Truth Seeker
New Member
Abstract
Pancreatic β-cell dysfunction is a hallmark event in the pathogenesis of type 2 diabetes. Injectable peptide agonists of the glucagon-like peptide 1 (GLP-1) receptor have shown significant promise as antidiabetic agents by virtue of their ability to amplify glucose-dependent insulin release and preserve pancreatic β-cell mass. These effects are mediated via stimulation of cAMP through β-cell GLP-1 receptors. We report that the Gαs-coupled receptor GPR119 is largely restricted to insulin-producing β-cells of pancreatic islets. Additionally, we show here that GPR119 functions as a glucose-dependent insulinotropic receptor. Unlike receptors for GLP-1 and other peptides that mediate enhanced glucose-dependent insulin release, GPR119 was suitable for the development of potent, orally active, small-molecule agonists. The GPR119-specific agonist AR231453 significantly increased cAMP accumulation and insulin release in both HIT-T15 cells and rodent islets. In both cases, loss of GPR119 rendered AR231453 inactive. AR231453 also enhanced glucose-dependent insulin release in vivo and improved oral glucose tolerance in wild-type mice but not in GPR119-deficient mice. Diabetic KK/Ay mice were also highly responsive to AR231453. Orally active GPR119 agonists may offer significant promise as novel antihyperglycemic agents acting in a glucose-dependent fashion.
THE GUT HORMONES glucagon-like peptide 1 (GLP-1) and gastric inhibitory polypeptide (GIP) promote normoglycemia acutely by enhancing glucose-stimulated insulin release and chronically by maintaining pancreatic β-cell mass (1, 2, 3, 4). These effects occur via β-cell-expressed, class B G protein-coupled receptors (GPCRs) that in turn mediate elevated intracellular cAMP. Decreased GIP responsiveness may contribute to the pathogenesis of type 2 diabetes, resulting in mixed enthusiasm toward GIP receptor agonists as a means to treat the disease (5, 6). By contrast, strategies that enhance GLP-1 receptor function have shown significant therapeutic promise. These strategies have so far consisted of injectable peptidic GLP-1 receptor agonists or, alternatively, blockade of endogenous GLP-1 metabolism through selective inhibition of dipeptidyl peptidase 4 (1, 7). Orally active, small-molecule GLP-1 receptor agonists have proven elusive, a feature that unfortunately is characteristic of class B GPCRs. Meanwhile, the spectrum of signaling peptides affected by inhibition of dipeptidyl peptidase 4 remains unclear and could potentially extend significantly beyond GLP-1 and GIP (8, 9). It is therefore worthwhile to search for therapeutic approaches that afford both the physiological selectivity of GLP-1 signaling and the opportunity for orally active treatment modalities.
GPR119 was recently identified as a class A, islet-enriched receptor that could potentially mediate the insulinotropic actions of lysophosphatidylcholine observed in vitro (10). However, the islet cell type expressing GPR119 was not identified, and a separate study suggested that oleoylethanolamide was a significantly more potent GPR119 agonist (11). Oleoylethanolamide and the related lipid amide anandamide have been identified as modulators of food intake (11, 12, 13), but there has been no evaluation of these lipid amides with regard to glucose homeostasis. Thus, the importance of GPR119 in glucose homeostasis is uncertain. Moreover, there have been no studies that address whether GPR119 has a sufficiently robust impact on islet biology to indicate its value as a target for therapeutic intervention in type 2 diabetes.
Here we show that GPR119 is a robust mediator of glucose-dependent insulin release and enhanced glucose homeostasis. Importantly, GPR119 is highly amenable to the development of potent, selective, small-molecule agonists. These data strongly suggest that GPR119 may be a therapeutically significant, glucose-dependent, insulinotropic receptor.
Materials and Methods
Cell lines
HEK293, RIN-5F, and HIT-T15 cells were obtained from the American Type Culture Collection (Manassas, VA). RIN-5F is a rat insulinoma cell line that does not express GPR119. GPR119-expressing RIN-5F stable lines were generated by cotransfection of pCDNA3.1 (Invitrogen, Carlsbad, CA) and a human GPR119 expression plasmid encoding amino acids 2—335 of the receptor with a hemagglutinin epitope tag at the N terminus. After selection in G418, clones were chosen for additional studies based on hemagglutinin immunofluorescence and responsiveness to GPR119 agonists.
Expression analysis
For quantitative RT-PCR of mouse GPR119, a fluorescence (6FAM)-labeled probe (5′-CTGCTCAACCCACTCATCTATGCCTATTGG-3′) and amplification primers (forward, 5′-CTCGGCGTGGGCAACTC-3′; reverse, 5′-CTGCAGTCGCACCTCCTTCT-3′) were generated for TaqMan analysis. A β-actin probe (fluorescent VIC-labeled) was used as internal control. Total RNA was prepared with RNA-Bee, DNase treated (DNA-free kit; Ambion, Austin, TX), and converted to cDNA (iScript cDNA synthesis kit; Bio-Rad, Hercules, CA). Relative expression levels of GPR119 were determined by using TaqMan 7900 HT Sequence Detection System (ABI Prism; Applied Biosystems, Foster City, CA) and normalized against β-actin internal control.
For immunofluorescence studies, a polyclonal antibody (AR361) was generated in rabbits using a BSA-conjugated synthetic peptide with sequences corresponding to residues 314—336 (C-terminal tail) of mouse GPR119 (RGPERTRESAYHIVTISHPELDG). Pancreata were obtained from 4% paraformaldehyde/PBS-perfused mice or rats and embedded in optimal cutting temperature (OCT) embedding medium (Tissue Tek, Torrance, CA). Pancreatic sections were immunostained with anti-GPR119 (AR361) in combination with a mouse monoclonal antiinsulin antibody (Santa Cruz Biotechnology, Santa Cruz, CA), or a goat antiglucagon antibody (Santa Cruz Biotechnology). Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Immunoreactive signals were detected by using Alexa Fluor-488 (Molecular Probes, Eugene, OR) or Cy3 (Jackson ImmunoResearch Laboratories, West Grove, PA) conjugated secondary antibodies and fluorescence microscopy.
In situ hybridization
Pancreas from male Sprague Dawley rat was immersed in 4% paraformaldehyde overnight at 4 C, immersed in 20% sucrose/PBS for 2 d at 4 C and then frozen in optimal cutting temperature compound in embedding molds over dry ice. Sections (12 μm thick) were collected on SuperFrost Plus slides (VWR, San Diego, CA; no. 48311-703). The following plasmids were used for probe synthesis. For GPR119, the full-length rat sequence was inserted into pCR4-TOPO (Invitrogen). For insulin, a 201-bp rat insulin cDNA fragment was inserted into pCRII-TOPO (Invitrogen). For preproglucagon, a 544-bp rat preproglucagon cDNA fragment was inserted into pCRII-TOPO. Sense and antisense 33P-radiolabeled GPR119 probes and digoxigenin-labeled preproglucagon probes were generated using standard protocols. Double in situ hybridization was performed with the addition of digoxigenin-labeled probes (500 ng/slide) and salmon sperm (20 μg/slide) together with the 33P-labeled probes. After the post-hybridization step, sections were washed in Tris/NaCl (TN) buffer containing 100 mM Tris (pH 7.5) and 150 mM NaCl for 5 min. Sections were then placed in 0.5% casein/TN blocking solution for 30 min and then incubated for 2 h with alkaline phosphatase (AP)-conjugated antidigoxigenin antibody (Roche, Indianapolis, IN; no. 1093274) in 0.5% casein/TN solution (1:300). Sections were then washed in TN three times and then in 100 mM Tris (pH 9.5), 100 mM NaCl, and 50 mM MgCl2 four times and incubated in color reaction [0.2 mg/ml levamisole, 3.4 μl/ml nitroblue tetrazolium (Roche no. 1383213), 3.5 μl/ml 5-bromo-4-chloro-3-indolyl phosphate (Roche no. 1383221) in 100 mM Tris (pH 9.5), 100 mM NaCl, and 50 mM MgCl2] for 20—30 min. Antibody was stripped off by incubating sections in 0.1 M glycine and 0.5% Triton X-100 for 10 min and washed in water. Sections were fixed in 2.5% glutaraldehyde for 1—2 h, washed with water, and then air dried. Once dried, sections were exposed to x-ray-sensitive film (Bio-Max; Kodak, Rochester, NY; no. IB-856 7232) for 5—7 d and dipped in photographic emulsion (Ilford Scientific K.5D emulsion; Polysciences, Warrington, PA; no. 17537), dried, and stored in a slide box with desiccant at 4 C for 6—8 wk. After development of dipped slides following manufacturer recommendations (Kodak D19), sections were washed extensively in water, air dried, and mounted with coverslips for microscopic examination.
In vitro assays
cAMP measurements were done with a Flash Plate adenylyl cyclase kit (NEN Life Science Products, Boston, MA) according to the supplier's protocol. Briefly, HEK293 cells were transfected with either empty vector DNA or GPR119 expression plasmid DNA (described above) using Lipofectamine (Invitrogen). After 24 h, transfected cells were harvested in GIBCO (Gaithersburg, MD) cell dissociation buffer (catalog item 13151-014) and resuspended in assay buffer (50% 1× PBS/50% stimulation buffer). Compounds were incubated with 105 cells per well for 60 min at room temperature. After another 2-h incubation with tracer in detection buffer, plates were counted in a Wallac (Gaithersburg, MD) MicroBeta scintillation counter. Values of cAMP per well were extrapolated from a standard cAMP curve that was included on each assay plate. cAMP assays in hamster insulinoma-derived HIT-T15 cells and GPR119-transfected RIN-5F stable lines were performed essentially in the same way but without transient transfection. In experiments designed to selectively lower endogenous GPR119 expression in HIT-T15 cells, cells were transfected with either a control RNA oligo with sequence derived from hamster GPR119 in sense orientation (5′-CUAUGCUGCUAUCAAUCUA-3′) (control) or in antisense orientation (5′-UAGAUUGAUAGCAGCAUAG-3′) [small interfering RNA (siRNA)]. Forty-eight hours after transfection, cells were assayed for GPR119 expression and for agonist-stimulated cAMP production as described above.
For assays measuring inositol phosphate accumulation, HEK293 cells in 96-well plates were transfected with the indicated G protein receptor expression plasmid. Additionally, cells were cotransfected with either an empty cytomegalovirus (CMV) vector or a Gαq/Gαi expression plasmid in which the terminal six residues of Gαq were replaced by the corresponding residues of Gαi. The receptor and G protein chimera were transfected at a molar ratio of 4:1, respectively. The following day, cells were incubated in 100 μl inositol-free/serum-free DMEM (GIBCO) containing 0.4 μCi [3H]myoinositol (Perkin-Elmer, Norwalk, CT). Cells were incubated overnight, after which the medium was replaced with 100 μl inositol-free/serum-free DMEM containing 10 μM pargyline (Sigma Chemical Co., St. Louis, MO) and 10 mM lithium chloride (Sigma). One hour later, the cells were lysed and inositol phosphates were isolated using chromatography on AG1-X8 formate resin (Bio-Rad). After additional purification through binding, four washes, and elution from a multiscreen filter plate (Millipore, Bedford, MA), eluted counts were quantified using a Wallac scintillation counter.
For in vitro insulin release, HIT-T15 insulinoma cells were plated in 24-well plates (2.5 × 105 cells per well) for insulin release assays. One day before the assay, culture media were changed to DMEM (3 mM glucose) with 10% dialyzed horse serum and 2.5% fetal bovine serum. On the next day, cells were washed twice with PBS and incubated for 1 h with test compounds in DMEM in the presence of 15 mM glucose in 0.25 ml HEPES-buffered Krebs-Ringer buffer. Supernatant insulin levels were determined using an Ultra Sensitive Insulin ELISA kit (Crystal Chem Inc., Downers Grove, IL). Insulin assays with RIN-5F stable lines were performed in essentially the same manner.
Insulin release assays with pancreatic islets were done using islets isolated from female Sprague Dawley rats (body weight, 175—185 g) or male C57BL/6 mice (body weight, ∼25 g). Insulin release was determined in static islet incubation as described previously (14). Briefly, groups of five islets each were placed in incubation wells. After a 30-min preincubation with HEPES-buffered Krebs-Ringer buffer (pH 7.4) containing 5 mM glucose, islets were transferred to wells containing 2 ml HEPES-buffered Krebs-Ringer buffer and different concentrations of glucose and test compounds. The studies were performed at 37 C in a water bath shaker with an atmosphere of 95% O2/5% CO2. Samples of incubation buffer were collected at 60 min for insulin determination using ELISA methodology.
In vivo experiments
C57BL/6 male mice were obtained from Harlan (Indianapolis, IN). KK/Ay and db/db male mice were purchased from The Jackson Laboratory (Bar Harbor, ME). All in vivo animal protocols were approved by the Animal Welfare Committee of Arena Pharmaceuticals, Inc., and in adherence with government regulations. For the oral glucose tolerance test, overnight fasted mice (n = 6 per treatment) were given either vehicle (80% polyethylene glycol 400/10% Tween 80/10% ethanol) or test compounds at desired doses via oral gavage. A glucose bolus was then delivered (3 g/kg orally or 2 g/kg ip). Plasma glucose levels were determined at desired time points over a 2-h period using blood (∼5 μl) collected from tail nick and a glucose meter. For insulin pharmacodynamic studies, vehicle or AR231453 was administered orally to fasted animals (n = 6 per treatment group and time point). After 30 min, a glucose bolus of 3 g/kg was administered orally. Blood was collected in heparinized blood collection tubes at desired time points. Plasma samples were obtained via centrifugation at 500 × g for 20 min and assayed for insulin as described above.
Statistical analysis
Statistical analysis was performed using t test and one-way ANOVA using either Microsoft Excel or Prism statistical methods.
Results
GPR119 is a β-cell GPCR
Quantitative TaqMan analysis indicated that GPR119 expression is greatly enriched in islets relative to whole pancreas (Fig. 1A⇓), consistent with previous reports (10). Moreover, significant GPR119 expression was present in the islets of hyperglycemic db/db mice (Fig. 1A⇓). To determine the predominant islet cell types expressing GPR119, we first performed in situ hybridization analysis on pancreatic sections. Autoradiography showed that antisense GPR119 hybridized to sections in an islet-like pattern, whereas no such signal was observed with the sense probe (Fig. 1B⇓, panels a and b). Examination of emulsion-dipped sections indicated significant hybridization throughout the core of pancreatic islets (Fig. 1B⇓, panels c and d), strongly suggesting that GPR119 is expressed in the β-cell population. These data also suggested that a subset of glucagon-producing cells might express GPR119, but this was difficult to quantify. To define further the islet cell types that express GPR119, polyclonal anti-GPR119 antisera were raised against a synthetic peptide derived from the C-terminal tail region of the mouse sequence. Immunofluorescence analysis of rat (Fig. 1C⇓) or mouse (see Fig. 7D⇓) pancreatic sections indicated that GPR119 colocalizes primarily with insulin. Moreover, the immunofluorescent signal was absent in pancreatic sections from GPR119-deficient mice (see Fig. 7D⇓). Collectively, these data indicate that GPR119 expression is largely restricted to the β-cells of pancreatic islets. However, it remains possible that there might be additional GPR119 expression in a subset of α-cells or in other islet cell types.
AR231453 is a GPR119-selective small molecule agonist
In transfected HEK293 cells, high-level expression of GPR119 stimulated a significant increase in cAMP levels relative to controls (Fig. 2A⇓). Such constitutive activity is occasionally seen with class A GPCRs (15) and shows that GPR119 couples efficiently to Gαs. In inositol phosphate assays, a constitutively active variant of the Gαq-coupled 5-hydroxytryptamine 2A receptor (16) was highly active, but GPR119 was inactive (Fig. 2B⇓). Thus, GPR119 exhibits poor coupling efficiency to Gαq. GPR119 also failed to produce inositol phosphate when coexpressed with a Gαq/Gαi chimera (Fig. 2B⇓), which has previously been shown to redirect Gαi-coupled receptors to stimulate inositol phosphate accumulation (17). The Gαi-coupled α2A adrenergic receptor functioned very efficiently in this assay (Fig. 2B⇓). Collectively, these data show that GPR119 couples selectively to Gαs.
To evaluate GPR119 function pharmacologically, we screened a library of approximately 55,000 small-molecule compounds. Based on the foregoing, a cAMP assay was used to identify GPR119 modulators. One hit was identified, from which we ultimately developed an agonist termed AR231453 (Fig. 3A⇓), which potently increased cAMP in transfected HEK293 cells (EC50 = 5.7 ± 1.6 nM; Fig. 3B⇓) and stably transfected RIN-5F insulinoma cells (not shown). AR231453 was inactive in cells transfected with empty vector (Fig. 3B⇓) or GLP-1 receptor (Fig. 3C⇓). Indeed, AR231453 was inactive at all other GPCRs tested (more than 230, including all known pancreatic islet receptors), indicating that it is highly selective for GPR119.
To test whether AR231453 stimulates cAMP production via endogenously expressed GPR119 in β-cells, we employed the hamster β-cell line HIT-T15, which expresses GPR119 at high levels (Fig. 3E⇑, inset). AR231453 potently stimulated cAMP accumulation (EC50 = 4.7 nM) with a maximal efficacy similar to that seen with forskolin (Fig. 3D⇑). Pretreatment of HIT-T15 cells treated with a hamster GPR119-selective siRNA markedly lowered receptor levels and correspondingly rendered AR231453 virtually inactive (Fig. 3E⇑).
GPR119 agonist stimulates insulin release in vitro
The observation that GPR119 couples to Gαs and is expressed in β-cells prompted us to test the hypothesis that GPR119 is an insulinotropic receptor. In this regard, AR231453 significantly enhanced insulin release in HIT-T15 cells, with an EC50 of 3.5 nM (Fig. 4A⇓), similar to its effect on cAMP (EC50 of 4.7 nM; see Fig. 3D⇑), and similar in efficacy to forskolin. To assess the GPR119 dependence of the insulinotropic effects seen with AR231453, we tested its activity in RIN-5F cells lacking GPR119 (RIN-5F/vector) and in RIN-5F cells stably transfected with human GPR119 (RIN-5F/hGPR119). AR231453 stimulated insulin release in RIN-5F/hGPR119 cells but not in RIN-5F/vector cells (Fig. 4B⇓). These studies indicate that AR231453 is insulinotropic specifically in GPR119-expressing β-cell lines. We next examined the effect of GPR119 agonist on glucose-stimulated insulin release in isolated rat and mouse islets. At 15 mM glucose, 300 nM AR231453 enhanced insulin release in rat islets similarly to GLP-1 (Fig. 4C⇓). However, this compound had no effect on islets incubated in 5 mM glucose. The insulinotropic effect of GPR119 agonists in islets is therefore glucose dependent. AR231453 also stimulated insulin release in isolated mouse islets at glucose concentrations ranging from 8—17 mM (Fig. 4D⇓). This indicates that the insulinotropic effect of GPR119 agonists requires only modest elevations of blood glucose.
GPR119 agonist improves glucose tolerance and enhances glucose-dependent insulin release in mice
AR231453 exhibited good oral bioavailability in C57BL/6 mice, attaining micromolar plasma levels for more than 2 h at the doses used here. AR231453 (20 mg/kg, orally) markedly improved oral glucose tolerance in a dose-dependent fashion, with efficacy similar to maximally effective doses of the sulfonylurea glyburide (Fig. 5A⇓). On average, AR231453 maximally inhibited the glycemic excursion by approximately 42% (21 independent experiments). When glucose was administered ip, AR231453 improved glucose tolerance somewhat less effectively (23%; three independent experiments) (Fig. 5B⇓).
Next, we examined the effects of AR231453 on glucose homeostasis in rodent models of type 2 diabetes. GPR119 expression was well maintained in β-cells of ZDF diabetic rats (at ages before β-cell loss), fa/fa rats, db/db mice, and KK/Ay mice (Fig. 5C⇑ inset, Fig. 1A⇑, and data not shown). AR231453 significantly improved oral glucose tolerance in 8-wk-old KK/Ay mice (Fig. 5C⇑) and in 6-wk-old db/db mice (Fig. 5D⇑). Similar effects were seen in rat diabetic animal models with other GPR119 agonists (data not shown). Thus, GPR119 agonists improve glycemic control in both normal and diabetic mice.
To test whether the improved glucose tolerance was due to enhanced insulin release, we performed an insulin pharmacodynamic analysis in mice (Fig. 6A⇓). Glucose administered alone increased plasma insulin concentration 2 min and onward and peaked at 20 min. Mice pretreated with AR231453 had markedly elevated insulin levels beginning from 5 min and peaked at 10 min after glucose bolus (Fig. 6A⇓, bottom), which preceded the observed improvement in glucose tolerance (Fig. 6A⇓, top). By contrast, AR231453 was not effective in an insulin tolerance test (not shown). Taken together, these data suggest that AR231453 acts primarily by enhancing glucose-dependent insulin release from pancreatic β-cells, which subsequently results in enhanced glucose disposal. The glucose-dependent nature of AR231453 function was further evaluated by administering very high doses (100 mg/kg) to fasted C57BL/6 mice (Fig. 6B⇓). Even under these conditions, AR231453 did not induce hypoglycemia or trigger insulin release. By contrast, glyburide, a sulfonylurea antidiabetic agent known to stimulate insulin release in a glucose-independent fashion, elicited marked hypoglycemia (Fig. 6B⇓, top) that was associated with enhanced insulin release (Fig. 6B⇓, bottom). Thus, the actions of the GPR119 agonist AR231453 are glucose dependent both in isolated rat islets (Fig. 4⇑, C and D) and in vivo.
AR231453 is inactive in GPR119-deficient mice
To demonstrate that the in vivo effects of AR231453 occur via GPR119, we generated GPR119-deficient mice (Fig. 7⇓). As in human, the gene for mouse GPR119 is located on the X chromosome (Unigene Cluster Mm.34953) (Fig. 7A⇓). Deletion of GPR119 was confirmed by Southern blot analysis (Fig. 7B⇓), RT-PCR analysis of pancreatic GPR119 mRNA (Fig. 7C⇓), and immunofluorescent analysis of GPR119 receptor on pancreatic sections from the knockout mice (Fig. 7D⇓). Islets from GPR119-deficient mice retained normal morphology (Fig. 7D⇓) and a normal response to glucose and GLP-1 (Fig. 7E⇓). These mice also had normal size, body weight, and fed/fasted blood glucose levels (not shown) and a typical response to glyburide in glucose tolerance tests (Fig. 7⇓, F and G). Thus, deletion of GPR119 does not grossly impact glucose homeostasis. However, isolated islets from these animals no longer responded to AR231453 (Fig. 7E⇓). AR231453 also had no effect on glucose tolerance in GPR119-deficient mice (Fig. 7G⇓), indicating that its observed in vivo activities were indeed mediated via GPR119.
Discussion
These data provide strong evidence that GPR119 is a significant modulator of glucose-stimulated insulin release in pancreatic β-cells. The GPR119-selective agonist AR231453 stimulated cAMP accumulation and insulin release in 1) transfected RIN-5F insulinoma cells, 2) HIT-T15 insulinoma cells that express the receptor endogenously, and 3) isolated rat and mouse islets. These effects were substantially the same as those elicited by forskolin or GLP-1, suggesting significant physiological relevance. In vivo, AR231453 stimulated insulin release and improved glucose tolerance with comparable efficacy to glyburide, a very robust stimulator of insulin release in rodents and humans (18). Importantly, the activity of AR231453 was particularly impressive in diabetic KK/Ay mice. All these effects are very likely occurring via GPR119. RIN-5F control cells lacking GPR119 are unresponsive to AR231453. In HIT-T15 cells, siRNA-mediated reduction in GPR119 levels was associated with virtually complete loss of responsiveness to AR231453. Finally, AR231453 does not enhance glucose-stimulated insulin release in islets derived from GPR119-deficient mice, nor does it improve glucose tolerance in GPR119-deficient mice. GPR119-deficient mice had no obvious defect in homeostasis, but this perhaps is not surprising because mice lacking either GIP receptor or GLP-1 receptor have very minor alterations in this regard (19, 20, 21).
Because we did not measure AR231453 exposures in GPR119-deficient mice, its inactivity in glucose tolerance tests performed in these mice could simply be due to strain-dependent differences in plasma exposure of the compound. However, this seems unlikely. In C57BL/6 mice, AR231453 achieves exposures more than 100-fold greater than its in vitro EC50. Additionally, the compound is effective in all other strains of mice used here. Numerous, structurally distinct compounds have been assessed in wild-type littermates and GPR119-deficient mice, with similar outcomes to those seen for AR231453 (not shown). Taken together, these data strongly suggest that the actions of AR231453 are mediated via GPR119 and firmly establish that GPR119 is a physiologically significant mediator of glucose homeostasis.
Although these studies support the hypothesis that GPR119 regulates glucose homeostasis by direct enhancement of pancreatic β-cell function, it is important to recognize that GPR119 may work via additional mechanisms. The effectiveness of AR231453 is reduced by almost 50% when glucose is administered ip, suggesting that its actions may also involve modulation of incretin signaling. This hypothesis warrants further investigation.
Our studies in islets and mice indicate that GPR119 stimulates insulin release in a glucose-dependent manner. Even when administering doses of AR231453 (100 mg/kg in C57BL/6 mice) that greatly exceed the maximal in vivo efficacy of this compound, there is no observed stimulation of insulin release or lowering of glucose under fasted conditions. By contrast, glyburide elicited significant insulin release and marked hypoglycemia in fasted mice. These data are entirely consistent with the well-established mechanism by which GIP and GLP-1 stimulate insulin release (22).
A key distinguishing feature of GPR119 is the demonstrated ability to develop potent, orally active, small-molecule agonists of this receptor. This is in contrast to Gαs-coupled, β-cell peptide receptors for GLP-1, GIP, vasoactive intestinal polypeptide, and pituitary adenylate cyclase-activating polypeptide (23), for which there are no known activators of this type. GPR119 is likely unique in this regard because it is the only class A, nonpeptide GPCR identified that mediates glucose-dependent insulin release via increased intracellular cAMP.
It was recently reported that activators of GPR119 may markedly reduce food intake in rodents (11). Our studies with a number of GPR119 agonists have been inconclusive. In general, a reduction in feeding behavior has been observed only at doses significantly higher than those required to impact glucose homeostasis (unpublished results). This conceivably could be due to comparatively lower agonist exposure in tissues such as the brain, which expresses GPR119 at low levels in the amygdala, thalamus, and perhaps other subregions. Alternatively, these effects might occur via a non-GPR119-based mechanism. Understanding the potential for GPR119 agonists in weight control will be an important avenue of future investigation, because this represents a clearly established mechanism by which glucose control can be improved.
In summary, we show here that GPR119 is a highly β-cell-restricted receptor that significantly enhances cAMP accumulation, glucose-dependent insulin release, and in vivo glucose control in normal and diabetic mice. Moreover, activation of GPR119 is very unlikely to induce hypoglycemia. Because the actions of GPR119 are functionally analogous to those of the GIP and GLP-1 receptors, we propose terming GPR119 GDIR, or glucose-dependent insulinotropic receptor. GPR119 agonists may constitute a promising new class of antidiabetic agents.
Source, Graphs and Figures: A Role for
Pancreatic β-cell dysfunction is a hallmark event in the pathogenesis of type 2 diabetes. Injectable peptide agonists of the glucagon-like peptide 1 (GLP-1) receptor have shown significant promise as antidiabetic agents by virtue of their ability to amplify glucose-dependent insulin release and preserve pancreatic β-cell mass. These effects are mediated via stimulation of cAMP through β-cell GLP-1 receptors. We report that the Gαs-coupled receptor GPR119 is largely restricted to insulin-producing β-cells of pancreatic islets. Additionally, we show here that GPR119 functions as a glucose-dependent insulinotropic receptor. Unlike receptors for GLP-1 and other peptides that mediate enhanced glucose-dependent insulin release, GPR119 was suitable for the development of potent, orally active, small-molecule agonists. The GPR119-specific agonist AR231453 significantly increased cAMP accumulation and insulin release in both HIT-T15 cells and rodent islets. In both cases, loss of GPR119 rendered AR231453 inactive. AR231453 also enhanced glucose-dependent insulin release in vivo and improved oral glucose tolerance in wild-type mice but not in GPR119-deficient mice. Diabetic KK/Ay mice were also highly responsive to AR231453. Orally active GPR119 agonists may offer significant promise as novel antihyperglycemic agents acting in a glucose-dependent fashion.
THE GUT HORMONES glucagon-like peptide 1 (GLP-1) and gastric inhibitory polypeptide (GIP) promote normoglycemia acutely by enhancing glucose-stimulated insulin release and chronically by maintaining pancreatic β-cell mass (1, 2, 3, 4). These effects occur via β-cell-expressed, class B G protein-coupled receptors (GPCRs) that in turn mediate elevated intracellular cAMP. Decreased GIP responsiveness may contribute to the pathogenesis of type 2 diabetes, resulting in mixed enthusiasm toward GIP receptor agonists as a means to treat the disease (5, 6). By contrast, strategies that enhance GLP-1 receptor function have shown significant therapeutic promise. These strategies have so far consisted of injectable peptidic GLP-1 receptor agonists or, alternatively, blockade of endogenous GLP-1 metabolism through selective inhibition of dipeptidyl peptidase 4 (1, 7). Orally active, small-molecule GLP-1 receptor agonists have proven elusive, a feature that unfortunately is characteristic of class B GPCRs. Meanwhile, the spectrum of signaling peptides affected by inhibition of dipeptidyl peptidase 4 remains unclear and could potentially extend significantly beyond GLP-1 and GIP (8, 9). It is therefore worthwhile to search for therapeutic approaches that afford both the physiological selectivity of GLP-1 signaling and the opportunity for orally active treatment modalities.
GPR119 was recently identified as a class A, islet-enriched receptor that could potentially mediate the insulinotropic actions of lysophosphatidylcholine observed in vitro (10). However, the islet cell type expressing GPR119 was not identified, and a separate study suggested that oleoylethanolamide was a significantly more potent GPR119 agonist (11). Oleoylethanolamide and the related lipid amide anandamide have been identified as modulators of food intake (11, 12, 13), but there has been no evaluation of these lipid amides with regard to glucose homeostasis. Thus, the importance of GPR119 in glucose homeostasis is uncertain. Moreover, there have been no studies that address whether GPR119 has a sufficiently robust impact on islet biology to indicate its value as a target for therapeutic intervention in type 2 diabetes.
Here we show that GPR119 is a robust mediator of glucose-dependent insulin release and enhanced glucose homeostasis. Importantly, GPR119 is highly amenable to the development of potent, selective, small-molecule agonists. These data strongly suggest that GPR119 may be a therapeutically significant, glucose-dependent, insulinotropic receptor.
Materials and Methods
Cell lines
HEK293, RIN-5F, and HIT-T15 cells were obtained from the American Type Culture Collection (Manassas, VA). RIN-5F is a rat insulinoma cell line that does not express GPR119. GPR119-expressing RIN-5F stable lines were generated by cotransfection of pCDNA3.1 (Invitrogen, Carlsbad, CA) and a human GPR119 expression plasmid encoding amino acids 2—335 of the receptor with a hemagglutinin epitope tag at the N terminus. After selection in G418, clones were chosen for additional studies based on hemagglutinin immunofluorescence and responsiveness to GPR119 agonists.
Expression analysis
For quantitative RT-PCR of mouse GPR119, a fluorescence (6FAM)-labeled probe (5′-CTGCTCAACCCACTCATCTATGCCTATTGG-3′) and amplification primers (forward, 5′-CTCGGCGTGGGCAACTC-3′; reverse, 5′-CTGCAGTCGCACCTCCTTCT-3′) were generated for TaqMan analysis. A β-actin probe (fluorescent VIC-labeled) was used as internal control. Total RNA was prepared with RNA-Bee, DNase treated (DNA-free kit; Ambion, Austin, TX), and converted to cDNA (iScript cDNA synthesis kit; Bio-Rad, Hercules, CA). Relative expression levels of GPR119 were determined by using TaqMan 7900 HT Sequence Detection System (ABI Prism; Applied Biosystems, Foster City, CA) and normalized against β-actin internal control.
For immunofluorescence studies, a polyclonal antibody (AR361) was generated in rabbits using a BSA-conjugated synthetic peptide with sequences corresponding to residues 314—336 (C-terminal tail) of mouse GPR119 (RGPERTRESAYHIVTISHPELDG). Pancreata were obtained from 4% paraformaldehyde/PBS-perfused mice or rats and embedded in optimal cutting temperature (OCT) embedding medium (Tissue Tek, Torrance, CA). Pancreatic sections were immunostained with anti-GPR119 (AR361) in combination with a mouse monoclonal antiinsulin antibody (Santa Cruz Biotechnology, Santa Cruz, CA), or a goat antiglucagon antibody (Santa Cruz Biotechnology). Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Immunoreactive signals were detected by using Alexa Fluor-488 (Molecular Probes, Eugene, OR) or Cy3 (Jackson ImmunoResearch Laboratories, West Grove, PA) conjugated secondary antibodies and fluorescence microscopy.
In situ hybridization
Pancreas from male Sprague Dawley rat was immersed in 4% paraformaldehyde overnight at 4 C, immersed in 20% sucrose/PBS for 2 d at 4 C and then frozen in optimal cutting temperature compound in embedding molds over dry ice. Sections (12 μm thick) were collected on SuperFrost Plus slides (VWR, San Diego, CA; no. 48311-703). The following plasmids were used for probe synthesis. For GPR119, the full-length rat sequence was inserted into pCR4-TOPO (Invitrogen). For insulin, a 201-bp rat insulin cDNA fragment was inserted into pCRII-TOPO (Invitrogen). For preproglucagon, a 544-bp rat preproglucagon cDNA fragment was inserted into pCRII-TOPO. Sense and antisense 33P-radiolabeled GPR119 probes and digoxigenin-labeled preproglucagon probes were generated using standard protocols. Double in situ hybridization was performed with the addition of digoxigenin-labeled probes (500 ng/slide) and salmon sperm (20 μg/slide) together with the 33P-labeled probes. After the post-hybridization step, sections were washed in Tris/NaCl (TN) buffer containing 100 mM Tris (pH 7.5) and 150 mM NaCl for 5 min. Sections were then placed in 0.5% casein/TN blocking solution for 30 min and then incubated for 2 h with alkaline phosphatase (AP)-conjugated antidigoxigenin antibody (Roche, Indianapolis, IN; no. 1093274) in 0.5% casein/TN solution (1:300). Sections were then washed in TN three times and then in 100 mM Tris (pH 9.5), 100 mM NaCl, and 50 mM MgCl2 four times and incubated in color reaction [0.2 mg/ml levamisole, 3.4 μl/ml nitroblue tetrazolium (Roche no. 1383213), 3.5 μl/ml 5-bromo-4-chloro-3-indolyl phosphate (Roche no. 1383221) in 100 mM Tris (pH 9.5), 100 mM NaCl, and 50 mM MgCl2] for 20—30 min. Antibody was stripped off by incubating sections in 0.1 M glycine and 0.5% Triton X-100 for 10 min and washed in water. Sections were fixed in 2.5% glutaraldehyde for 1—2 h, washed with water, and then air dried. Once dried, sections were exposed to x-ray-sensitive film (Bio-Max; Kodak, Rochester, NY; no. IB-856 7232) for 5—7 d and dipped in photographic emulsion (Ilford Scientific K.5D emulsion; Polysciences, Warrington, PA; no. 17537), dried, and stored in a slide box with desiccant at 4 C for 6—8 wk. After development of dipped slides following manufacturer recommendations (Kodak D19), sections were washed extensively in water, air dried, and mounted with coverslips for microscopic examination.
In vitro assays
cAMP measurements were done with a Flash Plate adenylyl cyclase kit (NEN Life Science Products, Boston, MA) according to the supplier's protocol. Briefly, HEK293 cells were transfected with either empty vector DNA or GPR119 expression plasmid DNA (described above) using Lipofectamine (Invitrogen). After 24 h, transfected cells were harvested in GIBCO (Gaithersburg, MD) cell dissociation buffer (catalog item 13151-014) and resuspended in assay buffer (50% 1× PBS/50% stimulation buffer). Compounds were incubated with 105 cells per well for 60 min at room temperature. After another 2-h incubation with tracer in detection buffer, plates were counted in a Wallac (Gaithersburg, MD) MicroBeta scintillation counter. Values of cAMP per well were extrapolated from a standard cAMP curve that was included on each assay plate. cAMP assays in hamster insulinoma-derived HIT-T15 cells and GPR119-transfected RIN-5F stable lines were performed essentially in the same way but without transient transfection. In experiments designed to selectively lower endogenous GPR119 expression in HIT-T15 cells, cells were transfected with either a control RNA oligo with sequence derived from hamster GPR119 in sense orientation (5′-CUAUGCUGCUAUCAAUCUA-3′) (control) or in antisense orientation (5′-UAGAUUGAUAGCAGCAUAG-3′) [small interfering RNA (siRNA)]. Forty-eight hours after transfection, cells were assayed for GPR119 expression and for agonist-stimulated cAMP production as described above.
For assays measuring inositol phosphate accumulation, HEK293 cells in 96-well plates were transfected with the indicated G protein receptor expression plasmid. Additionally, cells were cotransfected with either an empty cytomegalovirus (CMV) vector or a Gαq/Gαi expression plasmid in which the terminal six residues of Gαq were replaced by the corresponding residues of Gαi. The receptor and G protein chimera were transfected at a molar ratio of 4:1, respectively. The following day, cells were incubated in 100 μl inositol-free/serum-free DMEM (GIBCO) containing 0.4 μCi [3H]myoinositol (Perkin-Elmer, Norwalk, CT). Cells were incubated overnight, after which the medium was replaced with 100 μl inositol-free/serum-free DMEM containing 10 μM pargyline (Sigma Chemical Co., St. Louis, MO) and 10 mM lithium chloride (Sigma). One hour later, the cells were lysed and inositol phosphates were isolated using chromatography on AG1-X8 formate resin (Bio-Rad). After additional purification through binding, four washes, and elution from a multiscreen filter plate (Millipore, Bedford, MA), eluted counts were quantified using a Wallac scintillation counter.
For in vitro insulin release, HIT-T15 insulinoma cells were plated in 24-well plates (2.5 × 105 cells per well) for insulin release assays. One day before the assay, culture media were changed to DMEM (3 mM glucose) with 10% dialyzed horse serum and 2.5% fetal bovine serum. On the next day, cells were washed twice with PBS and incubated for 1 h with test compounds in DMEM in the presence of 15 mM glucose in 0.25 ml HEPES-buffered Krebs-Ringer buffer. Supernatant insulin levels were determined using an Ultra Sensitive Insulin ELISA kit (Crystal Chem Inc., Downers Grove, IL). Insulin assays with RIN-5F stable lines were performed in essentially the same manner.
Insulin release assays with pancreatic islets were done using islets isolated from female Sprague Dawley rats (body weight, 175—185 g) or male C57BL/6 mice (body weight, ∼25 g). Insulin release was determined in static islet incubation as described previously (14). Briefly, groups of five islets each were placed in incubation wells. After a 30-min preincubation with HEPES-buffered Krebs-Ringer buffer (pH 7.4) containing 5 mM glucose, islets were transferred to wells containing 2 ml HEPES-buffered Krebs-Ringer buffer and different concentrations of glucose and test compounds. The studies were performed at 37 C in a water bath shaker with an atmosphere of 95% O2/5% CO2. Samples of incubation buffer were collected at 60 min for insulin determination using ELISA methodology.
In vivo experiments
C57BL/6 male mice were obtained from Harlan (Indianapolis, IN). KK/Ay and db/db male mice were purchased from The Jackson Laboratory (Bar Harbor, ME). All in vivo animal protocols were approved by the Animal Welfare Committee of Arena Pharmaceuticals, Inc., and in adherence with government regulations. For the oral glucose tolerance test, overnight fasted mice (n = 6 per treatment) were given either vehicle (80% polyethylene glycol 400/10% Tween 80/10% ethanol) or test compounds at desired doses via oral gavage. A glucose bolus was then delivered (3 g/kg orally or 2 g/kg ip). Plasma glucose levels were determined at desired time points over a 2-h period using blood (∼5 μl) collected from tail nick and a glucose meter. For insulin pharmacodynamic studies, vehicle or AR231453 was administered orally to fasted animals (n = 6 per treatment group and time point). After 30 min, a glucose bolus of 3 g/kg was administered orally. Blood was collected in heparinized blood collection tubes at desired time points. Plasma samples were obtained via centrifugation at 500 × g for 20 min and assayed for insulin as described above.
Statistical analysis
Statistical analysis was performed using t test and one-way ANOVA using either Microsoft Excel or Prism statistical methods.
Results
GPR119 is a β-cell GPCR
Quantitative TaqMan analysis indicated that GPR119 expression is greatly enriched in islets relative to whole pancreas (Fig. 1A⇓), consistent with previous reports (10). Moreover, significant GPR119 expression was present in the islets of hyperglycemic db/db mice (Fig. 1A⇓). To determine the predominant islet cell types expressing GPR119, we first performed in situ hybridization analysis on pancreatic sections. Autoradiography showed that antisense GPR119 hybridized to sections in an islet-like pattern, whereas no such signal was observed with the sense probe (Fig. 1B⇓, panels a and b). Examination of emulsion-dipped sections indicated significant hybridization throughout the core of pancreatic islets (Fig. 1B⇓, panels c and d), strongly suggesting that GPR119 is expressed in the β-cell population. These data also suggested that a subset of glucagon-producing cells might express GPR119, but this was difficult to quantify. To define further the islet cell types that express GPR119, polyclonal anti-GPR119 antisera were raised against a synthetic peptide derived from the C-terminal tail region of the mouse sequence. Immunofluorescence analysis of rat (Fig. 1C⇓) or mouse (see Fig. 7D⇓) pancreatic sections indicated that GPR119 colocalizes primarily with insulin. Moreover, the immunofluorescent signal was absent in pancreatic sections from GPR119-deficient mice (see Fig. 7D⇓). Collectively, these data indicate that GPR119 expression is largely restricted to the β-cells of pancreatic islets. However, it remains possible that there might be additional GPR119 expression in a subset of α-cells or in other islet cell types.
AR231453 is a GPR119-selective small molecule agonist
In transfected HEK293 cells, high-level expression of GPR119 stimulated a significant increase in cAMP levels relative to controls (Fig. 2A⇓). Such constitutive activity is occasionally seen with class A GPCRs (15) and shows that GPR119 couples efficiently to Gαs. In inositol phosphate assays, a constitutively active variant of the Gαq-coupled 5-hydroxytryptamine 2A receptor (16) was highly active, but GPR119 was inactive (Fig. 2B⇓). Thus, GPR119 exhibits poor coupling efficiency to Gαq. GPR119 also failed to produce inositol phosphate when coexpressed with a Gαq/Gαi chimera (Fig. 2B⇓), which has previously been shown to redirect Gαi-coupled receptors to stimulate inositol phosphate accumulation (17). The Gαi-coupled α2A adrenergic receptor functioned very efficiently in this assay (Fig. 2B⇓). Collectively, these data show that GPR119 couples selectively to Gαs.
To evaluate GPR119 function pharmacologically, we screened a library of approximately 55,000 small-molecule compounds. Based on the foregoing, a cAMP assay was used to identify GPR119 modulators. One hit was identified, from which we ultimately developed an agonist termed AR231453 (Fig. 3A⇓), which potently increased cAMP in transfected HEK293 cells (EC50 = 5.7 ± 1.6 nM; Fig. 3B⇓) and stably transfected RIN-5F insulinoma cells (not shown). AR231453 was inactive in cells transfected with empty vector (Fig. 3B⇓) or GLP-1 receptor (Fig. 3C⇓). Indeed, AR231453 was inactive at all other GPCRs tested (more than 230, including all known pancreatic islet receptors), indicating that it is highly selective for GPR119.
To test whether AR231453 stimulates cAMP production via endogenously expressed GPR119 in β-cells, we employed the hamster β-cell line HIT-T15, which expresses GPR119 at high levels (Fig. 3E⇑, inset). AR231453 potently stimulated cAMP accumulation (EC50 = 4.7 nM) with a maximal efficacy similar to that seen with forskolin (Fig. 3D⇑). Pretreatment of HIT-T15 cells treated with a hamster GPR119-selective siRNA markedly lowered receptor levels and correspondingly rendered AR231453 virtually inactive (Fig. 3E⇑).
GPR119 agonist stimulates insulin release in vitro
The observation that GPR119 couples to Gαs and is expressed in β-cells prompted us to test the hypothesis that GPR119 is an insulinotropic receptor. In this regard, AR231453 significantly enhanced insulin release in HIT-T15 cells, with an EC50 of 3.5 nM (Fig. 4A⇓), similar to its effect on cAMP (EC50 of 4.7 nM; see Fig. 3D⇑), and similar in efficacy to forskolin. To assess the GPR119 dependence of the insulinotropic effects seen with AR231453, we tested its activity in RIN-5F cells lacking GPR119 (RIN-5F/vector) and in RIN-5F cells stably transfected with human GPR119 (RIN-5F/hGPR119). AR231453 stimulated insulin release in RIN-5F/hGPR119 cells but not in RIN-5F/vector cells (Fig. 4B⇓). These studies indicate that AR231453 is insulinotropic specifically in GPR119-expressing β-cell lines. We next examined the effect of GPR119 agonist on glucose-stimulated insulin release in isolated rat and mouse islets. At 15 mM glucose, 300 nM AR231453 enhanced insulin release in rat islets similarly to GLP-1 (Fig. 4C⇓). However, this compound had no effect on islets incubated in 5 mM glucose. The insulinotropic effect of GPR119 agonists in islets is therefore glucose dependent. AR231453 also stimulated insulin release in isolated mouse islets at glucose concentrations ranging from 8—17 mM (Fig. 4D⇓). This indicates that the insulinotropic effect of GPR119 agonists requires only modest elevations of blood glucose.
GPR119 agonist improves glucose tolerance and enhances glucose-dependent insulin release in mice
AR231453 exhibited good oral bioavailability in C57BL/6 mice, attaining micromolar plasma levels for more than 2 h at the doses used here. AR231453 (20 mg/kg, orally) markedly improved oral glucose tolerance in a dose-dependent fashion, with efficacy similar to maximally effective doses of the sulfonylurea glyburide (Fig. 5A⇓). On average, AR231453 maximally inhibited the glycemic excursion by approximately 42% (21 independent experiments). When glucose was administered ip, AR231453 improved glucose tolerance somewhat less effectively (23%; three independent experiments) (Fig. 5B⇓).
Next, we examined the effects of AR231453 on glucose homeostasis in rodent models of type 2 diabetes. GPR119 expression was well maintained in β-cells of ZDF diabetic rats (at ages before β-cell loss), fa/fa rats, db/db mice, and KK/Ay mice (Fig. 5C⇑ inset, Fig. 1A⇑, and data not shown). AR231453 significantly improved oral glucose tolerance in 8-wk-old KK/Ay mice (Fig. 5C⇑) and in 6-wk-old db/db mice (Fig. 5D⇑). Similar effects were seen in rat diabetic animal models with other GPR119 agonists (data not shown). Thus, GPR119 agonists improve glycemic control in both normal and diabetic mice.
To test whether the improved glucose tolerance was due to enhanced insulin release, we performed an insulin pharmacodynamic analysis in mice (Fig. 6A⇓). Glucose administered alone increased plasma insulin concentration 2 min and onward and peaked at 20 min. Mice pretreated with AR231453 had markedly elevated insulin levels beginning from 5 min and peaked at 10 min after glucose bolus (Fig. 6A⇓, bottom), which preceded the observed improvement in glucose tolerance (Fig. 6A⇓, top). By contrast, AR231453 was not effective in an insulin tolerance test (not shown). Taken together, these data suggest that AR231453 acts primarily by enhancing glucose-dependent insulin release from pancreatic β-cells, which subsequently results in enhanced glucose disposal. The glucose-dependent nature of AR231453 function was further evaluated by administering very high doses (100 mg/kg) to fasted C57BL/6 mice (Fig. 6B⇓). Even under these conditions, AR231453 did not induce hypoglycemia or trigger insulin release. By contrast, glyburide, a sulfonylurea antidiabetic agent known to stimulate insulin release in a glucose-independent fashion, elicited marked hypoglycemia (Fig. 6B⇓, top) that was associated with enhanced insulin release (Fig. 6B⇓, bottom). Thus, the actions of the GPR119 agonist AR231453 are glucose dependent both in isolated rat islets (Fig. 4⇑, C and D) and in vivo.
AR231453 is inactive in GPR119-deficient mice
To demonstrate that the in vivo effects of AR231453 occur via GPR119, we generated GPR119-deficient mice (Fig. 7⇓). As in human, the gene for mouse GPR119 is located on the X chromosome (Unigene Cluster Mm.34953) (Fig. 7A⇓). Deletion of GPR119 was confirmed by Southern blot analysis (Fig. 7B⇓), RT-PCR analysis of pancreatic GPR119 mRNA (Fig. 7C⇓), and immunofluorescent analysis of GPR119 receptor on pancreatic sections from the knockout mice (Fig. 7D⇓). Islets from GPR119-deficient mice retained normal morphology (Fig. 7D⇓) and a normal response to glucose and GLP-1 (Fig. 7E⇓). These mice also had normal size, body weight, and fed/fasted blood glucose levels (not shown) and a typical response to glyburide in glucose tolerance tests (Fig. 7⇓, F and G). Thus, deletion of GPR119 does not grossly impact glucose homeostasis. However, isolated islets from these animals no longer responded to AR231453 (Fig. 7E⇓). AR231453 also had no effect on glucose tolerance in GPR119-deficient mice (Fig. 7G⇓), indicating that its observed in vivo activities were indeed mediated via GPR119.
Discussion
These data provide strong evidence that GPR119 is a significant modulator of glucose-stimulated insulin release in pancreatic β-cells. The GPR119-selective agonist AR231453 stimulated cAMP accumulation and insulin release in 1) transfected RIN-5F insulinoma cells, 2) HIT-T15 insulinoma cells that express the receptor endogenously, and 3) isolated rat and mouse islets. These effects were substantially the same as those elicited by forskolin or GLP-1, suggesting significant physiological relevance. In vivo, AR231453 stimulated insulin release and improved glucose tolerance with comparable efficacy to glyburide, a very robust stimulator of insulin release in rodents and humans (18). Importantly, the activity of AR231453 was particularly impressive in diabetic KK/Ay mice. All these effects are very likely occurring via GPR119. RIN-5F control cells lacking GPR119 are unresponsive to AR231453. In HIT-T15 cells, siRNA-mediated reduction in GPR119 levels was associated with virtually complete loss of responsiveness to AR231453. Finally, AR231453 does not enhance glucose-stimulated insulin release in islets derived from GPR119-deficient mice, nor does it improve glucose tolerance in GPR119-deficient mice. GPR119-deficient mice had no obvious defect in homeostasis, but this perhaps is not surprising because mice lacking either GIP receptor or GLP-1 receptor have very minor alterations in this regard (19, 20, 21).
Because we did not measure AR231453 exposures in GPR119-deficient mice, its inactivity in glucose tolerance tests performed in these mice could simply be due to strain-dependent differences in plasma exposure of the compound. However, this seems unlikely. In C57BL/6 mice, AR231453 achieves exposures more than 100-fold greater than its in vitro EC50. Additionally, the compound is effective in all other strains of mice used here. Numerous, structurally distinct compounds have been assessed in wild-type littermates and GPR119-deficient mice, with similar outcomes to those seen for AR231453 (not shown). Taken together, these data strongly suggest that the actions of AR231453 are mediated via GPR119 and firmly establish that GPR119 is a physiologically significant mediator of glucose homeostasis.
Although these studies support the hypothesis that GPR119 regulates glucose homeostasis by direct enhancement of pancreatic β-cell function, it is important to recognize that GPR119 may work via additional mechanisms. The effectiveness of AR231453 is reduced by almost 50% when glucose is administered ip, suggesting that its actions may also involve modulation of incretin signaling. This hypothesis warrants further investigation.
Our studies in islets and mice indicate that GPR119 stimulates insulin release in a glucose-dependent manner. Even when administering doses of AR231453 (100 mg/kg in C57BL/6 mice) that greatly exceed the maximal in vivo efficacy of this compound, there is no observed stimulation of insulin release or lowering of glucose under fasted conditions. By contrast, glyburide elicited significant insulin release and marked hypoglycemia in fasted mice. These data are entirely consistent with the well-established mechanism by which GIP and GLP-1 stimulate insulin release (22).
A key distinguishing feature of GPR119 is the demonstrated ability to develop potent, orally active, small-molecule agonists of this receptor. This is in contrast to Gαs-coupled, β-cell peptide receptors for GLP-1, GIP, vasoactive intestinal polypeptide, and pituitary adenylate cyclase-activating polypeptide (23), for which there are no known activators of this type. GPR119 is likely unique in this regard because it is the only class A, nonpeptide GPCR identified that mediates glucose-dependent insulin release via increased intracellular cAMP.
It was recently reported that activators of GPR119 may markedly reduce food intake in rodents (11). Our studies with a number of GPR119 agonists have been inconclusive. In general, a reduction in feeding behavior has been observed only at doses significantly higher than those required to impact glucose homeostasis (unpublished results). This conceivably could be due to comparatively lower agonist exposure in tissues such as the brain, which expresses GPR119 at low levels in the amygdala, thalamus, and perhaps other subregions. Alternatively, these effects might occur via a non-GPR119-based mechanism. Understanding the potential for GPR119 agonists in weight control will be an important avenue of future investigation, because this represents a clearly established mechanism by which glucose control can be improved.
In summary, we show here that GPR119 is a highly β-cell-restricted receptor that significantly enhances cAMP accumulation, glucose-dependent insulin release, and in vivo glucose control in normal and diabetic mice. Moreover, activation of GPR119 is very unlikely to induce hypoglycemia. Because the actions of GPR119 are functionally analogous to those of the GIP and GLP-1 receptors, we propose terming GPR119 GDIR, or glucose-dependent insulinotropic receptor. GPR119 agonists may constitute a promising new class of antidiabetic agents.
Source, Graphs and Figures: A Role for
