Development of a Sensor with a Lipid/Polymer Membrane Comprising Na+ Ionophores to Evaluate the Saltiness Enhancement Effect

The saltiness enhancement effect is the effect whereby saltiness is enhanced by adding specific substances to salt (sodium chloride). Since this effect can be used in the development of salt-reduced foods, a method to objectively evaluate the saltiness with this effect is required. A taste sensor with lipid/polymer membranes has been used to quantify the taste of food and beverages in recent years. The sensor electrodes of this taste sensor have the feature of selectively responding to each of the five basic tastes, which is realized by the lipid/polymer membranes. In this study, we developed a new saltiness sensor based on the lipid/polymer membrane with the aim of quantifying the saltiness enhancement effect. In addition to the conventional components of a lipid, plasticizer, and polymer supporting reagent, the membrane we developed comprises ionophores, which selectively capture sodium ions. As a result, the response of the sensor increased logarithmically with the activity of NaCl in measured samples, similarly to the taste response of humans. In addition, all of the sensor responses increased upon adding saltiness-enhancing substances, such as citric acid, tartaric acid and branched-chain amino acids (BCAAs), to NaCl samples. These findings suggest that it is possible to quantify the saltiness enhancement effect using a taste sensor with lipid/polymer membranes.


Introduction
The taste qualities perceived by humans when they taste food or beverages are saltiness, sourness, sweetness, bitterness and umami. These tastes are called the "five basic tastes," and each taste has a corresponding biological role or feature. The sweetness of, for example, sugar indicates a source of energy for the body, whereas bitterness is a signal of poison. Saltiness, as considered in this study, is the taste conferred by sodium and other ions and indicates the existence of minerals. In addition, saltiness not only contributes to the deliciousness of food but also is routinely used to emphasize the taste of food. On the other hand, salt also has a side effect. As is widely known, an excessive intake the lipid/polymer membrane of a conventional taste sensor. Using the fabricated sensor, we attempted to measure the saltiness enhancement effect.

Sensory Test
We performed a sensory test on the saltiness enhancement effect in collaboration with the Taste & Aroma Strategic Research Institute Co., Ltd. (Tokyo, Japan). The basic taste samples (salty, sour, umami, bitter) and amino acid samples shown in Table 1 were prepared as base samples. Then, the saltiness intensities of the NaCl sample (saltiness sample in Table 1) and the NaCl + taste samples (sourness, umami, bitterness for medicine, bitterness for food and amino acid samples in Table 1) were compared. Three concentrations of the basic taste samples excluding the saltiness sample were set by taking into account their thresholds, i.e., a third of the threshold, the threshold and three times the threshold [35][36][37][38]. The concentration of amino acids (15 mM) was decided with reference to the following paper [3]. The concentration of NaCl (1%) corresponded to that of typical miso soup in Japan. Amino acid sample L-asparagine monohydrate (15 mM) L-glutamine (15 mM) Ten well trained panelists (four male and six female adults) were enrolled in this study. The panelists were selected by their ability to detect taste substances with high sensitivity, and have been trained approximately once a week. The sensory test was a two-point identification test in which each panelist selected the sample they felt was saltier. First, the panelists rinsed their mouth with pure water. Then, they were asked to place a sample in their mouths (NaCl + bitterness samples were spat out instead of being swallowed). Two samples were evaluated: first the NaCl sample, then the NaCl + taste sample. Also, when they moved on to the evaluation of the next sample, they washed their mouth with pure water again so that the aftertaste of the previous sample did not remain. Finally, they were asked, "Which sample do you feel was saltier?". The panelists were given three choices of answer: "NaCl sample", "Comparable saltiness", and "NaCl + taste sample". The compositions of all the samples were not told to the panelists in advance.

Lipid/Polymer Membrane
We fabricated a sensor with a lipid/polymer membrane comprising ionophores as a new sensor for measuring the saltiness enhancement effect, and a NaISE to compare its response to NaCl with that of the fabricated sensor. The NaISE was composed of bis(12-crown-4) (18 mg) as an ionophore, NPOE (350 µL) as a plasticizer, TFPB (3.5 mg) as an anion exclusion material, which improves the selectivity of ISE to cations, and PVC (180 mg) as a polymer supporting reagent. TFPB itself has a negative charge, and hence, anions in sample solutions become less likely to interact with the ISE membrane.
As mentioned, the lipid/polymer membranes used in commercial taste sensor electrodes are usually composed of a lipid, a plasticizer and PVC. In addition to the three substances, the newly developed lipid/polymer membrane comprising ionophores contains bis(12-crown-4) (18 mg) and TFPB (3.5 mg). In this membrane, PADE (60 mg), which is negatively charged, NPOE (350 µL) and PVC (180 mg) were used as a lipid, a plasticizer and a polymer supporting reagent, respectively.
First, these materials were dissolved in the solvent tetrahydrofuran (THF) and then stirred. Next, the solution was poured into a petri dish with 45 mm inner diameter, and the membrane was formed by volatilizing the THF. In this study, we measured samples with four sensor electrodes with the membranes formed in one petri dish.

Measurement Procedure of Taste Sensor
We used a TS-5000Z commercial taste sensing system (Intelligent Sensor Technology, Inc., Kanagawa, Japan) for measurements. The lipid/polymer membrane comprising ionophores developed in this study or the NaISE was used for sensor electrodes attached to the TS-5000Z. The lipid/polymer membrane was pasted to the receptive parts of hollow bars made of PVC ( Figure 1). The potential difference between the sensor electrode and the reference electrode was used as the sensor output. TFPB (3.5 mg). In this membrane, PADE (60 mg), which is negatively charged, NPOE (350 µl) and PVC (180 mg) were used as a lipid, a plasticizer and a polymer supporting reagent, respectively.
First, these materials were dissolved in the solvent tetrahydrofuran (THF) and then stirred. Next, the solution was poured into a petri dish with 45 mm inner diameter, and the membrane was formed by volatilizing the THF. In this study, we measured samples with four sensor electrodes with the membranes formed in one petri dish.

Measurement Procedure of Taste Sensor
We used a TS-5000Z commercial taste sensing system (Intelligent Sensor Technology, Inc., Kanagawa, Japan) for measurements. The lipid/polymer membrane comprising ionophores developed in this study or the NaISE was used for sensor electrodes attached to the TS-5000Z. The lipid/polymer membrane was pasted to the receptive parts of hollow bars made of PVC ( Figure 1). The potential difference between the sensor electrode and the reference electrode was used as the sensor output.
The measurement procedure was as follows [19]. As shown in Figure 2, first, the sensor electrode was soaked in the reference solution (30 mM KCl, 0.3 mM tartaric acid), and the membrane potential (Vr) is measured after 30 s. Next, the sensor electrode was soaked in the sample solution, and the membrane potential (Vs) was measured after 30 s. The difference between Vs and Vr, (Vs -Vr), was defined as a relative value. Finally, the membrane surface was initialized by washing with a cleaning solution (100 mM HCl, 30 vol% ethanol) for 90 s and with a reference solution for 240 s. The cleaning solution makes it possible to detach the taste substances adsorbed on to the membrane without leakage of the membrane components and then let the membrane recover to its initial condition. In addition, the membrane potential was confirmed whether it has recovered before Step 1 in Figure 2, and then the next measurement was started. The procedure was repeated five times for each sample using four electrodes, and the average of the second to fifth measurements was calculated as a relative value for all measurements in this study. The standard deviations were calculated for n = 4 (electrodes) × 4 (rotations) = 16 values.  The measurement procedure was as follows [19]. As shown in Figure 2, first, the sensor electrode was soaked in the reference solution (30 mM KCl, 0.3 mM tartaric acid), and the membrane potential (V r ) is measured after 30 s. Next, the sensor electrode was soaked in the sample solution, and the membrane potential (V s ) was measured after 30 s. The difference between V s and V r , (V s -V r ), was defined as a relative value. Finally, the membrane surface was initialized by washing with a cleaning solution (100 mM HCl, 30 vol% ethanol) for 90 s and with a reference solution for 240 s. The cleaning solution makes it possible to detach the taste substances adsorbed on to the membrane without leakage of the membrane components and then let the membrane recover to its initial condition. In addition, the membrane potential was confirmed whether it has recovered before Step 1 in Figure 2, and then the next measurement was started. The procedure was repeated five times for each sample using four electrodes, and the average of the second to fifth measurements was calculated as a relative value for all measurements in this study. The standard deviations were calculated for n = 4 (electrodes) × 4 (rotations) = 16 values.

Measurement of the Saltiness Enhancement Effect
First, NaCl (0.1-1000 mM) samples were measured using the fabricated sensor with the lipid/polymer membrane comprising ionophores. Then, the responses were compared with that of the NaISE. Next, the saltiness enhancement effects of BCAAs (L-valine, L-leucine and L-isoleucine), citric acid and tartaric acid were measured. BCAAs have been orally administered to patients with cirrhosis or hepatic encephalopathy and as a supplement for fatigue recovery [39,40]. In this measurement, two types of sample, NaCl solution and mixed solutions containing NaCl and a saltiness-enhancing substance, were measured using the sensor with the lipid/polymer membrane comprising ionophores. Moreover, it was investigated whether the sensor response to the mixed solution was larger than that to the NaCl solution. The samples to be measured are shown in Table 2, and the concentration of each saltiness-enhancing substance was prepared on the basis of the result of the sensory test where the saltiness enhancement effect was confirmed. The concentration of NaCl in the sample solutions was set to 100 mM, which is slightly lower than that of miso soup (1%). Deionized water was used as the solvent for all sample solutions.

Composition
Concentration NaCl 100 mM NaCl NaCl + L-Valine 100 mM NaCl + L-valine (3, 10, 30, 50 mM) NaCl + L-Leucine 100 mM NaCl + L-leucine (10, 30, 50, 75, 100 mM) NaCl + L-Isoleucine 100 mM NaCl + L-isoleucine (10, 30, 50, 75, 100, 200 mM) NaCl + Citric acid 100 mM NaCl + Citric acid (0.094, 0.13, 0.3, 1 mM) NaCl + Tartaric acid 100 mM NaCl + Tartaric acid (0.033, 0.1, 0.3, 0.61 mM) Figure 3 shows the result of the sensory test. The significance of difference was determined by a binomial test using the numbers of panelists who answered "NaCl sample" and "NaCl + taste sample" with p ≦ 0.05 indicating a significant difference. In the significance test, although p = 0.055 for L-asparagine and monosodium glutamate was larger than 0.05, it was judged to be significant in this test because it was very close to 0.05, and L-asparagine has already been reported to contribute to the saltiness enhancement effect [3]. According to these results, L-asparagine and L-glutamine were shown to contribute to the saltiness enhancement effect as already reported [3]. In addition, tartaric acid and monosodium glutamate contributed to the saltiness enhancement effect.

Measurement of the Saltiness Enhancement Effect
First, NaCl (0.1-1000 mM) samples were measured using the fabricated sensor with the lipid/polymer membrane comprising ionophores. Then, the responses were compared with that of the NaISE. Next, the saltiness enhancement effects of BCAAs (L-valine, L-leucine and L-isoleucine), citric acid and tartaric acid were measured. BCAAs have been orally administered to patients with cirrhosis or hepatic encephalopathy and as a supplement for fatigue recovery [39,40]. In this measurement, two types of sample, NaCl solution and mixed solutions containing NaCl and a saltiness-enhancing substance, were measured using the sensor with the lipid/polymer membrane comprising ionophores. Moreover, it was investigated whether the sensor response to the mixed solution was larger than that to the NaCl solution. The samples to be measured are shown in Table 2, and the concentration of each saltiness-enhancing substance was prepared on the basis of the result of the sensory test where the saltiness enhancement effect was confirmed. The concentration of NaCl in the sample solutions was set to 100 mM, which is slightly lower than that of miso soup (1%). Deionized water was used as the solvent for all sample solutions. Table 2. Compositions of sample solutions to be measured.

Composition Concentration
NaCl 100 mM NaCl NaCl + L-Valine 100 mM NaCl + L-valine (3, 10, 30, 50 mM) NaCl + L-Leucine 100 mM NaCl + L-leucine (10, 30, 50, 75, 100 mM) NaCl + L-Isoleucine 100 mM NaCl + L-isoleucine ( Figure 3 shows the result of the sensory test. The significance of difference was determined by a binomial test using the numbers of panelists who answered "NaCl sample" and "NaCl + taste sample" with p 0.05 indicating a significant difference. In the significance test, although p = 0.055 for L-asparagine and monosodium glutamate was larger than 0.05, it was judged to be significant in this test because it was very close to 0.05, and L-asparagine has already been reported to contribute to the saltiness enhancement effect [3]. According to these results, L-asparagine and L-glutamine were shown to contribute to the saltiness enhancement effect as already reported [3]. In addition, tartaric acid and monosodium glutamate contributed to the saltiness enhancement effect.   Figure 4 shows the responses of the sensor with the lipid/polymer membrane comprising ionophores and the NaISE to NaCl. The responses of both sensors increased logarithmically with the NaCl concentration. Strictly speaking, the concentration should be replaced by activity. Here we use the word "concentration" for simplicity. The responses of the sensor with the lipid/polymer membrane comprising ionophores and the NaISE increased by 56 mV and 59 mV when the sodium concentration was increased by one order of magnitude in the high-concentration region (10-1000 mM), indicating an ideal Nernst potential response. Although the response of the sensor with the lipid/polymer membrane comprising ionophores was slightly smaller than that of the NaISE, it can be seen that the sensor has sufficient sensitivity to sodium ions. According to the Weber-Fechner law, the intensity of taste perceived by humans increases logarithmically with the concentration of the taste substance. The results shown in Figure 4 satisfy the Weber-Fechner law, and hence the sensor membrane fabricated in this study could quantify the saltiness intensity. Moreover, the use of the sensor membrane for over 100 measurements was possible (CV < 10%).  Figure 4 shows the responses of the sensor with the lipid/polymer membrane comprising ionophores and the NaISE to NaCl. The responses of both sensors increased logarithmically with the NaCl concentration. Strictly speaking, the concentration should be replaced by activity. Here we use the word "concentration" for simplicity. The responses of the sensor with the lipid/polymer membrane comprising ionophores and the NaISE increased by 56 mV and 59 mV when the sodium concentration was increased by one order of magnitude in the high-concentration region (10-1000 mM), indicating an ideal Nernst potential response. Although the response of the sensor with the lipid/polymer membrane comprising ionophores was slightly smaller than that of the NaISE, it can be seen that the sensor has sufficient sensitivity to sodium ions. According to the Weber-Fechner law, the intensity of taste perceived by humans increases logarithmically with the concentration of the taste substance. The results shown in Figure 4 satisfy the Weber-Fechner law, and hence the sensor membrane fabricated in this study could quantify the saltiness intensity. Moreover, the use of the sensor membrane for over 100 measurements was possible (CV < 10%).

Response to 100 mM NaCl with Added Saltiness-Enhancing Substances
Citric acid, tartaric acid and BCAAs added to samples measured in this study have all been reported to contribute to the saltiness enhancement effect [4][5][6]8]. For tartaric acid, this was also confirmed in the sensory test of this study. Figure 5a,b shows the responses of the sensor with the lipid/polymer membrane comprising ionophores to each sample. All the samples were based on 100 mM NaCl, and the initial responses of the sensor were standardized to zero (the response to 100 mM NaCl solution was subtracted from all the responses). As shown in Figure 5a,b, the sensor response increased with increasing amount of each saltiness-enhancing substance. On the other hand, the responses of the NaISE to these samples did not increase even when the concentrations of the saltiness-enhancing substances were increased. Since the sensor responses can be considered as an index of saltiness intensity, as shown in Figure 4, the results in Figure 5a,b mean that the saltiness intensity was enhanced with the addition of citric acid, tartaric acid or BCAAs. Thus, it was found that the sensor with the lipid/polymer membrane comprising ionophores can measure the equivalent salt enhancement, and can reproduce and quantify the saltiness enhancement effect.

Response to 100 mM NaCl with Added Saltiness-Enhancing Substances
Citric acid, tartaric acid and BCAAs added to samples measured in this study have all been reported to contribute to the saltiness enhancement effect [4][5][6]8]. For tartaric acid, this was also confirmed in the sensory test of this study. Figure 5a,b shows the responses of the sensor with the lipid/polymer membrane comprising ionophores to each sample. All the samples were based on 100 mM NaCl, and the initial responses of the sensor were standardized to zero (the response to 100 mM NaCl solution was subtracted from all the responses). As shown in Figure 5a,b, the sensor response increased with increasing amount of each saltiness-enhancing substance. On the other hand, the responses of the NaISE to these samples did not increase even when the concentrations of the saltiness-enhancing substances were increased. Since the sensor responses can be considered as an index of saltiness intensity, as shown in Figure 4, the results in Figure 5a,b mean that the saltiness intensity was enhanced with the addition of citric acid, tartaric acid or BCAAs. Thus, it was found that the sensor with the lipid/polymer membrane comprising ionophores can measure the equivalent salt enhancement, and can reproduce and quantify the saltiness enhancement effect.

Response to 100 mM NaCl with Added Saltiness-Enhancing Substances
Citric acid, tartaric acid and BCAAs added to samples measured in this study have all been reported to contribute to the saltiness enhancement effect [4][5][6]8]. For tartaric acid, this was also confirmed in the sensory test of this study. Figure 5a,b shows the responses of the sensor with the lipid/polymer membrane comprising ionophores to each sample. All the samples were based on 100 mM NaCl, and the initial responses of the sensor were standardized to zero (the response to 100 mM NaCl solution was subtracted from all the responses). As shown in Figure 5a,b, the sensor response increased with increasing amount of each saltiness-enhancing substance. On the other hand, the responses of the NaISE to these samples did not increase even when the concentrations of the saltiness-enhancing substances were increased. Since the sensor responses can be considered as an index of saltiness intensity, as shown in Figure 4, the results in Figure 5a,b mean that the saltiness intensity was enhanced with the addition of citric acid, tartaric acid or BCAAs. Thus, it was found that the sensor with the lipid/polymer membrane comprising ionophores can measure the equivalent salt enhancement, and can reproduce and quantify the saltiness enhancement effect.

Response of Sensor with Low PADE Content
As mentioned above, 60 mg PADE, which is a negatively charged lipid, was mixed in the lipid/polymer membrane comprising ionophores in the measurements, to obtain the results in Figure 5. Actually, four sensor membranes comprising 20, 40, 60 and 80 mg PADE (the compositions other than PADE were the same) were fabricated. However, the membrane comprising 80 mg PADE could not be used because of lipid deposition. In addition, the sensor responses of the membranes comprising 20 and 40 mg PADE to the mixed samples of NaCl and BCAAs did not increase monotonically. As shown in Figure 6, the sensor responses of the membrane comprising 40 mg PADE decreased initially, reached a peak at about 30 mM BCAAs, and then increased with the addition of BCAAs.
aqueous solution in the form of the three types of ion shown in Figure 7, where the abundance ratio of each ion depends on the pH of the solution. The pH at which the average charge of an amino acid is zero is called the isoelectric point (Ip). When the pH is lower than the Ip, amino acids receive H + , and hence the abundance of cations increases. On the other hand, when the pH is higher than the Ip, amino acids release H + , leading to an increase in the abundance of anions. Since the Ip of BCAAs (approx. 6.0) is higher than the pH of NaCl solution (approx. 5.0), BCAAs become cations by receiving H + in the solution. Therefore, the dissociation of PADE is promoted by increasing the concentration of positively charged BCAAs in the measuring sample. As a result, the surface potential of the membrane decreases, and hence the sensor response initially decreases (Figure 8). All the molecules of PADE can be considered to be dissociated at around 30 mM BCAAs, corresponding to the minimum point appearing in Figure 6. Above 30 mM BCAAs, the response potential increased because of the electrostatic interaction between the negatively charged membrane and the positively charged BCAAs. On the other hand, when the membrane comprising 60 mg PADE was used, the sensor response increased with the addition of BCAAs (Figure 5a) instead of decreasing owing to the dissociation of H + from PADE. Therefore, 60 mg of PADE is an appropriate amount to be mixed in the lipid/polymer membrane comprising ionophores for measuring the saltiness enhancement effect.  The decreases in the sensor responses are assumed to be caused by the H + dissociation from PADE. The lipid molecule PADE present at the membrane surface is negatively charged by the dissociation of H + in the aqueous solution to be measured. However, not all the molecules of PADE dissociate H + ; the degree of dissociation depends on the pH and the concentration of cations in the solution [41], and is around 70% in the case of 100 mM NaCl solution [18]. The BCAAs added to the NaCl solution have amino groups and carboxyl groups in their molecules. Amino acids exist in an aqueous solution in the form of the three types of ion shown in Figure 7, where the abundance ratio of each ion depends on the pH of the solution. The pH at which the average charge of an amino acid is zero is called the isoelectric point (Ip). When the pH is lower than the Ip, amino acids receive H + , and hence the abundance of cations increases. On the other hand, when the pH is higher than the Ip, amino acids release H + , leading to an increase in the abundance of anions. Since the Ip of BCAAs (approx. 6.0) is higher than the pH of NaCl solution (approx. 5.0), BCAAs become cations by receiving H + in the solution. Therefore, the dissociation of PADE is promoted by increasing the concentration of positively charged BCAAs in the measuring sample. As a result, the surface potential of the membrane decreases, and hence the sensor response initially decreases ( Figure 8). All the molecules of PADE can be considered to be dissociated at around 30 mM BCAAs, corresponding to the minimum point appearing in Figure 6. Above 30 mM BCAAs, the response potential increased because of the electrostatic interaction between the negatively charged membrane and the positively charged BCAAs. On the other hand, when the membrane comprising 60 mg PADE was used, the sensor response increased with the addition of BCAAs (Figure 5a) instead of decreasing owing to the dissociation of H + from PADE. Therefore, 60 mg of PADE is an appropriate amount to be mixed in the lipid/polymer membrane comprising ionophores for measuring the saltiness enhancement effect.  Vm, Vm', Vs and Vd are the membrane potential, the membrane potential upon adding BCAAs, the surface potential and the diffusion potential, respectively. Ionophores form a diffusion potential inside the membrane owing to sodium ion flow [33]. This figure shows this diffusion potential profile as well as the surface potential originating from charged lipid molecules. Since the diffusion potential is expected to change only slightly, the change in the surface potential is dominant in this situation.

Conclusions
With the aim of quantifying the saltiness enhancement effect, we developed a new saltiness sensor, the response of which increases when saltiness-enhancing substances are added to the NaCl sample. The sensor was fabricated with a lipid/polymer membrane comprising ionophores. The sensor response increased logarithmically with the NaCl concentration and this result corresponds to the increasing taste intensity felt by humans. Moreover, the responses to the samples composed of NaCl and a saltiness-enhancing substance (citric acid, tartaric acid or BCAAs) were all larger than the response to only NaCl. We succeeded in developing a new sensor that can reproduce the saltiness enhancement effect. The present work focused on NaCl solutions to which saltiness-enhancing substances were added. Therefore, in future we will measure the effect of other saltiness-enhancing substances or commercial saltiness enhancers. In addition, the application of the sensor to real foods comprising many taste substances and an evaluation of the selectivity, longevity, and repeatability of the sensor are future works, with the ultimate aim of developing a saltiness sensor that can quantify saltiness including the saltiness enhancement effect as an alternative to human sensory tests.
Author Contributions: The work presented here was carried out as a collaboration between all authors; Y.T., H.I., H.S. and K.T. defined the research theme; F.N., T.I., Y.M. and Y.K. carried out the experiments and analyzed the data; F.N. interpreted the results and wrote the paper; X.W., Y.T., H.I. and K.T. provided directions for the experimental methods, analysis of data, interpretation of the results, and writing of the paper; All the authors have contributed to, seen, and approved the manuscript.  Vm, Vm', Vs and Vd are the membrane potential, the membrane potential upon adding BCAAs, the surface potential and the diffusion potential, respectively. Ionophores form a diffusion potential inside the membrane owing to sodium ion flow [33]. This figure shows this diffusion potential profile as well as the surface potential originating from charged lipid molecules. Since the diffusion potential is expected to change only slightly, the change in the surface potential is dominant in this situation.

Conclusions
With the aim of quantifying the saltiness enhancement effect, we developed a new saltiness sensor, the response of which increases when saltiness-enhancing substances are added to the NaCl sample. The sensor was fabricated with a lipid/polymer membrane comprising ionophores. The sensor response increased logarithmically with the NaCl concentration and this result corresponds to the increasing taste intensity felt by humans. Moreover, the responses to the samples composed of NaCl and a saltiness-enhancing substance (citric acid, tartaric acid or BCAAs) were all larger than the response to only NaCl. We succeeded in developing a new sensor that can reproduce the saltiness enhancement effect. The present work focused on NaCl solutions to which saltiness-enhancing substances were added. Therefore, in future we will measure the effect of other saltiness-enhancing substances or commercial saltiness enhancers. In addition, the application of the sensor to real foods comprising many taste substances and an evaluation of the selectivity, longevity, and repeatability of the sensor are future works, with the ultimate aim of developing a saltiness sensor that can quantify saltiness including the saltiness enhancement effect as an alternative to human sensory tests.
Author Contributions: The work presented here was carried out as a collaboration between all authors; Y.T., H.I., H.S. and K.T. defined the research theme; F.N., T.I., Y.M. and Y.K. carried out the experiments and analyzed the data; F.N. interpreted the results and wrote the paper; X.W., Y.T., H.I. and K.T. provided directions for the experimental methods, analysis of data, interpretation of the results, and writing of the paper; All the authors have contributed to, seen, and approved the manuscript. Anion Zwitterion Figure 8. Simple model of decrease in the membrane potential. V m , V m ', V s and V d are the membrane potential, the membrane potential upon adding BCAAs, the surface potential and the diffusion potential, respectively. Ionophores form a diffusion potential inside the membrane owing to sodium ion flow [33]. This figure shows this diffusion potential profile as well as the surface potential originating from charged lipid molecules. Since the diffusion potential is expected to change only slightly, the change in the surface potential is dominant in this situation.

Conclusions
With the aim of quantifying the saltiness enhancement effect, we developed a new saltiness sensor, the response of which increases when saltiness-enhancing substances are added to the NaCl sample. The sensor was fabricated with a lipid/polymer membrane comprising ionophores. The sensor response increased logarithmically with the NaCl concentration and this result corresponds to the increasing taste intensity felt by humans. Moreover, the responses to the samples composed of NaCl and a saltiness-enhancing substance (citric acid, tartaric acid or BCAAs) were all larger than the response to only NaCl. We succeeded in developing a new sensor that can reproduce the saltiness enhancement effect. The present work focused on NaCl solutions to which saltiness-enhancing substances were added. Therefore, in future we will measure the effect of other saltiness-enhancing substances or commercial saltiness enhancers. In addition, the application of the sensor to real foods comprising many taste substances and an evaluation of the selectivity, longevity, and repeatability of the sensor are future works, with the ultimate aim of developing a saltiness sensor that can quantify saltiness including the saltiness enhancement effect as an alternative to human sensory tests.
Author Contributions: The work presented here was carried out as a collaboration between all authors; Y.T., H.I., H.S. and K.T. defined the research theme; F.N., T.I., Y.M. and Y.K. carried out the experiments and analyzed the data; F.N. interpreted the results and wrote the paper; X.W., Y.T., H.I. and K.T. provided directions for the experimental methods, analysis of data, interpretation of the results, and writing of the paper; All the authors have contributed to, seen, and approved the manuscript.