- 學位論文
糖尿病的分子病因學之研究：糖尿病的遺傳流行病學與胰島細胞功能的研究
Molecular pathogenesis of diabetes:Study of diabetic genetic epidemiology and pancreatic beta cell functions
摘要
最近幾十年來全球糖尿病的發生率節節升高，花費在糖尿病及其相關併發症的治療費用亦有相當驚人的成長，是公共衛生的ㄧ大課題。總體而言，90至95%為第2型糖尿病，且此型糖尿病也有發生年齡年輕化的趨勢，兒童與青少年第2型糖尿病的發生率也較過去增加。至於第1型糖尿病的發生率也有逐年增加與年輕化的趨勢。 第2型糖尿病的病因目前認為是因為β細胞功能衰退與胰島素阻抗所造成的。最近有個accelerator hypothesis被提出， 認為孩童時期體重增加過多會加速第1型糖尿病的發生並增加其發生率，因此認為第1型與第2型糖尿病有共通的病因。此假說認為第1型糖尿病患者與第2型糖尿病患者類似，同樣有胰島素阻抗而使β細胞過度分泌胰島素，而加速β細胞功能的衰退。 第1型與第2型糖尿病都是多基因遺傳的疾病，且基因與環境之間的交互作用均會影響其發病的風險。截至目前為止，已有幾個研究報告第1型糖尿病與第2型糖尿病有家族內聚集的傾向，第 1型糖尿病患者其家族史中有第2型糖尿病患者的比例比非糖尿病患者的家屬為高。因此第1型與第2型糖尿病可能有共同的遺傳基因，值得去尋找可能的共同易罹病基因，找出高危險的個體，及早從飲食運動方面教育病人以預防糖尿病的發生。 胰島β細胞的胰島素分泌缺陷與周邊組織的胰島素阻抗是造成第2型糖尿病的兩個主要的原因。β細胞分泌胰島素功能障礙與胰島素阻抗在疾病發生的早期即存在，且會隨著時間更加惡化。從UKPDS的前瞻性研究發現，不管使用何種現有的治療，包括sulphonylureas，biguanides和胰島素，胰臟的β細胞功能均會隨著糖尿病罹病時間的增加而持續下降。所以如何預防並延緩胰島β細胞功能的衰退將是治療糖尿病患的一大課題。 因此本論文將分三部分: (1) 台灣人VDR基因多型性變化與第1型 糖尿病罹病風險的相關性研究 (2) 台灣人β2-adrenoreceptor(ADRB2)基因的Arg16Gly多型性變化與第2型糖尿病罹病風險和罹病年齡的相關性研究 (3) TZDs類藥物rosiglitazone改善胰島細胞分泌功能的分子機轉；藉由遺傳流行病學來找出糖尿病的易罹病基因，並透過研究改善胰島β細胞功能的分子機轉來延緩β細胞功能的衰退，希望能找到新的糖尿病藥物標的，以改善糖尿病的照護品質。 第一部分: 台灣人VDR基因多型性變化與第1型糖 尿病罹病風險的相關性研究 雖然vitamin D receptor(VDR)基因的位置並不包含在由系統性基因體研究所確認的與第1型糖尿病相關的位置上，但由過去的ㄧ些研究發現可推論維他命D與其受器(Vitamin D receptor,VDR)與第1型糖尿病的病因有關。而在遺傳流行病學方面的研究，則是在南印度族群其VDR基因BsmI限制酶切點的多型性變化與第 1型糖尿病有相關，但台灣人此基因多型變化與第1型糖尿病的關係則尚未被報告，所以第一部分的研究將探討台灣人VDR基因多型性變化是否會影響第1型糖尿病的罹病風險。利用PCR-RFLP方法，VDR的基因型(B/b for BsmI polymorphism，A/a for ApaI polymorphism，T/t for TaqI polymorphism)可被定出來。本研究共有 157 位第1型糖尿病患者。本研究發現帶有BB基因型得第1型糖尿病的勝算比為6.74，因此帶有BB的個體有較高的危險性得到第1型糖尿病可以解釋悖離Hardy-Weinberg equilibrium，也意謂第1型糖尿病的易罹病基因可能很接近VDR 基因的位置。此外，BsmI、ApaI與TaqI RFLP三種基因型之間並沒有linkage disequilibrium。BsmI與ApaI基因型出現頻率的分布在第1型糖尿病與正常控制組之間有顯著的差異，但TaqI則否。第1型糖尿病患者帶有B與A對偶值(alleles)、BB、Bb、AA與Aa的頻率較高。但是當校正p 值(for multiple comparison)後，兩組之間ApaI基因型的出現頻率不再有統計上有意義的差異，只剩下BsmI仍為有意義的差異。帶有BB基因型的個體有較高的機會罹患第1型糖尿病(勝算比 6.74)，而帶有AA基因型的患者也有較高的機會罹患第1型糖尿病(勝算比2.46)。第1型糖尿病患者的VDR基因型與是否存在GAD65 或ICA512 autoantibody並無關係。 第二部分: 台灣人β2-adrenoreceptor(ADRB2)基因的Arg16Gly多型性變化與第2型糖尿病罹病風險和罹病年齡的相關性研究 ADRB2 基因在人類雖然在編碼地區(coding region)發現有幾個核苷酸多型性變化 (polymorphisms)，但只有3個影響氨基酸編碼(amino acid coding)，包括少見的Thr164Ile變異，與兩個較常見的 Arg16Gly 及 Gln27Glu 變異。在瑞典婦女裡，Glu27被發現是肥胖症的危險因子之一， 帶有homozygous Glu27的個體與與帶有Gln27 allele的人相比有較高的肥胖症比例，較多的身體脂肪，較大的脂肪細胞體積，與較高的空腹胰島素濃度。在日本人方面，Glu27 allele被認為是一個肥胖症和糖尿病的危險因子。相反地，在另一個瑞典人的家族研究發現Gln27 allele不僅是糖尿病的一個危險因子，也與胰島素阻抗相關。這些不同研究結果的差異可能來自於不同的種族和性別。至於Arg16Gly之多型性變化(polymorphism)，在瑞典婦女及日本人的研究發現與肥胖沒有相關性。不過，Gly16 homozygotes的頻率在肥胖的日本婦女比在非肥胖的日本婦女更低；這些不同研究結果的差異可能來自於不同的種族和性別。但台灣人ADRB2之Gln27Glu與Arg16Gly的多型變化與第2型糖尿病的關係則尚未被報告，所以第二部分的研究將探討台灣人此基因多型性變化是否會影響第2型糖尿病的罹病風險與發病年齡的早晚。 在本研究中，Arg16Gly16與Gln27Glu的基因型頻率分布均符合Hardy-Weinberg equilibrium。非糖尿病組codon 16 arginine與glycine的allele frequency分別為52.7%與47.3%; 糖尿病組codon 16 arginine與glycine的allele frequency則分別為58.5%與41.5%。非糖尿病組codon 27 glutamine與glutamate的allele frequency分別為92.3%與7.7%；糖尿病組codon 27 glutamine與glutamate的allele frequency則分別為91.9% 與8.1%。至於兩組間codon 16基因型的分布也達統計學上有意義的差異，帶有Arg16/Arg16(wild type)的個體有較高的比例會罹患糖尿病，其勝算比為1.87(95% 信賴區間:1.34-2.40)。相反地，兩組間codon 27基因型的分布並無差異。無論是疾病組或控制組，codon 16與codon 27之間均無linkage disequilibrium，其p值分別為0.088與0.936。利用logistic regression model發現當調整了年齡、性別與身體質量指數之後，帶有Arg16/Arg16(wild type)的個體有較高的比例會罹患糖尿病，其勝算比為1.894。利用存活分析(survival analysis)也發現帶有Arg16/Arg16(wild type)的個體第2型糖尿病的發病年齡較早(p=0.007, according to log-rank test)。 第三部分: TZDs類藥物rosiglitazone改善胰島細胞分泌功能的分子機轉 過去的研究曾報告troglitazone(一種TZDs類藥物)可 保護β細胞免 於受脂質毒性的破壞，且可恢復β細胞分泌胰島素的能力。此外，給予ZDF rat口服一種新的PPARγ的N – aryl tyrosine activator(GW1929) 14天也可恢復其胰島素的雙相分泌。但是上述這些PPARγ活化劑對胰島細胞的作用都是因為改善葡萄糖毒性和脂質毒性所造成的次發性作用。但是至今thiazolidinedione類藥物對胰臟β細胞的直接作用則有不一致的結論，可能是因為不同的實驗條件所造成的。本研究中，我們首度利用大鼠胰臟灌流系統偵測到rosiglitazone可直接促進葡萄糖刺激的胰島素分泌。我們也發現當葡萄糖濃度為6或10 mM時，rosiglitazone可促進葡萄糖刺激的胰島素分泌，且此促進作用與藥物劑量及作用時間成正相關；當葡萄糖不存在時，只灌流rosiglitazone並無法刺激胰島素分泌。再者，rosiglitazone促進胰島素分泌的效果會被PI3 kinase的抑制劑，LY294002所阻斷，代表rosiglitazone促進胰島素分泌的作用需要PI3 kinase的參與；更重要的是，PI3 kinase抑制劑不會影響葡萄糖刺激的胰島素雙相分泌作用；由此可推論rosiglitazone與葡萄糖刺激胰島素分泌的作用可能是透過不同的訊息傳遞路徑。 雖然在大鼠胰島細胞與MIN6均有表現PPARγ蛋白(Rosen et al., 2003)，但是在大鼠胰臟灌流實驗中，rosiglitazone刺激胰島素分泌的速度非常快，只需幾分鐘即有刺激的效果，因此不太可能透過活化PPARγ蛋白而刺激胰島素的分泌。有趣的是，選擇性的β 細胞PPARγ基因剔除鼠，當以高脂肪食物餵食，給予rosiglitazone治療仍然可改善胰島素敏感性及葡萄糖耐受性，，這代表TZDs類藥物可透過非基因非PPARγ參與的途徑來促進葡萄糖刺激的胰島素分泌。 為了更進一步釐清rosiglitazone刺激胰島素分泌的分子機轉，我們探討AMPK是否參與此作用。過去已有報告指出TZDs類藥物可在骨骼肌、肝臟及脂肪組織活化AMPK，而我們的研究則是第一個報告rosiglitazone可在β細胞促進AMPK第172個胺基酸的磷酸化並提高AMPK 的活性。至於AMPK在β細胞如何被調控則仍不清楚，而在本研究則發現rosiglitazone活化AMPK的作用會被PI3 kinase的抑制劑所阻斷，代表PI3 kinas位於AMPK的上游來調控其活性。有研究也支持我們的發現，即在牛的主動脈內皮細胞中，PI3 kinase位於AMPK的上游參與其活化；但是在大鼠的骨骼肌細胞由AMPK所活化的葡萄糖運送並不需要PI3 kinase的參與，因此有必要進一步探討胰臟β細胞活化AMPK的訊息傳遞路徑。 KATP channel是由四個sulfonylurea receptor (SUR1)與四個inward rectifying K+ subunit (Kir6.2) 所組成。目前已知KATP channel的活性在葡萄糖刺激的生理反應下受到細胞內ATP/ADP 比例的調控 ，也可被某些會直接與SUR1結合的藥物所調控。雖然過去TZDs類藥物曾被報告在胰島素分泌細胞株經由抑制KATP channel的活性可促進葡萄糖刺激的胰島素分泌 ，但其詳細的分子機轉則還不太明瞭。而在本研究中，我們發現RSG須經由PI3 kinase 訊息傳遞路徑來抑制 KATP channel的活性。 是否AMPK可當作rosiglitazone-PI3K signaling與KATP channel之間的橋樑仍需進一步驗證。我們發現在大鼠胰臟灌流系統中以AMPK 活化劑，AICAR短時間的灌流不僅可刺激基礎胰島素的分泌，同時也會促進葡萄糖刺激胰島素的雙相分泌。 在大鼠胰臟灌流實驗中，我們觀察到PI 3 kinase 訊息傳遞路徑並沒有參與葡萄糖刺激胰島素分泌的生理現象，但對 RSG的促進胰島素分泌扮演重要角色 (Yang et al., 2001)。而在胰島細胞的電生理研究中，發現高濃度葡萄糖 (16.7 mM)關閉KATP channel的作用並不受PI3 kinase抑制劑的影響。所以由此可推論RSG與葡萄糖刺激胰島素分泌可能是經由不同的訊息傳遞路徑。 在本研究中我們發現AICAR可快速抑制KATP channel的電流，雖然介於AMPK與KATP channel的媒介並不清楚，蛋白質激酶催化的磷酸化可能是個重要的機制來調控離子通道的活性 。 Light等人曾報告protein kinase C可透過磷酸化Kir6.2 subunit上第180個胺基酸threonine (Thr180) 來提高KATP channel的活性。 Protein kinase A 也曾被報告可磷酸化Kir6.2 subunit上第372個胺基酸serine (Ser372)與 SUR1上第1572 個胺基酸(Ser1572)。另一方面， AMPK被發現可與cystic fibrosis transmembrane conductance regulator (CFTR)交互作用並降低其活性而抑制CFTR channel gating。此外，最近有報告指出心室心肌細胞過度表現constitutively active AMPK mutant可調控voltage-gated sodium channels。所以，我們可以合理地推測AMPK或許也可調節其他組織的不同離子通道。但是AICAR抑制KATP channel是否透過磷酸化Kir6.2或SUR1 subunit上的serine 或threonine仍需進一步的實驗來證明。 總結來說，本研究證明 rosiglitazone可經由PI3 kinase 活化AMPK，進而抑制 KATP channel 的鉀離子電流而促進胰島素分泌。由於 rosiglitazone 的作用很快速，因此 rosiglitazone促進胰島素分泌的作用可能是透過非基因且與PPARγ無關的機轉所造成。
並列摘要
Diabetes is a big challenge of global public health By the end of the 20th century the worldwide diabetes pandemic had affected an estimated 151 million persons, distributed among both developed and developing countries. At the societal level, the burden of diabetes worldwide is great and continues to grow. Overall, 90–95% of diabetes is type 2, and this form of the disease (historically considered rare in youth) is now increasingly affecting children and adolescents. While it is estimated that 30–50% of diabetes cases remain undiagnosed, there were approximately 30 million people worldwide diagnosed with diabetes in 1985. By 1995 this number had increased to 135 million (4.0% of the world population), and projections indicate there may be 300 million people with diabetes (5.4% of the world population) by 2025. Between 1995 and 2025, the number of people with diabetes will increase 42% (from 51 to 72 million) in industrialized countries, but will increase 170% (from 84 to 228 million) in industrializing countries. Diabetes is a lifelong condition that seriously affects a person’s quality of life. Individuals with the disease have to make major lifestyle changes and learn to live with monitoring blood glucose, using multiple drugs and injections, and dealing with complications of the disease and their treatment. Diabetes is a complex and multifactorial disease that is associated with considerable mortality, morbidity, and loss of quality of life. Diabetes is a leading cause of death, new cases of end-stage renal disease (ESRD), lower-limb amputations, blindness, and cardiovascular disease (CVD). Around 25% of people with diabetes in the United States have some visual impairment. Of those with disease for 15 years, 10% develop severe visual impairment and 2% become blind. Diabetes accounts for 35% of all new ESRD cases in the United States. The most common cause of death among people with diabetes is CVD, and approximately 50% of people with diabetes in the United States die from coronary heart disease. About 10% of diabetic individuals in the United States report having had a stroke and 20–30% have some peripheral vascular disease (PVD). Furthermore, people with diabetes have extraordinarily high levels of CVD risk factors, including hypertension, dyslipidemia, and obesity. Between 30 and 60% of Americans with diabetes have neuropathy, 50% have considerable physical disability, and diabetes is also associated with higher risk of dental disease and complications of pregnancy. The burden of diabetes complications is similarly large worldwide. Based on data from several countries, the World Health Organization (WHO) estimates that in year 2000, between 30 and 45% of people with diabetes worldwide have retinopathy, 10–20% have nephropathy, 20–35% has neuropathy and 10–25% has CVD. At the national level, diabetes exerts a substantial toll on the direct health care costs in all countries. A person with diabetes costs the health care sector 2.5 times more than a person without the disease. The estimated total annual direct health care cost of diabetes in 1998 in industrialized countries varied from US$ 0.54 billion in Denmark to US$ 60 billion in the United States. One study, estimated the annual direct cost of diabetes in India at US$ 2.2 billion. Diabetes is an inheritable disease Type 1 diabetes is an autoimmune disease that leads to the destruction of pancreatic β cells and insulin deficiency. It is common in childhood and adolescence but can occur at any age. Although most cases lack a family history, first-degree relatives have a higher risk of developing type 1 diabetes than does the general population. Within families, susceptibility depends on the degree of genetic identity with the proband. The highest risk is observed in identical twins. The disease concordance rate in twins can be up to 70% in studies with the longest follow-up period. Although siblings have, on average, a lower prevalence of approximately 6%, this rate is still higher than the 0.4% observed in the white population in the United States, confirming a significant familiar clustering. Empirically, type 2 diabetes mellitus has been considered a partially inheritable disease, as suggested either by the studies of life-time risk of around 40% for the development of the disease in non-diabetic offspring, siblings and dizygotic twins and increasing to 70% if both parents have type 2 diabetes mellitus or by the studies of the concordance of type 2 diabetes mellitus in dizygotic and monozygotic twins. The recurrence risk (λs) of type 2 diabetes for monozygotic twins has been estimated as high as 10 and around 3 in first-degree relatives. The best evidence that heredity plays an important role comes from the following observations: (i)Concordance rates for type 2 diabetes and its predecessor, impaired glucose tolerance, are consistently higher in monozygotic than in dizygotic twin pairs; (ii) sibling recurrence rates are consistently higher than population prevalence rates, although the reported excess is modest; (iii) groups of patients labeled as having type 2 diabetes include individuals suffering from unrecognized monogenic and digenic disorders; and (iv) certain common single-nucleotide polymorphisms (SNPs) appear to influence diabetes risk. Accelerator hypothesis : Type 1 and type 2 diabetes as the same disorder of insulin resistance, set against different genetic backgrounds The prevalence of diabetes is increasing rapidly in industrialized countries. Although most attention has focused on the increase in type 2 diabetes, there has been a parallel increase in type 1 diabetes, which requires explanation. Type 2 diabetes is believed to result from the loss of β cell function in association with insulin resistance. The “Accelerator Hypothesis” regards type 1 diabetes in the same way. Awareness of overlap between type 1 and type 2 diabetes is not new. There has long been interest in insulin resistance in type 1 diabetes, although related more to its implications for management and outcome than to its pathogenesis. The term “type one-and-a-half” diabetes, referring to the progression in some from type 2 to type 1 diabetes, was coined years ago and remains an area of lively debate. In a modern context, the increasing difficulty in distinguishing type 1 from type 2 diabetes in obese young people has given rise to the designation “double diabetes,” in which recognition is given to the coexistence of autoimmunity and insulin resistance. Insulin resistance upregulates the β cells metabolically and accelerates their loss through glucotoxicity. The tempo is normally slow. The “Accelerator Hypothesis” argues that people in whom type 1 diabetes develops are subject to the same weight increase, the same insulin resistance, the same metabolic upregulation, and the same acceleration in β cell loss as those with type 2 diabetes. They are, in addition, genetically susceptible to mounting an aggressive immune response to metabolically upregulated β cells. Depending on the genotype, this further accelerator can greatly increase the tempo of β cell loss. Those with type 1 diabetes nevertheless remain a subset of type 2 diabetes, sharing the same basic accelerator: insulin resistance. Indeed, the “Accelerator Hypothesis” predicts that if people inwhom type 1 diabetes would develop lacked the immunogenetic accelerator, they would still be at risk for type 2 diabetes at a later time. It is already known that people in whom type 1 diabetes develops are heavier in early childhood than nondiabetic people and tend to be taller. Moreover, the prevalence and titer of GAD antibodies are also related to BMI both in first-degree relatives of type 1 diabetic subjects and in the normal population. The “Accelerator Hypothesis” goes further and predicts that, among those who develop type 1 diabetes, the heavier children will do so at a younger age, in the same way that greater body mass accelerates the onset of type 2 diabetes. It goes on to suggest a mechanism whereby insulin resistance could interact with the type 1 diabetes susceptibility genotype to further accelerate β cell loss. Because only a defined subgroup of the population is genetically susceptible, the “Accelerator Hypothesis” predicts that increasing obesity in children would cause the age at presentation to decrease without necessarily changing lifetime risk. Recent epidemiological data suggest that this may be the case. The associated genes of type 1 diabetes Association studies and linkage analysis have been used to identify susceptibility loci. So far, three loci are well characterized, and much progress has been made in elucidating the mechanisms by which alleles at these loci modulate type 1 diabetes susceptibility. These loci are IDDM1, corresponding to the human leukocyte antigen (HLA) genes HLA-DR and HLA-DQ; IDDM2, corresponding to the insulin gene INS; and IDDM12, corresponding to the CTLA4 gene. A description of these loci follows. (1) IDDM1: DR and DQ regions of HLA The major susceptibility locus lies within he HLA complex on chromosome 6 and provides up to 40% to 50% of the inheritable diabetes risk. The HLA-DR and HLA-DQ loci in the class II region have the strongest influence on type 1 diabetes risk. Most patients carry the HLA-DR3 or HLA-DR4 class II antigens, and approximately 30% of these are DR3/ DR4 heterozygotes. The DR3/ DR4 genotype confers the highest risk, with a synergistic mode of action, followed by DR4 and DR3 homozygosity, respectively. In the white population, the HLA-DQ heterodimers encoded by DQA1*0301, DQB1*0302, DQA1*0501, and DQB1*0201 have the strongest association with diabetes. However, the HLA-DQ locus also harbors type 1 diabetes protective alleles. Among the commonest DR2 haplotypes observed in white , the DQA1*0102, DQB1*0602, and DRB1*1501 haplotype is negatively associated with type 1 diabetes in several populations. (2) IDDM2: insulin gene The IDDM2 locus maps to a variable number of tandem repeats (VNTR) located approximately 0.5 kilobase (kb) upstream of the insulin gene (INS). Two main classes of VNTR alleles are defined based on the number of repeats: short alleles (class I) and long alleles (class III); intermediate size (class II) are rare in the white population. Homozygosity for class VNTR I alleles is found in 75% to 85% of the patients, compared with a frequency of 50% to 60% in the general population, and it increases the likelihood that identical tween will be concordant for type 1 diabetes. The relative risk ratio of the I/I genotype versus I/III or III/III has been reported to the moderate (in the 3 to 5 range), and it accounts for approximately 10% of the familial clustering of type 1 diabetes. (3) IDDM12: CTLA4 gene. The 2q33 chromosomal region contains a cluster of genes coding for proteins involved in T-cell costimulation, such as CTLA4 (cytotoxic T-lymphocyte-associated 4), CD28, and ICOS. Linkage with CTLA4 came from a multiethnic collection of families from Spain, France, China, Korea, and Mexican Americans. The transmission disequilibrium test (TDT) revealed deviation for alleles at the (AT)n microsatellite and the A→G polymorphism. Additional evidence suggesting that CTLA4 is the etiologic variant at the IDDM12 locus came from another study of a multiethnic collection nof 178 simplex and 350 multiplex families. The TDT revealed association/linkage with threee markers within CTLA4 and two flanking markers on each side of CTLA4 but not with more distant markers near CD28. The associated genes of type 2 diabetes Genomewide linkage mapping for type 2 diabetes or related quantitative metabolic traits have provided a number of different chromosomal loci, among which the chromosomes 1q, 2q, 8p, 10q, 12q and 20q seem to be the most consistent, but so far only the NIDDM1 locus on chromosome 2q37, with the identification of calpain-10 (CAPN10) as a diabetes susceptibility gene, has been a successful example of positional cloning of a common gene for type 2 diabetes. It appears that the most consistent findings in nonmendelian type 2 diabetes are the two common coding variants: Pro12Ala of the PPAR-g gene and Glu23Lys of Kir6.2, both of which may confer a modest relative increased risk of diabetes (odds ratio about 1.2). However, due to the high frequency of the two risk alleles, they contribute with a large population attributable to diabetes risk. The latter gene variant was recently by a second-look shown to be associated with impaired glucose-induced insulin release during an oral glucose tolerance test, impaired glucose-induced suppression of glucagon secretion and to associate with type 2 diabetes. Oligogenic (digenic) inheritance of severe insulin resistance and type 2 diabetes has been reported in a human pedigree with double mutations in the PPP1R3 and PPAR-g genes. Interestingly, several other gene variants have inconsistently been shown to associate with type 2 diabetes including a non-coding intronic variant of the SUR1 gene, IVS15-3t/c, Trp64Arg of the b3-adrenergic receptor, Gly972Arg of the IRS-1, PP1ARE of the PPP1R3 and Gly482Ser of the PGC-1. The potential role of these susceptibility genes deserves further cautious investigations applying SNP genotyping and haplotyping in sufficiently powered and well-characterized study materials. β cell function declined gradually with time Patients with Type 2 diabetes are characterized by peripheral insulin resistance, aberrant pancreatic insulin secretion and enhanced hepatic glucose output. Pancreatic beta cell function has been shown to decline with increasing duration of diabetes, regardless the use of different existing therapies including sulphonylureas, biguanides and insulin as demonstrated in the prospective study of UKPDS. A longitudinal study in Pima Indian demonstrated that transition from NGT to IGT was associated with an increase in body weight, a decline in insulin-stimulated glucose disposal, and a decline in the acute insulin secretory response (AIR) to intravenous glucose, but no change in EGO. Progression from IGT to diabetes was accompanied by a further increase in body weight, further decreases in insulin-stimulated glucose disposal and AIR, and an increase in basal EGO. Thirty-one subjects who retained NGT over a similar period also gained weight, but their AIR increased with decreasing insulin-stimulated glucose disposal. Thus, defects in insulin secretion and insulin action occur early in the pathogenesis of diabetes. Intervention to prevent diabetes should target both abnormalities. The specific aims of this study Both type 1 and type 2 diabetes were considered as polygenic disease，and it has been reported that certain genetic polymorphism will be associated with incidence rate, age of onset or the decline of β cell function。Recently, “Accelerator Hypothesis” was postulated, and it was considered that both type 1 and type 2 diabetes was the same, which have insulin resistance associated hyper secretion of insulin, and then accelerate the β cell loss. Therefore, there maybe the same genetic susceptibility to warrant between type 1 and type 2 diabetes. Besides, the defects in β cell secretory function and insulin resistance existed early in the prediabetic stage, and they were aggravated with time. Therefore, it is very important to investigate how to prevent and delay the loss of β cell function. Therefore the thesis will be divided into three parts: (1) The association study between vitamin D receptor (VDR) gene polymorphism and susceptibility to type 1 diabetes in Taiwanese population (2) The association study between β2-adrenoreceptor (ADRB2) gene polymorphism and susceptibility to type 2 diabetes (3) The molecular mechanism of thiazolidinediones (TZDs) to improve the insulin secretory capacity of islet β cells. First part: The association study between vitamin D receptor (VDR) gene polymorphism and susceptibility to type 1 diabetes in Taiwanese population OBJECTIVE: Vitamin D and its receptor have been suggested to play a role in the pathogenesis of type 1 diabetes mellitus. We have therefore studied the influence of vitamin D receptor (VDR) gene polymorphisms on susceptibility to type 1 diabetes, and rates of glutamic acid decarboxylase (GAD65) autoantibody and islet cell autoantibody (ICA512) positivity. SUBJECTS AND MEASUREMENTS: One hundred and fiffty-seven type 1 diabetic patients and 248 unrelated normal controls were recruited for this study. Genomic DNA was extracted from peripheral blood leucocytes. All type 1 diabetic patients and controls were genotyped using polymerase chain reaction (PCR) and restriction fragment length polymorphism (RFLP), for three restriction sites in the VDR gene, BsmI, ApaI and TaqI. The x2 test was used to compare the frequency of the VDR gene polymorphisms in patients and normal controls. The association of VDR gene polymorphisms in type 1 diabetes with the presence of GAD65 and ICA512 autoantibodies were also examined using the χ2 test. RESULTS: The allele frequency of the BsmI and ApaI polymorphisms, but not TaqI polymorphism, differed between patients and controls (BsmI P=0.015; ApaI P= 0.018; TaqI P= 0.266). However, after correction for the three different polymorphisms tested, only the BsmI was significant (pc=0.045). CONCLUSIONS: Vitamin D receptor gene polymorphisms were associated with type 1 diabetes in a Taiwanese population. However, functional studies are needed to establish the role of the vitamin D receptor in the pathogenesis of type 1 diabetes mellitus. Second Part: The association study between β2-adrenoreceptor (ADRB2) gene polymorphism and susceptibility to type 2 diabetes OBJECTIVE: The significance of the association of amino terminal polymorphisms in β2-adrenoreceptor (ADRB2) with obesity and type 2 diabetes is controversial and differs among ethnic groups. In this study, the association of ADRB2 with risk and age of onset of type 2 diabetes has been examined in a Taiwanese population. DESIGN: The study design is a case–control study to investigate the impact of the two amino acid polymorphisms in ADRB2 . PATIENTS AND MEASUREMENTS: This study includes 130 patients with type 2 diabetes [female : male = 1 : 1, age: 52.4 ± 10.0 years; body mass index (BMI): 24.2 ± 2.9 kg/m2 ; mean ± SD] and 130 controlled subjects matched for gender, age and BMI with normal glucose tolerance (female : male = 1 : 1, age: 51.7 ± 10.6 years; BMI: 23.9 ± 2.7 kg/m2). The Arg16Gly and Gln27Glu polymorphisms of ADRB2 were determined by polymerase chain reaction-restriction fragment length polymorphism (PCR–RFLP) assays. The genotypic and allelic frequencies between two groups were compared and the relationship between the genotypes and clinical phenotypes was examined. RESULTS: A difference in genotypic frequency in the Arg16Gly polymorphism was noted between groups in this gender-, age- and BMI-matched case–control study (P= 0.039). Multivariate regression analysis revealed that the Arg16Gly polymorphism was the only independent factor for development of type 2 diabetes(P= 0.021). In addition, we utilized the log-rank test to compare the differences in age of onset between wildtype and nonwild-type polymorphisms. The Arg16Gly polymorphism was independently associated with age of onset in type 2 diabetes (P= 0.017). There was no difference in the Gln27Glu polymorphism between diabetic and control groups in this study. CONCLUSIONS: In a Taiwanese population, homozygosity of Arg16 in the ADRB2 gene was associated with a higher frequency (odds ratio 1.87, 95% confidence interval 1.34–2.40) for development of type 2 diabetes. Moreover, this polymorphism was also associated with an earlier onset of type 2 diabetes. However, the Glu27Gln polymorphism had no impact on either BMI or type 2 diabetes in a Taiwanese population. Third Part: The molecular mechanism of thiazolidinediones (TZDs) to improve the insulin secretory capacity of islet β cells To elucidate the direct effect of rosiglitazone (RSG), a new thiazolidinedione antihyperglycemic agent, on pancreatic insulin secretion, an in situ investigation by rat pancreatic perfusion was performed. At a basal glucose concentration of 6 mmol/l, RSG (0.045–4.5 μmol/l) stimulated insulin release in a dose-dependent manner. In addition, 4.5 μmol/l RSG potentiated the glucose (10 mmol/l)-induced insulin secretion. Both the first and second phases of glucose-induced insulin secretion were significantly enhanced by RSG, by 80.7 and 52.4%, respectively. The effects of RSG on insulin secretion were inhibited by a phosphatidylinositol 3-kinase (PI3K) inhibitor, LY294002. In contrast, the glucose-stimulated insulin secretion was not affected by LY294002. The potentiation effect of RSG on glucose-stimulated insulin secretion, in both the first and second phases, was significantly blocked by LY294002. These results suggest that RSG has a direct potentiation effect on insulin secretion in the presence of 10 mmol/l glucose, mediated through PI3K activity. The inability of LY294002 to inhibit glucose-induced insulin secretion suggests that different pathways are responsible for glucose and RSG signaling. To further elucidate the molecular mechanism of RSG on potentiating the glucose-stimulated insulin secretion, we found that rosiglitazone could activate AMP-activated protein kinase (AMPK), downstream of PI3K activation in pancreatic beta cells. Using a pharmacological activator of AMPK, 5-aminoimidazole-4-carboxamide ribonucleoside (AICAR), we showed that AICAR inhibited ATP-sensitive potassium channel conductance by whole cell voltage clamp techniques and potentiate glucose-stimulated insulin secretion in the primary rat islets. To confirm the role of this signaling pathway in insulin secretion in vivo, we also found that AICAR could stimulate insulin secretion in the presence of basal and high glucose concentration in isolated pancreatic perfusion studies. Taken together, we conclude that PI3K-dependent activation of the AMPK is required for the insulin secretory response induced by rosiglitazone in pancreatic beta cells. AMPK may serve as a potential therapeutic target of insulin secretion for the treatment of diabetes mellitus.