The pharmacokinetic properties of were amenable to
The pharmacokinetic properties of 12 were amenable to oral dosing allowing in vivo comparison to 6. Compared to a maximum efficacious dose of 6 (60mg/kg), 12 demonstrated improved glycemic control during OGTT in high-fat fed/STZ treated (HF/STZ) and BDF/diet-induced obesity (DIO) mouse models of type II diabetes. Significant increases in GLP-1 and GIP secretion were measured in HF/STZ mice treated with 12, while partial agonist 6 did not show significant increases in GLP-1. This engagement of the incretin effect has not been demonstrated by other Heparin sodium of GPR40 agonists, further illustrating the disparate pharmacology of GPR40 full agonists from partial agonists. The positive cooperativity of 6 and 12 that was shown in vitro was extended to in vivo studies where coadministration provided efficacy greater than either compound alone.
To improve the potency, specificity, and pharmacokinetic profile of 12, spirocyclic constraints were introduced. These modifications culminated in the discovery of spirocycle 13, a ligand that demonstrated an improved selectivity profile on a panel of 101 GPCRs, ion channels, transporters, and enzymes at 10μM. Compound 13 showed increased potency (EC50=0.081μM, Emax=101%) and half-life (t½=4.2h) in rats compared to 12. In a HF/STZ mouse model of diabetes, 13 displayed greater reduction of glucose excursion after OGTT at 30mg/kg than 12 at 60mg/kg. Further efforts were focused on improving the drug-like properties and the pharmacokinetic profile of 12 by introducing polar functionalities. A 3.3unit reduction in ClogP was achieved with compound 14 (ClogP=6.0), while maintaining similar potency (EC50=0.16μM, Emax=94%) to 12. Both rat clearance and half-life were significantly improved in compound 14 (Cl=0.06L/h/kg; t½=8.1h). Similar reductions in glucose excursion were observed with a 3mg/kg dose of compound 14 as with 60mg/kg of 12 after OGTT in the BDF/DIO mouse model of diabetes.
Conclusions The clinically validated role of GPR40 in lowering HbA1c by 3 has prompted excitement around the GPR40 partial agonist class of antidiabetics. However, given the role of GPR40 in sensing long-chain free fatty acids in the body, it is not surprising that many of the small-molecule partial agonists to date have possessed properties that fall outside of the range of typical drug-like molecules (such as logP<5). Given that the Phase III clinical trial evaluating the safety and efficacy of 3 was halted due to hepatotoxicity and that three other clinical trials of GPR40 partial agonists were halted for undeclared reasons (6, 8, and ASP5034), the next generation of GPR40 partial agonists will need to overcome the toxicity challenges that hindered the progress of the initial small molecules in this class. However, the robust clinical effects should ensure interest in this class going forward.
Introduction 2-Lysophosphatidylcholines (1-acyl-glycero-3-phosphocholines, 2-LPCs, LPCs), which maintain the acyl chain in sn-1 position, are the most abundant lysophospholipid in nature (D'Arrigo and Servi, 2010). Lipidomic analysis has revealed a correlation of lower plasma concentrations of LPCs with impaired glucose tolerance and obesity (Zhao et al., 2010, Barber et al., 2012). LPC 16∶0 is the most abundant species in human plasma (146 ± 37 μM) followed by LPC 18∶0 (56.5 ± 14.9 μM) and LPC 18∶1 (28.4 ± 12.5 μM) (Heimerl et al., 2014). Although the presence of LPCs in plasma was observed at the beginning of the twentieth century (Kihara et al., 2015), the original observation from Metz's laboratory on dose-dependent lysophospholipid-induced insulin secretion was shown already in 1986 (Metz, 1986) while the involvement of G protein coupled receptor for the effect of LPC on insulin secretion was identified as GPR119 in 2005 (Soga et al., 2005). GPR119 is preferentially expressed on β-cells of the islets of Langerhans but its expression has also been demonstrated in intestinal L- and K-cells, where its activation was associated with secretion of glucagon-like peptide 1 and glucose-dependent insulinotropic peptide (Overton et al., 2006, Sakamoto et al., 2006, Ahlkvist et al., 2013). GPR119 has been shown to bind a variety of lipid-derived ligands, as well as a range of small synthetic molecules. Recent literature data indicate that lysophospholipids also have the ability to interact with other pancreatic receptors regulating carbohydrate metabolism. GPR55 activated by lysophosphatidylinositols may be another attractive target in type 2 diabetes mellitus (T2DM) (Liu et al., 2016). Treatment of diabetic rats with lysophoshatidylinositol has been found to counteract the symptoms of diabetes such as high blood glucose, lower body weight, increase amplitude of slow wave in stomach smooth muscle, and to improve gastric emptying (Lin et al., 2014). Both GPR119 and GPR55 receptors are stimulated by endocannabinoids such as palmitoylethanolamide (PEA), oleoylethanolamide (OEA), arachidonoylethanolamide (anandamide, AEA), and 2-arachidonoylglycerol (2-AG) (Godlewski et al., 2009). We have recently proved that ligand specificity of GPR55 is much wider and our studies evidence that GPR55 is activated also by LPC (Drzazga et al., 2017). However, GPR40 (also known as the free fatty acid receptor 1 or FFAR1) is the best-studied of the cell-surface receptors on β-cells. GPR40 is the most potently activated by endogenous free fatty acids (FFAs) with medium and long (C12-C22) aliphatic chains, resulting in amplification of insulin secretion only in the presence of elevated glucose levels (Itoh et al., 2003, Itoh and Hinuma, 2005, Briscoe et al., 2003). The glucose dependency of insulin secretion makes this receptor an excellent target for developing efficacious therapies with a desired safety profile for use in the treatment of T2DM.