The discovery of a new nonbile acid modulator of Takeda G protein-coupled receptor 5: An integrated computational approach

Abstract

The Takeda G protein-coupled receptor 5 (TGR5), also known as GPBAR1 (G protein-coupled bile acid receptor), is a membrane-type bile acid receptor that regulates blood glucose levels and energy expenditure. These essential functions make TGR5 a promising target for the treatment of type 2 diabetes and metabolic disorders. Currently, most research on developing TGR5 agonists focuses on modifying the structure of bile acids, which are the endogenous ligands of TGR5. However, TGR5 agonists with nonsteroidal structures have not been widely explored. This study aimed at discovering new TGR5 agonists using bile acid derivatives as a basis for a computational approach. We applied a combination of pharmacophore-based, molecular docking, and molecular dynamic (MD) simulation to identify potential compounds as new TGR5 agonists. Through pharmacophore screening and molecular docking, we identified 41 candidate compounds. From these, five candidates were selected based on criteria including pharmacophore features, a docking score of less than 9.2 kcal/mol, and similarity in essential interaction patterns with a reference ligand. Biological assays of the five hits confirmed that Hit-3 activates TGR5 similarly to the bile acid control. This was supported by MD simulation results, which indicated that a hydrogen bond interaction with Tyr240 is involved in TGR5 activation. Hit-3 (CSC089939231) represents a new nonsteroidal lead that can be further optimized to design potent TGR5 agonists.

Keywords: INT‐777; TGR5; molecular docking; nonbile acid; pharmacophore.

Figure 1

Bile acid and nonbile acid modulators of farnesoid X receptor (FXR) and Takeda G protein‐coupled receptor 5 (TGR5).^{[ 1} , ^{11 ]}

Figure 2

(a, b) Interaction between INT‐777 and TGR5. (c) Core structure of cholic acid with differences in R¹, R², and R³. (R indicates functional group; dashed green line shows hydrogen bond interaction; dashed purple line indicates hydrophobic interaction).

Figure 3

(a) The three‐dimensional (3D) arrangements of the selected pharmacophore models: model‐24, model‐29, and model‐31. (b) The deletion of hydrophobic features in C13 of the three models.

Figure 4

Docking score distribution histogram of 1625 candidate molecules. The red box indicates compounds with equal or better (lower) values of docking score compared to the reference ligand INT‐777.

Figure 5

Best poses of the five hit compounds in the ligand‐binding domain (LBD) of Takeda G protein‐coupled receptor 5 (TGR5) (the top left corner image of all hits on the LBD of TGR5; dashed green line shows hydrogen bond interaction).

Figure 6

Two‐dimensional (2D) interaction diagrams of five hit compounds in the ligand‐binding domain of Takeda G protein‐coupled receptor 5 (TGR5). (Dashed green line shows a hydrogen bond interaction; dashed purple line indicates a hydrophobic interaction).

Figure 7

Takeda G protein‐coupled receptor 5 (TGR5) activation in transfected HepG2 cells. (a) All hit compounds at 10 µM. (b) Concentration‐response curves for the activation by lithocholic acid (LCA) and Hit‐3. EC₅₀ values were calculated using nonlinear fitting of concentration–response curves. Values are presented as means ± SD from three independent experiments performed in triplicates. *p < 0.05 Hit‐3 versus NT; ***p < 0.001 LCA versus Hit‐3.

Figure 8

Hydrogen bond occupancies (%) to Takeda G protein‐coupled receptor 5 (TGR5) amino acid residues important for the lithocholic acid (LCA) binding (occupancy > 10%). Occupancies expressed as mean ± SD over three independent replicas, analysis performed at 20–70 ns.