Conceived and designed the experiments: DCHL GS JDR JLC HT YJL JBH JL. Performed the experiments: DCHL JZ RZ FL KN MC TT EL JYLL XNL LT JBH. Analyzed the data: DCHL JDR GS GRT YJL JL XNL. Contributed reagents/materials/analysis tools: JBH. Wrote the paper: DCHL.
All authors were employed by and shareholders in Amgen during the study period. Patents filed or issued: U.S. Patent No. 7,649,110, U.S. Patent No. 7,816,367 and WO 2005/086661. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials.
Agonists of GPR40 (FFA1) have been proposed as a means to treat type 2 diabetes. Through lead optimization of a high throughput screening hit, we have identified a novel GPR40 agonist called AMG 837. The objective of these studies was to understand the preclinical pharmacological properties of AMG 837. The activity of AMG 837 on GPR40 was characterized through GTPγS binding, inositol phosphate accumulation and Ca2+ flux assays. Activity of AMG 837 on insulin release was assessed on isolated primary mouse islets. To determine the anti-diabetic activity of AMG 837
GPR40 (also known as FFA1 and FFAR1) is a free fatty acid-activated G protein-coupled receptor that is found on the surface of pancreatic β-cells, gastrointestinal enteroendocrine cells, immune cells and parts of the brain. Long chain saturated and unsaturated fatty acids stimulate GPR40, and evidence points to GPR40 being a mechanistic link to the well-known effects of fatty acids to acutely stimulate insulin and incretin secretion
The molecular mechanisms of GPR40-mediated signal transduction have been best studied in pancreatic beta cell lines (
Both small molecule agonists and antagonists of GPR40 have been described
GPR40 is detected in human islets samples from multiple donors
In this report, we describe the preclinical pharmacological characterization of a novel synthetic GPR40 agonist, AMG 837. AMG 837 stimulates glucose dependent insulin secretion in rodent islets in a GPR40-dependent manner.
A high throughput screen for GPR40 agonists resulted in the identification of a lead series of β-substituted phenylpropanoic acids that was further optimized to obtain AMG 837. AMG 837 features an alkyne at the β-position relative to the carboxylic acid and a substituted biaryl group remote to the acid that increases potency on GPR40 relative to the lead series (Houze JB
(A) The chemical structure of AMG 837 is shown. (B–D) The activity of AMG 837 in various GPCR assays was assessed as described in
Activity of AMG 837 on GPR40 was characterized in a variety of biochemical and cell-based assay using cell lines that stably or transiently expressed GPR40. Because GPR40 is coupled to the Gαq class of G-proteins, we measured agonist-stimulated [35S]-GTPγ binding using an antibody capture method
Activity of AMG 837 on GPR40 was further explored in cell-based functional assays for the second messengers inositol phosphate and intracellular Ca2+. These assays were done in the presence of a low concentration of HSA (human serum albumin; 0.01% w/v) in order to minimize the effect of binding of AMG 837 to albumin (further described below). AMG 837 stimulated inositol phosphate accumulation with an EC50 of 7.8±1.2 nM (n = 52) in the A9_GPR40 cell line (
species | Human | Mouse | Rat | Dog | Monkey | Human | Human | Human |
|
GPR40 | GPR40 | GPR40 | GPR40 | GPR40 | GPR41 | GPR43 | GPR120 |
|
13.5±0.8 | 22.6±1.8 | 31.7±1.8 | 71.3±5.8 | 30.6±4.3 | >10,000 | >10,000 | >10,000 |
CHO cells were co-transfected with expression plasmids of a given receptor along with the Ca2+ sensitive bioluminescent reporter aequorin, as described in
In order to explore whether AMG 837 was a full or partial agonist of GPR40, we compared the activity of AMG 837 to that of the naturally occurring fatty acid ligand, docosahexaenoic acid (DHA), in plasmid titration experiments. In our standard aequorin assay, 5 µg of GPR40 expression plasmid (GPR40 under the control of the CMV promoter) are used to transfect ∼10–12 million CHO cells. Under these conditions, AMG 837 behaved as a partial agonist on the GPR40 receptor when compared to DHA, with a maximal activity 85% that of DHA (
AMG 837 is 98.7% bound when incubated with human plasma, indicating extensive binding to plasma proteins. Consistent with this, we found that the EC50 of AMG 837 in the GPR40 aequorin assay was ∼180-fold less potent when tested in the presence of human serum (100% v/v, EC50 = 2,140±310 nM (n = 7)) compared to the assay in 0.01% HSA (
GPR40 is expressed predominantly in the β-cells of the pancreatic islet and activation of GPR40 improves glucose-stimulated insulin secretion (GSIS). We examined the activity of AMG 837 on isolated islets in order to understand the effect of the compound on a relevant primary cell type. On islets isolated from mice, AMG 837 stimulated insulin secretion with an EC50 of 142±20 nM (n = 3,
Islets were isolated from mice and the activity of AMG 837 on insulin secretion was determined. (A) The dose response relationship of AMG 837 and insulin secretion on mouse islets at 16.7 mM glucose was evaluated. (B) In order to determine whether the activity of AMG 837 was GPR40/FFA1 dependent, islets were isolated from GPR40 null mice (
GPR40 agonists have been reported to increase insulin secretion in a glucose- dependent manner, and similarly we found that the activity of AMG 837 was glucose dependent. Activation of GPR40 by AMG 837 did not result in potentiation of glucose stimulated insulin secretion (GSIS) at glucose concentrations ≤5.6 mM (
We next tested the ability of AMG 837 to improve glucose tolerance and stimulate insulin secretion in Sprague-Dawley rats. Sprague-Dawley rats were chosen since they are euglycemic, allowing AMG 837 to be tested at normal glucose levels and during the challenged state following a glucose bolus. AMG 837 displays excellent pharmacokinetic properties in multiple species (Houze JB
AMG 837 administration did not have any effect on glucose levels prior to the glucose tolerance test (30 minutes following AMG 837 administration). Following administration of glucose, plasma glucose levels were suppressed in an AMG 837 dose-dependent manner (
8-week old Sprague-Dawley rats were treated with a single bolus of AMG 837 (at 0.03, 0.1 and 0.3 mg/kg, n = 6/group) by oral gavage 30-minutes prior to an intraperitoneal glucose challenge at t = 0 minutes. (A) Blood glucose measurements were taken during prior to and following glucose challenge. Black circle = vehicle, blue triangle = 0.03 mg/kg AMG 837, green diamond = 0.1 mg/kg AMG 837 and purple square = 0.3 mg/kg AMG 837 (B) The glucose AUC (from −30 to 120 minutes) during the course of the experiments were calculated. (C) Plasma insulin levels were measured using ELISA. Black circle = vehicle, blue triangle = 0.03 mg/kg AMG 837, green diamond = 0.1 mg/kg AMG 837 and purple square = 0.3 mg/kg AMG 837 (D–D) Two successive glucose challenges were conducted in Sprague-Dawley rats following a single oral dose of vehicle (n = 4, black circle) or AMG 837 at 0.3 mg/kg (n = 4, purple diamond). AMG 837 was dosed at −30 minutes, and glucose was administered by
The improvement in post-prandial glucose was a result of an increase in glucose-stimulated insulin secretion. In animals treated with AMG 837, there was a dose-dependent increase of plasma insulin levels following the glucose challenge (
We further tested whether a single dose of AMG 837 could improve post-prandial glucose following consecutive glucose challenges. A single dose (0.3 mg/kg) of AMG 837 was administered to Sprague-Dawley rats followed by two intraperitoneal glucose challenges 3 hours apart. AMG 837 improved blood glucose levels during both glucose challenges (p<0.01,
We next tested the effect of AMG 837 in the insulin resistant Zucker fatty (
8-week old Zucker fatty rats were administered a single bolus of AMG 837 (at 0.3, 1 and 3 mg/kg, n = 6/group) by oral gavage 30-minutes prior to an intraperitoneal glucose challenge at t = 0 minutes. (A) Blood glucose during the IPGTT (black circle = vehicle, blue triangle = 0.3 mg/kg AMG 837, green diamond = 1 mg/kg AMG 837 and purple square = 3 mg/kg AMG 837) (B) Glucose AUC (from −30 to 120 minutes) during the IPGTT. (C) Plasma insulin levels during the IPGTT (black circle = vehicle, blue triangle = 0.3 mg/kg AMG 837, green diamond = 1 mg/kg AMG 837 and purple square = 3 mg/kg AMG 837). Statistical significance compared to vehicle treated animals is denoted by * (p<0.5), ** (p<0.01), *** (p<0.001) and **** (p<0.001) as determined by one-way or two-way ANOVA and colors match the corresponding groups in the figure legend.
In order to understand the effect of AMG 837 following multiple doses, AMG 837 was dosed at 0.03, 0.1 and 0.3 mg/kg by oral gavage daily for 21-days. Thirty minutes following the first dose, an IPGTT was performed. AMG 837 improved glucose levels during the IPGTT (
8-week old Zucker fatty rats were administered a single bolus of AMG 837 (at 0.03, 0.1 and 0.3 mg/kg, n = 6/group) by oral gavage 30-minutes prior to an intraperitoneal glucose challenge at t = 0 minutes. (A) Blood glucose during the IPGTT (black circle = vehicle, blue triangle = 0.03 mg/kg AMG 837, green diamond = 0.1 mg/kg AMG 837 and purple square = 0.3 mg/kg AMG 837). (B) Glucose AUC (from −30 to 120 minutes) during the IPGTT. (E) Plasma insulin levels during the IPGTT (black bar = vehicle, blue bar = 0.03 mg/kg AMG 837, green bar = 0.1 mg/kg AMG 837 and purple bar = 0.3 mg/kg AMG 837). Once daily dosing was continued for 21-days. On day 21, an IPGTT was performed in an identical manner to that on day 1. (C) Blood glucose (D) glucose AUC (from −30 to 120 minutes) and (F) plasma insulin levels were measured from the day 21 IPGTT. Figure legends are identical to those of the day 1 figures. (G) Body weights of the animals were followed through the course of the 21-day study; no difference in BW was observed between the groups. (H) Total plasma concentration of AMG 837 30-minutes following the final dose on day 21. Statistical significance compared to vehicle treated animals is denoted by * (p<0.5), ** (p<0.01) and *** (p<0.001) as determined by one-way or two-way ANOVA.
Administration of AMG 837 was continued daily for 21-days in order to test the effects of AMG 837 following multiple doses. A second IPGTT was performed 30 minutes following the final dose on day 21 and AMG 837 lowered glucose levels following glucose challenge (
The GPR40 (FFA1) receptor appears to be a viable and amenable target for small molecule development for use in treating type 2 diabetes. Several GPCRs expressed on the pancreatic β-cell, such as GPR119 and GPR40 have gained considerable interest as drug targets
In addition to the expression on the β-cell and enteroendocrine cells, GPR40 expression has also been described in osteoclasts
Several synthetic GPR40 agonists have been described in recent years, including GW9508, TAK-875, TUG-424 and others
AMG 837 is a partial GPR40 agonist that potently activated GPR40 in cell-based assays and isolated islets. AMG 837 did not potentiate insulin secretion in islets from GPR40 knockout mice (
One important aspect of any GPR40 agonist that may be used to treat a chronic condition such as type 2 diabetes is the potential for tachyphylaxis. Results from two experiments addressed this potential. First, a single dose of AMG 837 improved glucose levels following consecutive glucose challenges, indicating a lack of acute tachyphylaxis (
In conclusion, we have discovered and characterized a novel GPR40 (FFA1) agonist, AMG 837. Further studies of AMG 837 will delineate the biological functions of GPR40 and whether the GPR40 pathway can be modulated to treat human disease.
GPR40 was amplified from genomic DNA using standard PCR techniques and cloned into pcDNA3.1 (Invitrogen) or the retroviral vector pLPC (from Dr. David Mu). Constructs were verified by DNA sequencing. An A9 cell line stably transfected with pLPC-GPR40 (A9_GPR40) was created by retroviral infection of mouse A9 cells (ATCC catalog # CRL-1811) followed by selection on 2 µg/mL puromycin (Sigma-Aldrich) as described
AMG 837 (
A GTPγS binding assays using an anti-Gα-protein scintillation proximity assay format was employed essentially as described
A9_GPR40 cells were plated in 96-well plates containing 20,000 cells/well in DMEM containing 10% FBS. After the cells attached to the well surface, the media was replaced with inositol free DMEM containing 10% dialyzed FBS and 1µCi, mL 3H-myo-inositol and incubated for 16 hours. Compounds were diluted in HBSS/10 mM LiCl, pH7.4 in 0.01% HSA and added directly to cells. Following 1 hour incubation at 37°C, the media was replaced with 100 µl of 20 mM formic acid to quench the reaction. 50 µL of the extract was then added to 100 µL of SPA beads, incubated overnight, and measured on a TopCount the following day.
CHO cells were plated in 15 cm plates containing 8×106 cells/plate in DMEM/F12 containing 10% FBS. The following day, cells were transfected with 5 µg of GPR40 expression plasmid and 5 µg of aequorin expression plasmid (Euroscreen) complexed with 30 µL of Lipofectamine 2000. In plasmid titration experiments, the amount of GPR40 expression plasmid was reduced, but the total amount of DNA transfected was kept constant by adding in empty vector DNA. Sixteen to twenty-four hours post-transfection, cells were washed with PBS and detached from the plate with 2 mL trypsin (0.25% in HBSS). 28 mL of HBSS containing a desired amount of HSA (0.01% or 0.625% w/v, Sigma-Aldrich) or human serum (100% v/v, Sigma-Aldrich) was added to the detached cells and coelenterazine was added to final concentration of 1 µg/mL. Cells were allowed to incubate in coelenterazine containing buffer for 2 hours prior to assay. AMG 837 and DHA (Sigma-Aldrich) stock solutions were prepared in DMSO and then diluted in HBSS buffer containing the % HSA identical to that in which the cells were incubated in. Compounds were allowed to complex with HSA for 1 hr at 37°C. Aequorin activity was measured using a microlumat.
Islets were prepared from mouse pancreas following injection of collagenase into the common bile duct followed by purification on a histopaque gradient. Animals were euthanized by CO2 inhalation and the abdominal cavity was opened. The common bile duct was clamped just proximal to the duodenum and the pancreas was perfused with 3–5 mL of ice cold collagenase (0.67 mg collagenase/mL in HBSS containing 25 mM Hepes pH 7.4 and 1% penicillin/streptomycin). The inflated pancreas was excised and collagenase digestion was allowed to proceed for 20 minutes in a 37°C water bath. The digestion was quenched by addition of quenching buffer (10% FBS in HBSS containing 25 mM Hepes pH 7.4 and 1% penicillin/streptomycin). Islets were washed twice with quenching buffer following centrifugation at 300×g for 2 minutes. Islets were pelleted and resuspended in 10 mL of Histopaque 1119 (Sigma-Aldrich). Next, 10 mL of Histopaque 1077 was layered on top of the Histopaque 1119 layer, and 10 mL of quenching buffer was carefully layered on the very top. The tube was centrifuged at 1000×g for 30 minutes and the islets were isolated with a pipet and washed in culture media (RPMI modified, 10% FBS, 25 mM Hepes, 1% penicillin/streptomycin, pH 7.4, 37°C). Islets were allowed to culture for 48 hours in a cell incubator and were then handpicked under a dissection microscope and transferred to a 96-well transwell plate (Corning). Insulin secretion assays were performed in KRBH, pH7.4 (consisting of 129 mM NaCl, 4.8 mM KCl, 1.2 mM KH2PO4, 1.2 mM MgSO4, 10 mM HEPES, 2.5 mM CaCl2, 25 mM NaHCO3) and the insulin secreted into the supernatant was measured using an insulin ELISA (Alpco).
All procedures on animals were approved by the Amgen San Francisco Institutional Animal Care and Use Committee (approved protocol #11-04). Eight week old male Sprague-Dawley (SD) rats (Harlan, Indianapolis, Indiana) and 8-week old male Zucker fatty rats (
AMG 837 was formulated for oral dosing using 1% methylcellulose (CMC), 1% Tween 80 (Sigma-Aldrich, St. Louis, MO). For evaluation of AMG 837 following a single dose in rats, animals were fasted overnight and then randomized into dose groups based on their body weights. Thirty minutes after oral administration of their respective treatments, the animals received a 1 g/kg glucose challenge dose by intraperitoneal injection. Blood samples were collected at 0, 5, 15, 30, 60, and 120 minutes via tail vein after the glucose challenge. Glucose levels were monitored with a Glucometer (Elite XL). Plasma insulin was measured using a rat insulin ELISA kit (ALPCO Diagnostics, Windham, NH). For evaluation of AMG 837 in Zucker fatty rats, animals were randomized based on body weight and received either vehicle, 0.03 mg/kg, 0.1 mg/kg, or 0.3 mg/kg AMG 837 once daily for 21 days by oral gavage. Treatments were administered between 0900 and 1000 h during the light cycle. On days 1 and 21, an intraperitoneal glucose tolerance test (IPGTT) was performed as described above.
AMG 837 plasma concentrations were measured using a sensitive and selective LC/MS/MS method. Briefly, following the addition of internal standard, samples were extracted using protein precipitation. The resulting supernatants were dried, reconstituted and injected into a a triple quadruple LC/MS/MS instrument (API 3000, AB Sciex, Foster City, CA) for detection. Concentrations of AMG 837 in plasma samples were calculated using a calibration curve with a lower limit of quantitation of 1 ng/mL.
Data was expressed as mean ± SEM. One- or two-way ANOVA (GraphPad Prism) was used to assess statistical significance between control and treatments.
The authors wish to acknowledge Simon Wong, Peter Coward, Julio Medina, Bei Shan and Juan Jaen for comments on the manuscript and numerous helpful discussions.