代謝功能之生理指標與相關基因型:非臨床青少年之雙胞胎/手足研究

Abstract

本論文為一雙胞胎/手足研究，以下列三個子研究探討代謝功能之生理指標的遺傳度，及其與TCF7L2基因多型性的相關性。 研究一: 遺傳或環境因素對於青少年之脂締素、瘦素與身體質量指數的影響: 雙胞胎/手足多變項分析 瘦素與脂締素和肥胖有高度相關，然而，有無共同的遺傳或環境因素影響青少年之脂締素、瘦素、與身體質量指數仍屬未知。本研究目的在於評估遺傳或環境因素對青少年之脂締素、瘦素與身體質量指數的個別及共同的影響。本研究收集了228對，年齡為12到18歲，國高中之雙胞胎或手足配對（包含130對同卵雙胞胎，68對異卵雙胞胎，和30對手足配對）。採用結構方程式模型建構之多變項分析進行雙胞胎資料分析。結果顯示脂締素、瘦素、與身體質量指數在同卵雙胞胎之間的相關係數都高於異卵雙胞胎或手足配對的相關係數。脂締素受遺傳與共享環境的影響相當（各約40%）。相反的，瘦素與身體質量指數主要受遺傳因子的影響（分別是74%和89%）。多變項分析的結果顯示脂締素、瘦素與身體質量指數受到共同的潛在因子影響，身體質量指數相當於此共同的潛在因子。只有15%脂締素的變異、以及34%瘦素的變異，受到影響身體質量指數的共同遺傳因素所影響。本研究指出脂締素、瘦素與身體質量指數都受到遺傳因素影響，而且，脂締素、瘦素僅一小部份的變異，受到來自於影響身體質量指數之共同遺傳因子所影響。本研究為肥胖之脂締素與瘦素的複雜調控提供更多的理解。 研究二: 青少年雙胞胎/手足之胰島素敏感性遺傳度: 比較以口服葡萄糖耐量試驗為基礎的指標與以禁食狀態為基礎的指標 在許多代謝異常或代謝症候群中，胰島素抗性扮演重要的角色。本研究的目的在評估青少年雙胞胎或手足，其各種來自於禁食狀態或口服葡萄糖耐量試驗的胰島素敏感性指標，是否受基因與環境的影響。本研究收集了228對年齡為12到18歲，國高中之雙胞胎或手足配對，進行兩小時口服葡萄糖耐量試驗。採用Mx軟體，進行單變項結構方程式模型建構雙胞胎資料分析。結果顯示大部份胰島素指標之廣義遺傳度為中到高 (27%到78%)，且沒有性別的差異。以禁食狀態為基礎，且受ADE模式所解釋的指標中，受線性遺傳的影響約為25%，而廣義遺傳度高達55%。以口服葡萄糖耐量試驗為基礎，且受ADE模式所解釋的指標中，有63%到75%的變異，受到遺傳因子影響; 若以ACE模式所解釋的指標中，則顯著受線性遺傳因子影響。其中，以SIisOGTT指標廣義遺傳度高達75%。本研究發現在青少年的各種胰島素敏感度指標中，以口服葡萄糖耐量試驗為基礎的指標，其廣義遺傳度的估計比以禁食狀態的指標來得高。在尋找受胰島素影響的複雜性狀之易感受基因的研究上，本研究提供較為有遺傳性的指標。 研究三: TCF7L2基因多型性與胰島素抗性及代謝表現型的相關 在歐洲族群及台灣人的研究中發現，TCF7L2基因變異與糖尿病及胰島素分泌不良有很強的相關，然而，此相關的機轉仍屬未知。本研究收集台灣525位青少年雙胞胎或手足配對的檢體，針對TCF7L2基因中18個常見的單核苷酸多態性（SNP）位點進行基因定型。每位研究對象都進行標準口服葡萄糖耐量試驗（OGTT），同時測量量性代謝指標。結果發現，在此18個SNP中，rs290487對偶基因C顯著的增加口服葡萄糖耐量試驗之60，90，和120分鐘葡萄糖濃度（p = 0.001, 0.014和0.021），以及增加60和90分鐘胰島素濃度（p = 0.017和0.019），此結果顯示TCF7L2基因變異可能會增加胰島素抗性。本研究並未發現rs290487和胰島素分泌之測量值有顯著相關，此胰島素分泌之測量值，包括穩態模式評估法（HOMA）的胰島β細胞指標，和胰島素生成指標。另外，位於第四內含子（intron）的單核苷酸多態性位點rs10749127也和代謝症候群的特徵有相關，包括高收縮血壓和舒張血壓（p = 0.024和2x10-4），三酸甘油酯（p =7x10-4），尿酸（p = 0.011），以及可能增加身體質量指數（p = 0.059）。總結來說，在青少年的研究顯示，華人TCF7L2基因變異與增加胰島素抗性有關，而不是與胰島β細胞受損有關。這些發現加強Wnt訊息傳導在胰島素抗性與代謝症候群的角色。

Three studies were conducted to investigate the heritability of adiponectin, leptin, BMI, and insulin sensitivity, and the association between variants of TCF7L2 gene and insulin sensitivity in adolescent twins. Study 1. Genetic and Environmental Influences on Adiponectin, Leptin, and BMI among Adolescents in Taiwan: A Multivariate Twin/Sibling Analysis Circulating levels of leptin and adiponectin are closely associated with obesity. However, it is not known whether there are common shared genes or environment exerting influences on the levels of leptin, adiponectin, and BMI. We aimed to assess the relative contribution of genes and environment to adiponectin, leptin, and BMI individually as well as simultaneously to the three measures. Our subjects included a total of 228 twin/sibling pairs aged 12 to 18 (130 monozygotic twins, 68 dizygotic twins and 30 sibling pairs) were recruited from the middle schools. Multivariate analyses were applied to twin/sibling data using structural equation modeling. The results showed that intraclass correlations for adiponectin, leptin and BMI were higher in the MZ twins than those in the DZ/SP group. The relative contribution of genes to adiponectin (39%) was comparable to those of shared environment (40%). In contrast, leptin and BMI were influenced mostly by genes (74% and 89%, respectively). The multivariate genetic analyses showed that a latent factor underlying the three measures was identified, with BMI being equivalent to this latent factor. The BMI-dependent genetic factor explains only 15% and 34% of variation of adiponectin and leptin, respectively. These data indicate a differential contribution of genetic factors for the variation of adiponectin, leptin and BMI. More importantly, only a small portion of the genetic influences on adiponectin and leptin was attributed to BMI. Our findings provided more insight into the complex regulation of adiponectin and leptin in obesity. Study 2. Heritability of Insulin Sensitivity in Adolescent Twin/Sibling: OGTT-Based versus Fasting-Based Indices Insulin resistance plays a crucial role in many metabolic abnormalities and metabolic syndrome. The aim of the present study was to investigate whether the variation of a wide variety of fasting- and OGTT-based insulin sensitivity indices was attributable to genetic or environmental factors in a sample of healthy adolescent twin/siblings. We recruited a total of 228 twin/sibling pairs aged 12 to 18 from the middle schools for a 2h OGTT. Univariate twin analyses using structural equation modeling implemented in Mx software package were employed to estimate genetic and environmental contributions on fasting- and OGTT-based insulin sensitivity indices. Our results showed that moderate to high broad heritability (27%-78%) was estimated for all insulin sensitivity indices except for GSI. In addition, no significant sex differences were found. For all fasting-based indices that were better explained by ADE model, the additive genetic influences accounted for around 25% of the total variance, whereas the estimates of broad heritability were around 55%. For OGTT-based insulin sensitivity indices that were better explained by ADE model, 63%-78% of the total variance was attributed to genetic influences. For those indices that were better explained by ACE model, significant additive genetic influences were observed. Among them, the broad heritability of SIisOGTT index was as high as 75%. These findings revealed that among a wide variety of insulin sensitivity indices in non-referred adolescents, the broad heritability estimates were greater for OGTT-based than for fasting-based indices. The findings may help in the search for genes underlying the variation in complex traits that are affected by insulin sensitivity. Study 3. Association of the TCF7L2 Genetic Polymorphisms with Insulin Resistance and Related Metabolic Phenotypes in Chinese Adolescents TFC7L2 genetic variants are strongly associated with type 2 diabetes and impaired insulin secretion in European and Taiwanese. However, the mechanism underlying the association remained unknown in the Chinese population. Therefore, we genotyped 18 common single nucleotide polymorphisms (SNPs) of TCF7L2 gene in 525 non-clinical adolescent twin/sibling in Taiwan. Oral glucose tolerance test (OGTT) was performed and quantitative metabolic parameters were measured for each participant. Among the 18 SNPs, rs290487 C allele were significantly associated with elevated 60, 90, and 120-min glucose concentrations (p = 0.001, 0.014 and 0.021) and 60 and 90-min insulin concentrations (p = 0.017 and 0.019) during OGTT, which suggested increased insulin resistance. We did not observe significant association of rs290487 with measures of insulin secretion including the homeostasis model assessment of beta cell and the insulinogenic index. Another SNP rs10749127 located in intron 4 was also significantly associated features of the metabolic syndrome including elevated systolic and diastolic pressures (p = 0.024 and 2x10-4), triglycerides (p = 7x10-4), uric acid levels (p = 0.011), and possibly higher body mass index (p = 0.059). In conclusion, the risk-conferring TCF7L2 genetic variant in the Chinese population is associated with increased insulin resistance but not impaired beta-cell function in healthy adolescents. These findings underscore the emerging evidence implicating the role of Wnt signaling in insulin resistance and metabolic syndrome.