【作者】 张军霞；

【导师】 陈兵；

【作者基本信息】 第三军医大学 ， 内科学， 2014， 博士

【摘要】 背景：高血压病是最常见的肥胖相关性疾病。肥胖性高血压病因十分复杂，迄今尚未完全阐明。高尿酸血症(Hyperuricemia, HUA)是肥胖患者常见的代谢紊乱状态。近年来，有大量研究发现，血尿酸与多种心血管疾病和代谢性疾病密切相关，如高血压病、肥胖、2型糖尿病、代谢综合征等。这些疾病中，血尿酸与高血压发病机制之间的联系引起研究者极大的兴趣。尽管具体的机制尚不清楚，但来自流行病学、临床试验和动物实验的数据表明，高尿酸血症参与了肥胖性高血压的发病。文献报道，脂肪组织几乎表达肾素-血管紧张素系统(Renin-angiotensin system,RAS)的全部组分[1]。动物实验的研究数据表明，白色脂肪组织所表达的血管紧张素原(Angiotensinogen, AGT)和血管紧张素Ⅱ(angiotensin Ⅱ, AngⅡ)是循环池RAS的重要来源，在机体血压调控中发挥重要作用[2]。脂肪组织AGT缺乏的小鼠血浆AGT水平更低，收缩压随之降低[3]。研究发现，原发性高血压患者血浆肾素浓度与血尿酸水平明显正相关[4]。然而，血浆AGT浓度与血尿酸之间在肥胖的原发性高血压患者中是否存在关联，目前尚不清楚。迄今为止，研究者发现有多种因素与脂肪组织RAS的调控相关[5]，包括机体的营养状态、胰岛素、糖皮质激素、游离脂肪酸、雄激素、肿瘤坏死因子α(Tumor necrosisfactor alpha, TNFα)以及环磷腺苷(Cyclic adenosine monophosphate, cAMP)等。体外实验发现，尿酸可以对几种组织RAS发挥调控作用，包括永生化人系膜细胞(Immortalizedhuman mesangial cells, ihMCs)组织RAS[6]、人血管内皮细胞(Human vascular endothelialcells, HVECS)组织RAS[7]、血管平滑肌细胞(Vascular smooth cells, VSMCs)组织RAS[8]。肥胖患者脂肪组织RAS处于过度活跃状态[9]。针对这种有趣的现象，最常见的解释是：与体型正常的个体比较，肥胖患者体内脂肪含量明显增多。研究发现，尿酸与脂肪组织RAS均在肥胖性高血压发病中发挥重要作用，由此推断，尿酸很可能对脂肪组织RAS表达也发挥调控作用。众所周知，肥胖性高血压是一种氧化应激相关性心血管疾病[10]。多个研究证实，在一定条件下，尿酸可以诱导氧化应激发生。据报道[11]，高浓度尿酸作用于3T3-L1脂肪细胞，可以明显增加细胞内活性氧簇(Reactive oxygen species, ROS)水平。虽然发现尿酸是通过上调NADPH氧化酶(NADPH oxidase, NOX)活性诱导脂肪细胞氧化应激的，然而，研究者并未阐明尿酸增加NOX活性具体的分子机制。针对以上问题，本课题尝试通过体外实验和临床研究两个层面进行深入探索。首先，利用3T3-L1脂肪细胞作为实验模型，我们研究尿酸是否可以对脂肪组织RAS发挥调控作用。接下来，我们探讨脂肪组织RAS激活在尿酸诱导的氧化应激发病分子机制中的作用。最后，以324例未治疗的原发性高血压患者为研究对象，我们观察血尿酸与血浆AGT浓度是否存在关联，以及这种关联是否因肥胖状态而存在差异。以上研究将为阐明尿酸参与肥胖相关性心血管疾病(尤其是高血压病)的发病机制提供新的视角。方法：1.以不同浓度的尿酸(0、1、5、15mg/dl)作用未分化的前脂肪细胞48小时，或以高浓度尿酸(5、15mg/dl)作用于分化第4天的3T3-L1脂肪细胞48小时。实时RT-PCR检测AGT mRNA水平。2.不同浓度尿酸(0、1、5、15mg/dl)作用于分化成熟的3T3-L1脂肪细胞48小时，或以15mg/dl尿酸作用于成熟脂肪细胞不同时间(0、24、48、96小时)。部分细胞在15mg/dl尿酸联合2mM probenecid条件下孵育48小时。实时RT-PCR检测脂肪组织RAS基因表达，酶联免疫吸附实验(Enzyme linked immunosorbent assay, ELISA)检测培养液AngⅡ的蛋白表达量。3.成熟脂肪细胞在15mg/dl尿酸联合10-4M氯沙坦或10-4M卡托普利条件下孵育48小时。分别以实时RT-PCR和ELISA法检测AGT mRNA水平和AngⅡ的蛋白分泌量。以比色法定量检测NOX的酶活性。4.成熟脂肪细胞以高浓度尿酸(5、15mg/dl)干预48小时。另外，部分细胞在15mg/dl尿酸联合RAS抑制剂(10-4M氯沙坦或10-4M卡托普利)条件下孵育，部分细胞在15mg/dl尿酸联合10mM N-乙酰半胱氨酸(N-acetyl-L-cysteine, NAC)或200μMapocynin (一种NOX抑制剂)条件下孵育。荧光探针2’,7’-二氯双乙酸盐(2′,7′-dichlorofluorescin diacetate, DCFH-DA)或二氢罗丹明(Dihydrogen rhodamine123, DHR)或NBT法检测细胞内活性氧含量。定性检测采用荧光显微镜，定量检测采用荧光酶标仪。5.临床研究共纳入162例未经治疗的肥胖的男性高血压患者，同期纳入162例与之年龄相匹配的男性非肥胖高血压患者。人体学指标由专门护士测量，包括血压、身高、体重、腰围、臀围等。全自动生化分析仪检测空腹血糖、血尿酸、血肌酐和血脂水平。空腹胰岛素(Fasting insulin levels, FINS)水平以放射免疫法检测。稳态模型评估的胰岛素抵抗指数(Homeostasis model assessment of insulin resistance, HOMA-IR)用于衡量机体胰岛素抵抗程度。血浆AGT浓度以ELISA法检测。6.肥胖高血压组与非肥胖高血压组组间指标比较采用独立样本t检验或卡方检验。肥胖高血压组与非肥胖高血压组按尿酸三分位数分别再划分为三个亚组，采用单因素ANOVA进行组间比较，采用Bonferrroni’s post hoc检验进行两两比较。尿酸与其他参数的相关性采用偏相关分析(调整年龄、吸烟史、饮酒史)。肥胖高血压人群血尿酸对血浆AGT水平的独立影响以多元回归模型进行分析与评估。统计分析采用SPSS17.0软件，作图采用GraphPad prism5.0软件。结果：1.不同浓度的尿酸对未分化的前脂肪细胞AGT mRNA表达无明显作用(P>0.05)。然而，高浓度尿酸(5、15mg/dl)可以使分化中的3T3-L1脂肪细胞AGT mRNA表达明显增加(均P<0.05)。2.分化成熟的脂肪细胞在生理浓度尿酸(1mg/dl)作用下，AGT mRNA变化不明显(P>0.05)，但高浓度尿酸(5、15mg/dl)作用可以使脂肪组织RAS基因表达和AngⅡ蛋白分泌量明显上调(均P<0.05)。15mg/dl尿酸作用于脂肪细胞24小时，AGT mRNA变化不明显(P>0.05)；作用48小时或96小时，AGT mRNA表达和AngⅡ蛋白产量明显增加(均P<0.05)。而且，尿酸对脂肪RAS的调控作用具有浓度依赖性和时间依赖性特点。另外，阴离子转运子(Organic anion transporter, OAT)抑制剂probenecid可以明显减弱尿酸对脂肪组织RAS的活化作用。3.与单纯高尿酸组(15mg/dl)比较，10-4M氯沙坦或10-4M卡托普利联合高尿酸组AGT mRNA和AngⅡ蛋白水平均明显减少(均P<0.05)。并且，氯沙坦和卡托普利可以使高浓度尿酸激活的NOX酶活性显著下降(均P<0.05)。4.高浓度尿酸(5、15mg/dl)使脂肪细胞内ROS水平明显增加(均P<0.05)。RAS抑制剂氯沙坦和卡托普利可以明显减少高浓度尿酸作用下的脂肪细胞内ROS水平(均P<0.05)。而且，抗氧化剂NAC或NOX抑制剂apocynin联合高尿酸培养脂肪细胞，也能降低细胞内ROS水平(均P<0.05)。5.肥胖高血压患者腰围、腰臀比和体重指数显著升高。此外，与非肥胖高血压患者比较，肥胖高血压患者饮酒比例增高，甘油三酯、空腹血糖、血尿酸、AGT、FINS、HOMA-IR及收缩压升高(均P﹤0.05)。按尿酸三分位数将患者划分为三个亚组，结果显示，肥胖高血压患者最高尿酸三分位数组(血尿酸435.9~642.2μmol/L)与最低尿酸三分位数组(血尿酸282.7~373.6μmol/L)比较，AGT (P﹤0.001)、FINS (P=0.002)、HOMA-IR(P=0.007)均明显升高。然而，非肥胖高血压患者组按尿酸三分位数划分后，没有发现类似差异(P>0.05)。6.调整年龄、吸烟史、饮酒史后，偏相关分析结果显示，肥胖高血压患者血尿酸与AGT(r=0.437, P﹤0.001)、FINS(r=0.245, P=0.002)、HOMA-IR(r=0.237, P=0.003)明显相关。然而，非肥胖高血压患者血尿酸与AGT、FINS以及HOMA-IR无明显相关性(P>0.05)。多元逐步回归分析结果显示，交互变量“肥胖×尿酸”独立影响高血压患者血浆AGT水平(β=0.257, P﹤0.001)。结论：1.尿酸可以上调分化中和分化成熟的脂肪组织RAS表达，但对未分化的前脂肪细胞无此作用。尿酸必须通过尿酸盐转运子进入脂肪细胞内，才能发挥对脂肪组织RAS的调控作用。2.脂肪组织RAS过度激活参与了尿酸诱导的氧化应激发病的分子机制：尿酸上调脂肪组织RAS，增加AngⅡ产生，继而激活NOX活性、增加脂肪细胞内ROS水平，最终诱导氧化应激发生。3.高血压患者血尿酸水平与血浆AGT浓度明显正相关，这种关联具有“肥胖依赖性”的特点。此外，肥胖高血压患者血尿酸与空腹胰岛素以及胰岛素抵抗水平密切相关。更多还原

【Abstract】 Background:Hypertension is one of the most common findings in obese patients. Also, elevatedserum uric acid levels usually occur in obesity. In recent years, multiple evidences havedemonstrated that serum uric acid closely correlates with cardiovascular diseases andmetabolic diseases, such as hypertension, obesity, type2diabetes mellitus, and metabolicsyndrome. Among these diseases, the roles of uric acid in the pathogenesis of hypertensionattract more attention. Data from epidemiological studies, clinical trials and animalexperiments have indicated that hyperuricemia (HUA) contributes to the occurrence ofobesity-related hypertension, despite of the unclear underlying mechanisms.It is well documented that adipose tissue almost expresses all components of therenin-angiotensin system (RAS)[1]. It was found that angiotensinogen (AGT) andangiotensin Ⅱ(AngⅡ) derived from white adipose tissue contributed to circulation pool ofRAS[2], which in turn had an influence on blood pressure regulation. Mice with deficientAGT in adipose tissue had lower plasma AGT and decreased blood pressure[3]. In someobservational survey, plasma renin was found to be positively related to serum uric acidlevels in hypertensive patients[4]. However, almost no research pays attention to theassociation between plasma AGT concentration and serum uric acid levels in obese patientswith essential hypertension.Up to now, many factors were found to be associated with the regulation of adiposetissue RAS[5], including nutrition condition, insulin, glucocorticoid, free fatty acid,androgen, tumor necrosis factor alpha (TNFα), and cyclic adenosine monophosphate(cAMP). Uric acid was reported to exert effect on RAS expression in immortalized humanmesangial cells (ihMCs)[6], human vascular endothelial cells (HVECS)[7]and vascularsmooth cells (VSMCs)[8]. Obesity is characterized by overexpression of adipose tissue RAS[9]. This interesting phenomenon is commonly explained by increased fat mass in obesepatients. However, in view of the roles of both uric acid and adipose tissue RAS in obesity hypertension, it could be postulated that uric acid might also regulate the expression ofadipose tissue RAS.It is well known that obesity hypertension is one of oxidative stress-relatedcardiovascular diseases[10]. Experiments in vitro and in vivo have suggested that uric acidcould result in oxidative stress under some conditions. In particular, an increase inintracellular reactive oxygen species (ROS) was reported to be triggered by highconcentrations of uric acid in3T3-L1adipocytes[11]. Unfortunately, the molecular pathwayunderlying is not clear, through which uric acid upregulates NADPH oxidase (NOX)activity and then results in oxidative stress.Therefore, in the present study, we tried to solve the problems mentioned above bothin vitro experiments and in population study. First, we observed the effect of uric acid onadipose tissue RAS regulation using3T3-L1adipocytes as an experimental model. Next,we testified the hypothesis that adipose RAS could play a role in uric acid-inducedoxidative stress. Last but not the least, we investigated the association between serum uricacid levels and AGT concentration in untreated patients with obesity hypertension. Thesedata might provide new insights into the mechanism by which uric acid is involved incardiovascular diseases related with obesity, especially hypertension.Methods:1. The undifferentiated pre-adipocytes were cultured with uric acid (0、1、5、15mg/dl)for48hours. In addition, when the adipocytes were differentiating on the fourth day, uricacid (5、15mg/dl) were added into the medium for48hours. AGT mRNA was detected withreal time RT-PCR.2. The differentiated adipocytes were incubated with uric acid at differentconcentrations (0、1、5、15mg/dl) for48hours or with15mg/dl uric acid for differentperiods of time (0、24、48、96hours). In some plates, the adipocytes were cultured in thepresence of15mg/dl uric acid with or without probenecid (2mM) for48hours. The levelsof adipose RAS gene were detected with real time RT-PCR. The production of AngⅡprotein was determined by enzyme linked immunosorbent assay (ELISA).3. In the presence of15mg/dl uric acid, the differentiated adipocytes were culturedwith losartan (10-4M) or captopril (10-4M) for48hours. The levels of AGT mRNA andAngⅡ protein were measured by real time RT-PCR and ELISA, respectively. NOX activity was quantitatively detected with colorimetry method.4. The differentiated adipocytes were incubated under high concentrations of uricacid (5、15mg/dl) with or without RAS inhibitors (10-4M losartan or10-4M captopril) for48hours. Some plates were treated with10mM N-acetyl-L-cysteine (NAC) or200μMapocynin. The levels of intracellular ROS were detected using fluorescence probe2′,7′-dichlorofluorescin diacetate (DCFH-DA) or DHR (dihydrogen rhodamine123) or NBTassay. The fluorescence microscope was used for qualitative detection. Thefluorescence microplate was used for quantitative detection.5.162obese and162non-obese male patients with untreated essential hypertensionwere enrolled in the population study. Anthropometry indexes, including blood pressure,height, weight, waist circumference, and hip circumference, were measured by a speciallyassigned nurse. Biochemical indicators, including blood glucose, serum uric acid, creatinine,and blood lipid, were detected using a fully automatic biochemical analyser. Fasting insulin(FINS) levels were determined by radioimmunoassay method. Homeostasis modelassessment of insulin resistance (HOMA-IR) was used for assessment of insulin resistancestatus. Plasma AGT concentrations were assayed with ELISA method.6. Independent-samples t-test or chi-square test was used for comparison ofvariables between obesity hypertension group and non-obesity hypertension group.One-way ANOVA was used for comparing the differences among subgroups dividedaccording to uric acid tertiles. If a difference existed, further analysis was performed withBonferrroni’s post hoc test. The correlation coefficient between serum uric acid and othervariables was calculated with partial correlation analysis after adjustment for age, smokingratio and alcohol assumption ratio. Multiple variables regression analysis was used fordetermining the independent effect of uric acid on AGT levels in obese patients withhypertension. Statistical analysis was performed using SPSS17.0. Illustrations were drawnwith software GraphPad prism5.0.Results:1. Uric acid at different concentrations had no effect on AGT mRNA expression inundifferentiated3T3-L1adipocytes (P>0.05). However, uric acid, at high concentrations(5、15mg/dl), resulted in an increase in AGT mRNA in differentiating3T3-L1adipocytes(both P<0.05). 2. At physiological concentration (1mg/dl), uric acid played no role in the expressionof AGT mRNA in differentiated3T3-L1adipocytes (P>0.05). However, highconcentrations of uric acid (5、15mg/dl) significantly upregulated both RAS mRNAexpression and AngⅡ protein production (all P<0.05). AGT mRNA and AngⅡincreasedwhen differentiated adipocytes were cultured with15mg/dl uric acid for48and96hours(all P<0.05), but not for24hour. Moreover, the regulation effect of uric acid on adiposetissue RAS is in a dose-dependent and a time-dependent way. In addition, probenecid, akind of organic anion transporter (OAT) inhibitors, attenuated the effect of uric acid onadipose RAS regulation (P<0.05).3. When compared with15mg/dl uric acid, both10-4M losartan and10-4M captoprilblunted the increase in the over expression of AGT mRNA and AngⅡ protein induced byhigh concentration of uric acid (all P<0.05). Furthermore, RAS inhibitors couldsignificantly reduce NOX activity when mature adipocytes were cultured in15mg/dl uricacid (all P<0.05).4. The intracellular ROS was significantly increased by high levels of uric acid (5,15mg/dl) in mature3T3-L1adipocytes (all P<0.05). However, when cells at15mg/dl uricacid were treated with RAS inhibitors losartan or captopril simultaneously, an increased inintracellular ROS expression was preventable in differentiated3T3-L1adipocytes (allP<0.05). Moreover, both antioxidant NAC and NOX inhibitor apocynin could ameliorateROS excessive activation in adipocytes (all P<0.05).5. In addition to waist circumference, waist-to-hip ratio, weight and body mass index(BMI), there was a significant increase in alcohol consumption ratio, systolic bloodpressure (SBP), triglycerides, fasting blood glucose, serum uric acid, plasma AGT levels,fasting insulin, and HOMA-IR in obese patients compared to non-obese patients (all P﹤0.05). When patients were divided into three subgroups based on uric acid tertiles, it wasdemonstrated that AGT (P﹤0.001), FINS (P=0.002), and HOMA-IR (P=0.007) had amarked increase in the highest tertile (435.9~642.2μmol/L serum uric acid) compared tothe lowest tertile (282.7~373.6μmol/L serum uric acid). However, there was no obviouschange in levels of AGT, FINS, and HOMA-IR among non-obese patients when subdividedaccording to uric acid levels (P>0.05).6. After adjustment for age, smoking ratio and alcohol assumption ratio, partial correlation analysis showed that, in obesity group, serum uric acid positively associatedwith AGT, FINS, and HOMA-IR with correlation coefficient0.437(P﹤0.001),0.245(P=0.002), and0.237(P=0.003), respectively. However, AGT, FINS, and HOMA-IR werefound not to be related to serum uric acid levels in non-obese hypertensive patients(P>0.05). Furthermore, multiple variables analysis using stepwise regression modelindicated that obesity×uric acid (standardized coefficient0.257, P﹤0.001) independentlycontributed to plasma AGT levels in untreated hypertensive patients.Conclusions:1. Uric acid could upregulate adipose tissue RAS expression in differentiating anddifferentiated3T3-L1adipocytes, but not in undifferentiated preadipocytes. Uric acid mustenter into adiposytes via urate transporter to play its role.2. Uric acid could upregulate adipose tissue RAS, increase AngⅡproduction, andthen activate NOX, ultimately result in oxidative stress. These findings indicate that overactivation of adipose RAS is implicated in the pathogenesis of uric acid-induced oxidativestress in adipose tissue.3. Serum uric acid was found to be positively associated with plasma AGT levels inan obesity-dependent manner in essential untreated hypertensive patients. Also, serum uricacid obviously correlated with the elevated insulin levels and insulin resistance in obesehypertensive patients.更多还原