Quercetin suppresses inflammation by reducing ERK1/2 phosphorylation and NF kappa B activation in Leptin-induced Human Umbilical Vein Endothelial Cells (HUVECs)

Background High concentrations of plasma leptin and the release of pro-inflammatory cytokines in leptin-resistance in obesity have been reported to trigger endothelial dysfunction. The objective of this study was to elucidate the role of quercetin in modulating leptin-induced inflammation as assessed by the levels of Ob-Ra expression, ERK1/2 phosphorylation, NF-kappa B activation and TNF-alpha secretion in umbilical vein endothelial cells (HUVECs) in vitro. Findings HUVECs were exposed to either control levels (0 ng/ml) or 500 ng/mL leptin (L) for 48 hours, followed by control or 125 uM quercetin (Q) for another 6 h. The experimental groups were as follows: L0Q0, L0Q125, L500Q0, L500Q125. The presence of the short chain leptin receptor isoform Ob-Ra in HUVECs was determined by Western blot and immunocytochemistry analyses. Ob-Ra expression, ERK1/2 phosphorylation, NF-kappa B activation and TNF-alpha secretion were quantified by ELISA, and NF-kappa B activationby immunofluorescence staining. Our results showed that Ob-Ra expression, ERK1/2 phosphorylation and NF-kappa B activation increased significantly after 500 ng/mL leptin exposure (1.8x, 1.5x, 6.2x for Ob-Ra, ERK1/2 and NF-kappa B, respectively), but were reduced by addition of 125 uM quercetin (0.7x, 0.3x and 0.4x for Ob-Ra, ERK1/2 and NF-kappa B, respectively), and that quercetin could also partially suppress leptin-induced TNF-alpha secretion (3.8x) by 0.8x. Conclusion Exposure of HUVECs to leptin up-regulated Ob-Ra expression and elevated ERK1/2 phosphorylation and NFkB activation, and increased TNF-alpha secretion. These effects strongly suppressed by quercetin, with the exception of TNF-alpha which was partially suppressed. The findings might be of clinical significance, as endothelial dysfunction that could lead to cardiovascular disease is preventable, and quercetin is a natural compound found in various plants and fruits.

Obesity has become a global health problem, with the prevalence of overweight and obesity reaching critical levels throughout the world, including Indonesia. A national survey in 2007 in 12 Indonesian provinces showed that 18.8% of the population older than 15 years old are obese [1]. Obesity is a major risk factor for cardiovascular disease, hypertension, dyslipidemia and diabetes mellitus, all of which reduce both the quality of life and life expectancy. Obesity is associated with excessive adipose tissue accumulation due to excessive energy intake and insufficient energy expenditure [2], and is characterized by the alteration of leptin levels, a cytokine produced by adipocytes and one of the regulators of energy metabolism. Studies have shown that most obese patients are leptin resistant, and high leptin levels were observed in these individuals [3]. An association between leptin and increased cardiovascular risk has been reported [4], and is associated with increased levels of inflammatory factors exhibiting pro-atherogenic effects [5][6][7]. Obesity has also been considered as a state of low-grade inflammation [8]; previous research has shown that atherosclerosis is the result of chronic inflammation, and early atherosclerosis formation is induced by pro-inflammatory cytokines and other proteins produced by inflammatory cells [9,10].
In obesity-related high plasma leptin conditions, inflammation occurs when signal transduction pathways are activated, such as activation of NFκΒ, by the binding of leptin to its receptor (Ob-R), and subsequent release of the inflammation factors, for instance tumour necrosis factor alpha (TNFα) [11]. Our preliminary results revealed that 500 ng/ml leptin decreases cell proliferation and increases TNFα, monocyte chemoattractant protein-1 (MCP-1), and intracellular Ca 2+ levels in human umbilical vein endothelial cells (HUVECs) [12].
Quercetin, a flavonoid compound found in plants and fruits, has been reported to have anti-inflammatory effects [13], which are mediated through the inhibition of pro-inflammatory cytokines [14]. The aim of this study was to investigate the effect of quercetin in modulating the expression of Ob-Ra, phosphorylation of ERK1/2, activation of NFκB and secretion of TNFα in leptin-induced human umbilical vein endothelial cells (HUVECs) in vitro.

Samples
Human umbilical vein endothelial cells (HUVECs) were obtained from umbilical cords of patients that have undergone cesarean section in Dr. Syaiful Anwar Hospital, Malang, after obtaining informed consent. This research was approved by the institutional research ethical committee from the Faculty of Medicine, Brawijaya University, Malang.
The experimental groups were: L0Q0 without leptin and quercetin (control, with DMSO and methylcellulose only); L0Q125 without leptin but with 125 μM quercetin; L500Q0 with 500 ng/mL leptin but without quercetin; L500Q125 with 500 ng/mL leptin and 125 μM quercetin. The experiments were repeated 5 times for each group.

Ob-Ra Immunocytochemistry
HUVECs were fixed with methanol in glass slides, then gently rinsed with phosphate buffer solution (PBS). Normal human serum (1:10 dilution) (MPBio, USA) was applied and incubated for 20-30 minutes at 37°C. Rabbit anti-human Ob-Ra antibody (1:100) (Santa Cruz Biotech, USA) was applied and the specimens were incubated overnight at 4°C. Labelled anti rabbit IgG-SA-HRP (KPL, USA) was applied for 60 minutes, then substratechromogen solution 3,3′-diaminobenzidine was added, and the specimens were incubated for 10 minutes. Hematoxylin counterstaining was performed to stain the nucleus. The slides were then covered with cover slips.
ELISA for secreted TNFα (Bender MedSystem, Austria) was conducted using the supernatant of HUVECs culture.

NFκB Immunofluorescence
Sub-confluent HUVECs grown on cover slips were fixed with methanol at room temperature for 10 minutes, labeled with 10 μg/mL antihuman p50/p65 NFκB antibody (Thermo Scientific, USA) for 1 hour, washed 3 times with PBS, then incubated with 8 μg/mL FITC-labeled goat anti-rabbit IgG (Santa Cruz Biotechnology Inc, USA) for 1 hour. Bound antibody was detected using Fluo view 1000 confocal microscope (Olympus, Japan) equipped for epiluminescence.

Statistical analyses
Results were the average of at least five repeats. Differences between groups were determined by ANOVA, followed by Tukey's test (SPSS vs. 17). All data were shown as means ± SD, and p < 0.05 was considered statistically significant.

Ob-Ra was expressed in HUVECs
The short isoform of leptin receptor (Ob-Ra) was expressed in HUVECs, as demonstrated by Western blot analysis shown in Figure 1A. After leptin induction the expression of Ob-Ra was increased almost two folds as compared to control (band intensities are 1488 ± 355.4 vs. 2586 ± 380.8, without and with leptin exposure, p = 0.001). Immunocytochemistry staining confirmed the increased expression of Ob-Ra in leptin-induced HUVECs ( Figure 1B).

Quercetin inhibited leptin-induced NFκB activation
FITC immunolabeling was used to determine the NFκB activation. As shown in Figure 4A signal density was increased after HUVECs was exposed to 500 ng/mL leptin as compared to control. Quercetin alone did not p = 10 -4 ). Quercetin alone did not activate NFκB ( Figure 4C).

Discussion
We have demonstrated the presence of the short chain leptin receptor isoform Ob-Ra in HUVECs, confirming previous studies [16,17]. We were able to show that Ob-Ra expression was increased after leptin exposure. Leptin has been reported to differentially modulate the expression of its receptors in both dose-and tissuedependent manner [18]. When treated with high leptin concentration, the level of leptin receptor was increased, reported to be due to changes in receptor number that occur prior to gene expression changes [19], which was expected to happen post translationally [20]. We observed that up-regulated Ob-Ra expression was followed by elevated ERK1/2 phosphorylation, in line with previous reports that demonstrated ERK1 and ERK2 activation by the short leptin receptor isoform [21], and leptin-stimulated pro-inflammatory cytokine release abrogation by ERK 1/2 MAPK inhibitor U0126 [22]. As shown from our result, the elevated ERK1/2 phosphorylation was followed by increased NFκB activation and TNFα secretion, which was in agreement with a previous report that indicated leptin has proinflammatory action, involving pro-inflammatory cytokines TNFα through NFκB regulation [22].
Quercetin is one of the most widely distributed flavonoids in fruits, vegetables, tea and wine [23], reported to have anti-inflammatory properties, and might act as health-promoting substances [24,25]. In this study, we showed that 125 μM quercetin was able to downregulate Ob-Ra expression, reduce ERK1/2 phosphorylation, decrease NFκB activation, and partially suppress TNFα secretion in leptin-induced HUVECs. The antiinflammatory effect of quercetin has been described in various reports, among others were the ability to suppress ERK phosphorylation in 3 T3-L1 adipocytes [26] and NCI-H292 cells [27], inhibit NFκB activation in murine J774 macrophages [28] and bone marrowderived macrophages [29], VAT TNFα production in obese Zucker rats [30], reduce the activation of phosphorylated ERK kinase strongly in LPS-induced RAW 264.7 cells and inhibit NFκB activation through both stabilization of NFκB/IκB complex and suppression of proinflammatory cytokines including TNFα [31]. In PBMC, a 50% reduction of TNFα gene expression was observed after 24 h exposure to 50 μM quercetin [14]. A meta-analysis of long-term placebo-controlled human intervention trials reported that TNFα levels were decreased after flavonoid consumption, but only in a fixed model, and a higher dose or a longer duration intervention were not associated with a greater effect size [32]. Although preliminary, our data suggested that suppression of TNFα secretion by quercetin was associated with a b d c Figure 5 Quercetin partially suppressed leptin-induced TNFα secretion. TNFα secretion was markedly increased in leptin-induced HUVECs as compared to control, and was partially decreased after quercetin addition. L0Q0 control HUVECs without leptin and quercetin; L0Q125 without leptin, but with 125 μM quercetin; L500Q0 with 500 ng/mL leptin, but without quercetin; L500Q125 with 500 ng/mL leptin and 125 μM quercetin. Means without a common letter differ, at least p < 0.05.
inhibition of NFκB activation, as reported previously [28]. However, our result showed that quercetin suppression of TNFα secretion appeared to be partial, which we believe might be due to the short duration of quercetin exposure period. Although essential, unfortunately we can not perform further confirmation of TNFα suppression by TNFα mRNA assessment, since our samples were not prepared and stored for RNA analysis. This is the limitation of our study, besides that we only apply one quercetin concentration in our study, which might not be the optimum concentration. Further studies, such as measurements of TNFα mRNA and application of NFkB activation inhibitor, are still needed to clarify this discrepancy.
The suppression of inflammation by quercetin may have clinical significance in preventing cardiovascular disease induced by leptin-resistant in obesity. As reported previously, quercetin administration to obese Zucker rats improved dyslipidemia, hypertension, and hyperinsulinemia, all of which are cardiovascular risk factors [30]. A deeper assessements on the molecular mechanisms of quercetin action in suppressing leptin-induced inflammation are needed as high intake of plant-derived food rich in quercetin or the use of supplements of this flavonoid might be protective and could reduce cardiovascular risks. It might also be a promising area for the development of a flavonoid-based neutrapharmaceutical agents for the treatment of chronic inflammatory disease.

Conclusion
Leptin exposure of HUVECs resulted in up-regulated Ob-Ra expression and elevated ERK1/2 phosphorylation and NFkB activation, and increased TNF-alpha secretion. Despite strong suppression effect of quercetin on the increased Ob-Ra expression, ERK1/2 phosphorylation and NFkB activation, however, TNFα secretion was only partially suppressed. Therefore, further studies are still needed to elucidate the basis of this exception.