Abstract
Aim:
To examine the associations between fetal growth restriction (FGR) and DNA methylation of six growth-related genes in human placenta.
Materials & methods:
A total of 181 mother-newborn pairs (80 FGR cases and 101 controls) were enrolled in this case–control study. Placental DNA methylation was measured by bisulfite pyrosequencing.
Results:
DNA methylation levels of IGF2 and AHRR were positively associated with newborn birth weight and Quetelet's index, while DNA methylation levels of HSD11B2 and WNT2 were negatively associated with those fetal growth indicators. In addition, significantly elevated odds of FGR birth were associated with increasing DNA methylation of HSD11B2 and WNT2, and decreasing DNA methylation of IGF2.
Conclusion:
Our findings demonstrated that placental DNA methylation levels of IGF2, AHRR, HSD11B2 and WNT2 were associated with measures of fetal growth.
Keywords: : birth weight, case–control study, DNA methylation, fetal growth restriction, placenta
Background
Infants who fail to achieve their own growth potential are considered to have fetal growth restriction (FGR) [1]. Such infants have an increased risk for perinatal morbidity and mortality, as well as impaired postnatal growth, neurodevelopment and cognitive abilities [2]. FGR infants also have increased risk for certain diseases later in life, particularly those comprised in the metabolic syndrome [3–5]. Because of these immediate and long-term outcomes associated with FGR, understanding the molecular mechanisms involved in the loss of fetal growth potential is critically important.
Throughout in utero development, the placenta, through the production of various enzymes and hormones, plays an important role in controlling fetal growth and development [6]. Recent work has made it increasingly clear that proper epigenetic regulation of both imprinted and nonimprinted genes in placenta is important in placental development and fetal growth [7]. Investigating epigenetic alterations in placenta may add to our understanding of the molecular mechanisms behind reduced fetal growth in utero and increased disease susceptibility in later life.
Imprinted genes are highly expressed in placenta and have been shown to be indispensable for proper placental morphology and function [8,9]. IGF2, which is imprinted in tandem with H19, is the most intensively studied imprinted gene. Expression of IGF2/H19 is regulated by differential and coordinated methylation of two imprinting control regions (ICRs; ICR1 and ICR2) in chromosome 11 imprinting cluster [10]. Recently, investigators have reported an association between placental DNA methylation of these ICRs and fetal growth indices (e.g., birth weight, head circumference and thorax circumferences) [11–13].
In addition to imprinted genes, some nonimprinted genes in placenta are also involved in fetal growth, and DNA methylation of their promoters and exons has been shown to be associated with fetal growth indices. One gene to such effect is HSD11B2. HSD11B2 is a key gene involved in glucocorticoid metabolism [14]. High methylation of this gene in human placenta has been associated with reduced fetal growth [15,16].
Although a few studies have begun to examine the associations of placental DNA methylation with fetal growth, most of these studies lacked adjustment for potentially important confounders and thus few of them provided consistent results [11–13,17–18]. In addition, no study to date has examined the role of placental DNA methylation in the development of FGR. Thus, we conducted this case–control study to investigate how placental DNA methylation of six growth-related genes associates with measures of fetal growth, and further evaluate the risk of FGR birth in relation to placental DNA methylation.
Material & methods
Study population
This study is an extension of our previous case–control study, which was designed to investigate the association between prenatal environmental endocrine disruptors exposure and FGR [19]. Low birth weight (LBW) was defined as fetal birth weight <2500 g and gestational age (GA) ≥37 weeks [20]. Intrauterine growth restriction (IUGR) was defined as estimated fetal weight below the 10th centile for GA, based on ultrasound measurements of fetal biparietal diameter, head circumference, abdominal circumference and femur length [21–23]. Since in our study there is no significant difference in birth weight, birth length and Quetelet's index between IUGR infants and LBW infants (see Supplementary Table 1), infants with either IUGR or LBW were diagnosed as FGR cases, and equal numbers of healthy newborns were included as controls. During December 2011 and November 2013, 220 mother–newborn pairs, including 110 FGR cases and 110 healthy controls and their mothers, were enrolled in our case–control study. The mothers answered a detailed questionnaire concerning maternal weight, and height before pregnancy, maternal age, maternal smoking, drinking, dietary habits, etc., after delivery within 2 days. All subjects signed written informed consents approved by Fudan University's Institutional Review Board.
Placental samples were collected immediately after delivery. For each subject, eight biopsies of apparently normal tissue were collected (two from each of the four quadrants). All samples were taken from the maternal side of the placenta (the genes we studied mainly express on this side), 2 cm from the umbilical cord insertion site, free of maternal decidua. After washing three times in sterile phosphate-buffered saline, samples were immediately frozen in liquid nitrogen and stored at -80°C until analysis.
One hundred and eighty-one subjects donated placental samples, thus we enrolled those 181 subjects (including 80 FGR cases and 101 healthy controls) in this present study.
Quantification of placental DNA methylation
A pooled sample (120 mg) for each placenta that comprised four biopsies (two from the upper left-hand quadrant and two from the lower right-hand quadrant) was used for DNA extraction. Genomic DNA was extracted using the QIAmp DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Purified DNA was quantified using a ND1000 spectrophotometer (Nanodrop, DE, USA) and about 1 μg DNA was bisulfite modified using the EZ Methylation Gold-Kit (Zymo Research, CA, USA). Final elution was performed with 30 μl M-Elution Buffer.
Placental DNA methylation of six growth-related genes (HSD11B2, NR3C1, WNT2, IGF2, IL6 and AHRR) was quantitated using bisulfite-PCR and pyrosequencing. HSD11B2 and NR3C1 are two genes involved in glucocorticoid metabolism. WNT2, IGF2 and IL6 are crucial for normal development of the placenta. AHRR serves to inhibit aryl hydrocarbon receptor transcription, which is involved in mediating xenobiotic metabolism. All those studied genes are involved in metabolic and endocrine function of the placenta and their expression levels have been associated with fetal growth restriction [24–26]. The PCR primers and position of each CpG site are shown in Supplementary Table 2. Analysts were blind to all information concerning our subjects. To prevent batch effects from bisulfite treatments interfering with the analysis, samples were randomized across batches. Methylation level of each CpG dinucleotide was expressed as methylated cytosines over the sum of methylated and unmethylated cytosines. Each sample was tested in triplicate and the average was used for statistical analysis.
Some controls were included in every pyrosequencing run. To verify bisulfite conversion efficiency, a C outside a CG site was used in every assay as built-in control. After bisulfite treatment, the conversion of this C into T is expected to be 100%. We insert a C/T single-nucleotide polymorphism into the sequence to be analyzed that will result in a 100% T if conversion is efficient. To ensure that pyrosequencing sequenced the correct pattern, two wells were filled with oligonucleotide with a known sequence. Moreover, a human unmethylated (0%) standard and fully methylated (100%) standard were used as sample controls.
Statistical analysis
Statistical analyses were performed by the SPSS 16.0 statistical package (SPSS, Inc., IL, USA). Normally distributed continuous variables were expressed as mean ± standard deviation. Parametric t-test and χ2 test were used to compare demographic characteristics in the FGR mother-newborn pairs and controls. Potential differences in placental DNA methylation levels between FGR cases and controls were evaluated through Mann–Whitney U test.
Multiple linear regression models were used to model the associations of placental DNA methylation with fetal growth indices (birth weight, birth length and Quetelet index). Variables considered in regression models included infant gender, GA, birth type, maternal prepregnancy BMI, prenatal vitamin use, maternal age, maternal environmental tobacco smoke during pregnancy, maternal education and monthly income. A backward stepwise method (critical level of p < 0.20 for variable removal) was used to determine which variables should be included in final models.
Multivariate logistic regression was used to estimate the odds of FGR birth in association with placental DNA methylation. Variables considered in logistic regression models included GA, infant gender, maternal age, maternal prepregnancy BMI, maternal physical activity, environmental tobacco smoke during pregnancy, maternal education and monthly income. All potential confounders were added to models in a forward stepwise procedure and were finally included if they altered the association between placental DNA methylation and FGR birth by >10%. p < 0.05 was considered significant, and all statistical tests were two-sided.
Results
Population characteristics
Of the 220 subjects enrolled in our case–control study, 181 subjects (80 FGR cases, 101 controls) donated placental samples. There was no difference in characteristics between subjects who provided samples compared with those who did not (data not shown). The maternal and pregnancy characteristics of FGR and control subjects are listed and compared as shown in Table 1. Cases and controls differed with regard to birth weight, birth length, Quetelet index and GA. Cases and controls did not differ by maternal age, prepregnant BMI, infant gender, prenatal vitamin use, birth type, maternal education background and monthly income.
Table 1. . General characteristics of subjects by birth outcome status.
| Characteristics | Total (n = 181) | FGR (n = 80) | Control (n = 101) | p-value† |
|---|---|---|---|---|
| Birth weight (kg)‡ |
2.85 ± 0.61 |
2.29 ± 0.37 |
3.29 ± 0.34 |
<0.001** |
| Birth length (cm)‡ |
48.80 ± 2.31 |
46.97 ± 2.35 |
50.23 ± 0.74 |
<0.001** |
| Quetelet index (kg/m2)‡,§ |
11.84 ± 1.86 |
10.34 ± 1.33 |
13.02 ± 1.26 |
<0.001** |
| Gestational age (year)‡ |
38.07 ± 2.51 |
36.60 ± 2.93 |
39.25 ± 1.18 |
<0.001** |
| Maternal age (year)‡ |
26.73 ± 4.32 |
26.26 ± 4.82 |
27.11 ± 3.86 |
0.196 |
| Prepregnant BMI (kg/m2)‡ |
19.66 ± 2.18 |
19.65 ± 2.50 |
19.67 ± 1.92 |
0.965 |
| Gender: | 0.063 | |||
| – Male | 93 | 38 | 55 | |
| – Female |
88 |
42 |
46 |
|
| Prenatal vitamin use: | 0.302 | |||
| – Yes | 161 | 69 | 92 | |
| – No |
20 |
11 |
9 |
|
| Birth type: | 0.051 | |||
| – Vaginal delivery | 135 | 54 | 81 | |
| – Cesarean section |
46 |
26 |
20 |
|
| Maternal education: | 0.261 | |||
| – Middle school | 49 | 26 | 23 | |
| – High school | 49 | 18 | 31 | |
| – College |
83 |
36 |
47 |
|
| Monthly income (RMB): | 0.062 | |||
| – <5000 | 81 | 42 | 39 | |
| – ≥5000 | 100 | 38 | 62 |
†p-value calculated using the χ2-test and t-test.
‡Values are shown as mean ± standard deviation.
§Quetelet index was calculated as weight (in kg) divided by the square of height (in m).
**p < 0.01.
FGR: Fetal growth restriction; RMB: Ren Min Bi.
Placental DNA methylation levels
Four sites of HSD11B2, three sites of WNT2 and AHRR, two sites of IL6 and IGF2, and one site of NR3C1 were measured. In each assay, correlation between DNA methylation at the individual CpG site was examined. All correlations were above 0.7 in the assay of HSD11B2 and WNT2. Thus, the average methylation across four sites of the HSD11B2 locus and three sites of the WNT2 promoter was reported. In contrast, the one site evaluated in NR3C1, two sites in IGF2, two sites in IL6 and three sites in AHRR were each reported separately.
Mann–Whitney U test was used to compare the difference in DNA methylation between FGR cases and controls. As shown in Table 2, mean methylation across four CpG sites of HSD11B2 was significantly higher in FGR group than in control group (13.92 vs 13.25; p = 0.003). By contrast, DNA methylation at position 2 of IGF2 and position 1 of AHRR was significantly lower in FGR group than in control group (44.03 vs 46.35; p = 0.005; 55.84 vs 57.95; p = 0.021).
Table 2. . Distribution of placental DNA methylation levels (%).
| Gene name |
Median (25th–75th percentile) |
p-value† | ||
|---|---|---|---|---|
| Total (n = 181) | Case (n = 80) | Control (n = 101) | ||
|
HSD11B2 |
13.58 (11.77–14.77) |
13.92 (12.25–15.25) |
13.25 (11.30–14.50) |
0.003** |
|
NR3C1 |
63.96 (59.79–68.63) |
64.11 (58.90–68.34) |
63.79 (60.34–68.71) |
0.620 |
|
WNT2 |
12.98 (8.49–43.01) |
12.53 (8.63–43.45) |
13.46 (8.42–43.17) |
0.678 |
| IGF2: | ||||
| – Pos.1 | 58.55 (55.72–61.49) | 57.92 (54.65–61.27) | 58.66 (56.74–61.87) | 0.128 |
| – Pos.2 |
45.77 (40.18–48.60) |
44.03 (35.96–47.69) |
46.35 (42.98–49.25) |
0.005** |
| IL6: | ||||
| – Pos.1 | 6.13 (4.71–8.27) | 6.23 (4.82–8.28) | 6.10 (4.54–8.28) | 0.796 |
| – Pos.2 |
14.92 (11.34–18.51) |
14.84 (12.04–18.22) |
15.00 (10.89–18.53) |
0.948 |
| AHRR: | ||||
| – Pos.1 | 57.20 (52.49–61.99) | 55.84 (51.23–60.07) | 57.95 (53.95–63.48) | 0.021* |
| – Pos.2 | 73.78 (67.97–80.39) | 72.06 (66.60–77.93) | 75.02 (69.88–80.91) | 0.075 |
| – Pos.3 | 41.23 (36.81–46.39) | 41.01 (36.71–46.00) | 41.44 (37.04–46.72) | 0.673 |
†p value calculated using the Mann–Whitney U test.
*p < 0.05; **p < 0.01.
Pos.1: The first studied CpG site; Pos.2: The second studied CpG site; Pos.3: The third studied CpG site.
Placental DNA methylation levels & fetal growth indicators
Multiple linear regression models were used to model the associations of placental DNA methylation with fetal growth indicators. For WNT2, histogram of methylation data showed a bimodal distribution (Supplementary Figure 1). Thus, when we did regression analysis of WNT2, all subjects were divided into high and low methylation group.
Table 3 shows the relationship between placental DNA methylation and newborn birth weight. In adjusted model, DNA methylation of HSD11B2 and WNT2 (high methylation group) had significant negative association with newborn birth weight. Every 1% unit increase in DNA methylation was associated with 27 and 19 g reduction in newborn birth weight, respectively (β = -0.027; p = 0.033; β = -0.019; p = 0.034). In contrast, DNA methylation at position 1 of AHRR and position 2 of IGF2 had significant positive association with newborn birth weight. Every 1% unit increase in DNA methylation was associated with 8 and 10 g increase in newborn birth weight, respectively (β = 0.008; p = 0.026; β = 0.010; p = 0.027).
Table 3. . Crude and adjusted association between DNA methylation and birth weight.
| Methylation |
Birth weight |
|||
|---|---|---|---|---|
|
Crude model |
Adjusted model† |
|||
| β (95% CI) | p-value | β (95% CI) | p-value | |
|
HSD11B2 |
-0.048 (-0.083 to -0.012) |
0.009** |
-0.027 (-0.053 to -0.002) |
0.033* |
|
NR3C1 |
0.008 (-0.005–0.021) |
0.229 |
-0.003 (-0.012–0.006) |
0.557 |
| WNT2: | ||||
| – High | -0.018 (-0.040–0.005) | 0.121 | -0.019 (-0.037 to -0.002) | 0.034* |
| – Low |
0.011 (-0.014–0.037) |
0.387 |
0.010 (-0.005–0.025) |
0.198 |
| IGF2: | ||||
| – Pos.1 | 0.017 (0.001–0.034) | 0.043* | 0.009 (-0.002 to 0.020) | 0.125 |
| – Pos.2 |
0.010 (-0.001–0.020) |
0.072 |
0.008 (0.001–0.015) |
0.026* |
| IL6: | ||||
| – Pos.1 | -0.003 (-0.032–0.026) | 0.819 | -0.008 (-0.027–0.011) | 0.386 |
| – Pos.2 |
0.006 (-0.010–0.021) |
0.468 |
0.002 (-0.008–0.013) |
0.683 |
| AHRR: | ||||
| – Pos.1 | 0.020 (0.006–0.033) | 0.005** | 0.010 (0.001–0.019) | 0.027* |
| – Pos.2 | 0.010 (-0.001–0.022) | 0.076 | 0.006 (-0.001–0.013) | 0.086 |
| – Pos.3 | 0.001 (-0.012–0.012) | 0.956 | 0.004 (-0.003–0.012) | 0.250 |
Methylation level of each gene was modeled as the independent variable associated with birth weight as the dependent variable.
†Adjusted for gestational age, infant gender, birth type, prenatal vitamin use and environmental tobacco smoke during pregnancy.
*p < 0.05; **p < 0.01.
High: High WNT2 methylation group; Low: Low WNT2 methylation group; Pos.1: The first studied CpG site; Pos.2: The second studied CpG site; Pos.3: The third studied CpG site.
Table 4 shows the relationship between placental DNA methylation and newborn birth length. In both crude and adjusted model, DNA methylation of HSD11B2 was inversely associated with newborn birth length. Every 1% unit increase in HSD11B2 methylation was associated with a decrease of 0.118 cm in newborn birth length (β = -0.118; p = 0.023).
Table 4. . Crude and adjusted association between placental DNA methylation and birth length.
| Methylation |
Birth length |
|||
|---|---|---|---|---|
|
Crude model |
Adjusted model† |
|||
| β (95% CI) | p-value | β (95% CI) | p-value | |
|
HSD11B2 |
-0.162 (-0.302 to -0.022) |
0.023* |
-0.118 (-0.219 to -0.016) |
0.023* |
|
NR3C1 |
0.049 (-0.002–0.099) |
0.059 |
0.008 (-0.029–0.045) |
0.682 |
| WNT2: | ||||
| – High | -0.015 (-0.087–0.057) | 0.679 | -0.014 (-0.082–0.053) | 0.671 |
| – Low |
-0.002 (-0.104–0.101) |
0.976 |
-0.005 (-0.070–0.061) |
0.886 |
| IGF2: | ||||
| – Pos.1 | 0.035 (-0.032–0.102) | 0.301 | 0.008 (-0.039–0.056) | 0.730 |
| – Pos.2 |
0.028 (-0.013–0.068) |
0.183 |
0.018 (-0.011–0.046) |
0.226 |
| IL6: | ||||
| – Pos.1 | -0.013 (-0.128–0.102) | 0.824 | -0.033 (-0.133–0.047) | 0.411 |
| – Pos.2 |
0.032 (-0.030–0.095) |
0.309 |
0.008 (-0.035–0.052) |
0.702 |
| AHRR: | ||||
| – Pos.1 | 0.039 (-0.017–0.095) | 0.175 | 0.001 (-0.039–0.039) | 0.933 |
| – Pos.2 | 0.021(-0.026–0.068) | 0.380 | 0.005 (-0.027–0.037) | 0.772 |
| – Pos.3 | -0.023 (-0.071–0.024) | 0.332 | -0.004 (-0.036–0.029) | 0.827 |
Methylation level of each gene was modeled as the independent variable associated with birth length as the dependent variable.
†Adjusted for gestational age, birth type and environmental tobacco smoke during pregnancy
*p < 0.05.
High: High WNT2 methylation group; Low: Low WNT2 methylation group; Pos.1: The first studied CpG site; Pos.2: The second studied CpG site; Pos.3: The third studied CpG site.
Table 5 presents the regression results for placental DNA methylation and Quetelet index (calculated as weight [in kg] divided by the square of height [in m]). In adjusted model, evaluated DNA methylation was associated with smaller Quetelet index within high WNT2 methylation group (β = -0.093; p = 0.002), while DNA methylation at position 1 of AHRR was positively associated with the Quetelet index (β = 0.036; p = 0.024).
Table 5. . Crude and adjusted association between DNA methylation and Quetelet index.
| Methylation |
Quetelet index |
|||
|---|---|---|---|---|
|
Crude model |
Adjusted model† |
|||
| β (95% CI) | p-value | β (95% CI) | p-value | |
|
HSD11B2 |
-0.111 (-0.224–0.003) |
0.056 |
-0.060 (-0.148 to -0.029) |
0.184 |
|
NR3C1 |
0.011 (-0.029–0.052) |
0.581 |
-0.018 (-0.050–0.014) |
0.463 |
| WNT2: | ||||
| – High | -0.070 (-0.144–0.003) | 0.061 | -0.093 (-0.150 to -0.036) | 0.002** |
| – Low |
0.048 (-0.028–0.124) |
0.211 |
0.045 (-0.010–0.100) |
0.107 |
| IGF2: | ||||
| – Pos.1 | 0.057 (0.006–0.109) | 0.030* | 0.038 (-0.002–0.077) | 0.064 |
| – Pos.2 |
0.026 (-0.006–0.057) |
0.113 |
0.018 (-0.007–0.042) |
0.157 |
| IL6: | ||||
| – Pos.1 | 0.007 (-0.083–0.097) | 0.878 | -0.008 (-0.076–0.061) | 0.828 |
| – Pos.2 |
0.024 (-0.025–0.072) |
0.335 |
0.007 (-0.030–0.044) |
0.720 |
| AHRR: | ||||
| – Pos.1 | 0.067 (0.025–0.109) | 0.002** | 0.036 (0.005–0.067) | 0.024* |
| – Pos.2 | 0.033 (-0.003–0.069) | 0.070 | 0.019 (-0.007–0.045) | 0.152 |
| – Pos.3 | 0.006 (-0.031–0.042) | 0.755 | 0.015 (-0.012–0.041) | 0.267 |
Methylation level of each gene was modeled as the independent variable associated with Quetelet index as the dependent variable.
†Adjusted for gestational age, infant gender, birth type, prenatal vitamin use and environmental tobacco smoke during pregnancy.
*p < 0.05; **p < 0.01.
High: High WNT2 methylation group; Low: Low WNT2 methylation group; Pos.1: The first studied CpG site; Pos.2: The second studied CpG site; Pos.3: The third studied CpG site.
Although FGR newborns seemed to have shorter GA than control newborns, no significant association was found between placental DNA methylation and GA (data not shown).
Placental DNA methylation levels & relative odds of fetal growth restriction birth
Table 6 presents crude and adjusted odds ratios (OR) and 95% CI for the risk of FGR birth in relation to placental DNA methylation. In adjusted model, higher DNA methylation of HSD11B2 and WNT2 was significantly associated with higher relative odds of FGR birth. Every 1% unit increase in DNA methylation of HSD11B2 and WNT2 (high methylation group) was associated with 22.4 and 13.6% higher odds of FGR birth, respectively (OR: 1.224; p = 0.011; OR: 1.136; p = 0.037). In contrast, DNA higher methylation of IGF2 was associated with significantly lower odds of FGR birth (position 1, OR: 0.932; p = 0.041; position 2; OR: 0.947; p = 0.012).
Table 6. . Crude and adjusted odds ratios for placental DNA methylation and fetal growth restriction birth.
| Methylation |
Crude model |
Adjusted model† |
||
|---|---|---|---|---|
| OR (95% CI) | p -value | OR (95% CI) | p-value | |
| HSD11B2 | 1.208 (1.060–1.376) | 0.005** | 1.224 (1.047–1.430) | 0.011* |
|
NR3C1 |
0.989 (0.947–1.033) |
0.748 |
1.022 (0.967–1.079) |
0.444 |
| WNT2: | ||||
| – High | 1.060 (0.966–1.163) | 0.216 | 1.136 (1.007–1.281) | 0.037* |
| – Low |
0.979 (0.906–1.058) |
0.600 |
0.971 (0.872–1.081) |
0.971 |
| IGF2: | ||||
| – Pos.1 | 0.932 (0.879–0.988) | 0.018* | 0.932 (0.863–0.990) | 0.041* |
| – Pos.2 |
0.951 (0.917–0.987) |
0.008** |
0.947 (0.908–0.988) |
0.012* |
| IL6: | ||||
| – Pos.1 | 1.013 (0.922–1.113) | 0.794 | 1.020 (0.906–1.148) | 0.744 |
| – Pos.2 |
0.998 (0.949–1.050) |
0.947 |
0.995 (0.935–1.148) |
0.878 |
| AHRR: | ||||
| – Pos.1 | 0.947 (0.904–0.993) | 0.023* | 0.949 (0.895–1.006) | 0.079 |
| – Pos.2 | 0.967 (0.931–1.004) | 0.077 | 0.955 (0.910–1.002) | 0.060 |
| – Pos.3 | 0.992 (0.955–1.030) | 0.671 | 0.956 (0.909–1.006) | 0.083 |
†Adjusted for gestational age, prenatal vitamin use and environmental tobacco smoke during pregnancy.
*p < 0.05; **p < 0.01.
High: High WNT2 methylation group; Low: Low WNT2 methylation group; OR: Odds ratio; Pos.1: The first studied CpG site; Pos.2: The second studied CpG site; Pos.3: The third studied CpG site.
Discussion
Alterations to placental DNA methylation may result in altered expression of genes involved in metabolic/endocrine functions and may also affect the placenta's ability to transport water, gas, nutrients and waste products crucial for the proper growth and survival of the fetus [27]. In this analysis of an Asian population, we tested for the association of being born FGR with DNA methylation of IGF2, WNT2, HSD11B2, NR3C1, AHRR and IL6.
We observed significant differences in DNA methylation of IGF2 between FGR and control placental samples. Elevated DNA methylation was associated with higher newborn birth weight. The CpG site associated with birth weight is located in differentially methylated regions of IGF2, which may play a role in regulating gene expression. This observation is in keeping with the study by St-Pierre et al. In their study, they found DNA methylation at the same CpG site was positively associated with birth weight and thorax circumference without adjusting any confounders [11].
During development, the WNT pathway is involved in cell proliferation, fate, migration, polarity and cell death. Each of these mechanisms is crucial for normal development of the placenta to enable fetal growth [28]. Ferreira et al. demonstrated that high WNT2 promoter methylation is significantly associated with reduced WNT2 expression in placenta and low birth weight percentile in the neonate [17]. This is the only study to report an association of WNT2 methylation with fetal growth indices. In our study, we found the distribution of WNT2 methylation was bimodal, and WNT2 methylation was inversely associated with birth weight and Quetelet index within high methylation group. The bimodal distribution of WNT2 methylation might be driven by the SNPs within or near the promoter region [17].
Prenatal exposure to excess maternal glucocorticoids impaired fetal growth. HSD11B2, a key gene involved in glucocorticoid metabolism, interconverts hormonally active glucocorticoid to inactive cortisone [29,30]. In a cross-sectional study, Marsit et al. found that DNA methylation of HSD11B2 was inversely associated with HSD11B2 expression in placenta and with fetal growth indices in the neonate [15]. In line with this cross-sectional study and our previous small case–control study [16] we found placental DNA methylation was inversely associated with newborn birth weight and birth length. In addition to HSD11B2, NR3C1, the glucocorticoid receptor gene, is also a known mediator of glucocorticoid signaling [31]. Filiberto et al. demonstrated that DNA methylation of NR3C1 promoter region was inversely associated with large-for-gestational-age status in newborns [18], while in this present study, we did not find any association between NR3C1 promoter methylation and measures of fetal growth. This might be due to the different assays, which focused on different CpG sites within NR3C1 promoter.
Other genes highlighted in our study are AHRR and IL6. The relationship between DNA methylation of those two genes and fetal growth has not been addressed. AHRR is a known cancer susceptibility gene and a key regulator of the catabolism of xenobiotics [32,33]. Lower methylation of AHRR has been associated with higher levels of self-report or cotinine-based assessment of maternal smoking [34]. In our study, lower AHRR methylation was found to be associated with reduced fetal growth. Given maternal smoking is known to affect fetal growth, placental AHRR methylation might be one mechanism explaining the effect of maternal smoking on fetal growth. IL6 is a kind of proinflammatory cytokine, which is also thought to play a role in placental development and fetal growth. One study reported that FGR subjects had significantly more IL6 mRNA and protein in the placenta [26]. However, we did not find any difference in IL6 DNA methylation.
Our study is one of the first to show an association between placental DNA methylation of six growth-related genes and risk of FGR birth. A significant strength of this study is that we were able to adjust for potential confounders. Moreover, we used the population-based ascertainment of placental sample, which is a target tissue directly responsible for fetal growth. The limitations of this study include the relatively small sample size, which may restrict our statistical power. Additionally, owing to the lack of mRNA collection, we could not test the association between placental DNA methylation and gene expression.
Conclusion
Our findings demonstrated that placental DNA methylation levels of IGF2, AHRR, HSD11B2 and WNT2 were associated with measures of fetal growth. Further investigation into the role of DNA methylation in the development of FGR is warranted.
Future perspective
Placental DNA methylation may serve as an early epigenetic marker of maternal exposure and may be involved in the pathway leading to adverse birth outcomes and increased adult disease susceptibility. Elucidating the role of DNA methylation in the developmental origins of health and disease is the challenge in the next 5–10 years, with the goal of providing early interventions and prevention, and new treatment opportunities.
Executive summary.
Difference in DNA methylation between fetal growth restriction cases & controls
Mean methylation across four CpG sites of HSD11B2 was significantly higher in fetal growth restriction (FGR) group than in control group.
DNA methylation at position 2 of IGF2 and position 1 of AHRR was significantly lower in FGR group than in control group.
Placental DNA methylation levels & fetal growth indices
DNA methylation levels of IGF2 and AHRR were positively associated with newborn birth weight and Quetelet's index.
DNA methylation levels of HSD11B2 and WNT2 were negatively associated with newborn birth weight and Quetelet's index.
Placental DNA methylation levels & relative odds of fetal growth restriction
Significantly elevated odds of FGR birth were associated with increasing DNA methylation of HSD11B2 and WNT2, and decreasing DNA methylation of IGF2.
Supplementary Material
Footnotes
Financial & competing interests disclosure
The present work was in part supported by funding from the National Natural Science Foundation of China (grant 81072263 to Y Zhang) and from the NIH (R01ES020268, R01ES021357, R01NR013945, P30ES00002). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.
Ethical conduct of research
The authors state that they have obtained appropriate institutional review board approval or have followed the principles outlined in the Declaration of Helsinki for all human experimental investigations. In addition, for investigations involving human subjects, informed consent has been obtained from the participants involved.
References
Papers of special note have been highlighted as: • of interest
- 1.Longo S, Bollani L, Decembrino L, Di Comite A, Angelini M, Stronati M. Short-term and long-term sequelae in intrauterine growth retardation (IUGR) J. Matern. Fetal. Neonatal. Med. 2013;26(3):222–225. doi: 10.3109/14767058.2012.715006. [DOI] [PubMed] [Google Scholar]
- 2.Brodsky D, Christou H. Current concepts in intrauterine growth restriction. J. Intensive. Care Med. 2004;19(6):307–319. doi: 10.1177/0885066604269663. [DOI] [PubMed] [Google Scholar]; • Paper illustrated the impact of intrauterine growth restriction on later onset disease.
- 3.Ross MG, Beall MH. Adult sequelae of intrauterine growth restriction. Semin. Perinatol. 2008;32(3):213–218. doi: 10.1053/j.semperi.2007.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gluckman PD, Hanson MA, Cooper C, Thornburg KL. Effect of in utero and early-life conditions on adult health and disease. N. Engl. J. Med. 2008;359(1):61–73. doi: 10.1056/NEJMra0708473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Gillman MW. Developmental origins of health and disease. N. Engl. J. Med. 2005;353(17):1848–1850. doi: 10.1056/NEJMe058187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Robins JC, Marsit CJ, Padbury JF, Sharma SS. Endocrine disruptors, environmental oxygen, epigenetics and pregnancy. Front. Biosci. 2011;3:690–700. doi: 10.2741/e279. [DOI] [PMC free article] [PubMed] [Google Scholar]; • Revealed the important role of epigenetics in placenta development.
- 7.Nelissen EC, Van Montfoort AP, Dumoulin JC, Evers JL. Epigenetics and the placenta. Hum. Reprod. Update. 2011;17(3):397–417. doi: 10.1093/humupd/dmq052. [DOI] [PubMed] [Google Scholar]
- 8.Tycko B. Imprinted genes in placental growth and obstetric disorders. Cytogenet. Genome. Res. 2006;113(1–4):271–278. doi: 10.1159/000090842. [DOI] [PubMed] [Google Scholar]
- 9.Hemberger M. Epigenetic landscape required for placental development. Cell. Mol. Life. Sci. 2007;64(18):2422–2436. doi: 10.1007/s00018-007-7113-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Du M, Zhou W, Beatty LG, Weksberg R, Sadowski PD. The KCNQ1OT1 promoter, a key regulator of genomic imprinting in human chromosome 11p15.5. Genomics. 2004;84(2):288–300. doi: 10.1016/j.ygeno.2004.03.008. [DOI] [PubMed] [Google Scholar]
- 11.St-Pierre J, Hivert MF, Perron P, et al. IGF2 DNA methylation is a modulator of newborn's fetal growth and development. Epigenetics. 2012;7(10):1125–113. doi: 10.4161/epi.21855. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Cordeiro A, Neto AP, Carvalho F, Ramalho C, Doria S. Relevance of genomic imprinting in intrauterine human growth expression of CDKN1C, H19, IGF2, KCNQ1 and PHLDA2 imprinted genes. J. Assist. Reprod. Genet. 2014;31(10):1361–1368. doi: 10.1007/s10815-014-0278-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bourque DK, Avila L, Penaherrera M, Von Dadelszen P, Robinson WP. Decreased placental methylation at the H19/IGF2 imprinting control region is associated with normotensive intrauterine growth restriction but not pre-eclampsia. Placenta. 2010;31(3):197–202. doi: 10.1016/j.placenta.2009.12.003. [DOI] [PubMed] [Google Scholar]
- 14.Stewart PM, Murry BA, Mason JI. Type 2 11 beta-hydroxysteroid dehydrogenase in human fetal tissues. J. Clin. Endocrinol. Metab. 1994;78(6):1529–1532. doi: 10.1210/jcem.78.6.8200959. [DOI] [PubMed] [Google Scholar]; • The first study showing the association between placenta HSD11B2 methylation and fetal growth.
- 15.Marsit CJ, Maccani MA, Padbury JF, Lester BM. Placental 11-beta hydroxysteroid dehydrogenase methylation is associated with newborn growth and a measure of neurobehavioral outcome. PLoS ONE. 2012;7(3):e33794. doi: 10.1371/journal.pone.0033794. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Zhao Y, Gong X, Chen L, et al. Site-specific methylation of placental HSD11B2 gene promoter is related to intrauterine growth restriction. Eur. J. Hum. Genet. 2014;22(6):734–740. doi: 10.1038/ejhg.2013.226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ferreira JC, Choufani S, Grafodatskaya D, et al. WNT2 promoter methylation in human placenta is associated with low birth weight percentile in the neonate. Epigenetics. 2011;6(4):440–449. doi: 10.4161/epi.6.4.14554. [DOI] [PubMed] [Google Scholar]
- 18.Filiberto AC, Maccani MA, Koestler D, et al. Birthweight is associated with DNA promoter methylation of the glucocorticoid receptor in human placenta. Epigenetics. 2011;6(5):566–572. doi: 10.4161/epi.6.5.15236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zhao Y, Shi HJ, Xie CM, Chen J, Laue H, Zhang YH. Prenatal phthalate exposure, infant growth, and global DNA methylation of human placenta. Environ. Mol. Mutagen. 2015;56(3):286–292. doi: 10.1002/em.21916. [DOI] [PubMed] [Google Scholar]
- 20.Valero De Bernabe J, Soriano T, Albaladejo R, et al. Risk factors for low birth weight: a review. Eur. J. Obstet. Gynecol. Reprod. Biol. 2004;116(1):3–15. doi: 10.1016/j.ejogrb.2004.03.007. [DOI] [PubMed] [Google Scholar]
- 21.Hadlock FP, Harrist RB, Sharman RS, Deter RL, Park SK. Estimation of fetal weight with the use of head, body, and femur measurements – a prospective study. Am. J. Obstet. Gynecol. 1985;151(3):333–337. doi: 10.1016/0002-9378(85)90298-4. [DOI] [PubMed] [Google Scholar]
- 22.Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States national reference for fetal growth. Obstet. Gynecol. 1996;87(2):163–168. doi: 10.1016/0029-7844(95)00386-X. [DOI] [PubMed] [Google Scholar]
- 23.Chang TC, Robson SC, Spencer JA, Gallivan S. Identification of fetal growth retardation: comparison of Doppler waveform indices and serial ultrasound measurements of abdominal circumference and fetal weight. Obstet. Gynecol. 1993;82(2):230–236. [PubMed] [Google Scholar]
- 24.Mcternan CL, Draper N, Nicholson H, et al. Reduced placental 11 beta-hydroxysteroid dehydrogenase type 2 mRNA levels in human pregnancies complicated by intrauterine growth restriction: an analysis of possible me mechanisms. J. Clin. Endocrinol. Metab. 2001;86(10):4979–4983. doi: 10.1210/jcem.86.10.7893. [DOI] [PubMed] [Google Scholar]
- 25.Nishizawa H, Ota S, Suzuki M, et al. Comparative gene expression profiling of placentas from patients with severe pre-eclampsia and unexplained fetal growth restriction. Reprod. Biol. Endocrinol. 2011;9:107. doi: 10.1186/1477-7827-9-107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Street ME, Seghini P, Fieni S, et al. Changes in interleukin-6 and IGF system and their relationships in placenta and cord blood in newborns with fetal growth restriction compared with controls. Eur. J. Endocrinol. 2006;155(4):567–574. doi: 10.1530/eje.1.02251. [DOI] [PubMed] [Google Scholar]
- 27.Maccani MA, Marsit CJ. Epigenetics in the placenta. Am. J. Reprod. Immunol. 2009;62(2):78–89. doi: 10.1111/j.1600-0897.2009.00716.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Li F, Chong ZZ, Maiese K. Winding through the WNT pathway during cellular development and demise. Histol. Histopathol. 2006;21(1):103–124. doi: 10.14670/hh-21.103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Benediktsson R, Lindsay RS, Noble J, Seckl JR, Edwards CR. Glucocorticoid exposure in utero: new model for adult hypertension. Lancet. 1993;341(8841):339–341. doi: 10.1016/0140-6736(93)90138-7. [DOI] [PubMed] [Google Scholar]
- 30.Reinisch JM, Simon NG, Karow WG, Gandelman R. Prenatal exposure to prednisone in humans and animals retards intrauterine growth. Science. 1978;202(4366):436–438. doi: 10.1126/science.705336. [DOI] [PubMed] [Google Scholar]
- 31.Oakley RH, Cidlowski JA. The biology of the glucocorticoid receptor: new signaling mechanisms in health and disease. J. Allergy. Clin. Immunol. 2013;132(5):1033–1044. doi: 10.1016/j.jaci.2013.09.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Zudaire E, Cuesta N, Murty V, et al. The aryl hydrocarbon receptor repressor is a putative tumor suppressor gene in multiple human cancers. J. Clin. Invest. 2008;118(2):640–650. doi: 10.1172/JCI30024. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Opitz CA, Litzenburger UM, Sahm F, et al. An endogenous tumour-promoting ligand of the human aryl hydrocarbon receptor. Nature. 2011;478(7368):197–203. doi: 10.1038/nature10491. [DOI] [PubMed] [Google Scholar]; • The first study showing differential methylation sites across the genome in relation to maternal smoking during pregnancy using the 450K platform.
- 34.Joubert BR, Haberg SE, Nilsen RM, et al. 450K epigenome-wide scan identifies differential DNA methylation in newborns related to maternal smoking during pregnancy. Environ. Health. Perspect. 2012;120(10):1425–1431. doi: 10.1289/ehp.1205412. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.