【作者】 李敏；

【导师】 徐佩茹；

【作者基本信息】 新疆医科大学 ， 内科学， 2011， 博士

【摘要】 目的：探寻生活环境因素、体脂量和肥胖症相关基因(FTO)多态性、影响能量吸收的肠道共生菌与哈萨克族学龄儿童发生肥胖之间的关联性,初步探讨FTO基因与肠道菌群间的关联,完善我国不同民族肥胖数据库信息,为公共卫生部门采取一级病因防控提供依据。方法：本研究运用病例-对照研究策略：①以哈萨克族主要居住地伊犁哈萨克自治州为调查地点,按照随机整群抽样的方法,分别从城市、县城、乡村共抽取14所学校,由培训过的调查员对全部小学生进行体格测量(身高、体重、腰围、臀围、血压)、填写调查表(包括年龄、性别、家族史、个人史、饮食习惯、活动时间等),按照“中国学龄儿童青少年超重、肥胖筛查体重指数值分类标准”筛选出超重肥胖儿童作为病例组及正常体重儿童作为对照组,签署知情同意书后留取血样,检测肥胖相关代谢表型(采用全生化自动仪检测血糖、血脂,放射免疫法检测胰岛素浓度,血浆超敏C反应蛋白用免疫比浊法测定,采用稳态模型法评估胰岛素抵抗程度)；②按经典酚-氯仿-异戊醇法提取基因组DNA,用聚合酶链反应-限制性片段长度多态性技术检测FTO基因rs9939609、rs9930506多态位点,采用软件SHEsis进行配对连锁不平衡分析及单体型分析；③病例组及对照组儿童中符合大便采样要求的留取大便,提取肠道细菌总DNA,建立肠道主要菌群硬壁菌门、拟杆菌门的标准曲线,通过实时荧光PCR方法检测研究对象的肠道硬壁菌门/拟杆菌门比例。结果：①流行病学调查结果显示：7岁-13岁学龄儿童5360名(男2716名,女2644名),共检出超重儿童489名,肥胖儿童238名。肥胖的总检出率为4.40%(95%可信区间3.85%-4.95%),超重儿童总检出率为9.10%(95%可信区间8.71%-9.49%)。超重检出率男童(9.90%)与女童(8.30%)差异有统计学意义(X2=4.635,P=0.031),肥胖检出率男童(5.00%)与女童(3.90%)差异有统计学意义(X2=4.470,P=0.035)。各年龄组肥胖检出率(X2=46.121,P=0.000)、超重检出率(X2=47.106,P=0.000)差异均有统计学意义,并呈现随着年龄增长检出率下降趋势。超重及肥胖检出率从高到低分别是城市、县城、乡村,差异具有统计学意义(X2=116.075,P=0.000；X2=195.814,P=0.000)。哈萨克族7岁-13岁儿童肥胖组儿童腰围、臀围和腰臀比值明显高于超重、正常体重组儿童,按性别分层后差异仍有统计学意义。进一步做Pearson线性相关分析,腰围(rs=0.811,P=0.000)、臀围(rs=0.783,P=0.000)与BMI呈线性相关。BMI与收缩压(rs=0.468,P=0.000)、舒张压相关(rs=0.371,P=0.000)。运用Spearman线性相关分析腰臀比与血压(正常高值、高血压、正常血压)水平的关系,结果发现腰臀比与血压水平相关。BMI (rs=0.387, P=0.000)、腰围(r-=0.353,P=0.000)均与HOMA-IR呈正相关。有肥胖家族史的儿童发生超重或肥胖的风险是没有家族史的2.699倍；②FTO基因rs9939609、rs9930506的基因型频率和等位基因频率在哈萨克族学龄儿童群体中分布符合Hardy-Weinberg平衡；rs9939609多态位点基因型频率和等位基因频率的分布在哈萨克族学龄儿童超重/肥胖组和对照组中差异有统计学意义(X212.78,P=0.002);rs9930506位点多态的基因型频率和等位基因频率在哈萨克族学龄儿童两组中的分布差异没有统计学意义(x2=1.781,P=0.410)。rs9939609中携带AA等位基因的人群SBP、BMI、腰围、臀围、HDL、FIN、HMOA-IR与TT、AT基因型人群相比,差异有统计学意义；控制性别、年龄的影响后,这种差异仍存在；而每增加1个危险等位基因可引起BMI增加0.490(kg/m2),腰围增加0.530cm, HDL减少0.09mmol/L。rs9930506中携带GG等位基因的人群控制协变量性别、年龄的影响后仅与FPG, FIN、HMOA-IR相关。rs9930506每增加1个危险等位基因所引起血糖增加0.134mol/L。Logistic回归分析显示FTO基因rs9939609位点A等位基因(OR=0.546,95%CI0.397-0.752)、rs9930506位点G等位基因(OR=0.450,95%CI0.236-0.858)与哈萨克族学龄儿童超重肥胖有关联。FTO基因2个SNPs处于不完全连锁平衡,可形成4种单体型：AG、AA、TA和TG,单体型组合的估计频率在哈萨克族学龄儿童超重肥胖组和对照组中差异均有统计学意义；③Bacteroidetes数量在肥胖组、超重组和正常体重组三组间分别是(0.33±0.47)×107拷贝/ul、(1.37±1.76)×107拷贝/ul、(1.50±2.20)×107拷贝/ul,差异有统计学意义(F=5.836,P=0.004)；Firmicutes数量在肥胖组、超重组和对照组分别是(4.70±7.01)×107拷贝/ul、(2.05±3.01)×107拷贝/u1、(2.43±4.5)×107拷贝/ul,差异没有统计学意义(F=0.232,P=0.793)。肠道Bact/Firm比值在肥胖组、超重组和对照组三组分别是0.48±0.52、1.03±0.82、1.06±0.62,肥胖组的两大菌门比例(Bact/Firm)低于超重组、正常对照组,差异有统计学意义(F=6.483, P=0.002)。Bacteroidetes、Firmicutes、Bact/Firm比例在rs9939609、rs9930506不同基因型时差异没有统计学意义(P>0.05)。结论：①超重肥胖检出率与经济发展水平不匹配；超重及肥胖检出率存在男高女低特点,并随着年龄增长检出率下降；超重肥胖检出率从高到低仍然是城市>县城>乡村；哈萨克族学龄儿童BMI、腰围与HOMA-IR正相关；有肥胖家族史的儿童更易发生超重或肥胖;哈萨克族学龄儿童BMI、腰臀比均与其血压水平相关；②FTO基因rs9939609位点A等位基因、rs9930506位点G等位基因与哈萨克族学龄儿童超重肥胖有关联,很可能在该民族群体中有民族特异性肥胖易感单体型存在；③肠道菌群中的Bacteroidetes减少及Bact/Firm降低与哈萨克族学龄儿童超重/肥胖的发生有关系,FTO基因SNPs与肠道菌群间暂未发现关联。更多还原

【Abstract】 Objective:In order to explore the risk factors relevant with obesity in children from both gene level and environmental factor including microflora in intestinal, our group carried out case-control strategy to study all above factors in large Hazakh children sample in Xinjiang region. The results of the study will be helpful to reveal pathogenesis of obesity and supplement database information of obesity in different ethnics in China. And it can also be helpful to supply theory basis for causal prophylaxis and control obesity in childhood by public health branch. Methods:A three-step stratified sampling method was used. The northern region of the Tianshan mountains is the main living area of Kazakh ethnic populations in Xinjiang. At the initial stage, different counties of Kazakh residential areas were randomly selected from the region of Yili and Altay (NW Xinjiang). During the second stage of randomized sampling, several townships or districts were randomly selected from each of the counties and cities. At the final stage, children were randomly chosen from the primary schools which belonged to the townships or districts. A set of questionnaires, including demographic information, family history of obesity, outdoor activities, cultural levels of parents, eating speed, birth weight, hours of sleep and hours of watching TV, were collected from the participants. Three independent blood pressure measurements and anthropometric measuremenst, including height, weight, waist circumference and hip circumference, were obtained by trained and certified staff members using tstandard protocols. The method to determine overweight and obesity in this study was defined by the Group of China Obesity Task Force. With the purpose of exploring risk factors associated with overweight and obesity, blood glucose, total cholesterol, high density lipoprotein, low density lipoprotein and triglycerides were determined after overnight fasting using a modified hexokinase enzymatic method (Hitachi automatic clinical analyzer, Model 7060, Japan). High sensitity C-Reactive Protein (hs-CRP) and insulin also were determined by hs-CRP ELISA Kit and Radioimmunoassay Kit respectively assayed with kits from USCNLIFE and Peking north biotechnology research institute. To assessment the degree of insulin resistance, a homeostasis model (HOM) was calculated using the formula, HOMA-IR= (FPG×FIN)/22.5. Written informed consents were obtained from all study participants prior to data collection and measurements. The pattern of pairwise LD between the common (minor allele frequency> 5%) SNPs was measured by D’and r2, and all SNPs of FTO gene were futher genotyped in all subjects by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). Real-time fluorescence quantitative PCR (Q-PCR) was used to detecte Bacteroidetes, Firmicutes, and Bact/Firm ratio of DNA from stool of Hazakh children. Results:In total,5,360 Kazak subjects (male 2716 and femal 2644) were randomly enrolled from 14 schools which were predominantly Kazak. The detection rate of obesity and overweight was 4.40%(95%C73.85%～4.95%) and 9.10% (95%CI 8.71%～9.49%), respectively. The detection rate of obesity was difference in different age groups (x2=46.121, P=0.000) and in different gender (x2=4.470, P= 0.035). The detection rate of overweight was difference in different age groups (X2= 47.106, P= 0.000) and in different gender (X2=4.635, P=0.031). The tendency of detection rate of obesity was decrease along with age growing. Waist circumference, hip circumference and waist-to-hip ratio in obesity group were higher than overweight group and normal weight group.There was statistical significance among three groups in waist circumference, hip circumference and waist-to-hip ratio. Our data revealed that there existed a linear correlation between waist circumference, hip circumference and body mass index (BMI). And the similar correlation existed among BMI, SBP and DBP. Individuals with a greater waist-to-hip ratio were more prone to development of hypertension. The genotype distributions of the FTO gene variants were found to be in Hardy-Weinberg equilibrium in control group and case group. There was significant difference in frequencies of AA, AT and TT genotypes between controls and cases (X2=12.78, P=0.002), but the frequencies of AA, AG and GG genotypes between controls and cases was no significant difference (X2=1.781, P=0.410). In the case and control groups, SBP, BMI, waist circumference, HDL, FIN, HMOA-IR and hip circumference of the persons with AA genotype of rs9939609 were differences from those of TT or AT genotypes and the statistical significance was also observed (P<0.05). A allele of rs9939609 was associated with increased BMI with per-allele change of 0.490 kg/m2, increased waist circumference with per-allele change of 0.53cm and decreased HDL with per-allele change of 0.09mmol/L when compared with those not inheriting a risk allele. FPG, FIN and HMOA-IR of the persons with GG genotype of rs9930506 were differences from those of AA or AG genotypes elimination the influence of gender and age and the statistical significance was observed (P<0.05). FPG was increased 0.134mol/L in children who are homozygous for the risk allele(GG) comparing with those were not inherited risk allele. Logistic regression analysis showed that both the A allele of rs9939609 (OR=0.546,95%C70.397-0.752) and G allele of rs9930506 (OR=0.450, 95%C70.236～0.858) were an independent risk factor for overweight or obesity. Pairwise LD was carried out for SNPs respectively, rs9939609 and rs9930506 polymorphisms were in LD and most of the pairwise LD values of the polymorphisms were incomplete linkage equilibrium. Hyplotype based association study indicate that 4 haplotypes was reconsreucted by 2 SNPs and haplotype frequency in case group is different from control group in Kazakh population. The number of Bacteroidetes among obese group, overweight group and control group were (0.33±0.47)X 107copy/ul, (1.37±1.76)X107 copy/ul, (1.50±2.20) X107copy/ul and the difference was statistically significant (F=5.836, P=0.004).The amount of Firmicutes among above groups were (4.70+7.01) X107copy/ul, (2.05+3.01) X107copy/ul, (2.43+4.5) X107copy/ul and there were no difference among three groups (F=0.232, P=0.793).The ratio of Bact/Firm among three groups were (0.48±0.52), (1.03±0.82), (1.06±0.62) and the difference was statistically significant (F=6.483, P=0.002). There were no correlation between genetype of rs9930506 and rs9939609 with the number of Bacteroidetes, Firmicutes and ratio of Bact/Firm (P>0.05). Conclusions:The detection rate of overweight or obesity is mismatched with local economical level. The detection rate of overweight or obesity in boys is higher than girls and the tendency is decreasing along with children growing. And the rate of obesity or overweight from high to low is urban, rural counties and countryside. There is positive correlation between BMI and HOMA-IR. The same result exist between waist circumference and HOMA-IR.The results suggest that BMI, the ratio of waist-to-hip is correlated with blood pressure in Kazak children. Both genetype of AA in rs9939609 and GG in rs9930506 are correlate with overweight or obesity in Kazak children. And haplotype that are reconstructed by SNPs of FTO gene might be associated with obesity in Kazak populations. Both number of Bacteroidetes and ratio of Bact/Firm are correlation with overweight or obesity.Besides, there is no correlation between SNPs of FTO with intestinal micro flora.更多还原