Glycolipids from spinach suppress LPS-induced vascular inflammation through eNOS and NK-κB signaling

Biomedicine & Pharmacotherapy 91 (2017) 111–120

Available online at

ScienceDirect www.sciencedirect.com

Glycolipids from spinach suppress LPS-induced vascular inflammation through eNOS and NK-kB signaling Masakazu Ishiia,* , Tatsuo Nakaharab , Daisuke Arahob , Juri Murakamia,c, Masahiro Nishimuraa a b c

Department of Oral and Maxillofacial Prosthodontics, Kagoshima University Graduate school of Medical and Dental Science, Kagoshima 890-8544, Japan Maruzen Pharmaceuticals Co., Ltd., Hiroshima 729-3102, Japan Department of Oral and Maxillofacial Surgery, Kagoshima University Graduate School of Medical and Dental Science, Kagoshima 890-8544, Japan

A R T I C L E I N F O

Article history: Received 11 January 2017 Received in revised form 7 April 2017 Accepted 13 April 2017 Keywords: Glycolipids Anti-inflammation Endothelial cells eNOS NF-kB

A B S T R A C T

Glycolipids are the major constituent of the thylakoid membrane of higher plants and have a variety of biological and pharmacological activities. However, anti-inflammatory effects of glycolipids on vascular endothelial cells have not been elucidated. Here, we investigated the effect of glycolipids extracted from spinach on lipopolysaccharides (LPS)-induced endothelial inflammation and evaluated the underlying molecular mechanisms. Treatment with glycolipids from spinach had no cytotoxic effects on cultured human umbilical vein endothelial cells (HUVECs) and significantly blocked the expression of LPS-induced interleukin (IL)-6, monocyte chemoattractant protein-1 (MCP-1), vascular cell adhesion molecule-1 (VCAM-1), and intracellular adhesion molecule-1 (ICAM-1) in them. Glycolipids treatment also effectively suppressed monocyte adhesion to HUVECs. Treatment with glycolipids inhibited LPS-induced NF-kB phosphorylation and nuclear translocation. In addition, glycolipids treatment significantly promoted endothelial nitric oxide synthase (eNOS) activation and nitric oxide (NO) production in HUVECs. Furthermore, glycolipids treatment blocked LPS-induced inducible NOS (iNOS) expression in HUVECs. Pretreatment with a NOS inhibitor attenuated glycolipids-induced suppression of NF-kB activation and adhesion molecule expression, and abolished the glycolipids-mediated suppression of monocyte adhesion to HUVECs. These results indicate that glycolipids suppress LPS-induced vascular inflammation through attenuation of the NF-kB pathway by increasing NO production in endothelial cells. These findings suggest that glycolipids from spinach may have a potential therapeutic use for inflammatory vascular diseases. © 2017 Elsevier Masson SAS. All rights reserved.

1. Introduction Atherosclerosis is the major cause of cardiovascular and cerebrovascular disease, the leading causes of morbidity and mortality worldwide [1,2]. Vascular inflammatory processes play an important role in the initiation and progression of atherosclerosis [3,4]. The inflammatory response in the endothelium promotes leukocyte adhesion and increases vascular permeability via elevated expression of several cell adhesion molecules, including vascular cell adhesion molecule-1 (VCAM-1), intracellular adhesion molecule-1 (ICAM-1), and E-selectin, and via the

* Corresponding author at: Department of Oral and Maxillofacial Prosthodontics, Kagoshima University Graduate School of Medical and Dental Science, 8-35-1 Sakuragaoka, Kagoshima 890-8544, Japan. E-mail address: [email protected] (M. Ishii). http://dx.doi.org/10.1016/j.biopha.2017.04.052 0753-3322/© 2017 Elsevier Masson SAS. All rights reserved.

release of pro-inflammatory cytokines, such as interleukin (IL)-6, IL-8, and monocyte chemoattractant protein-1 (MCP-1) [5]. It is well known that activation of nuclear factor (NF)-kB plays a critical role in the inflammatory response by regulating the expression of various inflammatory mediators including cytokines, chemokines, and adhesion molecules [6]. In the inactive state, NF-kB is present in the cytoplasm by binding to the inhibitor of NF-kB (IkB). Upon activation by external stimuli, IkB is degraded, thus allowing NF-kB to translocate into the nucleus and induce the expression of inflammatory genes. Activation of NF-kB is an important process in the development and progression of endothelial dysfunction and atherosclerosis [7]. Thus, the inhibition of NF-kB signaling in the endothelium may play an important role in preventing atherosclerosis. It is well established that nitric oxide (NO) is a critical regulator of vascular homeostasis. NO promotes arterial vasodilation and inhibits platelet aggregation, monocyte adhesion to the

112

M. Ishii et al. / Biomedicine & Pharmacotherapy 91 (2017) 111–120

endothelium, and smooth muscle cell proliferation [8]. A number of studies have shown that increased NO production in the endothelium contributes to NF-kB inactivation and attenuates atherosclerosis [9,10]. In endothelial cells, NO is produced by enzymatic reduction of L-arginine to L-citrulline via endothelial nitric oxide synthase (eNOS). Therefore, promotion of eNOS activation and NO production may be an effective approach to control atherosclerosis. However, it has been reported that NO production by eNOS or inducible NOS (iNOS) have different effect on pathogenesis in atherosclerosis [11]. An excessive NO produced from iNOS leads to detrimental effects on vascular function. Several epidemiological studies have demonstrated that diets rich in vegetables and fruits decrease the risk of developing cardiovascular diseases [12,13]. The beneficial effects have been attributed to a variety of natural bioactive molecules present in vegetables and fruits. It has been reported that several bioactive molecules such as isoflavones, lycopene, or quercetin have antiinflammatory effects on the endothelium and have a preventative and therapeutic activity on atherosclerosis [14–16]. Spinach (Spinacia oleracea) is a green leafy vegetable recognized as a highly nutrient rich food. It is known to contain large amounts of vitamins, minerals, and other bioactive compounds such as carotenoids [17]. Previous studies showed that several bioactive compounds from spinach present various pharmacological activities including anti-obesity, hypoglycemic, anti-tumor, and anti-oxidant activities [18,19]. Furthermore, spinach contains several glycolipids such as monogalactosyl diacylglycerol (MGDG), digalactosyl diacylglycerol (DGDG) and sulfoquinovosyl diacylglycerol (SQDG) [20–22]. Previous studies reported that glycolipids from spinach show various bioactivities such as anti-angiogenic and anti-tumor effects [20–23]. Recently, some reports indicated that glycolipids also induce anti-inflammatory activities [24,25]. However, whether glycolipids from spinach have anti-inflammatory effects on vascular endothelial cells remains unclear. In the present study, we investigated the effects of glycolipids extracted from spinach on LPS-induced endothelial inflammation and evaluated the underlying molecular mechanisms. 2. Materials and methods 2.1. Materials Lipopolysaccharides (LPS), N-nitro-L-arginine methyl ester (L-NAME), and the PKH26 Red Fluorescent Cell Linker Kit were purchased from Sigma Aldrich Co. (St Louis, MO, USA). The following antibodies were obtained from Cell Signaling Technology (Danvers, MA, USA): anti-phospho-eNOS (Ser-1177), anti-eNOS, anti-NF-kB, anti-phospho-NF-kB (Ser-536), anti-ICAM-1, antiVCAM-1, and anti-b-actin. Anti-iNOS was purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Horseradish peroxidase (HRP)-conjugated anti-rabbit IgG antibody and an enhanced chemiluminescence (ECL) system were obtained from GE Healthcare (Buckinghamshire, UK). 2.2. Isolation of the glycolipid fraction from spinach Dried spinach was obtained from Kodama Foods Co., Ltd (Hiroshima, Japan) and 200 g was extracted with 70% aqueous ethanol under reflux for 2 h. After filtration, the extract was treated with carboraffin (Osaka Gas Chemicals Co., Ltd., Osaka, Japan), and then stirred under reflux for 1 h. The eluted solution was evaporated, and the glycolipid mixture (48.7 g) was obtained. The mixture was subjected to Diaion HP-20 column chromatography (Mitsubishi Chemical Co. Ltd.) and eluted with 70% ethanol, 95% ethanol, and then chloroform. The chloroform elution was applied to silica gel column chromatography (silica gel 60 F254,

Merck Millipore, San Diego, CA, USA), and the glycolipid fraction (425 mg) was obtained using a chloroform/methanol/H2O (40:10:0.1, v/v) mixture as eluent. 2.3. Cell culture Human umbilical vein endothelial cells (HUVECs) were purchased from Cambrex Bio Science Walkersville, Inc. (Walkersville, MD, USA). HUVECs were cultured in endothelial cell basal medium (EBM-2; Lonza, Walkersville, MD, USA) supplemented with EGM-2MV at 37
C and 5% CO2. Cells between passage 3 and 5 were used for the experiments. THP-1 cells (human acute monocyte leukemia cell line) were purchased from JCRB cell bank (Osaka, Japan). THP-1 cells were cultured in RPMI1640 medium (Life Technologies, Waltham, MA, USA) supplemented with 10% FBS and 1% antibiotics (Life Technologies). 2.4. Cell viability assay Cell viability was analyzed using the Premix WST-1 Cell Proliferation Assay System (Takara, Tokyo, Japan), according to the manufacturer's instructions. Briefly, 1 103 HUVECs were cultured in 96-well plates for 4 h and treated with glycolipids (0.1–100 mg/mL) or vehicle (DMSO) for 24 h or 48 h. Cells were incubated with Premix WST-1 for 30 min and the absorbance was measured at 440 nm using a microplate reader (Multiscan FC, ThermoFisher Scientific, Rockford, IL, USA). 2.5. Real time RT-PCR analysis Total RNA was isolated with ISOGEN (NIPPON GENE, Tokyo, JAPAN) from HUVECs treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (0.1–100 mg/mL) or vehicle (DMSO) at 2 h. Reverse transcription was performed with 1 mg of RNA, random primers and MMLV reverse transcriptase (ReverTraAce-a TOYOBO, Osaka, Japan). Quantitative real-time PCR was performed with the LightCycler480 System (Roche Diagnostics, location) and the THUNDERBIRD SYBR Green PCR kit (TOYOBO, Osaka, Japan). The following primers were used: IL-6: sense, 50 -GGCTGCAGGACATGACAACT-30 , and antisense, 50 -ATCT0 0 GAGGTGCCCATGCTAC-3 , MCP-1: sense, 5 -AGCAGCAAGTGTCCCAAAGA-30 , and antisense, 50 -TTTGCTTGTCCAGGTGGTCC-30 , ICAM-1: sense, 50 -GGCCACGCATCTGATCTGTA-30 , and antisense, 50 -ACTTCCCCTCTCATCAGGCT-30 , VCAM-1: sense, 50 TCGTGATCCTTGGAGCCTCA-30 , and antisense, 50 -AGGAAAAGAGCCTGTGGTGC-30 , GAPDH: sense, 50 -CGACCACTTTGTCAAGCTCA-30 , and antisense, 50 -AGGGGAGATTCAGTGTGGTG-30 . 2.6. Monocyte adhesion assay HUVECs were cultured at a density of 5  104 cells on a 24-well cell culture plate for 24 h and then treated with LPS (1 mg/mL) or PBS for 24 h. Labeling with PKH26 (Sigma) was performed according with manufacture's instruction. Briefly, THP-1 cells were centrifuged at 400  g for 5 min and washed twice with serum free RPMI1640 medium. The cells were suspended in 1 ml of Diluent C, and mixed with equal volume of PKH26 Dye solution. The cell/dye mixture was incubated at room temperature for 5 min, then 2 ml of FBS were added to the suspension to stop the staining reaction. After 1 min, the suspension was centrifuged at 400  g for 10 min and washed twice with 10% FBS containing RPMI1640 medium. PKH26 Red labeled THP-1 cells (2  105 cells) were then added to HUVECs and incubated with glycolipids (100 mg/mL) or vehicle (DMSO) for 1 h. Unbound cells were removed by washing with PBS while the remaining cultured cells were fixed with 4% paraformaldehyde for 10 min and then washed with PBS. The

M. Ishii et al. / Biomedicine & Pharmacotherapy 91 (2017) 111–120

adhering red fluorescent THP-1 cells were quantified using fluorescence microscopy (EVOS FL Imaging System, ThermoFisher Scientific). Adhering cells were counted in five randomly chosen fields from the duplicated wells. Each experiment was repeated 4 times. 2.7. Immunoblotting analysis HUVECs were treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for indicated times and lysed in RIPA lysis buffer (ThermoFisher Scientific). Cell lysates were subjected to immunoblotting. In some experiments, HUVECs were pretreated with L-NAME (10 mM) for 1 h and then treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) at 37
C for 1 h or 4 h. The following antibodies were used: anti-phospho-eNOS (Ser-1177), anti-eNOS, anti-iNOS, anti-NF-kB, anti-phospho-NF-kB (Ser-536), anti-ICAM-1, anti-VCAM-1, and anti-b-actin. 2.8. Immunofluorescence staining HUVECs were treated with treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 4 h, then cells were washed with PBS and fixed with 4% paraformaldehyde for 15 min and permeabilized with cold methanol at 20
C for 15 min. After washing with PBS, cells were incubated with 0.1% Triton X-100 and 1% BSA in PBS for 1 h. Then, cells were incubated with anti-NF-kB antibody at 4
C overnight, and subsequently incubated with Alexa Fluor 488 conjugated antirabbit antibody (ThermoFisher Scientific) at room temperature for 1 h. Nuclei were stained with the DNA binding dye 4,6-diamino-2phenylindole (DAPI) (Dojindo, Kumamoto, Japan) for 10 min. The number of NF-kB positive nuclei was counted and expressed as the percentage of the total number of nuclei at least 200 cells in different microscopic fields. The experiment was repeated 3 times.

113

2.10. Detection of reactive oxygen species (ROS) production ROS production was measured using CellROX green Flow Cytometry Assay Kit (ThermoFisher Scientific). HUVECs were treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 24 h, and then, cells were collected and suspended at a density of 5  105 cells/mL in EGM-2 MV medium. CellROX green regent at a final concentration of 500 nM was added to the cell suspensions and incubate at 37
C. After 60 min, the samples were analyzed by flow cytometry (Guava easyCyte, Merck, Darmstadt, Germany). The experiment was repeated 3 times. 2.11. Statistical analysis The data are presented as mean  S.D. values. Comparisons between 2 groups were performed using Student's t-test. Statistical analysis for multiple comparisons among the groups was performed using one-way ANOVA. p < 0.05 was considered to indicate a statistically significant difference. 3. Results 3.1. Effect of purified glycolipids on HUVECs In the present study, we isolated glycolipids from dried spinach using Diaion HP-20 column chromatography and silica gel column chromatography. We analyzed the components of the extracted solution by TLC. The chloroform elute fraction mainly contained DGDG and MGDG (Supplement Fig. 1). We used this fraction as glycolipids fraction in the following experiments. To examine the viability of HUVECs treated with glycolipids, cells were incubated with different concentrations of glycolipids for 24 h or 48 h. WST-1 assay was used to analyze cell viability. Glycolipids (0.1–100 mg/ mL) had no cytotoxic effect on HUVECs after 24 h or 48 h of incubation (Fig. 1).

2.9. Detection of NO Production of NO was assessed as the accumulation of total nitrate (NO3)/nitrite (NO2) in the culture medium using the Nitrate/Nitrate Colorimetric Assay kit (Dojindo), according to the manufacturer's protocol. After treatment with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 24 h, HUVEC culture supernatants were collected from the wells. Samples were mixed with an equal (1:1) volume of Griess reagent. The absorbance was measured at 540 nm using a 96-well microplate reader (Multiscan FC, ThermoFisher Scientific).

3.2. Glycolipids suppress pro-inflammatory cytokines and adhesion molecules expression Treatment with LPS (1 mg/mL) for 2 h significantly induced IL-6 and MCP-1 mRNA expression in HUVECs. Furthermore, LPS also enhanced ICAM-1 and VCAM-1 mRNA expression in HUVECs (Fig. 2). To investigate whether glycolipids suppress the expression of LPS-induced pro-inflammatory cytokines and adhesion molecules, HUVECs were co-treated with glycolipids (0.1–100 mg/mL) and LPS (1 mg/mL) for 2 h. 100 mg/mL glycolipids effectively

Fig. 1. Effect of glycolipids from spinach on HUVEC viability. HUVEC viability was assessed after treatment with glycolipids from spinach using the WST-1 assay. Treatment with glycolipids (0.1–100 mg/mL) had no cytotoxic effect on HUVECs after 24 h or 48 h of incubation. Results are presented as mean  S.D. (n = 3)..

114

M. Ishii et al. / Biomedicine & Pharmacotherapy 91 (2017) 111–120

Fig. 2. Glycolipids suppress expression of pro-inflammatory cytokines and adhesion molecules. HUVECs were treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 2 h, then mRNA expression level of IL-6, MCP-1, ICAM-1, and VCAM-1 was determined by real-time PCR (n = 3). Results are presented as mean  S.D. *p < 0.05 vs. LPS ()/glycolipids () group, #p < 0.05 vs. LPS (+)/glycolipids () group (n = 3).

suppressed LPS-induced IL-6, MCP-1, ICAM-1, or VCAM-1 mRNA expression (Fig. 2). 3.3. Glycolipids block LPS-induced monocyte adhesion to HUVECs We further investigated the inhibitory effect of glycolipids on LPS-induced ICAM-1 and VCAM-1 protein expression. ICAM-1 and VCAM-1 protein expression were increased after 4 h of LPS treatment, but were significantly decreased by treatment with 100 mg/mL glycolipids (Fig. 3A). Because the glycolipids suppressed LPS-induced expression of cell adhesion molecules in HUVECs, we investigated whether glycolipids also block LPS-induced monocyte adhesion to HUVECs. LPS treatment induced THP-1 cells adhesion to HUVECs, and treatment with 100 mg/mL glycolipids significantly reduced this adhesion (Fig. 3B). In addition, to confirm the effect of LPS and glycolipids on the viability of HUVECs, we assessed the cell number of HUVECs after treatment with LPS (1 mg/mL) and/or glycolipids (100 mg/mL). HUVECs were treated with LPS or PBS for 24 h, and then treated glycolipids or vehicle (DMSO). After 1 h, cell number was counted. There was no significant differences in the cell number between each group (Supplemental Fig. 2). Treatment with LPS (1 mg/mL) and glycolipids (100 mg/mL) had no effect on the viability of HUVECs.

3.4. Glycolipids suppress LPS-induced NF-kB activation NF-kB signaling is known to play a critical role in LPS-induced expression of adhesion molecules in HUVECs [10]. We evaluated whether glycolipids suppress the LPS-induced NF-kB activation in HUVECs. Phosphorylation of NF-kB was significantly increased after treating HUVECs with LPS for 1 h. In contrast, treatment with 100 mg/mL glycolipids effectively suppressed LPS-induced NF-kB phosphorylation (Fig. 4A). While NF-kB protein levels in total cell lysates of HUVECs treated with or without LPS and glycolipids were unchanged (Fig. 4A), LPS mediated NF-kB translocation from the cytosol to the nucleus (Fig. 4B). Treatment with glycolipids significantly inhibited LPS-induced nuclear translocation of NF-kB in HUVECs. 3.5. Involvement of eNOS and NF-kB pathway in glycolipid-mediated inhibition of adhesion molecule expression and monocyte adhesion to HUVECs Nitric oxide (NO) plays a critical role in vascular homeostasis [8]. NO has been shown to have anti-inflammatory and protective effects on vascular endothelial cells by inhibiting NF-kB activation [10]. To examine the participation of eNOS signaling in the observed inhibition of LPS-induced NF-kB activation and monocyte adhesion to HUVECs by glycolipids treatment, we assessed the

M. Ishii et al. / Biomedicine & Pharmacotherapy 91 (2017) 111–120

115

Fig. 3. Glycolipids inhibit LPS-induced monocyte adhesion to HUVECs by suppressing expression of adhesion molecules. (A) Immunoblotting with indicated antibodies was performed on HUVECs treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 4 h. Representative blots are shown. Relative expression levels were normalized to the b-actin protein signal. Results are presented as mean  S.D. *p < 0.05 vs. LPS ()/glycolipids () group, #p < 0.05 vs. LPS (+)/glycolipids () group (n = 3). (B) HUVECs were treated with or without LPS (1 mg/mL) for 24 h, and then incubated with glycolipids (100 mg/mL) or vehicle (DMSO) for 1 h. Representative photomicrographs for adhered THP-1 cells on HUVECs are shown. The adhering red fluorescent THP-1 cells were quantified by counting in five fields randomly chosen from each wells. Values are expressed as the mean adhered cell number per optical field. Results are presented as mean  S.D. (n = 4). *p < 0.05 vs. LPS ()/glycolipids () group, #p < 0.05 vs. LPS (+)/glycolipids () group. Scale bar, 200 mm.

effects of glycolipids on activation of eNOS in HUVECs by western blot analysis. Treatment with 100 mg/mL glycolipids for 1 h significantly increased the phosphorylation of eNOS (Fig. 5A). Glycolipids had no effect on eNOS protein levels. Furthermore, to investigate NO production in HUVECs treated with glycolipids, we measured NO levels in the cell culture supernatant. Treatment with 100 mg/mL glycolipids for 24 h significantly induced NO production in HUVECs regardless of LPS treatment (Fig. 5B). Previous studies showed that LPS treatment induce iNOS and suppress eNOS expression in endothelial cells [26,27]. An excessive NO produced from iNOS leads to detrimental effects on vascular function. We assessed whether glycolipids suppress LPS-induced iNOS expression in HUVECs. Treatment with LPS for 24 h induced iNOS expression, and glycolipids treatment significantly suppressed iNOS expression in HUVECs (Fig. 5C). NO can react with reactive oxygen species (ROS) to produce more powerful oxidants such as peroxynitrite, which is of further deterioration of atherosclerosis [28]. We therefore examined the effect of glycolipids on ROS clearance. HUVECs were treated with LPS and/or glycolipids for 24 h, and then ROS production levels were detected using CellROX green by flow cytometry. LPS treatment strongly enhanced CellROX green fluorescent intensity, but, treatment with glycolipids did not suppress LPS-induced ROS production (Fig. 5D). To further analyze whether eNOS signaling participates in the inhibition of LPS-induced NF-kB activation by glycolipids, HUVECs

were pretreated with the NOS inhibitor L-NAME, followed by treatment with LPS and glycolipids. Treatment with 10 mM L-NAME significantly attenuated glycolipids-induced phosphorylation of eNOS, and inhibited glycolipids-induced suppression of phosphorylation of NF-kB (Fig. 6A). To investigate whether eNOS signaling participates in the glycolipids-induced suppression of NF-kB nuclear translocation, HUVECs were pretreated with L-NAME, followed by treatment with LPS and glycolipids. Treatment with L-NAME inhibited glycolipid-mediated suppression of NF-kB nuclear translocation (Fig. 6B). We next examined the effect of eNOS inhibition on the expression of adhesion molecules and monocyte adhesion to HUVECs. Pretreatment with L-NAME attenuated the inhibitory effect of glycolipids on LPS-induced ICAM-1 and VCAM-1 protein expression (Fig. 6C). Furthermore, treatment with L-NAME abolished the glycolipids-mediated suppression of THP-1 cell adhesion to HUVECs (Fig. 6D). These data suggest the importance of eNOS and NF-kB pathway in the inhibition of the expression of adhesion molecules and monocyte adhesion to HUVECs by glycolipids treatment. 4. Discussion In the present study, we demonstrated that glycolipids from spinach suppress LPS-induced vascular inflammation. Glycolipids significantly blocked LPS-induced pro-inflammatory cytokine and adhesion molecule expression, leading to the suppression of

116

M. Ishii et al. / Biomedicine & Pharmacotherapy 91 (2017) 111–120

Fig. 4. Glycolipids suppress LPS-induced NF-kB activation. (A) Effects of glycolipids from spinach on LPS-induced NF-kB activation. HUVECs were treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 1 h. Representative blots are shown. Relative expression levels were normalized to the b-actin protein signal. Results are presented as mean  S.D. (n = 3). *p < 0.05 vs. LPS ()/glycolipids () group, #p < 0.05 vs. LPS (+)/glycolipids () group. (B) Immunofluorescence microscopy analysis for NF-kB translocation in HUVECs treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 4 h. Immunostaining with NF-kB antibody (green) and DAPI (blue). The ratio of NF-kB positive nuclei was normalized to total nuclei. Results are presented as mean  S.D. (n = 3). *p < 0.05 vs. LPS ()/glycolipids () group, #p < 0.05 vs. LPS (+)/glycolipids () group. Scale bar, 50 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

monocyte adhesion to HUVECs. In this study, we also showed that glycolipids inhibited LPS-induced NF-kB phosphorylation and nuclear translocation. Treatment with glycolipids significantly induced eNOS activation and NO production in HUVECs. Whereas, treatment with glycolipids significantly blocked LPS-induced iNOS expression in HUVECs. Furthermore, inhibition of eNOS signaling by L-NAME effectively attenuated glycolipids-induced suppression of NF-kB activation and adhesion molecule expression and abolished the suppression of monocyte adhesion to HUVECs mediated by spinach glycolipids. These results indicate that glycolipids suppress LPS-induced vascular inflammation through attenuation of the NF-kB pathway by increasing NO production in endothelial cells. Spinach (S. oleracea) is a green leafy vegetable, which is recognized as a highly nutrient rich food. It is known to contain

large amounts of vitamins, minerals, and other bioactive compounds such as carotenoids [17]. Previous studies showed that several bioactive compounds from spinach present various pharmacological activities including anti-obesity, hypoglycemic, anti-tumor, anti-oxidant activities [18,19]. It is known that the thylakoid membrane of higher plants contains several glycolipids such as MGDG, DGDG, and SQDG, and spinach is one of richest sources of glycolipids. In the present study, we isolated glycolipids from dried spinach, and the obtained extracted fraction contained mainly DGDG and MGDG. Previous studies have reported that glycolipids fractions containing DGDG, MGDG, or SQDG inhibit DNA polymerase activity, and suppress cancer cell growth [21,22]. However, in the present study, glycolipids had no cytotoxic effect on HUVECs. Bruno et al. indicated that MGDG, DGDG, or SQDG showed anti-inflammatory activity in croton-oil-induced ear

M. Ishii et al. / Biomedicine & Pharmacotherapy 91 (2017) 111–120

117

Fig. 5. Glycolipids activate eNOS signaling and promote NO production in HUVECs. (A) Immunoblotting with indicated antibodies was performed on proteins extracted from HUVECs treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 1 h. Representative blots are shown. Relative eNOS phosphorylation levels were normalized to the total eNOS protein signal. Results are presented as mean  S.D. (n = 3). *p < 0.05. (B) HUVECs were treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 24 h, and NO levels in the cultured medium were measured. Results are presented as mean  S.D. (n = 5). *p < 0.05. (C) Immunoblotting with iNOS antibody was performed on proteins extracted from HUVECs treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 24 h. Representative blots are shown. Relative iNOS levels were normalized to the b-actin protein signal. Results are presented as mean  S.D. (n = 3). *p < 0.05. (D) Representative histogram of green fluorescent intensity for CellROX green. HUVECs were treated with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 24 h, and CellROX green intensity was detected using flow cytometry. The gray line indicates LPS ()/glycolipids (), the green line indicates LPS ()/glycolipids (+), the pink line indicates LPS (+)/glycolipids (), and the blue line indicates LPS (+)/glycolipids (+). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

edema and carrageenan-induced paw edema models with less toxicity [29]. In addition, a recent study indicated that MGDG inhibited human gastric cancer cell growth, but DGDG and SQDG had no cytotoxic effects [30]. The difference between our results and previously reported studies may be owing to the kinds of cells or the components of the glycolipids fraction, therefore, glycolipids may have no cytotoxic effect on HUVECs. Previous studies reported that glycolipids show various bioactivities such as anti-angiogenic and anti-tumor effects [20–23]. Recently, some reports indicated that glycolipids induce anti-inflammatory activities. It has been reported that MGDG shows anti-inflammatory effects in articular cartilage cells [25,31]. In an in vivo study, glycolipids (MGDG and DGDG) extracted from

spinach protected 5-fluorouracil induced intestinal mucosal injury by suppressing inflammation and reactive oxygen species (ROS) production [24]. MGDG and DGDG also showed anti-inflammatory activity in mice croton-oil-induced ear edema and carrageenaninduced paw edema models [29]. Chronic vascular inflammation and subsequent adhesion of monocytes to the endothelium are essential events in the initiation and progression of atherosclerosis [3–5]. Therefore, inhibition of inflammatory responses and monocyte adhesion to the endothelium could be an effective strategy for preventing atherosclerosis. The present study shows that treatment with glycolipids from spinach significantly suppressed LPS-induced expression of IL-6 and MCP-1, which are critical inflammatory molecules in the

118

M. Ishii et al. / Biomedicine & Pharmacotherapy 91 (2017) 111–120

Fig. 6. eNOS and NF-kB pathway are involved in glycolipids mediated inhibition of adhesion molecule expression and monocyte adhesion to HUVECs. (A) Effect of L-NAME on inhibition of LPS-induced NF-kB activation by glycolipids. HUVECs were pretreated with or without L-NAME (10 mM) for 1 h, followed by treatment with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 1 h. Representative blots are shown. Relative eNOS phosphorylation levels were normalized to the total eNOS protein signal. Relative phospho-NF-kB and total NF-kB protein expression levels were normalized to the b-actin protein signal. Results are presented as mean  S.D. (n = 3). *p < 0.05 vs. LPS ()/glycolipids ()/L-NAME () group, #p < 0.05 vs. LPS (+)/glycolipids (+)/L-NAME () group, yp < 0.05 vs. LPS ()/glycolipids (+)/L-NAME () group. (B) Immunofluorescence microscopy analysis for NF-kB translocation in HUVECs pretreated with L-NAME (10 mM) for 1 h, followed by treatment with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 4 h. Immunostaining was done with NF-kB antibody (green) and DAPI (blue). The ratio of NFkB positive nuclei was normalized to total nuclei. Results are presented as mean  S.D. (n = 3). *p < 0.05 vs. LPS ()/glycolipids ()/L-NAME (+) group. Scale bar, 50 mm. (C) Immunoblotting with indicated antibodies was performed on HUVECs pretreated with or without L-NAME (10 mM) for 1 h, followed by treatment with LPS (1 mg/mL) or PBS in the presence or absence of glycolipids (100 mg/mL) or vehicle (DMSO) for 4 h. Representative blots are shown. Relative expression levels were normalized to the b-actin protein signal. Results are presented as mean  S.D. (n = 3). *p < 0.05 vs. LPS ()/glycolipids ()/L-NAME () group, #p < 0.05 vs. LPS (+)/glycolipids ()/L-NAME () group. y p < 0.05 vs. LPS (+)/glycolipids (+)/L-NAME () group. (D) HUVECs were treated with or without LPS (1 mg/mL) for 24 h, and then, treated with L-NAME (10 mM) for 1 h, followed by incubation with glycolipids (100 mg/mL) or vehicle (DMSO) for 1 h. The adhering red fluorescent THP-1 cells were quantified using fluorescence microscopy by counting five fields randomly chosen from the duplicated wells. Results are presented as mean  S.D. (n = 4). *p < 0.05 vs. LPS ()/glycolipids ()/L-NAME (+) group. Scale bar, 200 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

progression of atherosclerosis. It is well known that the interaction of monocytes with the endothelium is mediated by cell adhesion molecules such as VCAM-1 and ICAM-1 localized on the surface of activated endothelial cells [3–5]. We showed that treatment with glycolipids significantly suppressed LPS-induced ICAM-1 and VCAM-1 expression on HUVECs, and reduced the adhesion of LPS-induced THP-1 cells to HUVECs. These results indicate the

beneficial effect of glycolipids from spinach on vascular inflammation. Activation of NF-kB has been shown to play a critical role in the inflammatory response through regulating the expression of various inflammatory mediators including cytokines, chemokines, and adhesion molecules [6]. Previous reports indicate that several glycolipids inhibit NF-kB activation and show anti-inflammatory

M. Ishii et al. / Biomedicine & Pharmacotherapy 91 (2017) 111–120

effects. MGDG suppresses the inflammatory response in human articular cartilage cells through the inhibition of p38 and the NF-kB pathway [25]. Dilinolenoyl galactosyl glycerol (DLGG), a specific type of MGDG, prevents NF-kB nuclear translocation by suppressing LPS-induced IkBa phosphorylation and degradation in RAW 264.7 macrophages [32]. In addition, a recent study has shown that DLGG is a ligand for the peroxisome proliferator activated receptor (PPAR)-g [33]. PPAR-g activation by PPAR-g ligands has been shown to suppress inflammation through modulating NF-kB activation [34,35]. Consistent with these observations, our study demonstrates that treatment with glycolipids significantly inhibited LPS-induced NF-kB phosphorylation and NF-kB nuclear translocation in HUVECs. It is well established that NO is a critical regulator of vascular homeostasis. NO promotes arterial vasodilation, and inhibits platelet aggregation, monocyte adhesion to the endothelium, and smooth muscle cell proliferation [8]. eNOS deficiency has been reported to promote various cardiovascular diseases including atherosclerosis [36–38]. Thus, promotion of eNOS activation and NO production may be an effective approach to control atherosclerosis. Our present data show that treatment with glycolipids significantly increased phosphorylation of eNOS and NO production in HUVECs. Accordingly, there is a possibility that NO is a key regulator of the glycolipids-mediated anti-inflammatory effect on vascular endothelial cells. Recently, a number of studies have shown that promotion of eNOS activation and NO production in the endothelium contributes to NF-kB inactivation, and attenuates atherosclerosis [9,10]. Consistent with these observations, we demonstrated that eNOS inhibition by L-NAME effectively abrogated glycolipids-induced suppression of phosphorylation and nuclear translocation of NF-kB in HUVECs. Furthermore, eNOS inhibition abolished glycolipids-mediated suppression of adhesion molecule expression and monocyte adhesion to HUVECs. Although our present data suggest that NO production via eNOS activation by glycolipids plays an important role in the suppression of LPS-induced vascular inflammation, it was reported that multiple isoforms of NOS were expressed in normal and atherosclerotic endothelial cells [39]. It has been reported that NO production by eNOS or iNOS have different effect on pathogenesis in atherosclerosis [11]. An excessive NO produced from iNOS leads to detrimental effects on vascular function. In the present study, treatment with glycolipids significantly suppressed LPS-induced iNOS expression. A number of studies have shown that LPS treatment increased NO production in endothelial cells [26]. In our study, LPS treatment did not affect NO production. Previously, Bernardini et al. also reported that LPS treatment for 24 h did not induce any significant change in the total level of NO, despite the fact that LPS effects on the NOSs expression [27]. It is unclear why NO levels did not change, even though the iNOS expression was increased. It is necessary to elucidate the discrepancy in future studies. Activation of NF-kB plays an essential role for the induction of iNOS [40]. Treatment with glycolipids significantly inhibited LPS-induced NF-kB activation, therefore, iNOS suppression by glycolipids may be mediated via NF-kB dependent mechanism. These results indicate that glycolipids-induced attenuation of vascular inflammation was mediated at least in part by the suppression of iNOS. It is well established that ROS play a critical roles in the development of vascular disease [41]. In the present study, we examined the effect of glycolipids on ROS clearance. However, treatment with glycolipids did not suppress LPS-induced ROS production. In conclusion, our data demonstrate that glycolipids from spinach suppress LPS-induced vascular inflammation and monocyte adhesion to the endothelium through eNOS and the NF-kB dependent pathway. These findings suggest that glycolipids from

119

spinach may have a potential therapeutic use for inflammatory vascular diseases. Conflict of interest statement The authors declare that there is no conflict of interest. Acknowledgements This work was in part supported by Grant-in Aid for Scientific Research B (26293414) and Grant-in Aid for Exploratory Research (26670849), and Grant-in Aid for Scientific Research C (15K11169). We gratefully acknowledge the technical assistance of Yasuko Tanaka. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biopha.2017. 04.052. References [1] M.H. Criqui, V. Aboyans, Epidemiology of peripheral artery disease, Circ. Res. 116 (9) (2015) 1509–1526. [2] A. Daiber, S. Steven, A. Weber, V.V. Shuvaev, V.R. Muzykantov, I. Laher, H. Li, S. Lamas, T. Munzel, Targeting vascular (endothelial) dysfunction, Br. J. Pharmacol. (2016). [3] R. Ross, Atherosclerosis – an inflammatory disease, N. Engl. J. Med. 340 (2) (1999) 115–126. [4] Y. Mizuno, R.F. Jacob, R.P. Mason, Inflammation and the development of atherosclerosis, J. Atheroscler. Thromb. 18 (5) (2011) 351–358. [5] P. Libby, Inflammation in atherosclerosis, Nature 420 (6917) (2002) 868–874. [6] P.P. Tak, G.S. Firestein, NF-kappaB: a key role in inflammatory diseases, J. Clin. Invest. 107 (1) (2001) 7–11. [7] R.G. Baker, M.S. Hayden, S. Ghosh, NF-kappaB, inflammation, and metabolic disease, Cell Metab. 13 (1) (2011) 11–22. [8] U. Forstermann, W.C. Sessa, Nitric oxide synthases: regulation and function, Eur. Heart J. 33 (7) (2012) 837a–1837a 829–837. [9] L. Wang, X.M. Qiu, Q. Hao, D.J. Li, Anti-inflammatory effects of a Chinese herbal medicine in atherosclerosis via estrogen receptor beta mediating nitric oxide production and NF-kappaB suppression in endothelial cells, Cell Death Dis. 4 (2013) e551. [10] H.J. Hwang, T.W. Jung, H.C. Hong, H.Y. Choi, J.A. Seo, S.G. Kim, N.H. Kim, K.M. Choi, D.S. Choi, S.H. Baik, H.J. Yoo, Progranulin protects vascular endothelium against atherosclerotic inflammatory reaction via Akt/eNOS and nuclear factor-kappaB pathways, PLOS ONE 8 (9) (2013) e76679. [11] A. Chatterjee, S.M. Black, J.D. Catravas, Endothelial nitric oxide (NO) and its pathophysiologic regulation, Vascul. Pharmacol. 49 (4–6) (2008) 134–140. [12] M.D. Miedema, A. Petrone, J.M. Shikany, P. Greenland, C.E. Lewis, M.J. Pletcher, J.M. Gaziano, L. Djousse, Association of fruit and vegetable consumption during early adulthood with the prevalence of coronary artery calcium after 20 years of follow-up: the coronary artery risk development in young adults (CARDIA) study, Circulation 132 (21) (2015) 1990–1998. [13] M. Ruiz-Canela, M.A. Martinez-Gonzalez, Lifestyle and dietary risk factors for peripheral artery disease, Circ. J. 78 (3) (2014) 553–559. [14] S. Nagarajan, Mechanisms of anti-atherosclerotic functions of soy-based diets, J. Nutr. Biochem. 21 (4) (2010) 255–260. [15] P. Palozza, R. Simone, A. Catalano, G. Monego, A. Barini, M.C. Mele, N. Parrone, S. Trombino, N. Picci, F.O. Ranelletti, Lycopene prevention of oxysterol-induced proinflammatory cytokine cascade in human macrophages: inhibition of NFkappaB nuclear binding and increase in PPARgamma expression, J. Nutr. Biochem. 22 (3) (2011) 259–268. [16] W.M. Loke, J.M. Proudfoot, J.M. Hodgson, A.J. McKinley, N. Hime, M. Magat, R. Stocker, K.D. Croft, Specific dietary polyphenols attenuate atherosclerosis in apolipoprotein E-knockout mice by alleviating inflammation and endothelial dysfunction, Arterioscler. Thromb. Vasc. Biol. 30 (4) (2010) 749–757. [17] L. Jaime, E. Vazquez, T. Fornari, C. Lopez-Hazas Mdel, M.R. Garcia-Risco, S. Santoyo, G. Reglero, Extraction of functional ingredients from spinach (Spinacia oleracea L.) using liquid solvent and supercritical CO(2) extraction, J. Sci. Food Agric. 95 (4) (2015) 722–729. [18] J.L. Roberts, R. Moreau, Functional properties of spinach (Spinacia oleracea L.) phytochemicals and bioactives, Food Funct. 7 (8) (2016) 3337–3353. [19] S.H. Ko, J.H. Park, S.Y. Kim, S.W. Lee, S.S. Chun, E. Park, Antioxidant effects of spinach (Spinacia oleracea L.) supplementation in hyperlipidemic rats, Prev. Nutr. Food Sci. 19 (1) (2014) 19–26. [20] H. Iijima, N. Kasai, H. Chiku, T. Takeuchi, K. Kuramochi, S. Hanashima, S. Kobayashi, F. Sugawara, K. Sakaguchi, H. Yoshida, Y. Mizushina, Structure-

120

[21]

[22]

[23]

[24]

[25]

[26] [27]

[28]

[29]

[30]

[31]

M. Ishii et al. / Biomedicine & Pharmacotherapy 91 (2017) 111–120 activity relationship of a glycolipid, sulfoquinovosyl diacylglycerol, with the DNA binding activity of p53, Int. J. Mol. Med. 19 (1) (2007) 41–48. N. Maeda, T. Hada, C. Murakami-Nakai, I. Kuriyama, H. Ichikawa, Y. Fukumori, J. Hiratsuka, H. Yoshida, K. Sakaguchi, Y. Mizushina, Effects of DNA polymerase inhibitory and antitumor activities of lipase-hydrolyzed glycolipid fractions from spinach, J. Nutr. Biochem. 16 (2) (2005) 121–128. N. Maeda, Y. Kokai, S. Ohtani, H. Sahara, Y. Kumamoto-Yonezawa, I. Kuriyama, T. Hada, N. Sato, H. Yoshida, Y. Mizushina, Anti-tumor effect of orally administered spinach glycolipid fraction on implanted cancer cells, colon-26, in mice, Lipids 43 (8) (2008) 741–748. K. Matsubara, H. Matsumoto, Y. Mizushina, M. Mori, N. Nakajima, M. Fuchigami, H. Yoshida, T. Hada, Inhibitory effect of glycolipids from spinach on in vitro and ex vivo angiogenesis, Oncol. Rep. 14 (1) (2005) 157–160. A. Shiota, T. Hada, T. Baba, M. Sato, H. Yamanaka-Okumura, H. Yamamoto, Y. Taketani, E. Takeda, Protective effects of glycoglycerolipids extracted from spinach on 5-fluorouracil induced intestinal mucosal injury, J. Med. Invest. 57 (3–4) (2010) 314–320. V. Ulivi, M. Lenti, C. Gentili, G. Marcolongo, R. Cancedda, F. Descalzi Cancedda, Anti-inflammatory activity of monogalactosyldiacylglycerol in human articular cartilage in vitro: activation of an anti-inflammatory cyclooxygenase2 (COX-2) pathway, Arthritis. Res. Ther. 13 (3) (2011) R92. R. Fu, Z. Chen, Q. Wang, Q. Guo, J. Xu, X. Wu, XJP-1, a novel ACEI, with antiinflammatory properties in HUVECs, Atherosclerosis 219 (1) (2011) 40–48. C. Bernardini, F. Greco, A. Zannoni, M.L. Bacci, E. Seren, M. Forni, Differential expression of nitric oxide synthases in porcine aortic endothelial cells during LPS-induced apoptosis, J. Inflamm. (Lond.) 9 (1) (2012) 47. J. Vasquez-Vivar, D. Duquaine, J. Whitsett, B. Kalyanaraman, S. Rajagopalan, Altered tetrahydrobiopterin metabolism in atherosclerosis: implications for use of oxidized tetrahydrobiopterin analogues and thiol antioxidants, Arterioscler. Thromb. Vasc. Biol. 22 (10) (2002) 1655–1661. A. Bruno, C. Rossi, G. Marcolongo, A. Di Lena, A. Venzo, C.P. Berrie, D. Corda, Selective in vivo anti-inflammatory action of the galactolipid monogalactosyldiacylglycerol, Eur. J. Pharmacol. 524 (1–3) (2005) 159–168. C. Murakami, T. Kumagai, T. Hada, U. Kanekazu, S. Nakazawa, S. Kamisuki, N. Maeda, X. Xu, H. Yoshida, F. Sugawara, K. Sakaguchi, Y. Mizushina, Effects of glycolipids from spinach on mammalian DNA polymerases, Biochem. Pharmacol. 65 (2) (2003) 259–267. M. Lenti, C. Gentili, A. Pianezzi, G. Marcolongo, A. Lalli, R. Cancedda, F.D. Cancedda, Monogalactosyldiacylglycerol anti-inflammatory activity on adult articular cartilage, Nat. Prod. Res. 23 (8) (2009) 754–762.

[32] C.C. Hou, Y.P. Chen, J.H. Wu, C.C. Huang, S.Y. Wang, N.S. Yang, L.F. Shyur, A galactolipid possesses novel cancer chemopreventive effects by suppressing inflammatory mediators and mouse B16 melanoma, Cancer Res. 67 (14) (2007) 6907–6915. [33] H. Martin, T.K. McGhie, R.C. Lunken, A PPARgamma ligand present in Actinidia fruit (Actinidia chrysantha) is identified as dilinolenoyl galactosyl glycerol, Biosci. Rep. 33 (3) (2013). [34] M. Sasaki, P. Jordan, T. Welbourne, A. Minagar, T. Joh, M. Itoh, J.W. Elrod, J.S. Alexander, Troglitazone, a PPAR-gamma activator prevents endothelial cell adhesion molecule expression and lymphocyte adhesion mediated by TNFalpha, BMC Physiol. 5 (1) (2005) 3. [35] X. Ming, M. Ding, B. Zhai, L. Xiao, T. Piao, M. Liu, Biochanin A inhibits lipopolysaccharide-induced inflammation in human umbilical vein endothelial cells, Life Sci. 136 (2015) 36–41. [36] L.N. Zhang, D.W. Wilson, V. da Cunha, M.E. Sullivan, R. Vergona, J.C. Rutledge, Y. X. Wang, Endothelial NO synthase deficiency promotes smooth muscle progenitor cells in association with upregulation of stromal cell-derived factor-1alpha in a mouse model of carotid artery ligation, Arterioscler. Thromb. Vasc. Biol. 26 (4) (2006) 765–772. [37] O. Suda, M. Tsutsui, T. Morishita, A. Tanimoto, M. Horiuchi, H. Tasaki, P.L. Huang, Y. Sasaguri, N. Yanagihara, Y. Nakashima, Long-term treatment with N (omega)-nitro-L-arginine methyl ester causes arteriosclerotic coronary lesions in endothelial nitric oxide synthase-deficient mice, Circulation 106 (13) (2002) 1729–1735. [38] P.J. Kuhlencordt, R. Gyurko, F. Han, M. Scherrer-Crosbie, T.H. Aretz, R. Hajjar, M. H. Picard, P.L. Huang, Accelerated atherosclerosis, aortic aneurysm formation, and ischemic heart disease in apolipoprotein E/endothelial nitric oxide synthase double-knockout mice, Circulation 104 (4) (2001) 448–454. [39] J.N. Wilcox, R.R. Subramanian, C.L. Sundell, W.R. Tracey, J.S. Pollock, D.G. Harrison, P.A. Marsden, Expression of multiple isoforms of nitric oxide synthase in normal and atherosclerotic vessels, Arterioscler. Thromb. Vasc. Biol. 17 (11) (1997) 2479–2488. [40] C.L. Cooke, S.T. Davidge, Peroxynitrite increases iNOS through NF-kappaB and decreases prostacyclin synthase in endothelial cells, Am. J. Physiol. Cell Physiol. 282 (2) (2002) C395–C402. [41] A. Konior, A. Schramm, M. Czesnikiewicz-Guzik, T.J. Guzik, NADPH oxidases in vascular pathology, Antioxid. Redox Signal. 20 (17) (2014) 2794–2814.