Department of Epidemiology and Health Statistics, Xiangya School of Public Health, Central South University, Xiangya Road, Kaifu District, Changsha City, Hunan Province, 410078, China.
Hunan Provincial Key Laboratory of Clinical Epidemiology, Changsha City, Hunan Province, 410078, China.
Sci Rep. 2021 Oct 14;11(1):20455. doi: 10.1038/s41598-021-99836-2.
Genome-wide DNA methylation profiling have been used to find maternal CpG sites related to the occurrence of gestational diabetes mellitus (GDM). However, none of these differential sites found has been verified in a larger sample. Here, our aim was to evaluate whether first trimester changes in target CpG sites in the peripheral blood of pregnancy women predict subsequent development of GDM. This nested case-control study was based upon an early pregnancy follow-up cohort (ChiCTR1900020652). Target CpG sites were extracted from related published literature and bioinformatics analysis. The DNA methylation levels at 337 CpG sites of 80 GDM cases and 80 matched healthy controls during the early pregnancy (10-15 weeks) were assessed using MethylTarget sequencing. The best cut-off level for methylation of CpG site was determined using the generated ROC curve. The independent effect of CpG site methylation status on GDM was analyzed using conditional logistic regression. Methylation levels at 6 CpG sites were significantly higher in the GDM group than in controls, whereas those at another 6 CpG sites were significantly lower (FDR < 0.05). The area under the ROC curve at each methylation level of the significant CpG sites ranged between 0.593 and 0.650 for the occurrence of GDM. After adjusting for possible confounders, the hypermethylation status of CpG site 68167324 (OR = 3.168, 1.038-9.666) and 24837915 (OR = 5.232, 1.659-16.506) was identified as more strongly associated with GDM; meanwhile, the hypermethylation of CpG site 157130156 (OR = 0.361, 0.135-0.966) and 89438648 (OR = 0.206, 0.065-0.655) might indicate lower risk of GDM. The methylation status of target CpG sites in the peripheral blood of pregnant women during the first trimester may be associated with GDM pathogenesis, and has potential as a predictor of GDM.
全基因组 DNA 甲基化谱分析已被用于寻找与妊娠糖尿病 (GDM) 发生相关的母体 CpG 位点。然而,在更大的样本中,这些差异位点都没有得到验证。在这里,我们的目的是评估妊娠妇女外周血中靶 CpG 位点在妊娠早期的变化是否能预测随后发生 GDM。这项巢式病例对照研究基于一个早孕随访队列(ChiCTR1900020652)。从相关已发表的文献和生物信息学分析中提取靶 CpG 位点。使用 MethylTarget 测序评估了 80 例 GDM 病例和 80 例匹配健康对照在早孕(10-15 周)时 337 个 CpG 位点的 DNA 甲基化水平。使用生成的 ROC 曲线确定 CpG 位点甲基化的最佳截断值。使用条件逻辑回归分析 CpG 位点甲基化状态对 GDM 的独立影响。GDM 组中 6 个 CpG 位点的甲基化水平显著高于对照组,而另 6 个 CpG 位点的甲基化水平显著低于对照组(FDR<0.05)。显著 CpG 位点的每个甲基化水平的 ROC 曲线下面积在 0.593 到 0.650 之间,用于 GDM 的发生。在调整可能的混杂因素后,CpG 位点 68167324(OR=3.168,1.038-9.666)和 24837915(OR=5.232,1.659-16.506)的超甲基化状态与 GDM 更密切相关;同时,CpG 位点 157130156(OR=0.361,0.135-0.966)和 89438648(OR=0.206,0.065-0.655)的超甲基化可能表明 GDM 的风险较低。妊娠妇女在妊娠早期外周血中靶 CpG 位点的甲基化状态可能与 GDM 的发病机制有关,并可能作为 GDM 的预测指标。