Metabolic Syndrome Abolishes Glucagon-Like Peptide 1 Receptor Agonist Stimulation of SERCA in Coronary Smooth Muscle.

Metabolic syndrome (MetS) doubles the risk of adverse cardiovascular events. Glucagon-like peptide 1 (GLP-1) receptor agonists induce weight loss, increase insulin secretion, and improve glucose tolerance. Studies in healthy animals suggest cardioprotective properties of GLP-1 receptor agonists, perhaps partially mediated by improved sarco-endoplasmic reticulum Ca(2+) ATPase (SERCA) activity. We examined the acute effect of GLP-1 receptor agonists on coronary smooth muscle cells (CSM) enzymatically isolated from lean, healthy Ossabaw miniature swine. Intracellular Ca(2+) handling was interrogated with fura-2. The GLP-1 receptor agonist exenatide activated SERCA but did not alter other Ca(2+) transporters. Further, we tested the hypothesis that chronic, in vivo treatment with GLP-1 receptor agonist AC3174 would attenuate coronary artery disease (CAD) in swine with MetS. MetS was induced in 20 swine by 6 months' feeding of a hypercaloric, atherogenic diet. Swine were then randomized (n = 10/group) into placebo or AC3174 treatment groups and continued the diet for an additional 6 months. AC3174 treatment attenuated weight gain, increased insulin secretion, and improved glucose tolerance. Intravascular ultrasound and histology showed no effect of AC3174 on CAD. MetS abolished SERCA activation by GLP-1 receptor agonists. We conclude that MetS confers vascular resistance to GLP-1 receptor agonists, partially through impaired cellular signaling steps involving SERCA.

Introduction

Metabolic syndrome (MetS) is defined as the presence of three or more of the following five risk factors: obesity, hypertension, glucose intolerance, insulin resistance, and dyslipidemia (1). Obesity, MetS, and type 2 diabetes are all independent risk factors for cardiovascular disease. The presence of MetS doubles the risk of experiencing an adverse cardiovascular event (2). Thus, it is important to understand the effect of diabetes and MetS treatment modalities on cardiovascular health.

Insulin-sensitizing drugs and those that enhance insulin secretion from pancreatic β-cells have been used in the treatment of type 2 diabetes and MetS. Others, such as dipeptidyl peptidase-4 inhibitors and glucagon-like peptide 1 (GLP-1) receptor agonists (exenatide, liraglutide, AC3174, etc.), increase both insulin sensitivity and insulin secretion (3,4). Recently, insulin-sensitizing thiazolidinediones have been implicated in increased risk of fluid retention, LDL and triglyceride accumulation, heart failure, and myocardial infarction (5–7). Hence, there is great attention on cardiovascular outcomes of antidiabetic agents, and the U.S. Food and Drug Administration requires cardiovascular safety profiles for all new agents (8).

GLP-1 receptor agonists are attractive treatment options for MetS because GLP-1 is an endogenous hormone that functions in normal physiology to increase insulin sensitivity, biosynthesis, and secretion (3). Additionally, GLP-1 receptor agonists reduce myocardial infarct size in ischemia/reperfusion injury (9), improve cardiac function in chronic heart failure (10), and attenuate neointimal formation after vascular injury (11). At the cellular level, GLP-1 receptor agonists improve endothelial calcium homeostasis after simulated ischemia/reperfusion (12). Further, GLP-1 receptor agonists improve sarco-endoplasmic reticulum Ca2+ ATPase (SERCA) activity in multiple cell types (13,14). These studies provide rationale for a protective role of GLP-1 in the coronary vasculature.

A recent study revealed impaired GLP-1 stimulation of myocardial glucose uptake in human patients with type 2 diabetes and in MetS swine (15). This finding raises the question of potential resistance to GLP-1 in MetS and type 2 diabetes and highlights the need for rigorous examination of GLP-1 action on coronary artery disease (CAD), specifically in MetS. Given the stimulation of SERCA in macrophages and ventricular myocytes (13,14) and altered SERCA in coronary smooth muscle cells (CSM) in MetS and diabetes (16,17), potentially vasoprotective actions of GLP-1 could be mediated through cellular Ca2+ signaling. This study investigated the effect of GLP-1 receptor agonists on SERCA stimulation in CSM from lean swine and on CAD and CSM SERCA stimulation in MetS.

Research Design and Methods

Animals

All protocols involving animals were approved by the Institutional Animal Care and Use Committee at Indiana University School of Medicine and fully complied with animal use standards (18,19). Ossabaw miniature swine were separated into two treatment groups: placebo (n = 10) and GLP-1 receptor agonist AC3174 treated (n = 10; 0.25 µg/kg body weight subcutaneously, twice daily; Amylin Pharmaceuticals, San Diego, CA), which displays nearly identical pharmacokinetic and pharmacodynamics profiles as the marketed GLP-1 receptor agonist exenatide (4). CAD and MetS were induced in both treatment groups as described in the Supplementary Material. One animal in the placebo group died prior to the collection of end point data.

Isolation of CSM

Epicardial coronary arteries were cleaned of adherent tissue and CSM were isolated with a collagenase solution as previously described (16,17).

Measurement of Intracellular Ca2+

CSM were loaded with fura-2/AM, and whole-cell intracellular Ca2+ levels were measured as described in the Supplementary Material.

Acute In Vitro Exenatide Treatment

CSM from lean, healthy Ossabaw swine or a separate group of swine with MetS and CAD were treated with 100 nmol/L exenatide during assessment of intracellular Ca2+. The selective SERCA inhibitor cyclopiazonic acid (CPA; 10 μmol/L) was used as a negative control for SERCA function.

Intravenous Glucose Tolerance Testing

Intravenous glucose tolerance testing (IVGTT) was performed as previously described (16,20) and in the Supplementary Material.

Plasma Lipid, Electrolyte, and Enzyme Assays

Blood samples were obtained at time of IVGTT prior to intravenous injection of glucose. Lipid electrolyte and enzyme content were measured offsite (Antech Diagnostics, West Lafayette, IN).

Intravascular Ultrasound

After 12 months on the diet, intravascular ultrasound (IVUS) pullbacks were performed as described in the Supplementary Material.

Histology

Coronary arterial rings were placed in phosphate-buffered formalin at the time of euthanasia. Hematoxylin and eosin, Verhoeff-van Gieson, and trichrome staining were performed on sections of these rings. Plaque burden and collagen content were determined using commercially available software (ImageJ 1.48v; National Institutes of Health) as previously described (16).

Results

Acute In Vitro Exenatide Treatment in CSM From Lean, Healthy Swine

We investigated the acute effect of the GLP-1 receptor agonist, exenatide, on intracellular Ca2+ handling in CSM isolated from lean, healthy Ossabaw swine. Sarcoplasmic reticulum (SR) Ca2+ stores were released with caffeine in the absence of extracellular Ca2+ in the presence and absence of 100 nmol/L exenatide (Fig. 1), and the subsequent undershoot of cytosolic Ca2+ during recovery was assessed. Exenatide treatment did not alter resting cytosolic Ca2+ (Fig. 1) or Ca2+ influx through voltage-gated Ca2+ channels (Fig. 1). Exenatide elicited improved recovery of cytosolic Ca2+ below baseline levels, and this effect was completely ablated in the presence of the SERCA inhibitor CPA (Fig. 1). The GLP-1 receptor agonist liraglutide also demonstrated improved recovery of cytosolic Ca2+, which was prevented by the GLP-1 receptor antagonist, exendin (9-39) (Supplementary Fig. 1). In the presence of caffeine, SERCA buffering of cytosolic Ca2+ is functionally inhibited because of the much more rapid release of Ca2+ through ryanodine receptors on the SR membrane. Therefore, we assessed the rate of recovery to 50% recovery to baseline (t50) in the presence of caffeine as a measure of Ca2+ extrusion and/or Ca2+ uptake into caffeine-insensitive intracellular stores, and to 100% recovery to baseline (t100) in the absence of caffeine as another measure of SERCA activation. Exenatide significantly decreased time to 100%, but not 50%, recovery to baseline (Fig. 1).

Figure 1

Effect of acute exenatide treatment on CSM Ca2+ handling in CSM isolated from lean, healthy Ossabaw swine. A: Representative tracing of data from Ca2+ imaging protocol. Horizontal dashed line indicates baseline. (CAF, caffeine; control, n = 114 CSM; exenatide [Exen], n = 82 CSM; CPA, n = 68 CSM.) B: Resting cytosolic Ca2+. C: Area under the curve (AUC) for Ca2+ influx during high K+, corresponding to the shaded region in panel A. D: Undershoot of Ca2+ below resting levels, corresponding to the black vertical arrow in panel A. E: Time to half (t50) and total (t100) recovery from caffeine-induced rise in cytosolic Ca2+, corresponding to the horizontal black arrows in panel A. *, different from other treatments. P < 0.001.

Clinical Measurement of MetS in Ossabaw Swine

We then investigated the chronic effect of GLP-1 receptor agonism in vivo. To examine the effects of diet and GLP-1 receptor treatment on glucose metabolism, kidney function, and plasma electrolytes, blood profiles of these parameters were obtained (Supplementary Table 1) and IVGTTs were performed. Overall, feeding of an excess-calorie, atherogenic diet increased body weight, which was significantly attenuated by AC3174 treatment (Fig. 2). Blood pressure was not altered by diet or AC3174 treatment. Whereas fasting plasma glucose was not altered by diet or AC3174 treatment (Supplementary Table 1), IVGTT assessment of glucoregulation revealed improved glucose handling in AC3174-treated swine, compared with placebo (Fig. 2). This was corroborated by augmented plasma insulin levels after intravenous glucose challenge in AC3174-treated animals, compared with placebo (Fig. 2).

Figure 2

AC3174 attenuates weight gain and improves glucose handling in Ossabaw miniature swine. Closed circles, placebo treatment group; open circles, AC3174 treatment group. A: Weight gain in Ossabaw miniature swine. Inset timeline indicates time on placebo or AC3174 treatment, which corresponds with times in B–D and Supplementary Table 1. Time course of plasma glucose responses during IVGTT at beginning (B), 12 weeks (C), and 24 weeks (D) of AC3174 treatment. Time course of plasma insulin responses during IVGTT at beginning (E), 12 weeks (F), and 24 weeks (G) of AC3174 treatment. *, different from placebo. P < 0.05.

Intravascular Ultrasound and Histology Measurement of CAD

IVUS was used to examine severity of CAD in vivo and was confirmed by histology. Both wall coverage and percent plaque burden were assessed as previously described (16) in placebo and AC3174-treated pigs. AC3174 treatment did not alter CAD severity, as indicated by either wall coverage or percent plaque burden (Fig. 3). In vitro, histological examination of coronary arterial rings revealed no effect of AC3174 treatment on plaque burden (Fig. 3) or collagen deposition (Fig. 3).

Figure 3

AC3174 treatment does not attenuate CAD progression as measured by IVUS and histology. A: IVUS image of artery with CAD. C, catheter; L, lumen. Distance between dots is 1 mm. Original lumen traced in white dashed line. Actual lumen traced in white dotted line. Atherosclerotic plaque is indicated between dotted and dashed lines. B: The heart and major epicardial coronary arteries. CFX, circumflex; LAD, left anterior descending; RCA, right coronary artery. The RCA and CFX on the anterior aspect of the heart are shown by the solid lines and the arteries traversing to the posterior aspect of the heart are shown by the dashed lines. The IVUS catheter is shown in the RCA positioned for a pullback. C and D: Quantification of IVUS pullbacks performed in the RCA in both placebo (n = 6) and AC3174-treated (n = 3) swine. AC3174 did not attenuate CAD as measured by percent plaque burden (C) or percent wall coverage (D). Coronary arterial rings from placebo (n = 4; E–G) and AC3174-treated (n = 5; I–K) swine. A, adventitia; I, intima; L, lumen; M, media. A and E: Coronary artery rings stained with hematoxylin and eosin (H&E). F and J: Coronary artery rings stained with Verhoeff-van Gieson (VVG) stain for elastin. An overt atherosclerotic plaque is traced in panel H. G and K: Coronary artery rings stained with Masson’s trichrome (TC) for collagen. H: Graphical representation of plaque burden (P = 0.35). L: Graphical representation of total collagen area (P = 0.79).

Effect of Chronic, In Vivo GLP-1 Receptor Agonist Treatment on CSM Ca2+ Handling in MetS-Induced CAD

We have previously demonstrated alterations in CSM Ca2+ handling in MetS-induced CAD (16,17,21). We therefore investigated whether chronic, in vivo AC3174 treatment resulted in improved intracellular Ca2+ handling in CSM. Figure 4 shows the Ca2+ signaling protocol to assess baseline Ca2+ levels, SR Ca2+ store release assessed by the caffeine-induced Ca2+ peak in the absence of extracellular Ca2+, and SERCA activity assessed by subsequent recovery of cytosolic Ca2+ levels and undershoot of cytosolic [Ca2+] below baseline levels. Chronic AC3174 treatment did not alter intracellular Ca2+ handling in CSM from swine with MetS-induced CAD (Fig. 4).

Figure 4

Effect of GLP-1R agonists on CSM Ca2+ handling in MetS Ossabaw swine. A: Representative tracing of data from Ca2+ imaging protocol. Dashed line indicates baseline. B–D: Effect of chronic in vivo AC3174 treatment on CSM Ca2+ handling (Caf, caffeine; placebo, n = 4 swine; AC3174, n = 6 swine). B: Resting cytosolic Ca2+ levels. C: Caffeine-induced SR Ca2+ store release, corresponding to brackets in panel A. D: Undershoot of cytosolic Ca2+ levels below baseline, corresponding to black arrow in panel A. E and F: Effect of acute ex vivo exenatide treatment on Ca2+ handling in CSM from MetS Ossabaw swine (0 Ca, n = 38 CSM; exenatide [Exen], n = 24 CSM). E: Caffeine-induced SR Ca2+ store release, corresponding to brackets in panel A. F: Undershoot of cytosolic Ca2+ levels below baseline, corresponding to black arrow in panel A.

Effect of Acute, In Vitro Exenatide Treatment on CSM in MetS and CAD

Further, we assessed the acute, direct effect of GLP-1 receptor agonists on Ca2+ handling in CSM isolated from swine with MetS and CAD. Cells were exposed to the GLP-1 receptor agonist exenatide (100 nmol/L), which has an identical pharmacological profile as AC3174 (4) (Fig. 4), for 3 min prior to treatment with high K+. We again assessed caffeine-induced SR Ca2+ store release by peak Ca2+ response to caffeine and SERCA activity by the subsequent undershoot of cytosolic Ca2+ during recovery. Exenatide did not alter either SR Ca2+ store release or the undershoot of cytosolic Ca2+ below baseline (Fig. 4).

Discussion

The principle finding of this study is that MetS confers vascular resistance to the effects of GLP-1 receptor agonists. AC3174 did improve several cardiovascular risk factors, such as the metabolic factors, glucose tolerance, and insulin secretion, indicating that longer-term treatment may have indirect benefits on cardiovascular health. Treatment of MetS and type 2 diabetes is confounded by potential secondary and detrimental cardiovascular effects. Previous studies in lean, healthy animals indicate that GLP-1 receptor agonists provide a potential cardioprotective treatment for type 2 diabetes and MetS (9–11). In the current study, we examined possible Ca2+ regulatory mechanisms for GLP-1 receptor agonist action in CSM from lean, healthy Ossabaw swine. Our finding that GLP-1 receptor agonists exert a positive effect on SERCA activity in CSM from lean Ossabaw swine is in agreement with findings that GLP-1 receptor agonists enhance SERCA activity in other cell types, including endothelial cells (12), macrophages (13), and cardiomyocytes (14), and is the first study examining the effect of GLP-1 receptor agonists in CSM. The kinetics of the Ca2+ transient shown in Fig. 1 demonstrate a lack of exenatide effect on Ca2+ extrusion. We have previously shown these kinetic measurements to be an appropriate assay of Ca2+ extrusion (17). Reduced SERCA activity has been implicated in CSM proliferation and neointimal formation in the progression of CAD (22,23). One possible mechanism explaining this phenomenon is induction of endoplasmic reticulum (ER) stress. As the single means by which Ca2+ may enter the ER, SERCA is a crucial regulator of ER Ca2+ homeostasis, and SERCA inhibition elicits an ER stress response through depletion of ER Ca2+ (24). ER stress is associated with the development of atherosclerosis (13). Inhibition of ER stress through heightened activation of SERCA could provide a novel means by which to treat CAD.

We also examined the effect of GLP-1 receptor agonists on CAD progression in MetS. This study was needed because of recent evidence of MetS-induced resistance to the cardioprotective effects of GLP-1 (15). The improvement of systemic glucoregulation with AC3174 treatment provides essential positive evidence for GLP-1 receptor agonist action in the Ossabaw swine model of MetS and CAD. Further, we demonstrated that GLP-1 receptor agonist AC3174 has no effect on MetS-induced CAD. This is in contrast with the recent finding that GLP-1 prevents myocardial ischemia–reperfusion injury (9) and injury-induced neointimal hyperplasia (11), providing evidence that the MetS phenotype itself confers cardiovascular resistance to the beneficial effects of GLP-1. It is important to note that although AC3174 treatment did not attenuate CAD progression, the absence of adverse effects of AC3174 provide a cardiovascular safety profile for GLP-1 receptor agonists.

Additionally, we examined the acute, direct effect of GLP-1 receptor agonists on intracellular Ca2+ regulation in CSM from Ossabaw swine with MetS and CAD who had not received any treatment with GLP-1 receptor agonists. Here, we found that exenatide had no effect on Ca2+ regulation in CSM from swine with MetS and CAD. This finding is in contrast with other studies in which GLP-1 receptor agonism resulted in increased SERCA activity (12–14). A study in humans with type 2 diabetes revealed that exenatide treatment enhanced endothelial-dependent vasodilation (25), although it is important to note that blood cholesterol levels were controlled in these patients. The current study, revealing a lack of effect of GLP-1 receptor agonism on SERCA activity in the setting of MetS, corroborates the earlier finding that MetS ablated the effect of GLP-1 receptor agonists in myocardium and highlights the need for additional studies to investigate specific aspects of MetS (obesity, hyperinsulinemia, glucose intolerance, hypertension, and dyslipidemia) that may underlie resistance of the coronary vasculature to GLP-1 receptor agonists.