Taste Receptor Cell Responses to the Bitter Stimulus Denatonium Involve Ca2+ Influx Via Store-Operated Channels

Tatsuya Ogura, Robert F. Margolskee, Sue C. Kinnamon


Previous studies in rat and mouse have shown that brief exposure to the bitter stimulus denatonium induces an increase in [Ca2+]i due to Ca2+ release from intracellular Ca2+ stores, rather than Ca2+influx. We report here that prolonged exposure to denatonium induces sustained increases in [Ca2+]i that are dependent on Ca2+ influx. Similar results were obtained from taste cells of the mudpuppy, Necturus maculosus, as well as green fluorescent protein (GFP) tagged gustducin-expressing taste cells of transgenic mice. In a subset of mudpuppy taste cells, prolonged exposure to denatonium induced oscillatory Ca2+responses. Depletion of Ca2+ stores by thapsigargin also induced Ca2+ influx, suggesting that Ca2+store-operated channels (SOCs) are present in both mudpuppy taste cells and gustducin-expressing taste cells of mouse. Further, treatment with thapsigargin prevented subsequent responses to denatonium, suggesting that the SOCs were the source of the Ca2+ influx. These data suggest that SOCs may contribute to bitter taste transduction and to regulation of Ca2+ homeostasis in taste cells.


A variety of chemical compounds induce bitter taste responses via different mechanisms (see reviews inGilbertson et al. 2000; Glendinning et al. 2000). One pathway involves activation of the T2R/TRB G protein-coupled membrane receptors (Adler et al. 2000;Chandrashekar et al. 2000; Matsunami et al. 2000; Ming et al. 1998). T2Rs activate gustducin, a chemosensory-specific heterotrimeric G protein composed of α-gustducin (McLaughlin et al. 1992), and its partners, β3γ13 (Huang et al. 1999). α-Gustducin activates phosphodiesterase (PDE), causing decreases in intracellular cAMP, while its partners activate phospholipase C (PLCβ2), to produce inositol-1,4,5-trisphosphate (IP3) and diacylglycerol (DAG) (Huang et al. 1999; Rossler et al. 1998; Yan et al. 2001). While the physiological consequences of the reduced cAMP are not clear, IP3 binds to Type III IP3receptors (Clapp et al. 2001) and elicits a release of Ca2+ from intracellular stores (Akabas et al. 1988; Ogura et al. 1997). These studies were conducted with brief stimulus exposures in Ca2+-free solutions to demonstrate the involvement of intracellular stores in the [Ca2+]i response. In this study we used a prolonged application of the bitter stimulus denatonium in the presence of extracellular Ca2+ to determine if Ca2+ influx also contributes to bitter transduction.

We used isolated taste cells of mudpuppy, Necturus maculosus, as well as taste cells of transgenic mice expressing green fluorescent protein (GFP) under the control of the α-gustducin promoter (Wong et al. 1999). The rationale for using mudpuppy taste cells is that more than 80% of the taste cells respond to denatonium with an IP3-mediated release of Ca2+ from intracellular stores (Ogura et al. 1997), while <5% of mammalian taste cells respond to denatonium (Caicedo and Roper 2001). We report here that prolonged exposure to denatonium results in Ca2+ influx that is likely mediated by store-operated channels.


Taste cells from Necturus lingual epithelium (Ogura et al. 1997) and mouse circumvallate papillae (Gilbertson et al. 1993) were isolated as described previously and plated onto Cell Tak-coated coverslips for Ca2+ imaging. [Ca2+]i in taste receptor cells was measured using the Ca2+-sensitive dye fura-2 AM (Ogura et al. 1997). Briefly, cells were loaded with fura-2 AM (∼2 μM; Molecular Probes). Images were acquired with an intensified CCD camera through an oil-immersion objective lens (Fluor 40×, 1.3 NA, Nikon) of an inverted microscope. Images obtained with excitation at 350 and 380 nm were captured every 2 s to record fast responses, or at 5 s intervals during slow responses or under control conditions. Averaged Ca2+ levels over the entire cell area were plotted as F350/F380 versus time. Amphibian physiological saline contained the following (in mM): 112 NaCl, 2 KCl, 8 CaCl2, 3 HEPES, buffered to pH 7.2 with NaOH. Tyrode's solution contained the following (in mM): 140 NaCl, 5 KCl, 1 CaCl2, 1 MgCl2, 10 Na pyruvate, 10 glucose, 10 HEPES, buffered to pH 7.2 with NaOH. Ca2+-free solutions contained 1 mM EGTA without CaCl2. Denatonium was dissolved in extracellular saline and bath applied for periods ≤3 min. This prolonged stimulus paradigm is consistent with the prolonged time-intensity psychophysical profiles that have been obtained for bitter stimuli (Cubero-Castillo and Noble 2001).


Denatonium induces Ca2+ influx in taste cells

In mudpuppy taste cells, sustained application of 2.5 mM denatonium resulted in a transient Ca2+ response followed by a sustained phase that lasted more than several minutes (Fig. 1 A). When external Ca2+ was removed, the sustained phase disappeared, suggesting that Ca2+ influx was involved. When extracellular Ca2+ was returned to the medium, the Ca2+ influx returned (Fig.1 A). Similar responses were obtained in all denatonium-responsive taste cells tested (n = 12). In a subset of denatonium-responsive taste cells (5 of 12), prolonged stimulation resulted in oscillations of [Ca2+]i (Fig.1 B). The intensity and frequency of oscillations were variable among cells, with some cells showing dramatic responses and other cells showing little if any oscillation. The oscillations were superimposed on the elevated baseline, with a frequency of 1–5/min. The oscillatory response also disappeared in Ca2+-free solution (Fig. 1 B), suggesting that oscillations require the presence of extracellular Ca2+.

Fig. 1.

Dependency of the Ca2+ responses to denatonium on external Ca2+. Stimulation with denatonium induced a transient increase in [Ca2+]i followed by a sustained (a) or oscillatory (b) response in mudpuppy taste cells, and a sustained response in gustducin-expressing taste cells of mouse (c). Concentration of denatonium was 2.5 mM for (a) and (b), and 1 mM for (c). The sustained and oscillatory responses disappeared in Ca2+-free saline, suggesting that Ca2+influx is required. In the presence of denatonium, changing the solution from Ca2+-free to normal saline induced increases in [Ca2+]i.

To determine if Ca2+ influx is also involved in response to denatonium in mammals, we recorded Ca2+ responses from isolated taste cells of transgenic mice expressing GFP under the control of the α-gustducin promoter. Since denatonium activates α-gustducin, we recorded selectively from GFP expressing taste cells of circumvallate papillae. The results were similar to those of mudpuppy. Denatonium (1 mM) induced a transient increase in [Ca2+]i, followed by a sustained phase (Fig. 1 C), although only a subset of gustducin-expressing cells responded to denatonium (7 of 30 tested). These data indicate that only a fraction of gustducin-expressing taste cells express functional taste receptors for denatonium. In Ca2+-free solution, the sustained phase disappeared, suggesting that the sustained phase requires Ca2+ influx. The sustained phase reappeared when Ca2+ was returned to the bath in the presence of denatonium (Fig. 1 C).

Ca2+ store-operated channels are present in taste cells

As shown in Fig. 1, A and C, the sustained phase was generated slowly compared with the initial transient increases in [Ca2+]i due to release from Ca2+ stores. These data suggest that the sustained phase may be induced by Ca2+entry through Ca2+ store-operated channels, which are activated by Ca2+ store depletion, rather than by IP3 or other second messengers (Parekh and Penner 1997). To test this, we treated denatonium-responsive mudpuppy taste cells with thapsigargin (1 μM) in the absence of extracellular Ca2+ to deplete Ca2+ stores. After a transient elevation of intracellular Ca2+ due to passive leak from intracellular sores, intracellular Ca2+ levels returned to baseline. When extracellular Ca2+ was returned, a significant increase in [Ca2+]i occurred, due to Ca2+ influx (Fig.2 A; n = 19 cells tested). Denatonium stimulation following treatment with thapsigargin did not cause a further rise in intracellular Ca2+ (Fig. 2 A), suggesting that Ca2+ influx is not enhanced further by receptor-mediated increases in DAG and IP3. These data indicate that store-operated channels are present in mudpuppy taste cells; however, they do not unequivocally prove that they mediate the Ca2+ influx in response to bitter stimulation. In addition to βγ-activation of PLC, α-gustducin activates PDE to reduce intracellular cAMP, and Ca2+ influx through a cyclic nucleotide-suppressible cation conductance has been suggested to result from activation of α-gustducin (Kolesnikov and Margolskee 1995). If the decrease in cAMP is responsible for the Ca2+ influx, then cAMP should decrease the Ca2+ influx in response to denatonium. However, prolonged stimulation with 100 μM 3-isobutyl-1-methylxanthine (IBMX), a phosphodiesterase inhibitor, had no effect on the sustained Ca2+ response (n = 3 cells; data not shown). In addition, treatment with the PLC inhibitorU73122 blocked both the transient and the sustained increases in Ca2+ in response to denatonium (Ogura et al. 1997), showing that IP3-induced store depletion is required for the Ca2+ influx in the denatonium response.

Fig. 2.

Depletion of Ca2+ stores induces Ca2+ influx. Direct depletion of intracellular Ca2+ stores with thapsigargin also induced increases in [Ca2+]i when external Ca2+ was added in taste cells of mudpuppy (a) and gustducin-expressing taste cells of mice (b). The same cells showed sustained increases in Ca2+ in response to stimulation with bitter compound denatonium. These data strongly suggest that store-operated channels mediate the Ca2+influx in response to bitter stimulation.

We performed similar experiments with GFP expressing taste cells in transgenic mice. Depletion of Ca2+ stores by thapsigargin (1 μM) induced an increase in [Ca2+]i, which was absent in Ca2+-free saline and resumed when external Ca2+ was restored (Fig. 2 B). Similar responses to thapsigargin occurred in all GFP-expressing taste cells tested (n = 6), which suggests that depletion of Ca2+ stores induce Ca2+influx and the channel mediating the Ca2+ influx is present in all gustducin-expressing taste cells. These data taken together strongly suggest that store-operated channels are present in gustducin-expressing taste cells of mouse as well as taste cells of mudpuppy.


Our data present the first evidence that responses of mouse and mudpuppy taste cells to the bitter stimulus denatonium involve Ca2+ influx in addition to release of Ca2+ from intracellular stores. Also, we show that Necturus taste cells often generate oscillatory Ca2+ responses to denatonium, which also require Ca2+ influx. Although the Ca2+ influx is most apparent during the sustained phase, even the transient response is decreased in many taste cells in Ca2+ free (i.e., Fig. 1), suggesting that the Ca2+ influx begins during the Ca2+ release phase of the response. These data suggest that Ca2+ influx plays a role in bitter taste transduction, in that it contributes to the increase in intracellular Ca2+, which would regulate synaptic transmission to the afferent nerve. Further studies will be required to demonstrate direct evidence for the role of Ca2+influx via store-operated Ca2+ channels in taste transduction.

Experiments with thapsigargin show that store-operated Ca2+ channels are present in denatonium-responsive taste cells of mudpuppy and mouse. It is likely that the store-operated channels are responsible for the Ca2+ influx, since treatment with thapsigargin inhibited subsequent responses to denatonium. Involvement of voltage-gated Ca2+ currents in the Ca2+ influx is unlikely in mudpuppy taste cells, since denatonium hyperpolarizes these cells (Ogura et al. 1997). However, we cannot rule out the participation of other ion channels, particularly in mouse, since detailed pharmacological manipulation could not be performed due to the scarcity of bitter responses. One of the functions of store-operated Ca2+ channels is refilling the Ca2+ stores (Parekh and Penner 1997). Therefore Ca2+ influx in response to denatonium is likely also to be involved in long-term Ca2+ homeostasis. Rapid refilling of stores may be important for repetitive responses to bitter stimuli.

Recently, a specific transient receptor potential (TRP) channel, TRP-T, was identified in mammalian taste cells (Perez et al. 2002). TRP-T is co-expressed in taste cells with α-gustducin, γ13, PLCβ2; and Type III IP3 receptor (Clapp et al. 2001; Perez et al. 2001). Is TRP-T the store-operated channel that mediates bitter compound denatonium-stimulated Ca2+ influx? Recent evidence suggests that store-operated Ca2+ influx may be mediated by TRP channels (Birnbaumer et al. 2000). Although the mechanism of their activation is not clear, activation of PLC or Ca2+ store depletion appears to be required for their activation. Our data are consistent with these requirements. Further experiments will be required to demonstrate the precise role of TRP-T in Ca2+ influx following bitter stimulation.


R. F. Margolskee is an Associate Investigator of the Howard Hughes Medical Institute.

This work was supported by National Institute on Deafness and Other Communication Disorders Grants DC-00766 and DC-00244 to S. C. Kinnamon and DC-03055 and DC-03155 to R. F. Margolskee.


  • Address for reprint requests: T. Ogura, Dept. of Anatomy and Neurobiology, Colorado State University, Fort Collins, CO 80523 (E-mail: Tatsuya.Ogura{at}colostate.edu).


View Abstract