Respiration Physiology 129 (2001) 231 245 www.elsevier.

com/locate/resphysiol

Acid detection by taste receptor cells

John A. DeSimone a,*, Vijay Lyall a, Gerard L. Heck a, George M. Feldman b,c
a b

Department of Physiology, Virginia Commonwealth Uni6ersity, Richmond, VA 23298, USA Department of Medicine, Virginia Commonwealth Uni6ersity, Richmond, VA 23298, USA c McGuire Veterans Affairs Center, Richmond, VA 23249, USA Accepted 7 March 2001

Abstract Sourness is a primary taste quality that evokes an innate rejection response in humans and many other animals. Acidic stimuli are the unique sources of sour taste so a rejection response may serve to discourage ingestion of foods spoiled by acid producing microorganisms. The investigation of mechanisms by which acids excite taste receptor cells (TRCs) is complicated by wide species variability and within a species, apparently different mechanisms for strong and weak acids. The problem is further complicated by the fact that the receptor cells are polarized epithelial cells with different apical and basolateral membrane properties. The cellular mechanisms proposed for acid sensing in taste cells include, the direct blockage of apical K+ channels by protons, an H+-gated Ca2 + channel, proton conduction through apical amiloride-blockable Na+ channels, a Cl conductance blocked by NPPB, the activation of the proton-gated channel, BNC-1, a member of the Na+ channel/degenerin super family, and by stimulus-evoked changes in intracellular pH. Acid-induced intracellular pH changes appear to be similar to those reported in other mammalian acid-sensing cells, such as type-I cells of the carotid body, and neurons found in the ventrolateral medulla, nucleus of the solitary tract, the medullary raphe, and the locus coceuleus. Like type-I carotid body cells and brainstem neurons, isolated TRCs demonstrate a linear relationship between intracellular pH (pHi) and extracellular pH (pHo) with slope, DpHi/DpHo near unity. Acid-sensing cells also appear to regulate pHi when intracellular pH changes occur under iso-extracellular pH conditions, but fail to regulate their pH when pHi changes are induced by decreasing extracellular pH. We shall discuss the current status of proposed acid-sensing taste mechanisms, emphasizing pH-tracking in receptor cells. 2001 Elsevier Science B.V. All rights reserved.
Keywords: Acid base, acid detection; Amphibians, frog, salamander; Channels, ion, pH-gated; Mammals, rodents, primates, humans; Perception, taste, acid, submodalities; Receptor, taste detection; Taste transduction, perception

1. Introduction
* Corresponding author. Tel.: + 1-804-828-4489; fax: +1804-828-7382. E-mail address: jdesimon@hsc.vcu.edu (J.A. DeSimone).

Among the chemoreceptors in the oral cavity is a subset that respond vigorously to acids. These are taste receptor cells (TRCs) that establish the

0034-5687/01/$ - see front matter 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 3 4 - 5 6 8 7 ( 0 1 ) 0 0 2 9 3 - 6

232

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245

response patterns in the peripheral gustatory system that ultimately produce the conscious perception of sourness. Sourness is one of the primary taste sensations (Scott and Plata-Salaman, 1991; Smith and Frank, 1993). It evokes an innate rejection response in infants (Beauchamp et al., 1991) and generally remains aversive to humans and most other animals throughout life. Aversive behavior toward sourness is easy to rationalize considering that acids uniquely comprise the class of sour stimuli, and acids are potentially harmful substances. Since each stimulus evoking sour sensations produces dissociable hydrogen ions, it would at rst appear that TRCs are most likely extracellular pH detectors. On that basis the perception of sourness ought to be a graded function of stimulus pH. However, this proves not to be generally true. A poor correlation exists between the perception of sourness and stimulus pH, a nding that has been repeatedly conrmed in human psychophysical studies (Richards, 1898; Liljestrand, 1922; Beatty and Cragg, 1935; Ganzevles and Kroeze, 1987) and in human studies where the rate of acid-induced salivary secretion has been used as the index of sourness (Makhlouf and Blum, 1972). Recordings from gustatory afferents in rats have also been consistent in failing to show a strong correlation between stimulus pH and the neural response to particular acidic stimuli (Beidler, 1967; Beidler and Gross, 1971; Ogiso et al., 2000). These results beg the question: What specically do TRCs sense when presented with acidic stimuli? In the following we will review a surprising diversity of transduction mechanisms that have been proposed for the sour taste modality and an equally surprising species variability. The reason for the plethora of proposed transduction mechanisms is, in part, the consequence of the high chemical reactivity of protons. At low concentration protons affect ion trafc across epithelia at the level of ion channels in the apical and basolateral cell membranes and in the paracellular shunt pathways connecting the cells. In addition they may affect these and other functional elements (e.g. neutral ion exchangers) from both outside and inside the cells (Lyall and Biber, 1994).

2. Sour perception in humans One of the earliest observations in taste psychophysics is that acetic acid is perceived as more sour than HCl at the same pH, but that HCl is more sour than acetic acid at equimolar concentrations (Richards, 1898). It was soon recognized that at least part of this paradox derives from the fact that HCl is completely dissociated in solution while acetic acid is not. Beatty and Cragg (1935) used the perceived sourness of HCl to establish a scale against which other acids could be compared. On that scale it required lower concentrations relative to HCl for the diprotic tartaric acid to produce the same sourness. For acetic acid, higher concentrations relative to HCl were judged equisour. They showed, in addition, that sourness correlated better with titratable acid than with pH. This hypothesis was further explored by Makhlouf and Blum (1972) who used salivary ow rate as the measure of the sour response in humans. In their study of six weak acids, they concluded that relative potency was determined by the amount of acid titratable to pH 56. They also reported a poor correlation between salivary ow rate and stimulus pH. The salivary ow rate data were accurately represented as saturable functions of the acid concentration. The rank ordering of their parameters, K (acid concentrations giving half-maximal salivary ow rates) was the same as that found by Beidler (1967) in electrophysiological recordings from rat chorda tympani nerves. Ganzevles and Kroeze (1987) compared the perceived sourness of acetic acid in humans under two adaptation conditions. In one case the subjects tongues were adapted to water before tasting acetic acid, and in the second case the adapting solutions were dilute HCl solutions with pH values ranging from 2.79 to 3.13. The test acetic acid solutions had the same pH values as the adapting HCl solutions. The subjects reported no difference in the perceived sourness of acetic acid with either water or HCl adaptation. In addition there were no signicant differences in the perceived sourness of the acetic acid solutions compared with the same concentrations of acetic acid buffered to higher pH with sodium acetate. These results indicated that extracellular pH per

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245

233

se could not be the sour stimulus. If it were, the sour taste of acetic acid would have been completely eliminated by adaptation in isohydric HCl. A remarkable speculation appeared in a 1930 paper by Taylor et al. (1930), viz. all acid solutions which taste equally sour produce the same pH within the interior of the cell. This is an area that is currently being investigated and it appears that the speculation is fundamentally correct (Feldman et al., 2000; Lyall et al., 2000) Taylor et al. (1930) also noted a strong correlation between the hydrophobicity of a fatty acid and its ability to evoke a sensation of sourness. This was conrmed in subsequent studies by Gardener (1980). The possibility that apical epithelial Na+ channels or an apical Na+/H+ exchanger might be involved in human sour taste has been investigated by evaluating the effects of amiloride application to the tongues of subjects. In each case amiloride had no effect on the perception of sourness due to acid stimulation (Schiffman et al., 1983; Tennissen and McCutcheon, 1996; Ossebaard and Smith, 1995).

3. Species differences in sour taste mechanisms Insights into the cellular mechanisms of acid taste reception have been gleaned from a variety of animal models. Surprisingly many of the proposed mechanisms derived from studies on a given species fail to generalize fully to other species (Table 1). For that reason we will discuss these mechanisms according to species.

3.1. Necturus
The salamander, Necturus maculosus, has unusually large cells, a characteristic that has made it a favorite subject for intracellular recordings. Kinnamon and Roper (1988) used an isolated lingual epithelial preparation to make recordings from TRCs. The cells were stimulated focally near the taste pore by applying pressure to an extracellular pipette containing taste stimuli. Application of KCl to the taste pore region produced a depolarizing intracellular potential and a decrease in input resistance relative to the resting potential

Table 1

Species Necturus Necturus Bullfrog Bullfrog Mouse Rat Rat Hamster Hamster Monkey/Chimpanzee

Reference Kinnamon and Roper, 1988 Kinnamon et al., 1988 Miyamoto et al., 1988 Okada et al., 1993 Miyamoto et al., 1998 DeSimone et al., 1995 Lyall et al., 1997 Gilbertson et al., 1992 Stewart et al., 1998 Hellekant et al., 1997a,b

Preparation Isolated lingual epithelium Isolated TRCs Isolated lingual preparations Isolated lingual preparations Polarized taste bud preparations In vivo CT recording Isolated TRCs Intact lingual epithelium Isolated TRCs In vivo CT recordings

Technique Microelectrodes Loose-patch recordings Microelectrodes Microelectrodes Patch-clamp Voltage-clamp TRC pHi Loose patch clamp TRC pHi measurement Single unit recordings

Acid-effects Citric acid blocked apical K+ conductance Acid blocked K+ channels localized in the apical membrane HCl depolarized TRCs via apical H+ gated channels permeable to Ca2+ and Na+ DCCD blocked Acetic acid induced depolarization NPPB blocked citric acid induced depolarization CT responses to HCl are amiloride-and voltage insensitive Acid-induced decrease in TRC pHi is amiloride insensitive Amiloride blocked citric acid-induced current transients in TRCs Acid-induced decrease in TRC pHi is amiloride-insensitive N-bers respond poorly while H-bers respond best to acid stimulation

234

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245

( 50 to 60 mV) and baseline resistance, respectively. When the resting potential was made slightly hyperpolarizing by current injection, the KCl-induced depolarization was often accompanied by a single action potential. The KCl-induced depolarization was blocked by bath application of TEA. Application of citric acid focally also caused a depolarizing potential, often with an initial burst of action potentials, but in this case with increasing input resistance. TEA also blocked the citric acid responses. In a subsequent study Kinnamon et al. (1988) used a loose-patch recording method on isolated Necturus TRCs to map the distribution of voltage-gated K+ channels along the cell surface. They concluded that the K+ conductance was about 50-fold greater on the apical membrane relative to the basolateral membrane, suggesting that voltage-gated K+ channels may then have a role in the detection of K+ ions and acids. Insideout and outside-out patches from the apical membrane conrmed the high K+-selectivity of the apical membrane (PK/PNa =28) (Cummings and Kinnamon, 1992). The acid block of the K+ channels occurred only when acid was added to the bath perfusate with the channels in the outside-out conguration. The results indicate that for Necturus the apical membrane provides a high density of voltage-gated K+ channels that are conductive under resting conditions. These serve as chemical transducers for K+ and also for H+ ions because the latter are effective blockers of K+ channels. It appears, however, that the apical membranes of TRCs of many other species do not contain potassium channels (Miyamoto et al., 1996, 1998; Ye et al., 1994; Avenet and Lindemann, 1991 Furue and Yoshii, 1997) so it is unlikely that this mechanism generalizes extensively. It should be noted, however, that isolated TRCs from the tiger salamander (Ambystoma tigrinum) do show evidence of voltage-gated K+ channels that can be blocked by citric and acetic acids (Sugimoto and Teeter, 1991) in a manner analogous to those of Necturus. However, it is unknown if these channels also are concentrated in the apical membranes of the taste cells. The status of the K+-channel block mechanism for the tiger salamander is, therefore, not as clear cut as in the case of Necturus.

3.2. Frog
Miyamoto et al. (1988) made intracellular recordings from the bullfrog (Rana catesbetana) using a preparation that permitted control of the composition of the interstitial uid as well as the lingual supercial uid. Applying HCl to the tongue produced a dose-dependent depolarization of the intracellular potential with threshold at 0.01 mM. Depolarization was accompanied by a decrease in input resistance. The following changes in the composition of the interstitial uid had no effect on the depolarization evoked by 1 mM HCl: substituting choline for Na+, removing Ca2 + , substituting choline for Na+ and removing Ca2 + at the same time, substituting isethionate for Cl, increasing K+ to 100 mM, or adding tetrodotoxin. From these results they concluded that the HCl-induced receptor potential does not depend on changes in ionic conductances located on the basolateral membranes of the TRCs. In contrast to the effects of ion substitutions on the basolateral compartment, removing Ca from the lingual surface perfusate the amplitude of the HCl depolarizing potential was inhibited by 48%, and removing Na reduced it by 63%. The addition of Cd2 + or Co2 + inhibited the HCl response. Increasing the Ca2 + concentration to 10 or 20 mM signicantly increased the HCl-induced depolarization and made further reductions in input resistance. Measuring the response as a function of Ca2 + concentration with and without Na+ and making the inverse measurements of response as a function of Na+ concentration with and without Ca2 + gave in each case competitive inhibition kinetics. The reversal potential for the HCl-induced depolarization was estimated to lie between + 60 and + 90 mV, a range bracketing the Ca2 + and Na+ equilibrium potentials. The authors concluded that in the frog a major part of the acid induced depolarizing potential (putative receptor potential) occurred through proton-gated ion channels in the apical membrane permeable to Ca2 + and Na+. In a subsequent study Okada et al. (1993) showed that in the frog the depolarizing potential due to acetic acid could be blocked with 0.1 mM N,N%-dicyclohexylcarbodiimide (DCCD), an inhibitor of the channel component of the

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245

235

V-type proton pump (Lauger, 1991). The view that emerges from these studies is that both a proton-gated ion channel and a proton pump are the membrane transducers for acids in frogs. Frog TRCs also have apical membrane K+ channels. A 80 pS channel is found exclusively in the apical region while a 40 pS K+ channel is found at highest density in that region (Fujiyama et al., 1994). Whether these channels have a role in acid sensing is presently unclear. In a more recent study Kolesnikov and Bobkov (2000) have made patch clamp recordings from TRCs of Rana temporaria. Their data suggested that extracellular K+ may serve as a ligand activating ionic channels (potassium-activated (PA) channels). The inuence of different ions on the PA current reversal potential indicated that the responsible channels are mainly permeable to K+ and H+. Relative permeabilities were estimated as PH/ PK =3600. They reported that external K markedly increased the sensitivity of isolated TRCs to bath solution pH due to the activation of the PA channels suggesting their role in sour transduction. This interesting possibility will doubtlessly be further persued.

3.3. Mouse
Miyamoto et al. (1998) made patch clamp recordings from mouse (C57BL/6 and BALN/c) TRCs using a preparation that preserves the polarity of the taste bud. Injection of elastase followed by incubaton allowed for the removal of the lingual epithelium from the underlying muscle with the intact fungiform taste buds still attached to the epithelial sheet at their apical poles. Recordings were made from the basolateral membrane using the whole cell path clamp method while the taste bud was held in position by a slight negative pressure applied to a narrow strip of epithelial tissue. The preparation was constantly perfused with Ringers solution made to ow from the submucosal side toward the mucosal side. Stimulation pipettes were positioned near the taste pore. Under pressure a small volume of stimulus could be sent toward the pore where it could interact with the apical processes of the TRCs before the counter ow of Ringers

solution removed it. Results showed that 43% of the cells responded to 25 mM citric acid with a depolarization of 1030 mV from the resting potential. The acid-induced depolarization was partially suppressed reversibly by 5-nitro 2-(3phenylpropylamino)-benzoic acid (NPPB) a Cl channel blocker, suggesting the participation of a Cl ux in acid detection by TRCs. A possible role for Cl in acid detection by taste receptors was suggested by an earlier study on the transepithelial short-circuit current across isolated canine lingual epithelium (Simon and Garvin, 1985). Miyamoto et al. (1998) localized the NPPB-sensitive site to the basolateral side of the taste bud. The NPPB-sensitive part of the current had a reversal potential of approximately 0 mV, which was close to the Cl reversal potential of 4.6 mV. The NPPB-insensitive current had a reversal potential of + 35 mV. Since the citric acid solutions were Na+ free, this suggests that in addition to activating a basolateral Cl conductance, an apical membrane cation-selective channel permeable to Ca2 + , H+, or K+ is also activated. Both NPPB-sensitive and NPPB-insensitive responses caused an increase in conductance. However, in an earlier study in mouse where impalement microelectrode methods were used, two cell populations emerged (Tonosaki and Funakoshi, 1984). In one cell type HCl-induced depolarization occurred with increased conductance while in the other the conductance decreased. The reason for the apparent discrepancy may lie in the different recording methods and stimulating solution compositions employed.

3.4. Rats
Rats have been the preferred mammalian model for gustatory electrophysiology. Numerous recordings have been made from the rat anterior lingual receptive eld innervated by the chorda tympani (CT). Beidler (1967) made a comparative study of the CT response to 20 acids. The acids were rank ordered according to the concentration required to give the same magnitude CT response as that obtained with 5 mM HCl. It was apparent that the pH of the stimulus per se is not the determining factor in the ranking. For example, it

236

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245

required just 10 mM succinic acid to give the same response as 150 mM butyric acid, but the former had a pH of 3.08 while the latter had a pH of 2.84. The poor correlation between CT response and stimulus pH was further investigated by comparing responses to 0.1 M acetic acid and 0.1 M acetic acid buffered with 0.01 M sodium acetate. The responses were essentially the same even though the pH of the former was 2.88 and that of the latter 3.76. A similar study has recently been done by Ogiso et al. (2000). They obtained the CT response to acetic acid+Tris-acetate mixtures at a constant total concentration of 0.1 M over the pH range 3 7. At every pH the responses were greater than responses to HCl at the same pH. The authors concluded that the undissociated form of acetic acid must be a major factor in the response to acetic acid, since the free proton response (represented by the HCl response) amounted to only about 25% of the total response to acetic acid at pH 3. With impalement microelectrodes, studies of rat TRCs have yielded evidence for acid-induced depolarization with increasing conductance (Ozeki, 1971) and for depolarization with decreasing conductance (Sato and Beidler, 1982), results similar to the mouse. DeSimone et al. (1995) obtained rat CT responses to NaCl and HCl with the stimulated part of the anterior lingual receptive eld under voltage clamp. Responses to NaCl were enhanced at submucosa negative clamp voltages and suppressed at submucosa positive voltages. NaCl responses were also amiloride-sensitive. In contrast responses to HCl were voltage and amiloride-insensitive. The results support the view that taste transduction for NaCl is mediated through apical membrane epithelial sodium channels (ENaC) (Heck et al., 1984; Stewart et al., 1995; Kretz et al., 1999). However, the lack of amiloride sensitivity in the HCI responses indicated that ENaC is not involved in acid sensing. The lack of amiloride sensitivity in rat chorda tympani responses to HCl has also been noted in single unit recordings (Ninomiya and Funakoshi, 1988). Moreover, the lack of voltage sensitivity suggested that the apical membrane of the TRCs is not very H+ permeable. Recordings of the transepithelial potential during HCl stimulation showed an in-

creasing submucosa positive potential at the onset of the HCl that then reached a plateau. However, upon washout the potential did not return to prestimulus levels, but rather became more positive than during stimulation. This was consistent with a change in the ion-selectivity of the paracellular shunts, the low resistance pathways through the tight junctions between the TRCs. The submucosa positive direction of the change suggested that upon HCl stimulation the shunts were converted from cation to anion selective by diffusing protons. This, in turn, suggested that protons stimulate TRCs from the basolateral side. The lack of an inhibitory amiloride effect on the chorda tympani response to acids indicates that acid sensing in the anterior lingual taste buds (i.e. taste buds in fungiform papillae) does not involve H+-gated ion channels from the Na+ channel/degenerin family. A study by Ugawa et al. (1998), however, demonstrates that the degenerin, MDEG-1, is present in the circumvallate papilla TRCs of the rat. They screened a cDNA library from rat circumvallate papillae for members of the degenerin cation channel family and obtained 18 cDNA clones. When pooled cRNA corresponding to a mixture of these clones was expressed in Xenopus oocytes, no amiloride sensitive current was detected at pH 7.5 but at an extracelluar pH of 5.5 a large inward current developed that was partially blocked by amiloride. Expressing the clones individually showed that the low pH-induced amiloride-sensitive current was the unique property of MDEG-1. Ugawa et al. (1998) also showed that MDEG-1 mRNA is found in circumvallate TRCs and that MDEG-1 immunoreactivity is found only in TRCs. Mindful of the fact that taste studies show that acetic acid is a stronger sour stimulus than HCl at the same pH, they compared the inward currents across Xenopus oocyte membranes produced by HCl and acetic acid at pH 5.0. Acetic acid produced the greater current. While this would appear to conrm that MDEG-1 is a probable sour taste receptor, there remain important issues to be resolved. First, the H+-gated members of the Na+ channel/degenerin family are activated by extracellular (not intracellular) pH (Waldmann et al., 1997). That being the case it is

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245

237

difcult to explain why current ow through MDEG-1 for HCl and acetic acid in the oocytes would be different since the extracellular pH was the same for both of them. The results are more consistent with intracellular pH gating the current magnitude (cf Feldman et al., 2000; Lyall et al., 2000, and below). Second the acid stimuli were presented in solutions having the usual concentrations of salts consistent with extracellular uid (i.e. Na+ 140 mM) so that current carriers for MDEG-1 would be available. However, neural responses to acids are readily obtained in the absence of salts (Beidler, 1967), an observation that puts into doubt that MDEG-1 functions in situ as a sour taste receptor protein (see also Kinnamon et al., 2000).

3.5. Hamster
The hamster chorda tympani is more sensitive to saccharides than that of the rat and for that reason the hamster taste system has also been well studied (Frank et al., 1988). For acids and salts, however, the chorda tympani response of the hamster is similar to that of the rat. The integrated chorda lympani response of the hamster to NaCl is strongly inhibited by amiloride, and like the rat, the integrated neural response to HCl is affected little, if at all, by amiloride treatment (Herness, 1987; Hettinger and Frank, 1990; Stewart et al., 1998). Consistent with this, recordings from neurons in the hamster nucleus of the solitary tract (NST), the rst postsynaptic locus in the gustatory neuraxis, show no amiloride sensitivity to lingual stimulation with HCl (Smith et al., 1996). A detailed study of the effect of amiloride and its analogs on hamster chorda tympani responses to HC1 was made by Stewart et al. (1998). The amiloride analog, benzamil, a more specic blocker of ENaC than amiloride, was a potent blocker of the chorda tympani response to NaCl, but had no effect on the response to 1 mM HCl. Benzamils effect on responses to 10 mM HCI were not statistically signicant. However, three preparations gave unusually large responses to 10 mM HCl (up to twice the mean response level). In these cases benzamil caused up to 50% suppression, but in no cases was amiloride a

blocker of the HCl response at either 1 or 10 mM. Methylisobutylamiloride, a blocker of the Na+/ H+ exchanger, had no effect on either responses to NaCl or HCl. Stewart et al. (1998) also recorded the responses to NaCl and HCl under lingual voltage clamp conditions. Similar to the rat, the chorda tympani responses to NaCl were enhanced at submucosa negative clamp voltage (60 mV) and suppressed at positive voltage (+60 mV). Responses to both 1 and 10 mM HCl were clamp voltage insensitive, i.e. like the rat, responses to HCl do not appear to involve members of the Na channel/degenerin family in the apical membranes of TRCs. A reinvestigation of the effect of amiloride on the response to HCl in the NTS indicated that amiloride had no effect on the response in HCl-best single units (i.e. units ring with highest frequency to acids) (Boughter and Smith, 1998). However, in NaCl-best units, amiloride inhibited responses to NaCl, HCl, and citric acid. On this basis the authors suggest that coding for sour may involve inputs to the NST from both HCl-best and NaCl-best afferents (Boughter and Smith, 1998). One possible way that might occur is for H+ ions to stimulate receptor cells through the Na+ channel/degenerin family. Other explanations involve possible branching of chorda tympani inputs at the NST level. A nal resolution of this point awaits further research. Evidence for a direct role for ENaC in hamster acid taste responses was obtained by Gilbertson et al. (1992, 1993). Gilbertson et al. (1992) used a recording technique introduced by Avenet and Lindemann (1991). This consisted of a pipetteelectrode that t over a single fungiform papilla on the dorsal surface of an excised hamster tongue. The pipette-electrode was held in place by weak vacuum. An agar bridge recording electrode was placed in the posterior muscle exposed by the cut to remove the tongue. The pipette-electrode served as ground. Various taste stimuli were applied through an inow tube that traveled to within 50 mm of the patch pipette opening to the papilla surface. Recordings were made using a standard path clamp amplier and consisted of stimulus evoked currents with the potential between the electrodes held at zero voltage-clamp.

238

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245

NaCl and citric acid each produced rapid current transients (named action currents). Citric acid at 8 mM (pH 2.6) produced action currents of 7 10 pA at a frequency of 1 2 Hz. The frequency increased in a dose-dependent manner with decreasing pH with threshold between pH 5 and 6. The action currents to both NaCl and citric acid were blocked by amiloride in a dose dependent manner (Ki for citric acid and NaCl were 2.4 and 1.9 mM, respectively). The results suggested that perhaps both Na+ and H+ excite TRCs through ENaC. This is difcult to determine from these studies alone, since the biophysical interpretation of the action currents is unclear. In a subsequent paper, however, Gilbertson et al. (1993) used whole cell patch clamp methods on isolated hamster taste buds to observe the effects of citric acid. They applied citric acid, using a pressure injection pipette, to the isolated taste buds in a Na+ and K+-free medium. Citric acid application gave an inward current at clamp voltages between 80 mV and + 20 mV. The reversal potential was estimated to be near the equilibrium potential for H+ at extracellular pH 4.5. The calculation of the H+ equilibrium potential assumed a constant intracellular pH of 7.2. However, in a medium of pH 4.5 the intracellular pH will not remain at 7.2, but will assume a much lower pH (as much as 1 2 pH units lower) due to the pH-tracking property of TRCs (Lyall et al., 1997; Stewart et al., 1998). The H+ equilibrium potential may be, therefore, much lower than the reversal potential, suggesting that H+ ion may not be the sole current carrier. The inward currents were, however, blocked by 30 mM amiloride, suggesting a role for the Na+ channel/degenerin family in acid detection in isolated taste cells. However, chorda tympani responses to acids are not amiloride sensitive. This suggests that the putative proton-conducting channels are present in the basolateral membranes of TRCs. If these channels are functional in situ, they may not be blocked by topical application of amiloride on the lingual surface, since amiloride will have to reach basolateral membrane sites through the paracellular pathway where amiloride transport is likely to be restricted by low permeability. The functionality of the channels would depend on the acid stimuli reach-

ing them by the paracellular route, a possibility suggested by the behavior of the transepithelial potential in the presence of acids recorded in situ (cf DeSimone et al., 1995; Stewart et al., 1998).

3.6. Rhesus monkey and chimpanzee

Not surprisingly primates appear to be the best animal model for human taste. Hellekant et al. (1997a) have recorded taste single unit activity in both the chorda tympani and glossopharyngeal nerves of the rhesus monkey (Macaca mulatta). Of the chorda tympani units investigated, 15 units gave best responses to NaCl and were designated the N cluster. However, N cluster bers also responded moderately well to citric acid and aspartic acid. Eight of the bers, designated the H cluster, gave best responses to citric acid and aspartic acid. The 16 S cluster bers designated as S bers, responded best to sweet-tasting stimuli, which included saccharides (sucrose, fructose, glucose, and galactose), and many compounds such as aspartame, saccharine, dulcin, acesulfame-K, D-tryptophan and others which are also sweet tasting to humans. The S cluster also responded well to aspartic acid and citric acid. Four bers, designated the Q cluster, gave best responses to quinine and other substances with some bitter attributes. These bers did not respond well to acids. It is in the near congruence of the pattern of sweet sensitivity between many primates and humans that the primate model excels. Therefore, it appears that neural input evoked by acids from the chorda tympani to the NST is carried mainly by H-cluster bers with some input from both N-and S-cluster bers, but with virtually no input from the Q-cluster. A similar analysis by Hellekant et al. (1997a,b) on the taste responses from the glossopharyngeal nerve revealed a Mcluster that gave largest responses to monosodium glutamate with more moderate responses to citric acid, and aspartic acid, a Q cluster that responded well to quinine and other bitter tasting compounds, but not to acids, and a S-cluster that responded to sweeteners but not to acids. So taste information about acids from posterior lingual taste receptive elds travels with bers that also respond well to glutamate.

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245

239

Hellekant et al. (1997b) have also made single unit recordings from the chorda tympani of the chimpanzee (Pan troglodytes). Of 49 units surveyed, only 2 gave best responses to acids. Similar to the monkey a large proportion of the units (47%) were found to be sucrose-best. The S-cluster did not respond to acids. Twenty-nine percent of bers were NaCl-best units. This N-cluster was itself composed of three subclusters, (1) a Na-subcluster; (2) a NaK subcluster; and; (3) a M-subcluster. The Na-subcluster responded well to NaCl and LiCl, but not to KCl and acids. The Na+ and Li+ responses were strongly inhibited by amiloride. The Na K subcluster responded about equally to NaCl and KCl, and also gave small responses to acids. The M-subcluster responded well to monosodium glutamate and gave low responses to acids and some sweet-tasting stimuli. Quinine-best bers accounted for 20% of the population. The Q-cluster bers were, however, more broadly tuned than NaCl-best or sucrose-best bers and included robust responses to acids as well as bitter-tasting compounds. Unlike the rhesus monkey, the chimpanzee chorda tympani lacks a H-cluster, so most of the acid-evoked responses appear to travel with the Q-cluster. This, therefore, contrasts sharply with the monkey in which the Q-cluster was essentially acidinsensitive.

4. Evidence supporting a role for intracellular pH decrease in acid taste Undissociated acid molecules as well as free protons interact directly with the apical membranes of TRCs. This interaction docs not seem to require specic H+ receptors on the apical membranes of TRCs. This conclusion is based on the observation that CT nerve responses to acids are not inuenced by pretreatment of the tongue surface with protease (Ogiso et al., 2000) whereas that to sucrose are reduced effectively (Hiji, 1975). This implies that unlike sucrose sensing, membrane receptors or other surface proteins are not involved in acid detection. However, acid-induced effects on TRCs may not be restricted to the apical cell membrane. DeSimone et al. (1995)

have presented data suggesting that protons pass through the paracellular pathway and are buffered by the xed anionic sites as they do so. This raises the possibility that stimulus protons may also interact with the basolateral membranes of TRCs. It is not clear whether TRCs respond directly to a change in extracellular pH (pHo) or if a change in intracellular pH (pHi) of TRCs is involved in acid sensing. Several studies have identied direct effects of acids on TRC membrane conductances (Miyamoto et al., 1988, 1998; Gilbertson et al., 1992, 1993; Ugawa et al., 1998; Kolesnikov and Bobkov, 2000; Kinnamon et al., 2000; Stevens and Lindemann, 2000). The underlying hypothesis in these studies is that pH-gated changes in membrane conductances depolarize the receptor potential in TRCs leading to the release of neurotransmitter(s) and excitation of the taste nerves. As discussed in Section 2, the perception of sour in the case of weak organic acids is not a graded function of pHo. This has led to the hypothesis that weak organic acids can penetrate TRC membranes in the undissociated form and may dissociate intracellularly leading to cytoplasmic acidication. This hypothesis is supported by the observation that a strong correlation exists between lipid solubility (i.e. membrane permeability) and sourness of weak acids (Gardener, 1980). In addition, cytoplasmic acidication has been reported to reduce the electrical coupling between TRCs in Necturus (Bigiani and Roper, 1994) and to impair electrical communications via gap junctions (Spray and Bennett, 1985; Spray et al., 1981). To investigate the possibility that changes in TRC pHi are indeed involved in acid sensing Lyall et al. (1997) and Stewart et al. (1998) monitored the relationship between pHo and pHi in isolated rat and hamster circumvallate and fungiform taste bud fragments and single TRCs. In single isolated TRCs altering pHo by changes in external H+ ion concentration caused parallel changes in TRC pHi (Fig. 1). The changes in TRC pHi were independent of the anion concentration in the external buffer system. For example, in stimulus solutions containing the CO2/HCO 3

240

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245

Fig. 1. Pseudo-color ratio images of a single TRC loaded with BCECF. (A) Shows an optical section through a single rat fungiform TRC. Scale bar, 5 mm. (B) The same optical section is shown as a BCECF uorescence image, excited at the pH-sensitive wavelength, 490 nm. Note that the dye distribution pattern is a reection of the irregular shape of the TRC. The same optical section is shown as a ratio of BCECF uorescence pseudo-color images when excited alternately at 490 and 440 nm at the steady state pHo of 6.6 (C) and 7.7(D). Note that a decrease in pHo induced a parallel decrease in pHi both in the soma and apical process. This gure is reprinted from Lyall et al. (1997) with permission from The American Physiological Society.

buffer system, the changes in TRC pHi demonstrated a linear relationship to pHo whether pHo was altered by varying the PCO2 at constant external HCO, concentration or if pHo was altered 3 by varying the HCO concentration at constant 3 PCO2. The changes in TRC pHi were rapid and sustained (Fig. 2). No spontaneous recovery of

pHi was observed as long as pHo was maintained at acidic pH. The relationship between pHo and TRC pHi was observed to be linear with a slope near unity. Thus isolated TRCs demonstrate four main characteristics of a pH sensory cell. First, changes in pHo induce a proportionate response in pHi. Second, the changes in TRC pHi are

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245

241

rapid. Third, the changes in TRC pHi are maintained during the acidic stimulus. Fourth, the changes in pHi were observed in all regions of interest in the taste bud fragment suggesting that a signicant number of TRCs in a taste bud are involved in pH tracking (Stewart et al., 1998). The sustained changes in pHi seem to be a common characteristic of pH-chemosensory cells (Buckler et al., 1991; Ritucci et al., 1997). The above studies were made in isolated nonpolarized TRCs. In these experiments both apical as well as basolateral membranes of TRCs were exposed to changes in pHo. However, in the intact lingual epithelium, TRCs maintain their natural

Fig. 2. Effect of changing pHo on pHi. (A) A circumvallate TBF was perfused with HEPES-buffered Tyrode solution of pH 7.4, 6.7, and 7.9. The top horizontal bar represents the periods during which the TBF was perfused with solutions of different pHs. (B) The steady-state relationship between pHo and pHi in ve TBFs similar to that of (A). The line of best t determined by linear regression (S = 0.979 0.02; r =0.9969 0.002). This gure is reprinted from Lyall et al. (1997) with permission from The American Physiological Society.

polarity and are only exposed to acidic stimuli on the apical side. Therefore, before a rm conclusion can be made regarding pHi as the signal for acid transduction, measurement of TRC pHi must be made in polarized TRC preparations in which the apical and basolateral membranes remain segregated and the TRCs are exposed to acidic stimuli only on the apical membrane. In our preliminary studies (Feldman et al., 2000; Lyall et al., 2000) using an isolated lingual epithelium preparation mounted in a special microscopy chamber (Chu et al., 1995) apical application of acidic stimuli induced sustained decreases in TRC pHi. These data support the conclusion that a decrease in TRC pHi is a rst essential step in sour taste transduction. The studies of TRCs in the CO2/HCO buffer 3 system indicate that to achieve a decrease in TRC pHi it is not necessary to have changes in pHo. It is the entry of acid equivalents that decreases TRC pHi. For example increasing the concentration of both CO2 and HCO, simultaneously 3 while maintaining constant pHo decreased the TRC pHi (Lyall et al., 1997). Further a change from a HEPES-buffered solution (pH 7.4) to a solution buffered with 72 mM HCO; and 10% 3 CO2 (pH 7.4) caused a decrease in TRC pHi (Lyall et al., unpublished observations). In TRCs the hydration of CO2 to form H2CO3 is catalyzed by intracellular carbonic anhydrase (Komai et al., 1994; Daikoku et al., 1999; Goto et al., 2000). A CO2-induced decrease in pHi has been shown to be the primary signal for the activation of chemosensory neurons during hypercapnia (Ritucci et al., 1997, 1998; Wiemann et al., 1998; Dean et al., 1990; Pineda and Aghajanian, 1997; Buckler and Vaughan-Jones, 1994). The sustained changes in TRC pHi do not imply that TRCs are incapable of regulating pHi under resting conditions. Lyall et al. (1997), Stewart et al. (1998) reported that at constant pHo exposing the cells to short pulses of Na-acetate, Na-propionate, NH4Cl or PCO2 acidied TRCs. However, under these conditions TRCs demonstrated rapid spontaneous recovery of pHi to their resting values (Fig. 3). These results suggest that under resting conditions TRCs behave like many other cells and regulate pHi. Consequently, like

242

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245

5. Summary In summary there is a remarkable diversity of mechanisms that appear to have a role in the transduction of the sour taste stimulus at the level of the TRCs. In some cases the mechanisms appear to be species specic. Most of these mechanisms involve putative cell membrane entry pathways for H+ ions or ion pathways modulated by H+ ions. However, weak organic acids can permeate cell membranes as undissociated molecules, suggesting that intracellular pH changes may also be important in acid taste reception. This view is supported by psychophysical studies that demonstrate a poor correlation between extracellular pH and the perception of sourness in general. Imaging studies on isolated taste buds show that intracellular pH follows changes in extracellular pH according to a linear relation that resembles the characteristics of acid sensing cells in the brain and carotid bodies. Recent studies on taste buds still attached to the lingual epitheliurn, i.e. with maintained epithelium polarity, show that pH-tracking can be observed from the apical side. In the case of mammals further studies are needed to identify the specic entry pathways for H+ ions in taste receptor cell membranes that are involved in transduction. While it appears that a decrease in intracellular pH is an important early step in transduction, subsequent steps in the process leading to neurotransmitter release are still obscure and await additional research for clarication.

Fig. 3. Effect of weak acids. A fungi form TBF initially perfused with HEPES-buffered Tyrode solution was perfused with a similar Tyrode solution containing 30 mM sodium propionate (Na Pro). The top horizontal bar represents the period during which the TBF was exposed to sodium propionate. This gure is reprinted from Lyall et al. (1997) with permission from The American Physiological Society.

many other cells, TRCs must have pH regulatory mechanisms that are activated during intracellular acidication. In our preliminary studies we have identied Na+ H+-exchange activity in the basolateral membrane of polarized TRCs (Lyall et al., 2000) that is involved in regulating TRC pHi under resting conditions. The fact that TRCs can regulate their pHi, does not, however, seem to be consistent with sustained changes in TRCs pHi during acid stimulation and their role as pH-sensors. This apparent discrepancy can be reconciled if the pH regulatory mechanisms were to be inhibited during acid stimulation. This is indeed the case in chemosensory neurons (Ritucci et al., 1998), where sustained decreases in pHi during acid stimulation are related to inhibition of the Na+ H+ exchanger. Our preliminary studies suggest that basolateral Na+ H+ exchange activity in the TRCs is pH sensitive and is blocked at low pH (Lyall et al., unpublished observations). These data suggest that the pH-sensitivity of pH regulatory mechanisms of TRCs may be involved in acid transduction. This has now been directly demonstrated in chemosensory neurons where inhibition of the Na+ H+ exchanger results in a decrease in pH, and consequently, chemosensory neural activation (Wiemann et al., 1998).

Acknowledgements Supported by NIH grants DC-02422 and DC00122 and the Department of Veterans Affairs.

References
Avenet, P., Lindemann, B., 1991. Noninvasive recording of receptor cell action potentials and sustained currents from single taste buds maintained in the tongue: the response to mucosal NaCl and amiloride. J. Membr. Biol. 124, 33 41. Beatty, R.M., Cragg, L.H., 1935. The sourness of acids. J. Am. Chem. Soc. 57, 2347 2351.

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245 Beauchamp, G.K., Cowart, B., Schmidt, H.J., 1991. Development of chemosensitivity and preference. In: Getchell, T.V., Doty, R.L., Bartoshuk, L.M., Snow, J.B. Jr (Eds.), Smell and Taste in Health and Disease. Raven Press, New York, pp. 405 416. Beidler, L.M., 1967. Anion inuences on taste receptor response. In: Hayashi, T. (Ed.), Olfaction and Taste II. Pergamon Press, New York, pp. 509 534. Beidler, L.M., Gross, G.W., 1971. The nature of taste receptor sites. In: Neff, W.D. (Ed.), Contributions to Sensory Physiology. Academic Press, New York, pp. 97 127. Bigiani, A., Roper, S.D., 1994. Reduction of electrical coupling between Necturus taste receptor cells, a possible role in acid taste. Neurosci. Lett. 176, 212 216. Boughter, J.D. Jr, Smith, D.V., 1998. Amiloride blocks acid responses in NaCl-best gustatory neurons of the hamster solitary nucleus. J. Neurophysiol. 80, 362 372. Buckler, K.J., Vaughan-Jones, R.D., 1994. Effects of hypercapnia on membrane potential and intracellular calcium in rat carotid body type-1 cells. J. Physiol. London 478, 157 171. Buckler, K.J., Vaughan-Jones, R.D., Peers, C., Lagadic-Gossmann, D., Nye, P.C.G., 1991. Effects of extracellular pH, PCO2 and HCO; on intracellular pH in isolated type-1 3 cells of the neonatal rat carotid body. J. Physiol. London 444, 703 721. Chu, S., Brownell, W.E., Montrose, M.H., 1995. Quarititative confocal imaging along the crypt-to-surface axis of colonic crypts. Am. J. Physiol. 269, C1557 C1564. Cummings, T.A., Kinnamon, S.C., 1992. Apical K+ channels in Necturus taste cells. Modulation by intracellular factors and taste stimuli. J. Gen. Physiol. 99, 591 613. Daikoku, H., Morisaki, I., Ogawa, Y., Maeda, T., Kurisu, K., Wakisaka, S., 1999. Immunohistochemical localization of carbonic anhydrase isozyme II in the gustatory epithelium of the adult rat. Chem. Senses 24, 255 261. Dean, J.B., Bayliss, D.A., Erickson, J.T., Lawing, W.L., Millhorn, D.E., 1990. Depolarization and stimulation of neurons in nucleus tractus solitarii by carbon dioxide does not require chemical synaptic input. Neuroscience 36, 207 216. DeSimone, J.A., Callaham, E.M., Heck, G.L., 1995. Chorda tympani taste response of rat to hydrochloric acid subject to voltage-clamped lingual receptive eld. Am J. Physiol. 268, C1295 C1300. Feldman, G.M., Lyall, V., Ereso, G.L., Phan, D., Desai, N., Montrose, M.M., DeSimone, J.A., 2000. Lingual surface pH affects intracellular pH (pHi) in polarized taste receptor cells (TRCs). Chem. Senses 25, 686. Frank, M.E., Bieber, S.L., Smith, D.V., 1988. The organization of taste sensibilities in hamster chorda tympani nerve bers. J. Gen. Physiol. 91, 861 896. Fujiyama, R., Miyamoto, T., Sato, T., 1994. Differential distribution of two Ca2 + -dependent and-independent K+ channels throughout receptive and basolateral membranes of bullfrog taste cells. Pugers Arch. 429, 285 290. Furue, H., Yoshii, K., 1997. In situ tight-seal recordings of taste substance-elicited action currents and voltage-gated

243

Ba currents from single taste bud cells in the peeled epithelium of mouse tongue. Brain Res. 776, 33 139. Ganzevles, P.G.J., Kroeze, J.H.A., 1987. Effects of adaptation and cross-adaptation to common ions on sourness intensity. Physiol. Behav. 40, 641 646. Gardener, R.J., 1980. Lipid solubility and the sourness of acids: implications for models of the acid taste receptor. Chem. Senses Flavor 5, 185 194. Gilbertson, T.A., Avenet, P., Kinnamon, S.C., Roper, S.D., 1992. Proton currents through amiloride-sensitive Na channels in hamster taste cells, Role in acid transduction. J. Gen. Physiol. 100, 803 824. Gilbertson, T.A., Roper, S.D., Kinnamon, S.C., 1993. Proton currents through amiloride-sensitive Na+ channels in isolated hamster taste cells: enhancement by vasopressin and cAMP. Neuron 10, 931 942. Goto, T., Komai, M., Bryant, B.P., Furukawa, Y., 2000. Reduction in carbonic anhydrase activity in the tongue epithelium and submandibular gland in zinc-decient rats. Int. J. Vitam. Nutr. Res. 70, 110 118. Heck, G.L., Mierson, S., DeSimone, J.A., 1984. Salt taste transduction occurs through an amiloride-sensitive sodium transport pathway. Science 223, 403 405. Hellekant, G., Danilova, V., Ninomiya, Y., 1997a. Primate sense of taste: behavioral and single chorda tympani and glossopharyngeal nerve ber recordings in the rhesus monkey, Macaca mulatta. J. Neurophysiol. 77, 978 993. Hellekant, G., Ninomiya, Y., Danilova, V., 1999b. Taste in chimpanzees II: single chorda tympani bers. Physiol. Behav. 61, 829 841. Herness, M.S., 1987. Effect of amiloride on bulk ow and iontophoretic taste stimuli in the hamster. J. Comp. Physiol. A 160, 281 288. Hettinger, T.P., Frank, M.E., 1990. Specicity of amiloride inhibition of hamster taste responses. Brain Res. 513, 24 34. Hiji, T., 1975. Selective elimination of taste responses to sugars by proteolytic enzymes. Nature 256, 427 429. Kinnamon, S.C., Roper, S.D., 1988. Evidence for a role of voltage-sensitive apical K+ channels in sour and salt taste transduction. Chem. Senses 13, 115 121. Kinnamon, S.C., Dionne, V.E., Beam, K.G., 1988. Apical localization of K+ channels in taste cells provides the basis for sour taste transduction. Proc. Natl. Acad. Sci. 85, 7023 7027. Kinnamon, S.C., Price, M.P., Stone, L.M., Lin, W., Welsh, M.J., 2000. The acid-sensing ion channel BNCl is not required for sour taste transduction. ISOT 2000, Brighton, UK. Proceedings Abst. 104, pp. 80. Kolesnikov, S.S., Bobkov, Yu.V., 2000. Regulation of the conductance and resting potential by extracellular K+ in frog taste receptor cells. Membr. Cell Biol. 13, 321 335. Komai, M., Bryant, B., Takeda, T., Suzuki, H., Kimura, S., 1994. The effect of topical treatment with a carbonic anhydrase inhibitor, MK-927, on the response of the chorda tympani nerve to carbonated water. In: Kurihara, K., Suzuki, N., Ogawa, H. (Eds.), Olfaction and Taste XI. Springer-Verlag, New York, p. 92.

244

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245 Ritucci, N.A., Dean, J.B., Putnam, R.W., 1997. Intracellular pH response to hypercapnia in neurons from chemosensitive areas of the medulla. Am. J. Physiol. 273, R433 R441. Ritucci, N.A., Chambers-Kersh, L., Dean, J.B., Putnam, R.W., 1998. Intracellular pH regulation in neurons from chemosensitive and nonchemosensitive areas of the medulla. Am. J. Physiol. 275, R1152 R1163. Sato, T., Beidler, L.M., 1982. The response characteristics of rat taste cells to four basic taste stimuli. Comp. Biochem. Physiol. A 73, 1 10. Schiffman, S.S., Lockhead, E., Maes, F.W., 1983. Amiloride reduces the taste intensity of Na+ and Li+ salts and sweeteners. Proc. Natl. Acad. Sci. USA 80, 6136 6140. Scott, T.R., Plata-Salaman, C.R., 1991. Coding of taste quality. In: Getchell, T.V., Doty, R.L., Bartoshuk, L.M., Snow, J.B. Jr (Eds.), Smell and Taste in Health and Disease. Raven Press, New York, pp. 345 368. Simon, S.A., Garvin, J.L., 1985. Salt and acid studies on canine lingual epithelium. Am. J. Physiol. 249, C398 C408. Smith, D.V., Frank, M.E., 1993. Sensory coding by peripheral taste bers. In: Simon, S.A., Roper, S.D. (Eds.), Mechanisms of Taste Transduction. CRC Press, Boca Raton, pp. 295 338. Smith, D.V., Liu, H., Vogt, M.B., 1996. Responses of gustatory cells in the nucleus of the solitary tract of the hamster after NaCl or amiloride adaptation. J. Neurophysiol. 76, 47 58. Spray, D.C., Bennett, M.V., 1985. Physiology and pharmacology of gap junctions. Physiol. Rev. 47, 281 303. Spray, D.C., Harris, A.L., Bennett, M.V., 1981. Gapjunctional conductance is a simple and sensitive function of intracellular pH. Science 211, 712 715. Stevens, D.R., Lindemann, B., 2000. Rat vallate taste cells exhibit a hyperpolarization activated cationic current which is enhanced by sour stimuli. ISOT 2000, Brighton UK. Proceedings Abst. 106, p. 82. Stewart, R.E., Lasiter, P.S., Benos, D.J., Hill, D.L., 1995. Immunohistochemical correlates of peripheral gustatory sensitivity to sodium and amiloride. Acta Anat. Basel 153, 310 319. Stewart, R.E., Lyall, V., Feldman, G.M., Heck, G.L., DeSimone, J.A., 1998. Acid-induced responses in hamster chorda tympani and intracellular pH tracking by taste receptor cells. Am. J. Physiol. 275, C227 C238. Sugimoto, K., Teeter, J.H., 1991. Stimulus-induced currents in isolated receptor cells of the larval tiger salamander. Chem. Senses 16, 109 122. Taylor, N.W., Farthing, F.R., Berman, R., 1930. Quantitative measurements on the acid taste and their bearing on the nature of the nerve receptor. Protoplasma 10, 84 97. Tennissen, A.M., McCutcheon, N.B., 1996. Anterior tongue stimulation with amiloride suppresses NaCl saltiness, but not citric acid sourness in humans. Chem. Senses 21, 113 120. Tonosaki, K., Funakoshi, M., 1984. lntracellular taste cell responses of mouse. Comp. Biochem. Physiol. A 78, 651 656.

Kretz, O., Barbry, P., Bock, R., Lindemann, B., 1999. Differential expression of RNA and protein of the three pore-forming subunits of the amiloride-sensitive epithelial sodium channel in taste buds of the rat. J. Histochem. Cytochem. 47, 51 64. Lauger, P., 1991. Electrogenic Ion Pumps. Sinauer Associ ates, Inc, Sunderland, MA. 8 Liljestrand, G., 1922. Uber den Schwellenwert des saveren Geschmacks. Arch. Neerland Physiol. 7, 532 537. Lyall, V., Biber, T.U.L., 1994. Potential-induced changes in intracellular pH. Am. J. Physiol. 266, F685 696. Lyall, V., Feldman, G.M., Heck, G.L., DeSimone, J.A., 1997. Effects of extracellular pH, PCO2, and HCO on 3 intracellular pH in isolated rat taste buds. Am. J. Physiol. 273, C1008 1019. Lyall, V., Ereso, G.L., Phan, D., Desai, N., Montrose, M.M., DeSimone, J.A., Feldman, G.M., 2000. Na+ H+ exchange activity in the basolateral membrane of taste receptor cells (TRCs). Chem. Senses 25, 686. Makhlouf, G.M., Blum, A.L., 1972. Kinetics of the taste response to chemical stimulation: a theory of acid taste in man. Gastroenterology 63, 67 75. Miyamoto, T., Okada, Y., Sato, T., 1988. Ionic basis of receptor potential of frog taste cells induced by acid stimuli. J. Physiol. London 405, 699 711. Miyamoto, T., Miyazaki, T., Okada, Y., Sato, T., 1996. Whole-cell recording from non dissociated taste cells in mouse taste bud. J. Neurosci. Methods 64, 245 252. Miyamoto, T., Fujiyama, R., Okada, Y., Sato, T., 1998. Sour transduction involves activation of NPPB-sensitive conductance in mouse taste cells. J. Neurophysiol. 80, 1852 1859. Ninomiya, Y., Funakoshi, M., 1988. Amiloride inhibition of responses of rat single chorda tympani bers to chemical and electrical tongue stimulations. Brain Res. 451, 319 325. Ogiso, K., Shimizu, Y., Watanabe, K., Tonosaki, K., 2000. Possible involvement of undissociated acid molecules in the acid response of the chorda tympani nerve of the rat. J. Neurophysiol. 83, 2776 2779. Okada, Y., Miyamoto, T., Sato, T., 1993. Contribution of proton transporter to acid-induced receptor potential in frog taste cells. Comp. Biochem. Physiol. A 105, 725 728. Ossebaard, C.A., Smith, D.V., 1995. Effect of amiloride on the taste of NaCl, Na-gluconate and KCl in humans: implications for Na+ receptor mechanisms. Chem. Senses 20, 37 46. Ozeki, M., 1971. Conductance change associated with receptor potentials of gustatory cells in rat. J. Gen. Physiol. 58, 688 699. Pineda, J., Aghajanian, G.K., 1997. Carbon dioxide regulates the tonic activity of locus coeruleus neurons by modulating a proton-and polyamine-sensitive inward rectier potassium current. Neuroscience 77, 723 743. Richards, T.W., 1898. The relation of the taste of acids to their degree of dissociation. Am. Chem. J. 20, 121 126.

J.A. DeSimone et al. / Respiration Physiology 129 (2001) 231245 Ugawa, S., Minami, Y., Guo, W., Saishin, Y., Takatsuji, K., Yamamoto, T., Tohyama, M., Shimada, S., 1998. Receptor that leaves a sour taste in the mouth. Nature 395, 555 556. Waldmann, R., Champigny, G., Bassilana, F., Heurteaux, C., Lazdunski, M., 1997. A proton-gated cation channel involved in acid-sensing. Nature 386, 173 177.

245

Wiemann, M., Baker, R.E., Bonnet, U., Bingmann, D., 1998. CO2-sensitive medullary neurons: activation by intracellular acidication. Neuroreport 9, 167 170. Ye, Q., Heck, G.L., DeSimone, J.A., 1994. Effects of voltage perturbation of the lingual receptive eld on chorda tympani responses to Na+ and K+ salts in the rat: implications for gustatory transduction. J. Gen. Physiol. 104, 885 907.