Home | Help | Feedback | Subscriptions | Archive | Search | Table of Contents

This Article Abstract PDF (Full Text) PPT slides of all figures Alert me when this article is cited Citation Map Services Email this article Similar articles in this journal Similar articles in PubMed Alert me to new content in the JGP Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Lyall, V. Articles by DeSimone, J. A. Search for Related Content PubMed PubMed Citation Articles by Lyall, V. Articles by DeSimone, J. A. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB CARBON DIOXIDE D-MANNITOL HYDROCHLORIC ACID Social Bookmarking                      What's this?

¹ Department of Physiology, Virginia Commonwealth University, Richmond, VA 23298
² Department of Internal Medicine, Virginia Commonwealth University, Richmond, VA 23298

Correspondence to Vijay Lyall: vlyall{at}hsc.vcu.edu

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The relationship between cell volume and the neural response to acidic stimuli was investigated by simultaneous measurements of intracellular pH (pH_i) and cell volume in polarized fungiform taste receptor cells (TRCs) using 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF) in vitro and by rat chorda tympani (CT) nerve recordings in vivo. CT responses to HCl and CO₂ were recorded in the presence of 1 M mannitol and specific probes for filamentous (F) actin (phalloidin) and monomeric (G) actin (cytochalasin B) under lingual voltage clamp. Acidic stimuli reversibly decrease TRC pH_i and cell volume. In isolated TRCs F-actin and G-actin were labeled with rhodamine phalloidin and bovine pancreatic deoxyribonuclease-1 conjugated with Alexa Fluor 488, respectively. A decrease in pH_i shifted the equilibrium from F-actin to G-actin. Treatment with phalloidin or cytochalasin B attenuated the magnitude of the pH_i-induced decrease in TRC volume. The phasic part of the CT response to HCl or CO₂ was significantly decreased by preshrinking TRCs with hypertonic mannitol and lingual application of 1.2 mM phalloidin or 20 µM cytochalasin B with no effect on the tonic part of the CT response. In TRCs first treated with cytochalasin B, the decrease in the magnitude of the phasic response to acidic stimuli was reversed by phalloidin treatment. The pH_i-induced decrease in TRC volume induced a flufenamic acid–sensitive nonselective basolateral cation conductance. Channel activity was enhanced at positive lingual clamp voltages. Lingual application of flufenamic acid decreased the magnitude of the phasic part of the CT response to HCl and CO₂. Flufenamic acid and hypertonic mannitol were additive in inhibiting the phasic response. We conclude that a decrease in pH_i induces TRC shrinkage through its effect on the actin cytoskeleton and activates a flufenamic acid–sensitive basolateral cation conductance that is involved in eliciting the phasic part of the CT response to acidic stimuli.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Sour taste is a primary taste modality and is uniquely associated with acidic stimuli. It evokes an innate rejection response to extremely acidic or sour stimuli, limiting the ad libitum ingestion of acid (Beauchamp et al., 1991; Scott and Plata-Salaman, 1991). In this regard, along with the lungs and kidneys, which are the primary organs for acid secretion, acid sensing taste receptor cells (TRCs) in the oral cavity, by restricting the intake of acid, function as part of a multi-organ system to maintain acid–base homeostasis. Acids in aqueous solution yield H⁺, which can alter both the extracellular (pH_o) and intracellular pH (pH_i) of cells. A large number of enzymatic reactions, channels, transporters, and intracellular signaling events are modulated by changes in pH_i. It is, therefore, a special challenge to identify the specific channels, transporters, and other intracellular signaling events in TRCs that sustain acid taste transduction and ultimately the cortical events resulting in sour taste sensation. In spite of this complexity, some progress has been made in understanding the underlying mechanisms involved in sour taste transduction at the level of TRCs.

Recent studies indicate that the proximate signal for sour taste transduction is an acid-induced decrease in pH_i in a subset of TRCs (Lyall et al., 2001). In the case of strong acids, an apical adenosine 3'5'-cyclic monophosphate (cAMP)–sensitive, but Ca²⁺-insensitive H⁺ conductance allows H⁺ entry and the subsequent decrease in TRC pH_i (Lyall et al., 2002a; 2004a). However, H⁺-gated channels, such as the acid sensing ion channel (ASIC) in the apical membrane and both ASIC (Ugawa et al., 1998; Lin et al., 2002) and hyperpolarization-activated channels (HCN) (Stevens et al., 2001) in the basolateral membrane of TRCs, and TASK-2, a two pore domain K⁺ channel (Lin et al., 2004; Richter et al., 2004a) may also play a role in sour taste transduction. These channels in the basolateral membrane could be activated if H⁺ can cross tight junctions and decrease pH in the basolateral compartment. However, recent studies suggest that in mice ASIC2 is not required for acid taste (Richter et al., 2004b). In contrast, weak organic acids do not seem to have apical membrane receptors. They enter TRCs across the apical membrane by passive diffusion as lipid soluble undissociated neutral molecules. Once inside the cell they dissociate to generate H⁺ and decrease TRC pH_i (Lyall et al., 2001). TRCs also contain 5-nitro-2-(3-phenylpropylamino)-benzoate (NPPB)-sensitive stretch-activated Cl^– channels (Gilbertson, 2002). The presence of an NPPB-sensitive Cl^– channel activated by acid was demonstrated in mouse taste cells (Miyamoto et al., 1998). These results suggest a role for Cl^– channels in acid taste transduction. In addition to Cl^– channels, there is also evidence that NPPB-insensitive poorly selective cationic conductance in the apical receptive membranes of mouse TRCs may also be involved in sour taste transduction (Miyamoto et al., 1998). The above data suggest that sour taste transduction is mediated via multiple pathways. Similarly, in the case of other acid-sensitive cells, such as the central chemosensitive neurons, a multiple factors model has been proposed for acid signaling (Putnam et al., 2004).

Recent studies in our laboratory have focused on identifying specific cellular mechanisms that determine the neural response profiles to acidic stimuli. Similar to other taste stimuli, acidic stimuli elicit chorda tympani (CT) responses that are composed of two components with distinct temporal characteristics: a rapid transient increase in the phasic neural response that slowly declines to a quasi-steady state, defined as the tonic phase of the neural response. Treating rat tongue with specific membrane-permeable blockers of carbonic anhydrases (MK-417 or MK-507) inhibited both the phasic and tonic components of the CT response to CO₂ (Lyall et al., 2001, 2002b). This indicates that an acid-induced decrease in TRC pH_i is necessary to elicit both the phasic and the tonic components of the CT response to acid stimulation. In a subset of TRCs a decrease in pH_i is followed by an increase in intracellular Ca²⁺ ([Ca²⁺]_i) (Liu and Simon, 2001; Lyall et al., 2003; Richter et al., 2003). An increase in TRC [Ca²⁺]_i activates basolateral Na⁺-H⁺-exchanger-1 (NHE-1) and results in neural adaptation to both strong and weak acid stimuli (Lyall et al., 2002a, 2004a). Neural adaptation is related to the tonic part of the CT response to acidic stimuli. Therefore, a primary decrease in TRC pH_i, followed by a secondary increase in [Ca²⁺]_i, and the subsequent activation of basolateral NHE-1 are specific events related to the tonic part of the neural response only. However, at present a major unsolved problem in acid taste transduction is: how is the decrease in pH_i related to the phasic part of the neural response to acid stimulation?

In this study, we tested the hypothesis that a decrease in pH_i induces a decrease in TRC volume by altering the cell actin cytoskeleton leading to the activation of one or more membrane conductances that are involved in eliciting the phasic part of the CT response to acidic stimuli. This hypothesis is based on our previous studies in which hypertonic solutions of mannitol or cellobiose (Lyall et al., 1999) or ethanol (Lyall et al., 2005a,b) induced cell shrinkage and elicited a transient phasic CT response. This suggests that a decrease in cell volume is an intracellular signal for eliciting a transient neural response. Second, since only a transient phasic response was observed in ethanol solutions in the absence of permeable ions on the apical side, it would appear that cell depolarization involves changes in a basolateral membrane ion conductance or conductances uniquely associated with transient phasic responses (Lyall et al., 2005a,b). To test this hypothesis, we investigated the relationship between TRC volume and the CT response to acidic stimuli by simultaneously monitoring pH_i and cell volume in polarized rat fungiform TRCs loaded with 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF) using imaging techniques in vitro, and by rat CT taste nerve recordings in vivo. The results demonstrate that during acid taste transduction a decrease in TRC pH_i induces cell shrinkage through its effect on the cell actin cytoskeleton. A decrease in volume activates a flufenamic acid–sensitive nonselective cation conductance in the basolateral membrane of TRCs that is involved in eliciting the phasic part of the CT response to acidic stimuli. Some of the data have appeared in abstract form (DeSimone et al., 2005).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In Vitro Studies
Simultaneous Measurement of TRC Volume and pH_i Using BCECF
The animals were housed in the Virginia Commonwealth University animal facility in accordance with institutional guidelines. All in vitro and in vivo animal protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of Virginia Commonwealth University. Female Sprague-Dawley rats (150–200 g) were anesthetized by exposing them to the inhalation anesthetic, isoflurane (1.5 ml), in a desiccator. When rats were fully unconscious, a midline incision was made in the chest wall and the aorta severed. The tongues were then rapidly removed and stored in ice-cold Ringer's solution (R; Table I). The lingual epithelium was isolated by collagenase treatment. A small piece of the anterior lingual epithelium containing a single fungiform papilla was mounted in a special microscopy chamber as described earlier (Lyall et al., 2001).

Measurement of TRC [Ca²⁺]_i
Relative changes in [Ca²⁺]_i were monitored in polarized TRCs by loading the tissue with Fura-2-AM (Molecular Probes). The method for loading Fura-2 and recording temporal changes in FIR (F₃₄₀/F₃₈₀) was essentially similar to that used earlier for measuring [Na⁺]_i changes with SBFI (Lyall et al., 2002b, 2005a).

Solutions
The composition of the various solutions used in the in vitro experiments is given in Table I. However, in some experiments, the control solution (C) was made hypertonic by increasing the NaCl concentration from 150 to 500 mM. In some experiments, the NH₄Cl concentration was varied between 0 and 25 mM. To maintain constant osmolarity, an equivalent amount of NaCl or NMDGCl was replaced with NH₄Cl. Some solutions contained the following drugs: ionomycin (a Ca²⁺ ionophore; 10 µM), nigericin (a K⁺-H⁺ exchanger; 10 µM), phalloidin (F-actin modifier; 10 µM), cytochalasin B (G-actin modifier; 20 µM). All drugs were purchased from Sigma-Aldrich and were dissolved in DMSO. The stock solutions were then mixed with appropriate solutions to achieve the final concentration of the drugs used in the experiments.

Data Analysis
Changes in TRC pH_i were expressed as the mean ± SEM of n; where n represents the number of ROIs within the taste bud, M ± SEM (n). The changes in fluorescence intensity at 440 nm were expressed relative to the fluorescence intensity (F₄₄₀) under control conditions. The F₄₄₀ under control conditions for each ROI was taken as 100%. The data were also presented as the mean ± SEM from different tissue preparations (N). In this case N represented the number of polarized lingual preparations studied. Student's t test was employed to analyze the differences between sets of data.

Labeling of F- and G-actin
Fungiform taste bud fragments were harvested from isolated lingual epithelia as described before (Vinnikova et al., 2004). The isolated taste buds were placed on a CellTak (Sigma-Aldrich)-coated coverslip that formed the bottom of a perfusion chamber and were perfused with Ringer's solution (R; Table I; pH 7.4). The isolated TRCs were triple loaded with bovine pancreatic deoxyribonuclease (DNase-1) conjugated with Alexa Fluor 488, 4',6-diamidino-2-phenylin (DAPI), and rhodamine phalloidin (all from Molecular Probes) at room temperature. DNase 1 binds to monomeric G-actin, phalloidin selectively labels polymerized F-actin, and DAPI selectively binds DNA and is a nuclear stain. TRCs were treated with 0.5 µM DAPI, 10 U of rhodamine phalloidin (1.6 µM), and 0.3 µM DNase 1. DAPI and rhodamine phalloidin were dissolved in DMSO. The stock solution was then mixed with control solution (C; Table I; pH 7.4) to give the final concentration of 0.5 and 1.6 µM, respectively. A stock solution (161 µM) of DNase-1 conjugated with Alexa 488 was made in a solution containing 20 mM Tris (pH 7.6) + 50 mM NaCl + 50% glycerol + 1 mM dithiothreitol and 0.1 mg/ml bovine serum albumin and stored at –20°C. The stock solution was then mixed with control solution to give the final concentration of 0.3 µM. The labeled nuclei, F-actin, and G-actin were imaged using an LSM 510 Meta confocal microscope (Carl Zeiss MicroImaging, Inc.). For DAPI the excitation wavelength was 405 nm and the emitted light was detected using a bandpass filter (420–480 nm). For rhodamine phalloidin, the excitation wavelength was 543 nm and the emitted light was detected using a bandpass filter (550–600 nm). For DNase-1 conjugated with Alexa 488 the excitation wavelength was 488 nm and the emitted light was detected using a bandpass filter 505–530 nm. The isolated taste bud fragments were visualized by taking optical sections in the Z-axis when perfused with control solution at pH 7.4 and 6.7 (C; Table I), and in the presence of 10 µM cytochalasin B (pH 7.4) or 500 mM mannitol (pH 7.4).

In Vivo Studies
CT Nerve Recordings
Female Sprague-Dawley rats (150–200 g) were anesthetized by intraperitoneal injection of pentobarbital (60 mg/kg), and supplemental pentobarbital (20 mg/kg) was administered as necessary to maintain surgical anesthesia. The animal's corneal reflex and toe-pinch reflex were used to monitor the depth of surgical anesthesia. Body temperatures were maintained at 37° with a Deltaphase Isothermal PAD (Model 39 DP; Braintree Scientific, Inc.). The left CT nerve was exposed laterally as it exited the tympanic bulla and placed onto a 32G platinum/iridium wire electrode. An indifferent electrode was placed in nearby tissue. Neural responses were differentially amplified with an optically coupled Isolated Bio-Amplifier (ISO-80; World Precision Instruments). For display, responses were filtered using a band pass filter with cutoff frequencies 40 Hz–3 kHz and fed to an oscilloscope. Responses were then full-wave rectified and integrated with a time constant of 1 s. Integrated neural responses and current and voltage changes were recorded on a chart recorder and also captured on disk using Labview software (National Instruments) and analyzed offline. Stimulus solutions were injected into a Lucite chamber (3 ml; 1 ml/s) affixed by vacuum to a 28 mm² patch of anterior dorsal lingual surface. The chamber was fitted with separate Ag-AgCl electrodes for measurement of current and potential and served as inputs to a voltage-current clamp amplifier that permitted the recording of neural responses with the chemically stimulated receptive field under zero current clamp or voltage clamp. The clamp voltages were referenced to the mucosal side of the tongue. Neural responses were recorded and analyzed as described before (Lyall et al., 2005a,b).

Solutions
The composition of the various rinse and stimulating solutions used in the CT experiments is given in Table II. CT responses to 300 mM NH₄Cl, 300 mM NaCl, 100 mM NaCl, 1 M mannitol (M), 10 mM quinine (Q), and 20 mM HCl were recorded with reference to 10 mM KCl rinse. In some experiments, NaCl stimulating solutions contained 5 µM benzamil (Bz; a specific blocker of the epithelial sodium channel; ENaC). The tongue was stimulated with dissolved CO₂ at pH 7.4 (72 mM KHCO₃ + 10% CO₂). The corresponding rinse solution contained the same concentration of KCl as KHCO₃ and was buffered to pH 7.4 with HEPES. The tongue was stimulated with acetic acid solutions in which 10 mM acetic acid was buffered to pH 6.1 with K-acetate. The corresponding rinse solution contained the same concentration of KCl as K-acetate and was buffered to pH 6.1 with HEPES. In some experiments both rinse and acid stimuli contained, in addition, 0.5 or 1.0 M mannitol. CT responses to the above acidic stimuli were also monitored after topical application of flufenamic acid (40 µM), a specific blocker of shrinkage-activated nonselective cation channels (Koch and Korbmacher, 2000) and after application of G- and F-actin modifiers, cytochalasin B (20 µM) and phalloidin (1.2 mM), respectively. The above compounds were dissolved in DMSO and applied topically on the tongue for 30–45 min. In some experiments TRCs were loaded in vivo with a Ca²⁺ chelator, 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid tetrakis-(acetoxymethyl ester) (BAPTA-AM; Sigma-Aldrich). BAPTA-AM (30 mM) was dissolved in DMSO and applied topically to the tongue for 45 min.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In Vitro Studies
We first investigated the relationship between pH_i and cell volume in polarized TRCs using imaging techniques in vitro.

Effect of Increasing Basolateral Solution Osmolarity on TRC pH_i and Volume
In TRCs loaded with BCECF (Fig. 1 A), perfusing the basolateral membrane with 500 mM NaCl produced a rapid but transient increase in F₄₄₀ (a–b; dotted line), followed by a spontaneous decrease to below its resting value (b–c). Upon reperfusing control solution there was a rapid transient decrease in F₄₄₀ (c–d), followed by a spontaneous increase to a value slightly above the resting level (d–e). Increase in F₄₄₀ indicates decrease in cell volume. A decrease in cell volume will result in an increase in dye concentration inside the cells and increase in F₄₄₀ (a–b). The spontaneous recovery of F₄₄₀ (b–c) suggests that TRCs are capable of regulatory volume increase (RVI). A decrease in F₄₄₀ as a response to lowering NaCl concentration from 500 to 150 mM is consistent with an increase in TRC volume. An increase in cell volume will result in a decrease in dye concentration inside the cells and a decrease in F₄₄₀ (c–d). The spontaneous recovery of F₄₄₀ (d–e) suggests that TRCs are capable of regulatory volume decrease (RVD).

View larger version (43K): [in this window] [in a new window]   Figure 1. Effect of basolateral NaCl, NH4Cl, and Na-acetate on TRC pHi and volume. Lingual epithelia loaded with BCECF were initially perfused on both sides with control solution containing 150 mM NaCl (pH 7.4). During the time period shown by the top horizontal bar the basolateral membrane of polarized TRCs was perfused with (A) control solution containing 500 mM NaCl, (B) control solution containing 15 mM NH4Cl, (C) control solution containing 15 mM Na-acetate at pH 7.4, or (D) with Na+-free solution (0 NaBl) containing 150 mM NMDG-Cl or with Na+-free solution containing 135 mM NMDG-Cl + 15 mM NH4Cl (pH 7.4). Temporal changes in TRC pHi and volume were monitored as changes in FIR (F490/F440, solid line) and florescence intensity of the isosbestic wavelength 440 nm (F440, dotted line), respectively. The F440 under control conditions for each ROI was taken as 100%. Values are presented as mean ± SEM of n, where n = number of ROIs within the taste bud.

Effect of Weak Organic Acids and Bases on TRC pH_i and Volume
At constant pH_o and osmolarity, exposing the basolateral membrane of TRCs to weak organic acids or bases induces transient changes in pH_i that recover spontaneously (Vinnikova et al., 2004). We hypothesize that at constant pH_o and osmolarity, changes in TRC pH_i will be accompanied by parallel changes in cell volume. Changes in pH_i were induced by exposing the basolateral membrane of polarized TRCs to short pulses of NH₄Cl, Na-acetate, or CO₂.

Exposing the basolateral membrane to 15 mM NH₄Cl (NH₄Cl; Table I; pH 7.4) produced a rapid intracellular alkalinization (Fig. 1 B, a–b, solid line). NH₃ diffuses across the basolateral membrane into TRCs and binds to free intracellular H⁺ to form NH₄⁺ and increases pH_i. This was followed by a spontaneous decrease in pH_i (b–c) due to the slower entry of NH₄⁺ across the basolateral membrane and its dissociation to NH₃ + H⁺. In addition, one or more pH compensatory mechanisms in TRC membranes may be involved in assisting the recovery from alkaline pH_i. Upon NH₄Cl washout, there was a rapid decrease in pH_i to below its resting value (Fig. 1 B, c–d). This is due to the rapid exit of intracellular NH₃ and the conversion of intracellular NH₄⁺ to NH₃ + H⁺. The loss of NH₃ from the cells results in the accumulation of excess H⁺ inside the cells. The decrease in pH_i was transient and recovered spontaneously toward its resting value (Fig. 1 B, d–e). The alkalinization phase of the NH₄Cl pulse (a–b) was associated with a decrease in F₄₄₀ (Fig. 1 B, f–g, dotted line), indicating cell swelling. The spontaneous recovery of pH_i during the NH₄Cl pulse (b–c) was accompanied by a parallel recovery of cell volume (g–h). Upon NH₄Cl washout, pH_i decreased below its resting value and was accompanied by a rapid increase in F₄₄₀ (Fig. 1 B, h–i) that also increased (i) above its resting value (f), indicating rapid cell shrinkage. During spontaneous pH_i recovery (d–e) cell volume slowly increased toward its control value (i–j). The initial changes in pH_i, represented by a–b, b–c, c–d, and d–e, were linearly related to changes in F₄₄₀ (mean r² = 0.97 ± 0.04; n = 6). NH₄Cl pulses produced a similar relationship between pH_i and F₄₄₀ in five additional TRC preparations (see also Fig. 1 D and Fig. 2 A below).

View larger version (39K): [in this window] [in a new window]   Figure 2. Effect of basolateral NH4Cl concentration and pH on TRC pHi and volume. (A) A lingual epithelium loaded with BCECF was initially perfused on both sides with Na+-free solution containing 150 mM NMDG-Cl (pH 7.4). All basolateral solutions contained, in addition, 10 mM BaCl2, and 2 mM CPC was added to the apical solution. Temporal changes in TRC pHi and volume were monitored as changes in FIR (F490/F440, solid line) and fluorescence intensity of the isosbestic wavelength 440 nm (F440, dotted line) following exposure of the basolateral membrane to 25, 20, 15, 10, 5, and 1 mM NH4Cl (equivalent amount of NMDG-Cl replaced NH4Cl in the basolateral solutions). The F440 under control conditions for each ROI was taken as 100%. Inset, relationship between TRC pHi and percent change in F440 relative to 0 NH4Cl. Line of best fit was represented by the following equation: % change in F440 = –36.3 x pHi + 343.9; r2 = 0.99. Values are presented as mean ± SEM of n; where n = number of ROIs within the taste bud. (B) A lingual epithelium loaded with BCECF was initially perfused on both sides with control solution containing 150 mM NaCl (pH 7.4). During the time period shown by the top horizontal bar the basolateral membrane was perfused with control solution at pH 7.8, 6.7, and 7.4. Temporal changes in TRC pHi and volume were monitored as changes in FIR (F490/F440, solid line) and fluorescence intensity of the isosbestic wavelength 440 nm (F440, dotted line), respectively. The F440 under control conditions for each ROI was taken as 100%. Inset, relationship between TRC pHi and percent change in F440 in the initial 100 s between pHi 7.6 and 7.1. Line of best fit was represented by the following equation: % change in F440 = –14.5 x pHi + 203.2; r2 = 0.99. Values are presented as mean ± SEM of n; where n = number of ROIs within the taste bud.

Basolateral Na-acetate caused a rapid increase in F₄₄₀ (f–g, dotted line). In the initial 30-s period both pH_i and F₄₄₀ changed linearly with time (r² = 0.82 ± 0.05; n = 6). This was followed by a partial recovery of both pH_i (b–c) and F₄₄₀ toward baseline (g–h). Upon Na-acetate washout, there was a rapid transient decrease in F₄₄₀ (h–i) followed by partial recovery in cell volume (i–j). Upon Na-acetate washout, in the initial 1-min period, both pH_i and F₄₄₀ changed linearly with time (r² = 0.97 ± 0.04; n =6). In two additional TRC preparations Na-acetate pulses produced a similar relationship between pH_i and F₄₄₀. Similarly, a near linear relationship between TRC pH_i and F₄₄₀ was observed during basolateral CO₂ pulses (unpublished data).

Effect of Basolateral Na⁺ Removal on TRC pH_i and Volume
We hypothesize that at constant pH_o and osmolarity, a decrease in basolateral Na⁺ concentration will produce a decrease in TRC pH_i (Vinnikova et al., 2004) accompanied by a parallel decrease in cell volume.

Perfusing the basolateral membrane with Na⁺-free solution (NMDGCl, Table I, pH 7.4) produced a decrease in TRC pH_i (Fig. 1 D, a–b, solid line) due to the reversal of basolateral NHE-1 (Vinnikova et al., 2004). In two additional TRC preparations basolateral Na⁺-free solution also acidified pH_i and increased F₄₄₀. In the absence of Na⁺, exposing the basolateral membrane to a short NH₄Cl pulse (NMDG/NH₄Cl, Table I, pH 7.4) produced similar changes in pH_i as shown in Fig. 1 B. However, no spontaneous pH_i recovery occurred in the absence of Na⁺ (Fig. 1 D, d–e). This is because in the absence of Na⁺, NHE-1 is inhibited. Subsequently, perfusing with control solution (C; Table I; pH 7.4) produced a rapid increase in pH_i to its control value (e–f) due to the activation of NHE-1 (Vinnikova et al., 2004).

A decrease in pH_i induced by basolateral Na⁺ removal increased F₄₄₀ (Fig. 1 D, g–h, dotted line). Following Na⁺ removal, in the initial 1-min period, both pH_i and F₄₄₀ changed linearly with time (r² = 0.90 ± 0.01; n = 6), indicating that a decrease in pH_i causes cell shrinkage. In the absence of Na⁺, basolateral NH₄Cl pulse initially produced a rapid decrease in F₄₄₀ (h–i) and, upon NH₄Cl washout, increased F₄₄₀ (j) above its value just before the pulse (h). In the absence of Na⁺, no recovery was observed in either pH_i (d–e) or F₄₄₀ (j–k). Na⁺ addition produced pH_i recovery (e–f) and a decrease in F₄₄₀ (k–l). In the initial 2-min period, both pH_i and F₄₄₀ changed linearly with time (r² = 0.92 ± 0.04; n = 6), indicating that an increase in pH_i causes cell swelling.

Effect of Step Changes in Basolateral NH₄Cl Concentration on TRC pH_i and Volume
We hypothesize that at constant pH_o and osmolarity, step changes in basolateral NH₄Cl concentration will produce graded changes in TRC pH_i (Vinnikova et al., 2004) accompanied by parallel graded changes in cell volume. During step changes in NH₄Cl both the pH_i recovery and the NH₄⁺ loss from the cells were minimized. pH_i recovery was inhibited by perfusing the tissue with Na⁺-free solution (compare Fig. 1 D). NH₄⁺ loss through apical VR-1 variant cation channels and basolateral K⁺ channels was minimized by adding 2 mM cetylpyridinium chloride (CPC) and 10 mM BaCl₂ to the apical and basolateral solution, respectively (Vinnikova et al., 2004).

Exposing the basolateral membrane to 25 mM NH₄Cl (NMDG/NH₄Cl, Table I, pH 7.4) induced a rapid increase in mean pH_i from 6.58 to 6.98 (Fig. 2 A, solid line, point a) and decreased F₄₄₀ by 8%, indicating cell swelling. Decreasing the NH₄Cl concentration in a stepwise manner produced a graded decrease in pH_i (at time points b, c, d, e, and f) with minimal spontaneous pH_i recovery and a rapid and graded increase in F₄₄₀. At time points a, b, c, d, e, and f there was a linear relationship between pH_i and F₄₄₀ (Fig. 2, inset; r² = 0.98; n = 8). There was also a linear relationship between pH_i and F₄₄₀ at the time points represented by l, m, n, o, p, and q, during which small, but significant, spontaneous pH_i recovery was observed. This pH_i recovery is due to one or more Na⁺-independent pH recovery mechanisms in TRC membranes. Similar results were obtained in two additional TRC preparations.

Effect of pH_o on TRC pH_i and Volume
We hypothesize that changes in either apical or basolateral pH will induce changes in TRC pH_i (Lyall et al., 2001, 2002b) accompanied by parallel changes in cell volume. Elevating basolateral pH from 7.4 to 7.8 (Fig. 2 B) increased TRC pH_i (a–b, solid line) and cell volume (e–f, dotted line). Lowering basolateral pH from 7.8 to 6.7 decreased pH_i (b–c) and decreased cell volume (f–g). In the initial 100 s, changes in TRC pH_i between 7.6 and 7.1 produced a linear change in F₄₄₀ (Fig. 2 B, inset; r² = 0.99 ± 0.01; n = 4). It should be noted that changes in basolateral pH induced sustained changes in pH_i and cell volume (see also Fig. 6).

The relationship between basolateral pH, TRC pH_i, and cell volume shown in Fig. 2 B was not affected by replacing 50 mM NaCl with 50 mM KCl in the basolateral solution (unpublished data). An increase in basolateral K⁺ concentration is expected to depolarize the membrane potential, suggesting that the relationship between pH_i and cell volume is indifferent to changes in membrane potential. The relationship between basolateral pH, TRC pH_i, and cell volume was also not altered in the absence of Cl^– (unpublished data).

Similarly, during calibration of the pH using high K⁺ solutions plus nigericin (HK; Table I), changes in pH_i between 6.7 and 7.8 also demonstrated a linear relationship with changes in F₄₄₀ (r² = 0.96 ± 0.04; n = 6). In the presence of high K⁺, membrane potential depolarizes to 0 mV and nigericin causes equilibration of pH_i with pH_o. Under these conditions there is no pH gradient, and no pH or volume regulatory mechanisms are operative.

Perfusing the apical membrane with HCl (HCl, Table I, pH 3) induced a sustained decrease in resting TRC pH_i (Fig. 3 A, a–b) and a sustained decrease in cell volume (d–e). In the initial 100 s, changes in TRC pH_i produced a linear change in F₄₄₀ (r² = 0.97 ± 0.01; n = 7). Changes in pH_i (b–c) and volume (e–f) recovered upon reperfusing the apical membrane with control solution (pH 7.4). Similar results were obtained with acetic acid, pH 3.0 (unpublished data).

View larger version (35K): [in this window] [in a new window]   Figure 3. Effect of apical pH on TRC pHi and volume. A lingual epithelium loaded with BCECF was initially perfused on both sides with control solution containing 150 mM NaCl (pH 7.4). (A) During the time period shown by the top horizontal bar, the apical membrane was perfused with an unbuffered solution containing 1 mM HCl (pH 3). (B) During the time period shown by the top horizontal bar, the apical membrane was perfused with a HEPES-free solution in which 72 mM NaCl was replaced with 72 mM NaHCO3. The pH of this solution was adjusted to 7.4 by continuously bubbling with 10% CO2. Temporal changes in TRC pHi and volume were monitored as changes in FIR (F490/F440, solid line) and fluorescence intensity of the isosbestic wavelength 440 nm (F440, dotted line), respectively. The F440 under control conditions for each ROI was taken as 100%. In each case the values are presented as mean ± SEM of n, where n = number of ROIs within the taste bud.

Effect of Changes in [Ca²⁺]_i on TRC pH_i and Volume
Changes in [Ca²⁺]_i modulate TRC pH_i (Lyall et al., 2002a; 2004a). We hypothesize that changes in TRC [Ca²⁺]_i will also induce parallel changes in cell volume. In a polarized TRC preparation loaded with Fura-2, basolateral ionomycin (10 µM) produced a reversible increase in FIR (Fig. 4 A, F₃₄₀/F₃₈₀, a–b) and hence reversibly increased [Ca²⁺]_i. Ionomycin alkalinized resting TRC pH_i (Fig. 4 B, a–b, solid line) and decreased F₄₄₀ (d–e, dotted line), indicating cell swelling. In the initial 100 s following ionomycin treatment there was a linear relationship between the increase in pH_i and the decrease in F₄₄₀ (r² = 0.95 ± 0.05; n = 6). Similarly, upon ionomycin washout, pH_i decreased with an increase in F₄₄₀, indicating cell shrinkage. Ionomycin produced similar effects in two additional TRC preparations. Ionomycin induced intracellular alkalinization by increasing the rate of pH_i recovery from an NH₄Cl pulse (Fig. 4 C). Under control conditions, the mean pH_i recovery rate was 0.038 ± 0.003 pH_i/min (c–d) and was significantly increased to 0.11 ± 0.003 pH_i/min (g–h) after the ionomycin treatment (P < 0.01, n = 10, paired). An increase in [Ca²⁺]_i activates NHE-1 (Lyall et al., 2002a, 2004a).

View larger version (32K): [in this window] [in a new window]   Figure 4. Effect of changes in [Ca2+]i on TRC pHi and volume. (A) A lingual epithelium loaded with Fura-2 was initially perfused on both sides with control solution containing 150 mM NaCl (pH 7.4). During the time period shown by the top horizontal bar, the basolateral membrane was perfused with the control solution containing, in addition, 10 µM ionomycin. Changes in TRC [Ca2+]i were monitored as temporal changes in FIR (F340/F380). (B and C) Lingual epithelium loaded with BCECF was initially perfused on both sides with control solution containing 150 mM NaCl (pH 7.4). During the time period shown by the top horizontal bar the basolateral membrane was perfused with the control solution containing, in addition, 10 µM ionomycin. Temporal changes in TRC pHi and volume were monitored as changes in FIR (F490/F440, solid line) and fluorescence intensity of the isosbestic wavelength 440 nm (F440, dotted line). The F440 under control conditions for each ROI was taken as 100%. In C the spontaneous pHi recovery rates were monitored by the NH4Cl pulse technique before (c–d) and after ionomycin treatment (g–h). In each case the values are presented as mean ± SEM of n, where n = number of ROIs within the taste bud.

View larger version (26K): [in this window] [in a new window]   Figure 5. Labeling of F- and G-actin in an isolated fungiform taste bud fragment at pH 7.4 and 6.7. A confocal image of an optical plane of a fungiform taste bud fragment showing F-actin binding to rhodamine phalloidin (A and F, red), transmitted image (B and G), G-actin binding to pancreatic DNase 1 conjugated with Alexa 488 (C and H, green), staining of TRC nuclei with DAPI (D and I, blue), and an overlay of all images (E and J) are shown while the taste bud was perfused with control solution at pH 7.4 (A–E) or pH 6.7 (F–J).

View larger version (39K): [in this window] [in a new window]   Figure 6. Effect of phalloidin and cytochalasin B on TRC pHi and volume. A lingual epithelium loaded with BCECF was initially perfused on both sides with control solution containing 150 mM NaCl (pH 7.4). (A) At the time period shown by the top horizontal bar, both the apical and basolateral membranes were perfused with control solution containing 10 µM phalloidin (pH 7.4). In B, both sides of the epithelium were exposed to 20 µM cytochalasin B for 20 min (pH 7.4). Temporal changes in TRC pHi and volume were monitored as changes in FIR (F490/F440, solid line) and the fluorescence intensity of the isosbestic wavelength 440 nm (F440, dotted line), respectively, as a response to a decrease in the basolateral pH from 7.4 to 7.0. The F440 under control conditions for each ROI was taken as 100%. In each case the values are presented as mean ± SEM of n, where n = number of ROIs within the taste bud.

Effect of Hypertonic Mannitol on the CT Response to Acidic Stimuli
We preshrank TRCs by superfusing the tongue with the rinse solution (R) containing 1 M mannitol. Replacing the rinse with an isosmotic and iso-pH CO₂/HCO₃^– buffered solution (Table II; pH 7.4) elicited a CT response that is composed of two components with distinct temporal characteristics. A rapid transient increase in the phasic neural response (Fig. 7 A, a–b), which declined to a quasi-steady state after 2–5 s (b–c), defined as the tonic phase of the CT response (c–d). Repeatedly stimulating the tongue with dissolved CO₂ produced almost identical CT response profiles. Replacing the rinse (R) with another rinse containing 1 M mannitol (R+M; pH 7.4) produced a transient phasic response (e–f) that spontaneously decreased to rinse level. Replacing R+M with dissolved CO₂ solution containing mannitol (M+CO₂; pH 7.4) produced a CT response in which the phasic part of the neural response relative to the tonic part (h–i) was 65% smaller relative to control (b–c). For comparison, a set of transient phasic responses to CO₂ stimulation in the presence and absence of mannitol are shown in Fig. 7 B in an extended time scale. Hypertonic mannitol did not affect the tonic part of the CT response (i–j) relative to control (c–d). Repeatedly stimulating the tongue with M+CO₂ produced almost identical CT response profiles. Similar results were obtained with 10 mM acetic acid (Table II; pH 6.1) (unpublished data).

View larger version (27K): [in this window] [in a new window]   Figure 7. Effect of hypertonic mannitol on the CT response to CO2 and HCl. (A) The CT responses to dissolved CO2 (72 mM KHCO3 + 10%/90% CO2/O2; pH 7.4) were recorded relative to the rinse solution, R (72 mM KCl + 10 mM HEPES; pH 7.4). In the second part of the experiment the CT responses were recorded with both rinse (R+M) and the CO2 solution (M+CO2) containing, in addition, 1 M mannitol (M). (B) Inset shows a set of transient phasic responses to CO2 stimulation in the presence and absence of mannitol in an extended time scale. The magnitude of the peak phasic response was normalized to the phasic response elicited by stimulation with 300 mM NH4Cl. (C) The CT responses to 20 mM HCl + 10 mM KCl (R+HCl) were recorded relative to the rinse solution, R (10 mM KCl). In the second part of the experiment the CT responses were recorded with both rinse (R+M) and the HCl solution (R+M+HCl) containing, in addition, 1 M mannitol (M). The arrows indicate the time when the tongue was superfused with different solutions.

Effect of Hypertonic Mannitol on the CT Response to NaCl and Quinine
Stimulating the tongue with 10 mM KCl rinse solution containing 1 M mannitol (R+M; Table II) and then with the NaCl stimulating solution containing 10 mM KCl + 100 mM NaCl + 1 M mannitol (R+N+M) elicited a CT response (Fig. 8 A) in which both the phasic part of the response (g–h) and the tonic part of the response (i–j) were greater relative to their respective values (a–b) and (c–d) in the absence of mannitol. Repeatedly stimulating the tongue with R+N+M produced almost identical CT response profiles. Similar results were obtained with hypertonic cellobiose (Lyall et al., 1999). In contrast, hypertonic urea did not affect CT responses to NaCl (Lyall et al., 1999) or to acidic stimuli (unpublished data). These results indicate that in contrast to acidic stimuli, hypertonic mannitol increases the phasic and tonic components of the CT response to NaCl.

View larger version (31K): [in this window] [in a new window]   Figure 8. Effect of hypertonic mannitol on the CT response to NaCl, NH4Cl, and quinine. (A) The CT responses to 100 mM NaCl + 10 mM KCl (R+N) were recorded relative to the rinse solution, R (10 mM KCl). In the second part of the experiment the CT responses were recorded with both rinse (R+M) and the NaCl solution (R+N+M) containing, in addition, 1 M mannitol (M). (B) The CT responses to 100 mM NaCl + 10 mM KCl + 5 µM Bz (N+Bz), 10 mM KCl + 10 mM quinine (Q), or 0.3 M NH4Cl were recorded relative to the rinse solution, R (10 mM KCl) in the absence and presence of 1 M mannitol (M). In the second part of the experiment CT responses to the above stimuli were again recorded in the same animal after the topical lingual application of 20 µM cytochalasin B for 20 min. The arrows indicate the time when the tongue was superfused with different solutions.

Effect of F- and G-actin Probes on the CT Response to Acidic Stimuli
Cytochalasin B shifts the equilibrium from F-actin to G-actin and inhibits the pH_i-induced decrease in cell volume (Fig. 6 B). Superfusing the tongue with dissolved CO₂ solution containing 1 M mannitol decreased the magnitude of the phasic response relative to the tonic response (Fig. 9 A, f–g) compared with control (b–c). Cytochalasin B (Fig. 9 A, post-cytochalasin B) inhibited the phasic part of the CT response by >80% (j–k) relative to control (b–c). Cytochalasin B produced no effect on the tonic part of the CT response (k–l versus c–d). After cytochalasin B treatment, increasing the osmolarity of the CO₂ solution with 1 M mannitol (M+CO₂) completely eliminated the phasic part of the CT response (n–o) relative to control (f–g). Cytochalasin B also diminished the response to rinse + mannitol (R+M) relative to rinse alone (R). This indicates that interfering with the cell cytoskeleton also attenuates the osmotically induced decrease in TRC volume.

View larger version (26K): [in this window] [in a new window]   Figure 9. Effect of cytochalasin B and phalloidin on the CT response to CO2. The CT responses to dissolved CO2 (72 mM KHCO3 + 10%/90% CO2/O2; pH 7.4) were recorded relative to the rinse solution, R (72 mM KCl + 10 mM HEPES; pH 7.4) before and after the topical lingual application of (A) 20 µM cytochalasin B for 20 min or (B) 1.2 mM phalloidin for 30 min. The CT responses were recorded also with both rinse (R+M) and the CO2 solution (M+CO2) containing, in addition, 1 M mannitol (M). The arrows indicate the time when the tongue was superfused with different solutions. In B, the peak CT response data were normalized to the tonic 300 mM NH4Cl response in each animal (as described in the MATERIALS AND METHODS section) and are presented as the mean ± SEM of the relative peak phasic response from three animals. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

CT responses to 20 mM HCl were recorded under control conditions, after topical lingual application of 20 µM cytochalasin B for 20 min and then after treating the tongue with 1.2 mM phalloidin for additional 40 min. Cytochalasin B inhibited the phasic part of the HCl CT response relative to the tonic part (Fig. 10 A, peak [f] to tonic level) relative to control (peak [b] to tonic level). In the presence of 1 M mannitol (R+M+HCl), the post-cytochalasin B phasic response to HCl (peak [h] to tonic level) was almost eliminated relative to control (peak [d] to tonic level). Subsequently treating the tongue with phalloidin, both the magnitude of the phasic part of the CT response to HCl in the absence (Fig. 10 A; peak [j] to tonic level) and presence of mannitol (peak [l] to tonic level) increased to near control levels. The data from three such experiments are summarized in Fig. 10 B and indicate that phalloidin partially reverses the attenuation in the phasic part of the CT response to HCl alone and to stimulation with M+HCl (hatched bars) caused by cytochalasin B pretreatment (open bars). Thus mannitol and the decrease in pH_i produce additive effects on TRC volume.

View larger version (35K): [in this window] [in a new window]   Figure 10. Effect of cytochalasin B and phalloidin on the CT response to HCl. (A) The CT responses to 20 mM HCl + 10 mM KCl (R+HCl) were recorded relative to the rinse solution, R (10 mM KCl). CT responses were also recorded with both rinse (R+M) and the HCl solution (R+M+HCl) containing, in addition, 500 mM mannitol (M). CT responses were recorded under control conditions (Control), after topical lingual application of 20 µM cytochalasin B for 20 min (Post-cyto B), and after topical lingual application of 1.2 mM phalloidin for 30 min (Post-cyto B and phalloidin). (B) Summary of the effect of cytochalasin B and phalloidin on the CT response to HCl. The relative peak phasic responses (a–b, c–d, e–f, g–h, i–j, and k–l) from three animals were normalized to tonic 300 mM NH4Cl in each individual animal and presented as mean ± SEM for control (black bar), post-cytochalasin (white bar), and post-cytochalasin B and phalloidin treatment (hatched bar). *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Relationship between pH_i Decrease, Changes in Cell Cytoskeleton, Cell Shrinkage, and Membrane Conductance
We hypothesize that pH_i-induced changes in the cell cytoskeleton activate a membrane conductance in TRCs that is involved in the phasic response to acidic stimuli. To detect the presence of a membrane conductance activated during the phasic response, CT responses to HCl were monitored under lingual voltage clamp. Just before superfusing the tongue with 20 mM HCl, during the rinse (10 mM KCl) a transepithelial voltage of –60 or +60 mV was applied across the tongue (Fig. 11 A, arrows). We reasoned that any conductance that is activated during pH_i-induced cell shrinkage should be voltage sensitive, and hence, should affect the phasic part of the CT response to acid stimulation. Second, in TRCs that are preshrunken with hypertonic mannitol, this conductance will either not be activated by acid stimulation or its activation will be significantly reduced. Imposing a –60 mV transepithelial potential across the tongue (Fig. 11 A) did not affect the magnitude of the phasic response (–60 mV) relative to zero current clamp (0 mV). In contrast, imposing a +60 mV transepithelial potential enhanced the magnitude of the phasic part of the CT response (+60 mV) by almost twofold relative to 0 mV (0 mV). In the presence of 1 M mannitol, the phasic responses at 0, –60, and +60 mV were attenuated relative to control. In addition, hypertonic mannitol (M+HCl) almost completely inhibited the increase in the magnitude of the phasic part of the CT response at +60 mV applied transepithelial potential (Fig. 11 A). It should be noted that both the applied transepithelial potential and the hypertonic mannitol produced only minimal effects on the tonic part of the CT response to HCl. The magnitude of the peak phasic response to HCl was monitored for a range of voltages between –80 and +80 mV in three animals. The results demonstrate that the membrane conductance becomes activated at positive voltages (Fig. 11 B, •). Preshrinking TRCs with hypertonic mannitol inhibited the membrane conductance at all voltages (Fig. 11 B, ). This suggests the possibility that the membrane conductance activated during acid transduction is most likely activated due to pH_i-induced cell shrinkage.

View larger version (30K): [in this window] [in a new window]   Figure 11. Effect of voltage on the phasic response to HCl stimulation in the absence and presence of hypertonic mannitol. (A) The CT responses to 20 mM HCl were recorded relative to the rinse (R = 10 mM KCl) in the presence (M+HCl) and absence (HCl) of 1 M mannitol (M). Just before superfusing the tongue with HCl, during perfusion of the tongue with the rinse solution a transepithelial voltage of either –60 or +60 mV was applied across the receptive field (arrows). (B) Summary of the osmotic effect of mannitol on the CT response to HCl at each applied transepithelial voltage between –80 and +80 mV. The relative peak phasic responses to HCl from three animals were normalized to the tonic 300 mM NH4Cl response in each individual animal and presented as mean ± SEM at each voltage in the absence (•) and presence () of mannitol.

Topical lingual application of flufenamic acid (40 µM) for 20 min (Fig. 12) inhibited the phasic part of the CT response to HCl relative to the tonic part (l–m) compared with control (b–c). In the presence of 1 M mannitol, the post-flufenamic acid phasic response to HCl (r–s) was significantly smaller relative to control (h–i). This treatment affected neither the phasic nor the tonic responses to 0.3 M NH₄Cl (unpublished data). This suggests that at this concentration, flufenamic acid most likely does not uncouple gap junctions and does not produce nonspecific effects (Srinivas and Spray, 2003). Thus the effect of flufenamic acid inhibition of the phasic responses is modality specific. The observation that a longer time exposure to flufenamic acid is required to inhibit the membrane conductance suggests that this conductance is most likely localized in the basolateral membrane of TRCs.

View larger version (23K): [in this window] [in a new window]   Figure 12. Effect of flufenamic acid on the CT response to HCl. CT responses to 20 mM HCl were monitored in the absence and presence of 1 M mannitol (M) before (A) and after (B) treating the tongue with 40 µM flufenamic acid for 20 min. The arrows represent the time periods at which the tongue was superfused with different solutions.

View larger version (30K): [in this window] [in a new window]   Figure 13. Effect of applied voltage on the phasic response to HCl stimulation in the presence and absence of flufenamic acid. The CT responses to 20 mM HCl were recorded relative to the rinse (R = 10 mM KCl) before (A) and after (B) treating the tongue with 40 µM flufenamic acid for 20 min. Just before superfusing the tongue with HCl, during perfusion of the tongue with the rinse solution a transepithelial voltage of either –60 or +60 mV was applied across the receptive field. The arrows represent the time periods at which the tongue was superfused with HCl. (C) The CT responses to 20 mM HCl were recorded relative to the rinse (R = 10 mM KCl) before and after treating the tongue with 40 µM flufenamic acid for 20 min. Just before superfusing the tongue with HCl, during perfusion of the tongue with the rinse solution, a transepithelial voltage between –80 or +80 mV was applied across the receptive field. For each voltage step the magnitude of the peak phasic response was calculated. The peak CT response data were normalized to the tonic 300 mM NH4Cl response in each animal (as described in the MATERIALS AND METHODS section) and are presented as the mean ± SEM of the relative peak phasic response from three animals.