Associations between Placental Insulin-Like Growth Factor-1 Gene Expression, DNA Methylation and Intrauterine Growth Restriction

Intrauterine growth restriction (IUGR) is a common fetal development disorder which has great impact on neonatal health. Insulin-like growth factor-1 (IGF1) has an important role in regulating fetal growth. Whether IGF1 DNA methylation was associated with IUGR has not been studied. Placenta samples from IUGR (n = 27) and normal delivery (n = 29) were collected whereas basic information of mothers and infants were also collected. RT-PCR was performed to examine IGF1 transcriptions and bisulfite sequencing PCR was used for DNA methylation analysis. Gene expression analysis found IUGR had significantly lower IGF1 transcription compared to control group (IUGR: 0.330 ± 0.351; control group: 1.001 ± 0.800, t = 3.995, P < 0.001). CpG sites at the promoter region ofIGF1 were all highly methylated and there is no difference on DNA methylation rate between IUGR and control group (IUGR: 75%; control group: 81%; P = 0.09). methylation rate (r = 0.095, P = 0.063). Intrauterine fetal growth restriction placenta had significantly lower IGF1gene expression; however, IGF1 DNA methylation level was similar. A potential fetus gender difference was also found in IGF1 DNA methylation rate.


Introduction
Fetal growth restriction, also called intrauterine growth restriction (IUGR), is a common fetal development disorder which has great impact on neonatal health.
The incidence of IUGR in China was reported to be 8.77% which equated to 1.6 million of affected neonates per year, posing a great burden on public health [1]. IUGR refers to the restricted fetal growth which is unable to reach genetic growth potential owning to pathogenic factors from maternal, placental and fetal causes. Because fetal intrauterine growth and development is difficult to monitor, so ultrasound is used to predict IUGR. Ultrasound measurement of fetal growth meridians includes: biparietal diameter, head circumference, abdominal circumference, femoral length. The fetal weight at the corresponding gestational age was calculated according to the Hadlock criterion to diagnose whether it is lower than the 10th percentile of the same fetal age to predict IUGR [2], which has been demonstrated to be the important cause for neonatal comorbidity and mortality [3]. Children with IUGR are born smaller than gestational age (SGA), the term SGA will be used exclusively to describe newborns whose birth weight is less than the 10th percentile for gestational age [4]. SGA neonates have four times or ten times higher mortality risk than infants with birth weight of 2500 -2999 grams or 3000 -3499 grams, respectively [5]. More importantly, intrauterine growth retardation predisposes children to a greater risk for diseases later in life. Previous studies have shown that a compromised intrauterine environment not only increased the risk of IUGR, Preeclampsia (PE), Intrauterine virus infection and malnutrition all cause IUGR in premature infants, causes the fetus to be born SGA. These children have inadequate intrauterine nutrition reserves, there were more complications after birth, prone to feeding intolerance, infection and other causes, this causes slow growth after birth, but also the risk of diseases during adulthood and led to permanent change of physiology and metabolism through modulating developmental programming [6]. Researchers have pointed out that IUGR neonates had greater risk of intellectual and physical growth retardation, insulin resistance, obesity, and diabetes later in childhood and adulthood [7] [8] [9]. It was indicated by a previous study that the insulin-like growth factor 1 (IGF1) in the umbilical cord was responsible for the growth, nutrition transportation of fetus and the placenta [10]. Hence, genetic regulation of IGF1 may be important in modulating fetal growth and has a potential role in IUGR. IGF1 expression starts at early stage of embryo development [11]. To date, it has been acknowledged that IGF1's role on fetal development including regulating cell division, differentiation and apoptosis [12]. It also affects placental development and transportation and promotes anabolism of glucagon, protein and fats [13] [14] [15].
Previous studies also demonstrated that IGF1 axis mutation led to severe fetal growth retardation [16] [17].
DNA methylation is one of the earliest found genetic modulation approaches and affects gene expression. It was found that at least 50% of gene expression was negatively associated with DNA methylation rate [18] [19]. It remained possible that DNA methylation of IGF1 was associated with its gene expression and thus, affected fetal growth. In this study, we aimed to investigate the association between placental expression of IGF1, its DNA methylation and IUGR.  Table 1.

Patients
IUGR was diagnosed as a fetal weight less than the tenth percentile or two standard deviations of the average fetal weight of the same gestational age. All participants have signed consent forms.

Placenta Tissue Samples Collection
After the delivery, the placenta is collected from the central part of the mother's side, a small piece of placenta was dissected (1 cm × 1 cm × 1 cm) from the whole tissue by aseptic technique. The placenta tissue was then placed in RNAlater solution with diethyl pyrocarbonate (Ambion, Thermo Scientific, China) and stored in the 4˚C freezer for overnight. Long-term storage of placenta tissue was kept under −80˚C.

DNA and RNA Extraction
Fifty milligrams of placenta tissue was ground into mince and DNA and RNA were than extracted by the Gene and RNA extraction kit (BioTeke Corporation, Beijing, China) according to the manufacturer's instruction.

IGF1 Methylation Analysis
Two hundred nanograms of DNA from each placenta sample were processed by bisulfite using EZ DNA Methylation-Gold Kit M (Zymo Research, CA, United States) according to manufacturer's instruction. Processed DNA sample was kept under 20˚C before BSP (bisulfite sequencing PCR) procedure. PCR was performed on all processed DNA samples using TaKaRaEpiTaq TM HS kit (Takara, Dalian, China). MethPrimer software was used for prediction of methylation status at the promoter area of IGF1. [19] Nine CpG sites at the upstream region of P3 promoter region where CpG sites were clustered (−861 -−574) were selected for methylation analysis. The forward primer was 5'-TGAGGTTATTTTGTTTGATTTGTGTAG-3' and the reverse primer was 5'-AATTTCTACTTAACCCACCCACTTAC-3' (Sangon Biotech Co., Ltd., Shanghai, China).

DNA Cloning and Sequencing
The PCR product was connected to pMDR18-T and transferred into E. coli DH5α competent cells. Confirmed clones with PCR product were selected and plasmids were sequenced by Sangon Biotech Co., Ltd., Shanghai. Methylation

Statistical Analysis
All data analysis was performed in SPSS v17.0 (SPSS Inc, Chicago). Students' t test was used for comparing the mRNA expression level between IUGR and control groups. Comparisons on DNA methylation was performed by Chi-square Fisher's exact method. Spearman correlation analysis was used for the correlation between mRNA expression and DNA methylation. P level less than 0.05 was considered as statistically significant.

Results
We

Placental mRNA Expressions of IGF1
The mRNA transcripts of IGF1 from placenta from IUGR and those from control group were compared. Results showed that lower mRNA transcription of IGF1 was demonstrated in the IUGR group. The relative transcription levels of IUGR were 33% of that of control group (IUGR: 0.330 ± 0.351; control group: 1.001 ± 0.800) (Figure 1), which was statistically significant (t = 3.995, P < 0.001).

Methylation Analysis of IGF1
No CpGisland was found within the promoter region of IGF1as predicted by Methprimer software. Nine CpG sites upstream of the promoter (CpG1 to CpG9) where CpG sites were more accumulated were than selected for methylation analysis. BSP analysis showed that both IUGR and control group IGF1 were highly methylated (IUGR: 75%; control group: 81%; P = 0.09) (Figure 2 and Figure 3).
However, there as a significantly higher methylation rate found in the placenta of male fetus than female fetus in the control groups (male: 87%; female: 74%, P = 0.016). Similar trend was also found in the IUGR group where the male fetus had higher IGF1 methylation rate than female fetus (male: 82%; female: 69%, P = 0.012). Overall, placenta from male fetus had significantly higher rate for meIGF1 methylation than the female fetus ( Figure 4). In the IUGR group, the PE group methylation rate was 85%, the non-PE group methylation rate was 71%, there were significant differences between the two groups (P = 0.06).
X. J. Li et al.

Association between mRNA Transcription and DNA Methylation of IGF1
Although the mRNA transcriptions of IGF1 were significantly different between IUGR and control groups, there is no significant association between mRNA transcription levels and the methylation rates of IGF1 (r = 0.095, P = 0.063).

Discussion
This study demonstrated that the placenta from IUGR fetus had significantly lower IGF1 transcription levels. The IUGR placenta showed lower IGF1 transcription level) with a 66% reduction found in IUGR group compared to control group. A homozygous partial deletion of the IGF1 gene was demonstrated in a case with poor fetal growth and persistent postnatal physical growth retardation [20]. More recently a Greek study also demonstrated reduced gene expressions of IGF1 and its binding protein in IUGR placentas [21]. Our results are consistent with previous study and support the important role of IGF1 in maintaining normal fetal growth. However, our results demonstrated no significant correlation between IGF1 gene expression and its DNA methylation rate. (A decreased transcription level) was not associated with IGF1DNA methylation status. Instead, a potential fetus's gender effect may modulate DNA methylation rate. As suggested by Kaiser et al., DNA methylation and gene expression might not be a simple linear relationship, but rather a threshold relationship [22]. Higher methylation inhibits gene expression whereas low methylation state enables gene expression [23], although the level of expression might also be affected by other factors.
A previously published investigation by our group found that IGF1 expression reduced in infants with a birth weight higher than 4260 grams without changes in DNA methylation rate [24]. Our data also showed that IGF1 gene had no CpGisland where CpG sites are mostly demethylated [25]. This high methylation rate may change the position of nucleosome, thus affecting gene expression [26].
The results of our study argue against the association between IGF1expression and its DNA methylation, Similar results were also reported in other studies.
Turan et al.reported a lack of association between gene expression and DNA methylation in 23 growth related genes [27]. Similar trends were also found in E-cadherin, GADgene and DLC1 gene [28] [29] [30].  [32], which can combine with different protein or protein complex under different circumstances and change the relevant gene expressions. More recent studies also suggested that gender difference in DNA methylation was also affected by multiple factors [33] [34].
In addition, we found that the methylation rate of IUGR varies with different factors.
There were 8 cases of PE on among the children with IUGR, and these had higher rates of placental methylation [35]. Jane W et al. found that the IGF1 promoter region of the placenta was hypermethylated in PE patients, which is consistent with our study, PE is one of the important causes of IUGR [36]. The invasion of E trophoblast cells into the spiral artery was limited, and the failure of placental vascularization leads to placental ischemia and hypoxia [37]. Severe hypoxia alters the epigenetic characteristics of the human placental trophoblast in culture [38]. Therefore, IGF1of PE patients in the IUGR group showed hypermethylation. In conclusion, our study has demonstrated that intrauterine fetal growth retardation placenta had significantly lower IGF1 gene expression. There was no difference in DNA methylation between IUGR and control placentas, and DNA methylation was not associated with IGF1gene expression levels. The methylation rate of IUGR caused by different factors was different. IUGR is the result of a variety of factors, and the causes of fetal growth restriction are complex. Therefore, we cannot use a single mechanism to measure the disease. However, a fetal gender difference in IGF1 DNA methylation rate was noted; its significance in IGF1gene regulation needs further elucidation. Future study on placenta whole genome analysis could give more information on genetic network on fetal growth regulation.