Crown ether-thiourea conjugates as ion transporters

Abstract

Na⁺, Cl^‒ and K⁺ are the most abundant electrolytes present in biological fluids that are essential to the regulation of pH homeostasis, membrane potential and cell volume in living organisms. Herein, we report synthetic crown ether-thiourea conjugates as a cation/anion symporter, which can transport both Na⁺ and Cl⁻ across lipid bilayers with relatively high transport activity. Surprisingly, the ion transport activities were diminished when high concentrations of K⁺ ions were present outside the vesicles. This unusual behavior resulted from the strong affinity of the transporters for K⁺ ions, which led to predominant partitioning of the transporters as the K⁺ complexes in the aqueous phase preventing the transporter incorporation into the membrane. Synthetic membrane transporters with Na⁺, Cl^‒ and K⁺ transport capabilities may have potential biological and medicinal applications.

Key words： ion transport; thiourea; crown ether; symport

1 Introduction

A wide range of synthetic ion transporters have been developed during the past decades owing to their potential biomedical applications [1,2] in treating cancer and channelopathies like cystic fibrosis [3,4], Bartter syndrome [5] and Dent’s disease [6]. Anions transporters were designed to transport HCO₃^–, F^–, Cl^–, I^– and SO₄^2– [7–11], among which Cl^– is the most commonly targeted anion. Besides cystic fibrosis, dysregulation of Cl^‒ transport will also indirectly result in exocrine pancreatic failure, intestinal blockage and male infertility [12]. Hence, synthetic Cl^– transporters have been developed, either working as channels [13] or carriers [14]. Transmembrane cation transport also plays key roles in cellular processes. The concentrations of Na⁺ and K⁺ in cells are regulated with the concentration gradient generated by Na⁺/K⁺ ATPase. The concentration of Na⁺ is ~10 mmol·L⁻¹ inside and ~150 mmol·L⁻¹ outside, however, for K⁺, the extracellular concentrations (5 mmol·L⁻¹) are lower than those in the cytoplasm (150 mmol·L⁻¹) [15]. In biological systems, Na⁺ and K⁺ channels work synergistically in processes such as initiating and regulating action potentials in neurons and cells [16,17]. Mutations of Na⁺ channels are found to be closely linked to inherited epilepsy, migraine, periodic paralysis and chronic pain syndromes [18,19]. In addition, cardiac arrhythmias, deafness and epilepsy involved mutations of several critical K⁺ channels [20,21].

Due to the indispensable role of Na⁺, K⁺ and Cl^– in nearly every animal cell type [22,23], synthetic transporters for Na⁺, K⁺ and Cl^‒ have attracted significant attention. In addition to cation transporters and anion transporters, several researcher groups have developed ion-pair symporters [24–27], which contain at least one cation binding site and one anion binding site. Calix[4]pyrrole was reported by Gale and co-workers to transport CsCl as an ion-pair complex [24,25]. Smith and co-workers reported transporters containing an amide and an azacrown ether moieties to achieve the co-transport of NaCl or KCl [26]. Azacrown ethers were also combined with a urea or a squaramide motif to selectively transport KCl [27,28]. However, due to the high hydration energy of Na⁺ [29,30] and the higher affinity of 15-crown-5 and 18-crown-6 for K⁺ under certain conditions [29], the selective transport of Na⁺ over K⁺ is challenging.

Herein, we synthesized bilateral crown ether-thiourea conjugates Azo-2C5 and Azo-2C6 as ion transporters (Scheme 1), which were designed based on the following considerations: (i) crown ethers, benzo-15-crown-5 in Azo-2C5 and benzo-18-crown-6 in Azo-2C6, are employed to bind and transport Na⁺ and K⁺ across membrane [27,31,32] and (ii) N-amidothiourea groups, which exhibit enhanced anion binding ability in comparison to conventional N-alkyl and N-arylthiourea counterparts, by virtue of an anion-induced conformation change in the N‒N single bond and a multivalent hydrogen-bonding network [33,34]. Thus N-amidothiourea moieties are employed for Cl^– binding and transmembrane transport. (iii) An azobenzene group is utilized as linker to improve the lipophilicity of the transporters [35,36], facilitating ion transmembrane transport. Control compounds Ph-2C5 and Azo-C5 (Scheme 1) were designed to investigate the significance of azobenzene motif and bilateral structure for ion transport, respectively. We found that Azo-2C5 could transport Na⁺ and Cl^– across the membrane with high activity, while Azo-2C6 exhibited a higher Cl^– transport activity and a much lower Na⁺ transport activity owing to the mismatch of Na⁺ with 18-crown-6 [37]. Surprisingly, neither transporter showed K⁺ transport activity, and the presence of K⁺ at high concentrations substantially hindered the transport of Na⁺ and Cl^–. This unique phenomenon has been shown by ultra-violet-visible spectroscopy (UV-Vis) studies to originate from the predominant partitioning of the transporters in the aqueous phase when K⁺ was present in high concentrations.

Fig.1 Scheme 1 Molecule structures of synthetic transporters Azo-2C5, Azo-2C6, and control compounds Azo-C5, Ph-2C5.

Full size|PPT slide

2 Experimental

2.1 Materials

4,4'-Diaminoazobenzene, 1,1'-thiocarbonyldiimidazole, methyl 3,4-dihydroxybenzoate, bis[2-(2-chloroethoxy)ethyl]ether, pentaethylene glycol dichloride 1,4-phenylene diisothiocyanate and 4-phenylazophenyl isothiocyanate were purchased from J&K Scientific Ltd. 1-Palmitoyl-2-oleoylphosphatidylcholine (POPC) and dipalmitoylphosphatidylcholine (DPPC) were purchased from Avanti Polar Lipids. Triton X-100, valinomycin (Vln), carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP), calcein, 8-hydroxypyrene-1,3,6-trisulfonic acid trisodium salt (HPTS), 6-methoxy-N-(3-sulfopropyl)quinolinium (SPQ), sodium gluconate (NaGlc), potassium gluconate, N-methyl-D-glucamine chloride (NMDG-Cl), 4-(2-hydroxyerhyl)piperazine-1-erhanesulfonic acid (HEPES) were purchased from Sigma Aldrich Co. and used directly without purification. Dimethyl sulfoxide-D₆ was purchased from Cambridge Isotope Laboratories Inc. All other starting materials were obtained from Sinopharm Chemical Reagent Ltd.

2.2 Instrumentation

¹H and ¹³C nuclear magnetic resonance (NMR) spectra were recorded on Bruker AV500MHz, AV600MHz or AV850MHz spectrometers. High-resolution mass spectra were acquired on a Bruker micro TOF-Q-II mass spectrometer. Absorption spectra were recorded on Shimadzu UV-2700 UV-Vis spectrophotometer. Fluorescence spectra were obtained on a Horiba Fluorolog-3 spectrometer.

2.3 Ion transport study and the EC₅₀ measurement

2.3.1 Preparation of large unilamellar vesicles (LUVs)

LUVs (mean diameter 200 nm) of POPC loaded with pH-sensitive fluorescence dye HPTS (0.1 mmol·L⁻¹) were prepared by freeze/thaw cycles. A chloroform solution 50 mL) of POPC (10 mg) was slowly evaporated by using a rotary evaporator and dried under high vacuum at room temperature for 4 h. The obtained lipid film was rehydrated by mixing with 1 mL salt solution (100 mmol·L⁻¹ NaGlc, 10 mmol·L⁻¹ HEPES, 0.1 mmol·L⁻¹ HPTS, pH 7.0) and vortexing for 2 min. The mixtures were subjected to 0 freeze-thaw cycles and extruded through a 0.22 μm polycarbonate membrane at least 5 times. To separate the un-entrapped HPTS, the lipid suspension (0.76 mL, 0 mmol) was transferred to a size exclusion chromatography (stationary phase: Sephadex G-50; mobile phase: 100 mmol·L⁻¹ NaGlc, 10 mmol·L⁻¹ HEPES, pH 7.0) and eluted with the mobile phase to acquire a 5 mL lipid stock solution containing 2 mmol·L⁻¹ of lipid.

2.3.2 Na⁺ transport study

In the Na⁺ transport study, 0.1 mL of the lipid stock solution was added to 1.88 mL buffer solution (100 mmol·L⁻¹ NaGlc, 10 mmol·L⁻¹ HEPES, pH 8.0) to generate a pH gradient. DMSO solution (20 μL) with or without transporters was added at time 0. The pH changing inside the LUVs was monitored by fluorescence of HPTS (λ_ex = 454 nm, λ_em = 510 nm). At 300 s, 20% Triton X-100 10 μL) was added as detergent to lysate the LUVs for calibration. The fractional fluorescence intensity (I_f) was calculated based on Eq. (1):

(1)I_{\mathrm{f}}={\displaystyle \frac{R_{\mathrm{t}}-R_{\mathrm{0}}}{R_{\mathrm{f}}-R_{\mathrm{0}}}}\times \mathrm{100}\mathrm{\%},

where R_t is the fluorescence intensity at time t, R_f is the final fluorescence intensity obtained by the addition of detergent, R₀ is the fluorescence intensity at the start time. Different concentrations of transporter molecules were added to obtain a series of fractional fluorescence intensity (I_f). Fitting I_f vs molecule concentration according to q. (2):

(2)y=y_{\mathrm{0}}+(y_{\max }-y_{\mathrm{0}}){\displaystyle \frac{x^n}{K+x^n}},

where y is the value of I_f corresponding to the carrier molecule loaded at concentration x, y₀ is the I_f value measured when compound has not been added, y_max is maximum I_f value, n is the Hill coefficient, K is the EC₅₀ value. In this work, the concentration of LUVs is fixed at 100 μmol·L⁻¹.

The methodologies for tests of other transport activities are similar, which are described in detail in the Electronic Supplementary Material (ESM).

3 Results and discussions

3.1 Transport activities of Na⁺ and Cl^–

The ion transport activities of Azo-2C5 were tested in POPC based LUVs (mean diameter 200 nm) with encapsulated pH-sensitive fluorescence dye HPTS (Figs. 1 and 2). As the transporter contains both anion and cation recognition motifs, i.e., N-amidothiourea and crown ether moieties, the transport processes are studied by two separate experiments. In a Na⁺ transport assay, NaGlc (100 mmol·L⁻¹) was used (Fig. 1(a)) based on the consideration that gluconate is too hydrophilic to pass through the hydrophobic membrane even with an anion transporter [38]. In a Cl^– transport assay (Fig. 2(a)), NMDG-Cl solution (100 mmol·L⁻¹) was used as the inert NMDG⁺ rules out the possibility of cation transport [38]. The transport rate increased rapidly with increasing the concentrations of the transporter and the fluorescence response of HPTS reached a platform of 68% for Na⁺ (Figs. 1(b) and 1(c)) and 62% for Cl^– (Figs. 2(b) and 2(c)). The plots can be fitted well utilizing the Hill equation and the corresponding parameters to give half maximal effective concentrations (EC₅₀) and Hill coefficients (n) values. The EC₅₀ values for Na⁺ (0.12 μmol·L⁻¹) and Cl^– (0.12 μmol·L⁻¹) are nearly the same, and the Hill coefficients for both transport processes are slightly higher than 1. The Hill coefficients indicate that the transporter is more likely to function as a monomer during Na⁺ and Cl^– transport processes [39]. When NaCl was used as the internal and external salt (Fig. S1(a), cf. ESM), a similar EC₅₀ of 0.15 μmol·L⁻¹ and a similar Hill coefficient close to 1 were determined (Fig. S1(c)), thus a 1:1:1 (Azo-2C5:Na⁺:Cl^–) stoichiometry could be expected for the symport of Na⁺ and Cl^– (see below). No change in the transport rate of Na⁺ and Cl^– was observed with the addition of a H⁺ ionophore FCCP (Figs. S(2) and S3(b), cf. ESM), indicating that H⁺ transport by Azo-2C5 is likely not the rate-limiting process. A limited maximum activity of 70% was observed (Fig. 1), and we assumed that the aggregation of the transporters in solution could be responsible, which limited the delivery of the transporters to the vesicles [40].

Fig.2 (a) Schematic representation of quantitative measurement of Na⁺ transport conducted by using a pH gradient of 7 inside and 8 outside in LUVs (mean diameter 200 nm) encapsulating pH-sensitive dye HPTS. Inside LUVs: 0.1 mmol·L⁻¹ HPTS, 100 mmol·L⁻¹ NaGlc, 10 mmol·L⁻¹ HEPES, pH 7.0. Outside LUVs: 100 mmol·L⁻¹ NaGlc, 10 mmol·L⁻¹ HEPES, pH 8.0. Normalized fluorescence intensity (NFI) obtained by addition of different concentrations of Azo-2C5 ((b) 0–4 μmol·L⁻¹) and Azo-2C6 ((d and e) 0–0.6 μmol·L⁻¹). (c) Hill analysis of Na⁺ transport facilitated by Azo-2C5.

Full size|PPT slide

Compared with Azo-2C5, Azo-2C6 has a significantly reduced transport activity for Na⁺ (Fig. 1, 0.12 μmol·L⁻¹ for EC₅₀ of Azo-2C5; EC₅₀ of Azo-2C6 is not detectable), but a better transport activity for Cl^– (Fig. 2, 0.12 μmol·L⁻¹ for EC₅₀ of Azo-2C5; 0.05 μmol·L⁻¹ for EC₅₀ of Azo-2C6). The former is likely due to the mismatch between the cavity size of 18-crown-6 in Azo-2C6 (134–143 pm) and the size of Na⁺ (95 pm) [37]. The latter is ascribed to the higher binding affinity of Azo-2C6 with Cl⁻ than those of Azo-2C5 (Figs. S4 and S5, cf. ESM. For Azo-2C6 in acetonitrile, K₁₁ = 1915 L·mol⁻¹, K₁₂ = 424 L·mol⁻¹; for Azo-2C5 in acetonitrile, K₁₁ = 1250 L·mol⁻¹, K₁₂ = 341 L·mol⁻¹). An experiment employing FCCP (Fig. S3(c)) also indicated that H⁺ transport by Azo-2C6 is likely not the rate-limiting process. The reference compounds, Ph-2C5 and Azo-C5 display nearly no transport activity no matter what salt solution was used (Figs. S6–S8, cf. ESM). The inactivity of Ph-2C5 indicates the essential role of azobenzene scaffold in facilitating the ion transport, which is ascribed to the lower lipophilicity of benzene than azobenzene motif. Notably, unilateral compound Azo-C5 resembles a classic surfactant structure containing a polar head-group and a lipophilic tail, which is known to be detrimental to the transmembrane transport [41], underscoring the importance of bilateral structure of Azo-2C5 in ion transport.

Fig.3 (a) Schematic representation of quantitative measurement of Cl^‒ transport conducted by using a pH gradient of 7 inside and 8 outside in LUVs (mean diameter 200 nm) encapsulating pH-sensitive dye HPTS. Inside LUVs: 0.1 mmol·L⁻¹ HPTS, 100 mmol·L⁻¹ NMDG-Cl, 10 mmol·L⁻¹ HEPES, pH 7.0. Outside LUVs: 100 mmol·L⁻¹ NMDG-Cl, 10 mmol·L⁻¹ HEPES, pH 8.0. NFI obtained by addition of different concentrations of Azo-2C5 ((b) 0–2.0 μmol·L⁻¹) and Azo-2C6 ((d) 0–0.4 μmol·L⁻¹). (c,e) Hill analysis of Cl^– transport facilitated by (c) Azo-2C5 and (e) Azo-2C6.

Full size|PPT slide

To obtain direct evidence for Na⁺ and Cl^– transport, ²³Na NMR and SPQ experiments were conducted respectively. In the NMR experiment (Fig. S9, cf. ESM), DyCl₃ was added to discriminate the signals from intra- and extravesicular solutions by upfield shifting of outer ²³Na peak [32]. The NMR spectra at different incubation time confirmed the Na⁺ transport ability of Azo-2C5. In the SPQ experiment (Fig. 3), as the inert NMDG can not pass through the membrane, the SPQ experiment showed relatively slower influx of Cl^– when the outer salt solution was NMDG-Cl (Figs. 3(a) and 3b). Replacing the NMDG-Cl with NaCl provides the transportable cation and hence the fluorescence quenching was observed (Figs. 3(c) and 3(d)), suggesting the symport of Na⁺/Cl^–. SPQ experiments, together with the ²³Na NMR experiment, confirmed that Azo-2C5 can act as a symporter for NaCl.

Fig.4 (a,c) Schematic representation of SPQ experiment to verify the Cl⁻ transport. Inside LUVs (200 nm): 0.5 mmol·L⁻¹ SPQ, 100 mmol·L⁻¹ NaGlc, 10 mmol·L⁻¹ HEPES, pH 7.0. Outside LUVs: (a) 100 mmol·L⁻¹ NMDG-Cl or (c) NaCl, 10 mmol·L⁻¹ HEPES, pH 7.0. (b,d) NFI obtained by addition of different concentrations of Azo-2C5 (0–0.2 μmol·L⁻¹).

Full size|PPT slide

3.2 Influence of K⁺ on the transport activities

18-Crown-6 moiety has a cavity radius of 134–143 pm, which matches the radius of K⁺ (133 pm) [37], whereas the 15-crown-5 moiety has been used to construct K⁺-selective channels [42] by forming a sandwich complex with K⁺ [29] or construction of constitutional polarized ion channels [30]. Zeng synthesized a series of cation transporters containing 15-crown-5 or 18-crown-6 moiety as cation receptor, most of which showing high selectivity for K⁺ over Na⁺ [42–44]. Other crown ethers reported by Zhu [32,45], Barboiu [30,46] and Giuseppone [47] are also K⁺-selective transporters. Given those literature examples of crown ether-based K⁺ transporters, the inactivity of both Azo-2C5 and Azo-2C6 towards K⁺ transport (Fig. S8) is intriguing. Valkenier, Šindelář and co-workers have reported that the transporter’s extremely high affinity for NO₃^– results in the poor Cl^‒/NO₃^‒ exchange activity [7]. Similarly in our current work, the presence of K⁺ is shown to retard the Na⁺ transport. We thus examined the relationship between transport activity and concentrations of K⁺ (Fig. 4). We added KCl during Na⁺ transport. As shown in Fig. 4(b), the addition of KCl reduced the Na⁺ transport rate remarkably [29]. Similarly, addition of KCl also induced a decrease of Cl^– transport rate when Azo-2C5 or Azo-2C6 was used as Cl^– transporters (Fig. S10, cf. ESM). In the Cl^– transport experiment, even 1 mmol·L⁻¹ of KCl can dramatically reduce the transport rate. With a NaCl or NMDG-Cl concentration of 100 mmol·L⁻¹, the addition of 1 mmol·L⁻¹ KCl would exert nearly no influence on the concentration of Cl^–, and therefore the decrease of Na⁺ and Cl^– transport activity can be ascribed to the presence of K⁺. It is likely that because of strong K⁺ affinity, the concentration of the uncomplexed transporter is low compared with that of the K⁺ complex, leading to the transport process being rate-limited by the diffusion of uncomplexed transporters.

To further investigate the selectivity of the crown ether-based transporters, we performed the HPTS assay with different cations inside and outside in the presence of a pH gradient. Under the non-symmetric ionic conditions, a membrane potential would be generated if the internal and external ions have different transport rates. This would give rise to an internal pH change, the direction of which indicates the selectivity of the transporter under study [48]. With NaGlc (100 mmol·L⁻¹) outside and KGlc (100 mmol·L⁻¹) inside (Fig. S11, cf. ESM), the internal pH (indicated by the HPTS fluorescence intensity) underwent an initial decrease despite the pH outside being higher than the pH inside, and then slowly increases following the direction of the pH gradient. The initial decrease of internal pH against the pH gradient indicates a positive membrane potential generated with K⁺ outside and Na⁺ inside, leading to the conclusion that the transporters are selective for K⁺ over Na⁺. This is consistent with the hypothesis that reduced transport activity with K⁺ arises from the extremely strong K⁺ binding affinity. Other transport experiments conducted when applying both a pH gradient and an ion concentration gradient (Na⁺, K⁺, Figs. S12 and S13, cf. ESM) also demonstrate that the transporters can transport K⁺ faster than Na⁺, Cl^– and H⁺.

Fig.5 (a) Schematic representation of quantitative measurement of the influence of K⁺ on the Na⁺ transport activity conducted by addition of different amounts of KCl to extravesicular solution before the measurement; (b) Na⁺ transport activities of Azo-2C5 (0.3 μmol·L⁻¹) measured with (red line) and without (blue line) addition of KCl (100 mmol·L⁻¹) (KCl was added after the transport process proceeded for several seconds); (c) NFI obtained by addition of Azo-2C5 (0.3 μmol·L⁻¹) and premix of different amounts of KCl; (d) Point data of NFI measured with Azo-2C5 (0.3 μmol·L⁻¹) and different concentrations of KCl.

Full size|PPT slide

3.3 Quantitative measurement of K⁺ transport

Given the diminished activity of the transporters when substantial K⁺ was present outside the vesicles, we attempted to quantitatively measure the transport efficiency of K⁺ by using K⁺ only inside the vesicles. An HPTS assay was conducted with 100 mmol·L⁻¹ of KCl inside and 100 mmol·L⁻¹ NaCl outside. The driving force was a concentration gradient of K⁺, since that the efflux of K⁺ would generate a H⁺ influx to result in the difference between starting and final pH in the vesicles (Fig. 5). At the end of the transport experiment, the final fluorescence intensity of HPTS corresponding to the transport equilibrium was obtained by the addition of Vln (10 μmol·L⁻¹) and FCCP (10 μmol·L⁻¹). The results demonstrate comparable activities of the two transporters (EC₅₀ values of 0.45 μmol·L⁻¹ and 0.32 μmol·L⁻¹ for Azo-2C5 and Azo-2C6, respectively) and also the similar Hill coefficients (n = 2.12 for Azo-2C5, 2.14 for Azo-2C6). Therefore, K⁺ is assumed to be transported as a sandwich type 2:1 (transporter:K⁺) complex [29,39]. The anion component was found to have negligible impact on the transport rates, as the replacement of the chloride by gluconate salt led to similar EC₅₀ and n values (Fig. S14, cf. ESM, EC₅₀: 0.65 μmol·L⁻¹ for Azo-2C5, 0.48 μmol·L⁻¹ for Azo-2C6; n: 2.65 for Azo-2C5, 2.46 for Azo-2C6). Since K⁺ has been shown to inhibit the transport when present at high concentrations outside vesicles (Fig. 4), by swapping the intra- and extravesicular salts (i.e., with KCl outside and NaCl inside), no transport activity was observed (Fig. S15, cf. ESM), and even when a pH gradient was applied, only a very low activity was observed (Fig. S16 , cf. ESM).

Fig.6 (a) Schematic representation of quantitative measurement of K⁺ transport (from inside to outside) conducted by exerting concentration gradient of K⁺ in LUVs (mean diameter 200 nm) encapsulating pH-sensitive dye HPTS. Inside LUVs: 100 mmol·L⁻¹ KCl, 10 mmol·L⁻¹ HEPES, pH 7.0. Outside LUVs: 100 mmol·L⁻¹ NaCl, 10 mmol·L⁻¹ HEPES, pH 7.0. NFI obtained by addition of different concentrations of Azo-2C5 ((b) 0–2.5 μmol·L⁻¹) or Azo-C6 ((d) 0–2.5 μmol·L⁻¹). Hill analysis of K⁺ uniport facilitated by (c) Azo-2C5 or (e) Azo-2C6.

Full size|PPT slide

3.4 Ion transport mechanism

To better understand the unusual behavior of the transporters that depend on the identity of intervesicular and extravesicular ions, we first determined the K⁺ and Na⁺ binding affinities by ITC titrations. Both the compounds have higher affinities for K⁺ than Na⁺ in acetonitrile from ITC titration results (Figs. S17 and S18, cf. ESM, Table 1). Since significant transport activity was only observed when the external solution contains no K⁺, we hypothesized that the transporters have better deliverability to the membrane when K⁺ is present only inside the LUVs. To verify our assumption, the LUVs with diameters of 5 μm were prepared to measure the transporter loading capacity under different experimental conditions. The large size of the LUVs enables easy separation of the LUVs from the solution by centrifugation (Fig. S19, cf. ESM), facilitating the measurement of the transporter loading capacity by the UV-Vis absorption spectra. In the Na⁺ transport experiment (Fig. 6), the transporter loading capacity (indicated by the absorption spectra of inside LUVs) decreased to a fairly low degree after the addition of K⁺, which induced leaching of the transporters into the extravesicular solution. We attributed this to the strong affinity of the transporters for K⁺ leading to accumulation of the transporters in the extravesicular solution when the K⁺ was added, hence the low transporter loading capacity and poor transport activity. On the contrary, under the Na⁺-out-K⁺-in condition (Fig. S20, cf. ESM), the compounds remained in the isolated LUVs. Compared with the volume of the extravesicular solution, the intravesicular volume is much lower. Therefore, even with a high concentration of K⁺ inside the vesicles, a significant portion of the transporters can partition in the membrane and facilitate ion transport. By contrast, if the extravesicular solution contains K⁺, the transporters partition almost exclusively into the extravesicular solution, resulting in poor transport activity. The sum of individual absorption spectral profiles from inside and outside LUVs were comparable to the profiles of reference samples that were prepared in the same way without centrifugation, confirming the reliability of this method.

Fig.7 (a) Schematic representation of test conditions of the absorption spectra; (b, c) Absorption spectra of Azo-2C5 (4 μmol·L⁻¹) in lipid suspension (black line) and disperse inside (blue line, the transporters that disperse in and on the LUVs) and outside (red line, the transporters that disperse in the extravesicular solution). Absorption values of inside (blue line) and outside (red line) are added and plotted (green line) to compare with the unseparated one (black line). (b) Inside LUVs: 100 mmol·L⁻¹ NaGlc, 10 mmol·L⁻¹ HEPES, pH 7.0. Outside LUVs: 100 mmol·L⁻¹ NaGlc, 10 mmol·L⁻¹ HEPES, pH 7.0. (c) Inside LUVs: 100 mmol·L⁻¹ NaGlc, 10 mmol·L⁻¹ HEPES, pH 7.0. Outside LUVs: 100 mmol·L⁻¹ NaGlc, 100 mmol·L⁻¹ KCl, 10 mmol·L⁻¹ HEPES, pH 7.0.

Full size|PPT slide

Tab.1 Overview of the association constants and thermodynamic state functions for the 2:1 complexation of hexafluorophosphate salts with Azo-2C5 and Azo-2C6 as measured in acetonitrile by ITC at 30 °C

Complex	Ka/(L·mol–1)	n	ΔG/(kJ·mol–1)	ΔH/(kJ·mol–1)	TΔS/(kJ·mol–1)
Azo-2C5+Na+	7.53 × 104	1.613	‒28.96	‒4.571	24.39
Azo-2C5+K+	1.82 × 105	2.117	‒30.03	‒6.101	23.93
Azo-2C6+Na+	4.84 × 104	2.043	‒26.74	‒1.453	25.29
Azo-2C6+K+	1.46 × 105	2.395	‒29.48	‒4.231	25.25

The integrity of the vesicle membrane was determined by the calcein leakage assay. Only negligible leakage of calcein occurred over 5 min (Fig. S21, cf. ESM), confirming that the crown ether-thiourea conjugates are working as transporters rather than destroying the vesicles or forming large pores [49].

Synthetic transporters can be classified as ion-channels or mobile carriers based on their working mechanism in lipid bilayers [50]. In our experiments, both the transporters seem to function as carriers for cations and anions. We prepared LUVs with DPPC, a lipid with a phase transition temperature of 41 °C, to explore the transport mechanism. Na⁺ transport was completely suppressed at 31 °C, but switched on when the temperature reached 37 °C, then accelerating to a higher rate at 43 °C (Fig. 7). The transport of Cl^‒ and uniport of K⁺ showed similar temperature-dependence in the DPPC experiments (Figs. S22 and S23, cf. ESM), indicating that both the transporters are operating as ion carriers. The DPPC experiments together with the Hill analysis data suggest that the transporters work as mobile carriers rather than ion-channels [40].

Fig.8 (a) Schematic representation of DPPC experiments to verify the transport mechanism (ion carrier or channel). NFI obtained by addition of DMSO ((b) 20 μL) or DMSO solutions of Azo-2C5 ((c) 0.05 μmol·L⁻¹) at different temperatures ranging from 25 °C to 43 °C.

Full size|PPT slide

4 Conclusions

In summary, we have employed crown ether-thiourea conjugates with azobenzene linkers, i.e., Azo-2C5 and Azo-2C6, as mobile carriers to transport the physiological important Na⁺, Cl^– and K⁺ ions. Though both the transporters are capable of transporting Cl^– across the membrane with considerable activity, owing to the anion binding ability of the thiourea groups, Azo-2C5 is a much better transporter for Na⁺ as its crown ether cavity (benzo-15-crown-5) fits Na⁺ well. Thus Azo-2C5 can facilitate the symport of NaCl. Moreover, the transporters’ high affinity for K⁺ led to predominant partitioning of the transporters in the extravesicular solution when high concentration K⁺ exists outside the LUVs, leading to the suppressed ion transport activity. On the contrary, K⁺ efflux facilitated by the transporters can occur when K⁺ only exists inside the LUVs. These new features provide new insights into the design of cation/anion symporters and the detrimental impact of extremely strong ion binding affinity on the ion transport activity.

Acknowledgements

We greatly appreciate the support of this work by the National Natural Science Foundation of China (Grant Nos. 21820102006, 91856118, 21435003 and 21521004), the MOE of China through Program for Changjiang Scholars and Innovative Research Team in University (Grant No. IRT13036), and the Scientific and Technological Plan Project in Xiamen (Grant No. 3502Z20203025).

Electronic Supplementary Material

Supplementary material is available in the online version of this article at https://dx.doi.org/10.1007/s11705-021-2049-7 and is accessible for authorized users.

References

1 Wu X, Howe E N W, Gale P A. Supramolecular transmembrane anion transport: new assays and insights. Accounts of Chemical Research, 2018, 51(8): 1870–1879

2 Fyles T M. How do amphiphiles form ion-conducting channels in membranes. Lessons from linear oligoesters. Accounts of Chemical Research, 2013, 46(12): 2847–2855

3 Davis A P, Sheppard D N, Smith B D. Development of synthetic membrane transporters for anions. Chemical Society Reviews, 2007, 36(2): 348–357

4 Zhang Z, Chen J. Atomic structure of the cystic fibrosis transmembrane conductance regulator. Cell, 2016, 167(6): 1586–1597

5 Konrad M, Vollmer M, Lemmink H H, Van den Heuvel L P W J, Jeck N, Vargas-Poussou R, Lakings A, Ruf R, Deschenes G, Antignac C, . Mutations in the chloride channel gene CLCNKB as a cause of classic Bartter syndrome. Journal of the American Society of Nephrology, 2000, 11(8): 1449–1459

6 Dutzler R, Campbell E B, Cadene M, Chait B T, MacKinnon R. X-ray structure of a ClC chloride channel at 3.0 Å reveals the molecular basis of anion selectivity. Nature, 2002, 415(6869): 287–294

7 Valkenier H, Akrawi O, Jurček P, Sleziaková K, Lízal T, Bartik K, Šindelář V. Fluorinated bambusurils as highly effective and selective transmembrane Cl⁻/HCO₃⁻ antiporters. Chem, 2019, 5(2): 429–444

8 Clarke H J, Howe E N W, Wu X, Sommer F, Yano M, Light M E, Kubik S, Gale P A. Transmembrane fluoride transport: direct measurement and selectivity studies. Journal of the American Chemical Society, 2016, 138(50): 16515–16522

9 Roy A, Joshi H, Ye R, Shen J, Chen F, Aksimentiev A, Zeng H. Polyhydrazide-based organic nanotubes as efficient and selective artificial iodide channels. Angewandte Chemie International Edition, 2020, 59(12): 4806–4813

10 Busschaert N, Karagiannidis L E, Wenzel M, Haynes C J E, Wells N J, Young P G, Makuc D, Plavec J, Jolliffe K A, Gale P A. Synthetic transporters for sulfate: a new method for the direct detection of lipid bilayer sulfate transport. Chemical Science (Cambridge), 2014, 5(3): 1118–1127

11 Wu X, Judd L W, Howe E N W, Withecombe A M, Soto-Cerrato V, Li H, Busschaert N, Valkenier H, Perez-Tomas R, Sheppard D N, . Nonprotonophoric electrogenic Cl^– transport mediated by valinomycin-like carriers. Chem, 2016, 1(1): 127–146

12 Davis J T, Gale P A, Quesada R. Advances in anion transport and supramolecular medicinal chemistry. Chemical Society Reviews, 2020, 49(16): 6056–6086

13 Ren C, Zeng F, Shen J, Chen F, Roy A, Zhou S, Ren H, Zeng H. Pore-forming monopeptides as exceptionally active anion channels. Journal of the American Chemical Society, 2018, 140(28): 8817–8826

14 Spooner M J, Li H, Marques I, Costa P M R, Wu X, Howe E N W, Busschaert N, Moore S J, Light M E, Sheppard D N, . Fluorinated synthetic anion carriers: experimental and computational insights into transmembrane chloride transport. Chemical Science (Cambridge), 2019, 10(7): 1976–1985

15 Gokel G W, Mukhopadhyay A. Synthetic models of cation-conducting channels. Chemical Society Reviews, 2001, 30(5): 274–286

16 Yu F H, Catterall W A. Overview of the voltage-gated sodium channel family. Genome Biology, 2003, 4(3): 207

17 Goldin A L. Resurgence of sodium channel research. Annual Review of Physiology, 2001, 63(1): 871–894

18 Payandeh J, Scheuer T, Zheng N, Catterall W A. The crystal structure of a voltage-gated sodium channel. Nature, 2011, 475(7356): 353–358

19 Ryan D P, Ptacek L J. Episodic neurological channelopathies. Neuron, 2010, 68(2): 282–292

20 Jentsch T J. Neuronal KCNQ potassium channels: physiology and role in disease. Nature Reviews. Neuroscience, 2000, 1(1): 21–30

21 Sanguinetti M C, Tristani-Firouzi M. hERG potassium channels and cardiac arrhythmia. Nature, 2006, 440(7083): 463–469

22 Russell J M. Sodium-potassium-chloride cotransport. Physiological Reviews, 2000, 80(1): 211–276

23 Simon D B, Karet F E, Hamdan J M, Pietro A D, Sanjad S A, Lifton R P. Bartter’s syndrome, hypokalaemic alkalosis with hypercalciuria, is caused by mutations in the Na-K-2Cl cotransporter NKCC2. Nature Genetics, 1996, 13(2): 183–188

24 Tong C C, Quesada R, Sessler J L, Gale P A. Meso-Octamethylcalix[4]pyrrole: an old yet new transmembrane ion-pair transporter. Chemical Communications, 2008, (47): 6321–6323

25 Fisher M G, Gale P A, Hiscock J R, Hursthouse M B, Light M E, Schmidtchen F P, Tong C C. 1,2,3-Triazole-strapped calix[4]pyrrole: a new membrane transporter for chloride. Chemical Communications, 2009, 21(21): 3017–3019

26 Koulov A V, Mahoney J M, Smith B D. Facilitated transport of sodium or potassium chloride across vesicle membranes using a ditopic salt-binding macrobicycle. Organic & Biomolecular Chemistry, 2003, 1(1): 27–29

27 Lee J H, Lee J H, Choi Y R, Kang P, Choi M G, Jeong K S. Synthetic K⁺/Cl^‒-selective symporter across a phospholipid membrane. Journal of Organic Chemistry, 2014, 79(14): 6403–6409

28 Yu X H, Cai X J, Hong X Q, Tam K Y, Zhang K, Chen W H. Synthesis and biological evaluation of aza-crown ether-squaramide conjugates as anion/cation symporters. Future Medicinal Chemistry, 2019, 11(10): 1091–1106

29 Sun Z, Barboiu M, Legrand Y M, Petit E, Rotaru A. Highly selective artificial cholesteryl crown ether K⁺-channels. Angewandte Chemie International Edition, 2015, 54(48): 14473–14477

30 Gilles A, Barboiu M. Highly selective artificial K⁺ channels: an example of selectivity-induced transmembrane potential. Journal of the American Chemical Society, 2016, 138(1): 426–432

31 Li Y H, Zheng S, Legrand Y M, Gilles A, van der Lee A, Barboiu M. Structure-driven selection of adaptive transmembrane Na⁺ carriers or K⁺ channels. Angewandte Chemie International Edition, 2018, 57(33): 10520–10524

32 Chen S, Wang Y, Nie T, Bao C, Wang C, Xu T, Lin Q, Qu D H, Gong X, Yang Y, Zhu L, Tian H. An artificial molecular shuttle operates in lipid bilayers for ion transport. Journal of the American Chemical Society, 2018, 140(51): 17992–17998

33 Wu F Y, Li Z, Guo L, Wang X, Lin M H, Zhao Y F, Jiang Y B. A unique NH-spacer for N-benzamidothiourea based anion sensors. Substituent effect on anion sensing of the ICT dual fluorescent N-(p-dimethylaminobenzamido)-N′-arylthioureas. Organic & Biomolecular Chemistry, 2006, 4(4): 624–630

34 Li A F, Wang J H, Wang F, Jiang Y B. Anion complexation and sensing using modified urea and thiourea-based receptors. Chemical Society Reviews, 2010, 39(10): 3729–3745

35 Villa M, Bergamini G, Ceroni P, Baroncini M. Photocontrolled self-assembly of azobenzene nanocontainers in water: light-triggered uptake and release of lipophilic molecules. Chemical Communications, 2019, 55(79): 11860–11863

36 Du Z, Ren B, Chang X, Dong R, Peng J, Tong Z. Aggregation and rheology of an azobenzene-functionalized hydrophobically modified ethoxylated urethane in aqueous solution. Macromolecules, 2016, 49(13): 4978–4988

37 Otis F, Racine-Berthiaume C, Voyer N. How far can a sodium ion travel within a lipid bilayer? Journal of the American Chemical Society, 2011, 133(17): 6481–6483

38 Yang Y, Wu X, Busschaert N, Furuta H, Gale P A. Dissecting the chloride-nitrate anion transport assay. Chemical Communications, 2017, 53(66): 9230–9233

39 Vargas Jentzsch A, Emery D, Mareda J, Metrangolo P, Resnati G, Matile S. Ditopic ion transport systems: anion-π interactions and halogen bonds at work. Angewandte Chemie International Edition, 2011, 50(49): 11675–11678

40 Busschaert N, Wenzel M, Light M E, Iglesias-Hernandez P, Perez-Tomas R, Gale P A. Structure-activity relationships in tripodal transmembrane anion transporters: the effect of fluorination. Journal of the American Chemical Society, 2011, 133(35): 14136–14148

41 Valkenier H, Haynes C J E, Herniman J, Gale P A, Davis A P. Lipophilic balance—a new design principle for transmembrane anion carriers. Chemical Science (Cambridge), 2014, 5(3): 1128–1134

42 Ren C, Shen J, Zeng H. Combinatorial evolution of fast-conducting highly selective K⁺-channels via modularly tunable directional assembly of crown ethers. Journal of the American Chemical Society, 2017, 139(36): 12338–12341

43 Ren C, Chen F, Ye R, Ong Y S, Lu H, Lee S S, Ying J Y, Zeng H. Molecular swings as highly active ion transporters. Angewandte Chemie International Edition, 2019, 58(24): 8034–8038

44 Ye R, Ren C, Shen J, Li N, Chen F, Roy A, Zeng H. Molecular ion fishers as highly active and exceptionally selective K⁺ transporters. Journal of the American Chemical Society, 2019, 141(25): 9788–9792

45 Liu T, Bao C, Wang H, Lin Y, Jia H, Zhu L. Light-controlled ion channels formed by amphiphilic small molecules regulate ion conduction via cis-trans photoisomerization. Chemical Communications, 2013, 49(87): 10311–10313

46 Sun Z, Gilles A, Kocsis I, Legrand Y M, Petit E, Barboiu M. Squalyl crown ether self-assembled conjugates: an example of highly selective artificial K⁺ channels. Chemistry (Weinheim an der Bergstrasse, Germany), 2016, 22(6): 2158–2164

47 Schneider S, Licsandru E D, Kocsis I, Gilles A, Dumitru F, Moulin E, Tan J, Lehn J M, Giuseppone N, Barboiu M. Columnar self-assemblies of triarylamines as scaffolds for artificial biomimetic channels for ion and for water transport. Journal of the American Chemical Society, 2017, 139(10): 3721–3727

48 Wu X, Small J R, Cataldo A, Withecombe A M, Turner P, Gale P A. Voltage-switchable HCl transport enabled by lipid headgroup-transporter interactions. Angewandte Chemie International Edition, 2019, 58(42): 15142–15147

49 Wu X, Busschaert N, Wells N J, Jiang Y B, Gale P A. Dynamic covalent transport of amino acids across lipid bilayers. Journal of the American Chemical Society, 2015, 137(4): 1476–1484

50 Zheng S P, Huang L B, Sun Z, Barboiu M. Self-assembled artificial ion-channels toward natural selection of functions. Angewandte Chemie International Edition, 2021, 60(2): 566–597