Abstract
Studies have shown that endothelial insulin resistance induced by oxidative stress contributes to vascular dysfunction in metabolic disorders. Quercetin, a natural antioxidant, has been recently shown to exert protective effects on endothelial function. However, the effects of quercetin on endothelial insulin resistance and its underlying mechanism are unclear. Here, we found that chronic oral treatment of obese mice with quercetin increased vascular endothelial insulin sensitivity, accompanied by alleviated mitochondrial fragmentation as revealed by confocal imaging. In addition, western blot analysis showed that quercetin treatment suppressed the levels of dynamin-related protein 1 (Drp1) and phosphorylation at serine 616 in endothelial cells of obese mice. Mechanistically, quercetin specifically suppressed Drp1 phosphorylation at serine 616, whereas it showed little effects on the Drp1 level and its phosphorylation at serine 637 in cultured endothelial cells under oxidative stress. Furthermore, our results also showed that quercetin suppressed Drp1 phosphorylation at serine 616 by inhibiting PKCδ as revealed by western blot analysis. Knockdown of PKCδ with siRNA alleviated the protective effects of quercetin on endothelial-mitochondrial dynamics and insulin sensitivity. These results suggest that chronic oral treatment with quercetin exerts endothelial protective effects through inhibition of PKCδ and the resultant mitochondrial fragmentation.
Introduction
Obesity represents one of the major health problems in modern societies, which is characterized by insulin resistance and strongly associated with an increased risk of vascular diseases [1,2]. Evidence has shown that endothelial dysfunction, an early step in the pathogenesis of obesity and other metabolic disorders, is a central pathologic contributor to cardiovascular morbidity and mortality [3–5]. Recently, endothelial insulin resistance has garnered much attention in the study of endothelial dysfunction in metabolic disorders, including obesity [6–8]. Besides its essential role in modulation of metabolism, insulin also induces vasodilation by stimulating the production of NO from endothelial cells. Endothelial insulin resistance as manifested by decreased vasodilation in response to insulin has been shown to contribute to vascular dysfunction [6,9]. Evidence has shown that improvement of endothelial insulin sensitivity alleviates vascular abnormalities in both animal and clinical investigations [10].
It has been suggested that oxidative stress, as manifested by increased oxygen reactive species (ROS) generation or decreased anti-oxidative capacity, is a major cause of endothelial dysfunction in metabolic disorders [3,5,6,11,12]. Various sources of ROS within endothelial cells have been implicated in the pathogenesis of endothelial dysfunction, including xanthine oxidase, nicotinamide adenine dinucleotide phosphate oxidase, uncoupled endothelial NO synthase (eNOS), and the mitochondria [13–15]. Oxidative stress induces eNOS uncoupling with increased ONOO− formation and reduced NO bioavailability, leading to impaired endothelium-dependent relaxation [5,16,17]. It also impairs insulin signaling in endothelial cells through oxidation of multiple molecules [18,19]. Thus, rebalancing redox balance has become one of the most promising approaches to improve endothelial and vascular function [20–22], and natural antioxidants from the human diet have recently garnered much attention in the prevention of metabolic diseases.
Quercetin, the most abundant flavonoid in the human diet, exerts potent free radical scavenging and antioxidant activities [23]. Several studies have revealed that chronic oral treatment with quercetin restores endothelial function in metabolic disorders, such as hypertension and diabetes [24,25]. However, the effects of quercetin on endothelial insulin sensitivity and its underlying mechanism in obesity are unclear. Here, we found that quercetin improves endothelial insulin sensitivity through inhibition of mitochondrial fragmentation in obese mice.
Materials and Methods
Animals
All animal experiments were performed following International Guidelines for the Care and Use of Laboratory Animals and were approved by the Animal Care and Research Ethics Committee of Renmin Hospital of Pudong New District. C57 mice (male, 8 weeks old) were purchased from Charles River (Beijing, China) and were used in this study. The mice were fed with either a standard normal diet (ND) or a high-fat diet (HFD) for 16 weeks. The ND consists of 10% of calories derived from fatty acids, and HFD consists of 45% of calories derived from fatty acids. After 10 weeks of feeding with ND or HFD, mice were treated orally by gavage with vehicle (1% w/v methylcellulose) or quercetin (10 mg/kg body weight) once a day for 6 weeks as previously described [22–24]. The treatments ended 2 days before the subsequent experiments.
Functional assessment of mouse aortas
The descending aorta was carefully excised from the mice and cut into ring segments (1 mm long) as previously described [6,7]. Briefly, the contractile force was detected using a temperature-controlled myograph (model 610M; Danish Myo Technology, Copenhagen, Denmark) and incubated in physiological saline solution (PSS). After a 40-min equilibration, PSS with high KCl (60 mM) was used to test the viability of vascular smooth muscle. Aortic rings were precontracted with phenylephrine (PE; 10 μM). Endothelium-dependent vasodilation evoked by cumulative acetylcholine (Ach) (10−10 M to 10−5 M) or insulin (10−10 M to 10−6 M) and endothelium-independent vasodilation evoked by cumulative sodium nitroprusside (SNP) (10−10 to 10−5 M) were detected and expressed as the percentage of PE-induced contractile force.
NO detection
Total NO production was determined using NO assay kit (EMSNO; Invitrogen, Carlsbad, USA) in the culture medium of isolated aortas or cultured cells according to the protocols provided by the manufacturer.
Cell culture
Primary endothelial cells were isolated, as previously described [26]. Isolated endothelial cells or human umbilical vein endothelial cells (HUVECs; Cell Applications, San Diego, USA) were cultured in endothelial growth medium-2 (EGM-2) containing 5% fetal bovine serum and SingleQuot Kit supplements (Lonza, Redwood, USA) as previously described [6].
Fluorescence detection
Images were captured using an inverted confocal microscope (LSM 800; Zeiss, Jena, Germany). For the detection of intracellular ROS, DCFH (5 μM; Invitrogen) was loaded for 10 min and washed three times. DCF fluorescence was excited at 488 nm, and emission was collected at 540–625 nm. A solution of mitoSOX (5 μM; Invitrogen) was loaded for 20 min to detect mitoSOX fluorescence. mitoSOX fluorescence was excited at 488 nm, and emission was collected at 540–625 nm. Mitochondrial network was monitored using mitoTracker Green (0.5 μM; Invitrogen). The mitoTracker fluorescence was monitored by exciting at 488 nm and collecting at >500 nm. All experiments were performed at room temperature (22°C~25°C).
siRNA transfection
siRNA specifically targeting PKCδ mRNA and the negative control were designed and provided by GenePharma (Shanghai, China). The siRNA sequences are 5′-AGTACTTGGCAAAGGCAGC-3′ against PKCδ, and 5′-CAGTCGCGTTTGCGACTGG-3′ as scramble control. The siRNA was transfected into endothelial cells using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. The transfected cells were used for further experiments after 60 h of culture.
Western blot analysis
The protein level was detected by western blot assay as previously described [6]. Briefly, the membranes were probed with anti-mfn1, anti-mfn2, anti-OPA1, anti-Drp1, anti-p-Drp1 (Ser637), anti-p-Drp1 (Ser616), anti-Akt, anti-p-Akt, anti-eNOS, anti-p-eNOS, anti-fis1, anti-PKCδ, or anti-GAPDH antibodies (Abcam, Eugene, USA) overnight at 4°C, followed by incubation with the corresponding secondary antibodies (Abcam) at room temperature for 1 h. Blots were visualized using secondary antibodies conjugated with IRDye (LI-COR, Lincoln, Nebraska) and an Odyssey imaging system (LI-COR). The intensity was quantified by ImageJ (NIH, Bethesda, USA).
Statistical analysis
All values are presented as the mean ± SEM. Data were compared with two-way ANOVA test, with all ANOVA test, followed by an unpaired t test, as appropriate. Bonferroni’s correction for multiple comparisons was used. Differences were considered significant when P < 0.05.
Quercetin treatment improved endothelial insulin sensitivity in obese mice (A) Body weight of obese mice treated with quercetin. (B) Quercetin treatment decreased intracellular ROS levels in endothelial cells of the isolated aortas from obese mice. Intracellular ROS was detected by DCF fluorescence. Scale bar, 100 μm. The typical images and the quantified results are shown. (C) Quercetin treatment increased NO level in the isolated aortas from obese mice. (D) Quercetin treatment improved ACh-induced vasodilation in the isolated aortas from obese mice. (E) Quercetin treatment showed little effect on SNP-induced vasodilation in the isolated aortas from obese mice. (F) Quercetin treatment improved insulin-induced vasodilation in the isolated aortas from obese mice. (G) Quercetin treatment increased insulin-stimulated Akt and eNOS phosphorylation in endothelial cells isolated from obese mice. Data were analyzed using two-way ANOVA in B–G, followed by unpaired t test. n = 6. *P < 0.05, **P < 0.01 vs control. #P < 0.05 vs obesity.
Quercetin treatment attenuated mitochondrial fragmentation in aortic endothelial cells of obese mice (A) Quercetin treatment improved endothelial mitochondrial dynamics in isolated endothelial cells from obese mice. Mitochondrial morphology was detected using mitoTracker. Scale bar, 50 μm. (B) Quercetin treatment decreased mitochondrial ROS levels in endothelial cells of the isolated aortas from obese mice. Mitochondrial ROS was detected by mitoSOX fluorescence. Scale bar, 100 μm. The typical images and the quantified results are shown. (C) The typical images and quantified results of major mitochondrial dynamics-related proteins in isolated endothelial cells from obese mice treated with quercetin. Data were analyzed using two-way ANOVA in B and D, followed by unpaired t test. n = 6. *P < 0.05, **P < 0.01.
Results
Quercetin treatment improved endothelial insulin sensitivity in obese mice
To test the endothelial protective potential of quercetin in vivo, control and obese mice were treated with quercetin for 6 weeks. As shown in Fig. 1A, quercetin treatment displayed no effect on body weight in both control and obese mice. To investigate the effects of quercetin on endothelial cells, aortic rings were isolated. Quercetin treatment decreased intracellular ROS levels in endothelial cells of isolated aortas and increased NO levels in isolated aortas from obese mice (Fig. 1B,C). For vascular function, quercetin treatment improved endothelial-dependent vasodilation as evidenced by improved vasodilation induced by ACh (Fig. 1D). However, quercetin treatment showed little effect on SNP-induced vasodilation (Fig. 1E). Importantly, endothelial insulin resistance was observed in obese mice after 16 weeks of HFD treatment, but quercetin treatment improved endothelial insulin sensitivity as evidenced by improved vasodilation induced by insulin (Fig. 1F). Endothelial insulin sensitivity was further confirmed by detection of insulin signaling in isolated endothelial cells. Insulin-stimulated phosphorylation levels of Akt and eNOS were decreased in endothelial cells isolated from obese mice, which were increased in endothelial cells isolated from quercetin-treated obese mice (Fig. 1G). These results suggested that quercetin treatment improved endothelial function and insulin sensitivity in obese mice.
Quercetin inhibited Drp1 phosphorylation at serine 616 in cultured endothelial cells (A) Quercetin decreased Drp1 phosphorylation at serine 616 in a dose-dependent manner. The typical images and the quantified results are shown. (B) Quercetin treatment decreased both Drp1 expression and Drp1 phosphorylation at serine 616 in endothelial cells under oxidative stress. The typical images and the quantified results are shown. (C) Quercetin treatment attenuated mitochondrial fragmentation in endothelial cells under oxidative stress. Mitochondrial morphology was detected using mitoTracker. Scale bar, 50 μm. (D) Quercetin treatment improved insulin-induced NO bioavailability in endothelial cells under oxidative stress. (E) Quercetin treatment improved insulin signaling in endothelial cells under oxidative stress. The typical images and the quantified results are shown. Data were analyzed using two-way ANOVA in A–E, followed by unpaired t test. n = 6. *P < 0.05, **P < 0.01.
Quercetin inhibited Drp1 phosphorylation through inhibition of PKCδ (A) Quercetin decreased PKCδ expression in endothelial cells in a dose-dependent manner. The typical images and the quantified results are shown. (B) Quercetin treatment decreased PKCδ expression in endothelial cells under oxidative stress. The typical images the quantified results are shown. (C) PKCδ silencing with siRNA attenuated the inhibitory effect of quercetin on Drp1 phosphorylation at serine 616 in endothelial cells under oxidative stress. The typical images and the quantified results are shown. (D) PKCδ knockdown attenuated the improved NO bioavailability by quercetin in endothelial cells under oxidative stress. (E) PKCδ knockdown attenuated the improved insulin signaling by quercetin in endothelial cells under oxidative stress. The typical images and the quantified results are shown. Data were analyzed using two-way ANOVA in A–E, followed by unpaired t test. n = 6. *P < 0.05, **P < 0.01.
Quercetin treatment attenuated mitochondrial fragmentation in aortic endothelial cells of obese mice
To investigate the underlying mechanism of quercetin in the improvement of endothelial insulin sensitivity, endothelial morphology was detected in isolated endothelial cells. Notably, quercetin treatment improved endothelial mitochondrial morphology in isolated endothelial cells from obese mice (Fig. 2A). Mitochondria exhibited a long filamentous morphology in the endothelial cells of aortas from control mice, while they were fragmented in obese mice (Fig. 2A). In addition, quercetin treatment also attenuated mitochondrial ROS levels in aortas of obese mice (Fig. 2B). These results suggested that quercetin improved endothelial mitochondrial morphology and function. Next, the major mitochondrial dynamics-related proteins were detected in isolated endothelial cells from control and obese mice. The levels of mfn1, mfn2, fis1, OPA1, and phosphorylated Drp1 at serine 637 showed no significant change in endothelial cells of obese mice compared with that of control mice, while the levels of Drp1 and its phosphorylation at serine 616 were increased (Fig. 2C). It has been suggested that Drp1 phosphorylation at serine 616 promotes mitochondrial fission, whereas Drp1 phosphorylation at serine 637 inhibits mitochondrial fission. Importantly, quercetin treatment decreased both total Drp1 level and the level of phosphorylated Drp1 at serine 616 in obese mice (Fig. 2C). These data indicated that quercetin may regulate endothelial mitochondrial morphology through regulation of Drp1.
Quercetin inhibited Drp1 phosphorylation at serine 616 in cultured endothelial cells
To examine the effects of quercetin on Drp1 level and its phosphorylation, cultured endothelial cells were treated with quercetin. As shown in Fig. 3A, treatment with quercetin for 12 h showed no significant effect on total Drp1 and Drp1 phosphorylation at serine 637, but it decreased Drp1 phosphorylation at serine 616 in a dose-dependent manner. Oxidative stress is a major cause of endothelial dysfunction, thus the effects of quercetin on mitochondrial Drp1 were also detected in endothelial cells under oxidative stress. To exclude the effect of quercetin on Drp1 phosphorylation at serine 616 under control condition, a short-term treatment (2 h) with quercetin was given, in which condition quercetin (10 μM) showed no significant effect on Drp1 phosphorylation at serine 616 in endothelial cells without oxidative stress (Fig. 3B). H2O2 treatment (20 μM for 2 h) increased both Drp1 level and Drp1 phosphorylation at serine 616, while quercetin treatment decreased both Drp1 level and Drp1 phosphorylation at serine 616 in endothelial cells under oxidative stress (Fig. 3B). As a result, quercetin treatment also attenuated mitochondrial fragmentation in endothelial cells under oxidative stress (Fig. 3C). In addition, quercetin treatment improved endothelial function as evidenced by elevated insulin sensitivity, increased NO bioavailability, and activation of insulin signaling in endothelial cells under oxidative stress (Fig. 3D,E). These results indicated that quercetin may improve endothelial insulin sensitivity through inhibition of Drp1 phosphorylation at serine 616.
Quercetin inhibited Drp1 phosphorylation through inhibition of PKCδ
It has been reported that PKCδ is an upstream signal in the regulation of serine 616 phosphorylation of Drp1 and promotes mitochondrial fragmentation under oxidative stress [26]. Quercetin treatment for 12 h decreased PKCδ expression in endothelial cells under control conditions in a dose-dependent manner (Fig. 4A). In addition, H2O2 (20 μM for 2 h) increased PKCδ expression, and quercetin treatment for 2 h (10 μM) decreased PKCδ expression in endothelial cells under oxidative stress (Fig. 4B). Knockdown of PKCδ with siRNA inhibited Drp1 phosphorylation at serine 616 in response to H2O2 challenge and attenuated the effect of quercetin on the inhibition of Drp1 phosphorylation at serine 616 (Fig. 4C). Furthermore, silencing of PKCδ mRNA attenuated the protective effects of quercetin on endothelial cells under oxidative stress as shown by the unchanged insulin sensitivity in endothelial cells (Fig. 4D,E). These results suggested that quercetin improved endothelial insulin sensitivity through inhibition of PKCδ.
These results reinforced our notion that quercetin treatment improves endothelial mitochondrial dynamics and endothelial insulin sensitivity through inhibition of PKCδ and subsequent inhibition of Drp1 phosphorylation at serine 616.
Discussion
Evidence has shown that endothelial insulin resistance contributes to vascular dysfunction in various metabolic disorders [6,8]. Thus, therapeutic interventions that improve endothelial insulin sensitivity will attenuate vascular abnormalities. Here, we found that quercetin, a natural antioxidant from the human diet, improves endothelial insulin sensitivity in obese mice. Mechanistically, quercetin improves endothelial insulin sensitivity through inhibition of PKCδ and the resultant mitochondrial fragmentation. These results shed light on the potential therapeutic applications of quercetin in the intervention of vascular diseases.
One class of natural antioxidants garnering much attention is the flavonoids which are widely found in plant-based human diets. Among flavonoids, quercetin is the most abundant compound in food, which has been shown to exert antioxidant actions in vivo [23,27]. Recently, quercetin has been shown to exert endothelial protective effects in various metabolic diseases [24,25,28,29]. However, the effects of quercetin on endothelial insulin sensitivity and its underlying mechanism are unclear in obesity. Here, we found that chronic oral treatment of obesity with quercetin exerted significant protective effects of endothelial function, with increased endothelial insulin sensitivity and endothelium-dependent vasodilation. Insulin-induced vascular relaxation plays an essential role in maintaining both hemodynamic and metabolic homeostasis in healthy individuals [30,31]. In metabolic disorders, impaired insulin signaling in endothelial cells leads to decreased NO bioavailability and endothelial dysfunction. Thus, improvement of endothelial insulin sensitivity is a potential strategy for the intervention of vascular diseases. Quercetin, as a natural antioxidant, could be used to improve endothelial function in metabolic disorders.
Mitochondria, highly dynamic organelles, undergo cycles of fusion and fission to maintain their networks. Mounting evidence has shown that mitochondrial dynamics is essential in the regulation of mitochondrial function, and mitochondrial fragmentation has been shown to be involved in the induction of various pathological processes [32–34]. However, in contrast to other cell types, mitochondria content in endothelial cells is modest, and the major source of ATP production is independent of mitochondria [35,36]. Thus, mitochondria in endothelial cells have not been well studied previously. Recent studies have provided evidence that mitochondrial fragmentation is a contributor to endothelial dysfunction in metabolic disorders, including diabetes, atherosclerosis, and myocardial reperfusion injury [37–39]. Here, our data provided evidence that endothelial mitochondrial fragmentation contributes to vascular dysfunction in obesity and the improvement of mitochondrial dynamics protects endothelial cells against obesity. These findings suggested that although mitochondria contribute a little to endothelial energy metabolism, they are essential in the regulation of endothelial function possibly through regulation of redox and other signals.
Mitochondrial dynamics is regulated by a series of proteins, namely mfn1, mfn2, OPA1, Drp1, and fis1. Among these proteins, we found that quercetin suppressed Drp1 activity through inhibition of Drp1 phosphorylation at serine 616. Phosphorylation is one of the critical post-translational modifications of Drp1, which regulates its translocation to mitochondria [40]. Phosphorylation of Drp1 at serine 616 or dephosphorylation of Drp1 at serine 637 promotes mitochondrial fragmentation. Here, we found that Drp1 phosphorylation at serine 616 is regulated by quercetin, while Drp1 phosphorylation at serine 637 is not regulated by quercetin, suggesting that quercetin specifically regulates Drp1 phosphorylation at serine 616. Inhibition of Drp1 phosphorylation at serine 616 increased endothelial insulin sensitivity as evidenced by enhanced Akt and eNOS phosphorylation in response to insulin. Although the underlying mechanism of the Drp1 phosphorylation at serine 616 affecting endothelial insulin sensitivity remains unknown, the possible mechanisms may involve ROS, mitochondrial metabolites, and other factors. In addition, our data also showed that quercetin regulates Drp1 phosphorylation at serine 616 through inhibition of PKCδ. PKCδ is an upstream regulator of Drp1 phosphorylation at serine 616. These results suggested that PKCδ and Drp1 phosphorylation can be used as targets in the intervention of endothelial mitochondrial fragmentation in obesity.
Taken together, we found that chronic oral treatment with quercetin exerted endothelial protective effects in obese mice. Mechanically, quercetin suppressed Drp1 phosphorylation at serine 616 by inhibiting PKCδ, leading to improved mitochondrial dynamics and insulin sensitivity in endothelial cells of obese mice. These findings suggest that endothelial mitochondrial dynamics may be used as a target in the intervention of vascular dysfunction.
Funding
This study was supported by the grants from the Key Discipline Group Construction Project of Shanghai Pudong (No. PWZxq2017-02) and the Featured Clinical Discipline Project of Shanghai Pudong (Nos. PWRq2017-38 and PWYst2018-01).
