Fetal growth is associated with the CpG methylation of the P2 promoter of the IGF1 gene
Le Stunff et al. Clinical Epigenetics
Fetal growth is associated with the CpG methylation of the P2 promoter of the IGF1 gene
Catherine Le Stunff 0 3
Anne-Laure Castell 2
Nicolas Todd 0 3
Clémence Mille 0 3
Marie-Pierre Belot 0 3
Nathalie Frament 0 3
Sylvie Brailly-Tabard 1
Alexandra Benachi 5
Delphine Fradin 4
Pierre Bougnères 0 3
0 Institut National de la Santé et de la Recherche Médicale U1169, Bicêtre Hospital, Paris Sud University , Le Kremlin-Bicêtre , France
1 Service de BiologieMoléculaire et Hormonologie, Bicêtre Hospital, Paris Sud University , Le Kremlin-Bicêtre , France
2 Service de Médecine des Adolescents, Bicêtre Hospital, Paris Sud University , Le Kremlin-Bicêtre , France
3 Institut National de la Santé et de la Recherche Médicale U1169, Bicêtre Hospital, Paris Sud University , Le Kremlin-Bicêtre , France
4 CRCINA, INSERM U1232, Université de Nantes , Nantes , France
5 Service de Gynécologie-Obstétrique, Antoine Béclère Hospital, Paris Sud University , Clamart , France
Background: There are many reasons to think that epigenetics is a key determinant of fetal growth variability across the normal population. Since IGF1 and INS genes are major determinants of intrauterine growth, we examined the methylation of selected CpGs located in the regulatory region of these two genes. Methods: Cord blood was sampled in 159 newborns born to mothers prospectively followed during their pregnancy. A 142-item questionnaire was filled by mothers at inclusion, during the last trimester of the pregnancy and at the delivery. The methylation of selected CpGs located in the promoters of the IGF1 and INS genes was measured in cord blood mononuclear cells collected at birth using bisulfite-PCR-pyrosequencing. Results: Methylation at IGF1 CpG-137 correlated negatively with birth length (r = 0.27, P = 3.5 × 10−4). The same effect size was found after adjustment for maternal age, parity, and smoking: a 10% increase in CpG-137 methylation was associated with a decrease of length by 0.23 SDS. Conclusion: The current results suggest that the methylation of IGF1 CpG-137 contributes to the individual variation of fetal growth by regulating IGF1 expression in fetal tissues.
Newborn; Growth; Epigenetics; DNA methylation
Since human neonates are about the same size as the
birth canal, size at birth has to be a tightly regulated
trait. It is controlled by a number of interacting genetic
and environmental factors. Among these factors, IGF1
(insulin-like growth factor-1), IGF2 (insulin-like growth
factor-2), and insulin are three key players. The
expression of IGF1 and IGF2 genes in fetal tissues and placenta
is essential for fetal growth, as indicated by studies of
knock-out mice and genetic defects in human. Although
IGF2 is more abundantly expressed in fetal serum and
tissues than IGF1, IGF1 is more closely associated with
fetal growth in a majority of species. Insulin secreted by
the fetal ß cells is also a major growth factor during
intrauterine life [
In animal studies, IGF1 gene ablation reduces fetal
]. Fetal IGF1 promotes tissue growth by
stimulating anabolic events and DNA synthesis.
Circulating IGF1 is most often decreased in animal models of
fetal growth restriction, as well as in human fetal growth
restriction. Deletions or loss-of-function mutations in
the IGF1 gene cause intrauterine and postnatal growth
]. In mice, the deletion of the IGF2 gene
reduces fetal growth [
]. The effects on fetal growth are
additive to those of IGF1. IGF2 is an imprinted gene.
The expression of the maternal allele is silenced during
fetal life, leading to fetal growth restraint when the
paternal allele is mutated or deleted. Epimutations and
molecular alterations of the human chromosomal region
11p15.5, which harbors IGF2 [
], and paternally
inherited nonsense mutations of the IGF2 gene [
associated with the Silver–Russell syndrome, a syndromic
growth-retardation disorder. Heterozygous mutations in
the INS gene are associated with a reduction of fetal
While monogenic defects in the IGF1, IGF2, and INS
genes demonstrate the major role of these genes for fetal
growth, the genetics of the individual variability in
birthweight of normal neonates is complex, as for most
quantitative traits, and relies on numerous polymorphic variations
of the genome. Birth size is a multifactorial phenotype
resulting from many processes and gene expression
patterns operating during fetal development. Genetic variation
is said to account for 38–80% of birth weight variance, with
a considerable variability in the estimates [
genome-wide association study found 60 loci to be
associated with birth weight at genome-wide significance .
Despite their implication in monogenic disorders of fetal
growth, neither IGF1 nor INS loci were among these 60
loci, but pathways involved in insulin and IGF signaling
were said to be involved. As pointed out for most
quantitative traits studied in humans, the genetics of birth size face
the missing heritability problem. Indeed, in contrast with
the estimates of birth weight heritability [
], the 60
variants most recently identified could only explain 2% of the
variance in birth weight . When SNPs at the INS [
] and IGF1 loci [
] were examined specifically as
candidates for their contribution to the variation of size at
birth, they showed inconsistent results [
The genotype of the fetus is not the sole determinant of
fetal growth. It is likely that maternal genes that regulate
the environment of the womb are important determinants
of birth size. Maternal environment also has a major
influence, although most of the environmental factors and
mechanisms of reduced fetal growth remain largely
unknown. Non-inherited maternal factors found to influence
birth size include parity, mother weight at birth and before
pregnancy, weight gain of pregnancy, and maternal
smoking, but cannot explain the vast majority of cases of
intrauterine growth retardation (IUGR) [
nutrition, in industrialized populations seems to have only a
small effect on placental and birth weights [
even in affluent societies where mothers are well
nourished, many children are still born with unexplained small
fetal size. On the other side, extreme maternal nutrition,
such as the Dutch Hunger Winter famine, is associated
with a reduced birth weight [
As for all quantitative traits, epigenetic influences are
likely to contribute to the regulation of birth size at the
individual level. Few studies have yet explored the
epigenetics of fetal growth in humans. To our knowledge,
the only epigenome-wide association study carried out
yet found that among the 485,577 CpGs analyzed in
cord blood by the 450K Infinium BeadChip, 19 CpGs
were significantly associated with birthweight in a large
population-based cohort [
]. Another study of
methylation (performed on cord blood and placenta) restricted
to growth-related genes found no association between
intrauterine growth retardation and IUGR and the
methylation of CpGs at the IGF1, IGF2, and INS loci
(note that this study investigated only 3 CpGs located
within the IGF1 P1 promoter) [
The current study investigates the relationship
between birth length and the methylation of specific CpGs
of the IGF1 and INS genes. These CpGs located within
promoters were selected because their level of
methylation was known to control gene expression. This is the
case for IGF1 CpG-137, whose methylation is associated
with IGF1 gene transcription [
], circulating IGF1 [
and childhood growth [
]. This is also the case for
CpG-180 in the INS promoter region, which influences
INS gene transcription  and is associated with SNPs
as rs689 [
] known to predict birth size to some degree
]. We have not explored a potential association of
birth length with CpG methylation on the
nonimprinted allele of IGF2 because we did not have access
to parental genotypes.
One hundred and fifty-nine women of European
ancestry aged 20–40 years, with a singleton pregnancy, were
recruited at Antoine Béclère Maternity. The clinical
characteristics of the 159 newborns are presented in
Table 1. All were healthy neonates born after 37 weeks
of amenorrhea. None had IUGR, defined as a weight or
length below the 10th percentile for its gestational age
]. Children born before 36 weeks of gestation were
not included. Clinical data and biological samples were
collected at inclusion (< 3 months of pregnancy) and
around delivery. The main characteristics of the mothers
are presented in Table 1. All mothers filled a 142-item
questionnaire. All mothers were healthy and had a
normal nutrition; none consumed alcohol or drugs during
their pregnancy. All protocols were agreed by French
ethic boards (CODECOH DC-2013-2017, CPP
CO-14001, CCTIRS no. 14-124bis, CNIL no. 914,253). Cord
blood samples were taken within minutes of delivery,
immediately refrigerated at 4 °C and transported to
laboratory within 24 h.
Isolation of genomic DNA and bisulfite genomic conversion
Cord blood mononuclear cells (CBMC) were isolated
from fresh blood using a density gradient. Five
millimeters of fresh blood was mixed with 5 ml of NaCl
154 mM, and 5 ml of Lymphoprep solution (Eurobio,
Paris, France) was added to the diluted blood and
centrifuged for 20 min at room temperature at 800 g.
After centrifugation, the interphase containing CBMC
was carefully aspirated and the cells were mixed with
NaCl. The cell suspension was centrifuged at 300 g, and
the cell pellet washed with PBS.
Nucleic acids were extracted from CBMC using
Gentra Puregene blood kit (Qiagen, Hilden, Germany). Four
hundred nanograms of genomic DNA was treated with
EZ DNA Methylation-Gold Kit, according to the
manufacturer’s protocol (Zymo Research Corporation, CA,
USA). A bisulfite-PCR-pyrosequencing technique was
used to measure the CG methylation [
]. We improved
the resolution of this method from a handful of bases to
up to 100 nucleotides, with the ability to quantify
methylation in the same sample of blood cells with a
coefficient of variation (sd/mean) as little as 1–5% [
Pyrosequencing-based bisulfite PCR analysis
CpGs are denominated according to their position
versus each promoter transcription start site (TSS). At
the IGF1 locus, we studied 3 CpGs (-1044, -960, -919)
located within the P1 promoter and 5 CpGs (-232, -224,
-218, -207, -137) located within the P2 promoter
(Additional file 1: Figure S1). We previously have shown
that the methylation of the P1 promoter does not
influence IGF1 gene expression [
]. At the insulin locus, we
studied 2 CpGs (-206 and -180) proximal to the TSS
(Additional file 1: Figure S1). The bisulfite-treated
genomic DNA was PCR-amplified using unbiased primers
(Additional file 2: Table S1) and performed quantitative
pyrosequencing using a PyroMark Q96 ID
Pyrosequencing instrument (Qiagen). Pyrosequencing assay was
designed using MethPrimer
]. Biotin-labeled single
stranded amplicon was isolated according to protocol
using the Qiagen Pyromark Q96 Work Station and
underwent pyrosequencing with 0.5 μM of sequencing
primer. The methylation percent for each CpG within
the target sequence was calculated using PyroQ CpG
Genotyping of SNPs at the IGF1 and INS loci
Genotyping of SNP rs35767 at the IGF1 gene locus was
performed with TaqMan allelic discrimination using
Biosystems 7500 Fast Real-Time PCR System (Applied
Biosystems, Courtaboeuf, France). SNP genotyping assay
(ID: C_7999146_10) was purchased from Life
technologies (Saint Aubin, France).
Genotyping of SNP rs689 at the INS gene locus was
determined by the analysis of PCR products [
amplification was in 96-well microliter plates, each 50 μl
reaction contained DNA (100 ng), MgCl2 (1.5 mM), 10×
reaction Buffer (Thermo Scientific), dNTPs (2.0 mM
each), primers (1 μM each), and Taq Polymerase (1.
25 U, Thermo Scientific, Saint Aubin, France). Amplified
PCR products were digested with 1 unit of Hph1
enzyme (Thermo Scientific, Saint Aubin, France).
For genotyping quality, negative controls were
included in each PCR plate. Twenty percent of samples
were analyzed as duplicate for genotyping determination.
The Hardy-Weinberg equilibrium (HWE) was calculated
by computing the test for deviations in HWE and was
shown to be present across genotypes. Allele frequencies
were calculated and tested by test.
Measurement of serum IGF1 and insulin concentrations
IGF1 and insulin concentrations were measured in
serum samples from the cord blood of newborns.
IGF1 concentration was measured using a
chemiluminescent immunometric assay after pre-treatment
with acid using Immulite®2000 (Siemens Healthcare
Diagnostics Products Llanberis, UK). Insulin
concentration was measured using Liaison®Insulin (DiaSorin,
Birth weight (sds) and birth length (sds) were regressed on
methylation of CpG-137, sex, folate, supplement intake
before conception (yes/no), folate supplement intake
during pregnancy (yes/no), parity (primiparous/non
primiparous), maternal age, and smoking (yes/no) in multivariate
linear regression. Results are expressed as mean ± sd. All
statistical analyses were conducted using R 3.3.2.
Effects of smoking and folate on birth weight
Six percent of the newborns were from smoking
mothers. The mean birth length of neonates born to a
smoking mother was 0.62 SDS (95% CI 0.22–1.04) lower
than that of neonates born to a non-smoking mother (P
= 3.0 × 10−3), a result retrieved in the multivariate
analysis (P = 5.6 × 10−3). Thirty-two percent of women
took a folate supplement before conception. The
bivariate analysis showed a trend for increased birth weight
associated with folate intake before conception (P = 0.09)
, an association found strengthened in the multivariate
regression (P = 9.2 × 10−3). No association between birth
length or birth weight was found with a supplement of
folate during pregnancy.
Neither smoking nor folate intake were associated with
variations in CpG-137 methylation.
Epigenetic and genetic variation at the IGF1 locus
While methylation of CpG-218 located within the IGF1
P2 promoter correlated closely with methylation of
CpG137 (P = 9.4 × 10−15), methylation of CpGs -1044, -960,
and -919, located within the P1 promoter (Table 2),
showed no correlation with CpG-137 methylation
(Additional file 3: Figure S2). We found no
correlation between the rs35767 genotype and the
methylation of CpG-137 of the P2 or other studied CpGs
(Additional file 4: Figure S3A).
Birth length correlated negatively with methylation at
CpG-137 (r = 0.2, P = 3.5 × 10−4, Fig. 1). This finding was
confirmed after adjustment for several covariates: a 10%
increase in methylation at CpG-137 was associated with a
decrease of birth length by 0.23 SDS (95% CI 0.11–0.35;
P = 1.6 × 10−4). In contrast, methylation at CpG-137 was not
associated with birth weight (P = 0.36). The same result was
obtained in the multivariate regression framework (P = 0.15).
Cord blood IGF1 concentration showed no association with
methylation at CpG-137 (r = 0.06, P = 0.61).
Methylation at CpG-224 and CpG-218 correlated
negatively with birth length (P = 0.03 and P = 0.02,
respectively) (Additional file 5: Figure S4).
Epigenetic and genetic variation at the INS locus
Methylation of INS CpG-180 and CpG-206 showed no
correlation with birth length or birth weight. Birth
length and birth weight were not associated with rs689
alleles (Additional file 6: Figure S5).
As previously reported [
], we found that CpG-206
and CpG-180 methylation in the studied newborns was
strongly influenced by rs689 alleles (Additional file 4:
Figure S3B; P < 2.10−16).
The fetal genotype explains only a limited part of
intrauterine growth variability among individuals of the
general population [
], which leaves an important role to
maternal genotype, maternal environment, and fetal and
placental epigenetics. Animal studies have demonstrated
that environment can shape the epigenome, notably
during the intrauterine period when it has the greatest
plasticity. Epigenetic effects of the intrauterine environment
can thus influence the phenotype in later life. However,
it remains unclear as to how influential the fetal period
is in shaping the epigenome, whether different genomic
regions show varying sensitivities to this environment
during this period, and the extent to which this early
interaction is sensitive to genetic influences.
To our knowledge, few studies have examined the
association of CpG methylation with fetal growth. The first
study found an increased methylation of all CpG sites at
the IGF1 P1 promoter in the placenta of fetuses with
], associated with the previously reported
decreased IGF1 gene expression [
]. Another study used
the Illumina Infinium HM27 platform to profile CpG
methylation in CBMC in a small number of twin pairs
]. Array technology, however, uses CpGs that are not
necessarily those having specific effects on gene
expression and phenotypes. This was the case at the IGF1
locus for the array used in twins, which analyzed only 1
CpG of the P1 promoter and no CpG of the P2
promoter at the IGF1 locus, while it examined 4 CpGs at
the INS locus. Findings were that none of these CpGs
had their methylation level associated with twins’
birthweight. A third study of methylation restricted to
growth-related genes found no association between
IUGR and the methylation of CpGs at the IGF1, IGF2,
and INS loci in placenta or cord blood samples [
Instead, our study focused on CpGs located within the
regulatory region of IGF1 and INS genes, because these
two genes are known to be major contributors to fetal
growth. Some of these CpGs were selected because their
methylation was already known to be associated with gene
], the others because they were located
within neighboring regulatory regions. We found that
CpG-137 methylation was negatively associated with birth
length in normal neonates. The methylation of CpG-137
has a strong functional role upon IGF1 gene expression in
children’s PBMC and in the HEK293 cell line . It was
previously shown to be associated with postnatal growth
]. If methylation at CpG-137 in CBMC is a
proxy of the methylation in growth-promoting tissues of
the fetus (which has not been shown in the current study),
one can speculate that methylation at this CpG influences
IGF1 gene expression and IGF1 production in fetal tissues,
thus fetal growth. There is no evidence, however, that the
individual variation of CpG methylation reflects that in
growth-promoting tissues. While this is a major weakness
of our study, this weakness is shared with a majority of
studies on epigenetics in humans where blood cells are
used as proxies of physiological tissues, given that clinical
research does not have access to these tissues [
Another weakness of measuring CpG methylation in
CBMC is that it is a unique cell mixture containing
red blood cells in addition to other blood cells, and
not a well-characterized homogeneous cell type. Our
study was not able to estimate cell composition of
this mixture based on methods developed for adult
blood cells [
], nor to sort a specific cell category
from the cord blood sample.
Several hypotheses may account for the association
of methylation at CpG-137 measured at birth with
fetal growth. A first possibility is that the individual
variation in CpG-137 methylation is determined in
the post implantatory embryo, at time of the primary
shaping of the methylome. Alternatively, the
variation observed in the level of methylation of
CpG137 may result from yet unknown maternal signals
transmitted to the fetal tissues through the placenta
at a post-embryonic stage of intrauterine life. No
association of fetal growth was observed with the
other CpGs of the IGF1 locus, except for CpG-218,
another CpG of the P2 promoter.
Maternal smoking [
] or folate intake [
been shown in other studies to affect methylation of
certain CpGs, but was not found to affect the
methylation of CpG-137 or of other CpGs studied
Since insulin is a major growth factor in fetal life
and the INS VNTR has a direct effect on insulin
], it is conceivable that INS VNTR
variation influences early growth. This is why
CpG180 and CpG-206 located within the INS promoter
were selected for the current study, given that their
level of methylation affects the expression of the
insulin gene [
]. As reported previously  the
methylation of CpG-206 and CpG-180 was strongly
influenced by rs689 alleles. We found that neither
rs689 alleles nor CpG-206 or CpG-180 showed a
relationship with birth length or birth weight, but this
conclusion should await the observation of a large
number of neonates. Unlike Dunger et al. [
observed no association (Additional file 4: Figure S3)
between birth size and insulin VNTR classes or
rs689 (in complete linkage disequilibrium with
VNTR classes). This may be due to the fact that we
were not able to distinguish “non-changers” [
among the studied neonates: Non-changers is the
term used to describe infants that do not show a
catch-up growth after being born small for
In conclusion, fetal growth in normal neonates is
associated with the methylation of CpG-137 located
in the IGF1 P2 promoter. This observation supports
a significant role of IGF1 epigenetics in the
regulation of fetal growth that does not seem to be
dependent on cis-genetic variation at the IGF1
locus. Although small, this epigenetic effect is of an
order of magnitude comparable with that of many
genomic variants associated with human
Additional file 1: Figure S1. Schematic representation of the IGF1 and
INS loci. (A) The two IGF1 gene promoters are figured. CpGs are indicated
as lollypops (studied CpGs in white and non-studied CpG in black).
rs35767 is indicated by a black arrow. TSS are shown as broken arrows.
(B) INS promoter is figured. CpGs are indicated as lollypops (studied CpGs
in white and non-studied CpG in black). TSS is shown as broken arrow.
Location of primers are indicated by arrow for CpG methylation and
genotyping. Sequences of primers and location on chromosome are
provided in Additional file 2: Table S1. (PPTX 67 kb)
Additional file 2: Table S1. List of primers and location used in our
study. Sequences are given from 5′ to 3′. (DOCX 15 kb)
Additional file 3: Figure S2. Correlation matrix of methylation values
(%) at the CG located in the P1 and P2 promoters of the IGF1 gene in
newborns patients. Pearson correlation coefficient is indicated in bold,
and P value below. (PPTX 88 kb)
Additional file 4: Figure S3. Relationship between promoter CG
methylation and genotypes. (A) Methylation at CpGs-137 of the IGF1 P2
promoter is independent from the rs35767 genotypes. (B) Methylation at
CpGs-206 and CpG-180 in insulin promoter is closely dependent on rs689
alleles. (PPTX 242 kb)
Additional file 5: Figure S4. Relation between CpG methylation and
birth length (SDS) at the IGF1 promoter 1 and 2. (A) at IGF1 P1 promoter,
we observed no significant correlation of birth length with the studied
CpGs, (B) at IGF1 P2 promoter, only two CpGs other that CpG-137
showed a weak correlation with birth length. The correlation between
CpG-224 (%) and birth length (SDS) is described by the equation: Birth
length = − 0.014*[CpG-224 methylation] + 0.43 (r = 0.17, P = 0.03). The
correlation between CpG-218 (%) and birth length (SDS) is described by
the equation: Birth length = − 0.016*[CpG-218 methylation] + 0.58 (r = 0.2,
P = 0.02). (PPTX 752 kb)
Additional file 6: Figure S5. Relationship between insulin rs689 genotype
and birth weight (sds) and birth length (sds). Birth weight (sds) and birth
length (sds) are independent from the rs 689 genotypes. (PPTX 161 kb)
CBMC: Cord blood mononuclear cells; IGF1: Insulin-like growth factor 1;
IUGR: Intrauterine growth retardation
Fanny Lachaux did the monitoring of Epichild cohort of mothers and children.
We thank the nurses and midwives of the Antoine Béclère Maternity for their
motivation and help.
The study was supported by the INSERM-CEA-Paris Saclay U1169 Research
Unit and the EPICHILD ANR R13107KK.
Availability of data and materials
Data are available from the corresponding author on reasonable request.
PB designed the study, interpreted the data, and wrote the manuscript. CLS
did the experiments and interpreted the results. NT did the statistical analyses.
NF prepared the data bases for final analysis. ALC interpreted the results. CM
and MPB did the experiments. SBT did insulin and IGF1 measurements. AB and
DF recruited subjects and discussed clinical aspects. All authors read and
approved the final manuscript.
Ethics approval and consent to participate
We obtained informed consent from all participating mothers. All protocols
were agreed by French ethic boards (CODECOH DC-2013-2017, CPP CO-14-001,
CCTIRS no. 14-124bis, CNIL no. 914253).
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional affiliations.
1. Gatford KL , Simmons RA . Prenatal programming of insulin secretion in intrauterine growth restriction . Clin Obstet Gynecol . 2013 ; 56 ( 3 ): 520 - 8 .
2. Baker J , Liu JP , Robertson EJ , Efstratiadis A . Role of insulin-like growth factors in embryonic and postnatal growth . Cell . 1993 ; 75 ( 1 ): 73 - 82 .
3. Fowden AL . The insulin-like growth factors and feto-placental growth . Placenta . 2003 ; 24 ( 8-9 ): 803 - 12 .
4. Abuzzahab MJ , Schneider A , Goddard A , Grigorescu F , Lautier C , Keller E , Kiess W , Klammt J , Kratzsch J , Osgood D , et al. IGF-I receptor mutations resulting in intrauterine and postnatal growth retardation . N Engl J Med . 2003 ; 349 ( 23 ): 2211 - 22 .
5. Woods KA , Camacho-Hubner C , Bergman RN , Barter D , Clark AJ , Savage MO . Effects of insulin-like growth factor I (IGF-I) therapy on body composition and insulin resistance in IGF-I gene deletion . J Clin Endocrinol Metab . 2000 ; 85 ( 4 ): 1407 - 11 .
6. Bonapace G , Concolino D , Formicola S , Strisciuglio P. A novel mutation in a patient with insulin-like growth factor 1 (IGF1) deficiency . J Med Genet . 2003 ; 40 ( 12 ): 913 - 7 .
7. Netchine I , Azzi S , Houang M , Seurin D , Perin L , Ricort JM , Daubas C , Legay C , Mester J , Herich R , et al. Partial primary deficiency of insulin-like growth factor (IGF)-I activity associated with IGF1 mutation demonstrates its critical role in growth and brain development . J Clin Endocrinol Metab . 2009 ; 94 ( 10 ): 3913 - 21 .
8. Gicquel C , Rossignol S , Cabrol S , Houang M , Steunou V , Barbu V , Danton F , Thibaud N , Le Merrer M , Burglen L , et al. Epimutation of the telomeric imprinting center region on chromosome 11p15 in Silver-Russell syndrome . Nat Genet . 2005 ; 37 ( 9 ): 1003 - 7 .
9. Begemann M , Zirn B , Santen G , Wirthgen E , Soellner L , Buttel HM , Schweizer R , van Workum W , Binder G , Eggermann T. Paternally inherited IGF2 mutation and growth restriction . N Engl J Med . 2015 ; 373 ( 4 ): 349 - 56 .
10. Stoy J , Edghill EL , Flanagan SE , Ye H , Paz VP , Pluzhnikov A , Below JE , Hayes MG , Cox NJ , Lipkind GM , et al. Insulin gene mutations as a cause of permanent neonatal diabetes . Proc Natl Acad Sci U S A . 2007 ; 104 ( 38 ): 15040 - 4 .
11. Magnus P , Berg K , Bjerkedal T , Nance WE . Parental determinants of birth weight . Clin Genet . 1984 ; 26 ( 5 ): 397 - 405 .
12. Magnus P. Causes of variation in birth weight: a study of offspring of twins . Clin Genet . 1984 ; 25 ( 1 ): 15 - 24 .
13. Horikoshi M , Beaumont RN , Day FR , Warrington NM , Kooijman MN , Fernandez-Tajes J , Feenstra B , van Zuydam NR , Gaulton KJ , Grarup N , et al. Genome-wide associations for birth weight and correlations with adult disease . Nature . 2016 ; 538 ( 7624 ): 248 - 52 .
14. Little RE , Sing CF . Genetic and environmental influences on human birth weight . Am J Hum Genet . 1987 ; 40 ( 6 ): 512 - 26 .
15. Dunger DB , Ong KK , Huxtable SJ , Sherriff A , Woods KA , Ahmed ML , Golding J , Pembrey ME , Ring S , Bennett ST , et al. Association of the INS VNTR with size at birth . ALSPAC Study Team. Avon Longitudinal Study of Pregnancy and Childhood. Nat Genet . 1998 ; 19 ( 1 ): 98 - 100 .
16. Mitchell SM , Hattersley AT , Knight B , Turner T , Metcalf BS , Voss LD , Davies D , McCarthy A , Wilkin TJ , Smith GD , et al. Lack of support for a role of the insulin gene variable number of tandem repeats minisatellite (INS-VNTR) locus in fetal growth or type 2 diabetes-related intermediate traits in United Kingdom populations . J Clin Endocrinol Metab . 2004 ; 89 ( 1 ): 310 - 7 .
17. Johnston LB , Dahlgren J , Leger J , Gelander L , Savage MO , Czernichow P , Wikland KA , Clark AJ . Association between insulin-like growth factor I (IGF-I) polymorphisms, circulating IGF-I, and pre- and postnatal growth in two European small for gestational age populations . J Clin Endocrinol Metab . 2003 ; 88 ( 10 ): 4805 - 10 .
18. Frayling TM , Hattersley AT , Smith GD , Ben-Shlomo Y . Conflicting results on variation in the IGFI gene highlight methodological considerations in the design of genetic association studies . Diabetologia . 2002 ; 45 ( 11 ): 1605 - 6 .
19. Arends N , Johnston L , Hokken-Koelega A , van Duijn C , de Ridder M , Savage M , Clark A . Polymorphism in the IGF-I gene: clinical relevance for short children born small for gestational age (SGA) . J Clin Endocrinol Metab . 2002 ; 87 ( 6 ): 2720 .
20. Lunde A , Melve KK , Gjessing HK , Skjaerven R , Irgens LM . Genetic and environmental influences on birth weight, birth length, head circumference, and gestational age by use of population-based parent-offspring data . Am J Epidemiol . 2007 ; 165 ( 7 ): 734 - 41 .
21. Mathews F , Yudkin P , Neil A . Influence of maternal nutrition on outcome of pregnancy: prospective cohort study . BMJ . 1999 ; 319 ( 7206 ): 339 - 43 .
22. Stein AD , Ravelli AC , Lumey LH . Famine, third-trimester pregnancy weight gain, and intrauterine growth: the Dutch Famine Birth Cohort Study . Hum Biol . 1995 ; 67 ( 1 ): 135 - 50 .
23. Engel SM , Joubert BR , Wu MC , Olshan AF , Haberg SE , Ueland PM , Nystad W , Nilsen RM , Vollset SE , Peddada SD , et al. Neonatal genome-wide methylation patterns in relation to birth weight in the Norwegian Mother and Child Cohort . Am J Epidemiol . 2014 ; 179 ( 7 ): 834 - 42 .
24. Turan N , Ghalwash MF , Katari S , Coutifaris C , Obradovic Z , Sapienza C. DNA methylation differences at growth related genes correlate with birth weight: a molecular signature linked to developmental origins of adult disease ? BMC Med Genet . 2012 ; 5 : 10 .
25. Ouni M , Gunes Y , Belot MP , Castell AL , Fradin D , Bougneres P . The IGF1 P2 promoter is an epigenetic QTL for circulating IGF1 and human growth . Clin Epigenetics . 2015 ; 7 : 22 .
26. Ouni M , Castell AL , Rothenbuhler A , Linglart A , Bougneres P . Higher methylation of the IGF1 P2 promoter is associated with idiopathic short stature . Clin Endocrinol . 2016 ; 84 ( 2 ): 216 - 21 .
27. Kuroda A , Rauch TA , Todorov I , Ku HT , Al-Abdullah IH , Kandeel F , Mullen Y , Pfeifer GP , Ferreri K. Insulin gene expression is regulated by DNA methylation . PLoS One . 2009 ; 4 ( 9 ): e6953 .
28. Fradin D , Le Fur S , Mille C , Naoui N , Groves C , Zelenika D , McCarthy MI , Lathrop M , Bougneres P. Association of the CpG methylation pattern of the proximal insulin gene promoter with type 1 diabetes . PLoS One . 2012 ; 7 ( 5 ): e36278 .
29. Niklasson A , Albertsson-Wikland K . Continuous growth reference from 24th week of gestation to 24 months by gender . BMC Pediatr . 2008 ; 8 : 8 .
30. Le Stunff C , Fallin D , Schork NJ , Bougneres P. The insulin gene VNTR is associated with fasting insulin levels and development of juvenile obesity . Nat Genet . 2000 ; 26 ( 4 ): 444 - 6 .
31. Nawathe AR , Christian M , Kim SH , Johnson M , Savvidou MD , Terzidou V . Insulin-like growth factor axis in pregnancies affected by fetal growth disorders . Clin Epigenetics . 2016 ; 8 : 11 .
32. Iniguez G , Gonzalez CA , Argandona F , Kakarieka E , Johnson MC , Cassorla F. Expression and protein content of IGF-I and IGF-I receptor in placentas from small, adequate and large for gestational age newborns . Horm Res Paediatr . 2010 ; 73 ( 5 ): 320 - 7 .
33. Gordon L , Joo JE , Powell JE , Ollikainen M , Novakovic B , Li X , Andronikos R , Cruickshank MN , Conneely KN , Smith AK , et al. Neonatal DNA methylation profile in human twins is specified by a complex interplay between intrauterine environmental and genetic factors, subject to tissue-specific influence . Genome Res . 2012 ; 22 ( 8 ): 1395 - 406 .
34. Armstrong DA , Lesseur C , Conradt E , Lester BM , Marsit CJ . Global and genespecific DNA methylation across multiple tissues in early infancy: implications for children's health research . FASEB J . 2014 ; 28 ( 5 ): 2088 - 97 .
35. Houseman EA , Accomando WP , Koestler DC , Christensen BC , Marsit CJ , Nelson HH , Wiencke JK , Kelsey KT . DNA methylation arrays as surrogate measures of cell mixture distribution . BMC Bioinformatics . 2012 ; 13 : 86 .
36. Joubert BR , Haberg SE , Nilsen RM , Wang X , Vollset SE , Murphy SK , Huang Z , Hoyo C , Midttun O , Cupul-Uicab LA , 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 - 31 .
37. Caffrey A , McNulty H , Walsh H , Irwin R , Pentieva K. Effect of folic acid supplementation in pregnancy on gene specific DNA methylation in the child: evidence from a randomised controlled trial . Proc Nutr Soc . 2016 ; 75 ( OCE3 ): E132 .