HPTS assay.
8176 | Org. Biomol. Chem., 2014, 12, 8174–8179
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 2 Single channel currents of compound 1 in POPC liposomes. (a) Single channel ramp recording protocol from −100 mV to +100 mV with a
0 mV holding potential. (b) Representative trace showing that no single channel current could be recorded in symmetric 200 mM NMDG-Cl in the
presence of 5 µM compound 1. (c and d) Representative traces showing that no channel current could be recorded in DMSO control (c) but typical
single channel current was found in the presence of 5 µM compound 1 (d) in symmetric 200 mM KCl solution.
Fig. 3 (a) Current traces of the patch clamp experiment with inside-out configuration in the presence of 5 µM compound 1. Symmetric 200 mM
KCl was used. Channel currents were observed at both positive and negative holding potentials. K+ is transported through the channel mechanism.
(b) Current traces of the patch clamp experiment with inside-out configuration in the presence of 5 µM compound 1. Symmetric 200 mM NaCl was
used. Channel currents were observed at both positive and negative holding potentials. Na+ is transported through the channel mechanism. (c) I–V
relationship of the patch clamp experiment with inside-out configuration in the presence of 5 µM compound 1. 180 mM NaCl and 20 mM KCl were
used in the pipette solution. 180 mM KCl and 20 mM NaCl were used in the bath solution. The reversal potential shifts to −15 3 mV (n = 5). The
relative permeability of K+ to Na+ ion is 2.2.
self-assembled K+ channels on living cells.27 HEK293 cell line, potential is mainly established by intracellular K+ efflux via K+
commonly used in ion channel research, was used in this channels according to electrochemical driving force, because
study. Interestingly, the result showed that 10 µM compound 1 the permeability of K+ is much higher than that of Na+ or Cl−
significantly hyperpolarized the resting membrane potential in resting cells. When new K+ channels self-assembled from 1
(−13 1 mV, n = 7) of HEK293 cells (Fig. 4e). It is well known are present in cell membrane, K+ permeability should be
that intracellular solution is composed of high concentration enhanced. Then more K+ ions would move out of cells, leading
of K+ but of low concentrations of Na+ and Cl−. Reversely, to hyperpolarization of the cell membrane potential. From
extracellular solution has high concentrations of Na+ and Cl− these data, we conclude that small molecule 1 is capable of
but low concentration of K+. Normally, the resting membrane self-assembling into functional K+ ion channels in the living
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 8174–8179 | 8177

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 4 (a) Graphical representations of the safranin O membrane potential assay. (b) Representative traces showing that the application of 2.5, 5,
and 10 µM compound 1 hyperpolarized the membrane potential of liposomes. Inside vesicles: 100 mM KCl, pH 6.8; outside vesicles: 100 mM NaCl,
60 nM safranin O, pH 6.8. Compound 1 was added at the time indicated at the arrow. The concentrations are all final concentrations. At the end of
experiments, 10 µM melittin was added to disrupt the lipid membrane of liposomes. (c) Graphical representations of the safranin O membrane
potential assay. (d) Representative traces showing that 0.6 µM valinomycin hyperpolarized the membrane potential of liposomes while the later
application of 10 µM compound 1 further enhanced the hyperpolarization. Inside vesicles: 100 mM KCl, pH 6.8; outside vesicles: 100 mM NaCl, 60 nM
safranin O, pH 6.8. At 100 s, 0.6 µM (final concentration) valinomycin was applied. At 500 s, compound 1 was applied at 10 µM (final concentration).
At the end of experiments, 10 µM melittin was added to disrupt the lipid membrane of liposomes. (e) The representative trace showing the resting
membrane potential of HEK293 cells. 10 µM compound 1 application (at the arrow) markedly decreased the membrane potential of HEK 293 cells.
cell membrane and regulating resting membrane potential in
living cells.
We further explored the biological application of compound
1 by testing its effects on vascular contraction evoked by
phenylephrine (PE), an α1-adrenergic agonist. As shown in
Fig. 5, compound 1 relaxed PE-induced contraction in isolated
mice aorta in
a concentration-dependent manner and
maximal relaxation was achieved (68 4%) when its concen-
tration reached 30 μM. The relaxant effect of compound 1 on
PE-induced contraction was not affected by endothelium denu-
dation (maximal relaxation: 72
10%), suggesting a direct
action on the vascular smooth muscle cells. This is in contrast
to the behavior of Cl− channel-forming compound C11, which
relaxes muscle contraction induced by high K+ concentration
but not PE.26
Fig. 5 Concentration–response curves for the vasorelaxant effect of 1
on the mouse aortic rings preconstricted by 10 µM PE. Data are
expressed as mean SE.
The proposed K+ channel function of compound 1 also
explains the ability of 1 to relax the PE-induced contraction
(Fig. 5). It is known that one pathway of PE-induced vascular
contraction is to depolarize the membrane potential of vascu-
lar smooth muscle cells,28–30 leading to Ca2+ entry through
L-type Ca2+ channels.10 In most living cells, as the equilibrium
membrane potential of K+ is more negative than the resting
membrane potential, activation of K+ channels would result in
8178 | Org. Biomol. Chem., 2014, 12, 8174–8179
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
K+ efflux and hence membrane hyperpolarization. Conceivably,
synthetic K+ channel-induced hyperpolarization would deacti-
vate L-type Ca2+ channels and inhibit depolarization-mediated
Ca2+ entry in PE-induced contraction.
8 A. C. Hall, C. Suarez, A. Hom-Choudhury, A. N. Manu,
C. D. Hall, G. J. Kirkovits and I. Ghiriviga, Org. Biomol.
Chem., 2003, 1, 2973–2982.
9 B. Baumeister, N. Sakai and S. Matile, Angew. Chem., Int.
Ed., 2000, 39, 1955–1958.
10 K. M. Linn, M. G. Derebe, Y. Jiang and F. I. Valiyaveetil,
Biochemistry, 2010, 49, 4450–4456.
Conclusions
11 T. M. Fyles, Chem. Soc. Rev., 2007, 36, 335–347.
12 A. Cazacu, C. Tong, A. van der Lee, T. M. Fyles and
M. Barboiu, J. Am. Chem. Soc., 2006, 128, 9541–9548.
13 C. De Cola, S. Licen, D. Comegna, E. Cafaro, G. Bifulco,
I. Izzo, P. Tecilla and F. De Riccardis, Org. Biomol. Chem.,
2009, 7, 2851–2854.
14 W. Wang, C. R. Yamnitz and G. W. Gokel, Heterocycles,
2007, 73, 825–839.
15 Y. J. Jeon, H. Kim, S. Jon, N. Selvapalam, D. H. Oh, I. Seo,
C. S. Park, S. R. Jung, D. S. Koh and K. Kim, J. Am. Chem.
Soc., 2004, 126, 15944–15945.
To summarize, we have developed a small molecule 1 that not
only mediates selective transport of K+ ions in lipid membranes,
but also obviously functions as K+ channels in the living
system. Our data have convincingly proved that compound 1 is
capable of self-assembling into K+ channels and mediating K+
transport across the cell membrane. Therefore, this finding
may provide a new avenue in K+ channel drug development for
the treatment of severe human diseases related to malfunction
of natural K+ channels, such as long QT syndrome, episodic
ataxia, Anderson–Tawil syndrome, and familial hyperinsuline-
mia. Future work will be directed at understanding the self-
assembling behavior and exploring other biological appli-
cations of synthetic K+ channels.
16 P. J. Cragg, Sci. Prog., 2002, 85, 219–241.
17 R. S. Vieira-Pires, A. Szollosi and J. H. Morais-Cabral,
Nature, 2013, 496, 323–329.
18 P. Reib and U. Koert, Acc. Chem. Res., 2013, 46, 2773–
2780.
19 J. Montenegro, M. R. Ghadiri and J. R. Granja, Acc. Chem.
Res., 2013, 46, 2955–2965.
20 F. De Riccardis, I. Izzo, D. Montesarchio and P. Tecilla, Acc.
Chem. Res., 2013, 46, 2781–2790.
21 X. Li, B. Shen, X. Q. Yao and D. Yang, J. Am. Chem. Soc.,
2007, 129, 7264–7265.
Acknowledgements
This study was supported by The University of Hong Kong and
the Hong Kong Research Grants Council (HKU2/06C and
HKU8/CRF/10).
22 K. Kano and J. H. Fendler, Biochim. Biophys. Acta, 1978,
509, 289.
Notes and references
1 J. I. Vandenberg, M. D. Perry, M. J. Perrin, S. A. Mann, Y. Ke 23 X. Chen and R. W. Gross, Biochemistry, 1995, 34, 7356–
and A. P. Hill, Physiol. Rev., 2012, 92, 1393–1478. 7364.
2 J. C. Koster, M. A. Permutt and C. G. Nichols, Diabetes, 24 O. P. Hamill, A. Marty, E. Neher, B. Sakmann and
2005, 54, 3065–3072. F. J. Sigworth, Pflugers. Arch., 1981, 391, 85–100.
3 H. Misonou, D. P. Mohapatra and J. S. Trimmer, Neuro- 25 D. M. Engelman and T. A. Steitz, Cell, 1981, 23, 411–422.
toxicology, 2005, 26, 743–752.
4 H. Ouadid-Ahidouch and A. Ahidouch, J. Membr. Biol.,
2008, 221, 1–6.
5 M. Jung, H. Kim, K. Baek and K. Kim, Angew. Chem., Int.
Ed., 2008, 47, 5755–5757.
26 X. Li, B. Shen, X. Q. Yao and D. Yang, J. Am. Chem. Soc.,
2009, 131, 13676–13680.
27 H. Y. Kwan, B. Shen, X. Ma, Y. C. Kwok, Y. Huang,
Y. B. Man, S. Yu and X. Yao, Circ. Res., 2009, 104, 670–
678.
6 C. P. Wilson and S. J. Webb, Chem. Commun., 2008, 28 K. P. Minneman, Pharmacol. Rev., 1988, 40, 87–119.
4007–4009.
29 X. L. Chen and C. M. Rembold, Am. J. Physiol., 1995, 268,
H74–H81.
30 J. R. Docherty, Cell. Mol. Life Sci., 2010, 67, 405–417.
7 G. W. Gokel and I. A. Carasel, Chem. Soc. Rev., 2007, 36,
378–389.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 8174–8179 | 8179