Discovery of a novel oxime ether scaffold as potent and orally bioavailable free fatty acid receptor 1 agonists

Abstract

The free fatty acid receptor 1 (FFA1) plays a key role in amplifying glucose-stimulated insulin secretion in pancreatic β-cells. Most of the reported FFA1 agonists contain a biphenyl scaffold, which is associated with toxicity as verified by Daiichi Sankyo. Herein, we describe the systematic exploration of a non-biphenyl scaffold to improve the druggability of GW9508 (β-oxidation, F_sp³ = 0.13, tPSA = 58.5 Ų) directed by F_sp³ and tPSA values. All these optimizations ultimately led to the identification of compound 21, an unconventional agonist (EC₅₀ = 72.5 nM) bearing a methyl oxime ether scaffold. Moreover, compound 21 revealed improved drug-like properties (F_sp³ = 0.23, tPSA = 86.6 Ų) when compared to GW9508 (F_sp³ = 0.13, tPSA = 58.5 Ų) and an even higher binding efficiency index (BEI = 15.3) than TAK-875 (BEI = 14.3). Further pharmacological studies suggested that compound 21 has a considerable hypoglycemic effect in both normal and type 2 diabetic mice with a low risk of hypoglycemia. In addition, the docking study promoted our understanding of the ligand-binding pocket. This information might help towards the design of more promising new molecular entities.

---

1. Introduction

Type 2 diabetes mellitus (T2DM) is a worldwide epidemic characterized by impaired glucose homeostasis due to tissue resistance and insulin deficiency.^1,2 Despite an inspiring array of pharmacotherapies for T2DM, most of the oral antidiabetic agents are associated with undesirable side effects such as a high risk of hypoglycemia, body weight gain and gastric symptoms.^3–6 Hence, a novel drug with increased durability and safety in achieving preferable glycemic control is still an urgent need for T2DM.^7,8 The free fatty acid receptor 1 (FFA1, also known as GPR40), a new anti-diabetic target, has attracted considerable attention over the last decade and plays a key role in amplifying glucose-stimulated insulin secretion on pancreatic β-cells without the risk of hypoglycemia.^9–12 Therefore, the glucose concentration-dependent mechanism of FFA1 provides enormous potential for improving insulin levels with a decreased risk of hypoglycemia.

Recently, a variety of synthetic FFA1 agonists that contain acidic moieties have been reported in the literature (Fig. 1),^13–21 of which compounds TAK-875 and AMG-837 are both in clinical trials. However, in December 2013, Takeda decided to terminate the phase III studies of TAK-875 due to concerns about liver toxicity.^22,23 Subsequently, the researchers at Daiichi Sankyo have shown that the common biphenyl scaffold (red mark in Fig. 1) of the FFA1 agonists was crucial for toxicity.²⁴ Therefore, it is extremely urgent to hunt for a new non-biphenyl scaffold. We previously reported a series of phenoxyacetic acid derivatives, which successfully removed the biphenyl moiety by introducing a phenoxyacetamide linker.²⁵ Herein, we chose GW9508 (Fig. 1) as the non-biphenyl lead compound to explore novel FFA1 agonists because it has outstanding efficacy and a simple structure to avoid undue increase in molecular size and lipophilicity during the modification process. On the other hand, GW9508 was susceptible to β-oxidation at the phenylpropanoic acid moiety and is a rather flat molecule with a fraction of saturated carbons (F_sp³) of 0.13. The F_sp³ (the number of sp³ hybridized carbons/total carbon count) is a new index representing drug-likeness and is extensively used in optimization processes.^26–29 Lovering et al. pointed out that an increase in the F_sp³ value can result in improved clinical success.^30,31 Moreover, FFA1 is expressed in the brain and its function is unclear.¹⁰ However, a structurally related agonist with low polar surface area revealed significant central nervous system (CNS) exposure.³² Thus, we considered that improvement of the F_sp³ and tPSA values would be a rational approach to hunting for a new non-biphenyl scaffold with improved drug-like properties (Fig. 2). After comprehensive exploration of SAR information and utilization of molecular modeling, the potent lead compound 21 bearing an oxime ether scaffold and its binding mode were identified. In further pharmacological studies, compound 21 with improved F_sp³ (0.23) and tPSA (86.6 Ų) values compared to GW9508, revealed a low risk of hypoglycemia and outstanding hypoglycemic effects in both normal and type 2 diabetic mice.

Fig. 1 Selected examples of synthetic GPR40 agonists.

Fig. 2 Design strategy starting from GW9508 as an initial lead.

2. Results and discussion

2.1. Chemistry

The synthetic routes used to prepare target compounds 1–4 and 6–8 are summarized in Scheme 1. The intermediates 3a–b were prepared by chlorination of 2a–b, which were derived from the Williamson ether synthesis of 3-hydroxybenzyl alcohol and benzyl bromide or (2-bromoethyl)benzene in the presence of K₂CO₃. The commercial starting material 2-bromoacetophenone 4a was treated with 3-hydroxybenzyl alcohol to afford the desired product 5a, which was further converted to intermediate 6a via chlorination with thionyl chloride catalyzed by DMF. The intermediate 7a was synthesized upon mixing 4a and methoxyamine hydrochloride in DMSO. Condensation of 7a with 3-hydroxybenzyl alcohol and then chlorination provided intermediate 9a. Similar conditions to 9a provided intermediates 13a–c from intermediates 11a–c, formed from commercially available 10a–c upon treatment with methoxyamine hydrochloride. The dihydrobenzofuran intermediate 14a was synthesized via literature procedures.¹⁵ Condensation of the obtained chlorinated intermediates 3a–b, 6a, 9a or 13a–c with 14a using the Williamson ether synthesis followed by basic hydrolysis, afforded the desired carboxylic acids 1–4 and 6–8.

Scheme 1 Syntheses of the target compounds 1–4 and 6–8. Reagents and conditions: (a) BnBr or (2-bromoethyl)benzene, K₂CO₃, acetone, 45 °C, 12 h; (b) SOCl₂, CH₂Cl₂, DMF, 40 °C, 4 h; (c) methoxyamine hydrochloride, DMSO, rt, 8 h; (d) K₂CO₃, acetone, KI, 60 °C, 8 h and (e) LiOH·H₂O, THF/MeOH/H₂O, rt, 4 h.

The syntheses of target compounds 5 and 9–14 are depicted in Scheme 2. The commercial starting material 2-bromoacetophenone 4a was treated with 4-hydroxybenzyl alcohol to afford the common intermediate 15a, which was further converted to the chlorinated intermediate 16a. Intermediate 17a was achieved from 15a via condensation with hydroxylammonium chloride. Alkylation of intermediate 17a with the corresponding alkylating reagent obtained intermediates 18a–f and was followed by chlorination using thionyl chloride to yield the desired intermediates 19a–f. Subsequently, connecting the obtained chlorinated intermediates 16a or 19a–f with 14a in the presence of K₂CO₃, followed by basic hydrolysis, afforded the desired carboxylic acids 5 and 9–14.

Scheme 2 Syntheses of target compounds 5 and 9–14. Reagents and conditions: (a) 4-hydroxybenzyl alcohol, K₂CO₃, acetone, 45 °C, 12 h; (b) SOCl₂, CH₂Cl₂, DMF, 40 °C, 4 h; (c) hydroxylammonium chloride, DMSO, rt, 8 h; (d) RBr, K₂CO₃, acetone, 45 °C, 12 h; (e) K₂CO₃, acetone, KI, 60 °C, 8 h and (f) LiOH·H₂O, THF/MeOH/H₂O, rt, 4 h.

The syntheses of the target compounds 15–30 are detailed in Scheme 3. The preparation of intermediates 21a–c began with the conversion of 7a to 20a–c via condensation with substituted 4-hydroxybenzyl alcohol. Similar conditions to 21a–c provided intermediates 25a–m from intermediates 23a–m, formed from commercially available 22a–m upon treatment with methoxyamine hydrochloride. Condensation of the obtained chlorinated intermediates 21a–c or 25a–m with 14a using the Williamson ether synthesis followed by basic hydrolysis, afforded the desired carboxylic acids 15–30.

Scheme 3 Syntheses of target compounds 15–30. Reagents and conditions: (a) K₂CO₃, acetone, 45 °C, 12 h; (b) SOCl₂, CH₂Cl₂, DMF, 40 °C, 4 h; (c) methoxyamine hydrochloride, DMSO, rt, 8 h; (d) K₂CO₃, acetone, KI, 60 °C, 8 h and (e) LiOH·H₂O, THF/MeOH/H₂O, rt, 4 h.

2.2. FFA1 agonistic activity and SAR study

The synthetic compounds were investigated in Chinese hamster ovary (CHO) cells expressing human FFA1 using a fluorometric imaging plate reader (FLIPR) assay. As shown in Table 1, compound 1 was initially designed to increase the F_sp³ value (F_sp³ = 0.21) and appeared to diminish the FFA1 agonistic activity when compared with GW9508 (F_sp³ = 0.13). An approximately 2-fold increase in potency over compound 1 was obtained by attaching an additional methylene in compound 1 to achieve compound 2, which further improved the F_sp³ value (F_sp³ = 0.24). We speculated that the improvement of activity was associated with the larger rotational freedom of conformation in compound 2 to adopt the required conformation. With the dual purpose of improving the potency of compounds as well as defining the active conformation of the ligand to understand the nature of the binding pocket, the conformationally constrained analog 3 was designed and synthesized. Compound 3, with an increased tPSA value (82.0 Ų), revealed a significant improvement in potency when compared to the parent compound 2, though a dramatically decreased F_sp³ value was observed. Concerns about the potential metabolic instability associated with the ketone moiety led us to introduce an oxime ether scaffold based on Connexios' patent,³³ whereas the afforded compound 4 (tPSA = 86.6 Ų) appeared to diminish the agonistic activity. Interestingly, re-positioning the site of the substituent from the 3-position to 4-position obtained compounds 5 and 6, which exhibited a significant improvement in potency in comparison with the parent compounds 3 and 4, respectively. Unlike compound 4, the oxime ether analog 6 revealed better activity than the ketone analog 5, which suggested that the oxime ether occupied an additional binding site or induced a more appropriate conformation. Moreover, the bioisosteres, thiophene (7) and the non-aromatic substituent (8) were introduced to explore the importance of the phenyl ring in compound 6. The obtained compounds 7 and 8 displayed a drastic loss of activity, indicating that the phenyl ring of compound 6 was crucial for FFA1 agonistic activity.

Table 1 In vitro agonistic activities and selected parameters of the target compounds

Compd	R	Scaffold	Act^a% (300 nM)	Act^b% (100 nM)	Fsp^3^c	tPSA^d (Å^2)
a The mean value of agonist activity at a screening concentration of 300 nM was obtained from three independent experiments.b The mean value of agonist activity at a screening concentration of 100 nM was obtained from three independent experiments.c Fsp^3 = number of sp^3 hybridized carbons/total carbon count.d The topological polar surface area was calculated using ACD/I-Lab: https://ilab.acdlabs.com/iLab2/index.php.
TAK-875			76.01	65.32	0.35	99.1
GW9508			73.02	57.16	0.13	58.5
1		A	18.31	3.96	0.21	65.0
2		A	35.57	14.74	0.24	65.0
3		A	43.85	20.36	0.20	82.1
4		A	32.26	13.64	0.23	86.6
5		B	54.83	38.24	0.20	82.1
6		B	67.45	53.76	0.23	86.6
7		B	36.75	15.62	0.25	114.8
8		B	20.85	4.18	0.33	86.6

---

Based on the results described above, we therefore selected the oxime ether scaffold of compound 6 as our starting point for further modification. Our optimized efforts were directed to probe the best substituent for oxime ether (Table 2). For hydrophobic substituents, as the size of the substituent increased, the potency decreased (compound 6 > 9 > 10 > 12 > 11 > 13), implying that the steric effects of the substituent in this area might interrupt the desired interactions with the receptor. However, a bulky hydrophilic substituent (compound 14) was well tolerated despite the increased steric effects.

Table 2 Agonistic activities and selected index of the designed compounds

Compd	R	Act^a% (300 nM)	Act^b% (100 nM)	Fsp^3^c	tPSA^d (Å^2)
a The mean value of agonist activity at a screening concentration of 300 nM was obtained from three independent experiments.b The mean value of agonist activity at a screening concentration of 100 nM was obtained from three independent experiments.c Fsp^3 = number of sp^3 hybridized carbons/total carbon count.d The topological polar surface area was calculated using ACD/I-Lab: https://ilab.acdlabs.com/iLab2/index.php.
TAK-875		76.01	65.32	0.35	99.1
GW9508		73.02	57.16	0.13	58.5
6	Me	67.45	53.76	0.23	86.6
9	Et	45.63	26.87	0.26	86.6
10		36.57	17.38	0.28	86.6
11		25.67	5.35	0.31	86.6
12		29.73	7.79	0.31	86.6
13		20.75	4.03	0.19	86.6
14		40.29	21.36	0.22	129.6

---

Having identified a favorable methyl oxime ether scaffold with improved F_sp³ (0.23) and tPSA (86.6 Ų) values, we then shifted our focus on evaluating the various substituents on the terminal and middle benzene rings. We had concerns about the relatively high molecular weights that exist in most of the reported agonists such as TAK-875 (524.63) and AM-1638 (514.64). Therefore, besides the agonistic activity, the binding efficiency index (BEI, pEC₅₀/M_w),^34,35 a metric incorporating the affinity and molecular weight, was taken advantage of in the drug-like evaluation of the ligands (Table 3). From the results of the fluorine screening in the middle benzene ring (compound 15 and 16), the 3′-position seemed to be better than the 2′-position to retain FFA1 agonistic activity. However, a bulkier methoxy group at the 3′-position (17) showed a markedly reduced agonistic activity, indicating that the introduction of a substituent with steric effects in the 3′-position was unfavorable. Then, an initial modification of the terminal benzene ring with a methyl group (18, 19 and 20) demonstrated that substitution was tolerated in the para-position and was preferred in the meta-position. Gratifyingly, compound 21 (2-F) exhibited a marked improvement in potency when compared with compound 18 (2-Me). These activity differences could be attributed to the steric hindrance of the methyl group, which moves the tail moiety of the ligand into a different and less favorable conformation. Further modification at the meta-position suggested that the van der Waals radius of the substituent revealed a good correlation with the agonistic activity: the agonistic activity of compound 22 (3-F, 1.47 Å) > 24 (3-Cl, 1.75 Å) > 19 (3-Me, 1.80 Å) > 26 (3-CF₃, 2.20 Å).³⁶ To complete the SAR study, a variety of substitutions in the para-position of the terminal phenyl ring were evaluated. For the para-substituted compounds, the agonistic activity of 6 (4-H) > 23 (4-F) > 25 (4-Cl) > 20 (4-Me) > 27 (4-CF₃) demonstrated that steric effects in the para-position might influence the FFA1 agonistic activity. In contrast, a methoxy group in the para-position (compound 28) was shown to improve the affinity for FFA1, implying that the methoxy group has an additional interaction with the receptor. Interestingly, although fluorine in the ortho-position and para-position exhibited excellent activity, the disubstituted compound 29 revealed a markedly lower agonistic activity than its corresponding mono-substituted analogs 21 and 23. Consistent with the abovementioned finding, the 3,4-di-chlorinated compound 30 was also inferior to its mono-substituted analogs, suggesting that the polysubstituted groups may introduce an unfavorable electrical or steric interactions with FFA1. Among all of the tested compounds, compound 21, a most potent agonist with considerable BEI value (15.3) in this series, had a significant advantage when compared to GW9508 in terms of the F_sp³ and tPSA values despite a lower BEI value.

Table 3 In vitro agonistic activities and index of the designed compounds^a

Compd	R1	R2	Act^b% (100 nM)	EC50^c (nM)	BEI^d	Fsp^3^e	tPSA^f (Å^2)
a ND = not determined.b The mean value of agonist activity at a screening concentration of 100 nM was obtained from three independent experiments.c The EC50 values for FFA1 activity represent the mean of three determinations.d BEI = pEC50/Mw in kDa.e Fsp^3 = number of sp^3 hybridized carbons/total carbon count.f The topological polar surface area was calculated using ACD/I-Lab: https://ilab.acdlabs.com/iLab2/index.php.
TAK-875			65.32	29.6	14.3	0.35	99.1
GW9508			57.16	63.2	20.6	0.13	58.5
6	H	H	53.76	76.9	15.9	0.23	86.6
15	H	2′-F	31.25	ND		0.23	86.6
16	H	3′-F	46.73	ND		0.23	86.6
17	H	3′-MeO	21.65	ND		0.26	95.8
18	2-Me	H	10.56	ND		0.26	86.6
19	3-Me	H	40.18	ND		0.26	86.6
20	4-Me	H	27.56	ND		0.26	86.6
21	2-F	H	54.83	72.5	15.3	0.23	86.6
22	3-F	H	51.45	89.7	15.1	0.23	86.6
23	4-F	H	45.76	ND		0.23	86.6
24	3-Cl	H	43.57	ND		0.23	86.6
25	4-Cl	H	30.78	ND		0.23	86.6
26	3-CF3	H	32.38	ND		0.26	86.6
27	4-CF3	H	20.78	ND		0.26	86.6
28	4-MeO	H	51.67	88.9	14.8	0.26	95.8
29	2,4-Di F	H	25.78	ND		0.23	86.6
30	3,4-Di Cl	H	20.63	ND		0.23	86.6

---

2.3. Molecular modeling study

To comprehensively understand the interaction mode and SAR of our oxime ether series, we performed a molecular docking study of compound 6 using the X-ray structure of FFA1 (PDB accession code: 4PHU).³⁷ As shown in Fig. 3, compound 6 was nearly a perfect dock to the same binding pocket of TAK-875 and the docking study aligns with the SAR determined above. The residues Tyr91, Arg183 and Arg2258 were highly coordinated with head acid moiety forming three stable hydrogen bonds. Moreover, Trp174 was oriented nearly orthogonal to the plane of the dihydroisobenzofuran ring of compound 6 where it forms an additional edge-on interaction. Notably, the terminal benzene ring of compound 6 has a hydrophobic interaction with the Leu135 residue, which had no interactions with TAK-875. It was suggested that this hydrophobic interaction was crucial for the robust agonistic activity of compound 6 rather than the other non-phenyl compounds (such as 7 and 8). Moreover, the methyl group in the oxime ether moiety extends into hydrophobic portion of the lipid bilayer, which further stabilized the binding conformation of the ligand. Therefore, it showed better agonistic activity than the other types of substituent.

Fig. 3 Overlay of compound 6 and TAK-875 bound to FFA1. The key residues are labeled in red and the hydrogen bonding interactions are represented by yellow dashed lines.

2.4. Effect of compound 21 on the OGTT

Based on the favorable in vitro activity, the most potent compound 21 (10, 20 and 40 mg kg⁻¹) was selected for further pharmacological evaluation using an oral glucose tolerance test (OGTT) in normal ICR mice. Single oral doses of compound 21 significantly improved the glucose tolerance in a dose-proportional manner from 10 to 40 mg kg⁻¹ (Fig. 4A). Furthermore, compound 21 revealed a robust hypoglycemic effect at a dose of 40 mg kg⁻¹, a potency that was similar with that found with TAK-875 (20 mg kg⁻¹).

Fig. 4 (A) Effect of compound 21 on the OGTT in normal ICR mice and glucose load (3 g kg⁻¹) at 0 min. (B) The OGTT of compound 21 in fasting type 2 diabetic C57BL/6 mice and glucose load (2 g kg⁻¹) at 0 min. The values are reported as the mean ± SEM (n = 6). *P ≤ 0.05 and **P ≤ 0.01 compared to vehicle mice using the Student's t test. ^#P ≤ 0.05 compared to vehicle diabetic mice using the Student's t test.

To assess the pharmacological effects of compound 21 in the diabetic state, we performed an OGTT on STZ-induced type 2 diabetic C57BL/6 mice, a disease model with impaired glucose tolerance.^38,39 As shown in Fig. 4B, the hyperglycemia state was significantly controlled in compound 21 (40 mg kg⁻¹) treated mice. These results suggested that compound 21, with improved drug-like properties (F_sp³ = 0.23 and tPSA = 86.6 Ų), has great potential for reducing the plasma glucose excursion in both normal and type 2 diabetic mice.

2.5. Effects of compound 21 on the risk of hypoglycemia

Based on the positive result of the hypoglycemic effect, the risk of hypoglycemia in fasted normal ICR mice was subsequently evaluated using a high oral dose of compound 21 in contrast with the positive control glibenclamide (a sulfonylurea insulin secretagogue). As shown in Fig. 5, glibenclamide (15 mg kg⁻¹) treated mice displayed lowered blood glucose levels far below that observed for normal fasted ICR mice. On the contrary, compound 21 treated mice, even at an oral dose of 100 mg kg⁻¹, displayed only mildly decreased fasting glucose levels and the reduction was far less than that found with glibenclamide. Therefore, the research results show that compound 21 possesses a relatively low risk of hypoglycemia, a common adverse effect to a sulfonylurea insulin secretagogue.

Fig. 5 Hypoglycemic effects of compound 21 in fasted normal ICR mice at a high dose of 100 mg kg⁻¹. The values are reported as the mean ± SEM (n = 6). *P ≤ 0.05 and **P ≤ 0.01 compared to vehicle mice using Student's t test.

3. Conclusions

In conclusion, starting from the previously reported non-biphenyl FFA1 agonist GW9508, we systematically explored a series of non-biphenyl scaffolds directed by the F_sp³ and tPSA values to improve the drug-like properties of GW9508 (F_sp³ = 0.13, tPSA = 58.5 Ų). Subsequently, a systematic exploration of the SAR information in the optimal methyl oxime ether scaffold lead to the discovery of lead compound 21, a potent FFA1 agonist with robustly agonistic activity (EC₅₀ = 72.5 nM), improved drug-like properties (F_sp³ = 0.23, tPSA = 86.6 Ų), higher BEI value (15.3) than TAK-875 (BEI = 14.3) and outstanding antihyperglycemic effects in both normal and type 2 diabetic mice without the risk of hypoglycemia. Moreover, our docking study expanded the comprehensive understanding of the binding site and may help to design more active FFA1 agonists with non-biphenyl scaffolds.

4. Experimental section

4.1. Chemistry

Chromatographic purification was performed on a silica gel (200–300 mesh) and monitored using thin layer chromatography carried out on GF/UV 254 plates using UV light (254 and 365 nm). Melting points were determined on a RY-1 melting-point apparatus and were not corrected. NMR (nuclear magnetic resonance) spectra were obtained on a Bruker ACF-300Q instrument (300 MHz for ¹H NMR and 75 MHz for ¹³C NMR spectra) with tetramethylsilane as an internal standard. Chemical shifts are reported in parts per million (ppm) and coupling constants (J values) given in hertz (Hz). Elemental analyses were carried out on the Heraeus CHN-O-Rapid analyzer and have errors within ±0.4% for CHN elements. The LC/MS spectra were obtained on a Waters liquid chromatography-mass spectrometer system (ESI†). All starting materials and reagents were obtained from commercial suppliers and used without further purification. TAK-875 and GW9508 were synthesized via previous reported procedures.^15,21

The physical characteristics, ¹H NMR, ¹³C NMR, MS and elemental analysis data for all intermediates and target compounds are reported in the ESI.†

4.2. Molecular modeling

The molecular docking study of compound 6 was performed using MOE (version 2008.10, the Chemical Computing Group, Montreal, Canada). The X-ray crystal structure of FFA1 (PDB accession code: 4PHU) was retrieved from the Protein Data Bank. Prior to docking, water molecules and ligands were deleted except TAK-875. Then, the X-ray crystal structure was prepared with Protonate 3D and a Gaussian contact surface was drawn around the binding site of TAK-875. Then, the active site was isolated and the backbone was removed. The ligand poses were filtered using Pharmacophore Query Editor. The structure of compound 6 was docked into the binding pocket using the Pharmacophore method and then ranked with the London dG scoring function. For energy minimization of the ligand in the active site, MOE Forcefield Refinement was used and ranked with London dG scoring function.

4.3. Biological methods

4.3.1. Ca²⁺ influx activity of CHO cells stably expressing human FFA1 (FLIPR assay). CHO cells stably expressing human FFA1 (accession no. NM_005303) were plated at a density of 15k cells per well and incubated for 12 h in 5% CO₂ at 37 °C. Subsequently, the culture medium was removed and washed with Hank's balanced salt solution (100 μL). Then, the cells were incubated in loading buffer (recording medium containing 2.5 μg mL⁻¹ fluorescent calcium indicator Fluo 4-AM, 0.1% fatty acid-free BSA and 2.5 mmol L⁻¹ probenecid) for 1 h at 37 °C. Various concentrations of test compounds or γ-linolenic acid (Sigma) were added to the cells and the intracellular Ca²⁺ flux signals after the addition were monitored by FLIPR Tetra system (Molecular Devices) for 90 s. The agonistic activities of the test compounds on human FFA1 were expressed as [(A − B)/(C − B)] × 100 (increase of the intracellular calcium concentration (A) in the test compounds-treated cells and (B) in vehicle-treated cells and (C) in 10 μM γ-linolenic acid-treated cells). The EC₅₀ values of selected compounds were obtained with Prism 5 software (GraphPad).

4.3.2. Animals and statistical analysis of the data. Male ICR mice (18–22 g) and male C57BL/6 mice (18–22 g) were obtained from the Comparative Medicine Centre of Yangzhou University (Jiangsu, China). Mice were acclimatized for 1 week before the experiments were performed and allowed ad libitum access to standard pellets and water unless otherwise stated. The feeding room was maintained on a constant 12 h light/dark cycle with controlled temperature (23 ± 1 °C) and humidity (55% ± 5%). The vehicle was orally administered 0.5% carboxymethyl cellulose aqueous solution for all the animal studies. All animal experiments were performed in compliance with the relevant laws and institutional guidelines, and our experiments have been approved by the Institutional Committee of China Pharmaceutical University.

Statistical analyses were obtained with GraphPad software (GraphPad InStat version 5.00, San Diego, CA, USA). Unpaired comparisons were analyzed using the two-tailed Student's t-test.

4.3.2.1. Effect of compound 21 on the OGTT explored in male mice. Ten-week-old male ICR mice were fasted for 12 h, weighed, bled via the tail vein and randomized into five groups (n = 6). A single dose of the vehicle TAK-875 (20 mg kg⁻¹) or compound 21 (10 mg kg⁻¹, 20 mg kg⁻¹, 40 mg kg⁻¹) was orally administered 30 min before the oral glucose load (3 g kg⁻¹). Blood samples were collected from the tail vein immediately before drug administration (−30 min), 0 min (just before the glucose load) and at 15, 30, 45, 60 and 120 min post-glucose load. The blood glucose levels were measured using blood glucose test strips (SanNuo ChangSha, China).
4.3.2.2. Hypoglycemic effects of compound 21 explored in type 2 diabetic mice. Male C57BL/6 mice after 7 days adaptation were fed with a high-fat diet (45% calories from fat, obtained from Mediscience Ltd., Yangzhou, China) ad libitum for a further 4 weeks to induce insulin resistance and then injected intraperitoneally (i.p.) with a low dose of STZ (10 mL kg⁻¹; 80 mg kg⁻¹). The C57BL/6 mice were fed with a high-fat diet for another 4 weeks and mice with a fasting blood glucose level ≥11.1 mmol L⁻¹ were used as the type 2 diabetic mice model.^38,39

The type 2 diabetic C57BL/6 mice were fasted for 12 h, weighed, bled via the tail vein and randomized into 3 groups (n = 6), another group of normal fasting C57BL/6 mice was added as a negative control. A single dose of the vehicle TAK-875 (20 mg kg⁻¹) or compound 21 (40 mg kg⁻¹) was orally administered 30 min before the oral glucose load (2 g kg⁻¹). Blood samples were collected from the tail vein immediately before drug administration (−30 min), 0 min (just before the glucose load) and at 15, 30, 45, 60 and 120 min post-glucose load. The blood glucose levels were measured using blood glucose test strips (SanNuo ChangSha, China).

4.3.2.3. Effects of compound 21 on the risk of hypoglycemia. 10 weeks old male normal ICR mice were fasted overnight and randomized into 3 groups (n = 6). Compound 21 (100 mg kg⁻¹), glibenclamide (15 mg kg⁻¹) or vehicle was orally administered and blood was collected from the tail vein immediately before administration (0 min) and at 30, 60, 90, 120 and 180 min post-administration and the blood glucose levels were measured as described above.

Conflict of interest

The authors have no conflicts of interest to declare.

Acknowledgements

This study was supported by grants from the National Natural Science Foundation of China (Grants 81172932 and 81273376), the Natural Science Foundation of Jiangsu Province (Grant BK2012356), and the Fundamental Research Funds for the Central Universities, China Pharmaceutical University (Grant JKZD2013001).

References

1. R. A. DeFronzo, Diabetes, 2009, 58, 773–795.
2. G. Danaei, M. M. Finucane, Y. Lu, G. M. Singh, M. J. Cowan, C. J. Paciorek, J. K. Lin, F. Farzadfar, Y.-H. Khang, G. A. Stevens, M. Rao, M. K. Ali, L. M. Riley, C. A. Robinson and M. Ezzati, Global Burden, Metab Risk Factors, Lancet, 2011, 378, 31–40.
3. O. J. Phung, J. M. Scholle, M. Talwar and C. I. Coleman, J. Am. Med. Assoc., 2010, 303, 1410–1418.
4. M. A. Avery, C. S. Mizuno, A. G. Chittiboyina, T. W. Kurtz and H. A. Pershadsingh, Curr. Med. Chem., 2008, 15, 61–74.
5. S. A. Stein, E. M. Lamos and S. N. Davis, Expert Opin. Drug Saf., 2013, 12, 153–175.
6. H. E. Lebovitz, Nat. Rev. Endocrinol., 2011, 7, 408–419.
7. A. A. Tahrani, C. J. Bailey, S. Del Prato and A. H. Barnett, Lancet, 2011, 378, 182–197.
8. X. Xu, G. Wang, T. Zhou, L. Chen, J. Chen and X. Shen, Expert Opin. Drug Discovery, 2014, 9, 1047–1058.
9. Y. Itoh, Y. Kawamata, M. Harada, M. Kobayashi, R. Fujii, S. Fukusumi, K. Ogi, M. Hosoya, Y. Tanaka, H. Uejima, H. Tanaka, M. Maruyama, R. Satoh, S. Okubo, H. Kizawa, H. Komatsu, F. Matsumura, Y. Noguchi, T. Shinobara, S. Hinuma, Y. Fujisawa and M. Fujino, Nature, 2003, 422, 173–176.
10. C. P. Briscoe, M. Tadayyon, J. L. Andrews, W. G. Benson, J. K. Chambers, M. M. Eilert, C. Ellis, N. A. Elshourbagy, A. S. Goetz, D. T. Minnick, P. R. Murdock, H. R. Sauls, U. Shabon, L. D. Spinage, J. C. Strum, P. G. Szekeres, K. B. Tan, J. M. Way, D. M. Ignar, S. Wilson and A. I. Muir, J. Biol. Chem., 2003, 278, 11303–11311.
11. L. A. Stoddart, N. J. Smith and G. Milligan, Pharmacol. Rev., 2008, 60, 405–417.
12. P. Wellendorph, L. D. Johansen and H. Bräuner-Osborne, Mol. Pharmacol., 2009, 76, 453–465.
13. X. Wang, T. Zhao, B. Yang, Z. Li, J. Cui, Y. Dai, Q. Qiu, H. Qiang, W. Huang and H. Qian, Bioorg. Med. Chem., 2015, 23, 132–140.
14. Z. Li, X. Wang, X. Xu, J. Yang, W. Xia, X. Zhou, W. Huang and H. Qian, Bioorg. Med. Chem., 2015, 23, 7158–7164.
15. N. Negoro, S. Sasaki, S. Mikami, M. Ito, M. Suzuki, Y. Tsujihata, R. Ito, A. Harada, K. Takeuchi, N. Suzuki, J. Miyazaki, T. Santou, T. Odani, N. Kanzaki, M. Funami, T. Tanaka, A. Kogame, S. Matsunaga, T. Yasuma and Y. Momose, ACS Med. Chem. Lett., 2010, 1, 290–294.
16. S. Mikami, S. Kitamura, N. Negoro, S. Sasaki, M. Suzuki, Y. Tsujihata, T. Miyazaki, R. Ito, N. Suzuki, J. Miyazaki, T. Santou, N. Kanzaki, M. Funami, T. Tanaka, T. Yasuma and Y. Momose, J. Med. Chem., 2012, 55, 3756–3776.
17. N. Negoro, S. Sasaki, M. Ito, S. Kitamura, Y. Tsujihata, R. Ito, M. Suzuki, K. Takeuchi, N. Suzuki, J. Miyazaki, T. Santou, T. Odani, N. Kanzaki, M. Funami, T. Tanaka, T. Yasuma and Y. Momose, J. Med. Chem., 2012, 55, 1538–1552.
18. J. B. Houze, L. Zhu, Y. Sun, M. Akerman, W. Qiu, A. J. Zhang, R. Sharma, M. Schmitt, Y. Wang, J. Liu, J. Liu, J. C. Medina, J. D. Reagan, J. Luo, G. Tonn, J. Zhang, J. Y.-L. Lu, M. Chen, E. Lopez, K. Nguyen, L. Yang, L. Tang, H. Tian, S. J. Shuttleworth and D. C. H. Lin, Bioorg. Med. Chem. Lett., 2012, 22, 1267–1270.
19. S. P. Brown, P. J. Dransfield, M. Vimolratana, X. Jiao, L. Zhu, V. Pattaropong, Y. Sun, J. Liu, J. Luo, J. Zhang, S. Wong, R. Zhuang, Q. Guo, F. Li, J. C. Medina, G. Swaminath, D. C. H. Lin and J. B. Houze, ACS Med. Chem. Lett., 2012, 3, 726–730.
20. Y. Wang, J. Liu, P. J. Dransfield, L. Zhu, Z. Wang, X. Du, X. Jiao, Y. Su, A.-R. Li, S. P. Brown, A. Kasparian, M. Vimolratana, M. Yu, V. Pattaropong, J. B. Houze, G. Swaminath, T. Thanhvien, N. Khanh, Q. Guo, J. Zhang, R. Zhuang, F. Li, L. Miao, M. D. Bartberger, T. L. Correll, D. Chow, S. Wong, J. Luo, D. C. H. Lin and J. C. Medina, ACS Med. Chem. Lett., 2013, 4, 551–555.
21. S. C. McKeown, D. F. Corbett, A. S. Goetz, T. R. Littleton, E. Bigham, C. P. Briscoe, A. J. Peat, S. P. Watson and D. M. B. Hickey, Bioorg. Med. Chem. Lett., 2007, 17, 1584–1589.
22. K. Kaku, T. Araki and R. Yoshinaka, Diabetes Care, 2013, 36, 245–250.
23. M. S. Hedrington and S. N. Davis, Expert Opin. Invest. Drugs, 2014, 23, 1703–1711.
24. R. Takano, M. Yoshida, M. Inoue, T. Honda, R. Nakashima, K. Matsumoto, T. Yano, T. Ogata, N. Watanabe and N. Toda, Bioorg. Med. Chem. Lett., 2014, 24, 2949–2953.
25. Z. Li, X. Wang, X. Xu, J. Yang, Q. Qiu, H. Qiang, W. Huang and H. Qian, Bioorg. Med. Chem., 2015, 23, 6666–6672.
26. K. Hirata, M. Kotoku, N. Seki, T. Maeba, K. Maeda, S. Hirashima, T. Sakai, S. Obika, A. Hori, Y. Hase, T. Yamaguchi, Y. Katsuda, T. Hata, N. Miyagawa, K. Arita, Y. Nomura, K. Asahina, Y. Aratsu, M. Kamada, T. Adachi, M. Noguchi, S. Doi, P. Crowe, E. Bradley, R. Steensma, H. Tao, M. Fenn, R. Babine, X. Li, S. Thacher, H. Hashimoto and M. Shiozaki, ACS Med. Chem. Lett., 2016, 7, 23–27.
27. M. J. Waring, J. Arrowsmith, A. R. Leach, P. D. Leeson, S. Mandrell, R. M. Owen, G. Pairaudeau, W. D. Pennie, S. D. Pickett, J. Wang, O. Wallace and A. Weir, Nat. Rev. Drug Discovery, 2015, 14, 475–486.
28. H. R. Hoveyda, G. L. Fraser, G. Dutheuil, M. El Bousmaqui, J. Korac, F. Lenoir, A. Lapin and S. Noel, ACS Med. Chem. Lett., 2015, 6, 736–740.
29. H. R. Hoveyda, G. L. Fraser, M.-O. Roy, G. Dutheuil, F. Batt, M. El Bousmaqui, J. Korac, F. Lenoir, A. Lapin, S. Noel and S. Blanc, J. Med. Chem., 2015, 58, 3060–3082.
30. F. Lovering, J. Bikker and C. Humblet, J. Med. Chem., 2009, 52, 6752–6756.
31. F. Lovering, MedChemComm, 2013, 4, 515–519.
32. J. Liu, Y. Wang, Z. Ma, M. Schmitt, L. Zhu, S. P. Brown, P. J. Dransfield, Y. Sun, R. Sharma, Q. Guo, R. Zhuang, J. Zhang, J. Luo, G. R. Tonn, S. Wong, G. Swaminath, J. C. Medina, D. C. H. Lin and J. B. Houze, ACS Med. Chem. Lett., 2014, 5, 517–521.
33. J. M. Ranganath Rao, N. Arumugam, M. M. Ansari, C. Gudla, S. Pachiyappan, M. Ramalingam, J. George, G. F. Arul, Y. K. Bommegowda, S. K. Angupillai, R. Kottamalai, P. Jidugu and D. Shivanageshwara Rao, Connexios Pat., WO2012011125, 2012.
34. C. Abad-Zapatero and J. T. Metz, Drug Discovery Today, 2005, 10, 464–469.
35. C. Abad-Zapatero, Expert Opin. Drug Discovery, 2007, 2, 469–488.
36. A. Bondi, J. Phys. Chem., 1964, 68, 441–451.
37. A. Srivastava, J. Yano, Y. Hirozane, G. Kefala, F. Gruswitz, G. Snell, W. Lane, A. Ivetac, K. Aertgeerts, J. Nguyen, A. Jennings and K. Okada, Nature, 2014, 513, 124–127.
38. M. S. Winzell and B. Ahrén, Diabetes, 2004, 53, 215–219.
39. J. A. Koehler, L. L. Baggio, B. J. Lamont, S. Ali and D. J. Drucker, Diabetes, 2009, 58, 2148–2161.

---

Footnote

† Electronic supplementary information (ESI) available: The physical characteristics, ¹H NMR, ¹³C NMR, MS and elemental analysis data for all intermediates and target molecules. See DOI: 10.1039/c6ra07356e

---