GPR is present primarily in the pancreas and the

GPR119 is present primarily in the pancreas and the intestine. Activation of GPR119 increases insulin, GLP-1, GIP and PYY secretion. GPR119 agonists stimulate insulin release in a glucose-dependent manner. The glucose dependent insulin secretion (GIDS) mechanism makes GPR119 an attractive target for the treatment of type 2 diabetes with low risk for hypoglycemia. Consequently, GPR119 agonists have been studied extensively and several compounds have entered clinical trials for the treatment of diabetes. However, despite the promising effects in preclinical species (especially in rodents) and enormous efforts by both academia and many pharmaceutical companies, it is still a challenge to demonstrate robust glycemic control efficacy in humans. So far, no vasopressin receptor antagonist has been reported to progress beyond phase II clinic trials. To increase the probability of success in glycemic control efficacy, we were interested in developing a fixed-dose combination (FDC) of a GPR119 agonist with a DPP4 inhibitor such as sitagliptin. It has been reported that in rat models, such combinations provide greater glucose lowering effects than either the GPR119 agonist or the DPP4 inhibitor alone., Our previous preclinical candidate 1 () which had excellent potency, selectivity and efficacy, was unsuitable for a fixed dose combination with sitagliptin (QD) due to, (a) a very long half-life in humans and, (b) poor solubility in, for example, fasted-state simulated intestinal fluid (FaSSIF) which would result in a high likelihood of a need for the use of enabled formulations in a FDC setting with sitagliptin. Structure-activity relationship (SAR) and structure-property relationship (SPR) efforts that were focused on reducing the predicted human t and improving FaSSIF solubility, while maintaining good potency at activation of GPR119 led to the discovery of benzyloxy compound (). Compound possessed excellent GPR119 potency, selectivity, and significantly improved physicochemical properties compared to its precursor . The projected human half-life for compound was ∼15h which was in the target range for a QD dosing. Its projected human dose (QD), however, was ∼300mg which was suboptimal as it would need to be combined with a 200mg once daily dose of sitagliptin. Herein we report our continued efforts in this area to improve the GPR119 potency and to further improve the overall pharmacokinetic profiles to reduce the projected human dose within this series of GPR119 agonists. Compound is a benzyloxy analogue while compound is a phenoxy derivative. As we reported before, changing from a phenoxy analogue to the corresponding benzyloxy group resulted in a favorable drop in lipophilicity and decreased unbound volume of distribution (Vd). For example in rat, compound had Vd 1725, while compound had Vd=54L/kg which helped reduce its rat t (1.0h vs 14.7h). However, high intrinsic clearance Cl was observed for many benzyloxy analogues due to extensive benzylic oxidative metabolism. Half-life can be shortened by increasing Cl and/or by decreasing Vd of a compound. However, increasing Cl will also lead to an increased QD dose projection and potentially higher levels of metabolites. Therefore, although the benzyloxy series was promising in reducing t and improving FaSSIF solubility, SAR studies on the original phenoxy lead were revisited in an effort to keep the Cl and projected human dose in the targeted range ( <200mg projected human QD dose). As we had already demonstrated that, the cyclopropyl piperidine scaffold provides several advantages such as improved GPR119 potency and selectivity, we elected to keep this core intact. Extensive SAR studies were first performed on the left hand side of the molecule by replacing the pyridinyl sulfone group of with different substituted aromatic rings. Among the many structurally diverse derivatives synthesized, we were pleased to find that phenyl acetamides provided some promising new directions. The representative synthesis of the phenyl acetamides is illustrated in . (1,2) piperidinyl cyclopropyl alcohol was converted to phenylacetic acid via a Mitsunobu reaction with methyl 2-(4-hydroxyphenyl)acetate (or the corresponding F substituted phenol) followed by methyl ester hydrolysis. The detailed syntheses of intermediate with different piperidine capping groups have been described in Ref. . Finally, amide bond formation with EDC, HOBt or HATU afforded a variety of phenyl acetamide final products.