Nicorandil Attenuates LPS-Induced Acute Lung Injury by Pulmonary Endothelial Cell Protection via NF-κB and MAPK Pathways

Acute lung injury (ALI) is a devastating critical disease characterized by diffuse inflammation and endothelial dysfunction. Increasing evidence, including from our laboratory, has revealed that the opening of ATP-sensitive potassium (KATP) channels has promising anti-inflammation and endothelial protection activities in various disorders. However, the impacts of KATP channels on ALI remain obscure. In this study, we used nicorandil (Nico), a classic KATP channel opener, to investigate whether opening of KATP channels could alleviate ALI with an emphasis on human pulmonary artery endothelial cell (HPAEC) modulation. The results showed that Nico inhibited lipopolysaccharide- (LPS-) induced inflammatory response, protein accumulation, myeloperoxidase activity, and endothelial injury. In vitro, Nico reduced LPS-induced HPAEC apoptosis and the expression of cleaved-caspase-3, caspase-9, and CCAAT/enhancer-binding protein homologous protein (CHOP). Additionally, Nico inhibited inflammation by suppressing monocyte-endothelial adhesion and decreasing the expression of proinflammatory proteins. Moreover, Nico restored the expression and the distribution of adherens junction vascular endothelial- (VE-) cadherin. Further, Nico abolished the increase in intracellular reactive oxygen species (ROS) and the activation of NF-κB and mitogen-activated protein kinase (MAPK) in HPAECs. Glibenclamide (Gli), a nonselective KATP channel blocker, abrogated the effects of Nico, implying that opening of KATP channels contributes to the relief of ALI. Together, our findings indicated that Nico alleviated LPS-induced ALI by protecting ECs function via preventing apoptosis, suppressing endothelial inflammation and reducing oxidative stress, which may be attributed to the inhibition of NF-κB and MAPK signaling pathways.


Introduction
Acute lung injury (ALI) is a life-threatening disease with high incidence and mortality in critically ill patients [1]. Bacterial infections are the leading cause of ALI and may progress towards acute respiratory distress syndrome (ARDS), regarded as the severe form of ALI. It is generally agreed that the uncontrolled inflammation, endothelial injury, and oxidative damage participate in the pathogenesis of ALI, causing injured alveolar-capillary barrier, increased lung permeability, deficient gas exchange, and ultimately respiratory failure [2]. Although antibiotic therapy and intensive care are developing, the prognosis of ALI remains poor [3]. Herein, there is an urgent need for safe and effective therapeutic options in ALI treatment.
Recently, pulmonary endothelium has been considered as a key modulator and orchestrator of ALI. As a dynamic and active organ lining of the entire pulmonary vessel, adjacent endothelial cells (ECs) are tightly connected to each other via intercellular junctions, performing as a semipermeable barrier for the separation of pulmonary circulation from lung interstitium [4]. This barrier performs a regulatory role in lung homoeostasis by controlling the transport of fluids, proteins, and cells, as well as trafficking of inflammatory cells [5]. However, ECs are vulnerable to injury from noxious stimuli including circulating (e.g., lipopolysaccharide (LPS)), released (e.g. reactive oxygen species (ROS)), and physical factors, leading to barrier disruption and proinflammatory protein secretion [6]. Moreover, ECs play major roles in vascular remodeling by regulating the clotting system, vascular tone, and angiogenesis, which may determine the outcome of ALI [7]. Thus, the endothelial-targeted therapy may emerge as a promising strategy in ALI management.
ATP-sensitive potassium (K ATP ) channels, first identified in cardiac tissue, are a unique link between cellular metabolic status and electrical activity [8]. Recently, nicorandil (Nico), a classic K ATP channel opener (KCO), has been shown to exert strong endothelial protective and anti-inflammation potential. Recent evidence has indicated that Nico has therapeutic effects in pulmonary arterial hypertension (PAH) via attenuating vascular endothelial damage and inflammatory responses [9,10]. Nico inhibited neutrophil recruitment in carrageenan-induced pleurisy [11], alleviated oxidative stress in cyclophosphamide-induced lung injury [12], and decreased paw edema induced by carrageenan [13]. Additionally, K ATP channels have been found to be involved in numerous cellular processes, such as proliferation, apoptosis, migration, and angiogenesis. In vitro, Nico shows endothelial protective impacts when stimuli occur, such as hypoxia and high glucose [14,15]. Although Nico has received increasing attention in the treatment of several pulmonary diseases, it remains unclear whether it has benefits in ALI therapy.
In the current study, a murine model of ALI induced by intratracheal LPS administration and cultured human pulmonary artery endothelial cells (HPAECs) were applied to determine the efficacy of Nico and its underlying potential cellular and molecular mechanisms.

Materials and Methods
2.1. Animal. Specific pathogen-free male C57B/L mice (20-22 g) were obtained from Cavans Laboratories (Changzhou, China). The mice were kept in a temperature-controlled room under a 12 hr dark/light cycle and were provided with food and water ad libitum. All experimental protocols were performed according to the guidelines of the National Institutes of Health and Nanjing Medical University, and all procedures were approved by the Institutional Animal Care and Use Committee of Nanjing Medical University (Nanjing, China).

LPS-Induced ALI in Mice.
After anesthesia, the mice received an intratracheal instillation of LPS (5 mg/kg, Sigma-Aldrich, St. Louis, MO, USA) dissolved in 50 μL of normal saline or saline alone as the control. To investigate the preventive effects, the mice were treated with Nico (12.5, 25, and 50 mg/kg) (Sigma-Aldrich) for 3 consecutive days by gavage. On day 3, the mice were administrated with Nico at 1 hr before LPS challenge or were treated with Nico (50 mg/kg) at 1 hr before saline administration (n = 6 − 8 for each group). The mice were sacrificed at 24 hr after LPS administration (Figure 1). Bronchoalveolar lavage fluid (BALF) was collected by lavage with ice-cold phosphatebufferedsaline (PBS, 400 μL for 3 times; 85-90% of the lavage volume was recovered) via a tracheal catheter. After centrifugation, the supernatants were frozen at -80°C for further experiments. The lung lobes were collected for further analysis.
2.3. Lung Histology. The lung tissues were placed in 4% paraformaldehyde fixative for paraffin embedding. Then, a series of microsections (5 μm) were cut with a microtome and stained with hematoxylin and eosin (H&E). For each section, at least eight random areas were selected by a light microscope (DM2500, Leica, Wetzlar, Germany). The degree of lung injury was assessed by a semiquantitative scoring system.

Measurement of Protein Content and Myeloperoxidase
(MPO). Protein concentrations in BALF were determined by BCA Protein Assay Kit (Beyotime, Nantong, China). MPO activities were measured by a commercial test kit (Jiancheng Bioengineering Institute, Nanjing, China) in both BALF and lung homogenate. All the procedures were performed according to the manufacturer's protocol.
2.5. Cell Culture. HPAECs were purchased from Sciencell Research Laboratories (Carlsbad, CA, USA). HPAECs were cultured in endothelial cell medium-2 (ECM-2) (Sciencell, USA), consisting of 5% fetal bovine serum (FBS), 1% penicillin, streptomycin, and 1% endothelial cell growth factors. Cells were used in their fourth to sixth passages. After starvation for 6 hr, cells were pretreated with Nico for 1 hr before stimulation with LPS (1 μg/mL) for indicated time intervals. To investigate the interaction between KCOs and K ATP blockers on HPAECs, glibenclamide (Gli), a nonselective K ATP blocker, was added 30 min prior to the addition of other drugs at the concentration of 10 μmol/L. Gli was purchased from Sigma-Aldrich. Nico and Gli were diluted with dimethyl sulphoxide (DMSO), and LPS was diluted with double-distilled H 2 O. The concentration of DMSO was controlled below 0.1% (vol/vol) and did not decrease the biological viability of the cells. The human monocytic leukemia cell line U-937 cells were purchased from the Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences (Shanghai, China). The U-937 cells were cultured in RPMI 1640 (Gibco-BRL, Grand Island, USA) containing 10% FBS. HPAECs and U-937 cells were incubated at 37°C in a humidified incubator with 5% CO 2 .
2.6. Cell Viability Assay. Cell counting kit-8 (CCK8) assay was conducted to analyze the cytotoxicity of Nico on HPAECs. HPAECs were seeded into a 96-well plate at a density of 10,000/well overnight. Then, cells were treated with Nico (0. 2.11. ELISA. Concentrations of TNF-α and IL-1β in the lung tissues were measured with murine cytokine-specific ELISA kits (R&D Systems, Abingdon, UK). In brief, the lung homogenate was centrifuged at 3000 rpm for 20 min at 4°C, and the supernatant was analyzed using ELISA kits according to the manufacturer's instructions. The level of NADPH oxidase 4 (Nox4) in HPAECs was measured. The cells were treated as described above, and the ELISA was conducted according to the manufacturer's protocols (Yifeixue Biotechnology, Nanjing, China).

Determination of Intracellular ROS Production.
Intracellular ROS were detected using the 2′,7′-dichlorofluorescin diacetate (DCFH-DA) assay with a laser scanning confocal microscope (Leica, Wetzlar, Germany). Briefly, HPAECs were seeded into each well of a 24-well plate, cultured for 24 hr, and pretreated with Nico (100 μM) with or without Gli (10 μM) before exposure to LPS (1 μg/mL) for 1 hr. The cells were then incubated with 10 μM DCFH-DA for 30 min at 37°C in the dark. After being fixed with 4% paraformaldehyde for 10 min, the cells were washed three times with PBS before being photographed. For each culture, a minimum of 5 random fields was captured with an excitation setting of 488 nm.
2.14. Statistical Analysis. All data were presented as mean ± SEM. Comparisons between controls and samples treated with different drugs were determined by one-way analysis of variance (ANOVA). Least significant difference (LSD) post hoc tests were used to assess significant effects. P < 0 05 was considered statistically significant.  (Figure 2(a)). However, these effects of LPS were significantly inhibited by pretreatment with Nico in a concentration-dependent way.

Nico
To detect the role of Nico in LPS-mediated lung injury, lung sections stained with H&E were used for detecting histological changes at 24 hr after LPS challenge. Compared with the control group, the mice treated with LPS showed increased severity of lung injury correlated with greater diffuse alveolar damage, alveolar wall thickening (Figures 2(b) and 2(c)), and elevated BALF protein (Figure 2(d)). Pretreatment with Nico (25 mg/kg and 50 mg/kg) alleviated LPS-induced lung injury (Figures 2(b) and 2(c)) and significantly suppressed the increase of protein level in BALF after LPS challenge (P < 0 05) (Figure 2(d)), while 12.5 mg/kg Nico treatment did not result in such significant changes.

Nico Ameliorated LPS-Induced Lung Inflammation.
MPO activities in BALF and lung homogenate were detected to assess neutrophil infiltration in LPS-induced ALI. Compared with the LPS group, administration of Nico (25 mg/kg and 50 mg/kg) induced remarkable decreases in MPO activities (P < 0 05), while 12.5 mg/kg Nico (12.5 mg/kg) did not cause notable decreases (Figures 2(e) and 2(f)).

Nico Alleviated EC Injury in LPS-Challenged Mice.
To investigate the effects of Nico on LPS-induced EC injury in ALI, endothelial related proteins eNOS, iNOS, and VEcadherin were detected. Although the expression of iNOS was upregulated in the LPS group, Nico dose-dependently prevented the change (P < 0 05) (Figures 3(a) and 3(c)). As shown in Figures 3(a) and 3(d), western blotting analysis revealed significant downregulation of eNOS and VEcadherin in the lungs of LPS-challenged mice. Nico at the concentration of 25 mg/kg and 50 mg/kg alleviated LPSdecreased expression of VE-cadherin and eNOS in the lungs of mice (P < 0 05).

Nico Decreased LPS-Induced HPAEC Apoptosis.
To evaluate the possible cytotoxic potential of Nico (0.1, 1, 10, and 100 μM), cellular viability was determined on HPAECs. The results showed that there was no cellular toxicity induced by Nico between the concentrations of 0.1 μM and 100 μM (Figure 4(a)).
As indicated in Figure 4(b), apoptotic nuclei with chromatin condensation and nuclear fragmentation were more easy to be dyed and showed brilliant blue fluorescence, while nonapoptotic nuclei showed light blue fluorescence. Statistical analysis demonstrated that pretreatment with Nico (100 μM) alleviated LPS-induced HPAEC apoptosis. Moreover, the inhibitory effect of Nico on LPS-induced cell apoptosis could be abolished by Gli (P < 0 05). The findings were further confirmed by western blotting analysis. As shown in Figure 4(c), Nico potently decreased LPS-induced expression of c-caspase-3 and caspase-9, critical executioners of apoptosis, and suppressed the proapoptotic transcription factor CHOP. Gli blocked the beneficial effects of Nico (P < 0 05). Nico and Gli themselves did not trigger HPAEC apoptosis.

Nico Restrained LPS-Induced Proinflammatory Factor
Expression in HPAEC. Pulmonary ECs have been reported to release cytokines and adhesion molecules, leading to an inflammatory process and activation of leukocytes in the pathogenesis of ALI [18]. To further substantiate the anti-inflammatory mechanism of Nico, we studied its effects on LPS-induced expression of proinflammatory mediators in HPAECs. The results showed that Nico markedly blocked LPS-induced upregulation of TNF-α and partially decreased the LPS-induced COX-2 expression in mRNA level (P < 0 05) (Figure 5(a)). Additionally, Nico suppressed the LPS-induced increased expression of iNOS, while Nico recovered LPS-downregulated expression of eNOS in the protein level (P < 0 05) ( Figure 5(b)). Pretreatment of Gli, a nonselective K ATP blocker, eliminated the inhibitory effects of Nico on LPS-induced inflammatory responses (P < 0 05) ( Figure 5(b)). Nico itself could slightly induce the expression of eNOS, even though no statistical significance was observed (P = 0 072).

Nico Suppressed LPS-Induced Adhesion Molecule
Expression and Adhesion between Endothelial Cells and Monocytic Cells. HPAECs were treated with LPS for 12 hr, and then monocyte adhesion to ECs was determined. Compared to the untreated control cells, monocytic adhesion to LPS-stimulated ECs was significantly increased (P < 0 05) (Figure 6(a)). Nico pretreatment resulted in reduced adhesion of monocytes to HPAECs compared to the LPS-treated cells (P < 0 05) (Figure 6(a)). Moreover, Gli inhibited Nicoalleviated adhesion of monocytes to HPAECs (P < 0 05) (Figure 6(a)).
Expression of adhesion molecules on the surface of ECs is important for the association of monocytes in circulation. We detected the effects of Nico in regulating the expression of ICAM-1 and VCAM-1, two critical adhesion molecules, in HPAECs in response to LPS. Upon stimulation with LPS, the expression of ICAM-1 and VCAM-1 was significantly increased in HPAECs (P < 0 05) (Figure 6(b)). Pretreatment with Nico showed a reduction of ICAM-1 and VCAM-1 expression (P < 0 05) (Figure 6(b)). The addition of Gli abolished the effects of Nico (P < 0 05) (Figure 6(b)).

Nico
Recovered LPS-Induced HPAEC Loss of VE-Cadherin. VE-cadherin is the major component for the adherens junctions (AJs) between ECs, which plays a critical role in controlling vascular endothelial function and endothelial permeability [19]. Herein, the expression and distribution of VE-cadherin were investigated. The results from western blotting analysis showed that the expression of VE-cadherin was notably decreased by LPS challenge for  Caspase-9/훽-actin 24 hr. Pretreatment with Nico recovered the LPS-induced downregulation of VE-cadherin, while Gli could abolish the effects (P < 0 05) (Figure 7(a)). Besides, immunofluorescence staining confirmed that Nico protected HPAECs from LPS-induced loss of VE-cadherin at cell-cell junctions (Figure 7(b)). Gli abolished Nico-induced beneficial effects in HPAECs.

Nico Diminished LPS-Induced Oxidative Stress in HPAEC.
Previous studies have suggested a strong association with oxidative stress and ALI. As shown in Figure 8(a), LPS apparently promoted ROS production, while pretreatment with Nico decreased the intracellular ROS level. Gli inhibited the effect of Nico on regulating ROS production. We further investigated the mechanisms by which Nico blocked oxidative stress. As the most ubiquitous of NAD(P)H oxidases, Nox4 is a major source of LPS-induced ROS generation in ECs [20]. Since Nico is known to primarily bind to mitochondrial K ATP channel, we hypothesize that Nico may enhance mitochondrial antioxidative activity. Thus, MnSOD as a major antioxidant in mitochondria was detected. As shown in Figures 8(b) and 8(c), we found that the Nox4 level was significantly decreased after Nico treatment compared with the LPS group by the methods of western blotting and ELISA (P < 0 05). Additionally, LPS led to increased expression of MnSOD compared to the control group (P < 0 05). However, Nico pretreatment induced a higher expression of MnSOD compared to the LPS group (Figure 8(b)). The results were further confirmed by immunofluorescence staining (Figures 8(d) and 8(e)). The data suggested that Nico may prevent LPS-induced oxidative stress by regulating the key antioxidant systems in mitochondria.

Nico Suppressed LPS-Induced Activation of NF-κB and MAPK Pathways in HPAEC.
It is widely accepted that NF-κB pathway is crucial for LPS-induced inflammation in pulmonary ECs. Therefore, the phosphorylation of p65 and iκB-α was examined. The data confirmed the effects of LPS on activating the NF-κB p65 pathway in HPAECs within 30 min after stimulation. The activation was suppressed in the Nico-pretreated group (P < 0 05) (Figure 9(a)). Additionally, immunofluorescence staining showed that Nico inhibited LPS-induced NF-κB p65 nuclear translocation (Figure 9(b)). Pretreatment with Gli could block the alternation elicited by Nico.
The activation of MAPKs is involved in the regulation of oxidative stress and inflammatory responses. As shown in Figure 9(c), LPS trigged significant activation of p38, JNK, and ERK within 30 min after stimulation (P < 0 05) (Figure 9(c)). Pretreatment of Nico could blunt the activation of p38 and JNK induced by LPS (P < 0 05), but not the activation of ERK (Figure 9(c)). Gli could reverse the  suppression elicited by Nico, while no alternation was observed with Nico and Gli alone.

Discussion
It is well-known that ALI is featured as loss of alveolarcapillary membrane integrity, sustained inflammation, and excessive oxidative stress. As key components of alveolarcapillary barrier, lung ECs might be the first target of circulating soluble factors and inflammatory cells. The apoptosis of ECs causes loss of endothelial integrity, leading to increased lung microvascular permeability, alveolar edema, and leukocyte extravasations during ALI [2]. Mounting studies have confirmed that LPS treatment induces EC apoptosis. In the present study, exposure to LPS increased EC apoptosis, with the enhanced expression of c-caspase-3 and caspase-9, two critical proteases in programmed cell death, and CHOP, one vital transcription factor in endoplasmic reticulum stress-(ERS-) induced apoptosis. Activation of K ATP channels by Nico blunted LPS-induced HPAECs apoptosis.
Recently, CHOP has been regarded as a multifunctional molecule involved in apoptosis as well as in metabolic and inflammatory processes [21]. Dong et al. showed that the opening of K ATP channels attenuated astrocyte injury and inflammation via suppressing ERS including the expression of CHOP [22]. Sargsyan et al. reported that activation of K ATP channels improved β-cell function via reducing the expression of CHOP [23]. Our data showed that Nico decreased LPS-induced expression of CHOP, indicating that ERS may be involved in the protective effects of KCOs. Persistent inflammation is another feature of ALI, characterized by inappropriate activity of inflammatory cells and substantial release of inflammatory mediators. In response to LPS, pulmonary endothelium becomes activated resulting in increased expression of adhesion molecules, excessive production of cytokines, and uncontrolled interaction with inflammatory cells [24]. In our in vivo study, the activation of K ATP channels by Nico decreased LPS-induced cytokine release and effectively suppressed inflammatory cell infiltration. In our in vitro study, it showed that Nico pretreatment reduced the increased expression of inflammatory cytokines induced by LPS, whereas these effects partially disappeared upon treatment with the K ATP channel inhibitor. Moreover, adhesion molecules play vital roles in regulating recruitment of leukocytes into the endothelium [25]. Our findings indicated that K ATP channel could regulate the expression of adhesion molecules, as well as monocyteendothelial adhesions. Uncontrolled oxidative stress contributes to the development of ALI. LPS-induced overproduction of ROS, such as superoxide anion and hydrogen peroxide, can exacerbate tissue damage and inflammatory response [26]. Previous evidence has revealed that several antioxidant agents, such as vitamin D [27], propofol [28], and olmesartan [29], prevent ECs from injury induced by excessive oxidative stress. Thus, inhibition of oxidative stress may be one of the mechanisms to alleviate ALI. In ECs, Nox4 is a highly expressed NADPH oxidase among seven homologs (Nox1, Nox3, Nox4, Nox5, Duox1, and Duox2) of gp91phox (Nox2) which can directly affect the production of inflammatory cytokines [30]. Our results showed that LPS notably enhanced the production of ROS with an elevation of Nox4 expression, whereas Nico attenuated the oxidative stress in HPAECs. Interestingly, it showed that LPS upregulated the expression of MnSOD, one of the critical antioxidant enzymes, which may be a possible self-adjusting mechanism of HPAECs to withstand oxidative stress. It is generally accepted that LPS induces ROS production, which activates ECs causing overexpression of proinflammatory factors. However, these proinflammatory proteins may in turn exacerbate inflammatory responses, either direct or indirect participation in oxidative stress. Herein, there is a vicious feedback loop occurring due to cytotoxic activities of ROS during ALI. Our findings suggested that the anti-inflammation effects of K ATP channel activation may be partly due to its ROS scavenging activity, leading to a reduction in LPS-induced oxidative damage.
It is recognized that AJs are essential parts in paracellular pathways, and their disruption leads to interendothelial adhesion breakdown and endothelial barrier hyperpermeability. VE-cadherin is a major component of AJs and is used to correlate the destruction or enhancement of the endothelial monolayer barrier function [31]. LPS has been shown to induce the abnormal distribution and decrease expression of VE-cadherin [32]. Correspondingly, we found that LPS significantly decreased the expression of VE-cadherin in vivo and in vitro; besides, activation of K ATP channels may maintain the barrier function of ECs by restoring interendothelial AJs. A previous study showed that Nico preserved endothelial junctions of VE-cadherin by decreasing endothelin-1 in acute myocardial infarction and reperfusion model [33]. Additionally, VE-cadherin is a calciumdependent cell-cell adhesion molecule, while it is widely accepted that KCOs participate in calcium signaling modulation [34,35]. Thus, it is possible that inflammatory responses and calcium signaling may be involved in the regulation of K ATP channels in AJ expression and distribution.
Both NF-κB and MAPK pathways are involved in the pathology of ALI. When stimulated with LPS or ROS, NF-κB signaling cascade becomes activated by degradation and phosphorylation of iκB-α as well as nuclear translocation and phosphorylation of NF-κB p65 [36,37]. Additionally, inhibition of MAPK pathways, including p38, ERK, and JNK, can alleviate the transcription of proinflammatory mediators and oxidative stress [38]. Studies have shown that activation of K ATP channels prevents ECs from hypoxiainduced apoptosis through p38 MAPK and NF-κB pathways [14,39] and ameliorated excess oxidative stress through NF-κB signaling [40]. In terms of the current study, it suggested that the endothelial protection of K ATP channels is related to NF-κB and MAPK signaling pathways.
Our findings have confirmed that Nico suppresses LPSinduced ALI via endothelial protective effects. However, a previous study showed that endothelium may not be the target for KCOs to exert vasorelaxant effects [41]; accumulating data, including ours, have shown that K ATP channels are critical in regulating EC functions to shear stress and arterial vasomotion [42,43]. In mice, knockout of endothelial specific Kir6.1 leads to impaired vasorelaxation during hypoxia in the coronary circulation [44]. In HPAECs, KCOs are capable of reducing endothelial apoptosis, promoting the release of nitrogen oxide and preventing EC dysfunction from ischemia and hypoxia, which could be blocked by K ATP channel blockers [9,45,46]. Although the present study showed that K ATP channels played important roles in regulating EC functions, the mouse model with knockout of endothelial specific K ATP channels may be used for further studies to demonstrate the protective effects of K ATP channels on ECs in ALI.

Conclusions
The present study suggested that Nico attenuated LPSinduced ALI through preventing EC apoptosis, decreasing oxidative stress, and suppressing endothelial-mediated inflammation. These protective effects of Nico may be attributed to the activation of K ATP channels and its downstream NF-κB and MAPK signaling pathways. Taken together, our results provide direct evidence that Nico might be a candidate for the adjuvant therapy for ALI patients.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
The authors confirm that there are no conflicts of interest.