Introduction

Excessive salt intake is a major risk factor for development of hypertension and cardiovascular diseases. Modest reduction in dietary salt consumption substantially reduces cardiovascular events in the general population.¹ Currently, the World Health Organization recommends <5g/day dietary salt intake. In reality, average daily salt intake is >10 g in most populations worldwide.² Despite several salt reduction strategies, such as sodium restricted diet, use of salt substitute, and the addition of flavor to sodium reduced product,³ daily salt intake per capita has not declined in the past 2 decades in most countries,⁴ possibly because that salt intake is a complex phenomenon associated with dietary habits of ancestors, cultural background, changes in salt taste, and other factors.^5,6

Increased appetite and acceptance of salt-rich foods is more likely to be considered an acquired process, which is associated with eating too many salty foods in infancy.⁷ It has been reported that individuals on long-term high-salt diets have increased their preference for salt,⁸ but other reports indicate that the perception and preference of salt taste in adults have nothing to do with sodium intake,⁹ suggesting that the association between salt taste perception and changes in salt intake is uncertain. Through maintaining a moderate amount of electrolyte intake in our daily life, salt taste perception is characterized by information processing in the taste center of brain via afferent nerves after sodium ions activate taste receptors.³ Although there was no significant change in salt taste perception thresholds in hypertensive subjects,¹⁰ changes in salt perception beyond the threshold were associated with hypertension.¹¹

Recent studies have identified some molecules involved in salt taste perception.¹² Within the physiological range (10–150 mmol/L), salt solution is mainly perceived by the tongue epithelial sodium channel (ENaC), which plays a major role in salt preference.¹³ With the increase of salt solution concentration, the aversion-avoidance reaction would be activated to prevent excessive salt intake. Oka et al¹⁴ found that transient receptor potential channel M5 (TRPM5), a bitter taste sensor, is involved in this process. As a member of transient receptor potential (TRP) channel family selectively expressed in taste receptor cells, TRPM5 acts as an amplifier of bitter, sweet, and umami taste.¹⁵ TRPM5 channels open and allow Na⁺ to enter and depolarize taste receptor cells by activation of G protein-coupled T2Rs (type 2 taste receptors) and subsequent Phospholipase Cβ/PKC-mediated release Ca²⁺ from intracellular stores.^16,17 As the effector of bitter compounds, TRPM5 is selectively inhibited by triphenylphosphine oxide (TPPO).¹⁸ Our recent study indicated that the sensitivity to salt was reduced in the individuals with high daily salt intake, accompanied by a higher threshold for salt tolerance.¹⁹ However, it is not clear whether excessive salt intake is because of disturbance with ENaC-dependent salt perception or impairment of aversion to high-salt stimulation. Therefore, we hypothesized that chronic high-salt intake may either affect ENaC-dependent salt preference or jeopardize TRPM5-mediated aversive behavior to high salt, to enhance high-salt intake.

Here, we investigate the effect of high salt on the salt perception and aversive response and its underlying mechanism. We demonstrate that chronic high-salt intake hinders TRPM5-mediated rejection behavior to noxious salt consumption through inhibition of PKCα activity and PKC-dependent phosphorylation on TRPM5, but does not alter ENaC-dependent salt preference. This effect also appeared in human study. The present study suggests that manipulation of taste TRPM5 may present a promising intervention for salt sensitive hypertension.

Methods

The data that support the findings of this study are available from the author Zhiming Zhu (zhuzm@yahoo.com) on reasonable request. Detailed methods are available in the online-only Data Supplement Materials and Methods section.

Animal Experiments

C57BL/6J (wild-type [WT]) and TRPM5^−/− male mice were purchased from Jackson Laboratory (Bar Harbor, ME). The WT mice were randomized into 3 groups and were fed with normal diet (ND; 0.27% salt) or high-salt diet (HSD; 8% salt). The TRPM5^−/− mice were fed with HSD. All of the experimental procedures were performed in accordance with protocols approved by the Institutional Animal Care and Research Advisory Committee, Daping Hospital, Third Military Medical University.

Human Study

Between March 2017 and October 2017, we screened and enrolled 114 eligible hypertensive patients hospitalized in Daping Hospital and 40 healthy control subjects. The human study was conducted according to the principles of the Declaration of Helsinki and the protocol was approved by the Ethics Committee in Daping Hospital at Third Military Medical University. To be eligible for participation in the study, participants were required to provide written informed consent, and able to comply with all study procedures.

Statistical Analysis

In animal experiments, comparisons between groups were made using 2-tailed unpaired Student t test, 1-way ANOVA with Tukey post hoc test or 2-way ANOVA with Bonferroni multiple comparison post hoc test. Mann-Whitney nonparametric U test was used to analyze data in abnormal distribution. The results represented mean±SEM. In clinical study, the baseline characteristics of the participants were compared between the groups using the χ² test for categorical variables and the 2 sample t test for continuous variables. Human results were presented as the percentge or mean±SD. All analyses were conducted using SPSS software, version 16.0 (SPSS, Inc), or GraphPad Prism software, version 6.0 (GraphPad Software); a 2-sided P of <0.05 indicated statistical significance.

Results

High Salt or TRPM5 Mutant Did Not Alter ENaC-Dependent Salt Preference

First, we compared the sensitive perception of saltiness in our recruited hypertensive patients and healthy volunteers and found no significant difference between hypertension group and control group (P=0.698; Figure 1A). When all the hypertensive patients were divided into low-salt or high-salt group according to their 24-hour urinary sodium excretion (see section Materials and Methods), the salty perception selection percentage was also unchanged (P=0.692; Figure 1A). Next, to explore whether ENaC-dependent salt preference was involved in the enhanced salt intake, we tested the expression of αENaC in tongue epithelium and ENaC-dependent salt preference in mice fed with ND or HSD. The results showed that both mRNA and protein levels of αENaC were not affected by HSD (Figure 1B and 1C). Similarly, the taste preference to low-salt solutions containing NaCl from 10 to 150 mmol/L also exhibited no significant difference (Figure 1D). To determine whether TRPM5 would participate in the ENaC-dependent salt preference, we examined the taste preference for low-salt solutions in TRPM5 knockout (TRPM5^−/−) mice. The preference for low-salt solutions was not altered by TRPM5 knockout. (Figure 1E), Consistently, immunofluorescence staining also indicated that αENaC and TRPM5 were not colocalized in the tongue epithelium of WT mice (Figure 1F). These results indicate that TRPM5 did not participate in the salt perception within the physiological range.

• Download figure
• Download PowerPoint

Functional TRPM5 Was Abundantly Expressed in Tongue Epithelium

Next, the expression and distribution of TRPM5 were examined in various tissues of mice. As previously reported,²⁰ TRPM5 was highly expressed at both the mRNA and protein levels in tongue epithelium, with low level of expression in other tissues/organs of mice (Figure 2A and 2B). Immunofluorescent staining revealed a colocalization of TRPM5 with T2Rs, a taste receptor in tongue epithelium cells (Figure 2C). As TRPM5 activation leads to sodium influx and subsequent depolarization of the cell,²¹ we detected the response of TRPM5 to bitterness in isolated tongue epithelium of mice. We observed a significant increase of the intracellular sodium (Figure 2D) and calcium concentration (Figure S1A in the online-only Data Supplement) when treated by denatonium benzoate, a potent bitter tastant,²² and these phenomena were abolished by TPPO, a selective TRPM5 channel inhibitor,¹⁸ or intracellular Ca²⁺ chelator 1,2-Bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (Figure 2D). These results confirmed the localization and bioactivity of TRPM5 channels in the mice tongue epithelium.

• Download figure
• Download PowerPoint

We further compared electrophysiological characteristics of HEK-293 cells transfected with a plasmid-expressing TRPM5 versus control cells with whole-cell patch-clamp (Figure S1B and S1C). Nontransfected HEK-293 cells displayed a steady-state leak current and linear voltage-current response (Figure 2E). In contrast, transfected cells showed a voltage-dependent outwardly rectifying current (Figure 2E). The voltage-gated current in transfected cells was inhibited by TPPO (Figure 2E and 2F). Moreover, we found the average single channel conductance in these experiments was 28.93 pS (Figure S1D), generally consistent with the slope conductance reported previously.²³ In transfected cells, cucurbitacin E, a chemical isolated from bitter melon,²⁴ increased both inward and outward currents in a concentration-dependent manner (Figure S2A through S2E). At maximum effective concentration (10 μM), cucurbitacin E increased the probability of single channel openings with a 23 pS conductance and the amplitude, similar to that observed in a previous study.²³ Similarly, chlorogenic acid, an intense bitter-tasting constituent in coffee, also increased both outward and inward TRPM5 currents. Treatment with TPPO significantly attenuated the response of currents to bitter substances in these experiments (Figure S2E through S2H). Altogether, it suggests an important role of TRPM5 channel in sensing bitter substances by tongue epithelial cells.

High Salt Inhibited the TRPM5 Function in Dose-Dependent Manner

Next, we examined whether exposure to high salt (NaCl) affects response of TRPM5 channels to denatonium benzoate in mouse tongue epithelium. We found that at a range of 200 to 500 mmol/L NaCl, 3-hour exposure attenuated denatonium benzoate-induced increases in intracellular sodium as well as calcium in a dose-dependent manner, which was independent of osmotic pressure alteration (Figure 3A; Figure S3A). In addition, at 500 mmol/L, NaCl exposure resulted in a time-dependent decrease of intracellular sodium response to denatonium benzoate (Figure S3B and S3C). Similarly, in HEK-293 cells expressing exogenous TRPM5, acute increase of extracellular Na⁺ concentration activated TRPM5-mediated currents, especially inward currents (Figure S3D through S3F); however, 3-hour high-salt exposure inhibited TRPM5-mediated currents in a dose-dependent manner (Figure 3B and 3C). It also caused a progressive decrease of basal current density at −100 mV and +100 mV (Figure 3D). In the cell attached mode, 3-hour high-salt exposure decreased opening probability of TRPM5 channels from 0.143±0.014 to 0.035±0.005 (Figure 3E). To investigate whether the inhibitory effects of NaCl on TRPM5 activity was dependent on the salt-increased osmotic pressure, mannitol was used as an osmotic control. And the results showed that mannitol failed to affect the current density (Figure S3G through S3I), suggesting that the change of osmotic pressure by NaCl would not be involved in the inhibitory effects of high salt on TRPM5 activity.

• Download figure
• Download PowerPoint

High Salt Inhibited TRPM5 Activity Through Alteration of PKC-Dependent Phosphorylation

TRPM5-related bitter taste perception is mediated by phospholipase C β2. As a direct target of PLC signaling, PKC is involved in the signal transduction in cellular multiple biological function.^20,25 To investigate how high-salt intake affects the expression and activity of TRPM5, mice were subjected to normal diet (ND) or high-salt diet (HSD) for 3 months. Compared with mice on ND, both protein and mRNA expression levels of TRPM5 were significantly reduced in the tongue epithelium of mice receiving a HSD (Figure 4A). Mice on HSD displayed lower PKCα and p-PKCα (phosphorylated PKCα) protein levels in tongue epithelium (Figure 4B), but the expression levels of AMPKα and PKAα were similar to that in mice on ND (Figure S4A). As PKC is required to the normal physiological function of TRPM5,¹⁷ we next detected the direct interaction between TRPM5 and PKCα using immunofluorescence and immunoprecipitation. We observed an obvious colocalization of TRPM5 and PKCα in mouse tongue epithelium (Figure 4C). Immunoprecipitation test also showed a strong interaction between TRPM5 and p-PKCα, which was reduced by high-salt administration (Figure 4D). In isolated tongue epithelium, high NaCl also significantly inhibited PKC activity (Figure 4E). In HEK-293 cells, high salt inhibited PKCα mRNA expression through increasing sodium content but not chloride or osmotic pressure (Figure S4B). And NaCl did not affect the mRNA stability of PKCα either (Figure S4C). Intracellular sodium concentration was significantly elevated by PMA (phorbol 12-myristate 13-acetate), a membrane permeable PKC activator (Figure 4F), suggesting that PKC activation would activate TRPM5 channel. Consistently, the inward and outward currents during the inactivating phase in TRPM5-expressed HEK-293 cells was activated by PMA and blocked by a PKC inhibitor bisindolylmaleimide I (Figure S4D through S4F), but not affected by AMPK or PKA (Figure S4G through S4J). In addition, the inhibitory effect of 3-hour high-salt exposure on opening frequency of TRPM5 channel was also augmented by PMA (Figure S4K). These findings suggest the following: (1) chronic high-salt intake could desensitize PKCα and TRPM5 in tongue epithelium; (2) stimulating PKC could reactivate TRPM5 channel.

• Download figure
• Download PowerPoint

PKC is a family of protein kinase enzymes involved in controlling a series of cell behaviors by phosphorylating their target proteins.²⁶ TRPM5 shares many similarities with TRPM4 in structure and function, for they are both voltage-dependent sodium channels and play important roles in taste conduction.²⁷ As TRPM4 has been reported to be phosphorylated by PKC,²⁸ we asked whether the activity of TRPM5 could be altered by phosphorylation. We found 2 putative PKC motifs centered on Ser33 and Thr1052 in TRPM5 protein, which were relatively conserved in humans and rodents (Figure S5A). Based on the predicted results, we constructed the corresponding point mutation TRPM5 plasmids (S33A and T1052A) and tested the effect of PKC activation on these mutant TRPM5 channels by applying PMA. Preincubation with 1 mmol/L PMA for 1 hour significantly increased TRPM5 currents. Compared with WT controls, the effect of PMA was weakened in cells expressing the mutants T1052A, but not mutant S33A (Figure S5B through S5D). Furthermore, our immunoprecipitation test showed that the threonine phosphorylation level on TRPM5 was reduced in the tongue epithelium of mice on HSD (Figure S5E). These findings indicate that PKC-dependent T1052 threonine phosphorylation on TRPM5 was inhibited by long-term high-salt stimulation, which might be responsible for the impaired channel function.

TRPM5-Mediated Salty Perception Involved Central Gustatory Cortex

Taste receptor cells on the tongue transmit signal through multiple neural stations to the primary gustatory cortex,²⁹ where each taste is represented in distinct area.³⁰ Thus, we used fiber fluorometry to examine whether TRPM5 channel activity in tongue epithelium was associated with neural response in the primary taste cortex (Figure S6A). Consistent with a previous study,³¹ the bitter cortical areas in primary taste cortex detected by the optical fiber was exactly showed by Oregon Green 488 BAPTA-1 distribution in the brain sections (Figure S6B). The calcium signal in the bitter cortical field was predominately evoked in anesthetized mice by bitter tastant as well as high-salt solutions containing amiloride, which was used to selectively exclude the contribution of the ENaC-dependent low-salt pathway³² (Figure S6C and S6D). The amplitude of calcium wave evoked by bitter tastant was enhanced by high-salt solutions with amiloride and vice versa (Figure S6E and S6F). And inhibition of TRPM5 by TPPO on the tongue epithelium dramatically decreased the neural activity of bitter cortical field (Figure S6C through S6F), indicating that the TRPM5-mediated high-salt taste perception was manufactured by neurons in the cortical field. Also, high-salt-evoked neural activity was enhanced by cucurbitacin E, and inhibited by TPPO (Figure S6G and S6H), suggesting that high salt could elicit a bitter taste perception through TRPM5 channels in tongue epithelium. Next, we compared the high-salt-evoked neural activity in WT mice on ND or HSD and TRPM5^−/− mice. In comparison with mice on a ND, the amplitude of calcium signals evoked by high salt, with or without cucurbitacin E, in the bitter cortical field was significantly lower in mice on an HSD (Figure S6I). As expected, in TRPM5^−/− mice, high salt failed to evoke higher amplitude of calcium signals than water control (Figure S6I), suggesting that long-term high-salt intake impairs aversive taste perception via a TRPM5-dependent pathway.

Long-Term High-Salt Dietary Intervention Impaired Aversive Behavior and Enhanced High-Salt Consumption

Next, we examined whether the impaired TRPM5 function induced by high-salt dietary intervention affected the aversive behavior and high-salt consumption in mice. The body weight, systolic blood pressure (SBP) and heart rate were comparable among the mice on ND or HSD (Table S1). Compared with mice on ND, mice on HSD had reduced aversive response to high-salt solutions and bitter tastant (Figure 5A), but without any significant alteration in sweet and umami taste (Figure S7A and S7B). When fed with a mixed salt fodder containing 0.25%, 1%, 4%, and 8% NaCl, mice on HSD showed increased preference for 1% and 4% NaCl, whereas mice on ND preferred to fodder with a much lower salt content (Figure 5B). Meanwhile, the food intake was comparable among the 2 groups (Figure 5C). Both sodium intake and urinary sodium excretion were higher in mice on HSD than in mice on ND (Figure 5D and 5E) without affecting the ratio of sodium excretion/sodium intake (Figure 5F). Water intake and urine volume were also higher in mice on HSD than in mice on ND (Figure S7C and S7D).These results indicate that long-term high-salt intake impaired aversive behavior and enhanced high-salt consumption in mice.

• Download figure
• Download PowerPoint

Mice With Mutant TRPM5 Preferred High Salt and Developed Hypertension

To further confirm whether the reduced TRPM5 expression by high salt was critical to the enhanced salt consumption, the high salt–induced aversive response and daily salt intake were examined in TRPM5 knockout (TRPM5^−/−) mice fed with HSD. Brief licking response test revealed that TRPM5^−/− mice displayed a dramatically enhanced tolerance for high-salt solution compared with wild-type (WT) littermates (Figure 6A). Similarly, the results from 2-bottle test also showed a reduced aversion to high-salt solution in TRPM5^−/− mice (Figure 6B). The food and salt intake as well as urinary sodium excretion were obviously enhanced in TRPM5^−/− mice on HSD without change in sodium excretion/sodium intake ratio (Figure 6C and 6D). Both water intake and urine volume in TRPM5^−/− mice on HSD were significantly higher than WT mice on HSD (Figure S8A). Moreover, high-salt diet increased the blood pressure in TRPM5^−/− mice from 2 weeks to the end of experiment, but not in WT mice (Figure 6E). And there was no difference in heart rate between the 2 groups (Figure S8B). Twenty-four hour monitoring revealed a higher blood pressure throughout the circadian cycle in TRPM5^−/− versus WT mice on HSD (Figure 6F). The ratio of heart to body weight was also elevated by knockout of TRPM5 (Figure S8C). Echocardiography also demonstrated that knockout of TRPM5 promoted high salt–induced cardiac hypertrophy, as evidenced by a lower LVID, higher LVPW and IVST thickness in TRPM5^−/− mice (Figure S8D through S8G). These results suggest that the reduced TRPM5 expression by high-salt intake impaired the aversive response to high salt thus further aggravated high-salt consumption and subsequently lead to high salt–induced hypertension.

• Download figure
• Download PowerPoint

High-Salt Intake Impaired Bitter Taste Perception in Hypertensive Patients

Previous studies showed that Momordica charantia, a bitter melon extract, could reduce blood pressure in Dahl salt sensitive or Nω-nitro-l-arginine-methyl ester-induced hypertensive rats.³³ We recruited 114 hypertensive patients. In comparison to patients with low-salt intake (n=58), patients with high-salt intake had higher threshold of intolerability (P=0.037; Figure S9A). Importantly, patients with high-salt intake also had less-sensitive perception of bitterness (P=0.042 versus subjects with low salt intake; Figure S9A). All patients were further divided into low and high bitter sensitive perception groups. Patients with high bitter perception had lower salt intake (8.5±4.5 versus 10.3±4.4 g/day in subjects with low bitter perception; P=0.039; Figure S9B). There was also a tendency of higher systolic blood pressure in patients with low bitter perception (145±21 versus 141±23 mm Hg in patients with high bitter perception; P=0.334; Figure S9B). Moreover, patients with high-salt intake and low bitter perception had a higher blood pressure (systolic, 147±20 mm Hg; diastolic, 78±15 mm Hg; Figure S9B). These findings indicated that hypertensive state might be associated with impaired bitter taste perception-mediated salt intake increase.

Discussion

Excessive salt intake is a well-known risk factor for the development of hypertension. As exposure to sodium determines salt taste preferences,³⁴ people with a high-salt intake generally tend to prefer salt-rich food possibly, thus results in an overall failure of ordinary strategies focusing on reducing salt intake.⁴ In the current study, we present evidence to prove that the impairment of TRPM5 but not ENaC in the taste sensing cells on tongue epithelium is a chief culprit accounting for the reduced aversive response and enhanced salt intake in mice under high-salt loading. Similarly, our clinical study also showed that hypertensive patients with high-salt intake displayed a lower perception of bitter taste, suggesting that their aversion to noxious salt stimulation could be impaired (Figure S9C). Taken together, these findings indicate that impairment of TRPM5-dependent bitter taste mediates the enhancement of salt intake, thus facilitates the development of hypertension.

Salt intake is determined by salt taste preference. Several factors affect the salt taste preference, such as the salt deficiency in food, difference of cultural background,³⁵ the inadequate intake of animal protein,³⁶ and so on. In modern society, the extensive application of salt in food processing makes daily salt intake in human easily exceed the standard. Although increased salt taste preference was reported in hypertensive patients,⁸ this result was not confirmed by other studies.³⁷ Our clinical study also revealed the unaltered salt taste preference between hypertensive individuals and healthy controls. In addition, there are controversy reports on the correlation between salt preference and salt intake. One study showed that there was no correlation between salt taste preference and 24-hour urinary sodium excretion that reflects individual daily salt intake level,³⁸ whereas another population survey showed that individuals exhibited an increased salt taste preference level following increases in salt consumption.³⁹ After artificially reducing salt intake, salt preference also declined in human.⁴⁰ Currently, the molecular mechanisms of salt taste preference are involved in participation of sodium channel and TRP channel in tongue epithelium.^14,41 Different concentrations of salt solution can cause 2 distinct behavioral reactions. Low concentration of NaCl can produce salt preference, which is mediated by the ENaC in tongue epithelium,¹³ but high concentration of NaCl can evoke a strong aversive reaction in which TRPM5-mediated bitterness perception participates because TRPM5 knockout mice blunted their aversion to high concentration of salt solution.¹⁴ However, the effects of long-term high-salt intake on salt preference and aversion remains elusive. Although high-salt intake can enhance the expression and activity of ENaC in kidney,⁴² few study has investigated the effect of high-salt intake on the alterations of ENaC and TRPM5 in tongue epithelium. In this study, we provided ample evidence to prove that long-term high-salt intake significantly reduced TRPM5 expression and impaired its mediated aversion to noxious high-salt stimulation, but did not alter the ENaC expression and its mediated salt preference.

TRPM5 is a Ca²⁺-activated cation channel that mediates bitter, sweet, and umami tastes in type II taste cells.^20,43 TRPM5 is the final element in the signaling cascade activating G protein-coupled receptors that requires PLCβ2.^16,44 The second messengers diacylglycerol and inositol triphosphate are direct targets of PLC signaling, which activate PKC and mobilize calcium release from intracellular calcium stores, respectively.¹⁶ Intracellular Ca²⁺ opens TRPM5 channels and leads to sodium influx and cell depolarization.⁴⁵ TRPM5 channel is also regulated by PKC in a Ca²⁺-independent manner.¹⁷ In this study, both the expression and activity levels of TRPM5 or PKCα were dose-dependently inhibited by NaCl in tongue epithelium. Changes of TRPM5 and PKCα in response to NaCl exposure were accompanied by decreased capacity of both Na⁺ and Ca²⁺ uptake, suggesting that the reduced uptake of Ca²⁺ contributes the reduced PKC or TRPM5 activity. Besides depolarizing receptor cells, Na⁺ influx through TRPM5 might also promote Ca²⁺ uptake through Na⁺/Ca²⁺ exchange.^46,47 Therefore, inhibition of TRPM5 further decreases Ca²⁺ influx to cytoplasm and retrospectively suppresses PKCα activity, thus forming a vicious cycle. Similar to some reports performed in other cell types,^48,49 we also observed a lower intracellular Ca²⁺ level in tongue epithelial cells on high-salt loading, possibly because of the reduction of TRPM5-mediated Na⁺ influx. These results suggest that the breakage of cellular cation homeostasis by high salt would be a fundamental step inhibiting PKCα. In addition, since salt produces a variety of effects on the structure and function of enzymes,⁵⁰ it might also directly affect PKCα activity. Furthermore, we also observed a colocalization and physical interaction between TRPM5 and PKCα in the tongue epithelium, suggesting that PKCα might contribute to regulate TRPM5 activity. Mechanistically, we observed that the threonine phosphorylation level of TRPM5 was reduced in mice on HSD and the PKC-dependent threonine phosphorylation site T1052A mutant significantly inhibited TRPM5 channel activity. Thus, we speculate T1052 might be one of the target site that influenced by subdued PKC activity in high sodium circumstance, which warrants further investigation.

Stimuli to taste receptor cells in the tongue are transmitted to corresponding areas in the primary gustatory cortex.²⁹ Bitter taste mediated by TRPM5 is represented in the primary gustatory cortex, with taste quality encoded by distinct cortical fields.³⁰ Neuronal activity in the primary cortex mediates the perception of taste and drives associated behaviours.³¹ In our previous study, we demonstrated that spicy flavor enhances salt taste perception and could reduce salt intake and blood pressure.¹⁹ In the current study, we showed that TRPM5-mediated bitter taste is handled by neurons in this cortical field, and acute high-salt stimulation recruits a bitter taste perception in this field, which could be impaired by long-term high-salt intake.

Bitter taste is a signal that warns against potentially noxious chemicals.¹² From an evolutionary viewpoint, bitter taste serves to protect an organism from plant-based poisons.⁵¹ However, distaste for bitterness can be modulated or even reversed for selected foods through experience or conditioning.⁵²M. charantia, commonly known as bitter melon, is cultivated worldwide and used as a culinary vegetable as well as folk medicine. Bitter melon is rich in bioactive chemicals, including cucurbitane type triterpenoids, triterpene glycosides, phenolic acids, flavonoids, and essential oils. Recent experimental and clinical studies show that bitter melon extracts have potential therapeutic benefit in cardiometabolic diseases.⁵³ In our experiment, cucurbitane E, a bitter compound isolated from bitter melon, and chlorogenic acid, an intense bitter-tasting constituent in coffee, possess potent activation on TRPM5 channel. We further showed that long-term high-salt intake attenuates the TRPM5-mediated bitter taste perception and aversive response for high salt in mice. Thus, it is rationale to propose that application of such dietary bitter food might benefit the TRPM5-mediated aversive response to high salt and antagonize excessive salt intake.

Study Limitations

Our study has several limitations. Although 8% high-salt intake has been extensively applied in previous studies, it is hard to mimic the high-salt intake in human. Therefore, exploring the appropriate high-salt intervention in human deserves further investigation. In addition, it would be better to further validate our conclusion using tongue epithelial TRPM5 conditional knockout mice because of weakness of systemic gene deletion.

Perspectives

The current study suggested that long-term high-salt consumption impairs aversive response to concentrated salt by downregulating TRPM5 and attenuating response to bitter tastant. Mechanistic investigation suggested a critical role of PKCα and PKC-dependent threonine phosphorylation on TRPM5. This might be an alternative explanation why people prefer to salty food besides well-known environmental factors. We propose that restoring TRPM5 function would represent a novel strategy to antagonize excessive salt intake and salt sensitive hypertension.

Acknowledgments

We thank Prof. Bernd Nilius (Department of Cell Molecular Medicine, Laboratory Ion Channel Research, Campus Gasthuisberg, Leuven, Belgium) and Dr Yu Huang (Institute of Vascular Medicine and Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong, China) for their critical review of the manuscript. We thank Hongbo Jia, Lijuan Wang, and Tingbing Cao for their technical assistance.

Footnotes