| 摘要: | 背景：
全基因組關聯研究（GWAS）是一個強大的工具被用來識別與人類疾病相關的基因變異。GWAS主要針對單核苷酸多態性（SNPs）和臨床特徵，或疾病之間的關聯性分析。最近的GWASs已找到數百個與人類的複雜性疾病和性狀相關的基因變異。然而，複雜的遺傳特性可以採取多種形式，而GWASs只能部分解釋遺傳所佔之比率，因此往往導致遺傳失落(missing heritability)。中介分析是用於研究兩個變量之間之關係機制的一個重要的統計工具。它已被普遍使用在社會科學的研究。我們針對兩個常見的基因位點，分別為ABO和CDH13基因，想利用中介分析，以評估基因變異及相關表現型之間的關聯性是否會受到另一個表現型之抑制效應。
方法和結果：
樣本族群包括617個台灣人。針對五個ABO基因上的多型性和四個CDH13基因變異，使用限制性酶片段長度多型性(RFLP)以及TaqMan SNP進行基因檢測分型。之後進行中介分析。首先，結果顯示ABO血型之基因型均與可溶性內皮細胞選擇素(sE-selectin)之血中濃度有相關性（P = 3.5×10-36）。當把可溶性內皮細胞選擇素排除後分析，結果顯示A血型的人比非A血型的人具有顯著較低之高密度脂蛋白膽固醇（HDL-C）濃度，較高之三酸甘油酯（TG）濃度，以及較高之三酸甘油酯/高密度脂蛋白膽固醇（TG / HDL-C）比率（P = 0.009，P = 0.004和P = 0.001）。中介分析進一步揭示了內皮細胞選擇素在ABO血型基因型與TG / HDL-C比值之間的關聯性呈現抑制效應（索貝爾測試; P = 1.18×10-6）。其次，CDH13基因型和單倍型與脂聯素(adiponectin)之血中濃度有相關性（rs4783244之最低P = 1.95×10-11，單倍型ATTT之最低P = 3.78×10-13）。當把脂聯素排除後分析，結果顯示具有rs12051272之次要等位基因的人都有一個較良性之代謝綜合症。這些個體具有較高的胰島素敏感性、HDL-C濃度，以及較低之舒張壓、空腹血糖，TG濃度，並且代謝綜合症的風險亦較低（P <0.05）。中介分析也顯示了脂聯素在CDH13基因型和代謝綜合症及其相關表型之間的關聯性具有抑制效應（索貝爾測試; P <0.001）。
結論：
ABO和CDH13的遺傳基因變異分別具有獨立影響內皮細胞選擇素和脂聯素之血中濃度。同時血液中的內皮細胞選擇素和脂聯素表現出對遺傳基因變異和各種表型之間的關聯性有抑制效應。這些結果提供了進一步的證據:中介分析的抑制效應可能部分克服GWAS在複雜疾病研究所發生的遺傳失落問題
Background:
A genome-wide association study (GWAS) is a powerful tool to identify genetic variants associated with human diseases. A GWAS typically focuses on associations between single nucleotide polymorphisms (SNPs) and traits, such as diseases. Recent GWASs have identified hundreds of genetic variants associated with complex human diseases and traits. However, complex inheritance can assume numerous forms and GWASs can only partially explain heritability, a problem often resulting in missing heritability. Mediation analysis is an important statistical tool used to investigate the mechanism underlying relationships between two variables. It had been commonly used in social science research. We estimated the missing heritability of two genetic associations of two common gene loci, ABO and CDH13, using mediation analysis to assess the suppression effects of various phenotypes on the statistical association between genetic variants and related phenotypes.
Methods and Results:
The sample population included 617 enrolled Taiwanese subjects. Five ABO gene region polymorphisms and four CDH13 gene variants were selected and genotyped using a TaqMan SNP assay. Mediation analysis was further conducted in two genetic association studies. Firstly, the genetic-inferred ABO blood group genotypes were associated with sE-selectin levels (P = 3.5 × 10-36). After adjusting for sE-selectin levels, significantly lower high-density lipoprotein cholesterol (HDL-C) levels, higher triglyceride (TG) levels, and higher (TG/HDL-C) ratios were noted in individuals with blood group A than in non-A individuals (P = 0.009, P = 0.004, and P = 0.001, respectively). Mediation analysis further revealed a suppression effect (Sobel test; P = 1.18 × 10-6) of sE-selectin level on the association between genetic-inferred ABO blood group genotypes and the TG/HDL-C ratio. Secondly, CDH13 genotypes and haplotypes were associated with adiponectin levels (lowest P = 1.95 × 10−11 for rs4783244 and lowest P = 3.78 × 10−13 for haplotype ATTT). After adjusting for adiponectin levels, participants with the minor allele of rs12051272 were considerably associated with a more favorable metabolic profile. These individuals had higher insulin sensitivity, higher HDL-C levels, lower diastolic blood pressure, lower circulating levels of fasting plasma glucose levels, lower TG levels, and a lower risk of metabolic syndrome (all P < 0.05). The mediation analysis also revealed a suppression effect of adiponectin levels on the associations between CDH13 genotypes and metabolic syndrome and its related phenotypes (Sobel test; all P < 0.001).
Conclusion:
Genetic variants of ABO and CDH13 genes independently affect sE-selectin and adiponectin levels, respectively. Circulating levels of both sE-selectin and adiponectin exhibit a suppressive effect on the association between genetic variants and various phenotypes. These results provide further evidence for the suppression effects of mediation analysis that may partially overcome the missing heritability problem that occurs when complex diseases are studied using GWAS. |
