Adiponectin and cardiovascular risk factors on the regulation of nitric oxide synthase: Molecular and clinical studies
Date Issued
2005
Date
2005
Author(s)
DOI
zh-TW
Abstract
One of the most important progresses of cardiovascular medicine in the twentieth century is to uncover that endothelial cell dysfunction (ECD) develops in the very early phase of cardiovascular diseases. ECD is the first step in the cellular mechanisms leading to atherosclerosis. ECD, initially introduced to describe defective endothelium-dependent vasorelaxation, has now been broadened to encompass impaired anti-thrombogenic and anti-inflammatory properties; perturbed angiogenic capacity; inappropriate regulation of smooth muslce proliferative capacity and migratory properties. Since recent studies have pointed to the pivotal role of dysregulation of nitric oxide (NO) in the pathophysiology of ECD, it is important to investigate the molecular mechanisms that regulate endothelial nitric oxide synthase (eNOS).
Nitric oxide is generated from the conversion of L-arginine to L-citrulline by the enzymatic action of an NADPH-dependent NO synthase (NOS), which requires Ca2+/calmodulin, FAD, flavin mononucleotide (FMN), and tetrahydrobiopterin (BH4) as the cofactors. In the blood vessels, NO is produced from the endothelium by constitutive expression of the endothelial isoform of eNOS, which can be activated by mechanical stresses, such as shear-stress, and stimulation with agonists, such as bradykinin and acetylcholine. All major risk factors for atherosclerosis, such as hyperlipidemia, diabetes, hypertension, and smoking, are associated with impaired endothelium-dependent vasorelaxation. Although the underlining mechanisms of ECD are divergent, the most important one is the derangement of eNOS/NO pathway, which may be produced by impaired eNOS (either reduced activity or reduced expression of eNOS), decreased sensitivity to NO of the tissues, or increased degradation of NO by superoxide. Among the various mechanisms responsible for ECD, increased NO breakdown by superoxide is especially important as there is augmented production of superoxide in atherosclerotic vessels. Recently, it has been revealed that, under certain circumstances, eNOS would become dysfunctional and produce superoxide rather than NO. The pathophysiological role of eNOS dysfunction in vascular disorders, including atherosclerosis, has attracted much attention in past decades.
Recent studies indicate that eNOS is highly regulated by post-translational modifications, such as Akt- or AMPK-induced phosphorylation and interaction with several regulatory proteins, such as heat shock protein 90 (HSP90). The association between eNOS and HSP90 has recently been shown to be critical in the regulation of eNOS function. In the quiescent state, eNOS is associated with caveolin and remains cell membrane-bound and inactive. When stimulated with vascular endothelial growth factor (VEGF), the eNOS-caveolin complex would be disrupted by Ca(2+)/calmodulin and the association between eNOS and HSP90 promoted. The eNOS-bound HSP90 can then recruit VEGF-activated Akt to the complex to induce phosphorylation of eNOS. The binding of HSP90 to eNOS ensures the transition from the early Ca2+-dependent to the late phosphorylation-dependent activation of eNOS. Failure of this binding can cause eNOS uncoupling and increase eNOS-dependent superoxide anion production, leading to endothelial cell injury. Even though these observations suggest that the association of HSP90 with eNOS is critical in eNOS-associated NO production, few studies have focused on the effects of cardiovascular risk factors, such as hyperglycemia, on the eNOS/HSP90 protein interaction.
The cell experiment part of present investigation (Studies #1 and #2) is intended to investigate the effects of endothelial cell injuring factors, such as hyperglycemia and angiotensin II, on the regulation of eNOS. Study #1 was designed to investigate the effects of hyperglycemia on the regulation of eNOS. Hyperglycemia is the hallmark of diabetes mellitus. Poor glycemic control is correlated with increased cardiovascular morbidity and mortality. High glucose can trigger endothelial cell apoptosis by de-activation of eNOS. Yet, little is known about the molecular mechanisms that regulate eNOS activity during high glucose exposure. The present study was designed to determine the involvement of protein interaction between eNOS and HSP90 by immunoprecipitation in high glucose-induced endothelial cell apoptosis. The protein interactions of eNOS/HSP90 and eNOS/Akt were studied in cultured human umbilical vein endothelial cells (HUVECs) exposed to either control-level (5.5mM) or high-level (33mM) glucose for different durations (2, 4, 6 and 24 h). The results showed that the protein interactions between eNOS and HSP90 and between eNOS and Akt and the phosphorylation of eNOS were all up-regulated by high glucose exposure for 2 to 4 h. With longer exposures, these effects decreased gradually. During early hours of exposure, the protein interactions of eNOS/HSP90 and eNOS/Akt and the phosphorylation of eNOS were all inhibited by geldanamycin, an HSP90 inhibitor. High glucose-induced endothelial cell apoptosis was also enhanced by geldanamycin and was reversed by NO donors. LY294002, a phosphatidylinositol 3 (PI3) kinase inhibitor, inhibited the association of eNOS/Akt and the phosphorylation of eNOS but had no effect on the interaction between eNOS and HSP90 during early hours of exposure.
From our results we proposed that, during early phase of high glucose exposure, apoptosis in HUVECs can be prevented by enhancement of eNOS activity through augmentation of protein interaction between eNOS and HSP90 with recruitment of activated Akt. With longer high glucose exposure, dysregulation of eNOS activity would result in enhanced apoptosis. This finding can be correlated with the observations in animal studies in which endothelial-dependent vasorelaxation is evident in the early stage of diabetes mellius but it deteriorates in the later stage. Clinically, it has been shown that, in humans, the peripheral resistance is decreased and the blood flow increased in early diabetes. In nondiabetic individuals, acute exposure to high glucose also induces vasodilatation through increase in the secretion of endothelium-derived nitric oxide. In our previous study in which cultured human endothelial cells were exposed to high glucose, we demonstrated a biphasic response of eNOS expression, an early up-regulation and a later down-regulation, resulted in an imbalance between NO production and free radical generation. The present study showed that the association between eNOS and HSP90 is also biphasic. This finding provides a molecular basis for the effects of eNOS in the prevention of endothelial cells apoptosis during early phase of high glucose exposure. These observations may contribute to the understanding of the pathogenesis of vascular complications in diabetes mellitus.
In study #2, we investigated the effects of angiotensin II on the regulation of eNOS. In the past, angiotensin II was recognized as a hormone that control blood pressure through the regulation of renal salt and water metabolism, central nervous system mechanisms (thirst and sympathetic outflow), and vascular smooth muscle cell tone. More recently, angiotensin II was found to exert long-term effects on tissue structure, including cardiac hypertrophy, vascular remodeling, and renal fibrosis. Recent clinical observations indicated that the wide-spread use of angiotensin converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs) have resulted in remarkable clinical benefits in decreasing the incidence of stroke, diabetes mellitus, and end-stage renal disease in high risk patients. Studies also demonstrated that angiotensin II can induce endothelial cell apoptosis and ECD. Furthermore, we tested whether a newly found protein, adiponectin, can protect endothelial cells from angiotensin II-induced injury. It has long been well known that adipose tissue serves as a structure for triglyceride (TG) storage and free fatty acid/glycerol release in response to changes in energy demands. In recent years, adipose tissue has been found to participate in the regulation of energy homeostasis and also serves as an important endocrine organ in secreting a number of biologically active adipokines, such as free fatty acid, adipsin, leptin, plasminogen activator inhibitor-1, resistin, and TNFα. Adiponectin is one such adipokine that has recently attracted much attention. Adiponectin is abundantly expressed in adipose tissue. Clinical and animal studies have shown that adiponectin can promote liver and skeletal muscle fat metabolism and reduce insulin resistance. In addition to the metabolic effects, adiponectin was found to protect vessels from damage through various mechanisms. In the present study, we tested the hypothesis that adiponectin can prevent endothelial cell apoptosis induced by angiotensin II through promotion of the association between eNOS and HSP90.
Cultured HUVECs were treated with angiotensin II (2µM) to induce apoptosis. In the presence of globular adiponectin, apoptosis was inhibited in a dose-response manner. Angiotensin II- induced apoptosis was also inhibited by treatment with an NO donor and by combined treatment with both angiotensin II type 1 and type 2 receptor blockers. eNOS phosphorylation was markedly inhibited by angiotensin II treatment and this effect could be prevented by pre-treatment of the cells with either globular adiponectin or combined angiotensin II receptor blockers. These findings indicated that angiotensin II-induced HUVEC apoptosis is proceeded through angiotensin II receptor. HUVECs were pre-treatment with adiponectin or radicicol for 1 hour, and then subjected to angiotensin II exposure for 18 hours. Proteins from the cell lysates were subjected to Western blotting and immunoprecipitation for eNOS and HSP90. The results showed that the association between eNOS and HSP90 were significantly inhibited by angiotensin II. Pre-treatment of the cells with globular adiponectin could restore the eNOSHSP90 association and the phosphorylation of eNOS while radicicol, an HSP90 inhibitor, reversed the effects of adiponectin on eNOS and HSP90.
In our study, we showed that angiotensin II-induced human endothelial cell apoptosis can be prevented by adiponectin through promotion and stabilization of the association between eNOS and HSP90. From our study, we disclosed an interesting novel finding that the inhibitory effect of angiotensin II on the formation of eNOSHSP90 complex is mediated through angiotensin II receptor. One previous in vitro study has demonstrated that angiotensin II receptor can directly bind eNOS to exert its subsequent inhibitory activity. Although the molecular mechanisms of the detrimental effect of angiotensin II is still unclear, it is possible that, by stabilizing the eNOS/HSP90 complex, binding of angiotensin II to its receptor can be interfered and the angiotensin II-induced de-activation of eNOS activity can be prevented. This study provided new mechanisms for angiotensin II in its effects on endothelial cell injury and for adiponectin in the protective effect against vascular injury. The mechanism of adiponectin is not clear and further study is warranted (Difficult in understanding this sentence.).
The clinical observation part of present investigation (Studies #3 and #4) were designed to investigate the mechanisms of ECD in two groups of patients with high cardiovascular risk: familial primary hypercholesterolemia and morbid obesity. Familial primary hypercholesterolemia is associated with excessive premature cardiovascular mortality and morbidity. Animal and cell studies have proved that adiponectin can prevent and reverse vascular injuries. Clinical observations also have shown that low adiponectin level is associated with coronary artery disease and atherosclerosis. Although several studies have proved that decreased serum adiponectin is associated with dyslipidemia and its underling etiology has been attributed to insulin resisitance, the serum adiponectin status has never been studied in patients with familial primary hypercholesterolemia. The aim of our Study #3 was to measure serum adiponectin level in a group of young patients with primary familial hypercholesterolemia and determine its correlation with insulin resistant status. For simplicity, we choice a group of young patients (<30 year-old) without clinical manifestations of metabolic syndrome. Since genetic mutations were not confirmed in these patients, we called our study subjects as “familial related severe primary hypercholesterolemia” (FRSPH).
We included 23 young patients with FRSPH in Study #3. The inclusion criteria were as follows: 1. In the proband and at least 2 first-degree relatives, the low-density lipoprotein cholesterol (LDL-C) should be above 4.921mM, with the presence of tendinous xanthomas. 2. The serum high density lipoprotein (HDL-C ) and triglycerides (TG) should be in normal ranges. Subjects with any condition which may cause secondary hyperlipidemia, such as nephritic syndrome, obstructive liver disease, hypothyroidism, diabetes mellitus (DM) and subjects who were taking drugs which may affect lipid metabolism were excluded. Forty-six age- and- sex matched healthy controls were included for comparison. The serum adiponectin, fasting sugar, insulin , lipids, systolic and diastolic blood pressure (SBP and DBP) and anthropometrical indices, such as body weight, waist circumference and body height were obtained. The homeostasis model assessment (HOMA) was calculated to estimate the insulin resistant status. The results showed that compared with healthy controls, patients with FRSPH had a significantly lower mean serum adiponectin level (7.7±1.8 μg/ml vs. 10.1±4.3 μg/ml, p=0.013). The HOMA score was not different between theses two groups. After adjustment for HOMA and associated covariates, multiple linear regression analysis showed that patients with FRSPH are significantly associated with hypoadiponectinemia.
Our study showed that, the serum adiponectin levels are lower in the young FRSPH patients than in the age and sex-matched control subjects. This may contribute to the high incidence of cardiovascular diseases in this patient population and it is possible that increase in serum adiponectin may reduce the development of cardiovascular diseases. Further more, we found that in young FRSPH subjects with extremely high serum LDL-C levels, the serum adiponectin levels were even lower after adjustment of other insulin sensitivity markers. These observations indicate that the correlation between hypercholesterolemia and hypoadiponectinemia may be mediated through certain pathways independent of insulin resistance. It is well known that cytokines such as TNFα and Interleukin-6 can inhibit the secretion of adiponectin from adipocytes. It is postulated that extreme hypercholesterolemia in FRSPH patients may elicit secretion of cytokines which in turn decrease the secretion of adiponectin.
Study #4 investigated the status of nitric oxide production and the regulating mechanisms involved in patients with morbid obesity both before and after weight reduction surgery. In the developing countries, the prevalence of obesity is increasing rapidly in adults as well as in children. Obesity is demonstrated as an important risk factor for coronary heart disease, ventricular dysfunction, stroke and cardiac arrhythmias. Studies have demonstrated that, in children, obesity is independently associated with ECD. Weight reduction with very-low-calorie diet in obese adults was recently demonstrated to be able to reverse ECD. Reduced bioavailability of vascular NO is believed to contribute to obesity-associated ECD. Processes that can decrease NO bioavailability include impaired synthesis of NO and increased production of reactive oxygen species that convert NO to a pro-inflammatory mediator. Insights regarding these processes can be obtained by measuring metabolites of NO in blood. The plasma level of stable metabolites of NO (NOx), mostly contributed by vascular endothelial cells, have been adopted as a reliable measurement of NO production in the body. The present study was designed to measure the changes of NO production before and after gastric partition surgery in a group of morbidly obese patients. Our aims were to investigate the effects of obesity on the regulation of nitric oxide production and to delineate possible factors which might be associated with nitric oxide production. A group of age-and-sex matched health controls were also included for comparison. Body mass index, blood pressure, serum levels of lipids, high sensitivity C-reactive protein (hs-CRP), HOMA, adiponectin, total nitrite and nitrate (NOx), and 8-iso-prostaglandin F2α (8-iso-PGF2α, the lipid peroxidation product), were obtained in the healthy controls and in the obese subjects both before and after gastric partition surgery. We included 69 mobidly obese patients and 69 health controls. The results showed that at baseline, the serum levels of lipids, glucose, insulin, hs-CRP and 8-iso-PGF2α, and HOMA were all higher in the obese group than in the control group. The serum NOx levels were not different between the two groups (43.4±22.8 vs. 48.9±22.2μM/L, p=0.148). For all subjects, the baseline serum NOx levels were positively associated with male gender, serum triglyceride (TG) and adiponectin levels, as analyzed by multiple linear regression analysis. In obese patients, the baseline serum NOx was only positively associated with the serum TG levels. For healthy controls, the serum NOx levels were positively associated with serum adiponectin levels but negatively associated with HOMA. Three to six months after weight reduction surgery, the body weight, waist circumference and diastolic blood pressure decreased significantly. The serum NOx levels decreased from 43.3±22.8μM/L to 24.4±12.5μM/L(p<0.001). Other serum markers, including hs-CRP, 8-iso-PGF2α, fasting sugar, insulin, total cholesterol and triglyceride levels, decreased significantly while the serum adiponectin levels increased (4.5±3.6 vs. 6.4±3.4μg/mL, p<0.001). The changes of serum NOx levels after weight reduction surgery were positively associated with the changes of body mass index and serum TG levels.
This study showed that, in severely obese subjects, the NO production was down-regulated after weight reduction surgery, suggesting an excessive production and/or inactivation of NO in these patients before surgery. We proposed that, in the present study, the decreased NO production after weight reduction surgery might reflect the weight loss-associated down-regulation of iNOS and the reversal of NO overproduction. Our results also suggested that, in healthy subjects, the NO production is negatively associated with the insulin resistant status and positively associated with the serum adiponectin levels, while in obese state, the NO production is positively associated with the serum TG levels rather than the oxidative stress or inflammatory markers. These findings imply that mechanisms other than oxidative stress might contribute to the regulation of NO production in severely obese patients.
Significance and clinical implication
In conclusion, our cell studies showed that the protein association between eNOS and HSP90 can enhance the phosphorylation and efficiency of eNOS. Cardiovascular injuring factors such as long exposure to high glucose and angiotensin II can interfere with this association and lead to endothelial cell injury. Furthermore, we found that adiponectin can protect endothelial cells from injury through stabilization of the eNOS/HSP90 complex. These observations all pointed to the pivotal role of the dysregulation of eNOS/HSP90 association in the pathogenesis of ECD. Our studies provide a new direction for the associated studies on the regulation of eNOS and ECD.
We also investigated the pathogenesis of two groups of patients with high cardiovascular risk. We found that FRSPH patients without clinical evidence of insulin resistance have lower serum adiponectin levels. This finding not only provides a possible mechanism and treatment for the excessive mortality and morbidity of the patients, but also disclosed a new alternative pathway for the development of hypoadiponectinemia in these patients.
Finally, we studied the regulation of eNOS in the extremely obese patients before and after weight reduction surgery. We found that the production of NO decreased significantly after weight reduction, indicating that ECD in obese patients is contributed by de-activation of NO rather than decreased production. Also, we found that the changes of NO production after weight reduction are associated with changes of serum TG. This study discovered a possible novel mechanism for the pathogenesis of ECD in extremely obese patients.
Subjects
一氧化氮合成酵素
一氧化氮
內皮細胞功能失常
熱激蛋白
脂泌素
細胞凋亡
第二型血管張力素
肥胖
endothelial nitric oxide synthase
nitric oxide
endothelial cell dysfunction
heat shock protein
adiponectin
apoptosis
angiotensin II
hyperglycemia
familial hypercholesterolemia
obesity
SDGs
Type
text
File(s)![Thumbnail Image]()
Loading...
Name
ntu-94-D90421003-1.pdf
Size
23.31 KB
Format
Adobe PDF
Checksum
(MD5):ecd5a46237bfed77c7eeb57034f4d644