A novel antidiabetic drug, fasiglifam/TAK-875, acts as an ago-allosteric modulator of FFAR1

Abstract

Selective free fatty acid receptor 1 (FFAR1)/GPR40 agonist fasiglifam (TAK-875), an antidiabetic drug under phase 3 development, potentiates insulin secretion in a glucose-dependent manner by activating FFAR1 expressed in pancreatic β cells. Although fasiglifam significantly improved glycemic control in type 2 diabetes patients with a minimum risk of hypoglycemia in a phase 2 study, the precise mechanisms of its potent pharmacological effects are not fully understood. Here we demonstrate that fasiglifam acts as an ago-allosteric modulator with a partial agonistic activity for FFAR1. In both Ca(2+) influx and insulin secretion assays using cell lines and mouse islets, fasiglifam showed positive cooperativity with the FFAR1 ligand γ-linolenic acid (γ-LA). Augmentation of glucose-induced insulin secretion by fasiglifam, γ-LA, or their combination was completely abolished in pancreatic islets of FFAR1-knockout mice. In diabetic rats, the insulinotropic effect of fasiglifam was suppressed by pharmacological reduction of plasma free fatty acid (FFA) levels using a lipolysis inhibitor, suggesting that fasiglifam potentiates insulin release in conjunction with plasma FFAs in vivo. Point mutations of FFAR1 differentially affected Ca(2+) influx activities of fasiglifam and γ-LA, further indicating that these agonists may bind to distinct binding sites. Our results strongly suggest that fasiglifam is an ago-allosteric modulator of FFAR1 that exerts its effects by acting cooperatively with endogenous plasma FFAs in human patients as well as diabetic animals. These findings contribute to our understanding of fasiglifam as an attractive antidiabetic drug with a novel mechanism of action.

Figure 1. Partial agonist activity of fasiglifam is affected by FFAR1/GPR40 expression levels.

(A) The chemical structure of fasiglifam. (B and C) FFAR1 agonist activities of fasiglifam and free fatty acids (FFAs) in the intracellular Ca²⁺ mobilization assay using CHO cell lines expressing hFFAR1 (clone #104) (B) or mFFAR1 (C). Data are representative of three experiments. (D) hFFAR1 mRNA levels of hFFAR1-expressing CHO clones were evaluated by qRT-PCR. (E - H) Relative Ca²⁺ influx activities of γ-LA and fasiglifam in CHO clones #104 (E), #19 (F), #2 (G), and #4 (H) with various hFFAR1 expression levels. Error bars indicate s.e.m. (n = 3).

Figure 2. Fasiglifam exhibits positive cooperativity with γ-LA in intracellular Ca²⁺ influx and insulin secretion.

(A and B) Allosteric modulation of γ-linolenic acid (γ-LA) activity by fasiglifam (A) and fasiglifam activity by γ-LA (B) in the Ca²⁺ mobilization assay using hFFAR1/GPR40-expressing CHO cells (clone #2). (C and D) Positive cooperative effect of fasiglifam (Fas) on γ-LA activity (C) and of γ-LA on fasiglifam activity (D) in insulin secretion in mouse pancreatic β cell line, MIN6 cells. (E and F) Fasiglifam-induced insulin secretion in the absence and presence of γ-LA in pancreatic islets of wild-type (E) and FFAR1-knockout mice (F). All insulin secretion assays (C-F) were conducted in the presence of 16 mM glucose (G). All data are representative of at least two replicates. Error bars indicate s.e.m. (n = 3). *P<0.025 by one-tailed Shirley-Williams test, ^# P<0.05, ^## P<0.01 versus DMSO alone by Student’s t-test.

Figure 3. Insulinotropic effects of fasiglifam are attenuated by pharmacological reduction of plasma FFA levels in vivo.

(A) Effects of the lipolysis inhibitors acipimox (30 mg/kg) and fasiglifam (10 mg/kg) on plasma free fatty acids (FFAs) during the oral glucose tolerance test (OGTT) in N-STZ-1.5 rats. (B) Area under the curve (AUC) of plasma FFA during 0–120 min. Fas, fasiglifam. (C) Plasma glucose levels after coadministration of acipimox (30 mg/kg) and fasiglifam (10 mg/kg). (D) AUC of plasma glucose levels during 0–120 min. *P<0.05, **P<0.01 versus vehicle by Student’s t-test, ^$$ P<0.01 versus vehicle by Aspin–Welch test. (E) Plasma insulin concentrations after coadministration of acipimox and fasiglifam during OGTT. (F) Insulinotropic effects of fasiglifam (Fas) just before glucose load (time 0) shown in (E) in the absence and presence of acipimox. **P<0.01 versus vehicle, ^$ P<0.05 versus acipimox alone by Student’s t-test, followed by Bonferroni’s correction for four time point comparisons. Data represent mean ± s.e.m. (n = 6).

Figure 4. Fasiglifam does not exacerbate FFA-induced apoptotic signaling in MIN6 cells.

Caspase 3/7 activity in the mouse pancreatic β cell line MIN6 after 72-h exposure to 0.25–1 mM palmitic acid (A) or γ-linolenic acid (γ-LA) (B) in combination with fasiglifam (Fas, 0.1–10 µM). “PA+Fas” and “γ-LA+Fas” indicate “1 mM palmitic acid +10 µM fasiglifam” and “1 mM γ-LA +10 µM fasiglifam”, respectively. Data shown are mean ± s.e.m. (n = 3).

Figure 5. Point mutations of FFAR1/GPR40 differentially affect Ca²⁺ influx activities of fasiglifam and γ-LA.

(A) Relative cell surface expression levels of FLAG-tagged FFAR1 wild-type and mutant receptors in transfected HEK293T cells were determined using flow cytometric analysis (FACS). (B - J) Effects of FFAR1 point mutations on the Ca²⁺ influx activities of FFAR1 agonists. HEK293T cells were transiently transfected with mock vector (B), wild-type (C), S8A (D), Y91A (E), H137A (F), R183A (G), L186F (H), N244A (I), and R258A (J) constructs. Data are representative of three independent experiments. Error bars indicate s.e.m. (n = 3); γ-LA, γ-linolenic acid.

Figure 6. Schematic model of insulinotropic action of fasiglifam in cooperativity with endogenous FFAs in β cells FFAR1 activation by endogenous FFAs contributes to glucose-stimulated insulin secretion (GSIS) in vivo (left).

Fasiglifam treatment potentiates FFA-induced insulin secretion as an allosteric modulator, whereas FFAs also augment the activity of fasiglifam, showing reciprocal positive cooperativity (middle). Fasiglifam treatment in the absence of FFAs (nonphysiological) results in partial activation of FFAR1 and weak potentiation of GSIS (right).