The role of GABA in modulation of taste signaling within the taste bud

Abstract

Taste buds contain 2 types of GABA-producing cells: sour-responsive Type III cells and glial-like Type I cells. The physiological role of GABA, released by Type III cells is not fully understood. Here, we investigated the role of GABA released from Type III cells using transgenic mice lacking the expression of GAD67 in taste bud cells (Gad67-cKO mice). Immunohistochemical experiments confirmed the absence of GAD67 in Type III cells of Gad67-cKO mice. Furthermore, no difference was observed in the expression and localization of cell type markers, ectonucleoside triphosphate diphosphohydrolase 2 (ENTPD2), gustducin, and carbonic anhydrase 4 (CA4) in taste buds between wild-type (WT) and Gad67-cKO mice. Short-term lick tests demonstrated that both WT and Gad67-cKO mice exhibited normal licking behaviors to each of the five basic tastants. Gustatory nerve recordings from the chorda tympani nerve demonstrated that both WT and Gad67-cKO mice similarly responded to five basic tastants when they were applied individually. However, gustatory nerve responses to sweet-sour mixtures were significantly smaller than the sum of responses to each tastant in WT mice but not in Gad67-cKO mice. In summary, elimination of GABA signalling by sour-responsive Type III taste cells eliminates the inhibitory cell-cell interactions seen with application of sour-sweet mixtures.

Keywords: Gamma-aminobutyric acid; Glutamate decarboxylase; Sour; Sweet; Taste buds; Taste mixture.

Fig. 1

Krt5^Cre mice were useful to induce gene recombination in taste tissues. Detection of fluorescent proteins in taste tissues of Krt5^CreGad67^GFP/+Rosa26^{lsl−Tom/lsl−Tom} mice. Green: GFP fluorescence. Magenta: Tomato fluorescence. CV: circumvallate papillae. Scale bar, 10 µm

Fig. 2

Lack of expression of GAD67 protein in taste buds of Gad67-cKO mice. Immunohistochemical detection of GAD67 in taste buds of Gad67^GFP/+ mice (WT) and Krt5^CreGad67^GFP/flox mice (KO). A-D. Immunostaining for GAD67 and GFP expression in fungiform (A, B) and circumvallate papillae (C, D) of a Gad67^GFP/+ mouse (A, C) and a Krt5^CreGad67.^GFP/flox mouse (B, D). Summarized data are shown in Table 1. Green: GFP fluorescence, Magenta: immunoreactivity (IR) for GAD67. N = 3 animals. Scale bar, 10 µm (A, B) or 20 µm (C, D)

Fig. 3

Expression of taste cell markers was not impaired in taste buds of Gad67-cKO mice. Immunohistochemical detection of taste cell markers in taste buds of Gad67^GFP/+ mice (WT) and Krt5^CreGad67^GFP/flox mice (KO). Immunostaining for ENTPD2 and GFP expression in fungiform (FP) and circumvallate papillae (CV) of a Gad67^GFP/+ mouse (A) and a Krt5^CreGad67^GFP/flox mouse (B). Immunostaining for Gustducin and GFP expression in FP and CV of a Gad67^GFP/+ mouse (C) and a Krt5^CreGad67^GFP/flox mouse (D). Immunostaining for CA4 and GFP expression in FP and CV of a Gad67^GFP/+ mouse (E) and a Krt5^CreGad67^GFP/flox mouse (F). Summarized data are shown in Table 1. Green: GFP fluorescence, Magenta: immunoreactivity (IR) for ENTPD2, Gustducin or CA4. N = 3 animals. Scale bar, 10 µm

Fig. 4

Lack of GAD67 in taste buds did not affect gustatory nerve responses to single tastants. A. Sample recordings of chorda tympani nerve responses of WT (upper) and Gad67-cKO mouse (lower). Taste stimuli were NH₄Cl (100 mM), HCl (30 mM), citric acid (30 mM), sucrose (300 mM), MSG (300 mM), NaCl (300 mM), quinine (20 mM). B-H. Concentration–response relationships of chorda tympani nerve responses of WT mice (red circle) and Gad67-cKO mice (black triangle) for HCl (WT: n = 9, cKO: n = 10), citric acid (WT: n = 11, cKO: n = 9), acetic acid (WT: n = 8, cKO: n = 9), sucrose (WT: n = 10, cKO: n = 9), MPG (WT: n = 9, cKO: n = 9), NaCl (WT: n = 12, cKO: n = 11), quinine (WT: n = 8, cKO: n = 9). Gustatory nerve responses were normalized to the response to 100 mM NH₄Cl. Values indicated are means ± S.E.M. Statistical differences were analyzed by two-way ANOVA tests (Table 2)

Fig. 5

Lack of GAD67 in taste buds did not affect behavioral lick responses to single tastants. Number of licks of 30–1000 mM NaCl (A), 0.01–3 mM quinine (B), 30–1000 mM sucrose (C), 10–300 mM MSG (D), 1–100 mM HCl (E), 1–100 mM citric acid (F) and 10 µM capsaicin in the short-term (5 s) lick test. Red circle: WT mice (n = 7), Blue rectangle: Gad67^flox/flox mice (n = 11), Green diamond: Krt5^CreGad67^flox/flox mice (n = 9), Black triangle: Krt5^CreGad67^flox/floxTrpv1.^−/− mice (n = 7). Values indicated are means ± S.E.M. Statistical differences were analyzed by two-way ANOVA tests (Table 3) or one-way ANOVA with post hoc Tukey HSD test. ***: P < 0.001

Fig. 6

Gad67-cKO mice showed greater responses to sweet–sour mixtures. A. Sample recordings of chorda tympani nerve responses of WT (left) and Gad67-cKO mice (right). Taste stimuli were NH₄Cl (100 mM), sucrose (Suc; 500 mM), HCl (10 mM), sucrose + HCl (Mix; 500 mM, 10 mM, respectively). B. Chorda tympani nerve responses to glucose (Glc; 500, 1000 mM), sucrose (Suc; 500 mM), Sucralose (Sucra, 10, 20 mM), saccharin (Sac; 10 mM) and HCl (10 mM) in WT (red) and Gad67-cKO mice (black). C, D. Comparison between chorda tympani nerve responses to sweet–sour mixtures (500, 1000 mM glucose, 500 mM sucrose, 10, 20 mM sucralose or 10 mM saccharin + 10 mM HCl, left, light color) and sum of sweet and sour responses (right, dark color) in WT (C) and Gad67-cKO mice (D). Gustatory nerve responses were normalized to the response to 100 mM NH₄Cl. Values indicated are means ± S.E.M. Statistical differences were analyzed by Student’s t-test. *: P < 0.05, **: P < 0.01, ***: P < 0.001