Liraglutide Attenuates Myocardial Fibrosis via Inhibition of AT1R-Mediated ROS Production in Hypertensive Mice

Abstract

Background/Aims:

Glucagon-like peptide-1 receptor agonist liraglutide has been reported to exert cardioprotective effects, but its effect on cardiac fibrosis remains controversial. The aim of this study was to investigate the effects of liraglutide on cardiac fibrosis and potential mechanisms.

Methods:

C57BL/6 mice (3-month old) were randomly divided into control, hypertension, and hypertension + liraglutide groups. The hypertensive state was created by infusion of Ang II (100 ng/kg·min) for 4 weeks through subcutaneously implanted osmotic pumps. The control mice were infused with saline. Mice were also given vehicle or liraglutide (400 μg/kg·day). Blood pressure (BP), blood sugar, myocardial fibrosis, AT1R expression, and reactive oxygen species (ROS) levels were measured. To further elucidate the mechanisms of fibrosis, mouse cardiac fibroblasts were isolated and treated with liraglutide (300 nM/L) or losartan (10 μM) for 3 hours, followed by Ang II (10⁻⁷ M) for additional 12 hours. Reactive oxygen species production and expressions of collagen-1 and -3 were measured.

Results:

Liraglutide reduced BP and blood sugar but did not affect the body weight of the hypertensive mice. Liraglutide also inhibited collagen accumulation, AT1R expression, and ROS generation in the hearts of the hypertensive mice. In in vitro studies, pretreatment with liraglutide and losartan (as control) markedly inhibited Ang II-induced ROS production and collagen expression in the cultured cardiac fibroblasts.

Conclusion:

Liraglutide reduces myocardial fibrosis in the hypertensive mice, which appears to be dependent on at least in part inhibition of ROS production.

Introduction

Heart failure is a major health problem affecting 20 to 30 million people around the world. ¹ Myocardial fibrosis, the excess deposition of extracellular matrix in the ventricular wall, is a hallmark of mid- to late-stage heart failure, which occurs in long-standing hypertension. ² The persistent pressure overload causes excessive accumulation of collagens, mainly collagen type I and type III fibers, in the interstitium and the perivascular tissues of the heart. ³ The activation of renin-angiotensin system (RAS) and subsequent release of Ang II is thought to play a role in the development of hypertension and myocardial fibrosis via stimulation of AT1R. ^4,5

Development of fibrosis, which leads to a progressive increase of ventricular stiffness and impaired diastolic function, is also seen commonly with aging and also in patients with myocardial ischemia and diabetes. ^{6 -8} Despite the high prevalence of diastolic heart failure, there are currently no effective remedies to control or inhibit cardiac fibrosis and improve the clinical state. ⁹

Glucagon-like peptide 1 (GLP-1) is an incretin hormone that enhances insulin secretion through adenylyl cyclase, inhibits glucagon secretion, delays gastric emptying, and reduces postprandial hyperglycemia. ^10,11 Glucagon-like peptide 1 binding to its receptor (GLP-1-R), a G protein-coupled receptor in pancreatic β cells, not only controls blood sugar level but also activates various signaling pathways including extracellular signal-regulated kinase 1/2 (ERK1/2-), mitogen-activated kinase (MAPK), Phosphoinositide 3-kinase (PI3K), Wnt, cyclic Adenosine monophosphate (cAMP), and protein kinase A (PKA) that mediate cell proliferation, migration, apoptosis, and autophagy. ^{12 -14} Liraglutide is a long-acting GLP-1-R agonist developed by Novo Nordisk for the treatment of type 2 diabetes by increasing insulin secretion. Recent studies have shown cardioprotective effects of liraglutide in animal models of disease as well as in clinical studies. ^{15 -17}

Some studies have shown that liraglutide has the potential to suppress cardiac fibrosis under a series of pathological conditions. ^18,19 Zhang et al showed that liraglutide can inhibit collagen-I synthesis and cardiac fibrosis in rats through a reduction in the ratio of AT1/AT2 receptor expression and increasing angiotensin-converting enzyme 2 (ACE2) activity. ²⁰ Fandiño et al also observed that liraglutide could increase ACE2 expression in rats with food restriction but did not affect ACE1 expression. ²¹ In addition, liraglutide has been reported to decrease the plasma level of Ang II in patients with type 2 diabetes. ²² Wang et al ²³ showed that pretreatment with liraglutide protected myocytes from injury, reduced myocardial infarct size, improved cardiac function, and inhibited collagen synthesis in the mouse hearts following ischemia–reperfusion. In in vitro studies, liraglutide decreased glucose- and angiotensin II (Ang II)-induced collagen accumulation in cardiac fibroblasts. ²⁴ This study was designed to investigate the effect of liraglutide on myocardial fibrosis and the potential mechanisms in the hypertensive mice infused with Ang II.

Materials and Methods

Materials

C57BL/6 mice were obtained from Beijing Vital River Laboratory Animal Technology Co, Ltd. Osmotic pumps were purchased from ALZET International Distributors. Ang II peptides and AT1R antibody were purchased from Abcam. Collagen-1a, collagen-3a, and β-actin antibody were purchased from Santa Cruz Biotechnology, Inc, and the secondary antibodies were from ZSGB-Bio. Western blot Substrate and Immun-Blot polyvinylidene fluoride (PVDF) membrane were purchased from Thermo Fisher Scientific. Dihydroethidium (DHE) kit was purchased from Beyotime Biotechnology. Masson Trichrome Stain Kit (HT15) was purchased from Sigma-Aldrich.

Hypertension Model

C57BL/6 mice (3-month old; male) were fed with a standard diet and water ad libitum and kept in an experimental condition with a 12-h light-to-dark cycle at 25 °C ± 2 °C. The mice were allowed 7 days to adapt to the environment before the experiments. Thirty-two mice were randomly divided into 2 groups: hypertensive and hypertensive + liraglutide groups (n = 16/group). After anesthesia with sodium pentobarbital (80 mg/kg, IP), the mice were placed on an operating table, and hair on their back was shaved. A 1-cm midline skin incision was made in their back, and a pocket was created along the mouse back for pump insertion by separating the skin and subcutaneous tissue using clean surgical forceps. Mini-osmotic pump (with saline or Ang II infusion) was inserted into each pocket, and the pocket was closed. The pocket was large enough to allow some free movement of the pump but not so large that it would slip. After surgery, the mice were kept in a warm incubator until awake. Ang II (100 ng/kg/min) infusion was continued for 4 weeks. During the infusion period, mice were also injected intraperitoneally with vehicle or liraglutide (400 μg/kg body weight), respectively. Another group of (control) mice (n = 16) were implanted with osmotic pumps infused with the same volume of saline and administered intraperitoneally equal volume of vehicle as liraglutide solution. All animal experiments were approved by the Animal Use and Care Committee of Xinxiang Medical University and conformed to the Guide for the Care and Use of Laboratory Animals published by NIH (#85-23, revised in 1996).

Blood Pressure Measurement

Blood pressure (BP) was measured every week by a noninvasive tail-cuff method with a Softron BP-98A (Softron Co, Ltd).

Isolation and Culture of Cardiac Fibroblasts

Cardiac fibroblasts were isolated from C57BL/6 mice (n = 4) as per previously published protocol ²⁵ and cultured in Dulbecco’s modified Eagle’s medium supplemented with 5% fetal bovine serum at 37 °C under 5% CO₂. Experiments were performed with fibroblasts from passages 3 to 5.

Masson Trichrome Staining

Hearts were collected and cardiac sections were prepared as per previously published protocols. ²³ Five-micrometer-thick slides were stained with Masson Trichrome staining Kit as per the manufacturer’s instructions. The images were captured by a digital imaging system.

Dihydroethidium Staining

For the analysis of oxidative stress in the hearts, frozen cardiac sections were cut into 7-μm-thick slides and incubated with 5-μM DHE in the dark in a humidified box at 37 °C for 30 minutes. After washing with phosphate-buffered saline (PBS), the sections were covered with coverslips by antifade reagent with 4’,6-diamidino-2-phenylindole (DAPI) and imaged with a fluorescent microscope. For the analysis of oxidative stress in cells, cardiac fibroblasts were cultured in 24-well plates, pretreated with liraglutide for 2 hours, and then treated with Ang II (10⁻⁷ mol/L) together with liraglutide (300 nmol/L) for additional 12 hours. Subsequently, cells were incubated with 5-μM DHE in 37 °C incubator for 30 minutes and 10-μM DAPI for 5 minutes. The images were rapidly captured under a fluorescent microscope.

Immunostaining

The fixed left ventricle (LV) slides were blocked with 5% goat serum/1% bovine serum albumin (BSA) in PBS for 30 minutes at room temperature and then incubated with anti-AT1R antibody (1:400, vol/vol) in 1% BSA at 4 °C overnight. After washing twice with PBS, the sections were incubated with Texas Red (TR)-conjugated secondary antibody (1:1000, vol/vol) in 1% BSA for 30 minutes at room temperature. After washing, the sections were covered with coverslips by antifade reagent with DAPI. The images were captured by a digital imaging system.

Western Blotting

Protein extraction, electrophoresis, protein transfer, and blot blocking were performed as per recently published protocols. ²⁶ After that, blots were incubated with primary antibodies against AT1R, collagen-1a, collagen-3a, or β-actin in tris-buffered saline with Tween 20 (TBS-T) at 4 °C overnight on a shaker. After washing with TBS-T, blots were incubated with secondary antibodies for 1 hour at room temperature. After washing, blots were incubated with Luminol Reagents for 3 minutes and then exposed to Autorad Film.

Statistical Analysis

Statistical analysis was performed with SPSS 15.0 software. Data were presented as the means ± standard deviations from 4 to 8 independent experiments. Univariate comparisons of means were evaluated using 1-way analysis of variance with Tukey’s post hoc adjustment for multiple comparisons. P < .05 was considered statistical significant.

Results

Body Weight, BP, and Blood Sugar Levels

Body weight and BP were measured prior to the treatment and at the end of each week during the infusion period. Blood sugar in each group of mice was measured at the end of 4-week treatment. As shown in Figure 1A, there was no significant difference in body weight among 3 groups of mice in the duration of Ang II infusion and liraglutide treatment (P > .05). Systemic infusion of Ang II markedly increased BP (systolic blood pressure [SBP]), and administration of liraglutide markedly attenuated Ang II-induced increase of SBP (P < .05; Figure 1B). Systemic infusion of Ang II did not significantly affect blood sugar levels (P > .05), but the administration of liraglutide markedly decreased blood sugar levels (P < .05; Figure 1C).

Figure 1.

Effects of liraglutide on body weight, blood pressure, and blood sugar. A and B, Body weight and systolic blood pressure (SBP) of each group of mice at 0, 1, 2, 3, and 4 weeks; C, blood sugar levels of 3 groups of mice at the end of 4 weeks. n = 8. *P < 0.05 vs. hypertension group; ^# P < 0.05 vs. control and hypertension groups.

Liraglutide and Cardiac Fibrosis in Hypertensive Mice

Masson Trichrome staining (Figure 2A and B) showed that systemic infusion of Ang II for 4 weeks caused a marked fibrosis in mouse LV, and liraglutide markedly attenuated this phenomenon (P < .05). These data were further confirmed by Western blotting, which showed similar changes in collagen-3 expression in the LV of hypertensive mice following treatment with liraglutide (Figure 2C).

Figure 2.

Effects of liraglutide on cardiac fibrosis in hypertensive mice. (A) Masson staining showing collagen accumulation in the ventricles of each group of mice; (B) quantification of positivity of Masson staining in each group of mice; (C) Western blotting data showing collagen-3 expression in the ventricles of each group of mice. n = 8. *P < .05 versus control group; ^# P < .05 versus hypertension group.

Liraglutide and AT1R Expression and Reactive Oxygen Species Generation in the Hypertensive Mice

Previous studies have shown increased AT1R expression and reactive oxygen species (ROS) generation in the hypertensive murine hearts. ^27,28 In accordance with these reports, we observed an increase of AT1R expression (Figure 3) and ROS production in the mouse hearts following 4-week of Ang II infusion (Figure 4). More importantly, we found that treatment with liraglutide markedly inhibited Ang II-induced AT1R expression (Figure 3) and ROS production (Figure 4) in the hearts of hypertensive mice (P < .05).

Figure 3.

Effect of liraglutide on angiotensin II receptor type 1 (AT1R) expression in the ventricles of hypertensive mice. (A) Immunostaining showing AT1R expression in the ventricles of each group of mice; (B) quantification of images of AT1R immunostaining; (C) Western blotting data showing AT1R expression in the ventricles of each group of mice. n = 8. *P < .05 versus control group; ^# P < .05 versus hypertension group.

Figure 4.

Effect of liraglutide on ROS production in the ventricles of hypertensive mice. (A) DHE staining showing ROS levels in the ventricles of each group of mice; (B) quantification of images of DHE staining. n = 8. *P < .05 versus control group; ^# P < .05 versus hypertension group. DHE indicates dihydroethidium; ROS, reactive oxygen species.

In Vitro Studies

To further elucidate the mechanism of actions of liraglutide, we performed in vitro studies using mouse cardiac fibroblasts. As per previous reports, ²⁵ Ang II caused a marked increase in AT1R at transcriptional (Figure 5A) and protein (Figure 5B) levels in cardiac fibroblasts. Importantly, liraglutide attenuated Ang II-induced AT1R expression (P < .05). Liraglutide also significantly inhibited Ang II-induced ROS production (Figure 6; P < .05). Similar to the in vivo data, liraglutide markedly Inhibited Ang II-induced collagen-3 accumulation (Figure 7A) in cardiac fibroblasts (P < .05). As a control for liraglutide, we used AT1R blocker losartan and observed that it also markedly inhibited Ang II-induced collagen-3 expression in cardiac fibroblasts (P < .05). Liraglutide and losartan both also inhibited Ang II-induced collagen-1 accumulation in cardiac fibroblasts (P < .05; Figure 7B).

Figure 5.

Effect of liraglutide on angiotensin II (Ang II)-induced AT1R expression in cardiac fibroblasts. (A) AT1R mRNA expression; (B) AT1R protein expression. n = 4. *P < .05 versus control group; ^# P < .05 versus Ang II group.

Figure 6.

Effect of liraglutide on angiotensin II (Ang II)-induced ROS production in cardiac fibroblasts. (A) DHE staining showing ROS levels in different groups of cardiac fibroblasts; (B) quantification of images of DHE staining. n = 4. *P < .05 versus control group; ^# P < .05 versus Ang II group. DHE indicates dihydroethidium; ROS, reactive oxygen species.

Figure 7.

Effect of liraglutide on angiotensin II (Ang II)-induced collagen accumulation in cardiac fibroblasts. (A) Collagen-3 expression following treatments with Ang II, losartan (Los) plus Ang II or liraglutide plus Ang II; (B) collagen-1 expression following treatments with Ang II, losartan (Los) plus Ang II, or liraglutide plus Ang II. n = 4. *P < .05 versus control group; ^# P < .05 versus Ang II group.

Discussion

Cardiac fibrosis is a chronic process, characterized by myocardial stiffening and impaired diastolic function. It is a common problem associated with nearly all forms of heart disease and believed to be the final pathway causing heart failure. ² The abnormal secretion of collagen from cardiac fibroblasts is thought to be the basis of cardiac fibrosis. ²⁹ It is known that both hypertension and diabetes are independent risk factors for cardiac fibrosis, most likely due to the activation of RAS and release of Ang II. ^30,31 Prevention of cardiac fibrosis is an essential goal in the management of hypertensive heart disease, which is often associated with diabetes. ^31,32 Liraglutide, a long-acting GLP-1-R agonist has been reported to prevent myocardial fibrosis in patients and in some animal models of diabetes and myocardial infarction ^18,19 but not in all. ³³ In the present study, we asked if liraglutide would reduce cardiac fibrosis in hypertensive mice. Indeed, we observed that liraglutide markedly reduced cardiac fibrosis in mice that had been made hypertensive by chronic Ang II infusion.

Renin-angiotensin system activation is believed to be one of the most important mechanisms in controlling myocardial fibrosis in hypertension. Ang II is a major component of RAS, which stimulates collagen synthesis and secretion from fibroblasts and myofibroblasts through activating its type 1 receptor (AT1R). ³⁴ In accordance with previous reports, we observed an increase in AT1R expression (immunostaining and Western blotting) and ROS production in the mouse hearts following 28 days of infusion of Ang II. More importantly, treatment with liraglutide markedly inhibited Ang II-induced rise in SBP as well as an increase in AT1R expression and ROS production in the hearts of hypertensive mice (Figures 3 and 4). The in vitro data also showed that liraglutide markedly suppressed Ang II-induced AT1R expression and ROS production in primary mouse cardiac fibroblasts. A recent study by Zheng et al demonstrated that liraglutide attenuated cardiac fibrosis and improved heart function via blocking AT1R in rats. ³⁵ In this study, we further proved that the anti-diabetic drug liraglutide can prevent AT1R-ROS signaling. Indeed, AT1R-mediated ROS production in cardiovascular system and its relevance have been recently discussed. ³⁶ As control for liraglutide, we used a specific AT1R blocker losartan and showed that it prevented ROS generation and collagen-1 and -3 expressions. Losartan effectively reverses myocardial fibrosis in the murine hypertensive models. ^37,38 Based on the similarity of actions of losartan and liraglutide, it may be concluded that liraglutide acts as an anti-fibrosis drug works in part by inhibiting AT1R-ROS axis. The anti-fibrotic effects of liraglutide and losartan may ameliorate cardiac functions in the hypertensive mice. Actually, a recent study reported liraglutide and telmisartan (another AT1 blocker) both markedly improved cardiac performance of hypertensive rats. ³⁵

Cardiac fibroblasts account for approximately 25% of myocardial volume and 60% of all cells in the heart in the physiologic state. ³⁹ In the pathological state, cardiac fibroblasts proliferate, become activated in response to neurohormonal changes such as heightened levels of Ang II, and produce collagen. ⁴⁰ Our previous study showed that liraglutide could attenuate Ang II- and glucose-induced collagen formation in isolated cultured cardiac fibroblasts by inhibiting ERK1/2 and nuclear factor κB (NF-κB) pathways. ²⁴ Gaspari et al also observed that liraglutide suppressed cardiac fibrosis through inhibiting NF-κB expression and activity in obesity, hypertension, and age-induced murine models. ¹⁹

Despite advances in understanding the mechanisms of cardioprotective effects of liraglutide, its role in preventing cardiac fibrosis remains controversial. Some studies showed that liraglutide exerts cardioprotective effects in animal and clinical studies. ^15,16 Bizino et al reported that liraglutide could reduce the infarct volume and slow the progression of diabetic cardiomyopathy in patients with type 2 diabetes. ¹⁵ Noyan Ashraf et al reported that liraglutide not only reduced cardiac rupture and infarct size but also improved cardiac output. ¹⁶ Their subsequent study indicated that liraglutide could also attenuate high-fat diet-induced collagen deposition and perivascular fibrosis in the hearts of obese mice, independent of changes in body weight. ⁴¹ A previous study indicated that the cardioprotective effects of liraglutide at least in part depend on the inhibition of ROS production and inflammatory cytokines in the hearts following myocardial infarction. ²³ An in vitro study indicated that liraglutide protected against cardiomyocyte injury induced by H₂O₂ by preserving physiological levels of calcium. ⁴² This study also suggested that liraglutide prevented cardiac fibrosis in hypertensive mice in part by inhibiting AT1R-ROS pathway. It has been reported that AMPKα signaling exerts a direct inhibitory effect on ATR1 both in terms of the receptor’s expression and signaling. ^43,44 Liraglutide also inhibits cardiac remodeling via regulating PI3K/Akt1 and AMPKα signaling. ⁴⁵ Thus, the effects of liraglutide on myocardial fibrosis are multifactorial and partially related to activation of adenosine 5‘-monophosphate (AMP)-activated protein kinase-α (AMPKα) signaling.

Although several studies ^16,42,41 showed cardioprotective effect of liraglutide in mice or in vitro experiments, other studies suggest that liraglutide may not exert cardioprotection in large animal models. ^46,47 For instance, Kristensen et al and Mortensen et al both showed that administration of liraglutide did not affect infarct size and cardiac function of pigs with ischemia/reperfusion. ^46,47 More importantly, clinical trials also indicated that liraglutide did not reduce the rates of nonfatal myocardial infarction and hospitalizations for heart failure. ¹⁷ It is possible that the species specificity may be one of the main factors behind the controversy.

We showed that liraglutide markedly attenuated Ang II-induced increase of SBP. A recent study suggested BP-lowering effect of liraglutide in individuals, which is dependent on drug dose. ⁴⁸ We believe that the reduction in BP as well as lowering of blood sugar may be the physiologic mediators of the anti-fibrotic effect of liraglutide.

There are some limitations to this study. First, this work was performed in an animal model of hypertension and in cultured cardiac fibroblasts. Further clinical studies are needed to prove the protective effects of liraglutide on cardiac fibrosis in hypertension. Second, the present data emphasized the effect of liraglutide in Ang II-induced hypertension model, there are many other pathways of the genesis of hypertension. If the mechanisms of cardiac fibrosis are the same in other types of hypertension and if liraglutide affects all different mechanisms leading to cardiac fibrosis cannot be discerned from this study. Thus, future studies are required to clarify the specificity of the cardioprotective actions of liraglutide.