Acetylcholine is released from taste cells, enhancing taste signalling

Abstract

Acetylcholine (ACh), a candidate neurotransmitter that has been implicated in taste buds, elicits calcium mobilization in Receptor (Type II) taste cells. Using RT-PCR analysis and pharmacological interventions, we demonstrate that the muscarinic acetylcholine receptor M3 mediates these actions. Applying ACh enhanced both taste-evoked Ca²⁺ responses and taste-evoked afferent neurotransmitter (ATP) secretion from taste Receptor cells. Blocking muscarinic receptors depressed taste-evoked responses in Receptor cells, suggesting that ACh is normally released from taste cells during taste stimulation. ACh biosensors confirmed that, indeed, taste Receptor cells secrete acetylcholine during gustatory stimulation. Genetic deletion of muscarinic receptors resulted in significantly diminished ATP secretion from taste buds. The data demonstrate a new role for acetylcholine as a taste bud transmitter. Our results imply specifically that ACh is an autocrine transmitter secreted by taste Receptor cells during gustatory stimulation, enhancing taste-evoked responses and afferent transmitter secretion.

Key points

• Acetylcholine (ACh), a classical neurotransmitter, stimulates M3 muscarinic receptors on Receptor (Type II) taste bud cells

• ACh is synthesized by, and released from Receptor (Type II) taste bud cells during gustatory stimulation.

• This muscarinic autocrine feedback amplifies taste-evoked Ca²⁺ signals and enhances afferent neurotransmitter (ATP) release from Receptor (Type II) cells.

• Taste Receptor cells in mice lacking M3 muscarinic receptors display depressed sensitivity to gustatory stimulation

• The findings highlight a new signalling pathway in taste buds and may explain taste disturbances (i.e. side effects) associated with certain anticholinergic drugs.

Introduction

Mammalian taste buds utilize an array of neurotransmitters to analyse and transmit gustatory signals to the CNS in response to taste stimulation. While the full extent of these interactions is far from elucidated, roles have been revealed for ATP (Finger et al. 2005; Huang et al. 2007; Romanov et al. 2007), serotonin (Kaya et al. 2004; Huang et al. 2005), GABA (Cao et al. 2009; Dvoryanchikov et al. 2011; Huang et al. 2011), and noradrenaline (Huang et al. 2008a; Zhang et al. 2010), with many other candidate transmitters implicated. One taste transmitter that has not been extensively characterized to date is acetylcholine (ACh). Decades ago, acetylcholinesterase (AChE) was characterized in taste tissue and between individual taste cells (Macintosh, 1941), suggesting a role for ACh in taste buds. Later studies suggested the presence of muscarinic receptors in canine lingual epithelia (Simon & Baggett, 1992), and showed that ACh elicits calcium mobilization in a subset of taste cells (Ogura, 2002). Moreover, ACh receptors have been localized within the taste bud (Ogura & Lin, 2005; Eguchi et al. 2008; Oliveira-Maia et al. 2009; Rogachevskaja et al. 2010), implying that cholinergic signalling is utilized in taste transduction. There remains, however, no identification of which cells respond to ACh or the probable source(s) of ACh that evoke responses in taste cells. Consequently, there is yet no physiological role suggested for ACh during taste reception.

Taste buds consist of three principal cell types (Chaudhari & Roper, 2011), any one of which may represent a candidate target for ACh. Type I cells are glial-like in nature and ensheath the other two classes with thin lamellar processes. Type II cells express G-protein coupled receptors (GPCRs) for sweet, bitter and umami compounds, and communicate with afferent nerve fibres by secreting the excitatory neurotransmitter ATP through pannexin 1 channels (Huang et al. 2007; Romanov et al. 2007). Because of their role in transducing sweet, bitter and umami, these cells have been termed Receptor cells. Type III cells are the only cells that possess synapses, defined morphologically (Yang et al. 2000). Consequently, these cells have been termed Presynaptic cells. Presynaptic cells respond to sour (acid) taste stimulation (Richter et al. 2003; Huang et al. 2006; Huang et al. 2008b) as well as to ATP released by Receptor cells (Richter et al. 2003; Huang et al. 2006; Huang et al. 2007). Presynaptic cells secrete serotonin which inhibits taste responses in Receptor cells, thereby providing negative feedback during gustatory stimulation (Huang et al. 2009), possibly due to the more rudimentary non-synaptic release mechanism of ATP through pannexin channels, which may require fast desensitization due to their high conductivity. Thus, the taste bud functions as a complex unit, employing several transmitters in the generation of taste signals.

The function of ACh in taste buds is unresolved, though it plays a documented role in other peripheral chemosensory organs (Li & Matsunami, 2011; Krasteva et al. 2012). ACh has been hypothesized to be an efferent transmitter released from parasympathetic nerve fibres within the taste bud (Inoue et al. 1992). Indeed, evidence for cholinergic innervation of taste buds has been presented (Ogura et al. 2007). Alternatively, ACh may be stored and released from within the taste bud during gustatory stimulation. In this report, we employed scanning laser confocal calcium imaging and transmitter biosensors to examine the actions of ACh in mouse taste buds. Our results identify Receptor (Type II) taste cells as the source of ACh release in taste buds and highlight a cholinergic contribution to taste transduction.

Methods

Ethical approval

All experimental procedures were approved by the University of Miami Animal Care and Use Committee. Animals were killed via exposure to CO₂ followed by cervical dislocation.

Animals

Mice used in these experiments included C57BL/6 mice (wild-type); transgenic mice expressing enhanced green fluorescent protein (GFP) under control of the PLCβ2 promoter (PLCβ2-GFP mice) (Kim et al. 2006); transgenic mice expressing GFP under the GAD67 promoter (GAD-GFP mice) (Chattopadhyaya et al. 2004); and genetic knockout mice missing the muscarinic receptors M1 and M3, along with wild-type animals from the same mixed genetic background (a gift from Dr J. Wess, NIDDK, Bethesda, MD, USA). Receptor (Type II) taste cells selectively express GFP in PLCβ2-GFP mice, thereby facilitating their identification in living tissues (Kim et al. 2006). Similarly, the majority, but not all Presynaptic (Type III) cells fluoresce green in GAD-GFP mice (Tomchik et al. 2007).

Taste cell isolation

Lingual epithelium was peeled from the circumvallate papillae after injection with 2.5 mg ml⁻¹ dispase II (Roche), 1 mg ml⁻¹ trypsin inhibitor (Sigma) and 1 mg ml⁻¹ collagenase A (Roche). The epithelium was subsequently bathed in Ca²⁺-free Tyrode for 20 min, pinned, and whole taste buds collected with a fire polished micropipette. Taste buds were then transferred to a shallow recording chamber, secured with Cell-Tak (BD Biosciences, San Jose, CA, USA), and superfused with normal Tyrode solution.

Cellular biosensor technique

Acetylcholine biosensors consisted of CHO-K1 cells stably expressing M3 receptors (CRL-1981, ATCC, Manassas, VA, USA) and loaded with 5 μm Fura-2-AM (Rodriguez-Diaz et al. 2011). These cells are sensitive to ACh in the micromolar range (EC₅₀= 1.4 μm). Endogenous purinoceptors on the biosensors were desensitized by incubating the cells in 500 μm ATP for 60 min prior to the experiments (Huang et al. 2007). Biosensors did not respond to any of the taste stimuli utilized in this study, or to 5-HT (3 nm), noradrenaline (10 nm) or GABA (10 um), or to depolarization with 50 mm KCl. For ATP biosensors, we used Fura-2-loaded CHO cells expressing P2X2 and P2X3 receptors (Huang et al. 2007).

Experiments were performed on whole taste buds and on isolated taste cells. If the latter, taste cells were also loaded with Fura-2 and imaged along with biosensors. Biosensors were drawn onto a fire-polished micropipette with gentle suction and manipulated into contact with taste buds. Fura-2 images (F₃₄₀, F₃₈₀) were acquired ratiometrically using an Olympus (Tokyo, Japan) IX71 inverted fluorescence microscope with a 20× water immersion objective. Image analysis was carried out using Imaging Workbench 6.0 (Indec Biosystems, Santa Clara, CA, USA), with traces plotted as intracellular calcium concentration using a Ca²⁺ imaging calibration kit (Invitrogen, USA) (Grynkiewicz et al. 1985).

Lingual slice preparation

Lingual slices of 100 μm containing circumvallate papillae taste buds loaded with either Calcium Green Dextran (10 kDa), or Calcium Orange (both 2 mm in 50% DMSO, Invitrogen, Carlsbad, CA, USA) when GFP cellular markers was used were cut from blocks of lingual tissue using a vibrating blade microtome (Leica VT1000 S) (Caicedo et al. 2000). Slices were placed in a shallow recording chamber and attached to the coverslip base using Cell-Tak. Lingual slices were perfused at room temperature with Tyrode buffer containing elevated Ca²⁺ (8 mm; see below). Elevating Ca²⁺ improved the stability of the recordings and improved signal-to-noise ratio. We perfused the recording chamber at a rate of 2 ml min⁻¹ and focally applied taste stimuli to the apical tips of taste buds (see below), to ensure that taste solutions only contacted the apical pores of the taste buds, and not the parenchyma. Focal taste application and buffer perfusion at 2 ml min⁻¹ assured that tastant was entirely cleared from the bath in 15–30 s.

Scanning laser confocal Ca²⁺ imaging

In the lingual slice preparation, we imaged cells confocally to record Ca²⁺ transients in cells while maintaining sufficient z-resolution to ensure sampling from a single cell. All stimuli were dissolved in Tyrode solution. KCl and pharmacological agents were bath-applied. We applied taste stimuli to the taste pore via a 2 μm ‘puffer’ micropipette, using an MSC PLI-100 pressure source (Medical Systems, Greenvale, NY, USA). Taste stimuli were ejected by applying a pressure of 3 p.s.i. for 2 s. Responses from cells were recorded using an Olympus BX50WI laser scanning confocal microscope (Olympus Optical, Thornwood, NY, USA), scanning with argon and krypton lasers, and using the Fluoview image analysis package. Images were captured at 2 s intervals. Responses are presented as changes in relative fluorescence, ΔF/F₀: (F–F₀)/F₀. Photobleaching of the sample was corrected by plotting the gradual decline of signal over time (typically with an exponentially decreasing baseline) and subtracting this baseline from the response (Caicedo et al. 2000). Only cells that maintained consistent responses for the duration of experimentation were included in our analysis (see Data analysis, below). We identified taste cell types either physiologically or through GFP expression in PLCβ2-GFP or GAD-GFP mice.

RT-PCR analysis

RNA was extracted from isolated taste buds and from surrounding non-taste epithelium (DeFazio et al. 2006). Suitable positive control tissues were obtained and validated for genes of interest, with water acting as a negative control. Genomic contamination was removed from samples using DNase 1, and RNA was reverse transcribed with Superscript III (Invitrogen, Carlsbad, CA, USA). A sample the equivalent of 2 taste buds was used for each analysis, with all results run in triplicate. Amplification was carried out on an iCycler with reagents from Bio-Rad (Hercules, CA), for a total of 37 cycles for each gene of interest, with β-actin at 30 cycles and PLC-β2 at 35 cycles acting as controls.

Solutions and taste stimuli

Unless otherwise indicated, chemicals were purchased from Sigma (St Louis, MO, USA). Normal Tyrode solution contained the following (in mm): 145 NaCl, 5 KCl, 2 CaCl₂, 1 MgCl₂, 10 Hepes, 10 glucose, 10 sodium pyruvate, 5 NaHCO₃ pH 7.4, 310–320 mosmol l⁻¹. To prepare buffer with elevated calcium (for confocal calcium imaging) or with nominally zero calcium, we substituted equimolar sodium with calcium. Similarly, high-potassium Tyrode solution for depolarizing taste cells was prepared by substituting equimolar KCl (50 mm) for NaCl.

Taste stimuli used in these experiments consisted of a mixture of the following: an artificial sweetener, 100 μm SC45647 (2-[[[4-(aminomethyl)anilino]-[[(1R)-1 -phenylethyl]amino]methyl]amino]ethane-1,1-diol) (a gift from Dr G. Dubois, Coca Cola, Atlanta, GA, USA); bitter compounds, 30 μm cycloheximide and 200 μm denatonium; and umami compounds 200 mm monopotassium l-glutamate (MPG) with 1 mm inosine 5-monophosphate (IMP). Fluorescein (2 μm) was added to the tastant mixture when applied focally in the lingual slice preparation to track stimulation and to ensure consistent application of tastants. 1,1-Dimethyl-4-diphenylacetoxypiperidinium iodide (4-DAMP) was purchased from Tocris Bioscience (Ellisville, MO, USA) and MT7 from Peptides International (Louisville, KY, USA).

Data analysis

Baseline fluorescence was established for a minimum of 100 s before a stimulus was presented. We scored positive responses only if (1) the peak ΔF/F was at least two times the baseline fluctuation for that cell, (2) responses could be elicited at least twice in the same cell by the same stimulus and appeared approximately of the same amplitude, and (3) responses were cell-specific (i.e. were observed in specific cells, not in all dye-loaded cells in the field of view; and not recorded in background regions) (Caicedo et al. 2000). This assured that signals in the lingual slice preparation were not generated by tissue movement or a stimulus artifact. Data analyses were performed in Microsoft Excel, and Prism v.5 (Graphpad Software, Inc, La Jolla, CA, USA), with data displayed as in a manner appropriate for respective sample size (Drummond & Vowler, 2011).

Results

Acetylcholine elicits Ca²⁺ responses in taste receptor cells

Previous researchers have shown acetylcholine (ACh) or carbachol (CCh) elicits Ca²⁺ mobilization in mouse taste receptor cells (Ogura, 2002; Eguchi et al. 2008). We used a different approach – intact taste buds in a lingual slice – to examine this phenomenon in greater detail, and to localize the actions of ACh to apical (chemosensory) versus basolateral (synaptic) regions of taste buds.

Taste buds in lingual slices showed robust and repeatable Ca²⁺ responses to ACh (Fig. 1A), that were blocked with atropine (dashed line, 5 μm). Ca²⁺ responses were evoked by bath-applied ACh (which stimulates apical and basolateral regions of taste cells), but applying ACh focally onto the apical tips of taste cells was ineffective (Fig. S1). This finding suggests that ACh receptors are located on the basolateral membranes of taste cells, implicating a synaptic role in taste.

Figure 1. Acetylcholine (ACh) stimulates M3 muscarinic receptors on mouse taste bud cells.

Lingual slices of mouse circumvallate papillae containing intact taste buds were prepared as described (Caicedo et al. 2002; Dando & Roper, 2009) and were stimulated with bath-applied ACh. A, 1 μm ACh (arrow) mobilized Ca²⁺ in Receptor (Type II) taste cells (continuous line). In the presence of atropine (5 μm, dashed line), responses were much attenuated). Cells were loaded with Calcium Orange and identified as Receptor cells using GFP under the control of the PLCβ2 promoter. B, ACh-evoked responses (open circles) were blocked by atropine (5 μm) (filled circles), consistent with muscarinic receptor activation. P < 0.0001, Student's two-tailed paired t test. Grey bar and dotted line denote mean and 95% CI. C, blocking M3 receptors with 4-DAMP (5 μm, shaded area) inhibited ACh-evoked responses in taste Receptor cells. D, summary of experiments testing 4-DAMP (5 μm). P < 0.05, Student's two-tailed paired t test. E, MT7 (50 nm), an M1 antagonist, did not significantly affect ACh-evoked responses in Receptor cells. F, summary of data showing lack of effect of MT7 (50 nm) on ACh responses, P= 0.95, Student's two-tailed paired t test. A, C and E, ordinate = relative change in fluorescence (see Methods), B, D and F, ordinate =Δ[Ca²⁺]_i normalized to the level before antagonist. G, RT-PCR showed expression of M1, M3, and M5 ACh receptors in CV taste buds (TB), non-taste epithelium (NT), water (−, a negative control), and positive controls (+; M1, brain; M3, kidney; M5, brain). Only M3 was detected in taste buds at this sensitivity.

Consistent with previous reports (Ogura, 2002; Eguchi et al. 2008), we found ACh responses were blocked by atropine (5 μm) (Fig. 1B), indicating that ACh responses were generated by muscarinic ACh receptors. To date, there has been disagreement concerning which muscarinic receptors are responsible for the ACh sensitivity of taste buds (Ogura, 2002; Eguchi et al. 2008). Thus, we tested 4-DAMP (5 μm), an M3-selective receptor antagonist. 4-DAMP significantly reduced ACh-evoked calcium mobilization in taste cells (Fig. 1C and D). In contrast, the M1 antagonist MT7 (at 50 nm) had little to no effect (Fig. 1E and F). Conclusively, RT-PCR analysis of mRNA from isolated taste buds showed M3 but not M1 or M5 receptors are expressed in taste cells (Fig. 1G), consistent with the pharmacological testing. Collectively, the data show that ACh evokes Ca²⁺ mobilization in taste buds via M3 muscarinic receptors.

To localize the specific target(s) of ACh in taste buds, we isolated taste cells from transgenic mice expressing GFP in Receptor (Type II) taste cells (Kim et al. 2006). ACh-evoked Ca²⁺ mobilization was only recorded from GFP-expressing (Receptor) cells (8 of 19 cells). No ACh responses were recorded from cells lacking GFP (0 of 15 cells; P= 0.0045, Fisher's exact test).

ACh potentiates taste-evoked signals in Receptor (Type II) cells

We next investigated what role ACh might play during gustatory stimulation. Using the lingual slice preparation to study taste buds in a relatively intact environment (as opposed to taste buds or taste cells isolated in vitro), we recorded Ca²⁺ responses evoked by focal application of sweet, bitter, or umami taste stimuli to the taste pore. Such focal application of tastants reliably evokes Ca²⁺ mobilization in Receptor (Type II) taste cells (Tomchik et al. 2007; Dando & Roper, 2009). Importantly, taste-evoked responses to focal application were significantly enhanced by superfusing the preparation with ACh (1 μm) during taste stimulation (Fig. 2A and B).

Figure 2. ACh potentiates taste responses in taste buds.

Ca²⁺ mobilization in taste cells in the lingual slice preparation was assayed with laser scanning confocal Ca²⁺ imaging. Neurotransmitter secretion from isolated taste buds was measured with ATP biosensors. A, in the lingual slice preparation, taste-evoked Ca²⁺ mobilization was enhanced by perfusing the tissue with ACh (1 μm, present throughout shaded area). The traces show mean responses ± SEM from 15 cells. Arrows indicate focal application of taste stimulus. Note, perfusing ACh itself (shaded area) generated a small response, consistent with data in Fig. 1. This ACh response was taken into account by aligning the traces such that the maximum response to ACh was set to ΔF/F₀= 0, i.e. the baseline for taste-evoked responses was re-set to zero. Focally applied tastant responses were measured from that new baseline. Ordinate shows ΔF/F₀ from dye-loaded taste cells. B, summary of effects of ACh (1 μm) on taste-evoked responses. n= 15, P < 0.001, Student's two-tailed paired t test. Ordinate, Ca²⁺ responses normalized to the control stimulus. C, ATP biosensors were positioned near taste buds to detect taste-evoked neurotransmitter (ATP) release in the absence and in presence (shaded area) of 1 μm ACh. Traces show mean response ± SEM from 12 biosensors. Ordinate, [Ca²⁺]_i in ATP biosensors. D, summary of effects of ACh (1 μm) on taste-evoked ATP secretion, n= 15, P < 0.01, Student's two-tailed paired t test. Ordinate, biosensor responses normalized to control stimulus.

We then isolated taste buds from vallate papillae and tested whether ACh affects taste-evoked secretion of the taste neurotransmitter ATP (Finger et al. 2005; Huang et al. 2007; Romanov et al. 2007) using neurotransmitter biosensors (see Methods). Stimulating taste buds with a bitter/sweet taste solution elicited ATP release (Fig. 2C), and most importantly, this taste-evoked ATP release was significantly elevated by the presence of ACh (1 μm) (Fig. 2D).

ACh is synthesized in, and released from, Receptor (Type II) taste cells

We investigated potential source(s) of ACh to determine whether this transmitter might be intrinsic to taste buds. RT-PCR on mRNA isolated from taste buds showed that the biosynthetic enzyme choline acetyltransferase (ChAT) was expressed only in taste tissue and not in surrounding non-taste epithelium (Fig. 3A), indicating that ACh can be synthesized within the taste bud and may be stored there for release during taste reception.

Figure 3. Receptor (Type II) cells release ACh in response to taste stimulation.

A, RT-PCR showed mRNA for choline acetyltransferase (ChAT), the biosynthetic enzyme for ACh, is expressed at high levels in taste buds (TB), but not in the surrounding non-taste epithelium (NT). Controls, brain (+) and water (−). B, summary of data from 12 biosensor cells, testing ACh secretion from isolated taste buds in response to taste and KCl stimulation. Mean ± SEM shown. Bitter/sweet taste solution, but not KCl depolarization (50 mm) evoked ACh release. C, an ACh biosensor was positioned near an isolated Receptor (Type II) cell, identified using GFP-PLCβ2 mice (see Methods). D, as shown in C, a Receptor cell was apposed to an ACh biosensor cell. The traces show transmitter secretion was evoked by taste stimulation but not by depolarization with 50 mm KCl (bars). Responses in the isolated Receptor cell and biosensor were recorded concurrently (upper and lower traces, respectively). Ordinates indicate [Ca²⁺]_i in Receptor cell (upper trace) and in ATP biosensor (lower trace).

Acetylcholine biosensors are sensitive and specific detectors

To identify whether taste cells might indeed secrete ACh, we obtained CHO cells stably expressing highly sensitive muscarinic receptors and used these cells as ACh biosensors. ACh biosensors responded well to 1 μm ACh but were unaffected by taste stimulation, eliminating a potential source of artifacts (Supplemental Material, Fig. S2A). Furthermore, ACh-evoked responses in biosensors were not altered by the presence of taste stimuli (Fig. S2B). Because taste buds secrete several neurotransmitters, including ATP, noradrenaline, serotonin and GABA, we tested whether any of these transmitters stimulated ACh biosensors. None of these transmitters evoked a false positive (Fig. S2C and D). Parenthetically, purinoceptors endogenous to these biosensors had been desensitized, as described in Methods. Lastly, biosensor cells did not respond with Δ[Ca²⁺]_i to KCl depolarization (Figure S2D), in agreement with other biosensor cells constructed from genetically engineered CHO cells (Huang et al. 2007).

Stimulating taste buds with tastants reliably evoked biosensor responses, indicating ACh release. ACh biosensor responses could only be obtained when the biosensor was closely apposed to a taste bud; moving the biosensor even micrometres away from the taste bud eliminated taste-evoked biosensor responses. ACh release was observed when taste buds were stimulated with a taste mixture (sweet and bitter stimuli), but not when taste buds were depolarized with KCl (Fig. 3B). The taste mixture selectively stimulates Receptor (Type II) cells, whereas KCl depolarization is an effective stimulus only for Presynaptic (Type III cells) (DeFazio et al. 2006). These findings implicate Type II cells in the taste bud were the origin of taste-evoked ACh release.

To identify unambiguously which taste cells secrete ACh, we next isolated individual Receptor and Presynaptic cells from transgenic mice expressing GFP markers for these cell types and monitored taste-evoked ACh secretion with biosensors (Fig. 3C). Eight out of nine identified Receptor cells released ACh upon taste stimulation (Fig. 3D). In contrast, only one Presynaptic cell out of five elicited a biosensor response when the Presynaptic cell was stimulated (50 mm KCl) and even at that, the biosensor readout was weak, suggesting a false positive. The difference in incidence of ACh release from Receptor versus Presynaptic cells was significant at the P= 0.02 level (Fisher's exact test). These findings are consistent with the predominant source of ACh released during taste bud stimulation being mainly, if not exclusively Receptor (Type II) cells.

Cholinergic mechanisms during taste transduction

If ACh enhances ATP secretion (Fig. 2) and also if taste stimulation triggers Receptor cells to secrete ACh (Fig. 3), then one might expect that blocking cholinergic receptors during gustatory stimulation would lead to a reduced secretion of ATP. We tested this by measuring taste-evoked ATP secretion from taste buds before and after adding atropine to block muscarinic receptors. Indeed, on its own, atropine (5 μm) significantly and reversibly reduced taste-evoked ATP secretion (Fig. 4A and B). As a further test of the role of muscarinic enhancement of taste transmitter release and to verify that atropine was not directly interfering with transmitter secretion, we measured taste-evoked ATP secretion with and without atropine in taste buds isolated from genetically mutant littermates lacking M1 and M3 receptors (double knock out, DKO, mice) (Bymaster et al. 2003). Biosensor measurements showed that taste-evoked ATP secretion was on average 80% of that observed in wild-type animals in taste buds from double knockout mice. Importantly, atropine had no effect on taste-evoked ATP release in taste buds from M1/M3 DKO mice, in marked contrast to taste buds from wild-type mice (Fig. 4C–E).

Figure 4. ACh from Receptor cells enhances neurotransmitter (ATP) secretion during taste stimulation.

ATP biosensors were positioned near taste buds of wild-type (WT) and mutant mice lacking both M1 and M3 muscarinic receptors (KO) to monitor taste-evoked transmitter secretion, as in Fig. 2. A, in a WT mouse, taste-evoked ATP release was reduced by atropine (5 μm). Arrows indicate taste stimulation. Shaded area shows addition of atropine. (Parenthetically, atropine has no effect on ATP biosensors.) B, summary of effects of atropine in WT mice, from experiments as in A. n= 25, P= 0.02, Student's two-tailed paired t test. C, in an M1/M3 double knockout mouse, atropine (5 μm) did not affect taste-evoked ATP secretion. D, summary of experiments from M1/M3 double knockout mice, as in E. n= 32, P= 0.94, Student's two-tailed paired t test. E, comparison of actions of atropine on taste-evoked ATP release in wild-type (open bar) and double knockout (filled bar) mice; dashed line indicates response of same cell to taste without atropine. Mean+SEM shown by bar. Ordinates in A and C show [Ca²⁺]_i in biosensor cells, ordinates in B, D and E indicate biosensor responses normalized to control stimulus.

Discussion

Our findings show that ACh potentiates taste-evoked signals in, and augments taste-evoked transmitter (ATP) secretion from, Receptor (Type II) cells. Simon & Baggett (1992) first identified muscarinic receptors in lingual epithelium, with Ogura (2002) subsequently describing muscarinic Ca²⁺ mobilization in rat taste cells. However, distinctions between Receptor and Presynaptic cells had not been defined at that time and thus earlier studies were not able to more precisely identify the targets for ACh in taste buds. Further, muscarinic receptors in taste buds have alternately been reported as the subtype M1 (Ogura, 2002) or M3 (Eguchi et al. 2008) by immunocytochemistry and RT-PCR, respectively. The present findings, using pharmacology and RT-PCR, establish that taste Receptor cells respond to ACh via M3 receptors. Rogachevskaja and colleagues (2010) also recently reported that muscarinic receptors, most likely either M3 or M5, induced Ca²⁺ transients in mouse taste cells with an EC₅₀ of 2.5 μm. Their observations are in good agreement with our own, save for the fact that they electrophysiologically identified the cells as Type I, and not Type II taste cells.

The mechanism whereby activating muscarinic M3 receptors augments taste responses is not known. There are reports that muscarinic receptors interact with olfactory receptors to enhance odour-evoked responses (Firestein & Shepherd, 1992; Li & Matsunami, 2011). For example, Li & Matsunami (2011) co-applied odour stimuli with M3 ligands and showed that agonists increased, and antagonists attenuated odour-evoked responses in HEK 293T cells that were co-expressing ORs and M3, similar to our present findings using gustatory stimuli and native taste cells. Li & Matsunami reported that M3 formed stable functional complexes with the odorant receptor OR-S6, but further details of the mechanism of interaction remain to be reported.

Nicotinic receptors have been implicated in taste buds but have been studied mainly with reference to activation of bitter taste by nicotine (Dahl et al. 1997; Oliveira-Maia et al. 2009). We found no evidence that ACh elicited responses when applied to the apical (i.e. chemosensory) tips of taste cells. This could be attributed to the differing experimental conditions in which these studies were carried out, the taste field studied, or simply be due to the fact that as nicotine is a small, lipophilic molecule, it could conceivably reach additional receptors on the basolateral membranes of taste cells. Further, all the actions of ACh on taste cells were characterized by muscarinic agents and RT-PCR, with the presumed targets being basolaterally located synaptic receptors. The actions of nicotine as a bitter alkaloid taste may involve apical nicotinic receptors, and nAChR subunits have been reported in lingual tissues containing taste buds (Oliveira-Maia et al. 2009). Although we found no evidence for nicotinic synaptic receptors, we cannot completely rule out their presence. ACh has one of the shortest half-lives of neurotransmitters due to the exceptional efficiency of acetylcholinesterase. Choline, however, has no such restriction. As the M3 receptor subtype is sensitive to both ACh and to a lesser degree to choline (Kurzen et al. 2007), we cannot exclude the possibility of endocrine control of the taste bud through the action of choline. However, the vesicular acetylcholine transporter (VAChT) has been observed in TRPM5- and PLCβ2-positive cells (Eguchi et al. 2008). This would suggest that ACh is released via exocytosis from taste Receptor cells, in agreement with our results.

Exocytosis of ACh would likely be triggered by Ca²⁺ release from internal stores, as Receptor cells are known to lack voltage gated calcium channels in the mouse (Clapp et al. 2006). We cannot, of course, rule out cholinergic innervation of the taste bud from extrinsic sources such as autonomic input. Any such innervation would be eliminated during isolation of taste buds and thus not open to testing in our experiments. In support of that possibility, Ogura (2002) showed nerve fibres innervating taste buds were immunopositive for the vesicular acetylcholine transporter VAChT.

Our findings may help explain some of the alterations in taste perception associated with drugs used for treating asthma (Drewnowski, 1989). Unpleasant taste is a well-established side effect of atropine derivatives (e.g. ipratropium bromide) taken as inhaled bronchodilators (Newnham, 2001). If topically applied ipratropium bromide penetrates the lingual epithelium and into taste buds (which remains to be determined), it would block the endogenous action of ACh, reducing the sensitivity of Receptor (Type II) cells, and may be the source of such taste disturbances. Perversion of taste due to cholinergic dysfunctions or caused by use of cholinergic drugs might merit further study of peripheral gustatory sensory involvement.

Glossary

AChE

acetylcholine esterase

CCh

carbachol

ChAT

choline acetyltransferase

DKO

double knockout

GAD67

glutamate decarboxylase 67

GPCR

G-protein coupled receptor

M1/3/5

muscarinic receptor 1/3/5

MT7

muscarinic toxin 7

nAChR

nicotinic acetylcholine receptor

PLCβ2

phospholipase C, β2

TRPM5

transient receptor potential melastatin

VAChT

vesicular acetylcholine transporter

Author contributions

R.D. and S.R. conceived and designed these experiments, R.D. collected, and with S.R. analysed and interpreted, results, and R.D. and S.R. drafted the article. Both authors approved the final version of the manuscript.

Supplementary material

Supplementary Figure S1

Supplementary Figure S2