Synthesis of non-steroidal cholic acid analogues as potent and selective TGR5 allosteric agonists with antidiabetic effects.

In this study, we designed and synthesized 12 novel non-steroidal cholic acid (CA) analogues as potential Takeda G protein coupled receptor 5 (TGR5) allosteric agonists based on scaffold hopping strategy. Biological results showed that all final compounds were able to activate the receptor both in human TGR5 (hTGR5) and mouse TGR5 (mTGR5) by Glosensor cAMP accumulation assay. Remarkably, lactone A1 and lactam A10 displayed 48- and 24-fold higher potencies than that of CA in hTGR5 cAMP accumulation, respectively. Furthermore, target compounds exhibited higher selectivity for cAMP formation compared to β-arrestin2 recruitment. Interestingly, target compounds A1-10 positively modulated the functional activity and potency of lithocholic acid (LCA) in TGR5, indicating that A1-10 are TGR5 allosteric agonists. Molecular docking study displayed that both lactone in A5 and lactam in A10 formed H-bonds with the key amino acid residue Thr131, which are critical for TGR5 allosteric effect. Unexpectedly, none of target compounds A1-10 could activate farnesoid X receptor (FXR), indicating that final compounds are potent and selective TGR5 allosteric agonists. More importantly, both A1 and A10 exhibited significant glucose-lowering effects in the oral glucose tolerance test (OGTT) in mice. Taken together, we discovered two potent and selective TGR5 allosteric agonists A1 and A10, which could be employed as lead compounds to further study TGR5 allostery with potential antidiabetic effects.