Intrauterine exposure to hyperglycaemia may increase the risk of later-life metabolic disorders. Although the underlying mechanism is not fully understood, epigenetic dysregulation in fetal programming has been implicated. With regard to energy homoeostasis, PGC-1α (peroxisome-proliferator-activated receptor γ co-activator-1α, encoded by the PPARGC1A gene) plays a regulatory role in several biochemical processes. We hypothesized that maternal gestational glucose levels would positively correlate with DNA methylation of the PPARGC1A promoter in placental tissue. We undertook a cross-sectional study of 58 mothers who underwent uncomplicated Caesarean delivery in a university hospital. Maternal gestational glucose concentration was determined after a 75-g OGTT (oral glucose tolerance test) at 24–28 weeks of gestation. Placenta tissue and cord blood were collected immediately after delivery. Genomic DNA was extracted and thereafter bisulfite conversion was performed. After PCR amplification, the DNA methylation of the PPARGC1A promoter was quantified using a pyrosequencing technique. The protein level of PGC-1α was evaluated by Western blotting. For all participants as a whole, including the GDM (gestational diabetes mellitus) and normoglycaemia groups, the maternal gestational glucose level was positively correlated with placental DNA methylation, and negatively correlated with cord blood DNA methylation of the PPARGC1A promoter in a CpG site-specific manner. In the GDM group alone, the placental CpG site-specific methylation of the PPARGC1A promoter strongly correlated with gestational 2-h post-OGTT glycaemia. Epigenetic alteration of the PPAGRC1A promoter may be one of the potential mechanisms underlying the metabolic programming in offspring exposed to intrauterine hyperglycaemia.

CLINICAL PERSPECTIVES

  • Intrauterine exposure to hyperglycaemia may increase the risk of later-life metabolic disorders; epigenetic dysregulation may be involved in this process. Peroxisome-proliferator-activated receptor γ co-activator-1α (PPARGC1A) plays a critical role in energy homoeostasis. We hypothesized that maternal gestational glycaemia would correlate with DNA methylation of the PPARGC1A promoter in placental tissue.

  • We found that, among the 58 participants [including the gestational diabetes mellitus (GDM) and normoglycaemia groups], the maternal gestational glucose level was positively correlated with placental DNA methylation of the PPARGC1A promoter in a CpG site-specific manner. The correlation between gestational 2-h post-OGTT glycaemia and CpG site-specific methylation was stronger in the GDM group.

  • Epigenetic alteration in the PPARGC1A promoter may be one of the potential mechanisms underlying the metabolic programming in offspring exposed to intrauterine hyperglycaemia.

INTRODUCTION

It has been well recognized that early-life conditions play a key role in adult health [13]. Many studies have shown that intrauterine exposure to hyperglycaemia increases the risk for obesity and diabetes in adulthood [46]. Notably, adiposity and perturbations in glucose metabolism may manifest early in children whose mothers had gestational hyperglycaemia [79]. Most intrauterine hyperglycaemia is caused by GDM (gestational diabetes mellitus), whose prevalence is likely to grow in parallel with the obesity pandemic [10]. Offspring of GDM mothers have a higher risk of long-term metabolic derangements such as being overweight and the metabolic syndrome [11]. Although the mechanism underlying fetal metabolic programming by maternal hyperglycaemia is unresolved, the involvement of epigenetic dysregulation has been implicated [12,13].

An adverse intrauterine environment, complicated by either undernutrition [14] or overfeeding [15], may programme harmful metabolic consequences later in life through epigenetic modifications such as DNA methylation. For instance, maternal gestational glycaemia concentrations correlate with placental DNA methylation adaptations of LEP (leptin) [16,17], ADIPOQ (adiponectin, C1Q and collagen domain-containing) [18] and ABCA1 (ATP-binding cassette transporter A1) [19] genes. Although the tenet of fetal metabolic programming has been extensively explored in animal models [20,21], only a small number of studies on epigenetic modification have been conducted in humans, especially in those with GDM.

The placenta plays an indispensable role in fetal growth and development, and placental dysfunction can unfavourably affect the lifelong health of both mother and offspring [22,23]. As an organ central in nutrition flux from mother to fetus, the placenta can yield biochemical clues to the impact of GDM on the fetus [24]. Furthermore, given its accessibility, the placenta is especially suitable to examine fetal epigenetic adaptations to the intrauterine milieu [16,18,24]. Being highly responsive to various environmental factors including nutrient signals, epigenetic modifications could conceivably predispose to placental dysfunction [22]. And, in concordance with this premise, placental DNA methylation of LEP [16,17], ADIPOQ [18] and ABCA1 [19] genes are associated with maternal gestational glycaemia, inferring that impaired placental function is mediated by epigenetic modifications [22].

PGC-1α (peroxisome-proliferator-activated receptor γ co-activator 1α, encoded by the PPARGC1A gene) is a powerful orchestrator of energy homoeostasis and metabolism, serving as a critical node linking environmental cues to energy metabolism [25]. Mitochondrial dysfunction plays an important role in diabetes, and PGC-1α is critical in mitochondrial biogenesis [25,26]. Recently, alterations in DNA methylation of the PPARGC1A promoter have been demonstrated in islets [27] and muscle [28] of patients with Type 2 diabetes as well as in liver of patients with NAFLD (non-alcoholic fatty liver disease) [29]. In humans, hypermethylation of the PPARGC1A promoter may be induced by a high-fat diet in a birth-weight-dependent manner [30], suggesting that epigenetic modification of PPARGC1A underlies the molecular aetiology of future metabolic disorders. Indeed, a recent study showed that GDM in humans preferentially affects the DNA methylation of placenta and cord blood genes which have been implicated in metabolic derangement [31].

In the present study, we examined statistical associations between the placental DNA methylation of the PPARGC1A promoter and maternal glucose concentrations at 24–28 weeks of gestation. We hypothesized that higher maternal gestational glucose levels would significantly correlate with more extensive DNA methylation of the PPARGC1A promoter.

MATERIALS AND METHODS

Participants and procedures

Eighty-two women of Chinese Han ethnicity with singleton pregnancy were enrolled. All underwent uncomplicated Caesarean delivery at a University Hospital in south China, from August 2013 to October 2013. Mothers at 24–28 weeks of gestation underwent a 75-g OGTT (oral glucose tolerance test) using a standard protocol and the glycaemic data were obtained from medical records. According to the diagnostic criteria of the IADPSG (International Association of Diabetes and Pregnancy Study Groups) [32,33], GDM was diagnosed when any of the following plasma glucose values were present: (i) fasting: ≥5.1 mmol/l; (ii) 1 h: ≥10.0 mmol/l; and (iii) 2 h: ≥8.5 mmol/l. Mothers with GDM were treated solely with diet. Participants were excluded from the study if the mother was >40 or <20 years of age, or was previously diagnosed with diabetes (including GDM) or any disorder (such as polycystic ovarian syndrome, uncontrolled thyroid and liver disease or renal insufficiency) known to affect glucose metabolism. Additional exclusion criteria included: history of substance (alcohol, tobacco and/or other drugs) abuse during the current pregnancy, premature delivery (<37 weeks of gestation), or birth weight <2500 g. Some participants were excluded owing to disorders which could alter placental DNA methylation (nine pregnancies resulted from in vitro fertilization) or glucose metabolism (four mothers received dexamethasone in pregnancy, and two had hyperthyroidism). Missing data, such as unavailable maternal gestational glucose records or unobtained cord blood samples, excluded nine participants. In total, data from 58 enrolees was available for analysis. Only women undergoing uncomplicated Caesarean deliveries were enrolled to exclude the influence of the mode of delivery on DNA methylation.

Placenta tissue (fetal) from the chorionic villi and intervillous tissue and cord blood were collected immediately after delivery. The fetal side of the placental samples were obtained by dissecting approximately 1-cm square-shaped segments of approximately 0.5 cm thickness close to the umbilical cord. The samples were rinsed in normal saline at 4°C to remove excess maternal blood. Thereafter, samples were frozen in liquid nitrogen and stored in the freezer at −80°C. Blood glucose was determined using a glucose-oxidase/peroxidase method (Biosino Bio-Technology and Science), and the intra-assay and interassay CVs (coefficients of variation) were 2% and 3% respectively. The insulin level was quantified using a chemiluminescent immunoassay (Access Ultrasensitive Insulin, Beckman Coulter), and the intra-assay and interassay CVs were 2–2.6% and 3.5–4.5% respectively. The HOMA-IR (homoeostatic model assessment of insulin resistance) was calculated as: HOMA-IR=fasting glucose (mmol/l)×fasting insulin (m-unit/l)/22.5.

All study procedures were approved by the Ethical Committee of Tongji Hospital, Huazhong University of Science and Technology, and written informed consent was obtained from each participant in accordance with the Declaration of Helsinki (2008).

DNA extraction and bisulfite treatment

Placental genomic DNA was extracted using a DNeasy Blood and Tissue Kit (Qiagen), whereas cord blood DNA was isolated using a QIAamp DNA blood mini kit (Qiagen) following the manufacturer's instructions. After the assessment of DNA quantity and quality with a UV spectrophotometer (Eppendorf), bisulfite conversion was performed using an EZ DNA Methylation Gold kit (Zymo Research) to convert unmethylated cytosines (C) into thymine (T).

Pyrosequencing

A DNA sequence starting 1000 bp upstream of the TSS (transcription start site) of PPARGC1A was retrieved from the Transcriptional Start Sites Database (http://dbtss.hgc.jp/) with the following ID numbers: NM_013261; Archive EnsEMBL: Chromosome 4: 23756664–23905712; TSS: 23891700. The sequence was also verified in the UCSC Genome bioinformatics Database (http://genome.ucsc.edu/). Six CpG sites in the PPARGC1A promoter were selected for analysis on the basis of previous studies [27,28,30,34,35] (Figure 1). Four primer assays covering the CpG sites of interest were designed using PyroMark assay design software (version 2.0.1.15; Qiagen). Assay1 (forward primer: 5′-GGGTATTAGGGTTGGAATTTAATGT-3′; reverse primer: 5′-CTTCCTTCTAATTATTTCCATTTCTCTCA-3′; seq-uencing primer: 5′-GTATTTTAAGGTAGTTAGGGA-3′) covered the CpG sites located 841 and −816 bp upstream of the PPARGC1A TSS, assay2 (forward and reverse primers were the same as assay1; sequencing primer: 5′-GTTGATT-TGAGTAGAGTAGTA-3′) covered the CpG site −783 bp upstream of the PPARGC1A TSS, assay3 (forward primer: 5′-TTGAAAGGATGGGGTTTTGTG-3′; reverse primer: 5′-AAACTCCTCCACCCAAAATTC-3′; sequencing primer: 5′-GGAGTAAAGAAAATTGTAGTAAT-3′) covered the CpG sites −652 and −617 bp upstream of the PPARGC1A TSS, and assay4 (forward primer: 5′-AGGGGATTTTGGTTATTATATGGT-3′; reverse primer: 5′-AAACCAACTTTAAATACCACAAACTCTA-3′; sequencing primer: 5′-GGATTTTGGTTATTATATGGTTAG-3′) covered the CpG site −260 bp upstream of the PPARGC1A TSS. Bisulfite-modified DNA was amplified using PyroMark PCR Master Mix (Qiagen) following the manufacturer's protocol. Biotinylated PCR products were immobilized on streptavidin-coated Sepharose beads (GE Healthcare) by constant agitation for 15 min to separate the double strands of DNA. After annealing of sequencing primers to samples, a pyrosequencing reaction was conducted in a PyroMark Q96 ID instrument (Qiagen) using PyroMark Gold Q96 Reagents (Qiagen). Quantification of cytosine methylation levels at each target CpG site of the PPARGC1A promoter were carried out using Pyrogram software (Qiagen), and data were subjected to quality control.

PPARGC1A gene locus (chromosome 4p15.1) and the CpG sites investigated in the PPARGC1A promoter

Figure 1
PPARGC1A gene locus (chromosome 4p15.1) and the CpG sites investigated in the PPARGC1A promoter

The six CpG sites analysed are marked with a perpendicular line. Numbers in the diamond shapes represent Spearman's correlation coefficients between CpG sites in the PPARGC1A promoter in placenta.

Figure 1
PPARGC1A gene locus (chromosome 4p15.1) and the CpG sites investigated in the PPARGC1A promoter

The six CpG sites analysed are marked with a perpendicular line. Numbers in the diamond shapes represent Spearman's correlation coefficients between CpG sites in the PPARGC1A promoter in placenta.

Western blot analysis

Placental tissue protein was extracted with RIPA lysis buffer containing 1 mM PMSF (Beyotime). Lysates were fractionated by SDS/PAGE (10% polyacrylamide gel) and subsequently electroblotted on to PVDF membranes (EMD Millipore). After blocking in Western blocking buffer (Beyotime), membranes were incubated with primary antibodies including those against PGC-1α (1:250 dilution; Abcam) and β-actin (1:1000 dilution; Antgene) overnight at 4°C. After washing with TBST (TBS with Tween 20) for 45 min, the membranes were probed with the HRP Affinipure secondary antibodies (1:10000 dilution; Earthox) for 1 h at 37°C. The membranes were washed with TBST for 1 h, and then developed using an ECL detection kit (Boster) in a ChemiDoc™ MP System with Image Lab™ Software (Bio-Rad Laboratories). β-Actin was used as a protein loading control.

Statistical analysis

A Shapiro–Wilk test was used to determine the probability distributions of parameters. Normally distributed variables including maternal glycaemic concentration at OGTT, age, pre-gestational BMI (body mass index), gestational weight gain, cord blood glucose and insulin as well as HOMA-IR, offspring birth weight and gestational age at delivery between the GDM and normoglycaemic groups were evaluated using independent-sample Student's t test. The distribution in offspring parity and sex between the two groups was analysed using a χ2 test. The DNA methylation level of the PPARGC1A promoter, which was non-normally distributed data, was analysed using a Mann–Whitney test. Associations between the CpG site-specific methylation level of the PPARGC1A promoter in placenta and the variables of interest (maternal glycaemic results at OGTT, gestational weight gain, pre-gestational BMI and cord blood HOMA-IR) were analysed by Spearman's rank correlation coefficient (rs). Correlation between normally distributed variables was examined using Pearson's correlation test. When appropriate, the potential confounders (maternal age, pre-gestational BMI, gestational weight gain, cord blood HOMA-IR, and offspring birth weight, sex, parity and gestational age at delivery) were incorporated into the correlation analysis. Results were considered statistically significant when P<0.01 (two-tailed), and a ‘trend’ was defined as 0.01<P<0.05 (two-tailed). All statistical analyses were performed using SPSS software (version 13.0).

RESULTS

Twenty-four women were classified as having GDM, and 34 were normoglycaemic according to the diagnostic criteria of the IADPSG consensus [33]. Maternal and newborn demographic and descriptive information including metabolic characteristics are presented in Table 1. On average, the OGTT glycaemia (fasting, 1-h and 2-h post-OGTT) concentrations at 24–28 weeks of gestation and cord blood glucose concentrations were all slightly higher in the GDM group (P<0.01). Cord blood HOMA-IR also showed a similar trend (P<0.05). No significant difference was observed in the other maternal and newborn characteristics including multivitamin (folic acid) intake. The average CpG site-specific methylation levels of the PPARGC1A promoter in cord blood and placenta were similar between the groups (Supplementary Tables S1 and S2). The placental protein levels of PGC-1α were comparable between the two groups (Supplementary Figure S1). Besides, the protein level did not correlate with the CpG site-specific methylation level of the PPARGC1A promoter in placenta with the adjustment of potential confounders (P > 0.1). Additionally, the methylation levels of CpG sites −841, −816 and −617 were generally well correlated with the methylation levels of other CpG sites, whereas the methylation level of CpG sites −260 and −783 did not correlate well with the other placental CpG sites, indicating that the epigenetic modification on CpG sites −260 and −783 may affect the placental function differently.

Table 1
Maternal and newborn characteristics

Results are means±S.D. except for the offspring sex and parity which are expressed as a ratio. †0.05<P<0.1, *P<0.05, **P<0.01.

CharacteristicGDMNormoglycaemia
n 24 34 
Mother's age (years) 30.60±4.07 29.50±3.31 
Pre-gestational weight (kg) 56.47±9.00 53.90±7.50 
Pre-gestational BMI (kg/m221.40±2.89 21.36±2.50 
Gestational weight gain (kg) 15.43±5.50 18.13±5.95† 
Cord blood insulin (m-unit/l) 6.48±2.80 5.20±2.50† 
Cord blood glucose (mmol/l) 4.00±0.88 3.53±0.65* 
Cord blood HOMA-IR 1.19±0.64 0.83±0.41* 
2-h post-OGTT glucose (mmol/l) at 24–28 weeks of gestation 8.73±1.43 6.49±0.94** 
1-h post-OGTT glucose (mmol/l) at 24–28 weeks of gestation 10.50±1.41 7.78±1.18** 
Fasting plasma glucose (mmol/l) at 24–28 weeks of gestation 5.00±0.45 4.32±0.44** 
Birth weight (kg) 3.29±0.46 3.28±0.35 
Gestational age at delivery (days) 270.79±6.27 272.56±6.87 
Offspring sex (males/females) 9/15 16/18 
Parity (% primiparous) 79.20 73.50 
CharacteristicGDMNormoglycaemia
n 24 34 
Mother's age (years) 30.60±4.07 29.50±3.31 
Pre-gestational weight (kg) 56.47±9.00 53.90±7.50 
Pre-gestational BMI (kg/m221.40±2.89 21.36±2.50 
Gestational weight gain (kg) 15.43±5.50 18.13±5.95† 
Cord blood insulin (m-unit/l) 6.48±2.80 5.20±2.50† 
Cord blood glucose (mmol/l) 4.00±0.88 3.53±0.65* 
Cord blood HOMA-IR 1.19±0.64 0.83±0.41* 
2-h post-OGTT glucose (mmol/l) at 24–28 weeks of gestation 8.73±1.43 6.49±0.94** 
1-h post-OGTT glucose (mmol/l) at 24–28 weeks of gestation 10.50±1.41 7.78±1.18** 
Fasting plasma glucose (mmol/l) at 24–28 weeks of gestation 5.00±0.45 4.32±0.44** 
Birth weight (kg) 3.29±0.46 3.28±0.35 
Gestational age at delivery (days) 270.79±6.27 272.56±6.87 
Offspring sex (males/females) 9/15 16/18 
Parity (% primiparous) 79.20 73.50 

For the participants as a whole, the fasting OGTT glucose level was positively correlated with DNA methylation in CpG site −260 of the PPARGC1A promoter in placenta (P=0.002), and the result remained unchanged after further adjusting this statistical model for gestational weight gain, pre-gestational BMI, parity, cord blood HOMA-IR as confounders (Table 2 and Figure 2). Moreover, correlations remained significant after further adjusting for newborn birth weight, sex, gestational age at delivery and maternal age (P=0.001). No significant correlation between 1-h or 2-h post-OGTT glucose and CpG site-specific methylation of the PPARGC1A promoter in placenta was found. Nevertheless, after adjusting the statistical models for maternal gestational weight gain, cord blood HOMA-IR, pre-gestational BMI and parity, newborn birth weight, sex, gestational age at delivery and maternal age as potential confounders, 2-h post-OGTT showed a positive trend with the DNA methylation level of CpG site −816 (P=0.034, Table 2).

Spearman's correlation between maternal gestational OGTT glucose concentration and placental DNA methylation level in the PPARGC1A promoter in all participants

Figure 2
Spearman's correlation between maternal gestational OGTT glucose concentration and placental DNA methylation level in the PPARGC1A promoter in all participants

Values are adjusted for gestational weight gain, pre-gestational BMI, parity, cord blood HOMA-IR, newborn birth weight, sex and gestational age at delivery and maternal age (n=58).

Figure 2
Spearman's correlation between maternal gestational OGTT glucose concentration and placental DNA methylation level in the PPARGC1A promoter in all participants

Values are adjusted for gestational weight gain, pre-gestational BMI, parity, cord blood HOMA-IR, newborn birth weight, sex and gestational age at delivery and maternal age (n=58).

Table 2
Spearman's correlation coefficient between maternal gestational OGTT glucose concentration and placental PPARGC1A methylation level in all participants

* Indicates adjustment for gestational weight gain, pre-gestational BMI, parity and cord blood HOMA-IR; † indicates adjustment for gestational weight gain, pre-gestational BMI, parity, cord blood HOMA-IR, newborn birth weight, sex and gestational age at delivery and maternal age; correlations with p<0.05 are in bold, whereas correlations with P<0.01 are in bold and italic.

Fasting OGTT glucose level (n=58)2-h post-OGTT glucose level (n=58)
CpG siteMeasure**
−841 rs 0.11 0.10 0.08 0.16 0.27 0.26 
 P 0.41 0.45 0.55 0.24 0.05 0.06 
−816 rs 0.10 0.12 0.14 0.135 0.27 0.29 
 P 0.45 0.40 0.32 0.31 0.04 0.03 
−260 rs 0.40 0.40 0.41 0.20 0.18 0.18 
 P 0.002 0.002 0.002 0.13 0.19 0.20 
Fasting OGTT glucose level (n=58)2-h post-OGTT glucose level (n=58)
CpG siteMeasure**
−841 rs 0.11 0.10 0.08 0.16 0.27 0.26 
 P 0.41 0.45 0.55 0.24 0.05 0.06 
−816 rs 0.10 0.12 0.14 0.135 0.27 0.29 
 P 0.45 0.40 0.32 0.31 0.04 0.03 
−260 rs 0.40 0.40 0.41 0.20 0.18 0.18 
 P 0.002 0.002 0.002 0.13 0.19 0.20 

The correlation between maternal glycaemic status and placental DNA methylation of the PPARGC1A promoter in the GDM group was analysed further. The fasting OGTT glucose level showed a positive trend with DNA methylation of CpG site −841 (0.01<P<0.05, Table 3). 1-h post-OGTT glucose was not significantly associated with DNA methylation of any CpG site, although the relationship was improved substantially for CpG site −816 upon adjustment for gestational weight gain and cord blood HOMA-IR (P<0.01, Table 3), the association became insignificant after further adjusting for parity and pre-gestational BMI (P > 0.1, Table 3), again affirming that the four above-mentioned factors are relevant confounders. Notably, 2-h post-OGTT glucose was positively correlated with DNA methylation of CpG sites −841 and −816 (P<0.01), and the association was stronger with adjustment for maternal gestational weight gain, HOMA-IR, pre-gestational BMI and parity (P<0.005). This correlation remained unaltered after further adjusting for newborn birth weight, sex, gestational age at delivery and maternal age (P<0.005, Table 3 and Figure 3).

Table 3
Spearman's correlation coefficient between maternal gestational OGTT glucose concentration and placental PPARGC1A methylation level in the GDM group

* Indicates adjustment for gestational weight gain, pre-gestational BMI, parity and cord blood HOMA-IR; † indicates adjustment for gestational weight gain, pre-gestational BMI, parity, cord blood HOMA-IR, newborn birth weight, sex and gestational age at delivery and maternal age; correlations with P<0.05 are in bold, whereas correlations with P<0.01 are in bold and italic.

Fasting OGTT glucose level (n=24)1-h post-OGTT glucose level (n=24)2-h post-OGTT glucose level (n=24)
CpG siteMeasurement***
−841 rs 0.44 0.16 -0.03 0.22 0.52 0.25 0.52 0.68 0.68 
 P 0.03 0.50 0.92 0.31 0.01 0.30 0.009 0.000 0.001 
−816 rs 0.36 0.22 0.05 0.34 0.58 0.33 0.53 0.64 0.63 
 P 0.09 0.33 0.84 0.11 0.005 0.15 0.008 0.001 0.003 
Fasting OGTT glucose level (n=24)1-h post-OGTT glucose level (n=24)2-h post-OGTT glucose level (n=24)
CpG siteMeasurement***
−841 rs 0.44 0.16 -0.03 0.22 0.52 0.25 0.52 0.68 0.68 
 P 0.03 0.50 0.92 0.31 0.01 0.30 0.009 0.000 0.001 
−816 rs 0.36 0.22 0.05 0.34 0.58 0.33 0.53 0.64 0.63 
 P 0.09 0.33 0.84 0.11 0.005 0.15 0.008 0.001 0.003 

Maternal gestational OGTT glucose was correlated with placental DNA methylation level in the PPARGC1A promoter in the GDM group

Figure 3
Maternal gestational OGTT glucose was correlated with placental DNA methylation level in the PPARGC1A promoter in the GDM group

(A) Methylation at CpG site −841. (B) Methylation at CpG site −815. Values are adjusted for gestational weight gain, pre-gestational BMI, parity, cord blood HOMA-IR, newborn birth weight, sex and gestational age at delivery and maternal age (n=24).

Figure 3
Maternal gestational OGTT glucose was correlated with placental DNA methylation level in the PPARGC1A promoter in the GDM group

(A) Methylation at CpG site −841. (B) Methylation at CpG site −815. Values are adjusted for gestational weight gain, pre-gestational BMI, parity, cord blood HOMA-IR, newborn birth weight, sex and gestational age at delivery and maternal age (n=24).

Interestingly, in all participants, the methylation level of CpG site −841 had a significant negative correlation (P<0.01) with 2-h post-OGTT glycaemia; the 1-h post-OGTT glycaemia showed a negative trend (P<0.05) with adjustment for potential confounders (Table 4 and Figure 4). Besides, the methylation level of CpG sites −816 and −617 in PPARGC1A promoter in cord blood showed a negative trend (P<0.05) with 2-h post-OGTT glycaemia after adjusting for potential confounders (Table 4).

Spearman's correlation between maternal gestational OGTT glucose concentration and cord blood DNA methylation level in the PPARGC1A promoter in all participants

Figure 4
Spearman's correlation between maternal gestational OGTT glucose concentration and cord blood DNA methylation level in the PPARGC1A promoter in all participants

The correlation was adjusted for gestational weight gain, pre-gestational BMI, parity, cord blood HOMA-IR, newborn birth weight, sex and gestational age at delivery and maternal age (n=53).

Figure 4
Spearman's correlation between maternal gestational OGTT glucose concentration and cord blood DNA methylation level in the PPARGC1A promoter in all participants

The correlation was adjusted for gestational weight gain, pre-gestational BMI, parity, cord blood HOMA-IR, newborn birth weight, sex and gestational age at delivery and maternal age (n=53).

Table 4
Spearman's correlation coefficient between maternal gestational OGTT glucose concentration and cord blood PPARGC1A methylation level in all participants

† Indicates adjustment for gestational weight gain, pre-gestational BMI, parity, cord blood HOMA-IR, newborn birth weight, sex and gestational age at delivery and maternal age; correlations with P<0.05 are in bold, whereas correlations with P<0.01 are in bold and italic.

1-h post-OGTT glucose level (n=53)2-h post-OGTT glucose level (n=53)
CpG siteMeasurement
−841 rs −0.37 −0.35 −0.40 −0.40 
 P 0.006 0.02 0.003 0.007 
−816 rs −0.19 −0.23 −0.25 −0.30 
 P 0.17 0.12 0.08 0.04 
−617 rs −0.21 −0.26 −0.31 −0.35 
 P 0.13 0.08 0.02 0.02 
1-h post-OGTT glucose level (n=53)2-h post-OGTT glucose level (n=53)
CpG siteMeasurement
−841 rs −0.37 −0.35 −0.40 −0.40 
 P 0.006 0.02 0.003 0.007 
−816 rs −0.19 −0.23 −0.25 −0.30 
 P 0.17 0.12 0.08 0.04 
−617 rs −0.21 −0.26 −0.31 −0.35 
 P 0.13 0.08 0.02 0.02 

Since gestational weight gain appeared to be influenced by GDM diagnostic (as shown in Table 1), the correlation between DNA methylation and gestational weight gain was analysed in the normoglycaemic group alone, and the correlation was not significant with the adjustment of potential confounders (Supplementary Table S3), implying that the gestational weight gain may not contribute to the regulation of the DNA methylation of the PPARGC1A promoter in placenta.

The methylation of CpG sites −816 and −783 correlated positively with cord blood HOMA-IR, and after adjusting the statistical model for confounders (including newborn birth weights, sex, parity and gestational age at delivery, maternal ages, pre-gestational BMI, 2-h post-OGTT glucose level and gestational weight gain), only the correlation for CpG site −783 showed a positive trend (rs=0.319, P=0.024).

DISCUSSION

The maxim that the intrauterine environment programmes long-term susceptibility to metabolic disorders has been convincingly demonstrated [1,2,4,11]. Although the underlying molecular mechanism is unclear, it has been suggested that epigenetic alterations are involved. In the present study, a higher placental methylation level and a lower cord blood methylation level of the PPARGC1A promoter were associated with higher maternal glucose concentrations at 24–28 weeks of gestation, independent of potential confounders, in a CpG site-specific manner. Our results may explain, in part, the previous observations that offspring exposed to intrauterine hyperglycaemia are prone to develop, years later, metabolic diseases such as Type 2 diabetes [4,5,79,11,36].

Epigenetic modifications are mitotically stable, with marginal changes over time, and with long-lasting effects on health [14,37], although they also can be ephemeral and rapidly altered [38]. The epigenome is more responsive to intrauterine environmental cues than to those encountered later in life [39]. In addition, if epigenetic modifications transpire early in life, they may be more immutable, with more of an impact on health over the lifespan [39]. The DNA methylation level of several gene promoters in umbilical cord tissue or cord blood have been found to correlate with the child's percentage fat mass at 9 years of age [40,41]. These reports reinforce the notion that epigenetic modifications in response to intrauterine milieu may also transpire in other tissues such as adipose, and predispose to later metabolic sequelae. PGC-1α is highly responsive to environmental cues and hormonal signals, thereby it is poised to play a prominent role in energy homoeostasis [25]. Insofar as PGC-1α is expressed in the major insulin-target organs (skeletal muscle, liver and adipose), as well as in islets, all of which are exposed to a similar intrauterine biochemical milieu to that of fetal placenta, it is conceivable that the methylation levels of the PPARGC1A promoter in these organs, akin to that of placenta, are also affected. Indeed, maternal gestational glucose concentrations are significantly associated with insulin sensitivity in the child [9], as well as adult obesity and glucose intolerance [4,5,11,36]. The PPARGC1A promoter methylation level is similarly increased in the liver from patients with NAFLD [29], as well as in islets [27] and skeletal muscle [28] in adults with Type 2 diabetes, and its expression is associated with insulin sensitivity. Paradoxically, PPARGC1A promoter methylation was positively correlated with insulin sensitivity in first-degree relatives of patients with Type 2 diabetes [34].

Among the group as a whole, gestational fasting OGTT glycaemia correlated with placental methylation level of CpG site −260, and maternal 2-h post-OGTT glycaemia correlated with cord blood methylation level of CpG site −841 in the PPARGC1A promoter, yet the correlations were not found in the GDM group alone, indicating that the epigenetic modifications of placental CpG site −260 and cord blood CpG site −841 are sensitive to maternal gestational glycaemia alteration, but may be not good markers of GDM. These correlations also imply that there is likely no overt threshold for risk of hyperglycaemia. This is salient insofar as fetal growth may be affected as a continuum within the full range of maternal glucose concentrations [32]. Similarly, placental adiponectin gene DNA methylation is associated with maternal glycaemia across a wide range of concentrations encompassing normal and impaired glucose intolerance [18]. Perhaps this accounts for the previous reports that maternal gestational glycaemia correlates with the risk of diabetes in offspring even when the gravid mother has an upper range, yet normal, glucose tolerance [42].

Interestingly, the correlation trend between PPARGC1A methylation in cord blood and gestational glycaemia was antithetical to that of placenta, similar to the findings of others [19], supporting the notion that DNA methylation patterns of specific genes can vary among tissues. The disparate epigenetic signature of an organ or tissue may reflect their different functions [4346]. It also indicates that the DNA methylation pattern of PPARGC1A in the placenta or cord blood may not apply to the muscle, liver, adipose tissue or islet of the offspring.

When analysing the GDM group alone, the placental DNA methylation level of −841 and −816 strongly corresponded with gestational 2-h post-OGTT glucose concentrations, consistent with the findings that the placental LEP gene DNA methylation level correlates with gestational glucose concentration in the impaired glucose tolerance group [16]. The correlations suggest that the placental methylation alterations of the two CpG sites −841 and −816 of the PPARGC1A promoter may be more responsive to an intrauterine hyperglycaemic stress, and thus can serve as GDM indicators, and these epigenetic alterations are a possible underlying pathogenesis for the fetal metabolic programming of future diseases susceptibilities.

Consistent with a previous study [47], gestational maternal weight gain was also positively associated with newborn birth weight. Gemma et al. [48] reported that maternal pre-gestational BMI is associated with PPARGC1A promoter methylation in the umbilical cord, an observation not corroborated in the present study, but likely to be due to the different ethnicities, the metabolic backgrounds of the participants (no patients had GDM in the former study), tissue tested (umbilical cord compared with placenta), and assay methodology.

The protein level did not correlate with the CpG site-specific methylation level of the PPARGC1A promoter in placenta (P > 0.1). This could be due to the complex regulating network, in which PGC-1α protein expression could be delicately regulated by both transcriptional and post-translational mechanisms in response to various metabolic cues [49]. For example, histone modifications could also influence protein expression [22]. Indeed, some previous studies did not observe correlations between gene expression and DNA methylation [30,34,50].

The placenta establishes an intimate metabolic communication between fetus and mother, playing a critical role in fetal development and growth. The fact that placental development can affect fetal nutritional supplements and contribute further to adult diseases gave birth to the notion of ‘placental origins of adult disease’ [51,52]. Epigenetic modifications occur from pre-implantation throughout pregnancy, and proper epigenetic regulation is substantial for placental development and function [22]. A previous study revealed that chronic exposure to a hyperglycaemic environment can lead to persistent alterations in DNA methylation of metabolic genes in Schwann cells [53]. Placental DNA methylation is responsive to intrauterine hyperglycaemia, as demonstrated in the present study by the correlations between maternal gestational glycaemia and the DNA methylation of PPARGC1A as well as other genes involved in energy homoeostasis [16,18,19,54]. The alterations in placental DNA methylation can disturb placental function which may contribute to disease susceptibility in later life [22]. Furthermore, PPARGC1A methylation alterations have been demonstrated in diabetes [27,28] and NAFLD [29] patients as well as low-birth-weight individuals [30] and IUGR (intrauterine growth restriction) rats [55]. It is likely that the placental DNA methylation adaptation of the PPARGC1A promoter in response to GDM is a surrogate marker for, or responsible for, future metabolic diseases in offspring such as Type 2 diabetes and NAFLD [51,52]. For example, besides serving as a metabolic interface, the placenta has numerous endocrine functions. The alteration in DNA methylation of the PPARGC1A promoter infers that the PPARGC1A methylation patterns of fetal endocrine organs, such as islets, and the responsivity to insulin in muscle and adipose tissue may also have been disrupted. Although the alterations are probably tissue- and organ-specific, the epigenetic deregulation of PPARGC1A in these organs/tissues could disturb downstream metabolic signals such as mitochondrial density [28,29], eventually programing higher risk of Type 2 diabetes in offspring in later life.

One strength of the present study is the well validated technology for the determination of DNA methylation. Furthermore, the present study was conducted in a homogeneous population of Chinese Han ethnicity; as such, the findings may not be applicable to other populations. Additionally, all correlations have been adjusted for possible confounders to avoid specious conclusions. A limitation of the present study is that the expression of PPARGC1A mRNA in the placenta was not determined. Although we found a positive correlation between maternal gestational glycaemia and placental DNA methylation, we cannot conclude irrefutably that the maternal glucose concentrations lead to the epigenetic alterations. In addition, the maternal gestational glucose data from a single 70 g OGTT at 24–28 weeks of gestation may not adequately reflect maternal glycaemic status throughout pregnancy. Future studies to examine the long-term consequence of alterations in DNA methylation of the PPARGC1A promoter in offspring may verify the overarching premise: maternal gestational glycaemia programmes later-life metabolic risk in offspring via epigenetic adaptations.

In conclusion, maternal gestational glycaemic concentrations are positively associated with DNA methylation of the PPARGC1A promoter in the placenta, suggesting that the epigenetic modification in the PPARGC1A promoter may predispose the child to eventual metabolic derangements.

AUTHOR CONTRIBUTION

Xuemei Xie and Hongjie Gao contributed to the design of the study, acquisition, analysis and interpretation of data and drafting the paper. Wanjiang Zeng, Suhua Chen, Ling Feng, Dongrui Deng, Fu-yuan Qiao and Lihong Liao contributed to acquisition of data and revision of the paper. Kenneth McCormick and Qin Ning contributed to analysis and interpretation of data, and revising the paper critically. Xiaoping Luo contributed to design, interpretation of data and revising the paper critically. All authors approved the final version of the paper.

We are grateful to the mothers who agreed to participate in the study.

FUNDING

This study was supported by the National Natural Science Foundation of China [grant number 81170627] and Program for Changjiang Scholars and Innovative Research Team in University [grant number PCSIRT1131].

Abbreviations

     
  • BMI

    body mass index

  •  
  • CV

    coefficient of variation

  •  
  • GDM

    gestational diabetes mellitus

  •  
  • HOMA-IR

    homoeostatic model assessment of insulin resistance

  •  
  • IADPSG

    International Association of Diabetes and Pregnancy Study Groups

  •  
  • NAFLD

    non-alcoholic fatty liver disease

  •  
  • OGTT

    oral glucose tolerance test

  •  
  • PGC-1α

    peroxisome-proliferator-activated receptor γ co-activator 1α

  •  
  • TBST

    TBS with Tween 20

  •  
  • TSS

    transcription start site

References

References
1
Barker
 
D.J.
Hales
 
C.N.
Fall
 
C.H.
Osmond
 
C.
Phipps
 
K.
Clark
 
P.M.
 
Type 2 (non-insulin-dependent) diabetes mellitus, hypertension and hyperlipidaemia (syndrome X): relation to reduced fetal growth
Diabetologia
1993
, vol. 
36
 (pg. 
62
-
67
)
[PubMed]
2
Hales
 
C.N.
Barker
 
D.J.
 
Type 2 (non-insulin-dependent) diabetes mellitus: the thrifty phenotype hypothesis
Diabetologia
1992
, vol. 
35
 (pg. 
595
-
601
)
[PubMed]
3
Sarr
 
O.
Yang
 
K.
Regnault
 
T.R.
 
In utero programming of later adiposity: the role of fetal growth restriction
J. Pregnancy
2012
, vol. 
2012
 pg. 
134758
 
[PubMed]
4
Lawlor
 
D.A.
Lichtenstein
 
P.
Langstrom
 
N.
 
Association of maternal diabetes mellitus in pregnancy with offspring adiposity into early adulthood: sibling study in a prospective cohort of 280,866 men from 248,293 families
Circulation
2011
, vol. 
123
 (pg. 
258
-
265
)
[PubMed]
5
Damm
 
P.
 
Future risk of diabetes in mother and child after gestational diabetes mellitus
Int. J. Gynaecol. Obstet.
2009
, vol. 
104
 
Suppl. 1
(pg. 
S25
-
S26
)
[PubMed]
6
Segar
 
E.M.
Norris
 
A.W.
Yao
 
J.R.
Hu
 
S.
Koppenhafer
 
S.L.
Roghair
 
R.D.
Segar
 
J.L.
Scholz
 
T.D.
 
Programming of growth, insulin resistance and vascular dysfunction in offspring of late gestation diabetic rats
Clin. Sci.
2009
, vol. 
117
 (pg. 
129
-
138
)
[PubMed]
7
Krishnaveni
 
G.V.
Hill
 
J.C.
Leary
 
S.D.
Veena
 
S.R.
Saperia
 
J.
Saroja
 
A.
Karat
 
S.C.
Fall
 
C.H.
 
Anthropometry, glucose tolerance, and insulin concentrations in Indian children: relationships to maternal glucose and insulin concentrations during pregnancy
Diabetes Care
2005
, vol. 
28
 (pg. 
2919
-
2925
)
[PubMed]
8
Bunt
 
J.C.
Tataranni
 
P.A.
Salbe
 
A.D.
 
Intrauterine exposure to diabetes is a determinant of hemoglobin A1c and systolic blood pressure in pima Indian children
J. Clin. Endocrinol. Metab.
2005
, vol. 
90
 (pg. 
3225
-
3229
)
[PubMed]
9
Bush
 
N.C.
Chandler-Laney
 
P.C.
Rouse
 
D.J.
Granger
 
W.M.
Oster
 
R.A.
Gower
 
B.A.
 
Higher maternal gestational glucose concentration is associated with lower offspring insulin sensitivity and altered β-cell function
J. Clin. Endocrinol. Metab.
2011
, vol. 
96
 (pg. 
E803
-
E809
)
[PubMed]
10
Norman
 
J.E.
Reynolds
 
R.M.
 
The consequences of obesity and excess weight gain in pregnancy
Proc. Nutr. Soc.
2011
, vol. 
70
 (pg. 
450
-
456
)
[PubMed]
11
Clausen
 
T.D.
Mathiesen
 
E.R.
Hansen
 
T.
Pedersen
 
O.
Jensen
 
D.M.
Lauenborg
 
J.
Schmidt
 
L.
Damm
 
P.
 
Overweight and the metabolic syndrome in adult offspring of women with diet-treated gestational diabetes mellitus or type 1 diabetes
J. Clin. Endocrinol. Metab.
2009
, vol. 
94
 (pg. 
2464
-
2470
)
[PubMed]
12
Fraser
 
A.
Lawlor
 
D.A.
 
Long-term health outcomes in offspring born to women with diabetes in pregnancy
Curr. Diab. Rep.
2014
, vol. 
14
 pg. 
489
 
[PubMed]
13
Zinkhan
 
E.K.
Fu
 
Q.
Wang
 
Y.
Yu
 
X.
Callaway
 
C.W.
Segar
 
J.L.
Scholz
 
T.D.
McKnight
 
R.A.
Joss-Moore
 
L.
Lane
 
R.H.
 
Maternal hyperglycemia disrupts histone 3 lysine 36 trimethylation of the IGF-1 gene
J. Nutr. Metab.
2012
, vol. 
2012
 pg. 
930364
 
[PubMed]
14
Heijmans
 
B.T.
Tobi
 
E.W.
Stein
 
A.D.
Putter
 
H.
Blauw
 
G.J.
Susser
 
E.S.
Slagboom
 
P.E.
Lumey
 
L.H.
 
Persistent epigenetic differences associated with prenatal exposure to famine in humans
Proc. Natl. Acad. Sci. U.S.A.
2008
, vol. 
105
 (pg. 
17046
-
17049
)
[PubMed]
15
Plagemann
 
A.
Roepke
 
K.
Harder
 
T.
Brunn
 
M.
Harder
 
A.
Wittrock-Staar
 
M.
Ziska
 
T.
Schellong
 
K.
Rodekamp
 
E.
Melchior
 
K.
Dudenhausen
 
J.W.
 
Epigenetic malprogramming of the insulin receptor promoter due to developmental overfeeding
J. Perinat. Med.
2010
, vol. 
38
 (pg. 
393
-
400
)
[PubMed]
16
Bouchard
 
L.
Thibault
 
S.
Guay
 
S.P.
Santure
 
M.
Monpetit
 
A.
St-Pierre
 
J.
Perron
 
P.
Brisson
 
D.
 
Leptin gene epigenetic adaptation to impaired glucose metabolism during pregnancy
Diabetes Care
2010
, vol. 
33
 (pg. 
2436
-
2441
)
[PubMed]
17
Lesseur
 
C.
Armstrong
 
D.A.
Paquette
 
A.G.
Li
 
Z.
Padbury
 
J.F.
Marsit
 
C.J.
 
Maternal obesity and gestational diabetes are associated with placental leptin DNA methylation
Am. J. Obstet. Gynecol.
2014
, vol. 
211
 (pg. 
654.e1
-
654.e9
)
18
Bouchard
 
L.
Hivert
 
M.F.
Guay
 
S.P.
St-Pierre
 
J.
Perron
 
P.
Brisson
 
D.
 
Placental adiponectin gene DNA methylation levels are associated with mothers’ blood glucose concentration
Diabetes
2012
, vol. 
61
 (pg. 
1272
-
1280
)
[PubMed]
19
Houde
 
A.A.
Guay
 
S.P.
Desgagne
 
V.
Hivert
 
M.F.
Baillargeon
 
J.P.
St-Pierre
 
J.
Perron
 
P.
Gaudet
 
D.
Brisson
 
D.
Bouchard
 
L.
 
Adaptations of placental and cord blood ABCA1 DNA methylation profile to maternal metabolic status
Epigenetics
2013
, vol. 
8
 (pg. 
1289
-
1302
)
[PubMed]
20
Zinkhan
 
E.K.
Lang
 
B.Y.
Yu
 
B.
Wang
 
Y.
Jiang
 
C.
Fitzhugh
 
M.
Dahl
 
M.
Campbell
 
M.S.
Fung
 
C.
Malleske
 
D.
, et al 
Maternal tobacco smoke increased visceral adiposity and serum corticosterone levels in adult male rat offspring
Pediatr. Res.
2014
, vol. 
76
 (pg. 
17
-
23
)
[PubMed]
21
Sarr
 
O.
Gondret
 
F.
Jamin
 
A.
Le Huerou-Luron
 
I.
Louveau
 
I.
 
A high-protein neonatal formula induces a temporary reduction of adiposity and changes later adipocyte physiology
Am. J. Physiol. Regul. Integr. Comp. Physiol.
2011
, vol. 
300
 (pg. 
R387
-
R397
)
[PubMed]
22
Nelissen
 
E.C.
van Montfoort
 
A.P.
Dumoulin
 
J.C.
Evers
 
J.L.
 
Epigenetics and the placenta
Hum. Reprod. Update
2011
, vol. 
17
 (pg. 
397
-
417
)
[PubMed]
23
Jansson
 
T.
Powell
 
T.L.
 
Role of the placenta in fetal programming: underlying mechanisms and potential interventional approaches
Clin. Sci.
2007
, vol. 
113
 (pg. 
1
-
13
)
[PubMed]
24
Maccani
 
M.A.
Marsit
 
C.J.
 
Epigenetics in the placenta
Am. J. Reprod. Immunol.
2009
, vol. 
62
 (pg. 
78
-
89
)
[PubMed]
25
Handschin
 
C.
Spiegelman
 
B.M.
 
Peroxisome proliferator-activated receptor γ coactivator 1 coactivators, energy homeostasis, and metabolism
Endocr. Rev.
2006
, vol. 
27
 (pg. 
728
-
735
)
[PubMed]
26
Newsholme
 
P.
Haber
 
E.P.
Hirabara
 
S.M.
Rebelato
 
E.L.
Procopio
 
J.
Morgan
 
D.
Oliveira-Emilio
 
H.C.
Carpinelli
 
A.R.
Curi
 
R.
 
Diabetes associated cell stress and dysfunction: role of mitochondrial and non-mitochondrial ROS production and activity
J. Physiol.
2007
, vol. 
583
 (pg. 
9
-
24
)
[PubMed]
27
Ling
 
C.
Del Guerra
 
S.
Lupi
 
R.
Ronn
 
T.
Granhall
 
C.
Luthman
 
H.
Masiello
 
P.
Marchetti
 
P.
Groop
 
L.
Del Prato
 
S.
 
Epigenetic regulation of PPARGC1A in human type 2 diabetic islets and effect on insulin secretion
Diabetologia
2008
, vol. 
51
 (pg. 
615
-
622
)
[PubMed]
28
Barres
 
R.
Osler
 
M.E.
Yan
 
J.
Rune
 
A.
Fritz
 
T.
Caidahl
 
K.
Krook
 
A.
Zierath
 
J.R.
 
Non-CpG methylation of the PGC-1α promoter through DNMT3B controls mitochondrial density
Cell Metab.
2009
, vol. 
10
 (pg. 
189
-
198
)
[PubMed]
29
Sookoian
 
S.
Rosselli
 
M.S.
Gemma
 
C.
Burgueno
 
A.L.
Fernandez Gianotti
 
T.
Castano
 
G.O.
Pirola
 
C.J.
 
Epigenetic regulation of insulin resistance in nonalcoholic fatty liver disease: impact of liver methylation of the peroxisome proliferator-activated receptor γ coactivator 1α promoter
Hepatology
2010
, vol. 
52
 (pg. 
1992
-
2000
)
[PubMed]
30
Brons
 
C.
Jacobsen
 
S.
Nilsson
 
E.
Ronn
 
T.
Jensen
 
C.B.
Storgaard
 
H.
Poulsen
 
P.
Groop
 
L.
Ling
 
C.
Astrup
 
A.
Vaag
 
A.
 
Deoxyribonucleic acid methylation and gene expression of PPARGC1A in human muscle is influenced by high-fat overfeeding in a birth-weight-dependent manner
J. Clin. Endocrinol. Metab.
2010
, vol. 
95
 (pg. 
3048
-
3056
)
[PubMed]
31
Ruchat
 
S.M.
Houde
 
A.A.
Voisin
 
G.
St-Pierre
 
J.
Perron
 
P.
Baillargeon
 
J.P.
Gaudet
 
D.
Hivert
 
M.F.
Brisson
 
D.
Bouchard
 
L.
 
Gestational diabetes mellitus epigenetically affects genes predominantly involved in metabolic diseases
Epigenetics
2013
, vol. 
8
 (pg. 
935
-
943
)
[PubMed]
32
Metzger
 
B.E.
Gabbe
 
S.G.
Persson
 
B.
Buchanan
 
T.A.
Catalano
 
P.A.
Damm
 
P.
Dyer
 
A.R.
Leiva
 
A.
Hod
 
M.
Kitzmiler
 
J.L.
, et al 
International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy
Diabetes Care
2010
, vol. 
33
 (pg. 
676
-
682
)
[PubMed]
33
American Diabetes Association
Standards of medical care in diabetes: 2014
Diabetes Care
2014
, vol. 
37
 
Suppl. 1
(pg. 
S14
-
S80
)
[PubMed]
34
Gillberg
 
L.
Jacobsen
 
S.
Ribel-Madsen
 
R.
Gjesing
 
A.P.
Boesgaard
 
T.W.
Ling
 
C.
Pedersen
 
O.
Hansen
 
T.
Vaag
 
A.
 
Does DNA methylation of PPARGC1A influence insulin action in first degree relatives of patients with Type 2 diabetes?
PLoS ONE
2013
, vol. 
8
 pg. 
e58384
 
[PubMed]
35
Alibegovic
 
A.C.
Sonne
 
M.P.
Hojbjerre
 
L.
Bork-Jensen
 
J.
Jacobsen
 
S.
Nilsson
 
E.
Faerch
 
K.
Hiscock
 
N.
Mortensen
 
B.
Friedrichsen
 
M.
, et al 
Insulin resistance induced by physical inactivity is associated with multiple transcriptional changes in skeletal muscle in young men
Am. J. Physiol. Endocrinol. Metab.
2010
, vol. 
299
 (pg. 
E752
-
E763
)
[PubMed]
36
Yessoufou
 
A.
Moutairou
 
K.
 
Maternal diabetes in pregnancy: early and long-term outcomes on the offspring and the concept of “metabolic memory”
Exp. Diabetes Res.
2011
, vol. 
2011
 pg. 
218598
 
[PubMed]
37
Petronis
 
A.
 
Epigenetics as a unifying principle in the aetiology of complex traits and diseases
Nature
2010
, vol. 
465
 (pg. 
721
-
727
)
[PubMed]
38
Wong
 
C.C.
Caspi
 
A.
Williams
 
B.
Craig
 
I.W.
Houts
 
R.
Ambler
 
A.
Moffitt
 
T.E.
Mill
 
J.
 
A longitudinal study of epigenetic variation in twins
Epigenetics
2010
, vol. 
5
 (pg. 
516
-
526
)
[PubMed]
39
Lehnen
 
H.
Zechner
 
U.
Haaf
 
T.
 
Epigenetics of gestational diabetes mellitus and offspring health: the time for action is in early stages of life
Mol. Hum. Reprod.
2013
, vol. 
19
 (pg. 
415
-
422
)
[PubMed]
40
Relton
 
C.L.
Groom
 
A.
St Pourcain
 
B.
Sayers
 
A.E.
Swan
 
D.C.
Embleton
 
N.D.
Pearce
 
M.S.
Ring
 
S.M.
Northstone
 
K.
Tobias
 
J.H.
, et al 
DNA methylation patterns in cord blood DNA and body size in childhood
PloS ONE
2012
, vol. 
7
 pg. 
e31821
 
[PubMed]
41
Godfrey
 
K.M.
Sheppard
 
A.
Gluckman
 
P.D.
Lillycrop
 
K.A.
Burdge
 
G.C.
McLean
 
C.
Rodford
 
J.
Slater-Jefferies
 
J.L.
Garratt
 
E.
Crozier
 
S.R.
, et al 
Epigenetic gene promoter methylation at birth is associated with child's later adiposity
Diabetes
2011
, vol. 
60
 (pg. 
1528
-
1534
)
[PubMed]
42
Franks
 
P.W.
Looker
 
H.C.
Kobes
 
S.
Touger
 
L.
Tataranni
 
P.A.
Hanson
 
R.L.
Knowler
 
W.C.
 
Gestational glucose tolerance and risk of type 2 diabetes in young Pima Indian offspring
Diabetes
2006
, vol. 
55
 (pg. 
460
-
465
)
[PubMed]
43
Herzog
 
E.
Galvez
 
J.
Roks
 
A.
Stolk
 
L.
Verbiest
 
M.
Eilers
 
P.
Cornelissen
 
J.
Steegers
 
E.
Steegers-Theunissen
 
R.
 
Tissue-specific DNA methylation profiles in newborns
Clin. Epigenet.
2013
, vol. 
5
 pg. 
8
 
44
Armstrong
 
D.A.
Lesseur
 
C.
Conradt
 
E.
Lester
 
B.M.
Marsit
 
C.J.
 
Global and gene-specific DNA methylation across multiple tissues in early infancy: implications for children's health research
FASEB J.
2014
, vol. 
28
 (pg. 
2088
-
2097
)
[PubMed]
45
Ollikainen
 
M.
Smith
 
K.R.
Joo
 
E.J.
Ng
 
H.K.
Andronikos
 
R.
Novakovic
 
B.
Abdul Aziz
 
N.K.
Carlin
 
J.B.
Morley
 
R.
Saffery
 
R.
Craig
 
J.M.
 
DNA methylation analysis of multiple tissues from newborn twins reveals both genetic and intrauterine components to variation in the human neonatal epigenome
Hum. Mol. Genet.
2010
, vol. 
19
 (pg. 
4176
-
4188
)
[PubMed]
46
Volkmar
 
M.
Dedeurwaerder
 
S.
Cunha
 
D.A.
Ndlovu
 
M.N.
Defrance
 
M.
Deplus
 
R.
Calonne
 
E.
Volkmar
 
U.
Igoillo-Esteve
 
M.
Naamane
 
N.
, et al 
DNA methylation profiling identifies epigenetic dysregulation in pancreatic islets from type 2 diabetic patients
EMBO J.
2012
, vol. 
31
 (pg. 
1405
-
1426
)
[PubMed]
47
Monte
 
S.
Valenti
 
O.
Giorgio
 
E.
Renda
 
E.
Hyseni
 
E.
Faraci
 
M.
De Domenico
 
R.
Di Prima
 
F.A.
 
Maternal weight gain during pregnancy and neonatal birth weight: a review of the literature
J. Prenat. Med.
2011
, vol. 
5
 (pg. 
27
-
30
)
[PubMed]
48
Gemma
 
C.
Sookoian
 
S.
Alvarinas
 
J.
Garcia
 
S.I.
Quintana
 
L.
Kanevsky
 
D.
Gonzalez
 
C.D.
Pirola
 
C.J.
 
Maternal pregestational BMI is associated with methylation of the PPARGC1A promoter in newborns
Obesity
2009
, vol. 
17
 (pg. 
1032
-
1039
)
[PubMed]
49
Fernandez-Marcos
 
P.J.
Auwerx
 
J.
 
Regulation of PGC-1α, a nodal regulator of mitochondrial biogenesis
Am. J. Clin. Nutr.
2011
, vol. 
93
 (pg. 
884S
-
890S
)
[PubMed]
50
Guay
 
S.P.
Brisson
 
D.
Munger
 
J.
Lamarche
 
B.
Gaudet
 
D.
Bouchard
 
L.
 
ABCA1 gene promoter DNA methylation is associated with HDL particle profile and coronary artery disease in familial hypercholesterolemia
Epigenetics
2012
, vol. 
7
 (pg. 
464
-
472
)
[PubMed]
51
Guttmacher
 
A.E.
Maddox
 
Y.T.
Spong
 
C.Y.
 
The Human Placenta Project: placental structure, development, and function in real time
Placenta
2014
, vol. 
35
 (pg. 
303
-
304
)
[PubMed]
52
Barker
 
D.J.
Thornburg
 
K.L.
 
Placental programming of chronic diseases, cancer and lifespan: a review
Placenta
2013
, vol. 
34
 (pg. 
841
-
845
)
[PubMed]
53
Kim
 
E.S.
Isoda
 
F.
Kurland
 
I.
Mobbs
 
C.V.
 
Glucose-induced metabolic memory in Schwann cells: prevention by PPAR agonists
Endocrinology
2013
, vol. 
154
 (pg. 
3054
-
3066
)
[PubMed]
54
El Hajj
 
N.
Pliushch
 
G.
Schneider
 
E.
Dittrich
 
M.
Muller
 
T.
Korenkov
 
M.
Aretz
 
M.
Zechner
 
U.
Lehnen
 
H.
Haaf
 
T.
 
Metabolic programming of MEST DNA methylation by intrauterine exposure to gestational diabetes mellitus
Diabetes
2013
, vol. 
62
 (pg. 
1320
-
1328
)
[PubMed]
55
Xie
 
X.
Lin
 
T.
Zhang
 
M.
Liao
 
L.
Yuan
 
G.
Gao
 
H.
Ning
 
Q.
Luo
 
X.
 
IUGR with infantile overnutrition programs an insulin-resistant phenotype through DNA methylation of peroxisome proliferator-activated receptor-γ coactivator-1 α in rats
Pediatr. Res.
2015
, vol. 
77
 (pg. 
625
-
632
)
[PubMed]

Author notes

1

These authors are co-first authors.

Supplementary data