Glibenclamide attenuates myocardial injury by lipopolysaccharides in streptozotocin-induced diabetic mice

Background Sepsis is a common disease that continues to increase in incidence in the world. Diseases, such as diabetes mellitus, may make the situation worse. Diabetic patients are at increased risk for common infections. This study was designed to investigate the role of glibenclamide on myocardial injury by lipopolysaccharides (LPS) in streptozotocin induced diabetic mice (STZ-mice). Methods LPS was used to induce endotoxemia in STZ-mice. Heart rate and mean arterial pressure were measured by MPA-HBBS. Serum epinephrine level was measured by enzyme-linked immunosorbent assays (ELISA). Myocardial injury was examined by light and transmission electron microscope and TUNEL staining. Macrophage infiltration was measured by immunohistochemistry. Interleukin-1β (IL-1β) and tumor necrosis factor-α (TNF-α) levels in myocardial tissue and serum in STZ-mice, and in conditional medium of primary cultured peritoneal macrophages were determined by ELISA. Nalp3 and Caspase-1 protein levels were measured by Western blotting analysis. Results STZ administration decreased body weight and increased blood glucose in C57BL/6 mice. LPS injection caused decreases of heart rate and mean arterial pressure, and elevated serum epinephrine level in C57BL/6 mice. Compared with control mice without STZ treatment, LPS induced more severe myocardial injury and macrophage infiltration in STZ-mice, which was attenuated by pretreatment of glibenclamide. LPS stimulation enhanced the levels of IL-1β and TNF-α in both cardiac tissue and serum. Glibenclamide pretreatment significantly inhibited the serum levels of pro-inflammatory cytokines. Either high glucose or LPS increased the levels of IL-1β and TNF-α in the conditional medium of peritoneal macrophages. Glibenclamide treatment suppressed the increase of IL-1β level induced by high glucose and LPS. Furthermore, Nalp3 and Caspase-1 levels were markedly increased by high glucose plus LPS, and both proteins were significantly inhibited by glibenclamide treatment. Conclusions We conclude that glibenclamide could attenuate myocardial injury induced by LPS challenge in STZ-mice, which was possibly related to inhibiting inflammation through Nalp3 inflammasomes. Electronic supplementary material The online version of this article (doi:10.1186/s12933-014-0106-y) contains supplementary material, which is available to authorized users.


Introduction
Septic shock induced by bacteremia is one of the leading causes of death in critical patients. The mortality of septic shock ranges from 37% to 47% [1]. Activation of inflammatory factors in septic shock always occurs as a simultaneous immune response program initiated early in the course of the disease. Such as endotoxemia, this occurs frequently in septic shock, cause to hemorrhages, necrosis of the kidneys, and myocardial dysfunction. To the final stage, the progressive systemic organ failures may be developed due to interaction between the severe infection and the hyperactive inflammatory response during septic shock [2].
Lipopolysaccharide (LPS) is considered the principal cause responsible for the heart failure in sepsis shock. LPS may trigger acute and chronic inflammation, leading to immune cell activation and cytokine release [3]. In endotoxemia, hyperactivation of the immune response leads to the excessive production of various pro-inflammatory cytokines (IL-1β and TNF-α) and cellular injury [4], which also can result in a systemic inflammatory response and eventually lead to multiple organ failure and death. However, the precise mechanisms responsible for myocardial dysfunction in the setting of endotoxemia are not fully elucidated [5].
Diabetes mellitus is a group of metabolic disorders characterized by hyperglycaemia resulting from defects in insulin secretion, insulin action or both [6]. With a disease rate of 8.3% and cost of $174 billion, there is no debate that diabetes mellitus is a highly prevalent and costly lifelong disease [7]. It was well-established that all patients with diabetes mellitus were at increased risk for bacterial infections. Moreover, diabetes mellitus was associated with a poorer prognosis among the patients with bacteremia [8]. Therefore, further studies are warranted to manifest the relation between infection and diabetes mellitus, and to deliver more effective management of infections in diabetic patients [9].
Glibenclamide, an ATP-sensitive potassium channel (K ATP ) blocker, is the most widely used sulfonylurea drug for the treatment of type 2 diabetes mellitus in the United States [10]. It has been shown that glibenclamide suppressed neutrophil migration and chemotaxis during inflammatory responses via blocking K ATP channel [11]. Previously glibenclamide was reported to be able to reduce shock-induced overproduction of pro-inflammatory cytokines during simulated in vivo endotoxinaemia [12]. Importantly, Mohamed et al. reported that glibenclamide prevented activation of the Nalp3 inflammasomes [13]. Nalp3 is an essential component of inflammasomes triggered by pathogen-associated molecular patterns, dangerassociated molecular patterns, and crystalline substances [14][15][16][17]. Inflammasomes activate Caspase-1 for processing and secretion of the cytokines IL-1β and TNF-α [17].
Inappropriate Nalp3 activity has been incriminated in the pathogenesis of several diseases, including gouty arthritis, Alzheimer's and silicosis [18][19][20]. Thus, inhibition of the Nalp3 inflammasomes may offer considerable therapeutic promise in inflammatory-associated disease [13]. With LPS induced endotoxemia in STZ-mice, in the present study, we hypothesized that glibenclamide could attenuate myocardial injury through inhibiting inflammation by preventing activation of Nalp3 inflammasomes.

Animals
Male C57BL/6 mice weighing 20 g, about 7 weeks of age, obtained from the SLAC Laboratory Animal Co., Ltd. (Shanghai, China). The animals were maintained at 23°C ± 2°C under a cycle of 12 h light/12 h darkness with free access to food and water. All the animals used in this study received humane care in compliance with the institutional animal care guidelines and the Guide for Care and Use of Laboratory Animals published by the National Institutes of Health.

STZ-induced diabetic mice
Animals were intraperitoneally injected with a single dose of STZ (Amresco, USA) at 60 mg/kg body weight, dissolved in 0.1 mM sodium citrate buffer (pH 4.5) [21]. On the fifth day after STZ administration, whole blood was obtained from the mice tail vein and glucose levels were measured using the blood glucose monitoring system (MAJOR, Taiwan). For the present study, hyperglycemia is defined as a blood glucose measurement of 20 mM or higher. Citrate buffer-treated mice were used as a normoglycemic control (blood glucose < 12 mM). The STZ-mice were randomly divided into 3 groups: hyperglycemic mice treated with vehicle and glibenclamide-treated mice (per day, 5 or 20 mg/kg, i.g, ×14 d) [11,22,23].

Morphological analysis
For histopathological observation, the mice were sacrificed at 6 h after LPS injection. Heart tissues were fixed in 10% formalin, embedded in paraffin, sectioned, and then stained with hematoxylin and eosin (H&E) for morphological analysis. Transmission electron microscope studies were performed as previously described [5]. Tissues were fixed with 2.5% glutaraldehyde in 0.1 mol/ l cacodylate buffer, pH 7.4, for 1 h at 4°C. After rinsing in cacodylate buffer, tissues were postfixed in 1% cacodylate-buffered osmium tetroxide for 2 h at room temperature, and then dehydrated in a graded series of ethanol, transferred to propylene oxide, embedded in Epon-Araldite. Ultrathin sections (60-to-80-nm thick) were cut with a diamond knife, placed on formvar carbon-coated copper grids (200 mesh), and stained with uranyl acetate and lead citrate. Morphometric analyses were observed with a Hitachi H-800 Transmission Electron Microscope (Hitachi, Japan) [26,27].

Apoptotic analysis
Terminal deoxynucleotidyl-transferase-mediated dUTPnick-end labeling (TUNEL) staining was performed on formalin-fixed, paraffin-embedded sections with a commercial kit (Boehringer Mannheim, Mannheim, Germany) according to the instructions. Briefly the sections were first deparaffinized with xylene and ethanol and slides rinsed twice with phosphate buffered saline (PBS), treated 15 min with 20 mg/ml proteinase K (Boehringer Mannheim) in 0.1 mol/L Tris-HCl buffer (pH 7.4), then again rinsed twice with PBS. After adding the total volume (50 μl) of enzyme solution (TdT) to the remaining 450 μl of labeled solution (dUTP) to obtain 500 μl TUNEL reaction mixture, each sample was incubated with 50 μl TUNEL reaction mixture at 37°for 60 min and the slides rinsed three times with PBS. After drying, the sample was incubated with 50 μl converter-peroxidase (POD) at 37°C for 30 min and slides were rinsed three times with PBS. Then, 50 μl diaminobenzidine (DAB) substrate was added and the sample incubated for a further 10 min at 20°C before again rinsing slides three times with PBS. Omission of the TdT enzyme in the TUNEL reaction was used as a negative control and resulted in no staining. Apoptosis was evaluated by computer-assisted image analysis system (LEICA QUIPS, LEICA Imaging Systems LTD, England) and the results were calculated as the number of positive-staining nuclei per 1,000 cells. For these counts, 2,000 cells were randomly selected from each specimen [28].

Immunohistochemical staining
The infiltration of macrophages was assessed using immunohistochemical assays. Sections (2 μm) were dewaxed, incubated with 3% H 2 O 2 , blocking serum, and thereafter with a polyclonal antibody against CD68 (Wuhan goodbio technology CO., LTD, GB13067-1, China) at 1:500 dilution. The sections were rinsed with TBST and biotinylated secondary antibody against rabbit IgG (KPL, 074-1506) for CD68 in a 1:200 dilution. After rinsing with TBST, the sections were incubated with HRP-conjugated streptavidin solution (Dako). HRP labeling was detected using a peroxide substrate solution with 0.8 mmol/L DAB and 0.01% H 2 O 2 . The sections were counterstained with hematoxylin before being examined under a light microscope. Image Pro Plus 6.0 software was used to transfer the interesting area staining density to an integrated optical density (IOD) which reflected the staining level of the area of interest [29][30][31].

Measurement of IL-1β and TNF-α in heart tissue and serum
Mice were anesthetized with an intraperitoneal injection of sodium pentobarbital (100 mg/kg). Blood samples and myocardial tissues were collected immediately. To measure cytokine, approximately 50 mg of cardiac tissue was transferred to a tube, homogenized in 1,000 μl Phosphate Buffer Solution, and then centrifuged at 3,500 rpm for 15 min at 4°C temperature, after which the upper layer was collected for further analysis. Blood samples were centrifuged (1,500 rpm for 10 min) and then supernatants were stored as serum at − 80°C used for analysis. IL-1β and TNF-α were measured by using commercially available enzyme-linked immunosorbent assay (ELISA) kits.

Measurement of IL-1β and TNF-α in the supernatant of primary peritoneal macrophages
The concentrations of cytokine in the supernatant were measured by ELISA. Briefly, three mice were injected with 1 ml of thioglycollate broth (3% w/v) (Sigma, 70157) days before harvesting peritoneal macrophages by washing the cavities with 4 ml of PBS for two times. Four hours later, 24-well culture plates were washed with 1640 medium to remove lymphocytes. Peritoneal macrophages were resuspended in RPMI 1640 cell culture medium cultivated in plates for 24 h in a 37°C humidified incubator to allow macrophages to attach to the surface [32]. Macrophages were incubated with high glucose (33 mM), or high glucose + LPS (1 μg/ml) [29,33], or high glucose + LPS + glibenclamide (100 μM). Cells incubated without drug treatment served as control. The supernatants were collected 24 h later, and IL-1β and TNF-α were measured.

Western blot analysis
The lysates (50 μg of protein) were boiled for 10 min and electrophoresed on a 12% SDS-PAGE electrophoresis. The membranes were incubated in PBS containing 5% non-fat dry milk for 4 h at 25°C. The blots were then incubated for 4 h at 25°C with primary antibodies for Nalp3 (1:1,000; Abcam, HK) and Caspase-1 (1:200; SANTA CRUZ, USA), and then incubated with IRDye 800CW-conjugated goat anti-rabbit secondary antibody (1:10,000; Rockland, USA) for 1 hour at 25°C. The infrared fluorescence image was obtained using Odyssey infrared imaging system (Li-Cor Bioscience, Lincoln, NE), and the band were quantified by Image J software (NIH, USA).

Statistical analysis
Data are presented as mean ± SEM and analyzed by ANOVA followed by the Tukey-Kramer multiple comparisons test when appropriate. P < 0.05 was considered statistically significant.

Glibenclamide attenuated myocardial injury by LPS in STZ-mice
Histological analyses were performed 6 h after LPS stimulation. STZ + LPS mice displayed the most serious myocardial injury, including irregular arrangement, degeneration of cardiocytes and rupture of myocardial fibers. Glibenclamide treatment attenuated myocardial injury in STZ + LPS mice. Consistent with the results from light microscopy, transmission electron microscopic analysis showed similar phenomena: the mitochondria of cardiocytes in STZ + LPS mice showed vacuolization and irregular swelling, accompanied by partial myofibers dissolving. Glibenclamide treatment relieved all these changes (Figure 2, n = 3). In addition, in our study, LPS injection caused decreases of heart rate and mean arterial pressure, and elevated serum epinephrine level in mice (Additional file 2: Figure S2, n = 5).

Discussion
The major findings of this study are as follows: (i) glibenclamide attenuated LPS-induced myocardial injury in STZ-mice; (ii) glibenclamide reduced serum IL-1β and TNF-α induced by LPS in STZ-mice; (iii) glibenclamide inhibited Nalp3 and Caspase-1 expression by LPS + high glucose stimulation in cultured primary peritoneal macrophages. These together suggested that glibenclamide might protect against myocardial injury under inflammation in diabetes.
Sepsis is a common disease with a growing morbidity around the world. Severe sepsis and systemic inflammation are the leading causes of mortality in critically ill patients, resulting from a systemic oxidative-mediated inflammatory response to severe bacterial infection [34]. It is well known that acute infections lead to difficulty in controlling blood glucose and that infection is the most frequently documented cause of ketoacidosis during diabetes mellitus [35]. In this study, the endotoxemic mice Figure 5 Glibenclamide did not alter expression of IL-1β and TNF-α in cardiac tissue by LPS stimulation in streptozotocin (STZ) diabetic mice. Expression of IL-1β and TNF-α in cardiac tissue were assessed by ELISA 6 h after LPS injection (15 mg/kg, i.p). In non-STZ treated mice, LPS stimulation significantly increased both IL-1β and TNF-α level. In STZ treated mice, LPS stimulation induced greater levels of IL-1β and TNF-α than those in non-STZ treated ones, which were not modified by glibenclamide (5 mg/kg, i.g, × 14 d) pretreatment. *P < 0.05, **P < 0.01 vs control, # P < 0.05 vs LPS. Values are means ± SEM (n = 8 per group).

Figure 6
Glibenclamide inhibited serum IL-1β and TNF-α levels by LPS stimulation in streptozotocin (STZ) diabetic mice. Serum IL-1β and TNF-α levels were assessed by ELISA 6 h after LPS injection (15 mg/kg, i.p). In non-STZ treated mice, LPS stimulation significantly increased both IL-1β and TNF-α levels. In STZ treated mice, LPS stimulation induced greater levels of IL-1β and TNF-α than those in non-STZ treated ones, which were significantly inhibited by glibenclamide (5 mg/kg, i.g, × 14 d) pretreatment. *P < 0.05, **P < 0.01 vs control; # P < 0.05, ## P < 0.01 vs STZ + LPS, † P < 0.05 vs LPS. Values are means ± SEM (n = 8-10 per group). displayed hypotension, decreased heart rate, and elevated serum epinephrine level after LPS stimulation, which was consistent with previous reports [24,36,37]. Prevention or management of sepsis remains a barrier to the successful care of many surgical and traumatic patients with diabetes, needing novel therapies urgently [38]. Type 1 diabetes mellitus is a chronic, multifactorial autoimmune disease that involves the progressive destruction of pancreatic β-cells, ultimately resulting in the loss of insulin production and secretion [39]. This response includes the production of cytokines such as IL-1β that orchestrate the recruitment of inflammatory cells to the islets and mediate direct cytotoxic effects on βcells [40]. TNF-α is considered to be a possible therapeutic target because it was up-regulated in multiple rodentobesity models and TNF-α blunted insulin signaling in insulin targeting tissues. It is not only exacerbates inflammatory response through acting as a signal amplifier to induce other inflammatory cytokines production, but also contributes to myocardial hypertrophy and fibrosis, leading to left ventricular remodeling and dysfunction [41].
Experimental and clinical studies have suggested that diabetic state causes a specific diabetic cardiomyopathy independent of vascular complications. This cardiomyopathy is characterized by myocyte hypertrophy and fibrosis and may be responsible for the high incidence of cardiac dysfunction and mortality [42]. Ventricles from diabetic patients show accumulation of glycoproteins, collagen, and active fibroblasts [43]. Elevated inflammatory cytokines have been found in circulation and in the hearts of diabetic patients, contributing to heart failure. Cardiac overexpression of TNF-α has been associated with cardiac hypertrophy and fibrosis, as well with left ventricular dysfunction [44]. Moreover, up-regulation of inflammatory cytokines and chemokines by subcutaneous injection of LPS was significantly more rapid and more pronounced in the diabetic mice compared with normal mice [45]. These indicate that cardiac tissue is sensitive to inflammation and inflammatory response is stronger in diabetic condition. We observed significant elevations of IL-1β and TNF-α levels in serum and cardiac tissue upon LPS challenge in mice, suggesting that both of circulating and local inflammatory reaction may cause cardiac injury under diabetic and septic condition. In addition, we found that glibenclamide was able to inhibit the inflammatory cytokines secreted by peritoneal macrophages after LPS treatment. Interestingly, glibenclamide only inhibited the increases of IL-1β but not TNF-α in vitro. We considered this specific inhibition on IL-1β level by glibenclamide implicated the involvement of Nalp3, which regulates IL-1β but not TNF-α. There was an inconsistency of TNF-a data in serum and in the supernatant. We think that the situation in vivo is rather complex while the cell model in vitro is simple. The serum TNF-α levels might be influenced by many factors. Glibenclamide may lower the serum TNFα level in a macrophage-independent manner.
K ATP channels are a type of Kir constituted by heteromultimers of two kinds of proteins. Each channel is formed from four pore-forming Kir subunits (Kir 6.1 or Kir 6.2) complexes with four regulatory sulfonylurea receptor proteins (SUR1 in neuronal/pancreatic beta cells or SUR2 in cardiovascular cells). Glibenclamide is a sulfonylurea drug which binds to the SUR1 domain with 10 to 500 fold higher affinity than to the SUR2 domains [6]. Previous study demonstrated that glibenclamide was able to suppress NALP3 activation independently of K ATP [7]. Thus, in this work, we used glibenclamide as an inhibitor of NALP3.
The mechanism of glibenclamide attenuated LPSinduced inflammation is not fully elucidated. Transient receptor potential melastatin 4 (TRPM4), a calciumactivated non-selective cation channel, is functionally expressed in the heart. TRPM4 has been linked to diverse physiological functions, such as protection against Ca 2+ overload by cell membrane depolarization, modulation of Ca 2+ oscillations controlling cytokine production in T lymphocytes and mast cells, and dendritic cell migration [46,47]. Grand et al. reported that TRPM4 is inhibited by glibenclamide, a modulator of ATP binding cassette proteins (ABC transporters), such as the cystic fibrosis transmembrane conductance regulator (CFTR) [48]. It was also reported that TRPM4 inhibitors 9-phenanthrol and glibenclamide could attenuate LPS-induced endothelial cell death and hypoxia and reoxygenation-induced early after depolarizations [49]. Glibenclamide is a potent blocker of the ATP-modulated K + channel in insulin secreting cells and a broadly used antidiabetic drug. Interestingly, recent study showed that glibenclamide could decrease TNF-α and NF-κB activation after subarachnoid hemorrhage [50]. Moreover, glibenclamide reduced LPS-induced release of IL-1β、TNF-α and PAI-2 mRNA in a concentration-dependent manner through reducing the calcium entry by drug-induced depolarization of hypoxic monocytes in an ex vivo model of human endotoxinaemia under hypoxaemic conditions [12]. Koh et al. found that glibenclamide directly reduced the secretion of IL-1β by bone-marrow-derived macrophages in a dose-dependent fashion [51]. In the present study, we further demonstrated that glibenclamide could decrease serum IL-1β and TNF-α level induced by LPS in STZ-induced diabetic mice.
Nalp3 is a 'general sensor' for danger signals, representing an important caspase-1-containing inflammasomes and is activated by various pathogens [18], and damage-associated molecules and environmental irritants [40]. Activation of Nalp3 leads to oligomerization and recruitment of apoptosis-associated speck-like protein and pro-caspase-1, with auto-cleavage and activation of Caspase-1. Active Caspase-1 cleaves pro-IL-1β to active IL-1β, which, when secreted, can exert direct cytotoxic effects as well as recruit other inflammatory cells. In this work, we found that glibenclamide could inhibit Nalp3 inflammasomes and Caspase-1 induced by LPS + high glucose stimulation in cultured peritoneal macrophages, suggesting that the effects of glibenclamide on prevention of serum IL-1β and TNF-α might be related to inhibiting Nalp3 inflammasomes and Caspase-1. The mechanism how glibenclamide inhibits NALP3 inflammasomes and suppresses IL-1β secretion is an intriguing question. Lamkanfi et al. demonstrated that glibenclamide could inhibit the assembly of the NALP3 inflammasomes in response to stimulation with lipopolysaccharide (LPS) and adenosine triphosphate (ATP) [13]. The ability to Figure 8 Glibenclamide inhibited Nalp3 and Caspase-1 expression by high glucose and LPS stimulation in primary peritoneal macrophages. Nalp3 and Caspase-1 in macrophages were measured by western blotting. Macrophages were incubated for 24 h under high glucose (33 mM), or high glucose + LPS (1 μg/ml), or high glucose + LPS + glibenclamide (100 μM). Cells incubated without drug treatment served as control. High glucose stimulation significantly increased Caspase-1 expression. Nalp3 and Caspase-1 levels were markedly increased by high glucose + LPS challenge. Caspase-1 level was further increased by high glucose + LPS compared with LPS group. Both proteins were significantly inhibited by glibenclamide treatment. *P < 0.05, **P < 0.01 vs control; # P < 0.05, ## P < 0.01 vs Glucose + LPS; † † P < 0.01 vs LPS. Values are means ± SEM (n = 3 per group).
suppress NALP3 activation of glibenclamide was independent of its inhibitory effect on K ATP [13]. The authors speculated that glibenclamide acts the upstream of NALP3 and the downstream of P2X7 [13]. However, the precise molecular target of glibenclamide for its inhibitory effect on NALP3 inflammasome has yet to be identified.

Conclusions
In conclusion, our findings indicate that glibenclamide could ameliorate myocardial injury and reducing IL-1β and TNF-α possibly through inhibiting Nalp3 inflammasomes and Caspase-1 signaling under LPS-induced endotoxemia in STZ diabetic mice.

Additional files
Additional file 1: Figure S1. Glibenclamide did not modify body weight and blood glucose in normal mice. Compared with control, glibenclamide (5 and 20 mg/kg, i.g, ×14 d) administration did not change body weight and blood glucose. NS means no significant difference. Values are means ± SEM (n = 5 per group).
Additional file 2: Figure S2. Injection of LPS caused a decrease in heart rate and mean arterial pressure, and elevated serum epinephrine level in mice. Six hours after LPS (15 mg/kg) injection (i.p), heart rate (A) and mean arterial pressure (B) were measured by MPA-HBBS software. Serum epinephrine (C) level was assessed by ELISA. Mice were anesthetized with chloral hydrate (4%), heart rate and mean arterial pressure of endotoxemic mice decreased during LPS stimulation (heart rate and mean arterial pressure had been measured for 10 minutes). Serum epinephrine level was significantly increased in endotoxemic mice compared with control. *P < 0.05, **P < 0.01. Values are means ± SEM (n = 5 per group).