ER stress and unfolded protein response (UPR) signaling modulate GLP-1 receptor signaling in the pancreatic islets

Abstract

Insulin is essential for maintaining normoglycemia and is predominantly secreted in response to glucose stimulation by β-cells. Incretin hormones, such as glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide, also stimulate insulin secretion. However, as obesity and type 2 diabetes worsen, glucose-dependent insulinotropic polypeptide loses its insulinotropic efficacy, whereas GLP-1 receptor (GLP-1R) agonists continue to be effective owing to its signaling switch from Gs to Gq. Herein, we demonstrated that endoplasmic reticulum (ER) stress induced a transition from Gs to Gq in GLP-1R signaling in mouse islets. Intriguingly, chemical chaperones known to alleviate ER stress, such as 4-PBA and TUDCA, enforced GLP-1R's Gq utilization rather than reversing GLP-1R's signaling switch induced by ER stress or obese and diabetic conditions. In addition, the activation of X-box binding protein 1 (XBP1) or activating transcription factor 6 (ATF6), 2 key ER stress-associated signaling (unfolded protein response) factors, promoted Gs utilization in GLP-1R signaling, whereas Gq employment by ER stress was unaffected by XBP1 or ATF6 activation. Our study revealed that ER stress and its associated signaling events alter GLP-1R's signaling, which can be used in type 2 diabetes treatment.

Keywords: Endoplasmic reticulum stress; G protein-coupled receptor signaling; Glucagon-like peptide-1; Glucagon-like peptide-1 receptor; Type 2 diabetes.

Fig. 1

Acute ER stress triggers GLP-1R signaling use of Gq. (A-D) Isolated mouse islets were pretreated with tunicamycin (Tm, 1 μg/ml, 8 hours) and further administered with MDL-12330A (MDL, adenylyl cyclase inhibitor, 10 μM) (A, B) or YM-254890 (YM, Gq inhibitor, 200 nM) (C, D) along with exendin-4 (Ex4, 50 nM) and high glucose (Glc, 17 mM) for an additional 1 hour. (A, C) Insulin secretion and (B, D) its fold change. Data are presented as means ± SEM (n = 4). Statistical analyses were performed using either 1-way ANOVA followed by Holm-Sidak’s test (A, C) or unpaired t-tests with Welch’s correction (B, D). ns, non-significance. *P < .05, **P < .01, ***P < .001.

Fig. 2

4-PBA, a chemical chaperone, promotes GLP-1R agonist-induced insulin secretion and GLP-1R’s Gq utilization in ER stress-experiencing and db/db islets. (A, B) Islet insulin secretion. Isolated lean mouse islets were sequentially treated with 4-PBA (2.5 mM, 24 hours), tunicamycin (1 μg/ml, 8 hours), and high glucose (Glc, 17 mM, 1 hour) with or without Ex4 (50 nM), YM (200 nM), and MDL (10 μM). (C-E) Islets were isolated from 12-week-old db/db mice and then pretreated with 4-PBA (2.5 mM, 24 hours) before high glucose with indicated reagents. (C) Insulin secretion and (D, E) its fold change. Data are presented as means ± SEM (n = 4). Statistical analyses were performed using either unpaired t-tests (A, D, E) or 1-way ANOVA (B, C). ns, non-significance. *P < .05, **P < .01, ***P < .001.

Fig. 3

Ectopic expression of active XBP1 (XBP1s) in ER stress-experiencing islets enhances GLP-1R signaling use of Gs. (A-D) Isolated mouse islets were infected with adenovirus expressing red fluorescent protein (Ad-RFP) or XBP1s (Ad-XBP1s) and then pretreated with tunicamycin (8 hours) before introducing indicated reagents for 1 hour. The concentrations of reagents were the same as in previous experiments. (A, C) Islets’ insulin secretion and (B, D) its fold change. Data are presented as means ± SEM (n = 4). Statistical analyses were performed using either 1-way ANOVA (A, C) or unpaired t-tests (B, D). *P < .05, **P < .01, ***P < .001.

Fig. 4

Pharmacological modulation of XBP1s activity leads to altered GLP-1R’s Gs utilization in ER-stress-experiencing islets. (A-G) Isolated mouse islets were pretreated with IRE1α activator (IXA4, 20 μM, 24 hours) (A-D) or inhibitor (4μ8c, 32 μM, 32 hours) (E-G) and then tunicamycin (8 hour) before introducing indicated reagents for 1 hour. The concentrations of reagents were the same as in previous experiments. (A, C, E) Islets’ insulin secretion and (B, D, F, G) its fold change. Data are presented as means ± SEM (n = 4). Statistical analyses were performed using either 1-way ANOVA (A, C, E) or unpaired t-tests (B, D, F, G). ns, non-significance. *P < .05, **P < .01, ***P < .001.