The Effect of Glucagon-Like Peptide 1 Receptor Blockade on Glucagon-Induced Stimulation of Insulin Secretion

Type 2 diabetes is characterized by fasting and postprandial hyperglycemia due in part to bihormonal islet dysfunction. The effects of decreased and delayed insulin secretion are exacerbated by impaired glucagon suppression in response to food ingestion (1). This results in impaired suppression of endogenous glucose production (EGP) contributing to hyperglycemia.

However, glucagon is an important insulin secretagogue (2), at least at the concentrations found within the islet (3). α-Cell dysfunction as occurs in type 2 diabetes (1), in prediabetes (4), or in response to free fatty acid elevation (4) results in increased glucagon concentrations. This raises the possibility that abnormal α-cell function represents an effort to stimulate (failing) β-cell function in these situations.

Glucagon can signal through the glucagon-like peptide 1 receptor (GLP-1R) present on β-cells (5), but the net contribution to insulin secretion versus that through its cognate receptor is unknown. This signaling occurs despite the lower affinity of GLP-1R for glucagon (6) since intraislet concentrations of glucagon are sufficiently high to stimulate GLP-1R (3).

As part of a series of experiments examining α- to β-cell communication, Capozzi et al. (7) report that a β-cell–specific deletion of the glucagon receptor (Gcgr) does not alter glucagon-induced insulin secretion compared with islets from littermates. This suggests that Gcgr is unnecessary for glucagon-induced insulin secretion. In contrast, β-cell–specific deletion of the GLP-1R (Glp1r) significantly impaired glucagon-induced insulin secretion, with some additional effect of a Gcgr antagonist. Taken together, these data suggest that, in rodents, both glucagon and GLP-1R signaling mediate β-cell insulin secretion in response to glucagon, but signaling through the GLP-1R is more important. Similar results were observed in intact animals (8), although the glucagon dose administered (1 mg/kg) is significantly higher than that administered to humans as a test of β-cell secretion (2).

We therefore designed experiments to test the hypothesis that glucagon-stimulated insulin secretion is decreased by GLP-1R blockade with exendin-9,39 (a competitive antagonist). To do so, we studied otherwise healthy subjects on two occasions in the presence and absence of exendin-9,39. In one experiment, subjects were studied in the presence of hyperglycemia and a glucagon infusion that started at 0.4 ng/kg/min and increased hourly to a maximum rate of 1.2 ng/kg/min. In another experiment, the response to a 1-mg glucagon bolus was studied in the presence and absence of exendin-9,39. In these conditions, GLP-1R blockade had a small, but significant, (negative) effect on insulin secretion.

After approval from the Mayo Clinic Institutional Review Board, we recruited subjects for two separate studies through intramural advertising. Eligible subjects (25–65 years old, BMI <30 kg/m²) had no history of chronic illness or upper gastrointestinal surgery and were not taking medications that affect glucose metabolism. Those interested in participating were invited to the Clinical Research and Trials Unit (CRTU) for a screening visit. After written, informed consent was obtained, participants underwent a 2-h 75-g oral glucose tolerance test to characterize their glucose tolerance status (9). All subjects followed a weight-maintenance diet containing 55% carbohydrate, 30% fat, and 15% protein for at least 3 days prior to each inpatient study. Body composition was measured using DEXA (Lunar, Madison, WI).

Two study days, the saline day and the exendin day, were performed 2 weeks apart in random order. Participants were admitted to the CRTU at 1700 on the day before the study. After consuming a standard 10 kcal/kg caffeine-free meal, they fasted overnight. At 0530 the following morning, a dorsal hand vein was cannulated and placed in a heated Plexiglas box (55°C) for sampling of arterialized venous blood. The contralateral forearm vein was cannulated to allow tracer, glucose, and hormone infusions. At 0600 (−180 min), a primed (10 μCi), continuous (0.1 μCi/min) infusion of [3-³H] glucose commenced and continued until 0900 (0 min). At that time, the infusion decreased to mimic the anticipated fall of EGP to minimize changes in specific activity (10). In addition, a glucose infusion, also labeled with [3-³H] glucose, commenced, and the infusion rate varied to produce peripheral glucose concentrations of 160 mg/dL. A glucagon infusion at 0.4 ng/kg/min was also started at 0900 (0 min), increasing by 0.2 ng/kg/min every hour for a total of 5 h. On the saline day, saline was infused starting at 0900 and continued until the end of the experiment. On the exendin day, exendin-9,39 was infused at a rate of 300 pmol/kg/min at 0900, as before (11), for 5 h. The experiment ended at 1400 (300 min), when all infusions were stopped; participants consumed a late lunch and left the CRTU.

This was conducted as part of a separate study with 2 study days, a saline and an exendin day (300 pmol/kg/min), in random order. Participants completed a 3-h hyperglycemic clamp like that described above. At 1200 (180 min), a 1-mg glucagon bolus was administered and blood sampled frequently until 1230 (210 min), when all infusions stopped; participants consumed a late lunch and left the CRTU.

All blood was immediately placed on ice after collection, centrifuged at 4°C, separated, and stored at −80°C until assay. Plasma glucose concentrations were measured using a Yellow Springs glucose analyzer. Plasma insulin concentrations were measured using a chemiluminescence assay (Access Assay; Beckman Coulter, Chaska, MN). Plasma C-peptide was measured using a two-site immunoenzymatic sandwich assay (Roche Diagnostics, Indianapolis, IN). Glucagon was measured using a two-site ELISA (Mercodia, Winston-Salem, NC). Plasma [3-³H] glucose-specific activity was measured by liquid scintillation counting as before (12).

The oral minimal model was applied to the glucose, insulin, and C-peptide concentrations observed in response to the screening oral glucose tolerance test and used to estimate insulin sensitivity (13) and β-cell responsivity (14). Insulin secretion rates during the experiments were calculated by nonparametric deconvolution from peripheral C-peptide concentrations, incorporating age-associated changes in C-peptide kinetics (15).

Glucose appearance and disappearance were calculated using the steady-state equations of Steele et al. (16), in which the actual tracer infusion rate was used. The volume of distribution of glucose was assumed to be 200 mL/kg with a pool correction factor equal to 0.65. EGP was calculated by subtracting the glucose infusion rate from the tracer-determined rate of glucose appearance. All rates of infusion and turnover were expressed per kilogram of lean body mass.

All continuous data are summarized as means ± SEM. Area under the curve was calculated using the trapezoidal rule. A paired, two-way Student t test (parametric) or a Wilcoxon matched-pairs signed rank test (nonparametric) was used to examine differences between study days. Assuming the previously observed variation in C-peptide response to glucagon (2), 11 subjects provided 80% power to detect a 22% change. A P value <0.05 was considered statistically significant.

The deidentified data sets generated during the current study are available from the corresponding author on reasonable request.

A total of 22 subjects (11 in each experiment) were studied. All had normal fasting glucose and normal glucose tolerance. Their anthropometric characteristics are summarized in Supplementary Table 1.

Fasting glucose concentrations did not differ between the study days. Subsequently, at time 0, glucose infusion raised blood glucose concentrations throughout the study and did not differ between study days (8.8 ± 0.1 vs. 8.7 ± 0.1 mmol/L, saline vs. exendin, respectively; P = 0.37) (Fig. 1A).

Fasting glucagon concentrations did not differ between study days. In response to each hourly increment in glucagon infusion, glucagon concentrations rose hourly but did not differ between study days (Fig. 1B).

In response to hyperglycemia and glucagon infusion, insulin concentrations rose from fasting values. There was a tendency for lower insulin concentrations in the presence of exendin-9,39 infusion, but this trend did not reach statistical significance (Fig. 1C).

Like insulin, fasting C-peptide concentrations did not differ between study days (Fig. 1D). Exendin-9,39 decreased integrated C-peptide concentrations during the experiment (688 ± 67 vs. 622 ± 55 nmol/5 h; P = 0.02).

Fasting insulin secretion rate did not differ between study days (Fig. 1E). Exendin-9,39 infusion impaired the response to hyperglycemia and glucagon infusion, as evidenced by the integrated insulin secretion rate during the experiment (191 ± 23 vs. 171 ± 19 nmol/5 h; P = 0.02) (Fig. 1F).

The rates of glucose necessary to maintain hyperglycemia did not differ between study days (Fig. 2A). The specific activity (Fig. 2B) was stable for each step of the glucagon infusion, enabling reliable calculation of glucose turnover.

Fasting EGP did not differ between study days (Fig. 2C). Despite the presence of hyperglucagonemia, hyperglycemia and insulin secretion were sufficient to suppress EGP to a nadir rate that was maintained throughout the study. EGP did not differ between the saline and exendin days.

Rates of glucose disappearance (Fig. 2D) did not differ during fasting and in response to hyperglycemia and insulin secretion on either study day.

By design, glucose infusion rates before and after glucagon bolus were adjusted to ensure no difference in glucose concentrations between study days (Fig. 3A).

Glucagon stimulated insulin secretion, but both peak (2.4 ± 0.4 vs. 1.9 ± 0.3 nmol/L) and integrated (43 ± 7 vs. 29 ± 5 nmol/L/30 min) concentrations were decreased (P < 0.01) in the presence of exendin-9,39 (Fig. 3B). Similarly, peak (8.1 ± 0.7 vs. 7.1 ± 0.7 nmol/L) and integrated (182 ± 17 vs. 154 ± 13 nmol/L/30 min) C-peptide concentrations were also decreased (P < 0.01) on the exendin study day (Fig. 3C).

In healthy subjects without diabetes, hyperglycemia together with glucagon infusion stimulated insulin secretion. Blockade of the GLP-1R with exendin-9,39 produced a small but significant decrease in insulin secretion. Exendin-9,39 infused at 300 pmol/kg/min has previously been shown to inhibit GLP-1 signaling at GLP-1 concentrations ∼60 pmol/L (11). Given that the lower affinity of glucagon for the GLP-1R (6), the experimental conditions were considered sufficient to ensure blockade of glucagon-induced signaling via the GLP-1R. Of note, in these insulin-sensitive subjects with presumably intact glucose effectiveness (17), hyperglycemia and hyperinsulinemia were sufficient to maintain suppression of EGP and stimulation of glucose disappearance despite rising glucagon concentrations and a small decrease in insulin secretion.

These data suggest that although glucagon can stimulate the β-cell through the GLP-1R, the net contribution to insulin secretion via this mechanism is small, at least under the conditions extant during the current experiment, in which hyperglycemia suppressed EGP and (presumably) endogenous glucagon secretion. Using a graded glucagon infusion resulted in an approximately threefold increase in circulating glucagon concentrations. Over the final 4 h of the study (when glucose was constant), the symmetric percent change in insulin secretion rate at the end of the study from the basal rate was 122 ± 10%. This increment was significantly (P < 0.05) attenuated by exendin-9,39 (99 ± 9%), implying that glucagon-induced stimulation of the β-cell through the GLP-1R is significant at physiologic circulating concentrations of glucagon.

Although these results support the notion that glucagon stimulates insulin secretion, one limitation of the study is that we did not measure insulin secretion in response to hyperglycemia alone. However, a similar experiment using a hyperglycemic clamp for 2 h in healthy subjects increased the insulin secretion rate by 88 ± 2% (18).

Another limitation of the glucagon infusion study is that although it replicates the range of glucagon concentrations observed in humans (we have previously shown that 0.65 ng/kg/min replicates portal fasting concentrations after accounting for hepatic extraction [19,20]), we likely did not replicate peak glucagon concentrations within the islet. This might explain why our observations differ from the rodent experiments (8) in which exendin-9,39 abolished glucagon-induced insulin secretion. To address this issue, in a separate group of healthy subjects, we examined the response to a 1-mg glucagon bolus. As in the first experiment, we demonstrate that blockade of the GLP-1R with a competitive antagonist, exendin-9,39, attenuates the insulinotropic effect of glucagon significantly in healthy humans. Peripheral concentrations of glucagon (Supplementary Fig. 1) were far higher than those observed in the first experiment. Additional studies will be necessary to examine the contribution, if any, of GLP-1R–mediated signaling to insulin secretion during an oral challenge and whether this is more marked in people with type 2 diabetes.

Acknowledgments. The authors thank M.M. Davis (Endocrine Research Unit, Mayo Clinic, Rochester, MN) for the excellent editorial assistance.

Funding. This study was supported by the Mayo Clinic General Clinical Research Center (DK TR000135). A.V. is supported by National Institute of Diabetes and Digestive and Kidney Diseases grants DK78646, DK116231, and DK126206. C.D.M. was supported by MIUR (Italian Minister for Education) under the initiative “Departments of Excellence” (Law 232/2016).

Duality of Interest. A.V. is the recipient of an investigator-initiated grant from Novo Nordisk and has consulted for Zealand Pharma, Crinetics, and Rezolute. No other potential conflicts of interest relevant to this article were reported.

Author Contributions. R.A.F. researched data and ran the studies. A.M.E. researched data and ran the studies. A.A.W. researched data and ran the studies. M.C.L. undertook mathematical modeling of insulin secretion. C.C. and C.D.M. supervised the mathematical modeling, contributed to the discussion, and reviewed and edited the manuscript. A.V. designed the study, oversaw its conduct, researched data, and wrote the first draft of the manuscript. The order of first authors was determined by the time that each joined the project. A.V. is the guarantor of this work and, as such, had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. Parts of this study were presented in abstract form at the 82nd Scientific Sessions of the American Diabetes Association, New Orleans, LA, 3–7 June 2022.