The prostaglandin E2 EP3 receptor has disparate effects on islet insulin secretion and content in β-cells in a high-fat diet-induced mouse model of obesity

Abstract

Signaling through prostaglandin E₂ EP3 receptor (EP3) actively contributes to the β-cell dysfunction of type 2 diabetes (T2D). In T2D models, full-body EP3 knockout mice have a significantly worse metabolic phenotype than wild-type controls due to hyperphagia and severe insulin resistance resulting from loss of EP3 in extra-pancreatic tissues, masking any potential beneficial effects of EP3 loss in the β cell. We hypothesized β-cell-specific EP3 knockout (EP3 βKO) mice would be protected from high-fat diet (HFD)-induced glucose intolerance, phenocopying mice lacking the EP3 effector, Gα_z, which is much more limited in its tissue distribution. When fed a HFD for 16 wk, though, EP3 βKO mice were partially, but not fully, protected from glucose intolerance. In addition, exendin-4, an analog of the incretin hormone, glucagon-like peptide 1, more strongly potentiated glucose-stimulated insulin secretion in islets from both control diet- and HFD-fed EP3 βKO mice as compared with wild-type controls, with no effect of β-cell-specific EP3 loss on islet insulin content or markers of replication and survival. However, after 26 wk of diet feeding, islets from both control diet- and HFD-fed EP3 βKO mice secreted significantly less insulin as a percent of content in response to stimulatory glucose, with or without exendin-4, with elevated total insulin content unrelated to markers of β-cell replication and survival, revealing severe β-cell dysfunction. Our results suggest that EP3 serves a critical role in temporally regulating β-cell function along the progression to T2D and that there exist Gα_z-independent mechanisms behind its effects.NEW & NOTEWORTHY The EP3 receptor is a strong inhibitor of β-cell function and replication, suggesting it as a potential therapeutic target for the disease. Yet, EP3 has protective roles in extrapancreatic tissues. To address this, we designed β-cell-specific EP3 knockout mice and subjected them to high-fat diet feeding to induce glucose intolerance. The negative metabolic phenotype of full-body knockout mice was ablated, and EP3 loss improved glucose tolerance, with converse effects on islet insulin secretion and content.

Keywords: G protein-coupled receptor; beta cell (β‐cell); insulin secretion; pancreatic islet; prostaglandin.

Graphical abstract

Figure 1.

Baseline metabolic phenotype of 11-wk-old male WT and EP3 βKO C57BL/6J mice. Eleven-week old male WT and EP3 βKO mice were weighed and fasted for 4–6 h before glucose or insulin challenge. Body weight (A) and 4–6 h fasting blood glucose levels (B). C: blood glucose excursion after oral gavage with 1 g/kg sterile dextrose solution (left) and area-under the-curve (AUC) analyses (right). D: percent decrease in blood glucose levels after intraperitoneal challenge with 0.75 U/kg rapid-acting human insulin lispro (Humalog) (left) and AUC analyses (right). In A–C, n = 22–24 mice/group. In D, n = 15–17 mice/group. No statistically significant differences were found. EP3 βKO, EP3 knockout; WT, wild type.

Figure 2.

Weekly food intake, body weight, or random-fed blood glucose between WT and EP3 βKO mice fed a control diet or HFD for 16 wk. Weekly food intake (A), body weight (B), and random-fed blood glucose levels (C) of WT or EP3 βKO control or HFD-fed mice. *P < 0.05 and **P < 0.01 for WT HFD vs. WT control diet. #P < 0.05 and ####P < 0.0001 for EP3 βKO HFD vs. WT control diet (In B, for clarity, only the initial week where there was a statistically significant change and the final week are shown). In all panels, data were analyzed by two-way paired ANOVA with Sidak’s test post hoc. n = 11 or 12 mice/group. In A and C, all statistically significant differences are shown. EP3 βKO, EP3 knockout; HFD, high-fat diet; WT, wild type.

Figure 3.

Metabolic phenotype of WT and EP3 βKO mice after 16 wk of control diet or HFD feeding. Body weight (A) and 4–6 h fasting blood glucose levels (B) of 27-wk-old male mice fed a control or 45 kcal% HFD for 16 wk. C: blood glucose excursion after oral gavage with 1 g/kg sterile dextrose solution (left) and glucose area-under the-curve (AUC) analyses (right). In A, B, and C (right), data were analyzed by one-way ANOVA with Sidak’s test post hoc. In C (left), data were analyzed by two-way paired ANOVA with Sidak’s test post hoc. *P < 0.05, **P < 0.01, and ****P < 0.0001 vs. WT control or the indicated comparison. #P < 0.05 and ###P < 0.001 for WT HFD vs. WT control diet. In all panels, n = 10 or 11 mice/group. All statistically significant differences are shown. EP3 βKO, EP3 knockout; HFD, high-fat diet; WT, wild type.

Figure 4.

Expression of EP3, COX-2, and markers of inflammation, apoptosis, and replication in islets from WT and EP3 βKO mice fed a control diet or HFD for 16 wk. A: relative mRNA expression of EP3, its splice variants, and COX-2. n = 5 or 6 mice/group. B: relative mRNA expression of inflammation and apoptosis genes. n = 5 or 6 mice/group. C: relative mRNA expression of replication genes. n = 5 or 6 mice/group. The combined data in A–C were analyzed by multiple T test with Holm–Sidak test post hoc to correct for multiple comparisons. *P < 0.05. All statistically significant differences are shown. EP3 βKO, EP3 knockout; HFD, high-fat diet; WT, wild type.

Figure 5.

Insulin secretion, insulin content, and markers of β-cell mass in islets from control diet- and HFD-fed WT and EP3 βKO mice. A: insulin secreted as a percent of content in 1.7 mM glucose (1.7 G), 16.7 mM glucose (16.7 G), and 16.7 G plus 10 nM exendin-4 (16.7 G + Ex4) after 16 wk of control or HFD feeding. n = 5 or 6 mice/group. B: total islet insulin content from the data shown in A. N > 103 islets from 5 to 6 mice of each group. C: insulin secreted as a percent of content in 1.7 G, 16.7 G, and 16.7 G + Ex4 after 26 wk of control or HFD feeding. n = 4 or 5 mice/group. D: total islet insulin content from the experiments shown in C. N > 140 islets from 4 to 5 mice of each group. E: percent of Ki67-positive β cells as measured by immunofluorescence of pancreas section from mice fed the control or HFD for 16 wk. N = 2 sections from 3 to 4 mice/group. F: insulin-positive pancreas area as measured by immunohistochemistry of pancreas slide sections from mice fed the control or HFD for 26 wk. N = 2 sections from 3 to 5 mice/group (G). Islet size from the experiments shown in F. In A and C, ^aP < 0.05, 16.7 G vs. 1.7 G; ^bP < 0.05, 16.7 G + Ex4 vs. 16.7 G; ^cP < 0.05 vs. WT control diet; and ^dP < 0.05 vs. WT HFD. Data were compared by two-way ANOVA with Holm–Sidak test post hoc to correct for multiple comparisons. In B, D, and E–G, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Data were compared by one-way ANOVA with Holm–Sidak test post hoc to correct for multiple comparisons. All statistically significant differences are shown. EP3 βKO, EP3 knockout; HFD, high-fat diet; WT, wild type.