Ion transport across bilayer lipid membranes between two aqueous phases in the presence of iodide and triiodide ions

Abstract

I⁻ and I₃⁻ usually coexist in nature, and it is well-known that I₃⁻ is much more hydrophobic than I⁻. Both I⁻ and I₃⁻ play a crucial role in human physiological activities and can be applied to various medical applications, such as synthesis of medicine, antibiotics, etc. During the measurement of the ion-transport current of KI aqueous solution, I₃⁻ is spontaneously generated and causes an increase in ion permeation. However, the mechanism of facilitated ion transport remains unclear. In this study, the influence of I₃⁻ on the ion transport across bilayer lipid membranes (BLMs) was elucidated. Physically stabilized BLMs were formed using the track-etched membrane (TM), and the ion-transport current was measured by applying a membrane potential across BLMs. Under asymmetric ionic concentration conditions, the permeability of K⁺, I⁻, and I₃⁻ was evaluated. The permeability of I⁻ across BLMs was about 8 times higher than that of K⁺. In the presence of I₃⁻, the permeability of K⁺ across BLMs drastically increased. The permeability of K⁺ became 9 times higher than that of I⁻ in the presence of 50 µM I₃⁻. It is considered that I₃⁻ facilitated the transport of K⁺ across BLMs by serving as a carrier of K⁺ within BLMs.

---

Introduction

Ion transport across cell membranes constructs the cell-to-cell communication network and contributes to various physiological activities, such as neurotransmission, respiration, photosynthesis, metabolism, muscle elasticity, and mass transport.^1–4 The ion transport mechanism across cell membranes is mainly classified into three types: transport through the penetration holes of channels and pumps, free diffusion of ion carriers within cell membranes, and direct transport of hydrophobic ions within cell membranes.^5–8 Therefore, the study of electrophysiological signal interaction is significant, and a bilayer lipid membrane (BLM) is one of the popular base materials to proceed with experiments. So far, BLMs have been widely applied to evaluate the physicochemical properties of reconstituted biofunctional compounds such as channel/pump proteins and membrane-bound enzymes.^9–14 In particular, electrochemical measurements of the ion transport across artificial BLMs containing a target compound have been frequently conducted to date. To avoid the influence of other coexisting biological substances on the membrane transport, researchers use artificial BLMs and focus on the specific ion transport due to the compound itself.

Although BLMs are convenient materials for studying electrophysiology, the relatively large area of BLMs shows their physical fragility which limits their application.^15,16 Specifically, external stimuli such as physical shocks and electrical fluctuation can damage BLMs and cause unpredictable influence. To increase the strength of BLMs, most research groups have improved the base materials such as surface materials, miniaturized holes, sharpened the edges of holes, etc.^17–21 The author's research group has developed a simple method to form BLMs within the penetration holes of a track-etched membrane (TM), which is a kind of porous filter membrane.²² When the BLMs are formed within the penetration holes of the TM, it is possible to stabilize the BLMs against external stimuli and also increase the total membrane area.²²

Iodine is a crucial element in biological activities.^23–27 Iodine is an essential component to synthesize thyroid hormones. Iodine plays an important role in treating diseases, such as hyperthyroidism and Graves’ disease.²⁸ On the other hand, both excess and deficiency of iodine can lead to different physiological dysfunctions or diseases.^29–31 The povidone-iodine (PVP-I) solution is a popular chemical substance that is used for skin disinfection before or after operation.³² Incidentally, I⁻ and I₃⁻ usually coexist in the natural world, and iodine species such as I₂, I⁻, and I₃⁻ have been of interest as redox compounds.^33–35 Since I⁻ and I₃⁻ are much more hydrophobic than Cl⁻, it is thought that these ions play an important role in transporting ions and/or redox compounds within living bodies. It has been reported that I₃⁻ can form a complex with a halide anion and serve as a charge carrier.³⁶ Due to their hydrophobicity, the complexes are more soluble in the membrane interior.³⁷ The addition of I₃⁻ within the BLMs increased the ion-transport current.³⁸ On the other hand, the antiport of I⁻ and co-ions across the membrane has been reported.^39,40 In addition, when there is an appropriate redox couple in the cell system, it has been reported that the electron transport across the BLMs in the presence of I⁻ and I₃⁻ would be observed.^36–38 However, these charge-transport mechanisms remain unclear and lack direct evidence.

In this study, we aim to elucidate how I⁻ and I₃⁻ influence the ion transport across BLMs in the TM. In the presence of only KI, the antiport of K⁺ and I⁻ across the BLMs was elucidated. The influence of the addition of I₃⁻ was studied, and the mechanism of the facilitated ion transport by dilute I₃⁻ was clarified.

Experimental

Materials and chemicals

Lecithin (from egg, 99.5%), cholesterol (99.5%), n-decane (99.5%), potassium iodide (KI, 99.5%), sodium iodide (NaI, 99.5%), cesium iodide (CsI, 99.5%), potassium chloride (KCl, 99.5%), D(+)-glucose (98.0%), and ethanol (99.5%) were bought from Fujifilm Wako Pure Chemical Co. Calcium iodide (CaI₂, 99.5%) was bought from Kanto Chemical Co., Inc. All the chemicals purchased were of reagent grade and were used without further purification. A mixture of 100 mg of lecithin and 25 mg of cholesterol was dissolved in 5 mL of n-decane to prepare the BLM-forming solution. The BLM-forming solution was stored in the refrigerator at 4 °C until use and was disposed of after three days. The TM (1000M25/511P121/25) was bought from it4ip S.A. (Louvain-la-Neuve, Belgium). The thickness, aperture diameter, and aperture density of the TM were 14 µm, 12 µm, and 10⁵ cm⁻², respectively. All the aqueous solutions were prepared with high-purity water (ρ = 18.2 MΩ cm).

Instruments

Electrochemical measurements were conducted using a potentiostat/galvanostat (Hokudo Denko, HA1010mM1A), a function generator (Hokuto Denko, HB-305), and an A/D converter (Graphtec Corp., Midi Logger GL900). To maintain the stability of BLMs, all the electrochemical measurements were performed at 40 ± 2 °C using an incubator (IC101W, Yamato Scientific Co., Ltd).

Electrochemical measurements

A schematic diagram of a self-made electrochemical cell used to conduct the measurements is shown in Fig. 1. The cell was composed of two glass chambers with a hole (diameter: 8 mm∅) in the middle of the adhesive side. The TM was set between both the chambers after the adhesive sides were painted with vacuum grease (HIVAC-G, Shin-Etsu Chemical Co., Ltd). The BLMs in the TM were formed according to the method used by Chuang et al.²² First, TM was soaked completely in pure n-decane for 20 min. Then, the n-decane-impregnated TM was placed between the two glass chambers. After assembling the electrochemical cell, 40 µL of the BLM-forming solution was dropped on the TM. Next, the assembled cell was made to stand still for about 3 min to allow the lipid solution to penetrate the whole TM apertures. This dropping step was repeated three times in the BLM-formation process to ensure that all of the apertures are covered by the BLMs.

Fig. 1 Scheme of the self-made electrochemical cell. Two glass chambers were filled with the electrolyte solution and separated by a piece of TM, in which BLMs formed in its apertures.

Two self-made Ag|AgCl electrodes (Ag|AgCl|0.1 M KCl aq.) were used as the reference electrodes (RE1 and RE2) and separately put in W1 and W2. Likewise, two Pt wires were used as the counter electrodes (CE1 and CE2) and separately put in W1 and W2, respectively. A potential difference (E_W1–W2) was applied between RE1 and RE2, and the ion-transport current across the BLMs was recorded (I_W1–W2) between CE1 and CE2. Because the n-decane residue remains a little within the BLMs, we first scanned the E_W1−W2 between −200 and 200 mV at a potential scanning rate of 30 mV s⁻¹ for 10 min to remove the remaining n-decane from the BLMs. To prevent the formation of an osmotic pressure difference between W1 and W2, 0.5 M glucose was added to W1 and W2 in the case of asymmetric concentration systems.

The ion-transport current across the BLMs in TM

The I_W1–W2 values are composed of the charging current, I_C, and the ion-transport current, I_T, and I_W1–W2 can be expressed using the following equation:

I_W1–W2 = I_C + I_T. (1)

I_T is generated by the antiport of cations and anions across the BLMs.^22,41

Normal pulse voltammograms (NPVs) were measured by applying the potential pulse (E_W1−W2) and recording the I_W1–W2 value. The interval of the single potential step (t_int) was 60 s, composed of the initial background potential (0 V) for 20 s and the subsequent potential pulse (E_W1–W2) for 40 s, respectively. Within the single interval, the I_W1–W2 value was recorded at 58 s (38 s after the potential pulse): the time when the charging current was close to 0 (I_C ≈ 0) and the I_W1−W2 value was regarded as the steady-state ion-transport current. Therefore, the measured current was close to I_T (I_W1–W2 ≈ I_T).

Preparation of triiodide ion solution

A stock aqueous solution containing I₃⁻ was prepared according to the method reported by Jung et al.⁴² Although I₂ is usually insoluble in aqueous solution, soluble I₃⁻ is formed in the presence of abundant I⁻, as the following reaction states:

I_2(aq) + I⁻ ⇌ I₃⁻ (K_eq = 698).

The equilibrium constant (K_eq) of the reaction was reported to be around 698 at 25 °C and around 440 at 45 °C.^43,44 Therefore, in the presence of abundant I⁻, I₂ can be fully converted into I₃⁻, and the concentration of I₃⁻ depends on the concentration of I₂. In this study, the stock solution containing 0.1 M I₃⁻ was prepared by dissolving 0.1 M I₂ in the aqueous solution containing 1 M KI. Afterward, the stock solution was diluted in KI aqueous solution to prepare different concentrations of I₃⁻ (2, 5, 25, and 50 µM). The concentration of I₃⁻ was spectrophotometrically measured using a UV-Vis spectrophotometer (UV-1900i, Shimadzu Co.) at wavelengths of 288 and 352 nm to make sure that the method can fully convert I₂ into I₃⁻,^42,45 as shown in Fig. S1.

Results and discussion

Ion transport across BLMs between two aqueous phases containing KI and KI₃

The NPVs were measured to investigate the relationship between E_W1−W2 and I_W1−W2 in the ion transport across BLMs between W1 and W2 containing 0.1 M KI and M KI₃ under the steady-state conditions. Scheme 1 describes the cell composition as follows:

Scheme 1

Here, W1 and W2 are composed of 20 mL of 0.1 M KI aqueous solution in the presence of 0, 5, and 50 µM I₃⁻. The electrodes on the W1 side were set as the reference side.

Fig. 2(a) shows the NPVs for the ion transport across BLMs in the TM when both W1 and W2 contained only 0.1 M KI. The I_W1–W2 value increased with an increase in the applied E_W1–W2. The relationship between I_W1–W2 and E_W1–W2 was symmetrical about the origin (0 V, 0 A). When 200 mV was applied as E_W1–W2, the I_W1–W2 value was observed to be 0.2 ± 0.1 nA. In the presence of I₃⁻, the I_W1–W2 value drastically increased, as shown in Fig. 2(b). The I_W1–W2 values in the presence of 5 and 50 µM I₃⁻ at E_W1–W2 = 200 mV were 35 ± 2 and 70 ± 20 nA, respectively. In other words, after the addition of I₃⁻, the I_W1–W2 values increased about 150 and 300 times higher than those observed in the absence of I₃⁻, respectively. Although the concentrations of I₃⁻ were still lower than those of K⁺ and I⁻, the ion transport was dramatically facilitated by the addition of I₃⁻. The I_W1–W2 value was not proportional to the concentration of I₃⁻, which indicates that the I_W1–W2 value seems not to be caused by the transport of I₃⁻ itself. When NaI and CsI were used as electrolytes instead of KI, the ion transport phenomena exhibited exactly the same behaviour (Fig. S2 and S3). However, the facilitated transport by addition of I₃⁻ was not observed in the case of 0.1 M CaCl₂ (Fig. 3). This seems to be caused by the inhibition of I₃⁻ distribution due to the complex formation of Ca²⁺ with phospholipids.⁴⁶

Fig. 2 NPV for the ion transfer across the BLMs in the TM between W1 and W2 containing 0.1 M KI aqueous solution (a) in the absence of I₃⁻ (●) and (b) in the presence of 5 (○) and 50 (▲) µM I₃⁻.

Fig. 3 NPV for the ion transfer across the BLMs in the TM between W1 and W2 containing 0.1 M CaI₂ aqueous solution in the absence of I₃⁻ (●) and the presence of 50 µM I₃⁻ (○).

In the absence of I₃⁻, the magnitude of the ion-transport current does not depend on the temperature, as shown in Fig. S4. On the other hand, the magnitude of the ion-transport current increases with an increase in the temperature in the presence of I₃⁻.

Evaluation of the permeabilities of respective ions

The I_W1–W2–E_W1–W2 curve shown in Fig. 2 describes the steady-state ion transport across the BLMs within the TM apertures. Thus, both the distribution ratios of the transporting ions at the W1|BLM and BLM|W2 interfaces are estimated to be equal. In addition, the diffusion potential within the BLM (E_m) is expected to be very close to the E_W1–W2 value. Therefore, the relationship between E_m and I_W1–W2 under the steady-state conditions can be expressed using eqn (2).^47,48where T is the absolute temperature, d is the thickness of the BLM, z is the charge number of the transporting ion, F is the Faraday constant (96 485 C mol⁻¹), R is the ideal gas constant (8.314 J K⁻¹ mol⁻¹), D is the diffusion coefficient of the transporting ion within the BLM, and c is the concentration of the transporting ion of the surface region of the BLM (M1 and M2).

When W1 and W2 contain the same concentration of KI, the I_W1–W2–E_W1–W2 sigmoidal curve, as shown in Fig. 2, was symmetrical about the coordinate origin (0 V, 0 A). Conversely, when the concentration of W1 was different from that of W2, the I_W1–W2–E_W1–W2 curve shifted and became asymmetrical about the coordinate origin. For instance, Fig. 4(a) shows the I_W1–W2–E_W1–W2 curve when the KI concentration of W1 is 0.1 M and that of W2 is 0.001 M, namely, the concentration ratio of W2 to W1 (r, r = c_W2/c_W1) was 0.01. Notably, the E_W1–W2 value when the measured I_W1–W2 is zero, referred to as the zero-current potential (E_{W1–W2, i=0}), was the point where the I_W1–W2–E_W1–W2 curve intersected with the x-axis. E_{W1–W2, i=0} shifted in the negative direction compared to the case of r = 1. The negative shift was caused because the permeability of anion (I⁻) was higher than that of the cation (K⁺). Fig. 4(b) shows the I_W1–W2–E_W1–W2 curve when we added I₃⁻ (5 and 50 µM) into both W1 and W2 phases at the same concentration. The KI concentration of W2 slightly increased to 0.0015 M because I⁻ also exists in the stock solution of I₃⁻. In the presence of I₃⁻, the I_W1–W2 values noticeably increased and, surprisingly, the E_{W1–W2, i=0} shifted in the positive direction. This indicates that the permeability of K⁺ became much higher than that of the coexisting anions (I⁻ and I₃⁻) in the presence of I₃⁻. Moreover, the estimated E_{W1–W2, i=0} values in the presence of 5 and 50 µM I₃⁻ were almost the same, which means that the I₃⁻ concentration can hardly affect the E_{W1–W2, i=0} value. Since the concentration of I₃⁻ was much lower than those of K⁺ and I⁻, the supply of I₃⁻ from aqueous phases is still less than those of K⁺ and I⁻. Therefore, the ion-transport current from aqueous phases to BLMs is negligibly small, and the current due to the transport of K⁺ (and I⁻) is mainly observed. Therefore, it is considered that I₃⁻ only facilitated the ion transport of K⁺ across the BLMs as a charge carrier.

Fig. 4 NPV for the ion transport across the BLMs in the TM between W1 and W2. W1 contained 0.1 M aqueous solution of KI, and W2 contained 0.001 M aqueous solution of KI in the (a) absence of I₃⁻ (●) and (b) presence of 5 (○) and 50 (▲) µM I₃⁻.

Facilitation of K⁺ transport across bilayer lipid membranes by I₃⁻

For the electrolyte solution (MX) of W2 and W1, the relationship between measured E_{W1–W2, i=0} and r can be expressed using eqn (3).^22,39,41,48Here, α represents the ratio of the ion-transport current of cation M⁺ to that of anion X⁻ (α = I_M⁺/I_X⁻). Therefore, after considering the estimated E_{W1–W2, i=0} and the corresponding r, the ratio of ion permeabilities on both ions can be evaluated. Eqn (3) also indicates that E_{W1–W2, i=0} shows a linear relationship to ln(r) when α is constant. Moreover, α can be evaluated by calculating the slope of the regression line. As for the aqueous solution containing only KI, α can be expressed using eqn (4).^22,41Here, c_K+ and c_I⁻ represent the concentration of K⁺ and I⁻, respectively, and they should be equal. Meanwhile, β_KI represents the distribution coefficient of KI. Therefore, we can express α as the ratio of D of K⁺ and that of I⁻ (α = D_K⁺/D_I⁻).

On the other hand, in the presence of I₃⁻, α can be expressed using eqn (5).

Here, I_K⁺(KI) and I_K⁺(KI3) are the currents for the transport of K⁺ due to the roles of I⁻ and I₃⁻, respectively. According to the experimental conditions, the concentration of KI, c_KI, is far greater than that of KI₃, c_KI3 (c_KI ≫ c_KI3). In addition, the distribution coefficient of KI₃, β_KI3, is greater than that of KI, β_KI (β_KI3 > β_KI), due to its high hydrophobicity.³⁷ Therefore, eqn (5) can be simplified as follows (eqn (6)):

It is thought that the α value varies with the concentration of K⁺, I⁻, and I₃⁻ within the BLMs in the presence of KI₃.

The measurements under the conditions of asymmetric concentration systems were conducted with r = 1, 0.1, and 0.01. Scheme 2 describes the cell composition of the respective case.

Scheme 2

Here, 20 mL of 0.1 M KI aqueous solution was injected into W1 and 20 mL of 0.001 M, 0.01 M, or 0.1 M KI aqueous solution was injected into W2, simultaneously. After that, the same amounts of the aqueous solution containing 0.1 M I₃⁻ were added into W1 and W2 to make 0, 2, 5, 25, and 50 µM, respectively. The increase of the I⁻ concentration after adding I₃⁻ stock solution was a concern in the later calculation.

The curve of E_{W1–W2, i=0}versus ln(r) in the absence of I₃⁻ is shown in Fig. 5(a). It shows a linear relationship. The negative shift of E_{W1–W2, i=0} with a decrease in ln(r) indicates that D_K⁺ is lower than D_I⁻. On the other hand, Fig. 5(b) shows the curves of E_{W1–W2, i=0}versus ln(r) in the presence of I₃⁻. These are also linear relation but they are opposite to the curve in Fig. 5(a). In Fig. 5(b), the positive shift of E_{W1–W2, i=0} with a decrease in ln(r) indicates that the permeability of K⁺ is higher than that of I⁻.

Fig. 5 Relationship between the calculated zero-current potentials and the logarithm of the concentration ratio (a) in the absence of I₃⁻ (●) and (b) in the presence of 2 (●), 5 (○), and 50 (▲) µM I₃⁻.

Fig. 6 shows the relationship between the estimated α value and the concentration of I₃⁻. In the absence of I₃⁻, the α value was estimated to be 0.12 ± 0.08. In other words, D_I⁻ is originally 8 times higher than D_K⁺. In the presence of 50 µM I₃⁻, the α value was estimated to be 9 ± 4. This means that the permeability of K⁺ became 9 times higher than that of I⁻ in the presence of I₃⁻. We can roughly conclude that, owing to the facilitation by I₃⁻, the ion permeability of K⁺ becomes 72 times higher. Furthermore, the α value rapidly raised with the low concentration region of I₃⁻ and then became constant when the concentration of I₃⁻ was more than 25 µM.

Fig. 6 Relationship between the estimated α and the concentration of I₃⁻ in the aqueous solution.

The facilitation ability of I₃⁻ to the transport of K⁺ across BLMs has been confirmed. Fig. 7 shows a schematic diagram of the facilitated transportation of K⁺ across BLMs assisted by I₃⁻. Owing to the high hydrophobicity, most I₃⁻ ions exist inside the BLMs. Moreover, due to the high K_eq (K_eq = 698), I₃⁻ is hardly oxidized to I⁻ in the absence of another oxidant, so I₃⁻ is considered to be stable within BLMs.^43,44 When K⁺ is distributed from one aqueous phase to the BLM, K⁺ exists with I₃⁻ in the BLM to maintain the electroneutrality within the BLM. Finally, K⁺ is released from the opposite side of the BLM into another aqueous phase. The process is generally similar to the mechanism that facilitates the transport of target ions by carrier compounds.

Fig. 7 Overview of the facilitation of K⁺-transport across the BLM by I₃⁻.

Conclusions

Although it is usually recognized that BLMs are a barrier for the transport of hydrophilic ions, it is proven that the ion-transport current is caused by the anti-port of K⁺ and I⁻ in the presence of 0.1 M KI. In addition, the role of I₃⁻ in the facilitation of K⁺-transport within the BLMs was elucidated. By comparison with the K⁺-transport in the presence of only I⁻, the permeability of K⁺ within the BLMs became 72 times higher than that in the presence of 50 µM I₃⁻. The result indicates that I₃⁻ serves as a charge carrier of K⁺ within the BLMs. In the living organisms, it is thought that I₃⁻ plays a special role in charge transport across cell membranes.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data will be made available on request.

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d5cp02791h.

Acknowledgements

This study was supported by the 30th Botanical Research Grant from the ICHIMURA Foundation and a donation from Mr Nobuo Takeshige.

References

1. H. T. Tien, Prog. Surf. Sci., 1985, 19, 169–274.
2. P. Berben, A. Bauer-Brandl, M. Brandl, B. Faller, G. E. Flaten, A.-C. Jacobsen, J. Brouwers and P. Augustijns, Eur. J. Pharm. Sci., 2018, 119, 219–233.
3. T. Ma, M. Sato, M. Komiya, X. Feng, D. Tadaki and A. Hirano-Iwata, Chem. Lett., 2021, 50, 418–425.
4. D. Mrdenovic, Z. Su, W. Kutner, J. Lipkowski and P. Pieta, Nano. Adv., 2020, 2, 3467–3480.
5. W. D. Stein, Channels, Carriers, and Pumps: An Introduction to Membrane Transport, Academic Press, Inc, California, 1990.
6. R. B. Gennis, Biomembranes: Molecular Structure and Function, Springer-Verlag, New York, 1st edn, 1989.
7. Z. Su, X. Ran, J. J. Leitch, A. L. Schwan, R. Faragher and J. Lipkowski, Langmuir, 2019, 35, 16935–16943.
8. Z. Su, J. J. Leitch, S. Sek and J. Lipkowski, Langmuir, 2021, 37, 9613–9621.
9. S. Kubota, S. Ozaki, J. Onishi, K. Kano and O. Shirai, Anal. Sci., 2009, 25, 189–193.
10. S. Kubota, O. Shirai, Y. Kitazumi and K. Kano, Anal. Sci., 2016, 32, 189–192.
11. T. Yamaguchi, O. Shirai, Y. Kitazumi and K. Kano, Electroanalysis, 2018, 30, 304–309.
12. T. Yamaguchi, Y. Kitazumi, K. Kano and O. Shirai, Electroanalysis, 2020, 32, 1093–1099.
13. P. Mueller, D. O. Rudin, H. Ti Tien and W. C. Wescott, Nature, 1962, 194, 979–980.
14. Z. Su, J. J. Leitch and J. Lipkowski, Langmuir, 2022, 38, 8398–8406.
15. M. Hirano, D. Yamamoto, M. Asakura, T. Hayakawa, S. Mise, A. Matsumoto and T. Ide, Micromachines, 2020, 11, 1070.
16. T. Ide, T. Kobayashi and M. Hirano, Anal. Chem., 2008, 80, 7792–7795.
17. A. Hirano-Iwata, T. Taira, A. Oshima, Y. Kimura and M. Niwano, Appl. Phys. Lett., 2010, 96, 213706.
18. T. Mach, C. Chimerel, J. Fritz, N. Fertig, M. Winterhalter and C. Fütterer, Anal. Bioanal. Chem., 2008, 390, 841–846.
19. M. Mayer, J. K. Kriebel, M. T. Tosteson and G. M. Whitesides, Biophys. J., 2003, 85, 2684–2695.
20. A. Simon, A. Girard-Egrot, F. Sauter, C. Pudda, N. Picollet D'Hahan, L. Blum, F. Chatelain and A. Fuchs, J. Colloid Interface Sci., 2007, 308, 337–343.
21. Z. Aminipour, M. Khorshid, M. Bayoumi, P. Losada-Pérez, R. Thoelen, S. Bonakdar, H. Keshvari, G. Maglia, P. Wagner and B. Van der Bruggen, Phys. Status Solidi A, 2017, 214, 1700104.
22. W. Chuang, K. Sowa, Y. Kitazumi and O. Shirai, Electrochim. Acta, 2024, 497, 144488.
23. M. B. Zimmermann, J. Trace Elem. Med. Biol., 2008, 22, 81–92.
24. B. Saad, W. T. Wai, B. P. Lim and M. I. Saleh, Anal. Chim. Acta, 2006, 565, 261–270.
25. M. Hingorani, C. Spitzweg, G. Vassaux, K. Newbold, A. Melcher, H. Pandha, R. Vile and K. Harrington, Curr. Cancer Drug Targets, 2010, 10, 242–267.
26. S. Ravera, J. P. Nicola, G. Salazar-De Simone, F. J. Sigworth, E. Karakas, L. M. Amzel, M. A. Bianchet and N. Carrasco, Nature, 2022, 612, 795–801.
27. Z. Edis, S. Haj Bloukh, H. Abu Sara, H. Bhakhoa, L. Rhyman and P. Ramasami, Journal, 2019, 8, 182.
28. R. S. Bahn, H. B. Burch, D. S. Cooper, J. R. Garber, M. C. Greenlee, I. Klein, P. Laurberg, I. R. McDougall, V. M. Montori, S. A. Rivkees, D. S. Ross, J. A. Sosa and M. N. Stan, Endocr. Pract., 2011, 17, 456–520.
29. A. M. Leung and L. E. Braverman, Nat. Rev. Endocrinol., 2014, 10, 136–142.
30. M. B. Zimmermann, Semin. Cell Dev. Biol., 2011, 22, 645–652.
31. M. B. Zimmermann, Y. Ito, S. Y. Hess, K. Fujieda and L. Molinari, Am. J. Clin. Nutr., 2005, 81, 840–844.
32. P. L. Bigliardi, S. A. L. Alsagoff, H. Y. El-Kafrawi, J.-K. Pyon, C. T. C. Wa and M. A. Villa, Int. J. Surg., 2017, 44, 260–268.
33. S. E. J. O'Kane, G. Richardson, A. Pockett, R. G. Niemann, J. M. Cave, N. Sakai, G. E. Eperon, H. J. Snaith, J. M. Foster, P. J. Cameron and A. B. Walker, J. Mater. Chem. C, 2017, 5, 452–462.
34. D. L. Jacobs, M. A. Scarpulla, C. Wang, B. R. Bunes and L. Zang, J. Phys. Chem. C, 2016, 120, 7893–7902.
35. L. A. Reznikova, E. E. Morgunova and A. D. Davydov, Russ. J. Electrochem., 2002, 38, 659–662.
36. D. Mauzerall and A. Finkelstein, Nature, 1969, 224, 690–692.
37. K.-H. Klotz and R. Benz, Biochim. Biophys. Acta, 1995, 1235, 378–386.
38. C. J. Bender and H. T. Tien, Anal. Chim. Acta, 1987, 201, 51–58.
39. O. Shirai, T. Ohnuki, Y. Kitazumi and K. Kano, Electroanalysis, 2013, 25, 1823–1826.
40. H. Inoue, J. Membr. Sci., 2004, 228, 209–215.
41. K. Kimura, O. Shirai, Y. Kitazumi and K. Kano, J. Electroanal. Chem., 2016, 779, 131–136.
42. S.-H. Jung, J.-W. Yeon, Y. Kang and K. Song, Asian J. Chem., 2014, 26, 4084–4086.
43. D. A. Palmer, R. W. Ramette and R. E. Mesmer, J. Solution Chem., 1984, 13, 673–683.
44. M. Morrison, G. S. Bayse and A. W. Michaels, Anal. Biochem., 1971, 42, 195–201.
45. J. T. Håkedal and P. K. Egeberg, Analyst, 1997, 122, 1235–1238.
46. K. Balantič, V. U. Weiss, G. Allmaier and P. Kramar, Bioelectrochemistry, 2022, 143, 107988.
47. D. E. Goldman, J. Gen. Physiol., 1943, 27, 37–60.
48. J. Onishi, O. Shirai and K. Kano, Electroanalysis, 2010, 22, 1229–1238.

---