Metal–Organic Synthetic Transporters (MOST): Efficient Chloride and Antibiotic Transmembrane Transporters

Article abstract of DOI:10.1002/chem.201700847

We present the synthesis of two functionalized 2,4,7-triphenylbenzimidazole ligands and demonstrate the formation of their respective metal assemblies in phospholipid membranes. Anion transport experiments demonstrate the formation of metal–organic synthe

Full text of DOI:10.1002/chem.201700847

A Journal of
Accepted Article
Title: Metal-organic synthetic transporters (MOST): efficient chloride
and antibiotic transmembrane transporters
Authors: Julie Kempf and Andreea R Schmitzer
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: Chem. Eur. J. 10.1002/chem.201700847
Link to VoR: http://dx.doi.org/10.1002/chem.201700847
Supported by

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
Metal-organic synthetic transporters (MOST): efficient chloride
and antibiotic transmembrane transporters
[
a]
[a]
Julie Kempf, and Andreea R. Schmitzer*
Abstract: We present the synthesis of two functionalized 2,4,7-
triphenylbenzimidazole ligands and demonstrate the formation of
their respective metal assemblies in phospholipid membranes.
Anion transport experiments demonstrate the formation of metal-
organic synthetic transporters (MOST) directly in phospholipid
membranes. The formation of MOST in phospholipid membranes
results in efficient architectures for chloride transport. We also
demonstrate the insertion of these ligands and the formation of
their metal-organic assemblies in bacterial membranes; the use
of MOST makes the membranes of resistant bacteria more
permeable to antibiotics. We also demonstrate that a combination
of MOST with tetracycline lowers the sensitivity of resistant
bacteria to tetracycline by 60-fold.
transport and this has opened the door to the synthesis of a wide
[
8]
variety of synthetic ion transporters. Nevertheless, the design of
efficient synthetic ion transporters, and precise ion channels,
remains a challenge. It is essential not only to predict the self-
association of a compound into a pore-forming architecture, but
also to understand the interactions with the surrounding
membrane. For example, water or simple ions might interfere with
[
9]
and affect the integrity of the supramolecular architecture. For
this reason, the assembly of well-defined metal-organic channels
in phospholipid bilayers may be a way to induce transmembrane
transport.
Since functionalized platinum complexes have been shown to
[
10]
behave as metal-organic anion receptors in solution and many
anion receptors have been incorporated into the structure of
transmembrane ion transporters, it is surprising that only a few
MLAs or MOFs have been studied as ion transport candidates.
Introduction
2
+
[11]
Large MLAs such as Cu -based polyhedrons , metal-oxide
[
12]
3+,[13]
2+,[14]
2+ [15]
Metal–organic framework (MOF) materials are one of the most
exciting, high-profile developments in nanotechnology in the last
capsules and oligo-porphyrins with Rh
, Pd
or Zn ,
were studied for their ionophoric properties. The large pores
formed by the metal oligo-porphyrin assemblies were shown to be
very effective for the transport of large molecules such as
carboxyfluorescein or tetrabutylammonium cation across
phospholipid bilayers. Due to the complexity of the synthetic
pathway and the poor membrane insertion of these bulky
assemblies, smaller components were designed to form synthetic
ion channels by coordination with metals. The formation of the so-
[
1],[2]
twenty years.
Composed of metal nodes (metal ions, clusters,
chains or layers) connected by organic linkers, they show some
of the highest porosity known. This property is ideal for gas
[
3]
[4]
capture or storage and drug delivery applications. In particular,
nontoxic and biodegradable MOFs have been used for the
encapsulation and controlled delivery of a large number of
therapeutic molecules, including several challenging antitumor
and antiretroviral drugs such as busulfan, cidofovir and
called “Fujita” motif between a bipyridine and an amphiphilic
[
5]
[16]
azidothymidine triphosphate.
In addition, metal–ligand
palladium derivative was studied by Fyles et al.
However,
assemblies (MLAs) are known to be capable of undergoing
spontaneous reconstitution. This allows for the introduction of
new functions in specific environments and is very attractive for
mimicking the behavior of different biological systems involving
multiple unknown assemblies between this ligand and the metal
were formed and the desired molecular square was not observed
in the phospholipid membrane. Webb et al. also investigated
pyridyl-palladium coordination for the formation of a reversible
[
6]
[17]
metals.
metal-organic channel in a phospholipid bilayer.
Addition of
2
+
Pd to an amphiphilic cholate derivative resulted in a metal-ligand
assembly that spanned the phospholipid bilayer and activated the
proton/sodium transport process.
Ion transport across cell membranes is one of the most important
processes in living cells and is regulated by a wide variety of
[
7]
transporters such as channels, mobile carriers and pumps.
There is a growing interest in understanding the mechanism of ion
The assembly of discrete molecular units into extended networks
is generally observed during MOF synthesis. Consequently, this
synthetic approach allows reactions under mild conditions.
Generally, little is known about the metal- ligand species
preformed in solution before the assembly process. Therefore,
MOFs are typically obtained by means of solvothermal synthesis.
[a]
Dr. J. Kempf, Prof. Dr. A. Schmitzer
Département de Chimie, Université de Montréal
C. P. 6128 Succursale Centre-Ville, Montréal, Québec, H3C 3J7
[
18], [19]
(
Canada)
The similar in situ formation of a metal-organic synthetic
Fax : (+1) 514-343-7586
transporter (MOST) in the phospholipid bilayer has yet to be
demonstrated and exploited in transmembrane ion transport. This
is presumably due to aqueous/membrane partitioning of the
E-mail: ar.schmitzer@umontreal.ca
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.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
ligand and the metal, as well as the presence of other ions can Results and Discussion
interfere with the assembly of the MOST.
We have previously demonstrated that 2,4,7-triphenyl-
Synthesis
benzimidazole 1 inserts into phospholipid bilayers and then self-
[
20]
Ligands 2 and 3 were synthesized according to the previously
assembles into rods that promote chloride transport (Figure 1).
[
22]
reported procedure for 1.
A pyridyl moiety was added in
With the addition of supplementary donor atoms (pyridine and
carboxylate) on to this scaffold, we demonstrate herein how the
assembly of MOSTs in the phospholipid bilayer, can be used to
induce the transmembrane transport of ions and other molecules,
particularly antibiotics. This is particularly important, as infections
from resistant bacteria have become quite common, as more
pathogens become resistant to multiple antimicrobials.
Identification of new antibacterial agents and strategies as
alternative treatments is crucial to increase our ability to fight
position 2 of the benzimidazole unit and two carboxylic acids were
added in the para position of the aromatic rings to obtain ligands
2
and 3, respectively. The synthetic pathway is shown in Scheme
1
.
Insertion into the phospholipid membrane
The insertion of ligands 2 and 3 into the phospholipid membrane,
as well as the in situ formation of the MOST in a phospholipid
membrane, were first monitored by fluorescence. Ligands were
added to egg yolk phosphatidylcholine (EYPC) liposomes in a
sodium/phosphate buffer (pH = 7.2).
[
21]
infectious diseases.
ꢀ
Figureꢀ1.ꢀStructureꢀofꢀtheꢀ2,4,7-triphenylbenzimidazoleꢀ(1)ꢀandꢀitsꢀtwoꢀanaloguesꢀ
Schemeꢀ1.ꢀꢀSynthesisꢀofꢀtheꢀligandsꢀ1,ꢀ2ꢀandꢀ3.ꢀ
[
20]
As shown in Figure 2a, the maximum of fluorescence of 2 (in
dichloromethane) was shifted from 440 nm to 435 nm when
incorporated in the phospholipid bilayer, due to the stabilization of
the excited state by the presence of a more polar environment
from one molecule to another one spatially close can occur.
The same behavior was observed when PdCl was added to
ligand 3, but the fluorescence quenching, induced by the addition
2
2
+
of Pd ions, was less pronounced in this case suggesting a less
dramatic structural change in the self-assembly of ligand 3 due to
metal complexation (Figure 2b).
[
20]
(
probably close to the charged groups of the phospholipids).
2
After addition of an aliquot of a PdCl solution (in MeOH), a
significant decrease of the fluorescence intensity was observed.
This is likely the result of a re-organization of 2 in the membrane
and formation of a metal-ligand complex, where photon transfer
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
a)
Formation of the metal-ligand complexes
Evidence for the formation of the metal-ligand complexes and
their apparent stoichiometry were obtained by UV-vis titration.
These experiments were performed in DCM in order to mimic the
hydrophobicity of the phospholipid bilayer. Different amounts of
PdCl
2
were added to the ligand solution to monitor the
complexation. In the case of ligand 1, a 1:1 complex was formed
2
+
by complexation of Pd by the benzimidazole moiety, as
indicated by the shift of the absorption band corresponding to the
[
23]
n-π* transition, to shorter wavelengths
(see ESI, Figure S1).
When PdCl was added to ligand 2 (Figure 3a), coordination of
2
the more accessible pyridyl unit occurs, translated into a red shift
of the band corresponding to the n-π* transition at 318 nm.This is
[
24]
characteristic of the formation of pyridine:palladium complexes,
where the formation of a 2:1 complex is favored (Figure 3a). A 1:1
complex is formed upon addition of 1 equivalent of PdCl to ligand
(Figure 3b). The proposed structures of the palladium
complexes formed in an apolar environment are shown in Figure
.UV-vis titrations were also performed with HAuCl , AgBF
RuCl and Pd(OAc) and the stoichiometry of the complexes are
summarized in Table 1 (UV-vis spectra are provided in Figures
2
3
b)
4
4
4
,
3
2
2
+
3+
S2-S9). Ligand 2 forms 2:1 complexes with Pd and Au and 1:1
3
+
+
complexes with Ru and Ag . Ligand 3 forms complexes with a
1
2
:1 general stoichiometry, which may also correspond to higher
:2, 3:3 or n:n stoichiometries, with all the studied metals.
However, addition of other metals than palladium to ligand 3,
resulted in the formation of different complexes with the
benzimidazole unit, as a shift corresponding to the n-π* transition
band was observed (ESI Figures S3, S5, S7 and S9).
Formation of metal-ligand assemblies in the presence of
liposomes was also monitored by UV-vis. The UV-vis spectra of
both ligands
2 and 3 in the presence of liposomes are
indistinguishable from those obtained in DCM, confirming their
insertion in the apolar environment of the phospholipid
membrane. When 0.5 equivalents of PdCl were added to the
2
liposomes solution containing ligands 2 or 3, a red shift of the
band corresponding to the n-π* transition was observed. In the
case of ligand 2, the increase in the amount of palladium to one
equivalent did not induce any further change, compared to ligand
Figureꢀ2.ꢀFluorescenceꢀofꢀa)ꢀ
2
ꢀandꢀb)ꢀ
3
ꢀinꢀdichloromethaneꢀ(DCM)ꢀ(0.016ꢀmM,ꢀ
squares),ꢀinꢀliposomesꢀatꢀ25ꢀ°Cꢀ(15ꢀmol%,ꢀtriangle)ꢀandꢀafterꢀtheꢀadditionꢀofꢀ0.5ꢀ
equivalentsꢀ ofꢀ PdCl
crosses).ꢀ Intravesicularꢀ solution:ꢀ 500ꢀ mMꢀ NaCl,ꢀ 5ꢀ mMꢀ phosphateꢀ buffer.ꢀ
Extravesicularꢀsolution:ꢀ500ꢀmMꢀNaNO ,ꢀ5ꢀmMꢀphosphateꢀbufferꢀ(pHꢀ=ꢀ7.2)ꢀ
2
ꢀ (relativeꢀ toꢀ theꢀ concentrationꢀ ofꢀ 2ꢀ andꢀ 3ꢀ respectively,ꢀ
3
3
, where an increase of absorption at 323 nm was observed after
addition of 0.67 equivalents of PdCl (Figure 5). Unfortunately, at
higher concentrations of PdCl precipitation occurred in these
2
2
aqueous conditions, the 1:1 conditions not having been reached.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
Figureꢀ4.ꢀSchematicꢀrepresentationꢀofꢀtheꢀmetalꢀcomplexesꢀofꢀ1,ꢀ2ꢀandꢀ3ꢀwith
palladium.ꢀChlorideꢀionsꢀandꢀcoordinatedꢀsolventsꢀareꢀnotꢀshown.
ꢀ
Figureꢀ3.ꢀUV-Visibleꢀspectraꢀof:ꢀa)ꢀ
inꢀpresenceꢀofꢀvariousꢀconcentrationsꢀofꢀPdCl
TheꢀnumberꢀofꢀequivalentsꢀofꢀPdCl ꢀreportedꢀinꢀtheꢀlegendꢀareꢀrelativeꢀtoꢀtheꢀ
ligands.ꢀ
2
ꢀandꢀb)ꢀ
3
ꢀ(concentrationꢀ0.033ꢀmM)ꢀinꢀCH
2 2
Cl ꢀ
2
ꢀ(methanolꢀsolutionꢀatꢀ2.5ꢀmM).ꢀ
2
Table 1. Stoichiometry of the complexes formed by 1, 2 and 3 with different
metals
PdCl
2
Pd(OAc)
2
RuCl
3
HAuCl
4
AgBF
4
1
1:1
n.d
n.d
n.d
n.d
2
3
2:1
1:1
2:1
1:1
1:1
2:1
1:1
1:1*
1:1*
1:1*
*
Metal complex formed with the benzimidazole moiety
Figureꢀ 5.UV-visibleꢀ spectraꢀ ofꢀ
aqueousꢀsolutionꢀofꢀliposomesꢀ(5ꢀmM)ꢀinꢀtheꢀpresenceꢀofꢀdifferentꢀamountsꢀofꢀ
PdCl
2ꢀ andꢀ 3ꢀ (finalꢀ concentrationꢀ 0.05ꢀ mM)ꢀ inꢀ anꢀ
2
ꢀ
The palladium complex obtained upon the addition of 0.5
equivalent of PdCl to ligand 2 in acetonitrile was isolated as a
dark orange powder and characterized by NMR (Figure S11).
2
1
2
Upon complexation with PdCl a significant change in the H NMR
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
spectra was observed, since the signal of the ortho and meta
protons of the pyridine nitrogen were significantly shifted
downfield. Due to the lack of solubility of ligand 3 in acetonitrile
and of PdCl
2
in DMSO, the formation of the complex of 3 with
Pd(OAc) was monitored in DMSO. Addition of 0.1 equivalents of
2
1
Pd resulted in the appearance in the H NMR spectra of a sharp
peak corresponding to the carboxylic proton. However, upon
addition of 0.2 equivalents of Pd(OAc)
and surprisingly instant gel formation was observed when a 0.5
equimolar amount of Pd(OAc) was added. Gel formation was
2
a precipitate was formed
2
confirmed by the vial inversion technique. The fast gelation
process suggests that there is a rapid complexation between the
palladium and the carboxylates, giving rise to large aggregates.
Interestingly, ligand 2 gave the same supramolecular gelation
under the same conditions, upon addition of 0.5 equivalents of
Pd(OAc)
metallogels were studied through scanning electron microscopy
SEM) (Figure 6). Both palladium complexes form large fibers that
2
in DMSO. The morphological properties of these
(
exhibit fine structure and appear to be multi-layered. The fibers
are relatively straight with only slight twisting, which is more
pronounced in the case of 3. Regular shape indicates that the
fibers are formed by well-ordered molecular packing. The
smallest fibers have an approximate width of 40 nm. The small
fibers are layered together into thicker bundles of long fibers that
have an approximate width of 0.3 μm. The regular shape and the
extreme ratio between width and length in this system must arise
from a strong anisotropic growth process resulting in the possible
presence of 1-D or 2-D structures in the system.
[
20]
Figure 6. SEM images of hierarchically self-assembled gel
microstructures (dried samples): a) Pd (2) and b) Pd (3).
We previously reported the crystal structure of ligand 1.
2+
2+
Unfortunately, crystals suitable for X-ray analysis were obtained
only for ligand 2 by dissolving it in hot acetonitrile and subsequent
slow evaporation of the solvent (see ESI). The supramolecular
organization of 2 in the solid state reveals the formation of an
aromatic channel, where the repeating unit is a dimer (Figure 7a).
Starting with this structure, models were generated for the
Both palladium complexes form rectangular prism structures in
the phospholipid bilayer by the self-assembly of several palladium
dimers (in the case of 2) or oligomers (in the case of 3). Three
2
+
2
+
2+
monomers are required in the case of Pd (3) to span the entire
Pd (2), 3 and Pd (3) using the PM6/SCF-MO method (Figure
2+
2
+
2+
bilayer, while in the case of Pd (2) four monomers self-assemble
7
a). In a second step, the energy minimized Pd (2) and Pd (3)
complexes were placed in a phosphatidylcholine bilayer and a
00 ps molecular dynamics study was performed. As shown in
to form the length of the rectangular prism. These models can be
2
+
used to interpret the observed self-association of Pd (2) and
2
2
+
Pd (3) into long fibers and bundles. Importantly, the formation of
these types of self-assembled structures can be seen as the type
of organized repeating unit that have the potential to produce a
gel (Figure 7c) and help to rationalize the self-association of
metal-ligand assemblies in the polarized environment of the
phospholipid bilayer.
Figure 7b, both resulting palladium complexes possess porous
structures, compared to the self-assemblies of the ligands alone.
The p-p interactions observed for 2 in the crystal are no more the
governing interactions, but CH-p interactions become the major
interactions in the metal complexes. In the case of ligand 3,
oligomers are formed through the coordination of palladium to the
carboxylic acids, compared to the complexed palladium on the
2
+
pyridine groups in the case of 2. Pd (2) complex remains
2
+
embedded in the hydrophobic part of the bilayer, while Pd (3)
complex spans the entire bilayer, with the carboxylic acids
exposed to the hydrophilic portion of the phosphatidylcholine
layer.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
Chloride transport
[
20]
Having previously reported the anionophoric properties of 1, we
investigated the ability of 2, 3 and their respective metal
complexes to induce anion transport across
a synthetic
phospholipid membrane. Chloride transport studies were
performed in the same EYPC large unilamellar liposomes (LUVs).
The intravesicular solution of the liposomes was 5 mM phosphate
buffer with 500 mM NaCl (pH = 7.2) and the chloride efflux
mediated outside the liposomes was measured with a chloride-
selective microelectrode. Kinetic experiments were carried out by
varying the concentration of ligands from 2.5 mol% to 25 mol%
relative to the EYPC concentration (5 mM). The quantity of
chloride released in the extravesicular solution (5 mM phosphate
3
buffer and 500 mM NaNO , pH = 7.2) was directly correlated to
the efficiency of the compound to mediate chloride transport. At
the end of the experiment, a solution of Triton X was used to lyse
the liposomes and record the maximum of chloride efflux. Based
on these kinetic results, a Hill plot analysis was performed to
determine the half-maximum effective concentration of the
compounds after 500 seconds (EC50,500s) at 25 °C.
The structural modifications of the 2,4,7-triphenylbenzimidazole
scaffold affected the transport efficiency, as the EC50,500s values
were respectively 0.34 mM (6.9 mol% relative to EYPC
concentration) for 1, 1 mM (20.0 mol%) for 2 and 0.63 mM (12.6
mol%) for 3 (See ESI, Figures S14 – S19). In the case of ligand
1
, addition of palladium did not induce any major change in the
transport properties (see ESI Figure S20), however, a completely
different scenario occurred when metals were added to ligands 2
and 3. For example, the addition of 0.5 equivalent of PdCl
2
(
relative to the concentration of 2) at 270 s after the chloride efflux
was initiated by 2, resulted in the assembly of an active MOSTs
in the phospholipid bilayer and an increase in transport efficiency.
Initially, metals (0.5 equivalent relative to 2) that were previously
used to perform UV-vis titrations were studied for their efficiency
to form MOSTs in the phospholipid bilayer. An increase of chloride
2
+
3+
efflux was observed in presence of the Pd and Ru , compared
to 2 alone. Although HAuCl and AgBF were able to form a
complex with in apolar solvents, their solubility in the
4
4
Figureꢀ 7.ꢀ Molecularꢀ modeling:ꢀ a)ꢀ Crystalꢀ packingꢀ ofꢀ 2ꢀ andꢀ PM6/SCF-MOꢀ
2
2
optimizedꢀgeometriesꢀofꢀ3ꢀinꢀtheꢀgasꢀphase.ꢀb)ꢀPalladiumꢀcomplexesꢀPd (
+
2
)ꢀandꢀ
3)ꢀinꢀanꢀEYPCꢀbilayerꢀafterꢀaꢀ200ꢀpsꢀmolecularꢀdynamicsꢀsimulation.ꢀTheꢀ
2
Pd (
+
extravesicular aqueous solution limited their insertion in the
hydrophobic part of the phospholipid membrane and prevented
resultingꢀmetal-ligandꢀcomplexesꢀwereꢀextractedꢀfromꢀtheꢀphospholipidꢀbilayerꢀ
forꢀclarityꢀandꢀonlyꢀaꢀpartialꢀviewꢀofꢀtheꢀphospholipidꢀisꢀshown.ꢀc)ꢀProposedꢀ
hierarchicalꢀassociationꢀinꢀtheꢀgel.ꢀꢀ
the formation of the MOSTs. Consequently, only PdCl
2 3
and RuCl
were studied for further transport experiments.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
Figureꢀ8.ꢀꢀChlorideꢀeffluxꢀobtainedꢀafterꢀtheꢀadditionꢀofꢀ0.5ꢀequivalentsꢀofꢀdifferentꢀmetalꢀsourcesꢀ(relativeꢀtoꢀtheꢀconcentrationꢀofꢀligand)ꢀatꢀ270ꢀsꢀinꢀtheꢀpresenceꢀofꢀ
2.5ꢀmol%ꢀ(relativeꢀtoꢀtheꢀEYPCꢀconcentration)ꢀofꢀ ꢀ(left)ꢀandꢀ ꢀ(right)ꢀinꢀliposomesꢀatꢀ25ꢀ°C.ꢀTritonꢀXꢀwasꢀaddedꢀatꢀ525ꢀsꢀtoꢀlyseꢀtheꢀliposomes.ꢀ
1
2
3
2
+
3+
The addition of the Pd and Ru to 3 resulted in the same
increase of the chloride efflux rate, demonstrating the assembly
of the MOSTs in the phospholipid bilayer (Figure 8). For both
ligands, the most efficient MOSTs were assembled with
Table 2. Chloride efflux at 500 s obtained after the addition of different
concentrations of PdCl at 270 s in presence of 1, 2 and 3
2
2
palladium. Different amounts of PdCl were added at 270 s during
Ligand
No. of equiv. of PdCl
2
% Cl efflux
(measured at 500 s)
41 ± 3
the chloride transport experiment in presence of the three ligands
and the maximums of chloride efflux at 500 s are reported in Table
(
added at 270 s)
2
. Addition of the metals alone only induced a slight chloride efflux
0
0.5
1
(
Figures S26 and S31). It is important to underline that the
increase in chloride efflux is not due to the presence of a higher
concentration of chloride (displaced from the palladium complex
by the solvent), as only 0.015 mM chloride could be generated
1
50 ± 3
50 ± 5
from the 2.5 mM PdCl
2
(the highest concentration used),
0
31 ± 2
53 ± 2
74 ± 4
75 ± 5
41 ± 2
56 ± 4
69 ± 3
69 ± 5
compared to the 500 mM NaCl present in the extravesicular
solution.
0
.25
.5
2
0
1
2
+
No significant influence of the amount of Pd was observed in the
case of 1. A maximum of chloride efflux was obtained when 0.5
0
equivalents of PdCl
formation of a 2:1 stoichiometry in the MOST. Addition of 1
equivalent of PdCl to 3 (required to form the 1:1 complex
2
were added to ligand 2, showing the
0
.25
3
0
.5
2
previously observed by UV-vis) did not increase the chloride
efflux, suggesting the formation of a 2:1 complex inside the
phospholipid membrane. Consequently, the same experiment
1
was repeated with RuCl
metal-ligand assembly in the phospholipid membrane with 2 and
(see ESI, Figures S32 and S33). Thus, in the phospholipid
3
confirming the formation of the 2:1
Additional kinetic experiments were carried out to extract the
EC50,500s for each MOST. Transmembrane chloride transport was
3
2
+
3+
initiated by addition of 1, 2 and 3 and 0.5 equivalent of PdCl
70 s. The maximum of chloride efflux obtained at 500 s was
measured and a new Hill analysis was performed. The EC50,500s
2
at
membrane, only 0.5 equivalent of metal (Pd or Ru ) is
necessary to assemble active MOSTs. This result, put together
with the molecular modeling results (where a 3:2 complex was
formed in the phospholipid bilayer) and the adduct observed by
mass spectrometry (Figure S13), suggests the formation of stable
2
2
+
2+
obtained for Pd(1), Pd (2)and Pd (3) were respectively 0.35 mM
7.0 mol%), 0.47 mM (9.5 mol%) and 0.54 mM (10.9 mol%). The
(
2
+
chloride efflux induced by Pd (2) was increased from 31 to 75 %,
2
:1 complexes in solution and oligomers of higher stoichiometry
in the phospholipid bilayer.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
2
+
while that induced by Pd (3) was increased from 41 to 69 % (see
4 (details and graphics for these experiments available in Figures
S36 - S38). No drastic decrease of the chloride efflux was
observed in the presence of cholesterol, suggesting the formation
of porous transmembrane architectures for both ligands.
ESI, Figures S20 – S25).
We also performed chloride transport experiments with the
preformed metal-organic complex for ligands 2 and 3. Complexes
A slight decrease of the chloride efflux efficiency in the
2
+
2+
Pd (2) and Pd (3)were initially prepared in methanol and added
to the liposome solution. As shown in Table 3, the preformed
complexes did not induce the same chloride efflux as the MOST
assembled in situ in the phospholipid bilayer. This suggests that
the preformed complexes are not stable in the extravesicular
aqueous solution, as the chloride efflux is very similar to that
obtained with the ligand alone. To confirm this hypothesis, a new
EYPC/cholesterol liposomes for the metal assemblies was
observed and may be associated to the dynamics of the self-
assembly of the MOST. Indeed, the ligands may have reduced
mobility inside this rigidified membrane, which may slow down the
self-association process and thus, the chloride transport.
Table 4. Chloride efflux at 500 s obtained for ligands 1, 2 and 3, with or without
addition of PdCl
2
at 270 s, in EYPC or EYPC/cholesterol liposomes
2
aliquot of 0.5 equivalent of PdCl was added 270 s after the
2
+
addition of the preformed Pd (2). An increased chloride efflux
was observed, reaching the same maximum as the previous
ones, confirming that the initially preformed complex
disassembles in aqueous media; the ligand alone permeates the
EYPC /
cholesterol
EYPC
No. of equiv.
Ligand
of PdCl
2
% Cl efflux (at 500s)
% Cl efflux(at
5
00s)
bilayer and when additional PdCl
2
is added, the MOST is formed
directly in the phospholipid bilayer.
-
0.5
-
41 ± 3
50 ± 3
30 ± 3
74 ± 4
40 ± 2
40 ± 3
30 ± 3
60 ± 3
1
2
3
We also confirmed by UV-vis spectroscopy that only 15 % of the
2
+
preformed Pd (2) penetrates the phospholipid bilayer. The entire
formation of the complex can only be achieved through addition
of an additional aliquot of PdCl
2
(see ESI Figure S10). An
0.5
-
intermediate transport efficiency was obtained with the preformed
2
+
41 ± 2
67 ± 3
42 ± 3
64 ± 2
Pd (3), which confirms its higher stability in aqueous media.
0
.5
Table 3. Chloride efflux at 500s obtained for ligands 2, 3 and their respective
2
+
2+
preformed complex Pd (2) and Pd (3), with or without the addition of PdCl
2
at 270 s.
The anion selectivity of the MOST was also investigated. Chloride
transport was conducted in presence of different external buffer
solutions. The external nitrate ion was replaced by fluoride and
-
%
Cl Efflux (at 500 s)
Without PdCl
2
Upon addition of 0.5 eq. of sulfate. The size and the hydrophilicity of the external anions had
PdCl
2
at 270 s
2+
no influence on the chloride efflux in the presence of Pd (3)
-
+
MOST, suggesting that a symport Cl /cation mechanism was
2
30 ± 1
30 ± 2
41 ± 1
53 ± 3
74 ± 4
68 ± 5
67 ± 3
68 ± 2
-
+
-
dominant. However, the symport Cl /cation and the antiport Cl
/
-
2+
anion probably occurs at the same time for Pd (2) which saw its
efficiency slightly reduced in presence of fluoride and sulfate
Figure 9).
Preformed
2
+
Pd (2)
(
3
The formation of MOST in the phospholipid membrane to ensure
anion transport is an important finding and our hypothesis was
that a MOST could also disrupt ion homeostasis in living
organisms. As we previously reported that benzimidazolium
Preformed
2
+
Pd (3)
transmembrane
transporters
possess
antimicrobials
[
26]
In order to determine the mechanism of the chloride efflux induced
by the MOSTs, transport experiments were performed in 7/3
EYPC/cholesterol liposomes. Cholesterol rigidifies and orders the
membrane by increasing the energy barrier of the movement of
phospholipid inside the bilayer (rotation, lateral diffusion or
properties, the toxicity of the MOST to Gram-positive wild-type
and tetracycline-resistant bacteria was investigated. Our
hypothesis was that if MOST can be formed in the bacterial
membranes and they possess a porous structure, they could act
as the hydraphiles reported by Gokel et al. , making the
bacterial membrane more permeable to ions and antibiotics.
[
27]
[
25]
phospholipid flip-flop).
In the case of
a
transmembrane
channel, the transport efficiency is not influenced by the rigidity of
the membrane, whereas the chloride efflux in the case of a mobile
carrier is significantly reduced. The chloride efflux obtained in
EYPC and EYPC/cholesterol liposomes are summarized in Table
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
Figureꢀ9.ꢀChlorideꢀeffluxꢀobtainedꢀafterꢀtheꢀadditionꢀofꢀ0.5ꢀequivalentsꢀofꢀPdCl
3
2
ꢀ(relativeꢀtoꢀtheꢀconcentrationꢀofꢀtheꢀtransporter)ꢀatꢀ270ꢀsꢀinꢀtheꢀpresenceꢀofꢀ12.5ꢀ
ꢀinꢀliposomesꢀbathingꢀinꢀdifferentꢀbuffersꢀatꢀ25°C.ꢀTritonꢀXꢀwasꢀaddedꢀatꢀ525ꢀsꢀtoꢀlyseꢀtheꢀliposomes.
mol%ꢀ(relativeꢀtoꢀtheꢀEYPCꢀconcentration)ꢀofꢀa)ꢀ
2
ꢀorꢀb)ꢀ
Antimicrobial activity
Table 5. Antimicrobial activity (MIC) and transport chloride efficacy (EC50,500s
of 1, 2 and 3*
)
The bacterial membrane is a complex environment containing
different proteins and a peptidoglycan layer in addition to the
Bacteria
B. thuringiensis
MIC (µM)**
> 100
Liposomes
EC50,500s
(mol %)
6.9
[
28]
phospholipid bilayer.
The ability of a synthetic compound to
No. of equiv.
Ligand
penetrate the artificial phospholipid membrane of a liposome and
the one of living bacteria can be very different.
of PdCl
2
The formation of MOSTs in complex bacterial membranes, their
ability to disrupt ion homeostasis and the effect on the survival
rate of living bacteria were investigated. We first determined the
minimum inhibitory concentration (MIC) for ligands 1, 2 and 3
alone in Gram-positive wild-type Bacillus thuringiensis (HD73)
-
1
2
3
0
.5
> 100
7.0
-
> 100
> 100
20.0
9.5
0
.5
(
Table 5).
-
35 - 40
15 - 20
12.6
10.9
The first observation that can be made from this study is that the
efficiency of the three ligands to transport chloride in liposomes
0.5
(
EC50) cannot be correlated to their ability to kill Gram-positive B.
*
*
PdCl
2
was not toxic even at 100 µM (Figure S46)
thuringiensis bacteria. Ligand 3 is the only one possessing a MIC
of 35 - 40 µM against B. thuringiensis. These experiments were
*100 µM was the limit of solubility in the bacterial growth media
2
+
2+
also performed with the preformed complex Pd (2) and Pd (3),
but no difference was noted in terms of the MIC values. These
results were not surprising, since we previously observed that the
metal-ligand complexes were not stable in an aqueous
However, in the following two hours the doubling time decreases
to 77 ± 1 min, showing the disassembly of the metal-ligand
complex. A more pronounced effect can be observed after the
environment. However, if 0.5 equivalent of PdCl
2
was added after
addition of the second aliquot of PdCl
69 ± 1 min was observed. However, after 24 h of incubation, the
global bacterial growth was not affected by the presence of 2 and
2
, when a doubling time of
1
h of incubation of the bacteria with the ligand 3, the MIC of Pd(3)
2
complex was reduced 2-fold. This result demonstrates the
formation of the MOST in the bacterial membrane and its capacity
to alter its permeability.
2
+
Pd in solution. These kinetic experiments suggest that the
2
+
Pd (2) MOST forms rapidly in the bacterial membrane but is not
2
+
stable enough to persist and function over a long period of time.
Intrigued by the lack of antibacterial activity of 2 and its Pd
2
+
2+
complex after 24 h of incubation, we performed a kinetic
experiment on the bacterial growth (Figure 10).
The higher stability of the Pd (3) compared to Pd (2) can also
be inferred from these experiments, as a notable difference in the
doubling time can be observed upon the addition of the first
2
Addition of 0.5 equivalent of PdCl to 25 µM of 2 results in an
aliquot of PdCl
2
to 3. In the presence of ligand 3 alone, despite its
increase of the doubling time from 82 ± 1 min to 234 ± 1 min in
efficacy at the beginning, bacteria entirely grown after 24 hours of
the first two hours.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
incubation, where a complete inhibition of the bacterial growth
they form in the bacterial membrane. To mimic a resistant strain,
2
+
was observed with the Pd (3) complex.
we used
a
B. thuringiensis tetracycline-resistant strain,
possessing a plasmid that expresses an efflux pump in the
bacterial membrane that limits the access of tetracycline to the
Antibiotic resistance has become a major problem in the
[
29]
treatment of Gram-positive bacterial infections.
These
[30]
ribosome. The presence of this plasmid confers a resistance to
organisms are able to escape antibiotic activity through several
tetracycline up to 670 µM (See ESI, Figure S47).
mechanisms including β-lactamase production, altered penicillin-
2
+
binding
proteins,
aminoglycoside-modifying
enzymes,
Firstly, we determined that 3 and MOST Pd (3) at 25 µM were
[
29-
modification of the target site of the antibiotic and active efflux.
non-toxic to this strain (Figure 11). As a reminder, at the same
3
0]
2+
Therapeutic strategies to kill these resistant Gram-positive
concentration, both 3 and MOST Pd (3) were toxic to the wild-
bacteria include the use of a higher antibiotic dosage, the use of
alternative, non-conventional drugs, alone or in combination and
type B. thuringiensis, but the presence of the efflux pump affects
their capacity to remain into the bacterial membrane after 24 h.
[
31]
the development of new drugs.
Bacteria use natural efflux
pumps to eject foreign substances such as antibiotics. Based on
the ability of the MOST to make the membrane permeable, we
thought interesting to exploit the possibility to facilitate the
passage of small antibiotic molecules in the porous architectures
Figure 10. Optical density of B. thuringiensis obtained after several hours of incubation with different concentration of 2 and 3. On the right, 0.5 equivalent of PdCl
relative to the ligand concentration) was added after 1h and 5h of incubation of the ligand with bacteria. Doubling time G was defined as the time taken by bacteria
to double in number during the exponential phase. G= t/(3.3logb/B) with t= time interval in minutes, b= number of bacteria at the end of the time interval and B=
2
(
[
32]
number of bacteria at the beginning of the time interval.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
Conclusions
We
have
successfully
synthesized
two
2,4,7-
triphenylbenzimidazole derivatives and demonstrated the
formation of their respective metal assemblies in phospholipid
membranes. Anion transport experiments performed with these
ligands and their metal complexes allowed us to show their
efficiency as chloride transporters in liposomes. The formation of
2
+
2+
MOST Pd (2) and Pd (3) in the phospholipid membrane results
in efficient assemblies in terms of chloride transport. We also
demonstrated the insertion of the ligands and formation of their
metal-organic assembly in complex bacterial membranes and
successfully applied MOST to make bacterial membrane more
permeable to antibiotics. Finally, as a potential strategy to combat
2
+
resistant bacteria, we showed that a combination of 3 and Pd (3)
with tetracycline lowers the sensitivity of resistant bacteria to
tetracycline by 60-fold. Work is in progress in our group to
demonstrate the generality and the therapeutic interest to use
MOST in pathologically relevant resistant strains.
Figure 11. Tetracycline-resistant B. thuringiensis growth in presence of 5
µl DMSO (used as a blank to monitor the effect of the DMSO on the
bacterial growth) and 25 µM of
µl in DMSO) was added after 1 h of incubation with
growth was measured after 24 h of incubation at 37 °C.
3
(5 µl in DMSO). The PdCl
2
aliquot (2.5
3
. Maximum of bacteria
Experimental Section
2
+
Bacterial growth was monitored in the presence of 3 and Pd (3)
MOST at 25 µM in the presence of different tetracycline
concentrations, below the MIC (670 µM). For each tetracycline
concentration, a blank with only DMSO was performed to confirm
the bacterial growth under these conditions.
Materials. L-α-phosphatidylcholine was purchased from Avanti Polar
Lipids and was used without further purification. All chemicals were
purchased from Aldrich Chemicals in their highest purity and used without
further purification. CD
Isotopes.
3 6
OD and DMSO-d were purchased from CDN
2
+
As shown in Figure 11, the presence of 3 and Pd (3) MOST
Instrumentation. NMR experiments were recorded on 400 MHz Brüker
Avance 400 or Bruker Advance 700 spectrometer at 298 K. Chemical shifts
are reported in ppm relative to residual coupling constants are given in
Hertz (Hz). High-resolution mass spectra (HRMS) were recorded on a TSQ
Quantum Ultra (thermo Scientific) triple quadrupole with accurate mass
option instrument. Chloride efflux was recorded with a chloride selective
micro electrode.
decreased the tolerance of these resistant bacteria to tetracycline.
2
+
Pd (3) MOST appears to be more effective than 3, being more
permeable and inhibiting the bacterial growth at lower
concentrations of tetracycline (Figure 11). Even at 11 µM
tetracycline (60-fold lower than the MIC), the bacterial growth rate
2
+
is surprisingly reduced. The combination of 25 µM of Pd (3)
MOST and tetracycline (110 µM) completely inhibits the bacterial
growth, whereas the same concentrations of tetracycline alone
have no influence on the bacterial growth of this resistant strain.
Preparation of EYPC large unilamellar vesicles (LUVs): A phospholipid
film was formed by evaporating 1 ml chloroform solution containing 25 mg
of EYPC, under vacuum at 25°C during 2 hours. The lipid film was then
hydrated with 1 mL of a NaCl (500 mM) and phosphate buffer solution (5
mM, pH = 7.2). The obtained suspension was subjected to at least 8
freeze/thaw/vortex cycles (1 cycle = 1 minute at - 78°C followed by 1
minute at 35°C and 1 minute in vortex). The solution was then extruded
through a 100 nm polycarbonate membrane 21 times until the solution was
transparent and passed down a Sephadex G-25 column to remove
extravesicular NaCl. The liposomes were eluted with a solution containing
2
+
It may be that formation of the Pd (3) MOST in the bacterial
membrane may compensate for the tetracycline efflux induced by
the pump, allowing a faster influx or an increased amount of
tetracycline into the bacteria. This also suggests that MOSTs
possess a porous architecture, as tetracyclin, which is much
larger than chloride, can be transported into bacteria. This result
5
mM of phosphate buffer with 500 mM of NaNO3 (pH = 7.2). 4.3 mL of
liposomes solution were isolated after separation. The stock solution was
diluted to obtain mM lipid solution, assuming all EYPC was
incorporated into the liposomes.
points to
a
synergy between MOST and tetracycline,
demonstrating the importance and potential application of MOST
as a strategy to fight resistant bacteria.
a 5
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
Chloride transport assays with EYPC LUVs: A 60 μL aliquot of the
solution of EYPC LUVs (5 mM) was added to a 1.2 mL gently stirred buffer
solution containing 5 mM phosphate salt and 500 mM NaNO3 (pH = 7.2).
The chloride efflux was monitored as function of time by a chloride-
selective electrode. 7.5 µL of a solution of transporter at different
concentrations in MeOH were added to start the transport. At t = 600 s,
[8]
a) P. A. Gale, N. Busschaert, C. J. E. Haynes, L. E. Karagiannidis and I.
L. Kirby, Chem.Soc.Rev. 2014, 43, 205-241; b) N. Busschaert, C.
Caltagirone, W. Van Rossom and P. A. Gale, Chem. Rev. 2015, 115,
8
038-8155.
[
[
9]
a) N. Sakai and S. Matile, Langmuir, 2013, 29; b) S. Matile, A. Vargas
Jentzsch, J. Montenegro and A. Fin, Chem Soc Rev 2011, 40, 2453-
2474; c) T. Saha, S. Dasari, D. Tewari, A. Prathap, K. M. Sureshan, A.
K. Bera, A. Mukherjee and P. Talukdar, J Am Chem Soc 2014, 136,
1
00 µL of a Triton-X 5% solution were added to lyse all liposomes to obtain
the maximum of chloride efflux. Experiments were repeated in triplicate
and all the reported traces are the average of the three independent trials
1
4128-14135.
Minimal inhibitory concentration (MIC): 5 mL of Lysogeny Broth (LB)
medium were inoculated with Bacillus thuringiensis (HD73 strain). The
preculture was grown overnight at 37°C under agitation at 230 RPM, and
resuspended in 75 mL of a fresh LB medium. The culture was grown at
10] a) M. G. Fisher, P. A. Gale, M. E. Light and S. J. Loeb, Chem Commun
(Camb) 2008, 5695-5697; b) C. R. Bondy, P. A. Gale and S. J. Loeb, J
Am Chem Soc 2004, 126, 5030-5031; c) S. Ma, M. M. Smulders, Y. R.
Hristova, J. K. Clegg, T. K. Ronson, S. Zarra and J. R. Nitschke, J Am
Chem Soc 2013, 135, 5678-5684.
3
7°C during 2 more hours. After this time, the culture was rediluted in fresh
LB medium to obtain an OD600 of 0.1-0.2. Each well was filled with 195
µL bacterial cultures and 5 µL of DMSO or compounds in DMSO solution,
as the final volume in each well was 200 µL and the maximum
concentration of DMSO was 5%. The plates were agitated at 230 RPM in
a thermostated incubator at 37°C and the solution OD600 was monitored
over time.Every experiment was repeated in triplicate on independent
bacterial cultures. The MICs were determined as the minimal
concentration at which no bacterial growth was detected.
[
[
11] a) O. V. Kulikov, R. Li and G. W. Gokel, Angew Chem Int Ed Engl 2009,
4
8, 375-377; b) M. Jung, H. Kim, K. Baek and K. Kim, Angew Chem Int
Ed Engl 2008, 47, 5755-5757.
12] E. Mahon, S. Garai, A. Muller and M. Barboiu, Adv. Mater. 2015, 27,
5
165-5170.
Acknowledgements
[13] a) M. Boccalon, E. Iengo and P. Tecilla, J Am Chem Soc 2012, 134, 20310-
0313; b) F. De Riccardis, I. Izzo, D. Montesarchio and P. Tecilla, Acc
2
Chem Res 2013, 46, 2781-2790.
We gratefully acknowledge the Natural Sciences and Engineering
Research Council of Canada (NSERC), the Fonds Québécois de
la Recherche sur la Nature et les Technologies (FRQ-NT) and the
Université de Montréal for the financial support. We also thank Le
Centre Régional de RMN and Le Centre Régional de
Spectrométrie de masse of the Université de Montréal. We thank
prof. J.N. Pelletier for access to instruments in her laboratory and
helpful discussion of this manuscript.
[
[
14] U. Devi, J. R. Brown, A. Almond and S. J. Webb, Langmuir 2011, 27,
1448-1456.
15] A. Satake, M. Yamamura, M. Oda and Y. Kobuke, J Am Chem Soc 2008,
1
30, 6314-6315.
[
[
16] T. M. Fyles and C. C. Tong, New J. Chem. 2007, 31, 655-661.
17] a) C. P. Wilson and S. J. Webb, Chem Commun (Camb) 2008, 4007-
4
009; b) C. P. Wilson, C. Boglio, L. Ma, S. L. Cockroft and S. J. Webb,
Keywords: metal-organic assemblies • 2,4,7-triphenyl
benzimidazole • transmembrane transport• antibacterials •
tetracycline
Chemistry 2011, 17, 3465-3473.
[
[
18] C. Janiak, Dalton Trans. 2003, 2781-2804.
19] a) G. A. Santillan and C. J. Carrano, Inorg Chem 2008, 47, 930-939; b) M.
Eddaoudi, D. B. Moler, H. Li, B. Chen, T. M. Reineke, M. O'Keeffe and
O. M. Yaghi, Accounts of Chemical Research 2001, 34, 319-330.
[
[
[
1]
2]
3]
G. Ferey, Chem Soc Rev 2008, 37, 191-214.
J. R. Long and O. M. Yaghi, Chem Soc Rev 2009, 38, 1213-1214.
a) L. J. Murray, M. Dinca and J. R. Long, Chem Soc Rev 2009, 38, 1294-
[
[
[
20]
J. Kempf, N. Noujeim and A. R. Schmitzer, RSC Adv. 2014, 4, 42293-
42298.
1
314; b) R. E. Morris and P. S. Wheatley, Angew Chem Int Ed Engl 2008,
7, 4966-4981.
21] S. Medici, M. Peana, V. M. Nurchi, J. I. Lachowicz, G. Crisponi and M.
4
A. Zoroddu, Coordination Chemistry Reviews 2015, 284, 329-350.
[
[
[
[
4]
5]
6]
7]
P. Horcajada, R. Gref, T. Baati, P. K. Allan, G. Maurin, P. Couvreur, G.
22] a) N. Noujeim, K. Zhu, V. N. Vukotic and S. J. Loeb, Org. Lett. 2012,
Ferey, R. E. Morris and C. Serre, Chem Rev 2012, 112, 1232-1268.
2
484-2487; b) K. Zhu, C. A. O'Keefe, V. N. Vukotic, R. W. Schurko and
P. Horcajada, C. Serre, M. Vallet-Regi, M. Sebban, F. Taulelle and G.
S. J. Loeb, Nat Chem 2015, 7, 514-519.
Ferey, Angew Chem Int Ed Engl 2006, 45, 5974-5978.
[
23] N. Şırecı, Ü. Yilmaz, H. Küçükbay, M. Akkurt, Z. Baktir, S. Türktekın and
O. Büyükgüngör, Journal of Coordination Chemistry 2011, 64, 1894-
1902.
N. Ahmad, H. A. Younus, A. H. Chughtai and F. Verpoort, Chem Soc Rev
2
015, 44, 9-25.
E. Gouaux and R. Mackinnon, Science 2005, 310, 1461-1465.
[24]
a) A. Krogul, J. Cedrowski, K. Wiktorska, W. P. Ozimiński, J. Skupińska
and G. Litwinienko, Dalton Trans. 2012, 41, 658-666; b) D. Ray, J. T.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
Foy, R. P. Hughes and I. Aprahamian, Nature Chemistry 2012, 4, 757-
[29] E. Bouza and R. Finch, Clinical Microbiology and Infection 2012, 7, 3.
7
62.
[
30]
B. S. Speer, N. B. Shoemaker and A. A. Salyers, Clin Microbiol Rev
1992, 5, 387-399.
[
[
25] M. R. Krause and S. L. Regen, Acc. Chem. Res. 2014, 47, 3512-3521.
26] C. R. Elie, G. David and A. R. Schmitzer, J. Med. Chem. 2015, 58, 2358-
[31] J. M. Rybak, K. E. Barber and M. J. Rybak, Expert Opin Pharmacother
2013, 14, 1919-1932.
2
366.
[
[
27] J. L. Atkins, M. B. Patel, Z. Cusumano and G. W. Gokel, Chem Commun
Camb) 2010, 46, 8166-8167.
[32] T. J. Daskivich, M. M. Regan and W. K. Oh, J Urol 2006, 176, 1927-1937.
(
28] N. Malanovic and K. Lohner, Biochim Biophys Acta 2015.
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201700847
FULL PAPER
ꢀ
ꢀ
EntryꢀforꢀtheꢀTableꢀofꢀContentsꢀꢀ
ꢀ
FULL PAPER
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Formation of metal-organic synthetic
transporters (MOST) in phospholipid
Julie Kempf, Andreea R. Schmitzer*
membranes.
MOST
make
the
Page No. – Page No.
membranes of resistant bacteria more
permeable to antibiotics. As a potential
strategy to combat resistant bacteria,
we demonstrate that the combination
of MOST with tetracycline lowers the
sensitivity of resistant bacteria by 60-
fold.
Metal-organic synthetic transporters
(
MOST): efficient chloride and
antibiotic transmembrane
transporters
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
This article is protected by copyright. All rights reserved.