Title: Longitudinal DNA Methylation Profiles in Saliva of Offspring from Mothers with Gestational Diabetes: Associations with Early Childhood Growth Patterns
题目：妊娠糖尿病母亲后代唾液中的纵向 DNA 甲基化谱：与儿童早期生长模式的关联.

Authors: Teresa M Linares-Pineda^1,2,3, Alfonso Lendínez-Jurado^3,4,5,6, Alberto Piserra López⁷, María Suárez-Arana⁸, María Pozo³, María Molina-Vega¹, María José Picón-César^1,2,3, Sonsoles Morcillo^1,2,3
作者： Teresa M Linares-Pineda^1,2,3， Alfonso Lendínez-Jurado^3,4,5,6， Alberto Piserra López⁷， María Suárez-Arana ⁸， María Pozo³， María Molina-Vega¹， María José Picón-César^1,2,3， Sonsoles Morcillo^1,2,3

Affiliations:
背景：

¹ Department of Endocrinology and Nutrition, Virgen de la Victoria University Hospital, 29010 Málaga, Spain.
¹ Virgen de la Victoria 大学医院内分泌学和营养学系，29010 Málaga，西班牙。

² CIBER Pathophysiology of Obesity and Nutrition-CIBERON, Instituto de Salud Carlos III, 28029 Madrid, Spain
2 CIBER 肥胖和营养病理生理学-CIBERON，卡洛斯三世健康研究所，28029 西班牙马德里.

³ Biomedical Research Institute-IBIMA Plataforma BIONAND, 29010 Málaga, Spain
3 生物医学研究所-IBIMA Plataforma BIONAND， 29010 马拉加，西班牙.

⁴ Andalucía Tech, Universidad de Málaga, Campus de Teatinos s/n, 29071 Málaga, Spain.
⁴ Andalucía Tech， Universidad de Málaga， Campus de Teatinos s/n， 29071 马拉加， 西班牙。

⁵ Department of Pediatric Endocrinology, Regional University Hospital of Málaga, 29011 Málaga, Spain.
⁵ 马拉加地区大学医院儿科内分泌科，西班牙马拉加 29011。

⁶ Distrito Sanitario Málaga-Guadalhorce, 29009 Málaga, Spain.
⁶ Distrito Sanitario Málaga-Guadalhorce， 29009 马拉加， 西班牙.

⁷ Department of Cardiology, Virgen de la Victoria University Hospital, 29010 Málaga, Spain
7 Virgen de la Victoria 大学医院心脏病科，29010 马拉加，西班牙.

⁸ Department of Obstetrics and Gynecology, Regional University Hospital of Málaga, 29011 Málaga, Spain
⁸ 马拉加地区大学医院妇产科，29011 马拉加，西班牙

Corresponding author: Sonsoles Morcillo (sonsoles75@gmail.com; sonsoles.morcillo@ibima.eu)
通讯作者： Sonsoles Morcillo （sonsoles75@gmail.comsonsoles.morcillo@ibima.eu）

ABSTRACT
抽象

Background: The prevalence of obesity and type 2 diabetes mellitus (T2DM) is rising globally, particularly among children exposed to adverse intrauterine environments, such as those associated with gestational diabetes mellitus (GDM). Epigenetic modifications, specifically DNA methylation, have emerged as mechanisms by which early environmental exposures can predispose offspring to metabolic diseases. This study aimed to investigate DNA methylation differences in children born to mothers with GDM compared to non-GDM mothers, using saliva samples, and to assess the association of these epigenetic patterns with early growth measurements.
背景： 肥胖和 2 型糖尿病 （T2DM） 的患病率在全球范围内不断上升，尤其是在暴露于不利宫内环境的儿童中，例如与妊娠糖尿病 （GDM） 相关的环境。表观遗传修饰，特别是 DNA 甲基化，已成为早期环境暴露使后代易患代谢疾病的机制。本研究旨在使用唾液样本调查 GDM 母亲与非 GDM 母亲所生孩子的 DNA 甲基化差异，并评估这些表观遗传模式与早期生长测量的关联。

Methods: This study analyzed saliva DNA methylation patterns in 30 children (15 born to GDM mothers and 15 to non-GDM mothers) from the EPIDG cohort. Samples were collected at two time points: 8–10 weeks postpartum and at one year of age. Epigenome-wide analysis of over 850,000 CpG sites was conducted using the Illumina Methylation EPIC Bead Chip. Differential methylation positions (DMPs) were identified with the limma package, using a significance threshold of p < 0.01 and delta β ≥ 5%. Correlation analysis examined associations between methylation and growth variables (weight, height, BMI and annual growth) using Spearman tests.
方法： 本研究分析了 EPIDG 队列中 30 名儿童 （15 名 GDM 母亲所生，15 名非 GDM 母亲所生） 的唾液 DNA 甲基化模式。在两个时间点收集样本：产后 8-10 周和一岁时。使用 Illumina 甲基化 EPIC 微珠芯片对超过 850,000 个 CpG 位点进行了表观基因组范围分析。使用 limma 包鉴定差异甲基化位置 （DMP），使用 p < 0.01 和 delta β ≥ 5% 的显着性阈值。相关性分析使用 Spearman 检验检查甲基化与生长变量 （体重、身高、BMI 和年生长） 之间的关联。

Results: We identified 6,968 DMPs at the postpartum stage and 5,132 after one year, with 50 sites remaining differentially methylated over time, 16 of which maintained consistent methylation directionality. Functional analysis linked several of these DMPs to genes involved in inflammation and metabolic processes, including CYTH3 and FARP2, both implicated in growth and metabolic pathways. Significant correlations were found between specific CpG sites and growth-related variables such as weight, head circumference, height, and BMI.
结果： 我们在产后鉴定了 6,968 个 DMP，一年后鉴定了 5,132 个 DMP，其中 50 个位点随着时间的推移保持差异甲基化，其中 16 个保持一致的甲基化方向性。 功能分析将其中一些 DMP 与参与炎症和代谢过程的基因联系起来，包括 CYTH3 和 FARP2，两者都与生长和代谢途径有关。发现特定 CpG 位点与体重、头围、身高和 BMI 等生长相关变量之间存在显著相关性。

Conclusions: This study’s longitudinal design reveals stable DNA methylation patterns in saliva samples that differentiate GDM-exposed children from controls across the first year of life, highlighting the feasibility of saliva as a minimally invasive biomarker source. The persistence of these epigenetic signatures underscores their potential as early indicators of metabolic risk, offering valuable insights into the long-term impact of maternal GDM on child health. Although the use of saliva offers a practical and non-invasive tool for pediatric epigenetic research, further studies are necessary to validate these findings in larger populations.
结论： 本研究的纵向设计揭示了唾液样本中稳定的 DNA 甲基化模式，这些模式在出生后第一年将暴露于 GDM 的儿童与对照组区分开来，突出了唾液作为微创生物标志物来源的可行性。这些表观遗传特征的持续存在强调了它们作为代谢风险早期指标的潜力，为孕产妇 GDM 对儿童健康的长期影响提供了有价值的见解。尽管唾液的使用为儿科表观遗传学研究提供了一种实用且无创的工具，但还需要进一步的研究才能在更大的人群中验证这些发现。

Keywords: gestational diabetes, offspring, epigenetic, dna methylation, saliva, child growth
关键词：妊娠糖尿病， 后代， 表观遗传学， DNA甲基化， 唾液， 儿童生长

INTRODUCTION
介绍

Obesity and diabetes are increasing globally, becoming significant public health concerns, particularly due to the rising prevalence among children. It is estimated that 29% of boys and 26% of girls are overweight or obese(1). This means that these children are more likely to develop diabetes and other metabolic diseases later in life. Moreover, the offspring of mothers with obesity and/or diabetes are known to have an elevated risk of developing diabetes in the future (2). This is closely linked to the Developmental Origins of Disease (DOHAD) hypothesis and the theory of the first thousand days which suggest that exposure to an adverse environment during the periconceptional and/or intrauterine period increases the lifelong risk of metabolic and related diseases in offspring. During the first 1000 days of life, the period from conception to the first two years of a person's life, there is an adaptive capacity to different stimuli and social environments that can be determinant for the individual's future(3)(4). This timeframe represents a phase of heightened sensitivity, wherein nutritional, metabolic, and hormonal exposures can significantly influence developmental processes and future disease risk.
肥胖和糖尿病在全球范围内呈上升趋势，成为重大的公共卫生问题，特别是由于儿童患病率的上升。据估计，29% 的男孩和 26% 的女孩超重或肥胖（1）。这意味着这些孩子在以后的生活中更有可能患上糖尿病和其他代谢疾病。此外，已知患有肥胖和/或糖尿病的母亲的后代将来患糖尿病的 ri sk 升高 （2）。这与发育性疾病起源 （DOHAD） 假说和前一千天理论密切相关，该理论表明，在围孕期和/或宫内期间暴露于不利环境中会增加后代终生患代谢和相关疾病的风险。在生命的前 1000 天，即从受孕到生命的前两年期间，对不同刺激和社会环境的适应能力可能决定个人的未来（3）（4）。这个时间范围代表了一个高度敏感的阶段，其中营养、代谢和激素暴露会显着影响发展过程和未来疾病风险。

In recent years, significant research has focused on how gestational diabetes mellitus (GDM) influences the development of metabolic conditions like type 2 diabetes (T2DM) and obesity, both in mothers and their children. GDM is known to be associated with a range of adverse outcomes, including a higher likelihood of childhood obesity, and metabolic disorders in offspring (5). Among the mechanisms by which prenatal environmental factors can exert long-term influences on gene expression and health outcomes, epigenetics, particularly DNA methylation, has gained substantial attention. GDM can alter the intrauterine environment, affecting fetal growth and metabolic programming, potentially predisposing the offspring to chronic conditions like obesity and T2DM (6). Studies on cord blood samples have shown that children born to mothers with GDM exhibit distinct epigenetic modifications compared to those born to non-GDM mothers (7)(8). These epigenetic changes affect the regulation of genes involved in metabolic pathways, immune responses, and growth processes. For instance, altered DNA methylation patterns have been observed in genes related to insulin signalling, adipogenesis, and inflammation, which may contribute to the increased risk of metabolic disorders and obesity observed in these children. Furthermore, these epigenetic differences may serve as potential biomarkers helping identify children at risk for future metabolic diseases. Recent studies have explored the use of saliva samples for studying epigenetic markers in children, as saliva collection is non-invasive and practical(9)(10). Some studies have linked DNA methylation markers in saliva to birth weight(11).The use of saliva samples offers a valuable, accessible tool for studying epigenetic biomarkers in the infant population offering promising applications for longitudinal follow-up studies.
近年来，重要的研究集中在妊娠糖尿病 （GDM） 如何影响母亲及其儿童的 2 型糖尿病 （T2DM） 和肥胖等代谢状况的发展。已知 GDM 与一系列不良结局相关，包括儿童肥胖的可能性更高，以及后代代谢紊乱 （5）。在研究产前环境因素对基因表达和健康结果产生长期影响的机制方面，表观遗传学，特别是 DNA 甲基化，已经得到了广泛的关注。GDM 可以改变宫内环境，影响胎儿生长和代谢编程，可能使后代易患肥胖和 T2DM 等慢性病 （6）。对脐带血样本的研究表明，与非 GDM 母亲所生的孩子相比，患有 GDM 的母亲所生的孩子表现出明显的表观遗传改变 （7）（8）。这些表观遗传变化会影响参与代谢途径、免疫反应和生长过程的基因的调节。例如，在与胰岛素信号传导、脂肪生成和炎症离子相关的基因中观察到 DNA 甲基化模式的改变，这可能导致在这些儿童中观察到代谢紊乱和肥胖的风险增加。此外，这些表观遗传差异可能作为潜在的生物标志物，帮助识别有未来代谢疾病风险的儿童。研究探索了使用唾液样本研究儿童的表观遗传标记，因为唾液采集是无创且实用的 （9）（10）。 一些研究将唾液中的 DNA 甲基化标志物与出生体重联系起来 （11）唾液样本的使用为研究婴儿人群的表观遗传生物标志物提供了一种有价值的、可访问的工具，为纵向随访研究提供了有前途的应用。

Furthermore, epigenetic modifications are not static and can be influenced by postnatal environmental factors such as diet, physical activity, and exposure to toxins. This dynamic nature of the epigenome suggests that early interventions could potentially mitigate the adverse effects of prenatal exposures.
此外，表观遗传修饰不是静态的，会受到出生后环境因素的影响，例如饮食、身体活动和毒素暴露。表观基因组的这种动态性质表明，早期干预可能会减轻产前暴露的不利影响。

This study aims to explore the epigenetic differences between children born to mothers with GDM and those born to non-GDM mothers, focusing on DNA methylation patterns and their potential associations with childhood anthropometric measurements.
本研究旨在探讨 GDM 母亲所生孩子和非 GDM 母亲所生孩子之间的表观遗传差异，重点关注 DNA 甲基化模式及其与儿童人体测量的潜在关联。

METHODS
美多DS

Subjects
科目

Samples were drawn from the EPIDG cohort, comprising women with and without GDM. Details of this cohort have been published previously (12).This analysis focused on a subset of children born to mothers in this cohort. Briefly, pregnant women, with a positive O’Sullivan test, attending the Diabetes and Pregnancy unit at Hospital Universitario Virgen de la Victoria were recruited for the study. GDM was diagnosed according to National Diabetes Data Group (NDDG) criteria. The threshold values for the 100 g oral glucose tolerance test (OGTT) were 105 mg/dL for fasting glucose and 190 mg/dL, 165 mg/dL, and 145 mg/dL at 60, 120, and 180 minutes, respectively. A GDM diagnosis was confirmed when glucose levels met or exceeded these values at 2 or more time points(13). Women with a negative 100 gr OGTT were classified as controls.
样本来自 EPIDG 队列，包括患有和不患有 GDM 的女性。该队列的详细信息之前已发布 （12）。该分析侧重于该队列中母亲所生的孩子子集。简而言之，该研究招募了奥沙利文试验呈阳性、在维多利亚圣女医院糖尿病和怀孕科就诊的孕妇。根据国家糖尿病数据组 （NDDG） 标准诊断 GDM。100 g 口服葡萄糖耐量试验 （OGTT） 的阈值分别为空腹血糖 105 mg/dL 和 60 、 120 和 180 分钟时的 190 mg/dL、165 mg/dL 和 145 mg/dL。当葡萄糖水平在 2 个或更多时间点达到或超过这些值时，确认 GDM 诊断 （13）。OGTT 阴性 100 克的女性被归类为对照组。

Our analysis included 15 children from mothers with GDM and 15 from non-GDM mothers (control group or non-GDM). Subjects were selected based on their order of recruitment and whether they had attended both the postpartum visit (8-10 weeks after birth) and the one year follow-up. Saliva samples were collected from the children during both visits. Anthropometric data including weight, height, BMI, head circumference and corresponding percentiles and Z-scores, were obtained from the records obtained during well-child care visits at one month, four months and one year of age at their primary care centers. These percentiles and Z-score were calculated based on the WHO growth standards and they are specific for sex and age group(14). Annual growth was calculated by the difference between the SD of weight at one year and at one month of age (15)
我们的分析包括 15 例 GDM 母亲和 15 例非 GDM 母亲 （对照组或非 GDM） 的孩子。根据受试者的招募顺序以及他们是否参加了产后访视（出生后 8-10 周）和一年的随访来选择受试者。在两次就诊期间，都从儿童身上采集了唾液样本。人体测量数据包括体重、身高、BMI、头围和相应的百分位数和 Z 分数，是从其初级保健中心一个月、四个月和一岁儿童的健康托儿所访问期间获得的记录中获得的。这些百分位数和 Z 分数是根据 WHO 生长标准计算的，它们特定于性别和年龄组 （14）。年生长量通过一岁和一个月龄体重 SD 之间的差值计算 （15）.

All mothers gave their consent to participate in the study. The EPIDG study was approved by the Institutional review board at the Hospital Universitario Virgen de la Victoria, Spain.
所有母亲都同意参加这项研究。EPIDG 研究已获得西班牙维多利亚圣母大学医院机构审查委员会的批准。

DNA extraction, and bisulfite conversion
DNA 提取和亚硫酸氢盐转化

Saliva samples were collected using swaps sample collection and stored at room temperature; DNA was extracted within 7 days of its collection (following the manufacturing protocol). Swap saliva DNA was isolated using Qiamp DNA Blood Mini kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions for DNA in swaps. Quality and concentration of DNA was measured using Qubit 3.0 Fluorometer with Qubit dsDNA HS Assay Kit Fluorometer (Thermo Fisher Scientific, Waltham, MA, USA). All the DNA extracted was bisulfited treated with EZ DNA Methylation kit (Zymo Research) for posterior DNA methylation analysis. After quality control, methylation analysis proceeded with 29 samples: 14 from children in the GDM group and 15 from the control group.
使用 swaps 样品收集收集唾液样品 并在室温下储存;在收集后 7 天内提取 DNA（遵循生产方案）。根据制造商关于交换中 DNA 的说明，使用 Qiamp DNA 血液小量试剂盒（Qiagen， Hilden， Germany）分离交换唾液 DNA。使用 Qubit 3.0 荧光计和 Qubit dsDNA HS 检测试剂盒荧光计（Thermo Fisher Scientific，Waltham，MA，USA）测量 DNA 的质量和浓度。所有提取的 DNA 均用 EZ DNA 甲基化试剂盒 （Zymo Research） 进行亚硫酸氢盐处理，用于后验 DNA 甲基化分析。 质量控制后，对 29 个样本进行甲基化分析： GDM 组 14 例，对照组 15 例。

Epigenome wide DNA methylation analysis
表观基因组宽 DNA 甲基化分析

The Epigenome- wide DNA methylation analysis (EWAS) was performed in a total of 29 children at both the postpartum (T1) and one year follow up (T2 visits). More than 850.000 CpGs sites were interrogated with the infinitum Methylation EPIC Bead Chip Kit (Illumina, San Diego, CA, USA) following the Infinium HD Assay Methylation protocol, and raw data (idat files) were obtained from iS (Illumina) software.
在 产后 （T1） 和一年随访 （T2 访视） 中，共对 29 名儿童进行了表观全基因组 DNA 甲基化分析 （EWAS）。按照 Infinium HD 检测甲基化方案，使用无限甲基化 EPIC 珠芯片试剂盒（Illumina，San Diego，CA，USA）询问超过 850.000 个 CpGs 位点，原始数据（idat 文件）从 iS （Illumina） 软件获得。

Raw data were processed using R software (version 4.0.0) and Bioconductor packages like minfi (16). Quality control included signal detection, bead count, and probe filtering, which excluded non-CpG probes, probes with SNPs, multiple alignment locations, and X/Y chromosome probes (17). Normalization was done using beta-mixture quantile Normalization (BMIQ) method (18). To correct the differences in methylation resulting from differences in cellular heterogeneity, the Houseman correction was used (19)
使用 R 软件（版本 4.0.0）和 minfi （16） 等 Bioconductor 软件包处理原始数据。质量控制包括信号检测、磁珠计数和探针过滤，其中排除了非 CpG 探针、带 SNP 的探针、多个比对位置和 X/Y 染色体探针 （17）。使用 β 混合物分位数归一化 （BMIQ） 方法 （18） 进行归一化。为了纠正由细胞异质性差异导致的甲基化差异，使用了 Houseman 校正 （19）.

Methylation data analysis
甲基化数据分析

Differentially methylated positions (DMPs) were obtained using limma package (20) in R studio (4.0.4). To determine methylation levels, β-values and M-values were calculated. β-values represent the estimate of methylation level based on the ratio of the intensity of the methylation probe to the overall intensity. These values are used for representing results. On the other hand, M-values are obtained by logarithmically transforming the β-values and are used for performing the differential methylation analysis (21). The significance threshold for determining DMPs was set a p-value less than 0.01, deltaβ≥5%(17). DMPs were obtained from both times, and the models were adjusted for infant sex, maternal treatment (diet or insulin) and maternal weight gain. Furthermore, to avoid possible false positives in identifying DMPs, array inflation was calculated using the RaMWAS package in the R study. This package provides the lambda (λ) value, with the following ranges considered: λ between 0.98-1.10: no inflation, λ between 1.10-1.19: slight inflation, λ between 1.20-1.40: moderate inflation, λ >1.4: high inflation. A Venn diagram (Venny 2.1 tool) was used to identify common DMPs across time points (https://bioinfogp.cnb.csic.es/tools/venny/).
在 R studio （4.0.4 ） 中使用 limma 包 （20） 获得差异甲基化位置 （DMP ）。为了确定甲基化水平，计算 β 值和 M 值。β值表示基于甲基化探针强度与总强度之比的甲基化水平估计值。这些值用于表示结果。另一方面，M 值是通过对 β 值进行对数变换获得的，并用于执行差异甲基化分析 （21）。确定 DMP 的显着性阈值设置为小于 0.01 的 p 值，deltaβ≥5% （17）。从两个时间获得 DMP，并根据婴儿性别、母体治疗（饮食或胰岛素）和母体体重增加调整模型此外，为避免在识别 DMP 时可能出现的假阳性，在 R 研究中使用 RaMWAS 包计算阵列充气。此包提供 lambda （λ） 值，并考虑以下范围：λ 在 0.98-1.10 之间：无膨胀， λ 在 1.10-1.19 之间：轻微膨胀，λ 在 1.20-1.40 之间：适度膨胀， λ >1.4：高膨胀。 维恩图（Venny 2.1 工具）用于识别跨时间点的常见 DMP （https://bioinfogp.cnb.csic.es/tools/venny/）。

Functional analysis
泛函分析

To explore potential mechanism that may be altered in the offspring from mothers with GDM compared to the offspring from mother non-GDM, those DMPs that were maintained during time were used to perform a functional analysis.
为了探索与非 GDM 母亲的后代相比，患有 GDM 的母亲的后代可能发生改变的潜在机制，使用那些在一段时间内维持的 DMP 进行功能分析。

A string network (https://string-db.org/) was performed with the genes of interest. Due to the lack of protein interaction, we settled a maximum number of interactors on 100, and an interaction score >0.7. With the resulting net, Cytoscape (https://cytoscape.org/) was used to perform the clusterization of the net using MCODE app. Finally, ENRICHR (22) was used to figured out Kyoto encyclopedia of genes and genome (KEGG) pathways and the Gene ontology of each cluster.
对目标基因进行字符串网络 （https://string-db.org/）。由于缺乏蛋白质相互作用，我们将最大数量的相互作用子确定为 100，相互作用评分 >0.7。有了得到的网，Cytoscape （https://cytoscape.org/） 被用来通过 MCODE 应用程序执行网的聚类。最后，使用 ENRICHR （22） 找出京都基因和基因组百科全书 （KEGG） 通路以及每个簇的基因本体论。

Statistical analysis
统计分析

Non-parametric tests (U-Mann Whitney) were used to compare quantitative variables between groups due to small sample size and variable distribution. Spearman correlation analysis explored associations between methylation levels and anthropometric variables. The correlation matrix was obtained with corrplot and GGally packages in RStudio, the threshold was P<0.05.
由于样本量小且分布可变，采用非参数检验 （U-Mann Whitney） 比较组间定量变量。Spearman 相关分析探讨了甲基化水平与人体测量变量之间的关联。相关矩阵是在 RStudio 中使用 corrplot 和 GGally 包获得的，阈值为 P<0.05。

RESULTS
结果

Study population
研究人群

The characteristics of these subjects are shown in Table 1. No significant differences were observed in the anthropometric measurements of the children born to mothers with GDM compared to those born to non-GDM mothers at any of the visits, except for BMI at one year of age, where a significant difference was found. Additionally, there were no significant differences in gender distribution between the two groups. In the GDM group, 35.7% of the children were boys, compared to 33.3% in the non-GDM group. Similarly, no significant differences were observed in delivery type: 64.3% of children in the GDM group and 60% in the control group were delivered via cesarean section.
这些主题的特征如表 1 所示。在任何一次就诊中，与非 GDM 母亲所生的孩子相比，GDM 母亲所生孩子的人体测量均未观察到显著差异，但一岁时的 BMI 除外，其中存在显着差异。此外，两组之间的性别分布没有显著差异。在 GDM 组中，35.7% 的儿童是男孩，而非 GDM 组为 33.3%。同样，在分娩类型上未观察到显著差异： GDM 组 64.3% 的儿童和对照组 60% 的儿童通过剖宫产分娩。

Table 1: Characteristics of the study subjects
表 1：研究对象的特征

Differentiated methylated analysis and correlation.
差异化甲基化分析和相关性。

A total of 6.968 DMPs and 5.132 DMPs were found at the postpartum and after one year of follow-up, between offspring from GDM and non-GDM, respectively. The CpGs most differentiated by chromosome in each time are represented in the Manhattan plot (Figure 1). The inflation in both times was measured, in the case of postpartum the λ was 1 meaning no inflation, and we found a moderate inflation at one year follow-up analysis (λ= 1.36) (Additional file 1).
在产后和随访一年后，GDM 和非 GDM 后代中共发现 6.968 个 DMP 和 5.132 个 DMP。每次染色体分化最多的 CpG 在曼哈顿图中表示（图 1）。测量了两个时间的通货膨胀，在产后的情况下，λ 为 1 表示没有通货膨胀，我们在一年的随访分析中发现了适度的通货膨胀 （λ= 1.36） （附加文件 1）。

Among all the DMPs, 50 remained significantly differentiated over time. Of these, 16 CpGs maintained the same direction of DNA methylation levels across time points. Figure 2 displays the log2 fold change (log2FC), with an increase indicating hypermethylation and a decrease representing hypomethylation over time. This suggests that the difference in methylation levels between the two groups becomes more pronounced as time progresses but with the same trend.
在所有 DMP 中，有 50 个随着时间的推移仍然存在显著差异。其中，16 个 CpG 在不同时间点保持相同的 DNA 甲基化水平方向。图 2 显示了 log2 倍数变化 （log2FC），增加表示高甲基化，减少表示低甲基化。这表明随着时间的推移，两组之间甲基化水平的差异变得更加明显，但趋势相同。

Functional analysis
泛函分析

Of the 16 persistent DMPs, 8 were annotated in the Illumina EPIC database. We used these 8 proteins to make a functional analysis with String and ENRICHR. STRING was configured to give a network with a maximum number of interactors of 100. The resulting net was uploaded in cytoscape to perform a clusterization with the application MCODE. Four clusters were identified, two of which included some of our proteins of interest. The first cluster contained DEFB104A and was associated with the Staphylococcus aureus infection pathway and the NOD-like receptor signaling pathway, both of which are involved in the immune response to infection. The second cluster included FARP2 and CYTH3. This cluster was linked to several pathways, with the most notable being the phospholipase D signaling pathway and the VEGF signaling pathway, both of which are related to inflammatory processes. Additionally, although less significant, this cluster was also associated with type 2 diabetes and regulation of lipolysis in adipocytes (Figure 3).
在 16 个持续性 DMP 中，有 8 个在 Illumina EPIC 数据库中进行了注释。我们使用这 8 种蛋白对 String 和 ENRICHR 进行了功能分析。STRING 配置为提供最大交互器数为 100 的网络。将所得网络上传到 cytoscape 中，以使用应用程序 MCODE 进行聚类。 鉴定出四个簇，其中两个包含我们感兴趣的一些蛋白质。第一个簇包含 DEFB104A，并与金黄色葡萄球菌感染途径和 NOD 样受体信号通路有关，这两者都参与对感染的免疫反应。第二个集群包括 FARP2 和 CYTH3。该簇与多种途径有关，其中最引人注目的是磷脂酶 D 信号通路和 VEGF 信号通路，这两者都与炎症过程有关。此外，虽然不太重要，但该集群也与 2 型糖尿病和脂肪细胞脂肪分解的调节有关（图 3）。

Correlation analysis
相关分析

We conducted a correlation analysis to explore the relationship between DNA methylation marks and anthropometric variables related to childhood growth. 15 out of 16 CpG sites showed significant correlations with growth percentiles for weight, height, head circumference, BMI, and annual growth (Additional file 2). Some CpG sites were correlated with multiple variables at different time points, with notable differences between children exposed to GDM and controls
我们进行了相关性分析，以探讨 DNA 甲基化标记与儿童成长相关的人体测量变量之间的关系。16 个 CpG 位点中有 15 个与体重、身高、头围、BMI 和年生长的生长百分位数呈显著相关性 （附加文件 2）。 一些 CpG 位点与不同时间点的多个变量相关，暴露于 GDM 的儿童与对照组之间存在显著差异.

In children of mothers with GDM, methylation levels at cg00124849 measured postpartum showed a strong positive correlation with head circumference at 1 month (r=0.762, p=0.004), 4 months (r=0.720, p=0.008), and one year (r=0.794, p=0.006), whereas in controls, cg00124849 was negatively correlated with annual growth (r=-0.580, p=0.04). The cg08080145 showed consistent methylation levels over time. Notably, its methylation levels assessed at the postpartum visit and at the one-year follow-up were positively associated with height percentiles in offspring of mothers with GDM (Additional file 2). This relationship was observed across multiple time points, suggesting a potential long-term epigenetic influence on growth patterns in children exposed to maternal GDM
在患有 GDM 的母亲的孩子中，产后测量的 cg00124849 甲基化水平与 1 个月 （r=0.762，p=0.004）、4 个月 （r=0.720，p=0.008） 和 1 年 （r=0.794，p=0.006） 的头围呈强正相关，而在对照组中，cg00124849 与年生长呈负相关 （r=-0.580，p=0.04）。cg08080145 随时间显示一致的甲基化水平。值得注意的是，在产后访视和一年随访时评估的其甲基化水平与 GDM 母亲后代的身高百分位数呈正相关（附加文件 2）。在多个时间点观察到这种关系，表明对暴露于母体 GDM 的儿童的生长模式存在潜在的长期表观遗传影响.

Other sites, such as cg09608131, cg20963866, and cg19795817, were significantly associated with weight, head circumference, and annual growth only in the control group. cg20935223 showed a positive correlation with BMI at one year across all subjects (r=0.455, p=0.025), while cg02349186 showed a negative correlation (r=-0.510, p=0.011) (Additional file 2).
其他位点，如 cg09608131、cg20963866 和 cg19795817，仅在对照组中与体重、头围和年生长显著相关。 cg20935223 显示所有受试者一年的 BMI 呈正相关 （r=0.455，p=0.025），而 cg02349186 显示负相关 （r=-0.510，p=0.011） （附加文件 2）。

DISCUSSION
讨论

In this study, we observed distinct epigenetic marks, specifically DNA methylation patterns, in saliva samples from children born to mothers with gestational diabetes mellitus (GDM) compared to those from non-GDM mothers. Furthermore, several of these epigenetic differences persisted over the first year of life and were associated with anthropometric variables linked to childhood growth. Additionally, some of these DNA methylation sites are annotated in genes involved in pathways related to inflammation and type 2 diabetes, highlighting their potential role in early metabolic programming.
在这项研究中，我们与非 GDM 母亲相比，在妊娠糖尿病 （GDM） 母亲所生孩子的唾液样本中观察到不同的表观遗传标记，特别是 DNA 甲基化模式。此外，其中一些表观遗传差异在出生后的第一年持续存在，并且与与儿童生长相关的人体测量变量有关。此外，其中一些 DNA 甲基化位点在参与炎症和 2 型糖尿病相关通路的基因中被注释，突出了它们在早期代谢编程中的潜在作用。

Previous studies have documented DNA methylation changes in cord blood and placenta of offspring born to GDM mothers, suggesting that these epigenetic modifications may reflect developmental programming of disease mechanisms and their potential role as biomarkers risk for metabolic diseases(23) (24)(8)(2)
以前的研究已经记录了 GDM 母亲所生后代脐带血和胎盘的 DNA 甲基化变化，这表明这些表观遗传修饰可能反映了疾病机制的发育编程及其作为代谢 疾病风险生物标志物的潜在作用 （23）（24）（8） （2）.

While some research has associated DNA methylation profiles in children’s saliva with birth weight, prenatal maternal stress, and potential childhood obesity predictors, studies specifically targeting the offspring of GDM mothers are limited, with most investigations performed on cord blood samples (25)(11)(26) (27). For instance, Franzago et al evaluated DNA methylation levels in MC4R and LPL genes, in children born to mothers with obesity and GDM in different tissues, including placenta, maternal blood, and buccal swab samples. However, their analysis did not yield significant findings in saliva(28). A recent review carried out by Saucedo et al (29) examined the role of DNA methylation as a potential biomarker for monitoring fetal growth during pregnancy in women with GDM (29). They found discrepant results in more than 15 studies that evaluated the role of DNA methylation in relation to birth weight. Most of these studies were conducted using cord blood, placental tissue, and maternal blood samples, with none in buccal (oral) samples. These discrepancies are explained by multiple factors, including sample size, the techniques and approaches used, the specific regions studied, and the inclusion of potential confounding variables.
虽然一些研究将儿童唾液中的 DNA 甲基化谱与出生体重、产前母体压力和潜在的儿童肥胖预测因子相关联，但专门针对 GDM 母亲后代的研究是有限的，大多数研究都是对脐带血样本进行的 （25）（11）（26）（27）。例如，Franzago 等人评估了不同组织中肥胖和 GDM 母亲所生儿童的 MC4R 和 LPL 基因中的 DNA 甲基化水平 ，包括胎盘、母体血液和口腔拭子样本。然而，他们的分析在唾液中没有产生显着发现 （28）。 Saucedo 等人 （29） 最近进行的一项综述研究了 DNA 甲基化作为 GDM 女性妊娠期胎儿生长的潜在生物标志物的作用 （29）。 他们在超过 15 项研究中发现了不同的结果，评估了 DNA 甲基化与出生体重相关的作用。这些研究大多是使用脐带血、胎盘组织和母体血液样本进行的，没有使用口腔（口腔）样本。这些差异由多种因素解释，包括样本量、使用的技术和方法、研究的特定区域以及包含潜在的混杂变量。

Additionally, there is a scarcity of research examining the variability of these epigenetic marks over time and their association with children's growth. The only study to date on this topic focused on GDM, conducted by Emper’s group, found differential DNA methylation profiles in children born to mothers with obesity or GDM compared to controls, with differences maintained over the first year of life in whole blood samples, showing epigenetic signatures enriched in metabolic pathways (30). Our findings align with these results, revealing that 16 DMPs, in saliva, remained differentially methylated over the first year of life, in the same direction. Some of these CpGs were annotated in genes, such as FARP2 and CYTH3, which are related to pathways involved in inflammatory processes and T2DM. Specifically, CYTH3 has roles in Golgi apparatus function and ADP-ribosylation factor regulation, with genome-wide association studies (GWAS) suggesting its association with BMI and height(31)(32). FARP2 is a gene that enables guanylyl-nucleotide exchange factor activity, and it is also involved in Rac protein signal transduction and neuronal remodelling. Polymorphisms of this gene are related to HDL concentrations and to height (33)(34). In our study, the cg10177795 which is annotated to FARP2 gene, was correlated with head circumference and height percentiles, supporting previous studies
此外，缺乏研究这些表观遗传标记随时间的变化及其与儿童成长的关系。迄今为止，由 Emper 的小组进行的关于该主题的唯一一项针对 GDM 的研究发现，与对照组相比，肥胖或 GDM 母亲所生儿童的 DNA 甲基化谱存在差异，在全血样本中的第一年保持差异，显示出富含代谢途径的表观遗传特征 （30）我们的研究结果与这些结果一致， 揭示了唾液中的 16 个 DMP 在出生后的第一年以相同的方向保持差异甲基化。其中一些 CpG 在基因中被注释，例如 FARP2 和 CYTH3，这些基因与参与炎症过程和 T2DM 的通路有关。具体来说，CYTH3 在高尔基体功能和 ADP-核糖基化因子调节中发挥作用，全基因组关联研究 （GWAS） 表明它与 BMI 和身高有关 （31）（32）。FARP2 是一种启用鸟苷酰核苷酸交换因子活性的基因，它还参与 Rac 蛋白信号转导和神经元重塑。该基因的多态性与 HDL 浓度和身高有关 （33） （34）。在我们的研究中，注释为 FARP2 基因的 cg10177795 与头围和身高百分位数相关，支持以前的研究.

To date, most studies have focused on identifying potential epigenetic markers in newborns through cord blood samples. However, the use of saliva samples offers several advantages, including the ability to conduct longitudinal follow-ups and improved applicability in clinical practice due to their non-invasive collection method and ethical suitability for paediatric research. Increasing evidence supports the use of saliva-based methylation studies, with established links between DNA methylation patterns and maternal BMI, gestational glucose levels, and various metabolic parameters (35) (36)(37). The correlations observed in our study between CpG sites and child anthropometric measures underscore the potential of saliva-derived epigenetic markers as accessible indicators of growth patterns, supporting their utility in non-invasive metabolic health monitoring.
迄今为止，大多数研究都集中在通过脐带血样本识别新生儿的潜在表观遗传标志物上。然而，使用唾液样本具有几个优势，包括能够进行纵向随访和由于其非侵入性采集方法和对儿科研究的伦理适用性而提高在临床实践中的适用性。越来越多的证据支持使用基于唾液的甲基化研究，DNA 甲基化模式与母体 BMI、妊娠葡萄糖水平和各种代谢参数之间建立了联系 （35）（36）（37）在我们的研究中观察到的 CpG 位点和儿童人体测量测量之间的相关性强调了唾液衍生的表观遗传标志物作为生长模式的可访问指标的潜力，支持它们在非侵入性中的实用性代谢健康监测。

Our study presents several strengths that contribute to its significance and potential clinical applicability. Firstly, the use of saliva samples provides a minimally invasive method for analyzing DNA methylation patterns, which is particularly advantageous for longitudinal studies in pediatric populations. This approach also enhances ethical considerations and facilitates the collection of samples over time, allowing for the monitoring of stable epigenetic markers. Secondly, our longitudinal design enables the assessment of DNA methylation changes across critical early life stages, revealing several persistent epigenetic differences between children of mothers with and without GDM, thus highlighting potential biomarkers of early metabolic risk. Importantly, the association of these methylation sites with anthropometric measures, such as weight, height, and BMI, underscores the potential of these markers to serve as indicators of growth patterns. Additionally, by focusing on an underexplored sample type—saliva—our study fills a gap in the literature, as most prior research has been limited to cord blood, placental, or maternal blood samples. This study also presents some limitations, mainly the small sample size and the lack of validation in other cohort. More studies are needed to replicate and validate these findings in larger and diverse populations.
我们的研究提出了有助于其重要性和潜在临床适用性的几个优势。首先，唾液样本的使用提供了一种分析 DNA 甲基化模式的微创方法，这对于儿科人群的纵向研究特别有利。这种方法还加强了伦理考虑，并促进了样本的收集，从而可以监测稳定的表观遗传标志物。其次，我们的纵向设计能够评估关键早期生命阶段的 DNA 甲基化变化，揭示患有和没有 GDM 的母亲的孩子之间的几个持续表观遗传差异，从而突出早期代谢风险的潜在生物标志物。重要的是，这些甲基化位点与人体测量指标（如体重、身高和 BMI）的关联强调了这些标志物作为生长模式指标的潜力。此外，通过关注一种未被充分探索的样本类型——唾液——我们的研究填补了文献中的空白，因为大多数以前的研究都仅限于脐带血、胎盘或母体血液样本。这项研究也存在一些局限性，主要是样本量小和在其他队列中缺乏验证。需要更多的研究来在更大和多样化的人群中复制和验证这些发现。

In conclusion, our study identifies a different DNA methylation pattern between children born to mothers with GDM and non-GDM across time, in saliva samples. Some of these epigenetic marks
总之，我们的研究在唾液样本中确定了 GDM 母亲和非 GDM 母亲所生的孩子在不同时间上的不同 DNA 甲基化模式。其中一些表观遗传标记

are associated with key anthropometric measurements, such as weight, height, head circumference, annual growth and BMI, indicating potential use as biomarkers risk of childhood obesity in the future. Notably, this is one of the few studies to analyze DNA methylation over time using saliva samples, which offer an accessible, non-invasive method for longitudinal studies.
与关键的人体测量数据有关，例如体重、身高、头围、年生长和 BMI，表明可能用作未来儿童肥胖风险的生物标志物。 值得注意的是，这是为数不多的使用唾液样本分析 DNA 甲基化随时间变化的研究之一，这为纵向研究提供了一种可访问的非侵入性方法。

The findings contribute to a growing body of evidence supporting the utility of saliva samples in epigenetic research and underscore their potential as non-invasive biomarkers for monitoring the developmental impact of maternal GDM and other prenatal exposures.
这些发现有助于越来越多的证据支持唾液样本在表观遗传学研究中的作用，并强调了它们作为非侵入性生物标志物的潜力，用于监测孕产妇 GDM 和其他产前暴露对发育的影响。

Funding
资金

This study was supported by the Juan Rodés program from the “Instituto de Salud Carlos III” (JR20-00040 to MM-V), the PFIS program (FI19/00178 to TML-P), and the Nicolas Monardes Program from the “Servicio Andaluz de Salud, Junta de Andalucía”, Spain (RC-0008-2021 to SM). In addition, this study was supported by the “Centros de Investigación Biomédica en Red” (CIBER) of the Institute of Health Carlos III (ISCIII) (CB06/03/0018), and research grants from “Servicio Andaluz de Salud”, Junta de Andalucía (PI-0283-2018, PI-0419-2019) and from the ISCIII (PI18/01175, PI21/01864). Grants were co-funded by the European Regional Development fund (ERDF) “Una manera de hacer Europa”.
这项研究得到了“卡洛斯三世健康研究所”的 Juan Rodés 计划（JR20-00040 至 MM-V）、 PFIS 计划（FI19/00178 至 TML-P）和西班牙“安达卢西亚军政府安达卢西亚健康服务”的 Nicolas Monardes 计划（RC-0008-2021 至 SM）的支持。此外，这项研究还得到了卡洛斯三世卫生研究所 （ISCIII） 的“Centros de Investigación Biomédica en Red” （CIBER） 的支持，以及“安达卢西亚健康服务”、安达卢西亚军政府 （PI-0283-2018， PI-0419-2019） 和 ISCIII （PI18/01175， PI21/01864） 的研究资助。赠款由欧洲区域发展基金 （ERDF） “Una manera de hacer Europa” 共同资助。

Institutional Review Board Statement: The study was approved by the Institutional Review Board at the Hospital Universitario Virgen de la Victoria de Málaga, Spain, in accordance with the Declaration of Helsinki. The legal guardians are Dr. Sonsoles Morcillo and Dr. María José Picón.
机构审查委员会声明：根据赫尔辛基宣言，该研究由西班牙马拉加维多利亚大学医院机构审查委员会批准。法定监护人是 Sonsoles Morcillo 博士和 María José Picón 博士。

Informed Consent Statement: Informed consent was obtained from all subjects.
知情同意书：获得所有受试者的知情同意。

Data Availability Statement: The raw DNAm array data are available in the XXXXX repository, under the accession number XXXXX (The authors are currently in the process of arranging data deposition in public repositories and commit to having this completed by the time of article acceptance). The other datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.
数据可用性声明：原始 DNAm 阵列数据可在 XXXXX 存储库中获得，登录号为 XXXXX（作者目前正在公共存储库中安排数据沉积，并承诺在文章接受时完成这项工作）在当前研究期间生成和/或分析的其他数据集可应合理要求从通讯作者处获得。

Acknowledgments: The authors thank the women who participated in the EPIDG study.
致谢：作者感谢参与 EPIDG 研究的女性。

Conflicts of Interest: The authors declare no conflicts of interest.
利益冲突：作者声明没有利益冲突。

Authorship: TML-P: methodology, analysis, and wrote original draft; A-LJ: collected anthropometric variables and reviewed the results and final version of the manuscript; A-PL: contributed to the manuscript; M-P: collected sample from EPIDG cohort; M-SA: involved in data collection; MM-V: EPIDG funding acquisition, acquired data from the patients and collected samples; MJP: EPIDG funding acquisition and acquired data from the patients and collected samples; SM: EPIDG funding acquisition, designed the study, interpretation of the data and has substantively revised the work and she is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
作者身份： TML-P：方法、分析并撰写原始草稿;A-LJ：收集人体测量变量并审查结果和手稿的最终版本;A-PL：为手稿做出贡献;M-P： 从 EPIDG 队列中收集的样本;M-SA：参与数据收集;MM-V：EPIDG 资金获取，从患者那里获取数据并收集样本;MJP：EPIDG 资金获取并从患者和收集样本中获取数据;SM：EPIDG 获得资金，设计研究，解释数据，并对工作进行实质性修改，她是这项工作的担保人，因此，她可以完全访问研究中的所有数据，并对数据的完整性和数据分析的准确性负责。

REFERENCES
引用

1.  world health organization. Obesity and overweight. [WWW document]. URL https://www.who.int/es/news-room/fact-sheets/detail/obesity-and-overweight
1. 世界卫生组织。肥胖和超重。[WWW 文档]。URL https://www.who.int/es/news-room/fact-sheets/detail/obesity-and-overweight

2.  Hjort L, Martino D, Grunnet LG, et al. Gestational diabetes and maternal obesity are associated with epigenome-wide methylation changes in children. JCI insight 2018;3.
2. Hjort L、Martino D、Grunnet LG 等人。 妊娠糖尿病和母亲肥胖与儿童表观基因组范围的甲基化变化有关。 JCI 洞察 2018;3。

3.  Heindel JJ, Balbus J, Birnbaum L, et al. Developmental Origins of Health and Disease: Integrating Environmental Influences. Endocrinology 2015;156:3416–3421.
3. Heindel JJ、Balbus J、Birnbaum L 等人。 健康与疾病的发展起源：整合环境影响。 内分泌学 2015;156：3416–3421。

4.  Ms TM, Arefadib N, Deery A, Megan M, Ms K, West S. The First Thousand Days The First Thousand Days: An Evidence Paper – Summary Centre for Community Child Health. 2017.
4. TM 女士、Arefadib N、Deery A、Megan M、K、West S。第一千天 第一千天：证据文件 – 社区儿童健康总结中心。2017 年。

5.  Damm P, Houshmand-Oeregaard A, Kelstrup L, Lauenborg J, Mathiesen ER, Clausen TD. Gestational diabetes mellitus and long-term consequences for mother and offspring: a view from Denmark. Diabetologia 2016.
5. Damm P， Houshmand-Oeregaard A， Kelstrup L， Lauenborg J， Mathiesen ER， Clausen TD.妊娠糖尿病及其对母亲和后代的长期影响：来自丹麦的观点。糖尿病学 2016。

6.  Goyal D, Limesand SW, Goyal R. Epigenetic responses and the developmental origins of health and disease. J Endocrinol 2019;242:T105–T119.
6. Goyal D， Limesand SW， Goyal R. 表观遗传反应以及健康和疾病的发育起源。 J 内分泌 2019;242：T105–T119。

7.  Yang I V., Zhang W, Davidson EJ, Fingerlin TE, Kechris K, Dabelea D. Epigenetic marks of in utero exposure to gestational diabetes and childhood adiposity outcomes: the EPOCH study. Diabet Med 2018;35:612–620.
7. Yang I V.， Zhang W， Davidson EJ， Fingerlin TE， Kechris K， Dabelea D. 子宫内妊娠糖尿病暴露和儿童肥胖结局的表观遗传标志：EPOCH 研究。 糖尿病医学 2018;35：612-620。

8.  Howe CG, Cox B, Fore R, et al. Maternal Gestational Diabetes Mellitus and Newborn DNA Methylation: Findings From the Pregnancy and Childhood Epigenetics Consortium. Diabetes Care 2019;43.
8. Howe CG、Cox B、Fore R 等人。 母体妊娠糖尿病和新生儿 DNA 甲基化：妊娠和儿童表观遗传学联盟的发现。 糖尿病护理 2019;43。

9.  Langie SAS, Moisse M, Declerck K, et al. Salivary DNA Methylation Profiling: Aspects to Consider for Biomarker Identification. Basic Clin Pharmacol Toxicol 2017;121:93–101.
9. Langie SAS、Moisse M、Declerck K 等人。 唾液 DNA 甲基化分析：生物标志物鉴定需要考虑的方面。 基础临床药理学毒理学 2017;121：93-101。

10.  Lowe R, Gemma C, Beyan H, et al. Buccals are likely to be a more informative surrogate tissue than blood for epigenome-wide association studies. Epigenetics 2013;8:445–454.
10. Lowe R、Gemma C、Beyan H 等人。 对于表观基因组范围的关联研究，口腔可能是比血液信息量更大的替代组织。 表观遗传学 2013;8：445–454。

11.  Moccia C, Popovic M, Isaevska E, et al. Birthweight DNA methylation signatures in infant saliva. Clin Epigenetics 2021;13.
11. Moccia C、Popovic M、Isaevska E 等人。 婴儿唾液中的出生体重 DNA 甲基化特征。 临床表观遗传学 2021;13。

12.  Linares-Pineda TM, Peña-Montero N, Gutiérrez-Repiso C, et al. Epigenome wide association study in peripheral blood of pregnant women identifies potential metabolic pathways related to gestational diabetes. Epigenetics 2023;18.
12. Linares-Pineda TM、Peña-Montero N、Gutiérrez-Repiso C 等人。孕妇外周血的表观基因组范围关联研究确定了与妊娠糖尿病相关的潜在代谢途径。 表观遗传学 2023;18。

13.  Sweeting A, Wong J, Murphy HR, Ross GP. A Clinical Update on Gestational Diabetes Mellitus. Endocr Rev 2022;43:763–793.
13. Sweeting A、Wong J、Murphy HR、Ross GP。妊娠糖尿病的临床更新。内分泌修订版 2022;43：763–793。

14.  Anon. WHO Child Growth Standards based on length/height, weight and age. Acta Paediatr Suppl 2006;450:76–85.
14. 佚名。基于身长/身高、体重和年龄的 WHO 儿童生长标准。 Acta Paediatr 增刊 2006;450：76-85。

15.  Cooke R, Goulet O, Huysentruyt K, et al. Catch-Up Growth in Infants and Young Children With Faltering Growth: Expert Opinion to Guide General Clinicians. J Pediatr Gastroenterol Nutr 2023;77:7–15.
15. Cooke R、Goulet O、Huysentruyt K 等人。 生长迟缓的婴幼儿的追赶性生长：指导普通临床医生的专家意见。 J 儿科胃肠营养学杂志 2023;77：7-15。

16.  Aryee MJ, Jaffe AE, Corrada-Bravo H, et al. Minfi: A flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics 2014;30:1363–1369.
16. Aryee MJ、Jaffe AE、Corrada-Bravo H 等人。Minfi：用于分析 Infinium DNA 甲基化微阵列芯片的灵活而全面的 Bioconductor 套装。 生物信息学 2014;30：1363–1369。

17.  Maksimovic J, Phipson B, Oshlack A. A cross-package Bioconductor workflow for analysing methylation array data. F1000Research 2017;5.
17. 马克西莫维奇 J， 菲普森 B， 奥什拉克 A.用于分析甲基化阵列数据的跨包 Bioconductor 工作流程。F1000Research 2017;5.

18.  Wang Z, Wu XL, Wang Y. A framework for analyzing DNA methylation data from Illumina Infinium HumanMethylation450 BeadChip. BMC Bioinformatics 2018;19.
18. 王 Z， 吴 XL， 王 Y.用于分析来自 Illumina Infinium HumanMethylation450 BeadChip 的 DNA 甲基化数据的框架。BMC 生物信息学 2018;19。

19.  Michels KB, Binder AM, Dedeurwaerder S, et al. Recommendations for the design and analysis of epigenome-wide association studies. Nat Methods 2013;10:949–955.
19. Michels KB、Binder AM、Dedeurwaerder S 等人。对表观基因组范围关联研究的设计和分析的建议。 Nat 方法 2013;10：949–955。

20.  Ritchie ME, Phipson B, Wu D, et al. Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 2015;43:e47.
20. Ritchie ME、Phipson B、Wu D 等人。 Limma 为 RNA 测序和微阵列研究的差异表达分析提供支持。 核酸研究 2015;43：e47.

21.  Du P, Zhang X, Huang CC, et al. Comparison of Beta-value and M-value methods for quantifying methylation levels by microarray analysis. BMC Bioinformatics 2010;11.
21. Du P， Zhang X， Huang CC， et al. 通过微阵列分析量化甲基化水平的 Beta 值和 M 值方法的比较。 BMC 生物信息学 2010;11。

22.  Kuleshov M V., Jones MR, Rouillard AD, et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res 2016;44:W90–W97.
22. Kuleshov M V.、Jones MR、Rouillard AD 等人。 Enrichr：全面的基因集富集分析 Web 服务器 2016 更新。 核酸研究 2016;44：W90–W97。

23.  Finer S, Mathews C, Lowe R, et al. Maternal gestational diabetes is associated with genome-wide DNA methylation variation in placenta and cord blood of exposed offspring.
23. Finer S、Mathews C、Lowe R 等人。 母体妊娠糖尿病与暴露后代胎盘和脐带血的全基因组 DNA 甲基化变异有关。

24.  Elliott HR, Sharp GC, Relton CL, Lawlor DA. Epigenetics and gestational diabetes: a review of epigenetic epidemiology studies and their use to explore epigenetic mediation and improve prediction. Diabetologia 2019;62:2171–2178.
24. Elliott HR、Sharp GC、Relton CL、Lawlor DA。表观遗传学和妊娠糖尿病：表观遗传流行病学研究综述及其在探索表观遗传介导和改进预测方面的用途。糖尿病学 2019;62：2171–2178。

25.  Rushing A, Sommer EC, Zhao S, Po’E EK, Barkin SL. Salivary epigenetic biomarkers as predictors of emerging childhood obesity. BMC Med Genet 2020;21.
25. Rushing A， Sommer EC， Zhao S， Po'E EK， Barkin SL. 唾液表观遗传生物标志物作为新出现的儿童肥胖的预测因子。 BMC Med Genet 2020;21。

26.  Sharma R, Frasch MG, Zelgert C, et al. Maternal–fetal stress and DNA methylation signatures in neonatal saliva: an epigenome-wide association study. Clin Epigenetics 2022;14:1–18.
26. Sharma R、Frasch MG、Zelgert C 等人。 新生儿唾液中的母胎应激和 DNA 甲基化特征：一项表观基因组范围的关联研究。 临床表观遗传学 2022;14：1-18。

27.  Canouil M, Khamis A, Keikkala E, et al. Epigenome-Wide Association Study Reveals Methylation Loci Associated With Offspring Gestational Diabetes Mellitus Exposure and Maternal Methylome. Diabetes Care 2021;44:1992–1999.
27. Canouil M、Khamis A、Keikkala E 等人。表观基因组范围关联研究揭示了与后代妊娠糖尿病暴露和母体甲基化组相关的甲基化位点。 糖尿病护理 2021;44：1992–1999 年。

28.  Franzago M, Borrelli P, Nicola M Di, et al. From Mother to Child: Epigenetic Signatures of Hyperglycemia and Obesity during Pregnancy. Nutr 2024, Vol 16, Page 3502 2024;16:3502.
28. Franzago M、Borrelli P、Nicola M Di 等人。从母亲到孩子：怀孕期间高血糖和肥胖的表观遗传特征。 营养学 2024，第 16 卷，第 3502 页 2024;16：3502。

29.  Saucedo R, Ferreira-Hermosillo A, Robledo-Clemente M, Díaz-Velázquez MF, Valencia-Ortega J. Association of DNA Methylation with Infant Birth Weight in Women with Gestational Diabetes. Metab 2024, Vol 14, Page 361 2024;14:361.
29. Saucedo R， Ferreira-Hermosillo A， Robledo-Clemente M， Díaz-Velázquez MF， Valencia-Ortega J. 妊娠糖尿病妇女 DNA 甲基化与婴儿出生体重的关联。 Metab 2024，第 14 卷，第 361 页 2024;14：361。

30.  Alba-Linares JJ, Pérez RF, Tejedor JR, et al. Maternal obesity and gestational diabetes reprogram the methylome of offspring beyond birth by inducing epigenetic signatures in metabolic and developmental pathways. Cardiovasc Diabetol 2023;22.
30. Alba-Linares JJ、Pérez RF、Tejedor JR 等人。孕产妇肥胖和妊娠糖尿病通过在代谢和发育途径中诱导表观遗传特征，在出生后重新编程后代的甲基化组。 心血管糖尿病醇 2023;22。

31.  Locke AE, Kahali B, Berndt SI, et al. Genetic studies of body mass index yield new insights for obesity biology. Nat 2015 5187538 2015;518:197–206.
31. Locke AE、Kahali B、Berndt SI 等人。 体重指数的遗传研究为肥胖生物学提供了新的见解。 2015 5187538 2015 年国家;518：197–206。

32.  Yengo L, Vedantam S, Marouli E, et al. A saturated map of common genetic variants associated with human height. Nat 2022 6107933 2022;610:704–712.
32. Yengo L、Vedantam S、Marouli E 等人。 与人类身高相关的常见遗传变异的饱和图谱。 2022 6107933 2022 年国家赛;610：704–712。

33.  Graham SE, Clarke SL, Wu KHH, et al. The power of genetic diversity in genome-wide association studies of lipids. Nat 2021 6007890 2021;600:675–679.
33. Graham SE、Clarke SL、Wu KHH 等人。 遗传多样性在脂质全基因组关联研究中的力量。 2021 6007890 2021 年国家赛;600：675–679。

34.  Sinnott-Armstrong N, Tanigawa Y, Amar D, et al. Genetics of 35 blood and urine biomarkers in the UK Biobank. Nat Genet 2021 532 2021;53:185–194.
34. Sinnott-Armstrong N、Tanigawa Y、Amar D 等人。 英国生物库中 35 种血液和尿液生物标志物的遗传学。 国家热内特 2021， 532， 2021;53：185-194。

35.  Lecorguille M, Mcauliffe FM, Twomey PJ, et al. Maternal Glycaemic and Insulinemic Status and Newborn DNA Methylation: Findings in Women With Overweight and Obesity. J Clin Endocrinol Metab 2022;108:85–98.
35. Lecorguille M、Mcauliffe FM、Twomey PJ 等人。 母体血糖和胰岛素状态以及新生儿 DNA 甲基化：超重和肥胖女性的发现。 J Clin 内分泌代谢 2022;108：85-98。

36.  Oelsner KT, Guo Y, To SB-C, Non AL, Barkin SL. Maternal BMI as a predictor of methylation of obesity-related genes in saliva samples from preschool-age Hispanic children at-risk for obesity. BMC Genomics 2017;18:57.
36. Oelsner KT， Guo Y， To SB-C， Non AL， Barkin SL. 母体 BMI 作为有肥胖风险的学龄前西班牙裔儿童唾液样本中肥胖相关基因甲基化的预测因子。 BMC 基因组学 2017;18：57。

37.  Murata Y, Fujii A, Kanata S, et al. Evaluation of the usefulness of saliva for DNA methylation analysis in cohort studies. Neuropsychopharmacol reports 2019;39:301–305.
37. Murata Y、Fujii A、Kanata S 等人。评估唾液对队列研究中 DNA 甲基化分析的有用性。 神经精神药理学报告 2019;39：301-305。

Legend Figures
图例

Figure 1. A) Manhattan plot representing the most differentiated CpGs per chromosome at the postpartum visit. B) Manhattan plot representing the most differentiated CpGs per chromosome at the one year of follow-up visit.
图 1.A） 曼哈顿图代表产后访视时每条染色体分化程度最高的 CpG。B） 曼哈顿图代表一年随访时每条染色体分化最多的 CpG。

Figure 2. CpGs that maintained the same direction of DNA methylation levels throughout the year. Postpartum: postpartum visit (8-10 weeks), Year: 1 year after birth.
图 2.全年保持相同 DNA 甲基化水平方向的 CpG。产后：产后访视（8-10 周），年份：出生后 1 年。

Figure 3. Enriched signalling pathways according to KEGG.
图 3.根据 KEGG 丰富的信号通路。

Additional files
其他文件

Additional file 1: Supplementary Figure 1.pptx
附加文件 1补充图 1.PPTX

Title of data: Quantile-quantile (QQ) plots of observed and expected distributions of p-values in our cohort. λ is the genomic inflation factor.
数据标题：我们队列中 p 值观测和预期分布的分位数-分位数 （QQ） 图。λ 是基因组膨胀因子。

Description of data: graphs showing the inflation factor of data.
数据描述：显示数据通货膨胀因子的图表。

Additional file 2: Supplementary table 1.xls
附加文件 2：补充表 1.xls

Title of data: Correlation analysis between DNA methylation levels and anthropometric variables
数据标题：DNA 甲基化水平与人体测量变量之间的相关性分析

Description of data
数据描述: