Physiological activation of human and mouse bitter taste receptors by bile acids

Abstract

Beside the oral cavity, bitter taste receptors are expressed in several non-gustatory tissues. Whether extra-oral bitter taste receptors function as sensors for endogenous agonists is unknown. To address this question, we devised functional experiments combined with molecular modeling approaches to investigate human and mouse receptors using a variety of bile acids as candidate agonists. We show that five human and six mouse receptors are responsive to an array of bile acids. Moreover, their activation threshold concentrations match published data of bile acid concentrations in human body fluids, suggesting a putative physiological activation of non-gustatory bitter receptors. We conclude that these receptors could serve as sensors for endogenous bile acid levels. These results also indicate that bitter receptor evolution may not be driven solely by foodstuff or xenobiotic stimuli, but also depend on endogenous ligands. The determined bitter receptor activation profiles of bile acids now enable detailed physiological model studies.

Fig. 1. Chemical structures of the investigated bile acids.

The structural formulas of the eight bile acids cholic acid, chenodeoxycholic acid, deoxycholic acid, glycocholic acid, lithocholic acid, taurocholic acid, thaurolithocholic acid, and ursodeoxycholic acid in their deprotonated form at physiological conditions (pH ~ 7) are presented. Hydroxyl groups at positions 7 and 12 are highlighted in bold and blue. Structures were generated with ChemDraw software.

Fig. 2. Human bitter taste receptor responses to bile acids.

Fluorescence traces of HEK293T-Gα16gust44 cells transiently transfected with expression constructs of the human bitter taste receptors TAS2R1 (1), TAS2R4 (4), TAS2R14 (14), TAS2R39 (39), and TAS2R46 (46). Cells were exposed to cholic acid (CA), taurocholic acid (TCA), glycocholic acid (GCA), deoxycholic acid (DCA), lithocholic acid (LCA), taurolithocholic acid (TLCA), chenodeoxycholic acid (CDCA) and ursodeoxycholic acid (UDCA). Fluorescence changes were measured with an automated fluorometric imaging plate reader (FLIPR^TETRA). Fluorescence traces are negative control corrected. Applied bile acid concentrations are given in µM in brackets. Only traces of responsive receptors are shown. A scale bar is provided at the bottom right.

Fig. 3. Mouse bitter taste receptor responses to bile acids.

Fluorescence traces of HEK293T-Gα16gust44 cells transiently transfected with expression constructs of the mouse bitter taste receptors Tas2r105 (105), Tas2r108 (108), Tas2r117 (117), Tas2r123 (123), Tas2r126 (126), and Tas2r144 (144). Cells were exposed to cholic acid (CA), taurocholic acid (TCA), glycocholic acid (GCA), deoxycholic acid (DCA), lithocholic acid (LCA), taurolithocholic acid (TLCA), chenodeoxycholic acid (CDCA) and ursodeoxycholic acid (UDCA). Fluorescence changes were measured with an automated fluorescence plate reader (FLIPR^TETRA). Fluorescence traces are negative control corrected. Applied bile acid concentrations are given in µM in brackets. Only traces of responsive receptors are shown. A scale bar is provided at the bottom right.

Fig. 4. Concentration-response relationships of eight tested bile acids with human TAS2R1.

HEK293T-Gα16gust44 cells were transiently transfected with human TAS2R1 (triangle, blue) and an empty vector control (circle, blue). Individual data points are depicted accordingly by black symbols. Receptor activation was recorded by increasing fluorescence intensities upon Ca²⁺ - release using an automated fluorometric imaging plate reader (FLIPR^TETRA). For dose-response relationships, increasing concentrations of the bile acids cholic acid a), taurocholic acid b), glycocholic acid c), chenodeoxycholic acid d), deoxycholic acid e), lithocholic acid f), taurolithocholic acid g) and ursodeoxycholic acid h) were applied. The relative fluorescence intensities were mock subtracted and plotted against the bile acid concentration in µM (n = 3 biologically independent experiments). Data are presented as the mean ± standard deviation (STD). Beginning statistical significance (p < 0.01) is indicated by (*).

Fig. 5. Comparison of the efficacies of bile acids with prototypical TAS2R agonists.

Human bitter taste receptors TAS2R1 a), TAS2R4 b), TAS2R14 c), TAS2R39 d), and TAS2R46 e) activated by highest applied bile acid concentrations (cholic acid (CA), taurocholic acid (TCA), glycocholic acid (GCA), taurolithocholic acid (TLCA), chenodeoxycholic acid (CDCA), deoxycholic acid (DCA), lithocholic acid (LCA), ursodeoxycholic acid (UDCA)) are presented (n = 3). For comparison, maximal signal amplitudes (ΔF/F) obtained with control stimuli of the corresponding TAS2Rs were added. The control stimuli were: 1 mM picrotoxinin (P) for TAS2R1, 3 mM colchicine (C) for TAS2R4, 10 µM aristolochic acid (AA) for TAS2R14, 3 mM denatonium benzoate (DB) for TAS2R39 and 10 µM strychnine (S) for TAS2R46. Data are presented as the mean ± standard deviation (STD). Individual data points are depicted by black triangles.

Fig. 6. Concentration-response relationships for the activation of mouse Tas2rs.

HEK293T-Gα16gust44 cells were transiently transfected with the murine Tas2r105 (triangle, blue), Tas2r108 (square, blue) or Tas2r144 (diamond, blue) and an empty vector control (circle, blue). Individual data points are depicted accordingly by black symbols. Receptor activation was recorded by increasing fluorescence intensities upon Ca²⁺ - release using an automated fluorometric imaging plate reader (FLIPR^TETRA). For dose-response relationships, increasing concentrations of the bile acids lithocholic acid a) and taurolithocholic acid b) and c) were applied. The relative fluorescence intensities were mock subtracted and plotted against the bile acid concentration in µM (n = 3 biologically independent experiments). Data are presented as the mean ± standard deviation (STD). Beginning statistical significance (p < 0.01) is indicated by (*).

Fig. 7. 2D and 3D representations of the putative binding mode of lithocholic acid in the TAS2R1 binding site obtained with MM/GBSA refinement.

The 2D plot a) was generated using the Ligand Interaction Diagram tool available in Maestro (Schrödinger Release 2022-3) showing residues at 4 Å from the ligand. In the 3D representation b), the ligand is shown as blue ball&stick, polar residues in CPK-colored sticks and hydrophobic residues as orange sticks. Hydrogen bonds are shown as dashed magenta lines in both representations.