β-Cell Inactivation of Gpr119 Unmasks Incretin Dependence of GPR119-Mediated Glucoregulation

The prevalence of type 2 diabetes (T2D) has been rising steadily, with several hundred million individuals worldwide at risk for developing diabetes-associated complications (1). These complications include cardiovascular disease, neuropathy, nephropathy, and retinopathy and result in extraordinary human and economic costs (2). Although considerable progress has been made toward development of new drugs for the management of hyperglycemia, many subjects still experience difficulty in attaining treatment goals with existing therapies (3). It seems evident that improved understanding of the pathophysiology of diabetes, β-cell failure, and insulin resistance is central to the development of new more effective agents for the treatment of T2D.

The importance of the β-cell response for maintaining normoglycemia in the face of chronic nutrient excess and obesity-associated insulin resistance has refocused attention on how nutrient-derived signals augment insulin secretion. Indeed physiological characterization of the incretin effect led to the discovery of the glucoregulatory roles of glucose-dependent insulinotropic polypeptide (GIP), and later, glucagon-like peptide 1 (GLP-1). Both these peptides act directly through their cognate receptors to augment glucose-dependent insulin secretion from islet β-cells, supporting the development of GLP-1 receptor agonists and dipeptidyl peptidase 4 inhibitors for the treatment of T2D (4,5).

The validation of two incretin receptors as feasible drug targets has energized scientific efforts directed at understanding how β-cells sense and respond to nutritional cues to amplify insulin secretion. Although several hundred G-protein–coupled receptors (GPCRs) are expressed in islets and many are coupled to control of insulin secretion, meaningful progress in identification of new antidiabetic drug candidates based on functional characterization of islet GPCR expression and activity has been limited (6,7). Nevertheless, there remains intense interest in understanding the signals that couple the sensing of nutrients after meal ingestion to rapid augmentation of insulin secretion.

Incretin hormones represent prototypical examples of how ingested energy indirectly transmits signals to the endocrine pancreas via a gut–β-cell endocrine axis (8,9). Nevertheless, the digestion of nutrients yields a complex mixture of amino acids, carbohydrates, fatty acids, and metabolites, which in turn function as ligands that directly or indirectly promote insulin secretion (10). For example, GPR40, encoded by FFAR1, is a fatty acid–sensing receptor that increases insulin secretion in rodent and human islets, spurring development of investigational GPR40 agonists for the treatment of human T2D (11). Other nutrients, exemplified by the amino acid leucine, enhance insulin secretion through interaction with intracellular protein sensors and β-cell pathways, coupling nutrient sensing and fuel metabolism to control of ion channel activity, intracellular ion flux, and stimulation of exocytosis.

The complexity of nutrient sensing and control of the gut-islet axis is highlighted by demonstration that nutrient and chemosensing receptors may be expressed on both enteroendocrine cells (EECs) and β-cells, enabling overlapping mechanisms for nutrient regulation of insulin secretion. For example, GPR40 and the G protein–coupled bile acid receptor 1 (also known as TGR5) agonists promote GLP-1 secretion from EECs (12), and both GPR40 agonists and bile acids also directly enhance insulin secretion from rodent and human islets (13). Bile acids also signal through the farnesoid X receptor (FXR) to control GLP-1 secretion in EECs, and FXR agonists directly augment glucose-stimulated insulin secretion (GSIS) in isolated islets (14). Similarly, cannabinoid receptors control the secretion of gut hormones (8) and some cannabinoid receptor ligands directly augment insulin secretion from β-cells, depending on the species studied and receptor specificity (15).

Among the most extensively studied nutrient response receptors is GPR119, a G_αs-linked class A GPCR that recognizes oleoylethanolamide, lysophosphatidylcholine, and structurally related lipids. Activation of GPR119 leads to stimulation of cAMP accumulation and increased insulin secretion directly from islet β-cells (16,17). GPR119 agonists also indirectly augment glucose-dependent β-cell function through stimulation of incretin secretion from EECs (16,18–20). These functional properties of GPR119, coupled with its suitability as a target for development of orally available small molecule agonists, led to the development and clinical testing of multiple GPR119 agonists for the treatment of T2D. Unexpectedly, the efficacy of these drugs in human subjects with diabetes was substantially less than predicted from the results of preclinical studies (21). Here we reexamine the importance of GPR119 for β-cell function in mice, demonstrating that GPR119 activity in β-cells is dispensable for the physiological control of glucose homeostasis and the acute pharmacological glucoregulatory response to AR231453, a potent and selective GPR119 agonist. These findings prompt reevaluation of the functional importance of GPR119 within β-cells, with implications for understanding the pharmaceutical potential of the GPR119 axis for the treatment of metabolic disorders.

Animal experiments were performed according to protocols approved by the Mount Sinai Hospital and The Centre for Phenogenomics (TCP). Male C57BL/6 mice were purchased from the in-house colony at TCP. Gpr119^−/− mice were obtained from Arena Pharmaceuticals (San Diego, CA) (22) and maintained on a C57BL/6 background. Construction of targeting vector and generation of a floxed Gpr119 mouse was performed by InGenious Targeting Laboratory (Ronkonkoma, NY) using homologous recombination of mouse embryonic stem cells, and subsequent injection into blastocysts. The resulting mice were rederived at TCP on a mixed 129/B6 background. The mice were bred with FLP mice to remove the FRT-flanked Neo selection cassette and then backcrossed a minimum of seven times onto a C57BL/6J background. The floxed Gpr119 mice were subsequently crossed with Mip-CreERT mice (23) to generate mice with β-cell–specific inactivation of Gpr119 (Gpr119^{βcell−/−}). All mice were housed under specific pathogen-free conditions in ventilated microisolator cages and maintained on a 12-h light/dark cycle with free access to food and water unless otherwise noted. For a list of genotyping primers used, see Supplementary Table 1.

For analysis of β-cell apoptosis, 8–10-week-old age-matched Gpr119^−/− and littermate controls were fasted for 4 h before streptozotocin (STZ; 50 mg/kg/day) or vehicle (0.1 mol/L sodium citrate, pH 5) injection once daily for five consecutive days as previously described (24,25). Histology sections from formalin-fixed, paraffin-embedded whole pancreas were generated and analyzed as previously described (26–28). Serial sections were immunostained for insulin and either cleaved caspase-3 or TUNEL to identify apoptotic β-cells, or glucagon to identify α-cells. Slides were scanned and analyzed using the ScanScope CS system and software (Aperio Technologies, Vista, CA) and α-cell/β-cell area quantified as previously described (27,28).

Male mice were randomized to receive a regular chow (RC) diet (18% kcal from fat, Teklad Global Rodent Diet; Harlan Laboratories) or high-fat diet (HFD) (45% kcal from fat; Research Diets, Inc.) for a minimum of 12 weeks prior to experimentation.

After a 5-h daytime fast, glucose tolerance tests were performed as previously described (28,29). When specified, AR231453 (Ab141627; Abcam, Toronto, ON, Canada) or vehicle (80% PEG-400, 10% Tween-80, 10% ethanol) was provided by oral gavage 30 min prior to testing. Blood samples (50–100 µL) were collected and mixed with 5–10 µL of a chilled solution containing 5,000 kIU/mL Trasylol (Bayer, Toronto, ON, Canada), 32 mmol/L EDTA, and 0.01 mmol/L diprotin A (Sigma-Aldrich, St. Louis, MO), and plasma collected for subsequent assessment of total GLP-1 (mouse/rat total GLP-1 assay kit; Mesoscale Discovery, Gaithersburg, MD), total or active GIP (rat/mouse GIP [total] ELISA kit from Millipore, Billerica, MA; or mouse GIP ELISA kit from CrystalChem, Downer’s Grove, IL), insulin (Ultrasensitive Mouse Insulin ELISA; Alpco Diagnostics, Salem, NH), and glucagon (Milliplex endocrine assay from Millipore; or Glucagon ELISA from Mercodia, Uppsala, Sweden). Insulin tolerance tests were similarly performed in mice fasted for 5 h using a single intraperitoneal dose of insulin (lean mice, 1.2 units/kg; obese mice, 1.7 units/kg) (Humulin R; Eli Lilly and Company, Scarborough, ON).

For assessment of fat and lean mass body composition, a whole-body magnetic resonance analyzer was used (EchoMRI-100; Echo Medical Systems). Measures of O₂ consumption, CO₂ production, respiratory quotient, and physical activity were obtained using an Oxymax system (Columbus Instruments) (28).

First-strand complementary DNA was synthesized according to the manufacturer’s protocol (RNeasy Kit from Qiagen, Mississauga, ON, Canada; or Tri-Reagent from Molecular Research Center, Inc., Cincinnati, OH) using the SuperScript III reverse transcriptase synthesis system (Invitrogen, Carlsbad, CA) and random hexamers. Quantitative PCR was performed with the ABI Prism 7900 Sequence Detection System using the TaqMan Gene Expression Assays and TaqMan Universal PCR Master Mix (Applied Biosystems, Foster City, CA).

Pancreatic fragments, isolated from the same anatomical portion of each mouse pancreas, were homogenized in ice-cold acid-ethanol solution (0.18 mol/L HCl, 70% ethanol). Insulin and glucagon levels were determined in pancreatic extracts using the rat insulin RIA kit (Millipore, St. Charles, MO) and the mouse Glucagon ELISA (Mercodia).

Isolated islets were allowed to recover overnight at 37°C in RPMI containing 10% FBS. Medium-sized islets were then handpicked into a 0.275-mL chamber containing Krebs-Ringer bicarbonate HEPES buffer (KRBH) solution containing 0.1% (weight for volume) BSA and perifused as previously described (28). Perifusion flow rate was 200 μL/min for insulin measurements and 100 μL/min for glucagon measurements. Insulin concentrations are expressed as percent of total islet insulin for each sample and measured using a Rat Insulin RIA (Millipore). Glucagon concentrations are expressed as fold-change from baseline and measured by Glucagon ELISA.

After overnight recovery, groups of five islets were incubated in low-glucose KRBH medium, and then high-glucose KRBH medium for 1 h each, with or without AR231453, in triplicate. The supernatant from each 1-h incubation (2 mL) was collected for insulin measurement by RIA. Insulin values were normalized to total islet insulin levels from each group of islets.

All results are expressed as mean ± SD. Statistical significance was assessed by ANOVA and, where appropriate, two-tailed, unpaired Student t test unless otherwise noted using GraphPad Prism (version 5; GraphPad Software). A P value <0.05 was considered to be statistically significant.

To determine the importance of GPR119 for β-cell function, we assessed the metabolic phenotype(s) of whole-body Gpr119^−/− mice (22) under basal conditions (18% fat RC diet). No differences in body weight or fat or lean mass were observed in Gpr119^−/− versus littermate control Gpr119^+/+ mice (Fig. 1A and B). Moreover, oral glucose tolerance (Fig. 1C) and plasma insulin and incretin levels were similar in Gpr119^−/− versus Gpr119^+/+ mice (Fig. 1D–F). Conversely, intraperitoneal glucose tolerance was slightly improved but not significantly different in Gpr119^−/− mice (Fig. 1G), associated with increased circulating insulin levels (Fig. 1H). Insulin tolerance and the glucagon response to insulin-induced hypoglycemia was not different in Gpr119^−/− versus Gpr119^+/+ mice (Fig. 1I and J). To determine the susceptibility of Gpr119^−/− β-cells to apoptotic injury, we analyzed islets from mice after STZ administration. More cleaved caspase-3–immunopositive cells were detected in islets from Gpr119^−/− mice (Supplementary Fig. 1A and B); however, pancreatic insulin area and β-cell size were not different (Supplementary Fig. 1C and D). Collectively, these findings are consistent with previous reports demonstrating no major metabolic perturbations in RC-fed Gpr119^−/− mice (16,30).

We next examined whether germline loss of Gpr119 disrupts the adaptive metabolic response to HFD feeding. Although weight gain was comparable in Gpr119^−/− versus Gpr119^+/+ mice (Fig. 2A), fat mass (Fig. 2B) and circulating levels of leptin (Fig. 2C) and resistin (Fig. 2D) were reduced in Gpr119^−/− mice. Histological analysis of the pancreas revealed reduced β-cell (but not α-cell) area in HFD-fed mice (Fig. 2E). Food intake (Supplementary Fig. 2A) and the nocturnal respiratory exchange ratio (Supplementary Fig. 2B) were increased in HFD-fed Gpr119^−/− mice, without changes in oxygen consumption (Supplementary Fig. 2C) or locomotor activity (Supplementary Fig. 2D).

Both oral and intraperitoneal glucose tolerance were impaired in HFD-fed mice, to a lesser extent in Gpr119^−/− mice (Fig. 2F and K), associated with lower levels of glucose-stimulated insulin (Fig. 2G and L). Glucagon levels were not different after oral or intraperitoneal glucose administration (Fig. 2H and M). Circulating levels of total GLP-1 (Fig. 2I) were not different; however, total GIP levels (Fig. 2J) were lower in HFD-fed Gpr119^−/− mice. Moreover, basal rates of gastric emptying and motility were reduced, and fecal triglycerides trended lower in Gpr119^−/− mice (Supplementary Fig. 3A–D).

Insulin sensitivity was not different in RC-fed Gpr119^−/− mice but significantly improved in HFD-fed Gpr119^−/− mice (Fig. 2N). Consistent with findings of improved insulin sensitivity, fasting insulin levels were reduced in HFD-fed Gpr119^−/− mice (Fig. 2O); however, plasma glucagon levels were similar after insulin administration (Fig. 2P). Moreover, pancreas weight was lower and pancreatic insulin content was reduced in HFD-fed Gpr119^−/− mice (Supplementary Fig. 4A and B). Furthermore, pancreatic levels of Irs2, Gcgr, Glp1r, Gpr40, Gck, and Sp1 mRNA transcripts were lower in Gpr119^−/− mice (Supplementary Fig. 4C).

Mechanistic attribution of the modest differences in metabolic and islet phenotypes arising in whole-body Gpr119^−/− mice is challenging due to pleiotropic roles for GPR119 in control of β-cell function, gut hormone secretion, gastrointestinal motility, food intake, and lipid absorption (16,19,22,27,31,32). As GPR119 was originally described as an orphan β-cell receptor (33) coupled to direct stimulation of insulin secretion (16,18), we mated floxed Gpr119 mice with Mip-CreERT mice (23,34,35) to generate Gpr119^{βcell−/−} mice, enabling selective conditional tamoxifen-mediated reduction of Gpr119 expression in adult β-cells (Fig. 3A). Mip-CreERT littermates were used as controls (denoted as Gpr119^βcell+/+) for all studies analyzing Gpr119^{βcell−/−} mice (35). Levels of Gpr119 mRNA transcripts in Gpr119^{βcell−/−} islets and total pancreas were markedly reduced, whereas no decrease in Gpr119 mRNA expression was detected in mRNA from colon in Gpr119^{βcell−/−} mice (Fig. 3B).

No difference in body weight or composition was observed in Gpr119^{βcell−/−} mice (Fig. 3C and D). Furthermore, glucose tolerance, levels of plasma insulin, GLP-1, and GIP, insulin sensitivity, and the glucagon response to hypoglycemia were not different in Gpr119^{βcell−/−} versus Gpr119^βcell+/+ mice (Fig. 3E–L). Unlike findings in mice with germline disruption of Gpr119 (Supplementary Fig. 1), Gpr119^{βcell−/−} mice did not exhibit enhanced islet apoptosis after STZ administration (Supplementary Fig. 5A and B). Moreover, in contrast to the dysregulated expression of islet genes observed in Gpr119^−/− mice (Supplementary Fig. 4C), levels of Glp1r mRNA were increased but mRNA transcripts for Ins2, Gcg, Gcgr, Irs2, Gipr, and Gpr40 were not different in islet RNA from Gpr119^{βcell−/−} versus Gpr119^βcell+/+ mice (Supplementary Fig. 5C).

To determine whether loss of β-cell GPR119 impairs the adaptation to metabolic stress, we assessed Gpr119^{βcell−/−} versus Gpr119^βcell+/+ mice after HFD feeding. Body weight (Fig. 4A) and VO₂, VCO₂, respiratory exchange ratio, heat production, and activity were similar (Supplementary Fig. 6A–F); however, fat mass trended lower and lean mass was increased in HFD-fed Gpr119^{βcell−/−} mice (Fig. 4B). β-Cell and α-cell area were similar (Fig. 4C) and both oral and intraperitoneal glucose tolerance and plasma levels of insulin and glucagon were comparable in HFD-fed Gpr119^{βcell−/−} versus Gpr119^βcell+/+ mice (Fig. 4D–I). Insulin sensitivity was improved (Fig. 4J), with a trend toward higher glucagon levels (Fig. 4K), likely reflecting the lower relative glucose levels achieved after insulin administration in HFD-fed Gpr119^{βcell−/−} mice.

The surprising preservation of glucose tolerance in HFD-fed Gpr119^{βcell−/−} mice prompted us to further scrutinize insulin secretion in islets lacking GPR119. We first tested insulin secretory responses in islets isolated from lean 12-week-old Gpr119^−/− and Gpr119^+/+ mice. No genotype-dependent differences in GSIS were detected (Fig. 5A). Furthermore, both Gpr119^−/− and Gpr119^+/+ islets responded similarly to the GLP-1R agonist exendin-4 (1 nmol/L) and to 30 mmol/L KCl (Fig. 5A). To eliminate potential confounding effects arising in whole-body Gpr119^−/− mice, we assessed insulin secretion in perifused islets from Gpr119^{βcell−/−} mice. Insulin responses to glucose, exendin-4, and 30 mmol/L KCl were similar in islets from Gpr119^{βcell−/−} versus Gpr119^βcell+/+ mice (Fig. 5B). Moreover, to verify persistent knockdown of Gpr119 in islets ex vivo, we reassessed Gpr119 expression. Gpr119 mRNA transcripts were undetectable in RNA from Gpr119^−/− islets and markedly reduced in Gpr119^{βcell−/−} versus Gpr119^βcell+/+ islet mRNA, whereas Glp1r expression was unaffected (Fig. 5C and D).

As some but not all studies have demonstrated that GPR119 agonists directly stimulate insulin secretion from rodent islets (16,27,36), we assessed insulin release from perifused mouse islets in the presence or absence of AR231453, a highly selective GPR119 agonist that fails to lower glucose or stimulate insulin levels in Gpr119^−/− mice (16). No AR231453-dependent enhancement of insulin secretion was observed in wild-type (WT) islets (Fig. 6A), irrespective of the timing and dose of AR231453 exposure during the perifusion period (Supplementary Fig. 7A and B). Moreover, insulin release was not consistently increased after AR231453 treatment of static cultures of mouse islets (Fig. 6B). Furthermore, in separate experiments, AR231453 did not modify glucagon secretion from perifused islets; in contrast, glucagon levels rose robustly after exposure to arginine (Fig. 6C).

As AR231453 did not directly regulate insulin or glucagon secretion in islets, we reassessed its bioactivity in vivo. Administration of an oral dose (20 mg/kg) of the same batch of AR231453 used in islet studies (Fig. 6D) produced a rapid reduction in glycemic excursion and increase in plasma insulin levels after an oral glucose challenge in WT mice (Fig. 6E and F). Furthermore, AR231453 robustly lowered glycemic excursions (Fig. 7A) and increased plasma levels of insulin and GLP-1 (Fig. 7B and C), but not GIP (Fig. 7D), in Gpr119^{βcell−/−} as well as in Gpr119^βcell+/+ mice (data not shown). Taken together, these findings demonstrate that β-cell GPR119 is not required for the physiological control of glucose homeostasis or the acute insulinotropic response to pharmacological GPR119 agonism.

The expression of GPR119 in β-cells and EECs, together with the robust glucoregulatory actions of GPR119 agonists in preclinical studies, raised considerable enthusiasm for GPR119 as a target for T2D drug development. To better understand the physiology of GPR119, we first reevaluated the metabolic phenotype of mice with whole-body germline disruption of Gpr119. Consistent with previous reports (22,30), islet morphology and glucose homeostasis were not dysregulated in RC-fed Gpr119^−/− mice (16,30). Lan et al. (30) demonstrated that plasma insulin levels and lean body mass trended lower in Gpr119^−/− mice, consistent with our current findings. Moreover, we observed improved insulin and glucose tolerance with lower insulin levels in HFD-fed Gpr119^−/− mice. Our findings of increased lean body mass and improved insulin sensitivity (and hence reduced β-cell secretory demand) in Gpr119^−/− mice complicates ascertainment of the importance of GPR119 activity in β-cells (31).

The essential role(s) of GPR119 in control of the incretin axis further challenges interpretation of β-cell phenotypes arising in Gpr119^−/− mice. Several studies have demonstrated that the glucoregulatory actions arising from GPR119 activation are dependent in part on enhanced GIP and GLP-1 secretion (19,22,27). Moreover, whole-body loss of GPR119 leads to lower circulating levels of GLP-1 in both RC- and HFD-fed Gpr119^−/− mice, and levels of GIP also trended lower (30). Consistent with these findings, plasma levels of GLP-1 and GIP were lower after acute olive oil challenge in Gpr119^−/− mice (32), and plasma GIP levels were lower in HFD-fed Gpr119^−/− mice studied herein. Moreover, analysis of islet mRNA from HFD-fed Gpr119^−/− mice revealed reduced levels of Glp1r, Gcgr, and Gpr40 mRNA transcripts, and Gipr mRNA levels trended lower (Supplementary Fig. 4). Hence, interpreting changes in β-cell function arising in whole-body Gpr119^−/− mice is complex due to potential compensation arising from changes in plasma incretin hormones and altered islet expression of functionally related receptors.

To circumvent the potential confounding importance of GPR119 for organ formation and cell differentiation that might produce adaptive compensation in metabolically sensitive tissues, and to avoid perturbations within the gut GPR119 EEC system, we generated Gpr119^{βcell−/−} mice. We detected ∼80–90% knockdown of Gpr119 expression in islets from Gpr119^{βcell−/−} mice (Fig. 3A and Fig. 5C). This level of knockdown in total islets using the Mip-CreERT mouse to achieve recombination selectively in β-cells is consistent with substantial expression of islet Gpr119 in β-cells, and roughly correlates with the relative proportion of β-cells within mouse islets. The 10–20% residual Gpr119 mRNA expression in Gpr119^{βcell−/−} islets could be attributed to inefficiency of knockdown in β-cells, or more likely to residual Gpr119 expression in the non–β-cell islet population.

Surprisingly, Gpr119^{βcell−/−} mice did not exhibit overt metabolic defects when studied on RC or after prolonged HFD feeding. A modest reduction in fat mass, increased lean body mass, and improved insulin sensitivity in Gpr119^{βcell−/−} mice resembles phenotypes we described for Gipr^{βcell−/−} mice (28), and likely reflects a role for β-cell GPR119 as a component of the β-cell response to lipid ingestion, as recently highlighted in studies of EECs (32). Furthermore, the β-cell insulin secretory response to glucose was completely normal in Gpr119^−/− and Gpr119^{βcell−/−} islets. Equally surprisingly, we were unable to directly elicit insulin secretion using AR231453 in WT islets, at doses as high as 300 nmol/L (16), under perifused or static incubation conditions. Similarly, despite rare reports of GPR119 agonists regulating glucagon secretion in mice (36), we did not detect alteration of glucagon secretion using AR231453 in perifused islets. In contrast, the same batch of AR231453 robustly lowered glucose and increased GLP-1 and insulin levels after acute administration to Gpr119^{βcell−/−} mice. These findings, together with our previous studies elucidating the essential requirements for both incretin receptors in GPR119-dependent glucose control (27), strongly suggest that the β-cell GPR119 is not an essential target for the glucoregulatory activity of GPR119 agonists.

Our data reveal that GPR119 expression within murine β-cells is not functionally important for the β-cell response to hyperglycemia, or for transduction of a direct insulinotropic response to pharmacological GPR119 agonism (36). Despite an extensive literature demonstrating robust GPR119-dependent incretin and insulin secretion, associated with reduction of glycemia in preclinical studies, successful translation of pharmacological GPR119 agonism for the treatment of T2D in human subjects has not been realized (21,37). GPR119 agonists increase incretin secretion from human EECs ex vivo (12); however, these observations have not been as robustly recapitulated in clinical studies of human subjects. Although multiple GPR119 agonists have been tested in clinical trials, reported increases in gut hormones have been modest, and usually not associated with robust changes in insulin levels or glycemic excursion (21,37,38).

Another potential explanation for the failure of GPR119 agonists to effectively lower glycemia and HbA_1c in clinical studies may be related to species differences in islet cell expression of GPR119. Notably, single-cell transcriptomic analysis of individual human islet cells revealed a 10-fold higher level of GPR119 expression in isolated α-cells, relative to β- or δ-cells (39). Nevertheless, analysis of individual islet cell transcriptomes consistently detects GPR119 mRNA transcripts in both mouse and human β-cells, at levels approximating those obtained for the functionally related fatty acid receptor FFAR1 (40,41). Our findings, using isolated islets and Gpr119^{βcell−/−} mice, strongly suggest that β-cell GPR119 is dispensable for 1) the insulin response to glucose, 2) the adaptive islet response to HFD feeding, and 3) the acute glucoregulatory response to pharmacological GPR119 agonism. Collectively, these results support the evolving view that GPR119 functions primarily as an enteroendocrine lipid sensor (32,42) and may be much less important as a direct β-cell target for the stimulation of insulin secretion.

Funding. B.L.P. was supported by fellowships from TD Bank, the Canadian Diabetes Association (grant PF-3-15-4737-BP), and the National Institutes of Health (fellowship grant F32DK107050). This work was supported in part by grants from the Canadian Diabetes Association (OG-3-11-3270-DD), Canadian Institutes of Health Research (CIHR) (Institute of Nutrition, Metabolism and Diabetes grants 82700 and 123391), the Canada Research Chairs Program, and the Banting & Best Diabetes Centre-Novo Nordisk Chair in Incretin Biology (D.J.D.). J.E.C. was supported by a CIHR fellowship, and J.L.B. was supported by a Canadian Diabetes Association fellowship.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. B.L.P. and G.B.F. designed and performed the experiments and wrote the manuscript. J.E.C., J.L.B., and X.C. performed the experiments. D.J.D. designed the experiments and wrote the manuscript. D.J.D. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.