Supramolecular Transmembrane Anion Transport - American

Article Cite This: Acc. Chem. Res. XXXX, XXX, XXX−XXX

pubs.acs.org/accounts

Supramolecular Transmembrane Anion Transport: New Assays and Insights Xin Wu, Ethan N. W. Howe, and Philip A. Gale*

Downloaded via UNIV OF SUSSEX on August 1, 2018 at 03:39:33 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

School of Chemistry (F11), The University of Sydney, Sydney, NSW 2006, Australia CONSPECTUS: Transmembrane anion transport has been the focus of a number of supramolecular chemistry research groups for a number of years. Much of this research is driven by the biological relevance of anion transport and the search to find new treatments for diseases such as cystic fibrosis, which is caused by genetic problems leading to faulty cystic fibrosis transmembrane conductance regulator (CFTR) channels, which in turn lead to reduced chloride and bicarbonate transport through epithelial cell membranes. Considerable effort has been devoted to the development of new transporters, and our group along with others have been searching for combinations of organic scaffolds and anion binding groups that produce highly effective transporters that work at low concentration. These compounds may be used in the future as “channel replacement therapies”, restoring the flux of anions through epithelial cell membranes and ameliorating the symptoms of cystic fibrosis. Less effort has been put into gaining a fundamental understanding of anion transport processes. Over the last 3 years, our group has developed a number of new transport assays that allow anion transport mechanisms to be determined. This Account covers the latest developments in this area, providing a concise review of the new techniques we can use to study anion transport processes individually without resorting to measurement of exchange processes and the new insights that these assays provide. The Account provides an overview of the effects of anion transporters on cells and an explanation of why many systems perturb pH gradients within cells in addition to transporting chloride. We discuss assays to determine whether anionophores facilitate chloride or HCl transport and how this latter assay can be modified to determine chloride versus proton selectivity in smallmolecule anion receptors. We show how molecular design can be used to produce receptors that are capable of transporting chloride without perturbing pH gradients. We cover the role that anion transporters in the presence of fatty acids play in dissipating pH gradients across lipid bilayer membranes and the effect that this process has on chloride-selective transport. We also discuss how coupling of anion transport to cation transport by natural cationophores can be used to determine whether anion transport is electrogenic or electroneutral. In addition, we compare these new assays to the previously used chloride/ nitrate exchange assay and show how this exchange assay can underestimate the chloride transport ability of certain receptors that are rate-limited by nitrate transport.

1. INTRODUCTION The transmembrane transport of anions across lipid bilayer membranes is a challenge that brings supramolecular chemistry and medicinal chemistry together to develop potential future treatments for diseases such as cystic fibrosis that are caused by dysregulation of this process.1−4 Great strides have been made

in the design and synthesis of receptors that function as discrete molecular transporters at low transporter to lipid ratios while being structurally simple (i.e., having “druglike” structures that obey rules of thumb such as Lipiniski’s rule of five5). One class of transporters that exemplifies this is the family of o-phenylenediamine-based bisureas of general structure 1 (Figure 1).6 For example, compound 2 was found to be effective as a chloride transporter in a chloride/ nitrate exchange assay designed to measure the flux of chloride from 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) vesicles. In addition, this compound was shown to reduce the viability of a range of human cancer cell lines and to depolarize acidic compartments within melanoma A375 cells stained with acridine orange.6 Compound 3 was found to be effective at even lower transporter to lipid ratios7 and to facilitate chloride transport across epithelial cell membranes.8 Other classes of hydrogen-bond-donor anion transporters such as squaramides behave in a similar way.9 Compound 4 (Figure

Figure 1. Structures of compounds 1−4.

Received: June 4, 2018

© XXXX American Chemical Society

A

DOI: 10.1021/acs.accounts.8b00264 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 2. Transmembrane assays using POPC vesicles to assess (a) proton transport, (b) HCl transport, and (c) chloride selectivity.

1) has been shown to mediate the transport of chloride into FRT and HeLa cells (with concomitant transport of sodium through endogenous cation channels) but also to increase lysosomal pH, leading to disruption of the autophagic process.10 In both cases, the transporters function in cells to mediate chloride transport but also dissipate pH gradients. We investigated the mechanism of this pH dissipation process to see whether we could design transporters that would mediate chloride flux while not affecting the pH, as this process has been linked to apoptosis.11 Such compounds would be of use to electrophysiologists as a tool to study transport processes as well as being potential future channel replacement therapies.

Figure 4. Structures of tren-based tristhioureas 12−15.

hydroxypyrene-1,3,6-trisulfonic acid trisodium salt (HPTS), was used to monitor the pH inside the vesicles. Anion transporters dissipate the pH gradient via H+/Cl− symport or functionally equivalent OH−/Cl− antiport, leading to an increase of the internal pH and the resultant ratiometric fluorescence response of HPTS. However, previously mechanistic insights had not been available regarding how anion transporters facilitate proton transport. Furthermore, the Cl− transport activity could be underestimated in this assay if anion transporters are weak H+ or OH− transporters. As a first step in investigating pH dissipation triggered by anionophores, we modified the HPTS assay to investigate solely H+ or OH− transport (Figure 2a).13 In this assay, we replaced NaCl with sodium gluconate (100 mM) as the internal and external salt. Gluconate is a large and highly hydrophilic anion, and we can assume that it cannot be transported through the lipid bilayer. The vesicles were suspended in isotonic sodium gluconate solution. A pulse of tetrabutylammonium hydroxide (5 mM) was then added to the extravesicular solution to increase the pH to 8 and generate a pH gradient across the bilayer. Finally, the transporter was added to the suspension to start the experiment. Tetrabutylammonium is lipophilic and can diffuse across the lipid bilayer to balance the charge of any proton efflux facilitated by the added anionophore (the membrane is not permeable to protons in the absence of the anionophore under these conditions and time scales). We also devised a complementary assay to determine whether an anionophore could facilitate HCl cotransport (Figure 2b). In this assay, N-methyl-D-glucammonium chloride (NMDG-Cl, 100 mM) was used in place of sodium gluconate (100 mM), and N-methyl-D-glucamine (NMDG, 5 mM) was used in place of TBAOH (5 mM) as the base added to create a pH gradient. NMDG+ is a large hydrophilic cation that again we can assume cannot be transported through the bilayer. Upon creation of the pH gradient by addition of NMDG, the only transport process that can occur is cotransport of HCl through the bilayer, as transport of either H+ or Cl− alone under these conditions would result in the formation of a charge gradient across the membrane. The conditions used in

2. NEW ASSAYS FOR PROTON AND HCL COTRANSPORT AND TO MEASURE CHLORIDE SELECTIVITY IN TRANSPORT The HPTS assay has been routinely used by supramolecular chemists to determine the ion transport activity and selectivity

Figure 3. Structures of compounds 5−11, the weak acid protonophore CCCP, and the HCl cotransporting natural product prodigiosin.

of synthetic ion transport systems.12 In a typical experimental setup, vesicles containing an internal solution of NaCl (100 mM) buffered at pH 7 were suspended in an external solution of NaCl (100 mM), and the external pH was brought to ∼8 by addition of a NaOH pulse. A fluorescent probe, 8B


Article


Figure 5. Cation coupling assays with (a) valinomycin and (b) monensin.

gramicidin divided by the EC50 value in its presence. Hence, a transporter with S > 1 possesses a degree of chloride selectivity. We found that compounds 5−10 (Figure 3) had little chloride selectivity, with S values of 1.1, 2.1, 1.4, 1.0, 0.8, and 1.5, respectively. We also measured the chloride selectivity of N,N′-bis(3,5-trifluoromethylphenyl)squaramide (11), which also did not show any chloride selectivity (S = 0.8). However, when we started to examine series of compounds with increasing degrees of anion encapsulation, we did begin to see evidence of chloride-selective transport. The assay was applied to the series of tren-based tristhioureas 12−15 (Figure 4). Compound 12 has a chloride selectivity S of 14. The introduction of electron-withdrawing cyano groups results in loss of selectivity, with an S value of 0.9 for compound 13. Presumably this occurs because the more acidic compound functions as a weak acid protonophore in a similar fashion as CCCP in addition to transporting chloride. Replacing the cyanophenyl groups with n-pentyl groups restores the selectivity, with an S value of 39 for compound 14. Increasing the steric bulk of the substituent (tert-pentyl) increases the S value to 78 for compound 15. We found a similar effect with nonbridged cholapods14 having a lower selectivity for chloride than bridged cholaphanes15 synthesized by the Davis group at the University of Bristol. Compound 15 has the highest selectivity for chloride transport among this series of functionalized tren-based compounds. We used cationophore coupling assays to confirm the mechanism of transport of this compound. These assays, shown in Figure 5, are used to determine whether the anionophore functions as an electrogenic chloride transporter (i.e., transporting only chloride across the membrane, resulting in a net flow of charge) or an electroneutral H+/Cl− cotransporter (i.e., transporting both a proton and a chloride anion across the membrane, resulting in no net flow of charge) or if it can function in both capacities in a nonspecific manner. The assays employ vesicles that contain KCl with intra- and extravesicular solutions buffered at pH 7.4. Under the conditions of the assay, an anionophore alone will not

this assay are similar to those of the HPTS assays employed by Matile and co-workers.12 With both the proton transport assay and the HCl transport assay, we studied a variety of simple hydrogen-bond-donor anion transporters 5−10, the weak acid protonophore CCCP (which is a known transmembrane proton transporter), and the natural product prodigiosin (a highly effective transmembrane HCl cotransporter). By studies of these ionophores at different concentrations, Hill plot analyses were performed to obtain a Hill coefficient, which indicated the stoichiometry of the species mediating ion transport, and the effective concentration to reach 50% of maximum transport at 200 s (the EC50 value) to quantify the ion transport activity. We found that, as expected, CCCP dissipated a pH gradient in the proton transport assay but was inactive in the HCl cotransport assay. Conversely, again as expected, prodigiosin was not active in the proton transport assay but was active in the HCl cotransport assay with a very low EC50 value. We then tested compounds 5−10 and found them to be active in both assays, meaning that all of these compounds were capable of dissipating pH gradients across membranes via H+ and HCl transport processes, although with a range of EC50 values. Given that all of these simple systems were unselective in transporting both protons and chloride, we devised an assay in order to measure the chloride selectivity of a transporter (by modification of the HCl transport assay). We ran this assay again, but in the presence of gramicidin-D, a natural proton channel. A chloride-selective transporter may have a very high EC50 in the HCl transport assay, as it is not capable of effectively transporting HCl. However, in the presence of gramicidin-D, the proton channel will mediate the flux of protons while the chloride-selective transporter facilitates the diffusion of chloride across the membrane. Therefore, a hallmark of a chloride-selective transporter will be a high EC50 value in this assay in the absence of gramicidin and a low EC50 in its presence. We define the chloride selectivity S as the EC50 value of the transporter in this assay in the absence of C


Article


Figure 7. Electrogenic chloride transport as mediated by compound 15.

presence of monensin, this is evidence that the anionophore functions as a H+/Cl− cotransporter, as monensin functions as a K+/H+ exchanger. In the presence of a H+/Cl− cotransporter, the proton transfer processes balance out, resulting in overall KCl efflux (Figure 5b). If an anionophore facilitates chloride transport in the presence of either monensin or valinomycin, this is evidence that it can function as both a H+/Cl− cotransporter and a Cl− uniporter and hence is nonspecific in its transport mechanism. The results of coupling assays for prodigiosin, urea 5 and tristhiourea 15 are shown in Figure 6. Figure 6a shows that prodigiosin couples to monensin and not to valinomycin, confirming that the natural product acts as a H+/Cl− cotransporter but not a chloride uniporter. Figure 6b shows that urea 5 couples to both valinomycin and monensini.e., the urea can transport both chloride anions and HCl and thus is a nonspecific anionophore. Tristhiourea 15 couples only to valinomycin and not to monensin, indicating that compound 15 functions as an electrogenic chloride transporter (Figure 7). Preliminary studies showed that while compound 15 could facilitate the transport of chloride across epithelial cell membranes in FRT cells expressing the halide sensor YFP-H148Q/I152L, it did not completely depolarize the acidic compartments within human lung adenocarcinoma (A549) cells.13

3. MECHANISMS OF PROTON TRANSPORT In our studies, we found that most simple hydrogen-bonddonor anionophores were unselective and would transport HCl in addition to transporting chloride. This led us to suggest a number of potential mechanisms for proton transport by anionophores. For acidic compounds carrying electron-withdrawing substituents, there is evidence that the compounds function as weak acid protonophores in a similar fashion to CCCP and FCCP (Figure 8a).13 There is also evidence that less acidic receptors and some halogen-bonding transporters12b may function as hydroxide transporters (Figure 8b).13 However, we also considered a third mechanism involving the interaction of fatty acids present in the lipid bilayer16 with anion transporters (Figure 8c).17 In this process, the transporter binds to a deprotonated fatty acid carboxylate headgroup and facilitates its transport through the lipid bilayer. On the more acidic side of the membrane, the headgroup is protonated, and the now-neutral carboxylic acid headgroup can diffuse through the bilayer and be deprotonated on the less acidic side. Without the anion transporter, the deprotonated fatty acid diffuses very slowly across lipid bilayers because of the highly hydrated carboxylate headgroup. Thus, the combination of fatty acids and synthetic anion transporters may serve to dissipate pH gradients across lipid bilayers in a synergistic manner. To demonstrate this process, we prepared large unilamellar vesicles (LUVs) of POPC loaded with and suspended in 100 mM potassium gluconate solution and buffered at pH 7.0 with

Figure 6. Coupling between cationophores and anionophores to facilitate KCl efflux: (a) prodigiosin couples with monensin but not valinomycin; (b) compound 5 couples to both monensin and valinomycin; (c) compound 15 couples to valinomycin but not monensin.

transport chloride because there is no mechanism to balance the flow of negative charge out of the vesicle. Valinomycin (Figure 5a) is a cationophore that functions as an electrogenic potassium transporter. When both valinomycin and an electrogenic chloride transporter are added to the vesicles, KCl efflux will occur (Figure 5a). Hence, under these conditions chloride efflux is evidence in support of an electrogenic chloride transport mechanism. However, if an anionophore facilitates chloride efflux in this system in the D


Article


Figure 9. Intravesicular pH (as measured by HPTS fluorescence) in response to the addition of oleic acid (OA) at 1 min, valinomycin (Vln) at 2 min, compound 12 at 2.5 min (red trace only), and monensin (Mon) at 10 min. The POPC LUVs (mean diameter 200 nm) were loaded with and suspended in a 100 mM potassium gluconate (KGlc) solution buffered at pH 7.0 with 10 mM HEPES. All of the compounds were added as DMSO solutions. The vertical arrows indicate the additions of compounds, and the tilted arrows indicate the assignment of the HPTS response to the flip-flop of neutral OA or oleate ions. Compound concentrations are shown as compound-to-lipid molar ratios.

making the interior of the vesicle more acidic. At 2 min into the experiment, valinomycin was added, which allowed potassium to be transported through the membrane to balance the charge of any potential proton transport process. When nothing further was added to the vesicle suspension, the pH remained fairly stable over time (blue trace in Figure 9). However, when a synthetic anion transporter was added (in this case compound 12), deprotonated fatty acids were transported from the exterior to interior leaflet of the lipid, were protonated, and then diffused back across the membrane to dissipate the pH gradient (Figure 8c). At the end of the experiment in both cases, monensin was added to dissipate any remaining pH gradient. As the combination of fatty acids and synthetic anion transporters can facilitate transmembrane proton transport, we wished to investigate the effect of varying fatty acid concentrations on this process. Commercially available POPC contains varying levels of fatty acids such as palmitic acid (PA) and oleic acid (OA), but these can be sequestered from vesicles using bovine serum albumin (BSA). We used the assay shown in Figure 10 to follow the rate of pH dissipation from LUVs containing and suspended in potassium gluconate solution buffered with HEPES. HPTS was used to follow the pH inside the vesicle, and monensin was added at the end of the experiment to collapse the pH gradient, allowing calibration of the HPTS fluorescence. In the presence of 2 mol % oleic acid, there was little change in the pH over time. Similarly, when the vesicles were treated with BSA and then compound 12 was added, there was little change in the pH. However, when compound 12 was added to untreated vesicles (which contained fatty acid impurities), the pH gradient slowly dissipated (magenta trace in Figure 10). When compound 12 was added to vesicles containing 2 mol % OA, the pH gradient dissipated more quickly (red trace in Figure 10). These results are consistent with the fatty acid flip-

Figure 8. Anion transporters facilitating electrogenic (electrophoretic) H+/OH− transport leading to dissipation of an electrochemical proton gradient. (a) Proton transport via a weak acid protonophore (deprotonation−reprotonation) mechanism. (b) Hydroxide transport via reversible binding of OH−. (c) Proton transport via a fatty acid cycling mechanism in which the anion transporter facilitates the flip-flop of the carboxylate form of fatty acids present in the lipid bilayer.

HEPES buffer. The pH inside the vesicles was monitored with HPTS. A dimethyl sulfoxide (DMSO) solution of oleic acid was added to the vesicles after 1 min. The pH inside the vesicles dropped as most oleic acid partitioned into the lipid bilayer and the neutral carboxylic acid headgroup diffused through the bilayer to the inner leaflet, where it deprotonated, E


Article


Figure 12. Bar chart showing the 1/EC50 values for compounds 18− 20 from POPC LUVs (200 nm) loaded with NMDG-Cl (100 mM) and HPTS (1 mM) suspended in NMDG-Cl (100 mM). The H+/Cl− symport (or OH−/Cl− antiport) dose−response of 18−20 was measured by monitoring the HPTS fluorescence ratio after the addition of a 5 mM NMDG pulse (with an additional pulse of gramicidin-D (0.1 μM, 0.1 mol %) or oleic acid (2 μM, 2 mol %) as appropriate). The Hill equation was used to fit the dose−response curves for 18−20 alone, with gramicidin-D, and with oleic acid.

chloride versus H+ transport selectivities of compounds 12 and 15 under “fatty-acid-free” and “fatty-acid-rich” conditions. We found enhanced selectivity for both compounds in the absence of fatty acids, with S values of 160 and 690 for compounds 12 and 15, respectively. In the presence of 2 mol % oleic acid the selectivity diminished, with S values of 1.7 and 3.9, respectively, because of accelerated H+ transport. These results led us to investigate whether transporters with lower affinities for carboxylates than for chloride might function as chloride-selective transporters in the presence of fatty acids.18 Calix[4]pyrroles are a class of cyclic tetrapyrroles that function as anion receptors via the formation of four pyrrole NH···anion hydrogen bonds (Figure 11).19 The parent macrocycle 16 has been shown to function as a CsCl cotransporter, with the chloride anion binding to the pyrrole NH groups preorganizing the macrocycle into a cone conformation that can then bind the cesium cation via cation−π interactions with the calix[4]pyrrole cup.20 In the absence of cesium, this compound does not function as a chloride transporter. However, enhancing the affinity of the macrocycle for chloride, either by attaching electron-withdrawing substituents such as fluorine21 (compound 17) or by introducing additional CH hydrogen-bond-donor triazole groups as part of a strap22 (compounds 18−20), “switched on” chloride transport in the absence of cesium, and these compounds could facilitate chloride/nitrate across lipid bilayers. We re-examined the transport properties of the series of strapped calix[4]pyrroles 18−20 using the newly developed assays for measuring chloride transport selectivity and assessing the anion transport mechanism. Cationophore-coupled assays (as described above) were performed with compounds 18−20. It was found that the compounds would couple to valinomycin but not to monensin. Therefore, the compounds function predominantly as electrogenic chloride transporters and do not significantly facilitate HCl cotransport. We then employed a modified version of the chloride selectivity assay (Figure 2c) using either gramicidin-D or oleic acid to facilitate proton transport. The results of this assay are shown in bar chart form

Figure 10. (a) Electrogenic H+/OH− transport assay that follows the rate of pH dissipation from POPC LUVs (mean diameter ca. 200 nm). (b) H+ (or counter OH−) transport induced by 12 (or OA as a control) measured by the assay shown in (a) under the conditions shown in the figure. Valinomycin (0.05 mol %) was used in all cases to facilitate potassium transport in order to balance proton transport. The vesicles were treated with monensin (0.1 mol %) at 200 s to collapse the pH gradient for calibration of the HPTS fluorescence. Compound concentrations are shown as compound-to-lipid molar ratios.

flop mechanism proposed above, which requires both fatty acids and a synthetic anion transporter to be present to conduct protons and shows that the fatty acid concentration is important in determining the rate of pH gradient dissipation. The work also demonstrates that even traces of fatty acid impurities present in commercial lipid samples can dramatically accelerate the rates of pH gradient dissipation facilitated by anion transporters in HPTS assays, an effect that seems to have been overlooked in the past.

4. SELECTIVE ANION TRANSPORT IN THE PRESENCE OF FATTY ACIDS Given that anion transporters can facilitate the transport of protons via the fatty acid flip-flop mechanism, we examined the

Figure 11. Calix[4]pyrroles 16−20. F


Article

Accounts of Chemical Research Table 1. Summary of Cl− > H+/OH− Selectivities for 16 and 18−20 receptor

EC50 (mol %h)

EC50 Gra (mol %h)a

SGb

EC50 OA (mol %h)c

FOAd

SOA/Ge

SOA/G − SGf

16 18 19 20

− 10.9 4.1 2.7

− 0.6 0.4 0.2

− 18.2 10.3 13.5

− 11.2 3.6 1.6

− 1.0 1.1 1.7

− 18.7 9.0 8.0

−g 0.5 −1.3 −5.5

g

g

g

g

g

g

EC50 in the presence of gramicidin-D. This value shows the total H+/Cl− symport (OH−/Cl− antiport) activity possible with no rate-limiting H+/ OH− transport. The Gra concentration added was optimized at 0.1 mol % to prevent this from having a limiting effect. bThe Cl− > H+/OH− selectivity value SG was calculated by dividing EC50 in the absence of Gra by EC50 Gra. SG > 1 indicates Cl− selectivity. cEC50 in the presence of oleic acid (2 mol %). This value helps to show the effect of the fatty acid on the selectivity results. dFactor of overall H+/Cl− symport enhancement in the presence of OA. FOA is calculated by dividing EC50 in the absence of Gra or OA by EC50 OA. FOA > 1 suggests that the receptor has affinity for OA− head groups, enhancing pH gradient dissipation. eCl− > H+/OH− selectivity value in the presence of oleic acid. SOA/G is calculated by dividing EC50 OA by EC50 Gra. SOA/G > 1 indicates Cl− selectivity. fIf SOA/G − SG ≈ 0, there is no selectivity loss in the presence of OA. gNo transport ability in this assay; no Hill analysis results or selectivity factors were calculated. hConcentration of receptor with respect to lipid. a

Table 2. Overview of the Association Constants and Thermodynamic State Functions for the 1:1 Complexation of Tetraethylammonium (TEA) Salts with 16 and 18−20 As Measured in Acetonitrile by Isothermal Titration Calorimetry at 303 K chloridea

acetateb

receptor

Ka (M−1)

ΔG (kJ mol−1)

ΔH (kJ mol−1)

TΔS (kJ mol−1)

Ka (M−1)

ΔG (kJ mol−1)

ΔH (kJ mol−1)

TΔS (kJ mol−1)

16 18 19 20

× × × ×

−30.1 −37.2 −37.9 −40.2

−42.3 −49.7 −52.7 −50.4

−12.8 −12.1 −14.6 −10.0

× × × ×

−33.2 −29.7 −36.2 −35.4

−45.3 −41.4 −51.0 −42.6

−12.1 −11.7 −14.8 −7.3

1.9 2.6 3.5 8.7

5c

10 106 106 106

5.2 1.3 1.7 1.2

5

10 105 106 106

a

From ref 20. bFrom ref 16. cThe host solution was titrated into the guest solution.

across the lipid bilayer, with the chloride efflux measured using a chloride-selective electrode. At the end of the experiment, the vesicles are lysed with detergent to calibrate the experiment. This assay relies on the assumption that nitrate transport is not rate-limiting. This is a reasonable assumption to make, as nitrate is more lipophilic than chloride. However, it is also true that nitrate is more challenging to bind with hydrogen-bonddonor anion receptors than chloride because of its lower charge density. We therefore decided to compare the chloride/ nitrate exchange assay with two newer assays: an HPTS assay (Figure 13b) and a valinomycin-coupled osmotic assay (Figure 13c), in which KX cotransport leads to water efflux and a change in the shape of the vesicles, which can be monitored by following light scattering using a fluorimeter.23 We tested six compounds in all three assays (Figure 14). Three of the compounds are planar (4, 5, and 8), while two (12 and 22) are tripodal and hence can encapsulate an anionic guest to a higher degree. A bipodal analogue of these compounds (21) was also tested. One advantage of the more modern assays is that they can be run for a single anion, and hence, it is possible to get separate EC50 values for chloride and nitrate transport. In fact, the two new assays gave consistent results for the anion transport activity and selectivity, although the exact EC50 values were not identical, presumably because of the different experimental conditions used. The results from the HPTS assay are summarized for chloride and nitrate in Figure 15, which plots the reciprocal of EC50 for each compound and anion (hence, longer bars represent more effective transport). The red dashed line shows that compound 4 has the highest activity in facilitating the slower anion uniport process, which explains its highest activity (lowest EC50 value) in the chloride/nitrate exchange assay. The chloride transport properties of the tripodal compounds are underestimated by the chloride/nitrate exchange assay, as nitrate transport in these cases is rate-limiting. The results demonstrate the

in Figure 12. The blue bars represent the 1/EC50 values of the three calixpyrroles in the absence of gramicidin-D or oleic acid (the longer the bar the better the transporter). The 1/ EC50 values were all small, as the compounds are not effective HCl cotransporters. However, in all cases where the proton channel gramicidin-D was added, the 1/EC50 values increased significantly (green bars), as gramicidin-D can facilitate proton transport while the calixpyrroles facilitate chloride transport. Interestingly, when oleic acid was added instead of gramicidinD, the 1/EC50 value for compound 18 did not change significantly (red bars). This shows that in the presence of fatty acids, this calixpyrrole does not facilitate proton transport and remains as a Cl− > H+ selective transporter, unlike the tripodal thioureas 12 and 15. As the strap across the calixpyrrole gets longer, there is an increasing difference between the 1/EC50 values in the absence and presence of fatty acids, meaning that the calixpyrroles with the larger straps are beginning to lose chloride selectivity, presumably because of their ability to facilitate fatty acid flip-flop, as the larger anion binding cavities can accommodate the fatty acid carboxylate headgroup more effectively. These results are summarized in Table 1, and the association constants for 16 and 18−20 are summarized in Table 2. These results show that compound 18 has the highest selectivity for chloride versus acetate among the three strapped calix[4]pyrroles. These results together show that chlorideselective transport is possible even in the presence of significant quantities of fatty acids.

5. NEW VERSUS OLD ANION TRANSPORT ASSAYS Up until the development of assays that couple anion transport to either cation or proton transport, a commonly used assay to assess chloride transport was the chloride/nitrate exchange assay. In this assay (shown in Figure 13a), vesicles containing sodium chloride are prepared and suspended in sodium nitrate solution. An anionophore may facilitate the exchange of anions G


Article


Figure 15. Histogram showing the activities of compounds 4, 5, 8, 12, 21, and 22 facilitating chloride uniport (green bars) and nitrate uniport (blue bars) as determined by the HPTS assay. Activities are expressed as reciprocals of EC50 values (at 200 s).

A further study of fluorinated urea and thiourea analogues of compound 1225 showed that significantly more information about the anion transport mechanism and the effectiveness of transporters for transporting particular anions can be gathered using these new techniques, including EC50 values for the transport of anions such as bicarbonate and sulfate.26

6. CONCLUSION New assays in which anion uniport is coupled either to metal cation transport or proton transport have allowed us to gain a deeper insight into anion transport processes and anion transport selectivity. The role that fatty acids play in mediating proton transport in the presence of anionophores has been elucidated, and we have shown that in order to achieve chloride over proton transport selectivity, carboxylate complexation should be disfavored. In light of these findings, new challenges present themselves. If a greater degree of structural organization is required around the anion binding site in an anionophore to achieve chloride-selective transport, can we make chloride-selective transporters “druglike”?5 Can we make transporters that are selective for other biologically relevant anions such as bicarbonate? Can we make transporters that are

Figure 13. Schematic illustrations of the (a) the classical Cl−/NO3− exchange assay, (b) the HPTS assay, and (c) an osmotic assay.

reliability of the HPTS and osmotic assays in evaluating anion transport activity and selectivity. Although this information might also be assessed by conductance and reversal potential measurements,24 those methods are often not readily available to supramolecular chemists.

Figure 14. Planar squaramide 4, urea 5, and thiourea 8 anion transporters, encapsulating tren-based transporters 22 and 12, and dipodal analogue 21. H


Article


University of Sydney, The Royal Society, the Wolfson Foundation, and the EPSRC for funding.

gated by membrane potential? We are seeking to answer these and other questions.

■

■

AUTHOR INFORMATION

ABBREVIATIONS Glc, gluconate; Gra, gramicidin-D; HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; HPTS, 8-hydroxypyrene1,3,6-trisulfonic acid, trisodium salt; LUV, large unilamellar vesicle; NMDG, N-methyl-D-glucamine; POPC, 1-palmitoyl-2oleoyl-sn-glycero-3-phosphocholine

Corresponding Author

*E-mail: [email protected]. ORCID

Xin Wu: 0000-0002-7715-8784 Ethan N. W. Howe: 0000-0003-4611-2318 Philip A. Gale: 0000-0001-9751-4910

■

Notes

REFERENCES

(1) Gale, P. A.; Davis, J. T.; Quesada, R. Anion Transport and Supramolecular Medicinal Chemistry. Chem. Soc. Rev. 2017, 46, 2497−2519. (2) Gale, P. A.; Howe, E. N. W.; Wu, X.; Spooner, M. J. Anion receptor chemistry: Highlights from 2016. Coord. Chem. Rev. 2018, DOI: 10.1016/j.ccr.2018.02.005. (3) Davis, A. P.; Sheppard, D. N.; Smith, B. D. Development of synthetic transporters for anions. Chem. Soc. Rev. 2007, 36, 348−357. (4) Matile, S.; Vargas Jentzsch, A.; Montenegro, J.; Fin, A. Recent synthetic transport systems. Chem. Soc. Rev. 2011, 40, 2453−2474. (5) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Delivery Rev. 2001, 46, 3−26. (6) Moore, S. J.; Haynes, C. J. E.; González, J.; Sutton, J. L.; Brooks, S. J.; Light, M. E.; Herniman, J.; Langley, G. J.; Soto-Cerrato, V.; Pérez-Tomás, R.; Marques, I.; Costa, P. J.; Félix, V.; Gale, P. A. Chloride, carboxylate and carbonate transport by ortho-phenylenediamine-based bisureas. Chem. Sci. 2013, 4, 103−117. (7) Karagiannidis, L. E.; Haynes, C. J. E.; Holder, K. J.; Kirby, I. L.; Moore, S. J.; Wells, N. J.; Gale, P. A. Highly effective yet simple transmembrane anion transporters based upon ortho-phenylenediamine bis-ureas. Chem. Commun. 2014, 50, 12050−12053. (8) Dias, C. M.; Li, H.; Valkenier, H.; Karagiannidis, L. E.; Gale, P. A.; Sheppard, D. N.; Davis, A. P. Anion transport by ortho-phenylene bis-ureas across cell and vesicle membranes. Org. Biomol. Chem. 2018, 16, 1083−1087. (9) Busschaert, N.; Kirby, I. L.; Young, S.; Coles, S. J.; Horton, P. N.; Light, M. E.; Gale, P. A. Squaramides as Potent Transmembrane Anion Transporters. Angew. Chem., Int. Ed. 2012, 51, 4426−4430. (10) Busschaert, N.; Park, S.-H.; Baek, K.-H.; Choi, Y. P.; Park, J.; Howe, E. N. W.; Hiscock, J. R.; Karagiannidis, L. E.; Marques, I.; Félix, V.; Namkung, W.; Sessler, J. L.; Gale, P. A.; Shin, I. A synthetic ion transporter that disrupts autophagy and induces apoptosis by perturbing cellular chloride concentrations. Nat. Chem. 2017, 9, 667− 675. (11) Cheung, S.; Wu, D.; Daly, H. C.; Busschaert, N.; Morgunova, M.; Simpson, J. C.; Scholz, D.; Gale, P. A.; O’Shea, D. F. Real-Time Recording of the Cellular Effects of the Anion Transporter Prodigiosin. Chem. 2018, 4, 879−895. (12) (a) Matile, S.; Sakai, N. The Characterization of Synthetic Ion Channels and Pores. In Analytical Methods in Supramolecular Chemistry, 2nd ed.; Schalley, C. A., Ed.; Wiley-VCH: Weinheim, Germany, 2012; pp 711−742. (b) Vargas Jentzsch, A.; Emery, D.; Mareda, J.; Nayak, S. K.; Metrangolo, P.; Resnati, G.; Sakai, N.; Matile, S. Transmembrane anion transport mediated by halogen-bond donors. Nat. Commun. 2012, 3, 905. (13) Wu, X.; Judd, L. W.; Howe, E. N. W.; Withecombe, A. M.; Soto-Cerrato, V.; Li, H.; Busschaert, N.; Valkenier, H.; Pérez-Tomás, R.; Sheppard, D. N.; Jiang, Y.-B.; Davis, A. P.; Gale, P. A. Nonprotonophoric Electrogenic Cl− Transport Mediated by Valinomycin-like Carriers. Chem 2016, 1, 127−146. (14) Valkenier, H.; Davis, A. P. Making a Match for Valinomycin: Steroidal Scaffolds in the Design of Electroneutral, Electrogenic Anion Carriers. Acc. Chem. Res. 2013, 46, 2898−2909.

The authors declare no competing financial interest. Biographies Xin Wu received his Ph.D. degree in 2016 under the supervision of Professor Philip A. Gale (then at the University of Southampton) and Prof. Yun-Bao Jiang (Xiamen University). He is now a postdoctoral research associate in Professor Gale’s group in the School of Chemistry at the University of Sydney. His research interests include small-molecule membrane transporters, anion receptor chemistry, and amphiphile self-assembly. He received the A. P. de Silva Early Career Award during the MSMLG meeting in 2016. Ethan N. W. Howe obtained his B.Sc. (Hons.) and Ph.D. degrees in Chemistry from the University of New South Wales in 2010 and 2014, respectively. He moved to the University of Southampton in 2015 as an EPSRC Postdoctoral Research Fellow and then returned to Sydney in 2017 as a postdoctoral researcher at the University of Sydney under the guidance of Professor Philip A. Gale. His research interests include cooperativity in supramolecular interactions and transmembrane ion flux. Philip A. Gale is Professor of Chemistry and Head of the School of Chemistry at the University of Sydney. He received his D.Phil. degree from the University of Oxford in 1995 and then moved as a Fulbright Scholar to the University of Texas at Austin. He returned to Oxford in 1997 as a Royal Society University Research Fellow and then moved to a lectureship at the University of Southampton in 1999, where he became Professor of Supramolecular Chemistry in 2007. From 2010 to 2016 he served as Head of Chemistry at Southampton. He received a D.Sc. degree from the University of Oxford in 2014 and moved to the University of Sydney in 2017. His research focuses on the supramolecular chemistry of anions and has been recognized by a number of awards including the 2004 RSC Bob Hay Lectureship, the 2005 RSC Corday Morgan Medal and Prize, a Royal Society Wolfson Research Merit Award (2013), the 2014 RSC Supramolecular Chemistry Award, and the 2018 International Izatt−Christensen Award in Macrocyclic and Supramolecular Chemistry.

■

ACKNOWLEDGMENTS P.A.G. thanks all of his co-workers and collaborators at the Universities of Sydney and Southampton and elsewhere (their names appear in the references) and especially thanks Professors Tony Davis and David Sheppard at the University of Bristol, Professor Ricardo Pérez-Tomás at the University of Barcelona, Professor Kate Jolliffe at the University of Sydney, Professor Jonathan L. Sessler at the University of Texas at Austin, Professor Injae Shin at Yonsei University, Professor ́ Félix Yun-Bao Jiang at Xiamen University, and Professor Vitor at the University of Aveiro for their ongoing collaborations. We thank the Australian Research Council (DP180100612), The I


Article

Accounts of Chemical Research (15) Judd, L. W.; Davis, A. P. From cholapod to cholaphane transmembrane anion carriers: accelerated transport through binding site enclosure. Chem. Commun. 2010, 46, 2227−2229. (16) Gensure, R. H.; Zeidel, M. L.; Hill, W. G. Lipid raft components cholesterol and sphingomyelin increase H+ /OH− permeability of phosphatidylcholine membranes. Biochem. J. 2006, 398, 485−495. (17) Wu, X.; Gale, P. A. Small Molecular Uncoupling Protein Mimics: Synthetic Anion Receptors a Fatty Acid-Activated Proton Transporters. J. Am. Chem. Soc. 2016, 138, 16508−16514. (18) Clarke, H. J.; Howe, E. N. W.; Wu, X.; Sommer, M.; Yano, M.; Light, M. E.; Kubik, S.; Gale, P. A. Transmembrane Fluoride Transport: Direct Measurement and Selectivity Studies. J. Am. Chem. Soc. 2016, 138, 16515−16522. (19) Gale, P. A.; Sessler, J. L.; Král, V. Calixpyrroles. Chem. Commun. 1998, 1−8. (20) Tong, C. C.; Quesada, R.; Sessler, J. L.; Gale, P. A. mesoOctamethylcalix[4]pyrrole: an old yet new transmembrane ion-pair transporter. Chem. Commun. 2008, 6321−6323. (21) Gale, P. A.; Tong, C. C.; Haynes, C. J.; Adeosun, O.; Gross, D. E.; Karnas, E.; Sedenberg, E. M.; Quesada, R.; Sessler, J. L. Octafluorocalix[4]pyrrole: A Chloride/Bicarbonate Antiport Agent. J. Am. Chem. Soc. 2010, 132, 3240−3241. (22) Yano, M.; Tong, C. C.; Light, M. E.; Schmidtchen, F. P.; Gale, P. A. Calix[4]pyrrole-based anion transporters with tunable transport properties. Org. Biomol. Chem. 2010, 8, 4356−4363. (23) Yang, Y.; Wu, X.; Busschaert, N.; Furuta, H.; Gale, P. A. Dissecting the chloride−nitrate anion transport assay. Chem. Commun. 2017, 53, 9230−9233. (24) Hille, B. Ionic Channels of Excitable Membranes; Sinauer Associates: Sunderland, MA, 2001. (25) Busschaert, N.; Wenzel, M.; Light, M. E.; Iglesias-Hernández, P.; Pérez-Tomás, R.; Gale, P. A. Structure−activity relationships in tripodal anion transporters: the effect of fluorination. J. Am. Chem. Soc. 2011, 133, 14136−14148. (26) Jowett, L. A.; Howe, E. N. W.; Wu, X.; Busschaert, N.; Gale, P. A. New insights into the anion transport selectivity and mechanism of tren-based tris-(thio)ureas. Chem. - Eur. J. 2018, 24, 10475−10487.

J


Supramolecular Transmembrane Anion Transport - American

Recommend Documents