IGF1 gene is epigenetically activated in preterm infants with intrauterine growth restriction

Background IGF1 is a key molecule in the regulation of growth and metabolism. Low IGF1 secretion is known to cause growth restriction in childhood, as well as deregulated lipid metabolism, cardiovascular disease, and diabetes in adulthood. The IGF1 gene P2 promoter is highly methylated, resulting in low secretion of IGF1 in small infants and children. However, it is unknown when this methylation occurs. The aim of study was to clarify the point when this epigenetic program occurs during intrauterine development. We analyzed 56 preterm infants born before 32 weeks of gestation, including 19 intrauterine growth restriction (IUGR) infants whose birth weights were lower than − 2SD calculated by the Japanese datasets. We extracted genomic DNA from whole blood at birth; methylation of the six CpG sites in the IGF1 P2 promoter was analyzed by the bisulfite amplicon method using the MiSeq platform. Results In contrast to term infants and children, the methylation of all six CpG sites positively correlated with body weight and body length at birth. IGF1 P2 promoter methylation levels were significantly reduced in all six CpG sites in infants with IUGR. Conclusions These findings indicated that the IGF1 gene is epigenetically activated before 32 weeks of gestation in infants with IUGR and that the activated gene may become suppressed after this time point. This study may provide new insights to prevent the onset of adult diseases and to aid in nutritional management for preterm birth infants in neonatal intensive care units.

low IGF1 and deregulated lipid metabolism, cardiovascular disease, diabetes, and altered metabolic profile of diabetic patients in adulthood have been reported [9]. IGF1 is detectable in many fetal tissues from the first trimester, suggesting that it plays a role in early fetal development. However, increased levels of circulating IGF1 during the period of fetal life that corresponds to the third trimester, especially after 32 weeks of gestation, suggest that IGF1 is more important in the later months of fetal growth [10].
Although parental height correlates with offspring height [11], research on the association between the IGF1 gene and human stature indicated that associated genetic factors, such as allelic polymorphism and DNA mutations, only explained a minor part of the expected heritable fraction [12]. On the other hand, epigenetic modifications, such as DNA methylation, can contribute to alteration of gene expression in a heritable manner without affecting the underlying genomic sequences. This hypothesis may explain instances of heritable adaptations to changing environmental conditions, such as human stature [13].
There are several transcripts of the IGF1 gene from different transcriptional start sites. Class 1 transcripts have initiation sites on exon 1 and are driven by the P1 promoter, while class 2 transcripts use exon 2 as a leader exon and are driven by the P2 promoter [14][15][16]. IGF1 transcripts initiating at P1 are constitutively expressed in many tissues, while transcripts initiating at P2 are expressed primarily, but not exclusively, in the liver [17].
Ouni et al. examined whether CG methylation of the two promoters (P1 and P2) of the IGF1 gene is a potential epigenetic contributor to the individual variation in circulating IGF1 and stature in growing children. Their observations introduced epigenetics as an individual determinant of child growth and serum IGF1 level. Six CpG sites, CG108, CG137, CG207, CG218, CG224, and CG232 in the P2 promoter of the IGF1 gene, especially CG137, are the first epigenetic quantitative trait loci reported in humans [17]. The methylation of these six sites is associated with growth hormone responsiveness of children presenting with short stature at 8-9 years of age [18]. Moreover, Stunff et al. reported that CG137 methylation is associated with birth length in term newborn infants [19].
To determine whether individual variation in CpG site methylation is determined in the postimplantation embryo at the time of primary shaping of the methylome or as a result of maternal signals transmitted through the placenta at a postembryonic stage of intrauterine life, we investigated the methylation status of the IGF1 P2 promoter including CG137 in preterm (< 32 weeks of gestation) infants with and without intrauterine growth restriction (IUGR).

Participants
This study was approved by the Juntendo University Ethics Committee and conducted in accordance with the principles of the Declaration of Helsinki. We analyzed 56 preterm infants born before 32 weeks of gestation, including 19 infants with IUGR whose birth weights were lower than − 2SD calculated by the Japanese datasets (http://jpse.umin.jp/taikakubirthlongcrossv1.xlsx); all were admitted to the neonatal intensive care unit (NICU) of Juntendo University Shizuoka Hospital between July 2017 and August 2019, and their parents had provided written informed consent for inclusion in the study.

Sample collection
A sample of 10 μL of whole blood was taken and frozen immediately at the time of admission to the NICU as part of routine blood sampling and stored at − 80°C until analysis. Genomic DNA was extracted from the samples in August 2019.

Statistics
The results were analyzed by chi-square test, Mann-Whitney U test, and Spearman's correlation analysis. Hierarchical cluster analysis of individual methylation patterns of six CpG sites on each haplotype was performed. All statistical analyses were performed using SPSS Statistics, version 24•0 (IBM Corp., Armonk, NY, USA). In all analyses, p < 0.05 was considered to indicate statistical significance. The z scores of birth weight and length were calculated by subtracting the mean from the raw score, then dividing the difference by the population standard deviation using the Japanese datasets (http// jpse.umin.jp/taikakubirthlongcross1.xlsx).

Participant characteristics
The backgrounds of participants are shown in Table 1. All participants were Japanese. The gestational age ranged from 25 weeks + 1 day to 31 weeks + 6 days among infants in the IUGR ( 8,39.0] cm). Body height at birth was significantly shorter among infants in the IUGR group (p = 0.003). Fourteen of 19 infants in the IUGR group and 24 of 37 infants in the non-IUGR group were boys; there was no significant difference in sex distribution between groups. Regarding maternal contribution to intrauterine growth, the following factors were analyzed: maternal information; frequency of pregnancy by in vitro fertilization (IVF), frequency of twin pregnancy (twin), and frequency of hypertensive disorders of pregnancy (HDP) diagnosed  The differences between groups were analyzed by Mann-Whitney U test or chi-square test by an obstetrician. HDP was diagnosed by hypertension (systolic BP ≥ 140 and/or diastolic BP ≥ 90 mmHg) and proteinuria (spot urine protein/creatinine ratio ≥ 30 mg/ mmol). Among infants in the IUGR group, one IVF, one twin, and nine HDP were observed. Among infants in the non-IUGR group, six IVF, 10 twin, and three HDP were observed. There were no significant differences in frequency of IVF and twin pregnancies (p = 0.241, p = 0.202, respectively). Nine of the 19 infants in the IUGR group had HDP, which was significantly higher than the rate in the non-IUGR infant group (p = 0.001).
To assess the postnatal influence of IGF1 secretion, we investigated the frequency of infantile complications that may be associated with low IGF1 concentration, such as bronchopulmonary dysplasia (BPD; defined as the need for additional oxygen after the age of 28 days), retinopathy of prematurity (ROP; defined as the need for treatment by an ophthalmologist), and poor brain development (defined as abnormal brain magnetic resonance imaging findings (MRI) at the time of discharge). BPD was observed in 12 infants in the IUGR group and 18 infants in the non-IUGR group. ROP was observed in two infants in the IUGR group and five infants in the non-IUGR group. Abnormal MRI findings were observed in one infant in the IUGR group and 10 infants in the non-IUGR group. There was no difference between the two groups with regard to clinical complications during the neonatal period (BPD p = 0.086, ROP p = 0.749, MRI p = 0.241).

Discussion
In contrast to term infants and children, IGF1 P2 promoter methylation levels were significantly reduced in infants with IUGR. Our findings suggested that preterm infants can adapt to intrauterine adverse nutritional environments via epigenetic modification of DNA in the early developmental stages, and that > 32 weeks of gestation is the critical time window for programming of later body growth through IGF1 gene methylation. Children's heights are known to be highly correlated with their parents' heights [11]. Therefore, programming of growth-related genes, including IGF1, may occur at least in part in the early embryonic stage, such as the postimplantation stage. The results of the present study showed that IGF1 secretion is epigenetically regulated to be high in IUGR infants born at < 32 weeks of gestation, in contrast to term newborn infants and children. This finding indicated that methylation of the IGF1 gene has the ability to respond to in utero malnutrition before 32 weeks of gestation and may be reset at 32-37 weeks of gestation. Epigenetic reprogramming has been Fig. 2 Correlation of methylation status between the six CG sites. Correlation coefficients were calculated by Spearman's correlation analysis. All correlations were significant (p < 0.001) demonstrated in mammals at distinct, key developmental stages, such as in the zygote, in the primordial germ cells, and during early body development. There have been a few reports that epigenetic reprogramming occurs after organogenesis, such as glucocorticoid receptor gene methylation by the postnatal environment in preterm infants [21].
HDPs are associated with low birth weight and represent a major cause of maternal and/or perinatal mortality and morbidity. The mechanisms underlying HDPassociated complications are poorly understood, but epigenetic patterns contribute to the pathogenesis of fetal complications. Epigenome-wide DNA methylation analysis in cord blood of term infants revealed that   Haplotype methylation analysis of individuals. C represents a methylated CpG site, and T represents an unmethylated CpG site. CCCCCC means CG107, CG137, CG207, CG218, CG224, and CG232 were methylated, respectively. The frequency of each haplotype was shown as a heatmap. Hierarchical cluster analysis revealed a cluster surrounded by green where IUGR infants were dominant and relatively high TTTTTT and low CCCCCC were seen. There were two clusters surrounded by blue where non-IUGR infants were dominant and relatively high CCCCCC and low TTTTTT were seen HDP were associated with DNA methylation at 43 CpG sites [22]. The IGF1 P2 promoter region was not covered by the analysis. Thus, it is unclear whether it was differentially methylated in infants with maternal HDP. The present study indicates that maternal HDP may not have a large effect on IGF1 P2 promoter methylation.
Haplotype methylation analysis revealed that methylation change tended to occur simultaneously at several CpG sites. It seems to be important for future study.
A limitation of the present study was its lack of data regarding serum IGF1 concentrations. This was due to the large sample volume required from infants with a low body weight of 342-1900 g and the risk of postnatal hypovolemia. It is known that serum IGF1 concentrations have been positively correlated with birth weight and length in preterm infants [23]. Therefore, the IGF1 concentration may have been lower in the IUGR group than in the non-IUGR group in the present study due to intrauterine malnutrition status, despite epigenetic activation of the IGF1 gene.
After preterm birth, there is a decline in serum IGF1 concentration that is induced by undernutrition, deficient nutrition utilization [24], hypoxia [25], inflammation [26], and genetic factors, as well as hormones (e.g., thyroid hormone and cortisol). A prolonged duration of low serum IGF1 in extremely premature infants is strongly associated with increased risks of multiple major neonatal morbidities, which have a significant impact on long-term health [27][28][29], including BPD, ROP, and poor brain development. There were no differences in rates of these complications between the two groups in the present study. Therefore, the effect of low methylation of the IGF1 gene may have been reduced by severe conditions to produce IGF1 after preterm birth.
Tissue differences in epigenomic status represent the most important limitation of epigenetic studies in human subjects, including the present study. However, the tissue differences in IGF1 gene methylation are relatively well characterized. It has been shown that the methylation levels were comparable in white blood cells, peripheral blood mononuclear cells, and CD4 T lymphocytes. The methylation levels were approximately 50% lower for most P2 CGs in the liver and growth plates, which are the main sites of IGF1 production [17]. Therefore, the use of white blood samples in the present study was presumed to cause few problems related to tissue differences.
The evaluation of IGF1 gene methylation after 32 weeks of gestation in the uterus or after preterm birth may be important for perinatal management of infants with IUGR. Further, longitudinal investigations are needed along with the development of methods for analyzing the fetal epigenome using cell-free DNA in maternal blood.
Studies examining the epigenetic status of the IGF1 gene in deregulated lipid metabolism, cardiovascular disease, and diabetes are limited. IGF1 P1 promoter methylation levels were increased in type 2 diabetes patients compared with normal glucose tolerance subjects, while serum IGF1 levels were lower in type 2 diabetes [30]. Additional epigenetic studies of IGF1 P2 promoter methylation dynamics are needed to understand these interactions further.

Conclusions
It has been suggested that > 32 weeks of gestation may be the critical time window for programming of later body growth through IGF1 gene methylation.
This study may provide new insights to prevent the onset of adult diseases and to aid in nutritional management for preterm birth infants in neonatal intensive care units.