Molecular Tetrahedrons as Selective and Efficient Ion Transporters via a Two-Station Swing-Relay Mechanism

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE1 Aug 2021Molecular Tetrahedrons as Selective and Efficient Ion Transporters via a Two-Station Swing-Relay Mechanism Ning Li, Jie Shen, Gerome Kusuma Ang, Ruijuan Ye and Huaqiang Zeng Ning Li The NanoBio Lab, The Nanos, Singapore 138669 Google Scholar More articles by this author , Jie Shen The NanoBio Lab, The Nanos, Singapore 138669 Google Scholar More articles by this author , Gerome Kusuma Ang Raffles Institution, Singapore 575954 Google Scholar More articles by this author , Ruijuan Ye Institute of Advanced Synthesis, Northwestern Polytechnical University, Xi’an, Shaanxi 710072 Yangtze River Delta Research Institute, Northwestern Polytechnical University, Taicang, Jiangsu 215400 Google Scholar More articles by this author and Huaqiang Zeng *Corresponding author: E-mail Address: [email protected] Institute of Advanced Synthesis, Northwestern Polytechnical University, Xi’an, Shaanxi 710072 Yangtze River Delta Research Institute, Northwestern Polytechnical University, Taicang, Jiangsu 215400 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.020.202000475 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail The traditional approach to utilizing an ion-relay mechanism for ion transport requires three or more ion-relay stations. Herein, we describe a novel strategy, incorporating a swing action to realize a minimal ion-relay mechanism via only two relay stations. This swing-relay mechanism was achieved using a class of crown ether-appended, long-armed molecular tetrahedrons ( MTs). These MTs comprise ion-relaying crown units attached to a rigid tetrahedral core via flexible alkyl linkers, which act as the mobile arms and endow the crown units with great mobility to swing. Driven by the ionic concentration gradient and supported by the mobile arms, two crown units located in the membrane’s opposite hydrophilic regions swing toward each other to complete a single ion-relay step in the center of the membrane to enable fast ion conduction across. Generally, 18-crown-6-containing MT6s exhibited higher activities and better K+/Na+ selectivities than 15-crown-5-containing MT5s, with the most active MT6-C10 displaying an EC50 (K+) of 0.9 μM (i.e., 0.9 mol % relative to lipid) and the most selective MT6-C4 showing K+/Na+ selectively of 6.3. Finally, five control MTs were rationally designed to establish the action of the novel unimolecular two-station swing-relay as an efficient yet unprecedented mechanism of synthetic ion transporters. Download figure Download PowerPoint Introduction Precisely regulated ion flow across the cell membrane plays a crucial role in numerous biological processes, and thus, the perturbation of this process is closely associated with many physiological disorders and subsequent human diseases.1,2 Understanding how the naturally evolved ion transporters function and designing artificial versions of comparable or even exceeding performance have attracted extensive worldwide research interest due to their scientific significance and great potential in the biomedical marketplace.3–8 Passive transmembrane transport of ions is mainly accomplished by channel- or carrier-mediated mechanisms, in which ions pass through the membrane via a preformed passage spanning across or via reversible binding with ionophores mobilized in the lipid bilayer.9–25 The cross-membrane ion transporters, which do not use a conventional channel or carrier mechanisms, are an exciting class of artificial membrane transporters and to date remain largely unexplored, with only a few reports available in the open literature.26–29 In 2018, Chen et al.27 reported a rotaxane-derived molecular shuttle for K+ ion transportation, with an EC50 (K+) value equivalent to 3 mol % of lipid present. Featuring the molecular machine concept with much simpler molecular designs, two alternative highly active synthetic ion transporters were also demonstrated very recently, namely molecular swing28 and molecular ion fisher29 that facilitate ion transport by swinging and fishing actions, respectively. Although both molecular swings and ion fishers share certain structural similarities, they differ vastly in their ion transport activities and selectivities, foreshadowing the versatile potential of structurally fine-tuned swings and fishers in the field of synthetic ion transporters. A conventional ion-relay mechanism for ion transport requires three or more relay stations to fully span the hydrophobic membrane region, with at least one station residing around the center of the membrane where ions experience the highest energetic penalty.18,30–37 During the ion-relay process, where the preceding station captures and relays the ion to the next station, all stations do not move significantly. We envisioned that the number of relay stations could be reduced to two if they are customized to be capable of swinging along the membrane axis back and forth between the membrane’s hydrophilic region and its center.26 In this work, we disclose a unique class of crown ether-appended tetrahedron-shaped ion transporters called molecular tetrahedrons ( MTs) (Figures 1a–1c), which are indeed appropriately sized to enable highly efficient ion transport via a swing-relay mechanism minimally mediated by just two ion-relay stations and a single ion-relay step (Figure 1d). The modularly designed nanosized MTs consisted of a rigid tetrahedral core, flexible alkyl linkers of appropriate lengths, and ion-relaying crown ether units at the tetrahedron vertices. With the tetrahedron core serving as the anchor and the flexible linkers as the rope (Figures 1b and 1c), the peripheral crown ether-based relay stations become capable of swinging, relaying, and eventually conducting ions across lipid membranes (Figure 1d). Such a unique molecular design, incorporating three-dimensional (3D)-shaped mobile arms provides a novel approach and a new ion-relay principle to construct highly efficient ion transporters toward interesting applications relative to membrane transport. Figure 1 | (a) General chemical structure of the MTs. (b) Computationally optimized structure of MT6-C10 at the ωB97X/6-31G* level and its physical dimension from the CPK model. (c) Side-view snapshot of MD-simulated configuration of MT6-C10 in lipid bilayer membrane surrounded by an aqueous environment. (d) Schematic illustration of the proposed two-station swing-relay mechanism involving a head-to-head meet of two crown units for ion relay in the center of the membrane. MTs, molecular tetrahedron; CPK, Corey–Pauling–Koltun; MD, molecular dynamics. Download figure Download PowerPoint Experimental Methods Synthesis of MTs In this study, MTs were prepared via a dual-step synthetic procedure, exemplified by the preparation of MT6-C10 in the following discussion. First, a suspension of 1,10-dibromodecane (3.00 g, 10 mmol), 4′-carboxybenzo-18-crown-6 (356 mg, 1.0 mmol), and potassium carbonate (276 mg, 2.0 mmol) in acetonitrile (20 mL) was heated at 85 °C overnight. Removal of the solvent in vacuo gave the crude product, purified by flash column chromatography using ethyl acetate to afford the pure tetrahedron arm product purified by preparative thin layer chromatography (silica gel layer, 20 cm × 20 cm (L × W), 1 mm in gel thickness).29 Then a suspension of the prepared tetrahedron arm (201 mg, 0.34 mmol), potassium carbonate (96.7 mg, 0.70 mmol), and a 4,4′,4″,4‴-methanetetrayltetrabenzoic acid (34.7 mg, 0.070 mmol) was dissolved in anhydrous dimethylformamide (10 mL) and heated to 100 °C overnight. Next the reaction mixture was filtered to obtain a clear orange solution. Removal of the solvent in vacuo gave the crude product, which was purified by preparative thin-layer chromatography using methanol/dichloromethane (8/100) to afford a pure pale yellow solid product of MT6-C10. All other MTs were synthesized using similar procedures. Synthesis and characterization details are found in Supporting Information Figure. Molecular dynamic simulation A membrane builder in CHARMM-GUI was used to build the initial protein/membrane complex structure. The protocol comprised six steps, sequentially performed in the following order: objects reading, objects orientation, system size determination, building a lipid bilayer, assembling lipid bilayer, and system equilibrium. The MT6-C10 molecule was placed in the center of the membrane made up of 256 phosphatidylcholine (POPC) molecules, with the removal of lipid molecules within 1 Å from the MT6-C10 molecules. Then the membrane was placed in a box with dimensions of 93 × 93 Å in width and 99 Å in height. Subsequently, 15,061 water molecules were placed on the top and bottom sides of the membrane (~ 7500 on each side) for each system. Counter KCl ions were added to produce an ion concentration of 0.15 M. During molecular dynamics (MD) simulations, the pressure was maintained at 1 bar. After equilibration steps, the simulation’s production run was performed for 100 ns, and the structure at the 100th ns trajectory was used for analyzing the orientation of the MT6-C10 molecule. Further details are presented in Supporting Information Figure. 8-Hydroxypyrene-1,3,6-trisulfonic acid assay to evaluate ion transport performance Egg yolk l-α-phosphatidylcholine (EYPC; 1.0 mL, 25 mg/mL in CHCl3; Avanti Polar Lipids, Alabaster, AL) was loaded in a round-bottom flask, and the solvent was removed under reduced pressure at 30 °C. The resultant film was dried under high vacuum overnight at room temperature and hydrated with 4-(2-hydroxyethyl)-1-piperazine-ethane sulfonic acid (HEPES) buffer solution (1.0 mL, 10 mM HEPES, 100 mM NaCl, pH = 7.0) containing a pH-sensitive dye 8-hydroxypyrene-1,3,6-trisulfonic acid (HPTS; 1.0 mM) at room temperature for 45 min to give a milky suspension. Then the mixture was subjected to 10 freeze–thaw cycles: freezing in liquid N2 for 1 min and heating at 55 °C in a water bath for 2 min. The vesicle suspension was extruded through polycarbonate membrane (0.1 μm) to produce a homogeneous suspension of large unilamellar vesicles (LUVs) of ∼120 nm in diameter with HPTS encapsulated inside. The extravesicular HPTS dye was separated from the LUVs using size-exclusion chromatography (stationary phase: Sephadex G-50, GE Healthcare, Smyrna, GA; mobile phase: 10 mM HEPES buffer with 100 mM NaCl). The mobile phase was diluted to yield an LUV stock solution of 5.0 mL of 6.5 mM and stored at a 4 °C refrigerator until use. The HPTS-containing LUV suspension (30 μL, 6.5 mM in 10 mM HEPES buffer containing 100 mM NaCl at pH = 7.0) was added to a HEPES buffer solution (1.96 mL, 10 mM HEPES, 100 mM MCl at pH = 8.0, where M+ = Li+, Na+, K+, Rb+, and Cs+) to create a pH gradient for ion transport study. Then a solution of MTs or control compounds in dimethyl sulfoxide (DMSO) was injected into the suspension under gentle stirring. Upon addition, the emission of HPTS was monitored immediately at 510 nm with excitations at both 460 and 403 nm, recorded simultaneously for 300 s using a fluorescence spectrophotometer (Model F-7100; Hitachi, Tokyo, Japan). At t = 300 s, an aqueous solution of Triton X-100 (20 μL, 20% v/v) was injected to induce a maximum change in dye fluorescence emission. The final transport trace was obtained as a ratiometric value of I460/I403 and normalized based on the ratiometric value of I460/I403 after the addition of Triton using the equation If = [(It − I0)/(I1 − I0)], where If = fractional emission intensity, It = fluorescence intensity at time t, I1 = fluorescence intensity after the addition of Triton X-100, and I0 = initial fluorescence intensity. The fractional change RM+ was calculated for each curve using the normalized value of I460/I403 at 300 s before the addition of Triton, with the ratio of a blank set as 0 and that of Triton as 1. By fitting the fractional transmembrane activity, RM+, versus transporter concentration using the Hill equation: Y = 1/(1 + (EC50/[C])n), we obtained the EC50 values and Hill coefficient n. CF dye leakage assay LUVs containing 5(6)-carboxyfluorescein dyes were prepared via similar procedures with that in the HPTS assay, except for the hydration step, wherein the lipid film was hydrated with HEPES buffer solution (1.0 mL, 10 mM HEPES, 100 mM NaCl, pH = 7.5) containing 5(6)-fluorescein (CF; 50 mM) at room temperature for 45 min. Details are presented in Supporting Information Figure S24. Then the CF-containing LUV suspended in LUV buffer (30 μL, 6.5 mM in 10 mM HEPES buffer containing 100 mM NaCl at pH = 7.5) was added to a HEPES buffer solution (1.96 mL, 10 mM HEPES, 100 mM NaCl at pH = 7.5) to create a concentration gradient for the CF dye transport study. A solution of MT6-C10 (3.5 μM) or natural pore-forming peptide melittin in DMSO at different concentrations was then injected into the suspension under gentle stirring. Upon adding MTs or pore-forming peptide molecules, the emission of the CF dye was monitored immediately at 517 nm with excitations at 492 nm for 300 s. At t = 300 s, an aqueous solution of Triton X-100 (20 μL, 20% v/v) was added immediately to destroy the CF dye gradient completely. Results and Discussion Molecular design and synthesis In designing MT ion transporters, a few criteria were borne in mind, including (1) specific functional groups for selective ion capturing, (2) the movable and lipid-compatible moieties for ion transportation, and (3) appropriate physical dimension that would span across the entire membrane lipid bilayer. In the literature, crown ethers have been widely reported on their binding affinity to alkali metal ions,38,39 and have frequently been utilized in constructing ion transporters,18,19,33–37 thereby representing a good option as the selective ion-capturing group. We accomplished the synthesis of MTs by reacting 4-carboxybenzo-crown ether with dibromoalkane of appropriate lengths to construct the tetrahedron arm first, followed by reacting the arm with a 4,4′,4″,4‴-methanetetrayltetrabenzoic acid as the tetrahedral core to establish the desired tetrahedron configuration (Figure 1a and Supporting Information Scheme S1). The flexible alkyl chain was selected primarily because of its structural simplicity and flexibility, as well as its excellent compatibility with the hydrophobic lipid tails in the membrane. Via this facile dual-step synthetic scheme, six-related MTs were modularly prepared using three types of alkyl linkers (n-C8H16, n-C10H20, and n-C12H24) and two classes of crown units (18-crown-6 and 15-crown-5; Figure 1a). These compounds were characterized by 1H and 13C NMR spectra and high-resolution mass spectra (see Supporting Information Figure S25–S42 for more details). As an example of the MT family, the computationally optimized molecular structure of MT6-C10, containing 18-crown-6 units and n-C10H20 linkers, agreed with the expected tetrahedron configuration (Figure 1b). The distance between the opposite tetrahedron edges, excluding the crown units was calculated to be 35 Å at its most extended state, according to the Corey–Pauling–Koltun (CPK) model. The distances were 32 and 38 Å for MT6-C8 and MT6-C12, respectively. All these values were comparable with the typical thickness of the membrane hydrophobic region, hence, confirming the appropriate physical dimension of these MTs to span across the bilayer membrane. To further validate the molecular design, MD simulation of an all-atom MT6-C10 model in the lipid bilayer, performed for 100 ns in an aqueous environment. The stabilized snapshot revealed a favorable configuration of MT6-C10 sandwiched in the membrane, with two crown ethers at the top membrane–water interface and the other two at the bottom (Figure 1c). Considering the alkyl linker’s structural flexibility and crown ether’s ion binding affinity, these rationally designed MTs might perform as efficient ion transporters to move cations across the membrane in the presence of an ionic concentration gradient. A hypothetical swing-relay mechanism for ion transport In view of the unique tetrahedral molecular design, we hypothesized that the ion transport is dominated by an unconventional swing-relay mechanism that employs two of the four crown units, one from the top and one from bottom side, to transport a cation across the membrane. The step-by-step illustration of the proposed swing-relay action is shown in Figure 1d. The critical step involves a head-to-head meet of two ion-binding crown units at the central membrane line, where the highest energetic penalty for cations resides. Starting from the MD-simulated configuration as the initial state, one of the crown ethers at the top membrane–water interface first captures an ion (Step 1, Figure 1d). Driven by the ionic concentration gradient, this ion-carrying arm swings toward the membrane center (Step 2, Figure 1d). When it meets another arm that also swings stochastically toward the membrane center, ion-relay takes place between the two vicinal crown ethers (Step 3, Figure 1d). The ionic gradient further drives the new ion-carrying arm to swing back to its more extended state (Step 4, Figure 1d), followed by ion release (Step 5, Figure 1d). Subsequently, the system returns to the initial state to start the next swing-relay cycle. In this process, each of the two operational arms, on average, must only carry the ion across no more than half of the hydrophobic membrane region. Therefore, it should be of much higher efficiency than the other possibility of nonrelaying swing-only transport, in which a single-arm carries the ion across the entire membrane width, as described later with MT6-Cr4 (Figure 2). It is also worth emphasizing that both crown units at the top membrane–water interface are active for ion capturing, and they can convey their captured ions to either of the crown ethers at the bottom. In other words, there exist four equivalent active ion-relay pathways. Figure 2 | Molecular design of the control compounds MT6-Cr1 and MT6-Cr2 with rigid linkers linearly locked by intramolecular H-bonding, MT6-Cr3 that only contains two operational mobile arms, MT6-Cr4 featuring three indolyl-containing arms in green as lipid anchors and only a single mobile 18-crown-6-containing arm, and MT6-Cr5 with two indolyl-containing lipid anchors and two mobile 18-crown-6-containing arms. Note that in terms of physical dimensions, MT6-Cr1 is almost identical to MT6-C6, and MT6-Cr2 to MT6-Cr5 are almost identical to MT6-C10. Download figure Download PowerPoint To validate the swing-relay proposition, specific control MT compounds were designed and prepared, mainly with a focus on (1) verifying the indispensable role of structural flexibility of MT linkers, (2) confirming the synergistic effect among the mobile tetrahedron arms, and (3) probing the role played by the alternative nonrelaying swing-only in ion conduction (see Figure 2 and later discussions). High ion transport activity by MT6s Ion transport activities of these MTs were evaluated experimentally using the well-established HPTS assay constructed upon LUVs (Figure 3a), with an extravesicular environment of 100 mM MCl (M+ = Li+, Na+, K+, Rb+, and Cs+) at pH 8.0. The intravesicular region contains 100 mM NaCl and 1.0 mM pH-sensitive HPTS dye at pH 7.0. Driven by the pH gradient, cross-membrane ion transport is always coupled with an intravesicular pH change that can be monitored macroscopically by the HPTS fluorescence intensity. Figure 3 | (a) Schematic illustration of the pH-sensitive HPTS assay for the ion transport study. Different extravesicular salts are used to facilitate the comparison of ion transport activities. (b) Fractional K+ transport activities of various MTs determined over 300 s at 3.5 μM. The normalized fluorescence intensity RK+ is calculated using the formula RK+ = (IK+ − I0)/(ITriton − I0), in which IK+ and ITriton are the values of the I460 nm/I403 nm ratio at 300 s before and after the triton addition, respectively, and I0 is the background intensity. (c) M+ ion transport selectivity of the most active K+-transporting MT6-C10, determined > 300 s at 3.5 μM with an extravesicular environment of 100 mM MCl (M+ = Li+, Na+, K+, Rb+, and Cs+). [Total lipid] = 97.5 μM. HPTS, 8-hydroxypyrene-1,3,6-trisulfonic acid; MTs, molecular tetrahedron. Download figure Download PowerPoint Using the HPTS assay, we found that MT6s transport K+ ions much faster than other alkali metal ions, with MT6-C10 exhibiting the highest fractional K+ transport activity of 100% at 3.5 μM (Figure 3b). The ion selectivity follows the order K+ > Rb+ > Na+ > Cs+ > Li+ (Figure 3c and Supporting Information Figure S1), which agrees with the Eisenman sequence IV.40 This suggests that the observed selectivity could be ascribed to ion recognition by the crown units instead of ion dehydration energy or ionic radius. Interestingly, most MT5s transport Na+ faster than the other cations, following the order Na+ ≥ K+ > Rb+ > Li+ > Cs+, and the best performer, MT5-C12, exhibiting a fractional activity of 85.2% at 40 μM (see Supporting Information Figure S2). Then we performed Hill analyses to determine the EC50 values of these MTs in transporting K+ and Na+ ions, which are most relevant to physiological settings (see Supporting Information Figures S3–S8). The ion transport selectivity can also be quantified by simply taking the reciprocal ratio of the EC50 values. In general, MT6s exhibited both high activity (EC50 = 0.9–1.6 μM for K+, corresponding to 0.9–1.6 mol % relative to lipid) and likely excellent K+/Na+ selectivity [EC50(Na+)/EC50(K+) = 7.1–15.3] (Table 1). However, this EC50-based selectivity might be inaccurate in situations where the transporters are poorly soluble or interact unfavorably with the membrane. In this regard, we recently proposed the use of RK+/RNa+ ratio (see Supporting Information for details) as an alternative yet reliable index for evaluation of ion transport selectivity (Figure 3c and Table 1), provided that the normalized fractional fluorescence intensity RM+ is obtained at the transporter concentration where the transport activity for the most active ion reaches > 90% over the 300 s data recording period.29 Recently, the reliability of using such RM+ ratios to derive ion transport selectivity has also been verified by Zeng et al.41 Using this index, all MT6s were still found to be K+ selective, with the K+/Na+ selectivity falling in the range of 2.5–3.3 (Table 1). In comparison, MT5-C12, being the most active in the Na+-selective MT5s family, showed an EC50(Na+) value of 8.8 μM and a Na+/K+ selectivity of 2.2. Table 1 | Experimentally Determined EC50 (μM), EC50 (Na+)/EC50 (K+), and RK+/RNa+ Values for MTs MT5-C12 MT5-C10 MT5-C8 MT6-C12 MT6-C10 MT6-C8 MT6-C6 MT6-C4 MT6-C2c EC50 (Na+)a 8.8 ± 0.3 23.5 ± 2.2 40.1 ± 1.7 11.3 ± 0.3 13.1 ± 0.4 21.4 ± 0.3 45.4 ± 1.3 >32 — EC50 (K+)a >40 36.6 ± 2.5 37.3 ± 0.8 1.6 ± 0.1 0.9 ± 0.1 1.4 ± 0.1 9.0 ± 0.8 1.8 ± 0.1 3.8 ± 0.2 EC 50 ( Na + ) EC 50 ( K + ) <0.2 0.6 1.1 7.1 14.6 15.3 5.0 >17.8 — RK+/RNa+b 0.4 0.7 1.1 2.5 3.1 3.3 1.9 6.3 — Note: HPTS, 8-hydroxypyrene-1,3,6-trisulfonic acid. Bold values refer to the most active and most selective molecular tetrahedrons. a[Total lipid] = 97.5 μM. bSee Supporting Information Figure S1–S11 for more details. c MT6-C2 shows a peculiar Na+-transport curve in HPTS assay; its activity cannot be confidently determined. We elucidated the additional impact of the alkyl linker length toward ion transport properties of the better performing MT6s series by synthesizing three additional MT6s with shorter linkers (i.e., MT6-C6, MT6-C4, and MT6-C2) and assessed their ion transport performance using the same pH-sensitive HPTS assay. While all of them demonstrated K+ selectivity and generally highly active, their ion transport activities were inferior to that of MT6-C10 and did not monotonously decrease with alkyl linker length (Table 1). Given the sandwich-like configuration in the lipid bilayer (Figure 1c), some MT6s of intermediate size might have possessed the most appropriate physical dimension to have their crown units most stably docked at the membrane–water interfaces to attenuate the ability to swing. This would make the arm curling energetically less favorable, compared with other longer or shorter ones, and hence, largely sabotage their ion transport ability. Experimentally, we identified one such special case to be MT6-C6, which exhibited a much lower K+ transport activity [EC50(K+) = 9.0 μM] than the other five MT6s [EC50 (K+) = 0.9–3.8 μM, Table 1]. The alkyl linker length also imposed a significant influence on the ion transport selectivity. Using the aforementioned RM+ ratio-based approach, a good K+/Na+ selectivity of 6.3 was determined for MT6-C4, which stood out from the rest of MT6 members (RK+/RNa+ = 1.9–3.3). The K+/Na+ selectivity of this most selective MT6-C4 and also the most active MT6-C10 in the MT6s family was also verified qualitatively by the membrane potential experiments with Safranin O (see Supporting Information Figure S12 for details). Similar linker length dependency of ion transport activity and selectivity was also observed for the MT5s, although their overall performances were poor (Table 1). Transport mechanism study Further, we prepared a group of structurally related yet functionally specific control MTs to show that it is the swing-relay action of the flexible arms that accounts for highly efficient ion transport performance. First, since the ion-relaying action depends largely on the arm curling/swinging to facilitate the head-to-head meet of crown units, we envisioned that similar-sized MTs with rigid arms, which can only undergo minimal movement via limited covalent bond rotation in the membrane, are incapable of transporting K+ or Na+ ions, regardless of their arm lengths. To this end, we prepared two control MTs, namely MT6-Cr1 and MT6-Cr2, by replacing the flexible alkyl linkers with rigid ones, which were locked linearly by intramolecular hydrogen bonding (Figure 2). These controls possess almost identical physical dimensions to those of MT6-C6 and MT6-C10 and could readily span across the lipid membrane, but as expected, it turned out to elicit negligible Na+ or K+ transport activities experimentally, even at concentrations as high as 10 μM (see Supporting Information Figure S13). The lack of detectable activities of MT6-Cr1 and MT6-Cr2 pointed further to a limited contribution to ion transport by simple molecular rotation, much like a windmill. It is worth mentioning that only MT6-Cr2 with a rigid linker showed visible precipitates by naked eyes among all MTs and the control compounds tested at 10 μM, with the UV–vis measurement revealing at least 50% MT6-Cr2 incorporation into LUVs at 10 μM (see Supporting Information Figure S14). This further suggests that >50% of other soluble tetrahedrons might have been incorporated into LUVs. In situ dynamic light scattering (DLS) measurements were carried out to monitor the formation of precipitates of the rigidly armed MT6-Cr1 and MT6-Cr2 at varying concentrations in the LUV buffer. As shown in Supporting Information Figure S15, the addition of MT6-Cr1 into the LUV buffer with concentrations up to 10 μM did not produce any precipitates, indicating that all of the compounds penetrated the LUV membrane. In contrast, MT6-Cr2 at a high concentration of 10 μM showed extra DLS peaks at ∼600 and 1000 nm, but not at lower concentrations. This result agrees well with the naked eye observation, showing no visible precipitates at up to 7.5 μM. Thus, undoubtedly, MT6-Cr1 and MT6-Cr2 could insert readily into the LUV membrane lipid bilayer at concentrations up to 7.5 μM but unable to transport the alkali metal ions across the membrane. Such inactivity also confirms the limited contribution (if any) from the possible molecular tumbling of the MTs in the lipid bilayer to convey ions across. Second, the swing-relay ion transport mechanism should also dictate some synergistic effects among the four flexible arms. We verified this effect by fabricating another membrane-span control compound, MT6-Cr3, which only possessed two operational arms, with the other two arms terminated merely by a methyl group. We compared MT6-Cr3 with MT6-C10 that had all four arms (Figure 2). Using the Hill analysis, the EC50(K+) value of MT6-Cr3 was determined to be 4.6 μM (Figure 2 a