【作者】 曾转萍；

【导师】 陈清；

【作者基本信息】 南方医科大学 ， 流行病与卫生统计学， 2014， 博士

【摘要】 研究背景和目的：2型糖尿病也称成人糖尿病,患者以胰岛素分泌相对不足、胰岛功能缺陷和高血糖为主要症状,发病病因和机制非常复杂。2型糖尿病近年来无论患病率还是发病人数均呈上升趋势,已经成为继癌症、心血管疾病后的第三位重大公共卫生问题。该病发病隐匿,早期症状较轻微,并且多数以散发为主,因此大部分2型糖尿病病人是在发病多年后出现并发症(如肾衰、失明、糖尿病足、伤口愈合慢、心血管疾病等)才被诊断出来。并发症的治疗不但耗时长、治疗费用昂贵、而且给国家带来沉重的疾病负担,据世界各国对2型糖尿病疾病负担的预测,糖尿病带来的经济负担呈逐年增加的趋势。卫生组织提出,如果没有效的预防和控制策略,全球糖尿病的疾病负担会持续上升。中国糖尿病的预防形势也非常严峻。从近30年糖尿病的患病率来看,1980年我国的糖尿病患病率为0.67%,1994年糖尿病的患病率为2.5%,而到2007-2008年中国的糖尿病患病率已经升至9.7%,目前中国已经成为2型糖尿病人数最多的国家之一,由于糖尿病导致的疾病负担非常严重,中国糖尿病的控制刻不容缓。广州市位于中国南部,2型糖尿病的患病率与北京、上海等城市水平相当,是2型糖尿病的重灾区。由于广州市经济较为发达,人口众多,人群中可能影响糖尿病发生的饮食因素、行为因素以及其他的环境因素方面差别可能较大,因此,研究广州市特定人群2型糖尿病的危险因素,可为特定人群2型糖尿病的循证预防提供科学依据。目前认为2型糖尿病是基因与环境因素共同作用导致的疾病。在基因研究方面已经发现了超过50个基因与2型糖尿病的发生相关,但是基于已经发现的基因与2型糖尿病的归因危险度较小,而且不同民族和地区2型糖尿病的易感基因有差异,新的易感基因的发现对于人群2型糖尿病的预防依然重要。HDAC基因与SIRT1基因均属于脱乙酰蛋白酶类基因,新近文献报道HDAC基因能影响小鼠的血糖代谢,但是否与人群2型糖尿病发生相关尚需实验证明。SIRT1基因已经被证明能够影响Puma印第安人2型糖尿病的发生,基于2型糖尿病基因普遍存在种族差异,与中国人群2型糖尿病的发生是否有关联值得进一步研究；此外,SIRT1基因与HDAC基因能否共同影响2型糖尿病的发生对糖尿病的预防也至关重要。环境因素方面,流行病学研究发现超重或肥胖人群发生2型糖尿病相对风险增加；越来越多的流行病学研究指出增加体力活动可以降低2型糖尿病的发病率。饮酒与2型糖尿病的发生呈U型的曲线,吸烟不管主动还是被动吸烟均增加2型糖尿病的发病风险。另外,也有研究表明不同的饮食模式与2型糖尿病的发病关系不同。SIRT1基因被认为与饮食和脂肪代谢有关,这些因素是否与和HDAC基因共同作用,增加2型糖尿病的发病风险值得研究。本研究的目的是分析HDAC, SIRT1基因及其与环境因素的相互作用对2型糖尿病发病风险的关系,为高危人群的发现、疾病的控制策略的制定以及未来针对性的靶点药物的研发提供科学依据。方法：采用病例对照研究方法选择广州市某农村社区人群中的2型糖尿病病人为病例,并以同社区非糖尿病的人群为对照。病例的纳入标准为空腹血糖≥7.0mmol/L的受试者,从空腹血糖介于3.1-6.0mmol/L的研究对象中抽取对照。所有纳入研究的病例均为新发现的病例,不管病例还是对照以前均没有2型糖尿病和糖耐量异常的疾病史。病例和对照均来自同一社区。排除标准为近期患有感冒、慢性炎症、急性疾病、代谢性疾病和感染性疾病的受试者。采用Hapmap数据库进行中国汉族人群易感基因位点的选择。选择标准为中国北京汉族人群,最小等位基因频率(MAF)>0.05,连锁不平衡参数D’=1,r2>0.8。结果选取了HDAC基因rs2530223、rs11741808、rs2547547、和rs1741981共4个位点；SIRT1基因rs4746720、rs10509291、rs2236319和rs10823116共4个位点。研究对象特定基因SNPs的变异采用Taqman MGB探针法对易感基因进行检测。PCR方法被用于易感基因位点的扩增,易感基因的全序列采用测序的方法进行。贝克曼680全自动生化仪被用于研究对象的血糖、总胆固醇、甘油三酯等血清学项目的检测。血压是通过台式血压计结合听诊器由有经验的护士通过测量得出的。另外,身高和体重也是由有经验的护士通过测量得出的。研究对象的人口学特征、饮酒、吸烟、饮食因素等则通过调查员与研究对象面对面问卷调查得出,问卷调查是一对一进行的。SPSS13.0和Stata12.0被用于统计学分析。两独立样本t检验或χ2检验被用于对病例组和对照组进行均衡性检验；两独立样本t检验被用于对符合正态分布的连续变量进行统计学检验,χ2检验被用于分类变量的统计学检验中。利用非条件Logistic回归模型分析环境因素及基因多态性与2型糖尿病的关系。校正因素为性别、年龄、BMI、高血压等。基因之间的相加交互作用；高甘油三酯、高总胆固醇和高BMI与基因之间的相加交互作用用叉生分析方法。相乘交互作用通过多因素Logistic回归分析纳入乘积项进行。以P<0.05为差异有统计学意义。结果：共有568名研究对象参加了研究。284名糖尿病人的平均年龄为65.64±8.71岁,284名对照的平均年龄为66.83±8.91岁,两组年龄差别没有统计学意义(t=-1.605,P=0.109)。研究对象中女性有392名,病例和对照均为196人,两组性别差别没有统计学意义(χ2=0.000,P=1.000)。婚姻、教育程度和经济状况在两组间差别也没有统计学意义(χ2=0.428,P=0.513；χ2=1.464,P=0.226和Z=-0.667,P=0.505)。临床特征与2型糖尿病研究方面,2型糖尿病人的甘油三酯为1.91mmol/L (0.60-30.22mmol/L),正常对照为1.21mmol/L (0.50-7.39mmol/L),差别有统计学意义(Z=-7.888,P<0.001)；2型糖尿病人的总胆固醇含量为5.50mmol/L (2.40-12.60mmol/L),正常对照为5.20mmol/L(2.80-8.10mmol/L),差别有统计学意义(Z=-3.401,P=0.001)；2型糖尿病人的BMI为(23.77±3.50)kg/m2,正常对照为(22.86-3.51)kg/m2,两组BMI差别有统计学意义(t=3.507,P=0.003)。行为因素与2型糖尿病研究方面,研究对象中糖尿病组吸烟者有73人,占25.80%,对照组吸烟者有72人,占25.44%,吸烟与2型糖尿病差别没有统计学意义(χ2=0.009,P=-0.923)。在是否参加锻炼方面,2型糖尿病组每天锻炼半小时以上占50.35%,正常对照每天锻炼半小时以上的占27.56%,参加锻炼与2型糖尿病差别有统计学意义(χ2=39.685,P<0.001)。在定期检查方面,2型糖尿病组有规律检查的占32.04%,正常对照有规律检查的占17.79%,两组在定期检查方面差别有统计学意义(χ2=19.796,P<0.001)。2型糖尿病组有40.14%的对象有慢性病史,正常对照有30.04%有慢性病史,两组在慢性疾病史方面差别有统计学意义(χ2=6.355,P=0.012)。饮食因素与2型糖尿病研究方面,2型糖尿病组爱吃酸味食品的占11.27%,正常对照爱吃酸味食品的占19.72%,两组爱吃酸味食品构成比差别有统计学意义(χ2=7.745,P=0.005)。2型糖尿病组爱吃辣味食品的占3.17%,正常对照爱吃辣味食品的占7.04%,两组爱吃辣味食品差别有统计学意义(χ2=4.397,P=0.036)。2型糖尿病组有24.30%的对象爱喝牛奶,正常对照有35.91%爱喝牛奶,两组爱喝牛奶方面差别有统计学意义(χ2=9.111,P=0.003)。在饮食的口味上,2型糖尿病组口味偏咸的占24.65%,正常对照口味偏咸的占16.55%,两组差别有统计学意义(χ2=36.128,P<0.001)。在每天吃蔬菜方面,2型糖尿病组每天吃得少的占11.62%,正常对照每天吃得少的占8.10%,两组吃蔬菜差别有统计学意义(χ2=30.968,P<0.001)。在食用油方面,2型糖尿病组常吃动物油者占13.38%,正常对照常吃动物油占2.11%,两组常吃动物油方面差别有统计学意义(χ2=25.227,P<0.001)。在爱吃白肉方面,2型糖尿病组占43.66%,正常对照组占58.66%,两组爱吃白肉方面差别有统计学意义(χ2=12.756,P<0.001)。基因多态性与2型糖尿病的研究方面,HDAC基因的rs2530223、rsl1741808、rs2547547和SIRT1基因的rs4746720位点与2型糖尿病有关联。rs11741808与2型糖尿病的OR为0.54(95%CI=0.36-0.81)；rs2547547与2型糖尿病的OR为1.72(95%CI=1.13-2.64)；rs2530223与2型糖尿病的OR为1.96(95%CI=I.04-3.68),rs4746720与2型糖尿病的OR为1.42(95%CI=1.02-1.98)。分层分析显示在BMI≥23kg/m2,高甘油三酯和高血压的人群中携带rs11741808AG基因型发生2型糖尿病的几率比携带AA基因型的低,OR分别为0.50(95%CI=0.27-0.91)、0.38(95%CI=0.20-0.71)和0.43(95%CI=0.24-0.76)。在总胆固醇正常人群中,携带rs11741808AG基因型与携带AA基因型的研究对象比较,发生2型糖尿病的风险降低,OR为0.42(95%CI=0.25-0.70)。对于rs2547547,在总胆固醇正常和甘油三酯正常人群中,携带AG基因型的对象与携带AA基因型的对象比较发生2型糖尿病的风险增加,OR分别为1.92(95%CI=1.17-3.15)和2.24(95%CI=1.28-3.94)。对于rs4746720,高甘油三酯组,携带CC或者TT基因型发生2型糖尿病的风险比携带CT基因型的对象增加,OR为1.85(95%CI=1.06-3.23)；在吃红肉组,携带CC或者TT基因型发生2型糖尿病的风险比携带CT基因型的对象增加(OR=1.43；95%CI=1.01-2.02)。同样,吸烟或者吃甜食的对象,携带CC或者TT基因型发生2型糖尿病的风险高于携带CT基因型者(OR=2.22,95%CI=1.21-4.06; OR=1.65,95%CI=1.10-2.47)。多因素分析结果表明高血压、高甘油三酯、高总胆固醇、BMI≥23kg/m2和rs2547547AG基因型携带者发生2型糖尿病的风险增加,OR分别为2.23(95%CI=1.53-3.27)、2.92(95%CI=1.98-4.31)、1.50(95%CI=0.97-2.32)、2.89(95%CI=1.98-4.21)和1.93(95%CI=1.14-3.27)。相反,rs2530223TT基因型vsrs2530223CT+CC基因型、rs2530223CT基因型vs rs2530223TT+CC基因型和rs11741808AG基因型vs rs11741808AA+GG基因型发生2型糖尿病的风险降低,OR分别为0.42(95%CI=0.20-0.87)、0.33(95%CI=0.16-0.69)和0.51(95%CI=0.32-0.81)。多因素分析结果也表明,喝牛奶、喝豆浆、吃白肉和每天吃蔬菜和低盐饮食发生2型糖尿病的几率也降低,OR分别为0.51(95%CI=0.29-0.88).0.43(95%CI=0.26-0.74)、0.51(95%CI=0.32-0.83)、0.21(95%CI=0.10-0.44)、0.28(95%CI=0.12-0.65)和0.35(95%CI=0.21-0.51)。而吃红肉、高盐饮食、BMI≥23kg/m2,吃动物油和rs4746720CC+TT基因型与rs4746720CT基因型比较,发生2型糖尿病的概率增加,OR分别为2.89(95%CI=1.38-6.01)、2.73(95%CI=1.61-4.64)、3.47(95%CI=2.28-5.28)、27.91(95%CI=9.24-84.32)和1.61(95%CI=1.06-2.44)。关于HDA基因与脂肪代谢的关系,我们发现rS11741808位点与高甘油三酯、BMI≥23kg/m2和高总胆固醇间没有统计学关联,其OR(95%CI)分别为1.35(95%CI=0.90-2.01)、1.00(95%CI=0.67-1.50)和0.96(95%CI=0.61-1.51)。同样我们也没有发现rs2547547及rs2530223位点与脂肪代谢有统计学关联。SITR1基因rs4746720位点与脂肪代谢也没有统计学的差异。结论：(1) HDAC基因rs11741808、rs2547547及rs2530223多态性与2型糖尿病有关：rs11741808AG基因型发生2型糖尿病的几率比携带AA基因型的低；rs2547547AG基因型发生2型糖尿病的风险比携带AA基因高；rs2530223TT基因型发生2型糖尿病的几率比rs2530223CT+CC基因型低。(2) HDAC基因的rs1741981位点的多态性与2型糖尿病发生没有关联。(3) S1RT1基因的rs4746720位点的多态性与2型糖尿病发生有关联:rs4746720CC+TT基因型与rs4746720CT基因型比较,发生2型糖尿病的概率增加。(4) SIRT1基因的rs10509291, rs2236319和rs10823116的多态性与2型糖尿病发生没有关联。(5)经常喝牛奶、爱吃蔬菜、爱吃酸味食品等是中国汉族人群2型糖尿病发生的保护因素。(6)经常吃红肉、高盐饮食、爱吃动物脂肪等是中国汉族人群2型糖尿病发生的危险因素。(7)高血压、高甘油三酯、高总胆固醇和BMI≥23kg/m2是2型糖尿病的危险因素。(8)未发现研究的基因位点间、基因间以及基因与环境因素间与2型糖尿病的关联存在交互影响。更多还原

【Abstract】 Objectives:Type2diabetes mellitus (T2DM) is a metabolic disorder that is characterized by high blood sugar in the context of insulin resistance and relative lack of insulin. But the real pathogenesis and mechanism of T2DM is not known now. T2DM is rising all over the World.In2000, the number of T2DM was171million, however, there will be366million in2030. This increase will strain health systems already facing a high burden of t T2DM and its complication, such as renal failure,lose sight,diabetic foot and angiocarpy disease.From the estimation of WHO, the burden of T2DM will increase rapidly in world if we haven’t effective measure. For example, the burden of T2DM is24,500million in America in2012. And this burdent was higher41%than2007. T2DM combined with cancer and cardiovascular disease had been the three serious diseases in the world.In these thirty years, the prevenlence of T2DM is increase rapily in China too. The prevenlence of T2DM was0.67%in1980and2.5%in1994. However, the rate was up to9.7%in2007-2008. China has been the most number of T2DM in World. During the high disease burdent of T2DM, it needed to play some mesures on T2DM control and prevention in China too. Guangzhou is located in the south of China. The prevance of T2DM is higher than other middle and west city. Because the economic in Guangzhou is higher and abundant of diatery ways and other environment risk factors, risk factors of T2DM in Guangzhou are needed further research.T2DM is a complicated disease and relate to both genetic and environmental factors interact to produce hyperglycemia.There were over50genes had been found associated with T2DM, but the population attribute risk is very small in former research, new risk genes of T2DM also worth further research.There are no data regarding the possible role of the single nucleotide polymorphism (SNP) of class I Histone deacetylases (HDACs) in T2DM. Although SIRT1gene has been reported that it can increase the risk of Puma Indian.However, we haven’t found data in Chinese Han population. Furthermore, HDAC gene and SIRT1gene are deacetylases, the interaction of the two genes and T2DM worth further research.For environment risk factors, epidemiology data show that heavy weight or obesity is the risk factor of T2DM. Phicical activity including walking, exercise, etal will decrease the risk of T2DM.The relation of dringing and T2DM is U shape, little or heavy dringing had little risk of T2DM than normal dringing.Active or passive smoking will increase the risk of T2DM.Furthermore, unhealthy diatery mode will increase T2DM, however little fat and high carbohydrate will decrease T2DM. From the above, we know that T2DM is a complicated disease in which both genetic and environmental factors interact to produce hyperglycemia. We designed this study to examine whether polymorphisms of HDAC gene. SIRT1gene and other factors can be implicated in this disease.Methods:A community-based, case-control study was conducted, with a total of568subjects (284patients and284controls) enrolled. Eight polymorphisms of HDAC1(rs1741981), HDAC3(rs11741808, rs2547547, rs2530223) and SIRT1(rs4746720, rs10509291, rs2236319, rs10823116) were examined by the use of TaqMan technology. Dietary data were collected by an inquiring officer through a face-to-face method. Subjects’ body weights, heights and blood pressure were measured and recorded by a nurse. The levels of fasting serum glucose, triglyceride and total cholesterol was measured by a specialist using a Beckman Coulter AU680. We investigated the gene locus on the linkage disequilibrium and haplotype block analysis of the HapMap project data. SPSS13.0and Stata12.0were used to do the statistic.In demographic characteristics, the differences between the cases and controls were tested using the t test for continuous variables or χ2test for categorical variables; Multivariate logistic regression models were used to assess the effects of other factors and genotype on T2DM, controlling for the demographic characteristics such as gender, age, BMI, blood pressure, family income et al. The interaction between clinical factors and genotypes on T2DM were evaluated by multiplicative models.Results:There were no significant differences in the age or sex distribution between the control and the case (P values are0.109and1.000, respectively). Marital status, education and economic status had no significant differences too (P values are0.513,0.226and0.505).Compared with control, the case subjects showed higher levels of triglyceride and cholesterol, as well as BMI value (P<0.001, P=0.001and P=0.003, respectively).Behavior risk factors and T2DMThere were73(25.80%)smokers in type2diabetes mellitus group,72(25.44%) smokers in control group, no significant differences between the two groups (χ2=0.009, P=0.923). Moreover, there were50.35%population in type2diabetes mellitus group doing exercise over0.5h per day, however, in control group only27.56%population doing exercise over0.5h per day, the differences had statistical significant (x2=39.685, P<0.001).Compare with control, exam at regular interval, chronic disease history population have higher risk of type2diabetes mellitus.(χ2=19.796, P<0.001; x2=6.355, P=0.012, respectively).Diatary factors and T2DM11.27%case group population like to eat sour food,19.72%control group population like to eat sour food, there are significant differences between the two groups (x2=7.745, P=0.005). Otherwise, eating piquancy food, dringing milk, eat vegetable, plant oil and eat white meat will decrease the risk of T2DM (x2=4.397, P=0.036;x2=9.111, P=0.003;x2=30.968,P<0.001;χ2=25.227,P<0.001,χ2=12.756, P<0.001, respectively)We found significant association with risk of T2DM for three SNPs of HDAC3, including rs11741808[odds ratio (OR)=0.54,95%confidence interval (CI)=0.36-0.81], rs2547547[OR=1.72,95%CI=1.13-2.64], and rs2530223[OR=1.96;95%CI=1.04-3.68]. We found significant association with risk of T2DM for rs4746720CC+TT genotype compared with CT genotype [OR=1.42,95%CI=1.02-1.98].Subgroup analysis showed that BMI>23kg/m2, high triglyceride and high blood pressure, together with the rs11741808AG genotype, were associated with a significantly decreased risk for T2DM, with an OR of0.50(95%CI=0.27-0.91),0.38(95%CI=0.20-0.71) and0.43(95%CI=0.24-0.76) compared with AA genotype, respectively. In population with normal total cholesterol, the AG genotype was associated with a significantly decreased risk of T2DM risk, with an OR of0.42(95%CI=0.25-0.70) when compared with the persons of the AA genotype. For rs2547547, in population with normal total cholesterol and triglyceride, the AG genotype was associated with a significantly increased risk of T2DM, with an OR of1,92(95%CI=1.17-3.15) and2.24(95%CI=1.28-3.94) when compared with population carrying AA genotype. Further, we performed stratification analyses for SIRT1to explore the role of the polymorphism in the subgroup population. For rs4746720, subjects with high triglyceride harboring the CC or TT genotype had a significantly increased risk of type2DM [OR=1.85,(95%CI=1.06-3.23)] compared with subjects of the CT genotype. In subjects eat red meat more, the CC or TT genotype significantly increased type2DM risk[OR=1.43,(95%CI=1.01-2.02)] compared with subjects of the CT genotype. In analysis of the effect of sugary food or smoking, individuals with the CC or TT genotype of rs4746720had a significantly increased risk of type2DM compared with individuals carrying the CT genotype[OR=2.22,(95%CI=1.21-4.06)], and [OR=1.65,(95%CI=1.10-2.47)]. Multiple factors tests showed that high blood pressure, high triglyceride, high total cholesterol, BMI≥23and rs2547547AG compared with rs2547547AA+GG had higher risk of T2DM, with OR of2.23(95%CI=1.53-3.27),2.92(95%CI=1.98-4.31),1.50(95%CI=0.97-2.32),2.89(95%CI=1.98-4.21),1.93(95%CI=1.14-3.27), respectively. However, compared with rs2530223CT+CC, rs2530223TT+CC and rs11741808AA+GG, rs2530223TT, rs2530223CT and rs11741808AG had lower risk of type2DM, with OR of0.42(95%CI=0.20-0.87),0.33(95%CI=0.16-0.69),0.51(95%CI=0.32-0.81), respectively. Milk, soy, white meat, vegetables and low-salt diet decrease risk of T2DM, with OR of0.51(95%CI=0.29-0.88),0.43(95%CI=0.26-0.74),0.51(95%CI=0.32-0.83),0.21(95%CI=0.10-0.44),0.28(95%CI=0.12-0.65),0.35(95%CI=0.21-0.51) respectively. Red meat, salty food, BMI≥23kg/m2, use of animal fat and rs4746720CC+TT compared with rs4746720CT yielded higher risks of T2DM, with OR of2.89(95%CI=1.38-6.01),2.73(95%CI=1.61-4.64),3.47(95%CI=2.28-5.28),27.91(95%CI=9.24-84.32) and1.61(95%CI=1.06-2.44) respectively.We found no significant association between rs11741808and high triglyceride, BMI or high total cholesterol, the OR were1.35(95%CI=0.90-2.01),1.00(95%CI=0.67-1.50),0.96(95%CI=0.61-1.51), respectively. Moreover, there were no association between rs2547547, rs2530223and clinical character like BMI, high triglyceride, high total cholesterol too. And there were no association between SIRT1and clinical character like BMI, high triglyceride, high total cholesterol too.Conclusion:(1)The variance of HDAC (rs2530223, rs11741808and rs2547547) contribute to an increased prevalence of T2DM.(2)The variance of HDAC (rs1741981) has no relation with T2DM.(3) The variance of SIRT1(rs4746720) contribute to an increased prevalence of T2DM.(4) The variance of SIRT1(rs10509291, rs2236319and rs10823116) has no relation with T2DM.(5)Dringing milk, eat vegetable, eat sour food, et al were protect factors of T2DM.(6)Eating red meat, salt food and eating animal oil were risk factors of T2DM.(7)High blood pressure, high triglyceride, high total cholesterol an BMI≥23kg/m2were risk factors of T2DM.(8)No interaction relations were found between gene, environmental factors and T2DM.更多还原