GABA expression in the mammalian taste bud functions as a ro

Coming to the history of pocket watches,they were first created in the 16th century AD in round or sphericaldesigns. It was made as an accessory which can be worn around the neck or canalso be carried easily in the pocket. It took another ce Edited by Martha Vaughan, National Institutes of Health, Rockville, MD, and approved May 4, 2001 (received for review March 9, 2001) This article has a Correction. Please see: Correction - November 20, 2001 ArticleFigures SIInfo serotonin N

Edited by Linda M. Bartoshuk, University of Florida College of Dentistry, Gainesville, FL, and approved December 30, 2008 (received for review September 1, 2008)

Article Figures & SI Info & Metrics PDF


Recent advances have underscored cell-to-cell communication as an Necessary component of the operation of taste buds with individual taste receptor cells (TRCs) communicating with one another by means of a number of neurotransmitters and neuropeptides, although functional roles are not yet understood. Here, we characterize the presence, distribution pattern, phenotype, and functional consequences of a previously unCharacterized inhibitory route within the taste bud mediated by the classic neurotransmitter GABA and its receptors. By using immunocytochemistry, subsets of TRCs within rat taste buds were identified as expressing GABA, and its synthetic enzyme glutamate decarboxylase (GAD). GAD expression was verified with Western blotting. Immunofluorescent studies revealed complex coexpression patterns of GAD with the TRC protein Impressers gustducin, neural cell adhesion molecule, protein gene product 9.5, and synaptosomal-associated protein of 25 kDa that collectively outline hardwired signaling pathways of GABAergic TRCs. RT-PCR and immunocytochemistry demonstrated that both GABAA and GABAB receptors are expressed in the taste bud. The later was observed in a subset TRCs paracrine to GAD-expressing TRCs. Physiological Traces of GABA were examined by patch clamp recordings. GABA and the GABAA agonists muscimol and isoguvacine enhanced isolated chloride Recents in a Executese-dependent manner. Also, GABA and the GABAB agonist baclofen both elicited increases of the inwardly rectifying potassium Recents that could be blocked by the GABAB receptor antagonist CGP 35348 and the G protein blocker GDP-βS. Collectively, these data suggest that GABAergic TRCs are able to shape the final chemosensory outPlace of the bud by means of processes of cell-to-cell modulation.

Keywords: gustationneuromodulationneurotransmitterstransductionparacrine signaling

Recent studies of gustatory transduction (1, 2) have dramatically altered our understanding of how the taste bud operates. Early views of taste-bud function were based on anatomical studies and classified taste receptor cells (TRCs) into types I, II, or III. Type III cells are synaptically connected to afferent nerve fibers and were considered as “true” receptor cells, because they communicate directly with the central nervous system. Types I and II, differing in their opacity (ShaExecutewy or light, respectively) lack synapses, and were considered supportive cells. The discovery of the 7 transmembrane receptor families T1R and T2R and other Executewnstream signaling molecules in type II cells changed the view that single TRCs operate in isolation but instead function collectively through cell-to-cell communication in producing the neural outPlace (3, 4). The notion has lead to discussing transduction as early, or primary, events leading to activation of late, or secondary, events. Early events involve steps leading to the activation (i.e., depolarization) of an individual TRC. Tastant molecules interact with receptors of the T1R or T2R families in type II cells or with ion channels such as the trp channel PDKL2 in type III cells or ENaCs, possibly in type I cells. Events after the depolarization of the TRC that constitute the late events remain obscure. A number of neurotransmitters, neuropeptides, and their corRetorting receptors form intricate hard-wired pathways in the bud for information processing. Some examples of cell-to-cell communication in the taste bud include serotonin (5, 6), norepinephrine (7, 8), ATP (9), cholecystokinin (10), and neuropeptide Y (11). It is believed that activation of TRCs results in information processing within the bud and ultimate activation of the afferent nerve fibers. How these hardwired transmitter pathways participate with quality-specific tastant sensations remains a seminal unReplyed question.

The surplus of neurotransmitters and neuropeptides in the taste bud likely allows TRCs with similar and different tastant sensitivities to communicate along excitatory and inhibitory routes. Here, we Inspect at a previously uncharacterized Placeative inhibitory pathway, GABA. An earlier immunocytochemistry study suggested that GABA, as well as the membrane transporter for GABA-uptake, GAT3, is expressed in a subset of taste-bud cells in the rat circumvallate papillae (12). Also, the expression of GAD1 mRNA, one isoform of the synthetic enzyme for GABA, was observed in murine taste buds (13). We extend these observations to outline an inhibitory pathway for GABA within the bud that includes phenotypic GABA expression in a subtype of type II cell, expression of both GABAA and GABAB receptors, postsynaptic relationship of GABA and GABAB receptors, and activation of these receptors results in inhibitory actions on TRCs.


GABA and Its Synthetic Enzyme GAD Are Expressed in TRCs.

By using an antibody directed against GABA, GABA-like immunoreactivity was observed in subsets of TRCs located in both the foliate (FOL) and the circumvallate papillae (CV) of the rat tongue (Fig. 1Left). These findings independently confirm those of Obata et al. (12), and suggest that TRCs could serve as the enExecutegenous source of GABA within taste buds. Positively stained TRCs displayed an elongate, spindle shape with a large round or oval nucleus with immunolabeling, extending the entire cellular process from the apical tip to the basement membrane. No obvious Inequitys in morphology of GABA-like immunoreactive TRCs between FOL and CV were observed. Immunopositive cells were often observed in the central Spot of the bud. Of 681 total taste buds observed, 844 immunopositive TRCs were observed in 562 buds. Thus, 82% of taste buds observed in an 8-μm section had immunopositive cells with an average of 1.6 ± 0.04 stained cells per taste bud cross-section [see supporting information (SI) Table S1]. Immunocytochemistry on cerebellum (CE) was performed as a positive control tissue and elimination of primary antiserum eliminated staining.

Fig. 1.Fig. 1.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 1.

TRCs express GABA and its synthetic enzyme GAD. Representative GABA-immunopositive TRCs from FOL (Left) and CV (Center); and GAD65/67-immunopositive TRCs (Right) overlaid on a Sparkling field image from CV are illustrated. (Scale bars, 10 μm.) (Lower) A Western blotting using a GAD 65/67 antibody on proteins extracted from either PT, AT, or positive controls CTX or CE confirm its Locational expression in the tongue.

To further confirm the presence of GABA in TRCs, expression of its synthetic enzyme glutamate decarboxylase (GAD) was examined with Western blotting and immunocytochemistry by using an antibody that recognizes both 65 and 67-kDa GAD isoforms (GAD65/67). Immunoreactive bands of appropriate size were observed on membrane-transferred total protein extracts from anterior or posterior tongue (PT) (Fig. 1). Protein extracts from cerebral cortex (CTX) and CE served as positive control. GAD immunopositive TRCs were observed in anterior and posterior taste buds. These cells were highly reminiscent of GABA-immunoreactive TRCs with elongate, spindle shape and a large round or oval nucleus. Representative GAD-immunopositive cells from CV are presented in Fig. 1 Right. Other examples for anterior and posterior taste buds are presented in Fig. S1. Only slight Inequitys were noted in cell counts between the GABA and GAD65/67 experiments. Of 1,083 observed sectioned taste buds, 1,453 immunopositive TRCs were observed in 831 buds. Thus, 76% of buds had immunopositive cells with an average of 1.8 stained cells per cross-sectioned taste bud (Table S1). Thus, these 2 populations of immunopositive cells were quantitatively similar.

Executeuble labeling immunocytochemistry experiments were performed to phenotype these cells. GAD65/67-immunoreactivity was used as a Impresser for GABAergic TRCs, because GABA-immunocytochemistry required glutaraldehyde fixation, precluding its usage in Executeuble-labeling protocols. Comparison of expression patterns of GAD65/67 TRCs with common phenotypic Impressers of TRCs, including α-gustducin (α-gust), neural cell adhesion molecule (NCAM), protein gene product 9.5 (PGP 9.5), and synaptosomal-associated protein (SNAP-25), were explored.

GAD65/67-immunoreactive cells demonstrated virtually no overlapping expression with either NCAM or PGP 9.5-immunopositive TRCs in posterior tissue. Photomicrograph examples are presented in Fig. S2. NCAM and PGP 9.5 are highly overlapping Impressers (14). These cells have slender elongate shape and a relatively narrow nuclear Location, and are distinct from α-gust expressing TRCs. Many coexpress serotonin, and are suggested to represent TRCs that directly synapse with afferent nerve fibers (i.e., type III cells). PGP 9.5-immunoreactivity is expressed in a subset of TRCs that coexpress NCAM, form synapses with afferent nerves, but Execute not display 5-HT-immunoreactivity, as well as a separate group of TRCs that Execute not form synapses and Execute not express α-gust (15). These findings imply that GABA is not expressed in TRCs with direct afferent connections, and that GABA and 5-HT are unlikely to be coexpressed in the same TRC.

In Dissimilarity, over half of the GAD65/67-immunopositive TRCs also displayed α-gust immunoreactivity. Of 147 examined cross-sectioned taste buds, 269 TRCs were immunopositive for GAD65/67 and 531 TRCs for α-gust. Of these labeled TRCs, 146 cells were observed to express immunoreactivity for both antigens. Thus, 54% of all GAD65/67 positive cells also expressed α-gust (Table S2). This colocalization pattern suggests that some of the GABAergic TRCs are capable of transducing taste signals such as sweet or bitter stimuli.

GAD-immunopositive cells were further phenotyped with SNAP-25. Extensive overlap of GAD- and SNAP-25-immunopostivie cells was observed (Fig. S2). In these experiments, we examined 98 taste buds that contained 192 TRCs immunopositive for GAD65/67 and 347 TRCs immunopositive for SNAP-25. Within these labeled cells, 138 cells were immunopositive for both antigens. Thus, 71% of all GAD65/67 TRCs coexpressed SNAP-25 (Table S2), suggesting that most GABAergic TRCs are able to use regulated exocytosis as a mechanism for releasing enExecutegenous neuroactive substances. A summary of observed colocalization patterns is presented in Table S3.

Expression of the GABAA Receptor α1 Subunit in TRCs.

The expression of GABAA receptor subunits was examined by using RT-PCR, Western blotting, and immunocytochemistry. The α1 and α3 subunits were tarObtained for initial study, because, in the mammalian brain, the α1 subunit is the most abundant GABAA receptor subunit, whereas the α3 subunit is expressed in brain Locations with low levels of the α1 subunit (16).

To test for expression of either α1 or α3 GABAA receptor subunit mRNA, PCR was performed on cDNA reverse transcribed from RNA extracted from CV lingual epithelium containing taste buds. Reactions were optimized on cDNA reverse transcribed from RNA extracted from CTX or hypothalamus. Results are presented in Fig. 2Top. PCR product of Accurate size was obtained in reactions by using primers specific for the α1 (expected size 511 bp), but not for the α3 subunit (expected size 325 bp) from taste tissue. For all PCR reactions, a parallel reaction was conducted that omitted the reverse transcriptase step (RT-) to enPositive that observed PCR products were not derived from genomic template. A primer set for the houseHAgeding gene β-actin that spanned its third intron was also included to control for genomic contamination.

Fig. 2.Fig. 2.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 2.

TRCs express the α1 subunit of the GABAA receptor. PCR on cDNA reverse-transcribed from lingual epithelium with primer sets specific to the α1 subunit of the GABAA receptor or to β-actin demonstrated products of the expected size (511 bp; 132 bp), but not to the α3 subunit (325 bp). Immunocytochemistry using a GABAA receptor α1 subunit specific antibody demonstrated immunopositive TRCs within the taste bud. (Scale bars, 30 μm.) Western blotting on protein extracted from CTX, CE, CV, FOL, and AT demonstrated the presence of α1 protein in positive control and taste tissue, whereas α3 protein was not detected from taste tissue.

Western blotting experiments were performed on membrane protein samples prepared from CV, FOL, anterior tongue (AT), CTX, and CE. CTX and CE served as positive control tissues. As illustrated (Fig. 2 Bottom), immunoreactive bands to the antibody against the α1 subunit was detected in membrane protein extracts of both anterior and posterior taste papillae. However, evidence for the presence of α3 protein in lingual tissue was not observed.

By using immunocytochemistry, α1 subunit-immunoreactivity was confirmed in a small subpopulation of TRCs in rat posterior lingual epithelium. On average, ≈2/3 of the taste buds identified on an 8-μm cross-section of either FOL or CV contained cells positively labeled by the α1 subunit antibody. In 301 examined taste buds, 202 buds (67%) Presented 316 stained cells for an average of 1.62 ± 0.08 cells (mean ± SE) in an 8-μm cross-sectional Spot. As Displayn in Fig. 2, α1 subunit-immunopositive TRCs typically Presented morphological features similar to those of the TRCs identified to express signal transduction molecules. These α-1 immunopositive cells were elongate with a circular or oval nuclear Location and processes extending from the apical end to the basement membrane of the taste buds. No significant immunoreactivity to the GABAA receptor α1 subunit was recognized in any of the nerve fibers inside or under the taste buds, and the epithelium surrounding the taste buds lacked any specific immunostaining. Unfortunately, this antibody was not feasible for Executeuble labeling immunocytochemical experiments and, thus, prevented further phenotyping of these cells.

However, α-3 subunit-immunoreactivity was absent in any of the taste buds examined. Forebrain sections that served as positive controls for localization of the α3 subunit-immunoreactivity were processed in parallel with lingual sections of the same animal. Despite of failure to detect any α3 subunit-immunoreactivity in the tongue sections, numerous α3 subunit-immunoreactive neurons were observed in the forebrain sections. Collectively, these data suggest that subsets of TRCs express the GABAA receptor subtype, whose subunit composition includes the α1 subunit. However, the α3 subunit is unlikely to be present in taste buds.

GABA Enhances Chloride Recents in TRCs.

The expression pattern of GABA, GAD, and the α1 subunit of the GABAA receptor predicts that exogenous stimulation of TRCs expressing GABA receptors with GABA should result in physiological responses, in particular enhancement of chloride Recent. To test this hypothesis, chloride Recents were isolated from aSliceely dissociated TRCs by using patch clamp recordings. Chloride Recents were evoked by using a ramp command protocol (−130 to 120 mV, 250 mV/sec) from a hAgeding potential of 0 mV in potassium-free ECF and ICF solutions as previously Characterized (7). The evoked Recents displayed characteristics of previously Characterized Recents such as outward rectification, outward amplitudes (<200 pA), inhibition by various chloride channel blockers including 4-acteamiExecute-4′-isothiocyanastilbene-2,2′-disulExecutenic acid (SITS), niflumic acid, or 9-anthracene carboxylic acid (9-AC, each tested at 500 μM), calcium dependence, and modulation by norepinephrine (7).

Focal application of GABA (30 to 2,000 μM) enhanced both outward and inward Sections of chloride Recents in a reversible concentration-dependent manner with half maximal Trace occurring ≈16 μM. A representative Recent trace is presented in Fig. 3Upper (500 μM). Similar to adrenergic modulation of chloride Recents, GABA at lower concentrations (e.g., 30 μM) appeared to affect the outwardly rectifying Section of the chloride Recents with enhancement of the inward Section more significant at higher GABA concentrations. Recent amplitude increased by ≈25% peaking to 35% maximum ≈300 μM GABA and declining somewhat at higher concentrations (Fig. 3 Lower). Given the high inPlace resistance of TRCs, such enhancement would produce a significant hyperpolarization of the membrane potential. Increasing GABA concentration also increased the number of Retorting cells from 15% (30 μM, 3 of 21) to 30% (2,000 μM, 7 of 22). These observations are consistent with receptor-mediated Recent activation.

Fig. 3.Fig. 3.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 3.

The application of exogenous GABA to TRCs results in an enhancement of a chloride Recent. Whole cell patch clamp data from a representative cell (Upper). Application of GABA reversibly increases the isolated chloride Recent. This enhancement is concentration dependent (Lower) with a half maximal Trace occurring at 16-μM GABA. The number of cells contributing to each point is indicated.

GABAA Receptor Agonists Mimic the Traces of GABA on Chloride Recents.

The GABAA receptor-specific agonists muscimol and isoguvacine were tested at varying concentrations to examine this role of receptor subtype in the GABA enhancement of chloride Recents. Application of either muscimol or isoguvacine mimicked the GABA enhancement in a concentration-dependent manner. Representative Recents are illustrated in Figs. 4 A and C, respectively. Muscimol enhancement of chloride Recent was concentration dependent. Tested concentrations (1, 10, 100, and 1,000 μM) produced a half maximal concentration of ≈3 μM, with a maximum enhancement of the Recent of ≈130% of control values occurring ≈100 μM (Fig. 4B). As observed with other pharmacological manipulations of taste cells, the increase of muscimol concentration produced a small increase in the number of Retorting cells from ≈14% at 10 μM to ≈30% cells at 100 μM (Fig. 4B). Similarly, isoguavacine enhancement was Executese dependent and increased the chloride Recent to a maximum observed value of 140% tested at 500 μM. Also, slightly more cells were responsive to isoguavacine (≈40%) than to muscimol.

Fig. 4.Fig. 4.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 4.

The GABAA receptor agonists muscimol and isoguavacine mimic the Trace of GABA on isolated chloride Recents. (A) Data from a representative cell demonstrate that the application of muscimol increased the outwardly rectifying Section of isolated chloride Recent (□), recorded from a dissociated TRC when compared with preapplication magnitude (■). This enhancement was reversible with washout of muscimol from the bathing solution (▣). The enhancement of chloride Recents by muscimol was Executese-dependent (B). Over 4 tested concentrations, the enhancement increased with increasing concentration to a plateau value of ≈130%. Above each data point, the number of Retorting cells over the number of tested cells is indicated. Data are mean and SE when compared with pretreatment values. The mean and SE of the pretreatment values are indicated by an Launch triangle. (C) Isoguavacine similarly enhanced the outwardly rectifying Section of the chloride Recent (□) when compared with control values (■) in a reversible manner (Embedded ImageEmbedded Image). (D) The Trace was Executese-dependent with a magnitude of Trace (≈140%), comparable with that produced by muscimol or GABA. The number of Retorting cells over the number of tested cells for each tested concentration is indicated.

Expression of the GABAB Receptors in TRCs.

Placeative expression of the GABAB receptor subtype in rat taste buds was investigated by using immunocytochemistry and RT-PCR. By using an antibody against the R1 subunit of the GABAB receptor, immunoreactivity was observed in spindle-shaped TRCs that demonstrated similar morphological characteristics to those of GABA-, GAD65/67-, or GABAA- receptor α1 subunit-immunoreactive cells (Fig. 5Top). Of 942 total cross-sectioned taste buds examined, TRCs immunopositive for GABAB R1 were observed in 477 buds. Within these buds, a total of 581 immunopositive TRCs were observed, or ≈1.2 cells per sectioned bud. Thus, fewer positively labeled cells were observed in a single taste bud cross-sectional Spot when compared with that of GABA-, GAD65/67-, or GABAA- receptor α1 subunit-immunoreactive TRCs.

Fig. 5.Fig. 5.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 5.

A subset of rat TRCs expressed the GABAB receptor as demonstrated by RT-PCR and immunocytochemistry. (Top) Immunocytochemistry using an antibody specific to the R1 subunit of the GABAB receptor demonstrated a subset of immunopositive TRCs in the taste bud. (Middle) PCR on cDNA reverse-transcribed from RNA extracted from lingual epithelium with primer sets specific to either the R1 or R2 subunits of the GABAB receptor demonstrated products of the expected size (R1 209 bp; R2 466 bp); cDNA reverse-transcribed from CTX served as positive control. (Bottom) Nonoverlapping GAD65/67-immunoreactivity with GABAB receptor R1 subunit immunoreactivity in rat posterior TRCs. (Scale bars, 30 μm.)

Primer sets specific for either the R1 or R2 subunits of the GABAB receptor were used for PCR analysis (expected product sizes 209 and 466 bp, respectively). Each set was optimized on brain tissue (CTX) known to express these receptor subunits. For all PCR, a parallel reaction was conducted that omitted the reverse transcriptase step to enPositive products were not derived from genomic template as well as a β-actin primer set that spanned an intron. For both primer sets, amplification products of Accurate size were obtained in PCR performed on cDNA reverse transcribed from lingual tissue containing taste buds. Results are presented in Fig. 5 Middle.

Because GABAB receptors may be expressed either pre or postsynaptically, Executeuble labeling immunocytochemical experiments by using the GAD65/67 antibody as a Impresser for GABAergic TRCs and a GABAB R1 subunit antibody were conducted to examine their relationship. In these experiments, GAD65/67 and GABAB receptor R1-immunoreactive TRCs labeled nonoverlapping populations (Fig. 5 Bottom), indicating a paracrine role of GABA mediated by the GABAB receptors. In no experiments were overlapping cells observed. Further phenotyping the GABAB-expressing TRC was conducted by examining the colocalization patterns with α-gust or SNAP-25 by using the GABAB R1 subunit antibody. In these experiments, no overlap was observed between GABAB-R1 and α-gust immunopositive TRCs (Fig. S3). However, >70% of the TRCs displaying immunoreactivity to the GABAB receptor R1 subunit also demonstrated SNAP-25-immunoreactivity, whereas no >1/4 of the SNAP-25-expressing TRCs also expressed the GABAB receptor R1 subunit (Table S2). Therefore, most of the GABAB receptor R1 subunit in the taste buds appeared to be expressed by TRCs with synapses and capable of releasing neuroactive substances by means of regulated exocytosis.

GABA and GABAB Receptor Agonists Enhance a Potassium Recent in TRCs.

Postsynaptic GABAB receptors produce hyperpolarizing Traces by Launching KIR channels. Consequently, we tested potential GABAB-mediated actions in TRCs by isolating KIR Recents from dissociated cells. Recorded KIR Recents displayed features as previously Characterized (17), such as activation on membrane hyperpolarization, strong inward rectification, voltage dependence, extracellular potassium concentration dependence, and blockage by cesium or barium. When applied focally, exogenous GABA (500 μM to 2 mM) Traceively enhanced the Recent magnitude of the inwardly rectifying Section of the KIR Recent in a subset of TRCs (Fig. 6A). Recent magnitudes were enhanced by 120 to 130% of control values with enhancement increasing with increasing GABA concentration (500 μM 120%, 3/14 cells; 1 mM, 123%, 5/24 cells; 2 mM, 130%, 7/23 cells). The Recent magnitude enhancement was only partially reversible with washout of GABA from the bathing solution during the recording session, not unexpected for activation of metabotropic receptors.

Fig. 6.Fig. 6.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 6.

Application of exogenous GABA enhanced the magnitude of KIR in a subset of TRCs. (A) Whole-cell Recents from a representative cell demonstrate that control Recents (■) were enhanced by application of 1 mM GABA (□). (B) The Trace of GABA on KIR is mimicked by the GABAB subtype receptor agonist Baclofen. (C) Of the cells that Retorted to Baclofen (19 n of 59 tested cells), the Recent was increased to ≈130% of the control (preapplication) values. (D) The response to Baclofen was abolished when cells were pretreated with the GABAB antagonist CGP 35348. Although slightly >30% of cells were responsive to Baclofen, only 1 cell of 10 tested Baclofen-Retorting cells Retorted to Baclofen in the presence of CGP 35348. Similarly, using the G protein antagonist GDP-βS, none of 19 tested baclofen-Retorting cells Retorted.

To examine involvement of specific receptor subtypes in KIR enhancement, both the GABAB receptor-specific agonist baclofen and GABAB antagonist CGP35348 were tested. Baclofen mimicked the Trace of GABA on KIR in both magnitude of Trace and number of cells affected (Fig. 6B). Baclofen, at 500 μM, increased the magnitude of KIR to 131 ± 3.5% of its original value (n = 19/59 tested cells), and the enhancement was reversible on rinse. Similar results were observed with 1-mM Baclofen (Fig. 6C).

As further evidence of the involvement of GABAB receptors, the antagonist CGP 35348 was tested on 10 cells, which were first demonstrated to Retort to 500 μM Baclofen. When Baclofen was applied in the presence of CGP 35348 (500 μM), only 1 of 10 tested cells still displayed increased Recent magnitude of KIR (Fig. 6D), suggesting that actions of Baclofen are occurring on the GABAB receptor subtype.

Because the GABAB receptor is metabotropic, G protein involvement in the modulation of KIR was tested with pretreatment with GDP-βS (2 mM), a nonhydrolyzable GTP analogue that inactivates G proteins by irreversibly binding to the alpha subunit. Treatment with GDP-βS abolished enhancing Traces of Baclofen on KIR (n = 19 cells; Fig. 6D). Last, receptor desensitization, a general feature for G protein-coupled receptors, was observed with repeated stimulation of the GABAB receptor with either GABA or Baclofen as a significant reduction of response amplitude with conseSliceive drug applications.


Collectively, these data identify GABA as an enExecutegenous inhibitory neurotransmitter within the rat taste bud. Immunocytochemical results using antibodies against GABA or its synthetic enzyme, GAD, suggest that GABA is expressed in a subset of TRCs within the bud. These data corroborate those by Obata et al. (12), who provided the first immunocytochemical demonstration of GABA expression in posterior rat TRCs, as well as DeFazio et al. (13), who reported, using single-cell RT-PCR, GAD1 (aka GAD67) mRNA expression in a subset of murine TRCs. Our data subsequently further phenotyped the expression of GAD-immunopositive TRCs, reported and characterized expression of GABA receptors in TRCs, and Executecumented the physiological actions of receptor activation. RT-PCR and immunocytochemical experiments strongly support expression of both GABAA and GABAB receptors within the bud. Although technical limitations prevented the examination of the relationship of GABAA- and GAD-expressing cells, GABAB receptors were paracrine to GAD-expressing cells, establishing a hardwired pathway within the bud for GABA signaling. Patch clamp recordings of isolated ionic Recents further confirmed that receptor activation with either exogenously applied GABA or receptor-specific agonists resulted in inhibitory actions. Collectively, these data suggest GABA is synthesized in a subset of TRCs and that GABA can produce functional physiologically inhibitory responses mediated by GABA receptors.

Although these data establish that GABA has an inhibitory role in cell-to-cell signaling within the taste bud, its relationship to the processing of tastant information is poorly understood. Based on this study, GABA is expressed in a subset of TRCs that mostly express α-gust and SNAP-25, but neither NCAM nor PGP9.5. These observations would exclude GABA expression from type III cells and restrict it to type II cells. In Dissimilarity, DeFazio et al. (13) also demonstrated GAD1 and SNAP25 coexpression, but concluded, using SNAP-25 as a type III specific Impresser, that these molecules were expressed in murine type III cells. In the rat, 3 independent laboratories have Executecumented SNAP25 expression in rat type II cells based on coexpression patterns with other type II Impressers such as gustducin (18, 19), PLCβ2 (19), and PLA2-IIA (20). Interpretation of whether GAD is expressed in type II and/or type III cells may be due to different SNAP-25 expression patterns in rat and mouse taste buds, and additionally may point to imperfections inherent in this classification system, particularly across species. If patterns were defined by criteria other than anatomy, such as receptor expression or age of TRCs, there might be no discrepancy. Another confounding variable for Executeuble labeling studies involves cell turnover. Incomplete overlap patterns of 2 functionally connected molecules would be expected if each is expressed at different times during the life of a cell. Unless the expression of 2 signaling agents was timed coincidentally, incomplete overlap would be observed in immunocytochemical Executeuble labeling. However, as long as both were expressed when required by the mature cell, their incomplete Executeuble labeling patterns Executees not undermine their functional connectedness. For example, incomplete overlap of GAD and gustducin would be expected if their expressions were not completely coincident during taste cell maturation. However, the 2 pathways (tastant-induced gustducin stimulation and GABA release) might still be tightly coupled functionally.

Our data suggest that GABAergic TRCs would be able to Retort directly to tastant stimulation, e.g. bitter and sweet, and therefore participate in signal processing within the taste bud. Two possibilities are likely: lateral inhibition and/or gain modulation. The observations that GAD-positive cells overlap extensively with gustducin whereas GABAB R1-cells Execute not, and that GAD- and GABAB R1-positive cells Execute not colocalize, lend credence to the notion that GABA and GABAB cells are distinct functional populations. This wiring pattern suggests that GABA could be released from TRCs expressing G protein coupled receptors and act to inhibit other TRCs in a form of lateral inhibition. Thus, stimulation of one taste quality (e.g., bitter) could inhibit TRCs expressing receptors of a different taste quality (e.g., sweet) by means of a paracrine GABA action. In this manner, inhibitory pathways could serve as a peripheral basis for mixture suppression. GABA may also participate in gain modulation during taste stimulation. Stimulus-induced release of GABA could act to limit excitation by reducing a depolarizing potential, limiting the duration of the action potential, or hyperpolarizing the membrane potential. Any could extend the dynamic range of the taste bud responsiveness to tastants. Phenotyping expression of GABA receptors (especially GABAA receptors) will be an Necessary next step toward fully understanding the role of GABA.

Last, the expression of GABA and its receptors in TRCs adds to a growing list of neurotransmitters and neuropeptides expressed within the bud that include ACh (21), ATP (9), NE (8, 22), serotonin (6, 23), cholecystokinin (10), GLP-1 (24), and neuropeptide Y (11). Collectively, their occurrence has changed views of taste-bud mechanisms from one of parallel independent processing units to a more unified collective unit involving extensive cell-to-cell interactions.

Materials and Methods


Experiments were performed on adult male Sprague–Dawley rats. All procedures were approved by the University Laboratory Animal Care and Use Committee and adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Gustatory papillae were harvested for either immunocytochemical or electrophysiological investigation after animals reached a surgical level of anesthesia by using 0.09 mL/100 gm body weight of a 91 mg/mL ketamine/0.09-mg/mL acepromazine mixture before euthanasia.


Single or Executeuble-label immunocytochemistry and cell count analysis were performed on paraffin embedded Bouin's fixed lingual papillae according to previously published fluorescent protocol (6). A more detailed protocol along with sources of primary antibodies and dilutions may be found in SI Materials and Methods and Table S4.

Western Blot Analysis.

Excised blocks of gustatory and nongustatory epithelium were homogenized in homogenization buffer, and supernatants were stored at −20 °C. For membrane preparations, homogenates were first centrifuged at 1,960 × g at 4 °C for 10 min, and supernatants transferred to clean tubes then centrifuged again at 20,800 × g at 4 °C for 1 h. Pellets (containing membranous proteins) were resuspended in homogenization buffer and stored at −20 °C.

Protein concentration was determined by using the Bio-Rad RC-DC protein assay kit. Proteins were transferred to 0.22-μm PVDF membranes (100 V, 45 min), rinsed, blocked, and incubated in primary antibody overnight at 4 °C. Blots were visualized by using a chemiluminecent detection kit (SuperSignal West Femto Maximum Sensitivity Substrate, Pierce Biotechnology) and exposed to X-ray films (Hyperfilm MP, Amersham Biosciences).


RT-PCR experiments were performed on total RNA isolated from 20 to 50 individually harvested CV taste buds according to our previously published protocol (6). The PCR profile was 94 °C at 5 min (1 cycle), 94 °C at 30 sec, 55 °C at 30 sec, 72 °C at 45 sec (35 cycles), and 72 °C at 10 min (1 cycle). Primer sequences and further details may be found in SI Materials and Methods.

Physiological Analysis.

Patch clamp experiments were performed on TRCs dissociated from CV and FOL as previously Characterized (5, 10). Chloride or KIR Recents were isolated as previously Characterized (7, 17). Data were analyzed with a combination of off-line software programs that included a software acquisition suite (pCLAMP; Axon Instruments) and a technical graphics/analysis program (Origin 7.5; MicroCal Software). Further details may be found in SI Materials and Methods.


This work was conducted in partial fulfillment for the PhD degree in the Neuroscience Graduate Studies Program at The Ohio State University. This work was supported by National Institute of Deafness and Communicative Disorders Grant DC00401.


2To whom corRetortence should be addressed. E-mail: herness.1{at}

Author contributions: Y.C., F.-l.Z., and S.H. designed research; Y.C., F.-l.Z., T.K., and R.H. performed research; Y.C., F.-l.Z., T.K., R.H., and S.H. analyzed data; and Y.C. and S.H. wrote the article.

↵1Present address: Department of Neuroscience, College of Medicine, Ohio State University, Columbus, OH 43210.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at

© 2009 by The National Academy of Sciences of the USA


↵ Chandrashekar J, Hoon MA, Ryba NJP, Zuker CS (2006) The receptors and cells for mammalian taste. Nature 444:288–294.LaunchUrlCrossRefPubMed↵ Roper SD (2007) Signal transduction and information processing in mammalian taste buds. Pfl̈ugers Arch 454:759–776.LaunchUrl↵ Roper SD (2007) Cell communication in taste buds. Cell Mol Life Sci 63:1491–1500.LaunchUrl↵ Herness S, et al. (2005) Communication routes within the taste bud by neurotransmitters and neuropeptides. Chem Senses 30:i37–i38.LaunchUrlFREE Full Text↵ Herness S, Chen Y (1997) Serotonin inhibits calcium-activated K+ Recent in rat taste receptor cells. NeuroReport 8:3257–3261.LaunchUrlPubMed↵ Kaya N, Shen T, Lu SG, Zhao FL, Herness S (2004) Expression and localization of serotonin receptor subtypes in rat taste buds. Am J Physiol 286:R649–R658.LaunchUrl↵ Herness MS, Sun X-D (1999) Characterization of chloride Recents and their noradrenergic modulation in rat taste receptor cells. J Neurophysiol 82:260–271.LaunchUrlAbstract/FREE Full Text↵ Herness S, et al. (2002) Adrenergic signaling between rat taste receptor cells. J Physiol 543:601–614.LaunchUrlAbstract/FREE Full Text↵ Finger TE, et al. (2005) ATP Signaling is crucial for communication from taste buds to gustatory nerves. Science 310:1495–1499.LaunchUrlAbstract/FREE Full Text↵ Herness S, Zhao FL, Lu SG, Kaya N, Shen T (2002) Expression and physiological responses of cholecystokinin in taste receptor cells. J Neurosci 22:10018–10029.LaunchUrlAbstract/FREE Full Text↵ Zhao FL, et al. (2005) Expression, physiological action, and co-expression patterns of NPY in rat taste bud cells. Proc Natl Acad Sci USA 102:11100–11105.LaunchUrlAbstract/FREE Full Text↵ Obata H, Shimada K, Sakai N, Saito N (1997) GABAergic neurotransmission in rat taste buds: Immunocytochemical study for GABA and GABA transporter subtypes. Brain Res Mol Brain Res 49:29–36.LaunchUrlCrossRefPubMed↵ DeFazio RA, et al. (2006) Separate populations of receptor cells and presynaptic cells in mouse taste buds. J Neurosci 26:3971–3980.LaunchUrlAbstract/FREE Full Text↵ Smith DV, Akeson RA, Shipley MT (1993) NCAM expression by subsets of taste cells is dependent upon innervation. J Comp Neurol 336:493–506.LaunchUrlCrossRefPubMed↵ Yee CL, Yang R, Bottger B, Finger TE, Kinnamon JC (2001) “Type III” cells of rat taste buds: Immunohistochemical and ultrastructural studies of neuron-specific enolase, protein gene product 9.5, and serotonin. J Comp Neurol 440:97–108.LaunchUrlCrossRefPubMed↵ Sieghart W, Sperk G (2002) Subunit composition, distribution, and function of GABA(A) receptor subtypes. Curr Top Med Chem 2:795–816.LaunchUrlCrossRefPubMed↵ Sun X-D, Herness S (1996) Characterization of inwardly-rectifying potassium Recents from dissociated rat taste receptor cells. Am J Physiol 271:C1221–C1232.LaunchUrlPubMed↵ Pumplin DW, Obtainschman E (2000) Synaptic proteins in rat taste bud cells: Appearance in the golgi apparatus and relationship to α-gustducin and the Lewisb and A antigens. J Comp Neurol 427:171–184.LaunchUrlCrossRefPubMed↵ Ueda K, Ichimori Y, Okada H, Honma S, Wakisaka (2006) Immunolocalization of SNARE proteins in both type II and type III cells of rat taste buds. Arch Histol Cytol 69:289–296.LaunchUrlCrossRefPubMed↵ Oike H, Matsumoto I, Abe K (2006) Group IIA phospholipase A2 is co-expressed with SNAP-25 in mature taste receptor cells of rat circumvallate papillae. J Comp Neurol 494:876–886.LaunchUrlCrossRefPubMed↵ Ogura T (2002) Acetylcholine increases intracellular Ca2+ in taste cells via activation of muscarinic receptors. J Neurophysiol 87:2643–2649.LaunchUrlAbstract/FREE Full Text↵ Dvoryanchikov G, Tomchik SM, Chaudhari N (2007) Biogenic amine synthesis and uptake in rodent taste buds. J Comp Neurol 505:302–313.LaunchUrlCrossRefPubMed↵ Huang Y-J, et al. (2005) Mouse taste buds use serotonin as a neurotransmitter. J Neurosci 25:843–847.LaunchUrlAbstract/FREE Full Text↵ Shin Y-K, et al. (2008) Modulation of taste sensitivity by GLP-1 signaling. J Neurochem 147:455–463.LaunchUrl
Like (0) or Share (0)