Placenta
Volume 31, Issue 3 , Pages 197-202, March 2010

Decreased Placental Methylation at the H19/IGF2 Imprinting Control Region is Associated with Normotensive Intrauterine Growth Restriction but not Preeclampsia

  • D.K. Bourque

      Affiliations

    • Department of Medical Genetics, University of British Columbia, Vancouver, BC, Canada
    • Child & Family Research Institute, University of British Columbia, Vancouver, BC, Canada
    • ,
  • L. Avila

      Affiliations

    • Department of Medical Genetics, University of British Columbia, Vancouver, BC, Canada
    • Child & Family Research Institute, University of British Columbia, Vancouver, BC, Canada
    • ,
  • M. Peñaherrera

      Affiliations

    • Department of Medical Genetics, University of British Columbia, Vancouver, BC, Canada
    • Child & Family Research Institute, University of British Columbia, Vancouver, BC, Canada
    • ,
  • P. von Dadelszen

      Affiliations

    • Child & Family Research Institute, University of British Columbia, Vancouver, BC, Canada
    • Department of Obstetrics and Gynaecology, Division of Maternal-Fetal Medicine, University of British Columbia, Vancouver, BC, Canada
    • ,
  • W.P. Robinson

      Affiliations

    • Department of Medical Genetics, University of British Columbia, Vancouver, BC, Canada
    • Child & Family Research Institute, University of British Columbia, Vancouver, BC, Canada
    • Corresponding Author InformationCorresponding author. Department of Medical Genetics, University of British Columbia, Child & Family Research Institute, 950 West 28th Avenue, Vancouver, BC V5Z 4H4, Canada. Tel.: +1 604 875 3229.

Accepted 5 December 2009. published online 11 January 2010.

Article Outline

Abstract 

Many genes exhibiting genomic imprinting, parent-of-origin differences in gene expression, are involved in regulating placental and fetal growth. The goal of the present study was to assess whether abnormal regulation of imprinted genes is associated with intrauterine growth restriction (IUGR) and/or preeclampsia (PET).

Methods

Genomic DNA was extracted from at least two whole villi samples from control (N=22), IUGR (N=13), PET (N=17), and PET+IUGR (N=21) placentas. Methylation was assessed using the Illumina GoldenGate Methylation Cancer Panel I array and Pyrosequencing and MS-SNuPE assays.

Results

The 11p15.5 ICR1 (associated with H19 and IGF2) methylation showed considerable intra-placental variability. Nonetheless, average methylation at this site was significantly decreased in normotensive IUGR placentas (p<0.001), but not in any other group. Methylation at ICR2 (KvDMR1; associated with CDKN1C and other maternally expressed 11p15.5 genes) was not significantly altered in any group and no significant changes in expression levels were observed in the genes controlled by this region. There were no significant methylation changes observed in any candidate imprinted gene evaluated by the Illumina array. LINE-1 methylation, a marker of whole genome methylation, was also similar in all groups.

Conclusions

Reduced methylation of ICR1 is associated with normotensive IUGR but not IUGR associated with preeclampsia, suggesting a different etiology of IUGR in this group. A reduction in placental IGF2 could be an adaptive response to restrict fetal growth in the presence of abnormal placentation or a response to poor fetal growth itself.

Keywords: Methylation, IGF2, 11p15.5, H19, Placenta, IUGR, Imprinting, Preeclampsia

 

Back to Article Outline

1. Introduction 

Abnormal placental development is responsible for a wide-range of pregnancy complications including infertility, miscarriage, maternal preeclampsia (PET) and intrauterine growth restriction (IUGR). Preeclampsia affects approximately 5% of pregnancies and can be fatal for both the mother and fetus. The fetal syndrome of preeclampsia may manifest as IUGR [1] and both normotensive IUGR and preeclampsia are associated with a deficiency of extravillous trophoblast (EVT) invasion leading to incomplete remodeling of the maternal spiral arteries [2], [3]. This in turn limits the transfer of nutrients and wastes between the fetus and mother. Placentas from pregnancies associated with maternal preeclampsia also have areas of syncytial knots (clusters of preapoptotic/apoptotic nuclei) and areas of necrosis associated with loss of the syncytial trophoblast microvillous membranes (STBM). Excess STBM are observed in early-onset preeclampsia but not in normotensive IUGR [4]. Although confined placental trisomy can contribute to placental insufficiency in some cases [5], [6] the initiating cause for abnormal trophoblast development in most cases is largely unknown and likely heterogeneous in etiology.

Many imprinted genes, those exhibiting parent-of-origin differences in gene expression, are intimately involved with the regulation of embryonic growth and placental development [7], [8] and disruption of imprinting in mouse models can result in abnormal placental development and fetal growth [9]. Although the placenta plays a critical role in coordinating fetal growth and development, regulation of imprinted gene expression appears to be less stable in the placenta than in the fetus itself. Preimplantation culture of mouse embryos can lead to loss of placental imprinting at multiple genes and is affected by the culture media used [10], [11]. This instability and relaxation of methylation may aid the placenta in adapting to changing physiological conditions. It has thus been hypothesized that sporadic loss-of-imprinting errors could also occur in human placentas and contribute to abnormal fetal growth. However, it has also been suggested that imprinting may be less maintained in human, as compared to mouse placentae [12].

Over 50 imprinted genes have been identified in humans that are distributed in distinct clusters that are regulated by a common imprinting control region (ICR) [13]. Two clusters of imprinted genes within chromosome 11p15.5, each regulated by a separate imprinting control region (ICR1 and ICR2), have been implicated in fetal and placental growth. The paternally expressed insulin-like growth factor 2 (IGF2) and maternally expressed H19 genes are coordinately regulated by a differentially methylated CTCF binding region known as imprinting control region 1 (ICR1) [14]. The H19 gene codes for a non-translated RNA of unclear function, while IGF2 has been implicated in several growth disorders. ICR1 is hypomethylated leading to repression of IGF2 expression in approximately one-third of patients with Silver Russell syndrome (SRS), a syndrome associated with pre- and post-natal growth deficiency [15], whereas it is hypermethylated leading to an increase in expression of IGF2 in some cases of pre- and post-natal overgrowth diagnosed as Beckwith–Wiedemann syndrome (BWS) [16], [17]. Complete loss of Igf2 expression in the mouse placenta results in severe placental and fetal growth restriction. Recently, a decrease in IGF2 expression was reported in small for gestational age placentas as compared to control placentas [18].

Imprinting control region 2 (KvDMR1), located centromeric to ICR1, is normally methylated on the maternal allele. Loss of methylation at this region has been reported in BWS patients with normal ICR1 methylation [19] and a maternally inherited duplication with gain of methylation of this region has been reported in a patient with SRS [20]. Loss of methylation at ICR2 can result in decreased cyclin-dependent kinase inhibitor 1C (CDKN1C) expression. Loss of Cdkn1c expression in fetal mice correlates with some phenotypes of BWS and preeclampsia in humans [21] and approximately 40% of BWS familial cases involve loss of function mutations in CDKN1C [22].

Despite improved understanding of the fundamental role that is played by genomic imprinting in the regulation of placental function in the mouse, current knowledge of imprinting in human placental development is poor [23]. The goal of the present study is to assess the role of aberrant DNA methylation associated with imprinted genes, particularly involving ICR1 and ICR2 within 11p15.5, in human placentas from pregnancies associated with preeclampsia and/or IUGR.

Back to Article Outline

2. Methods 

2.1. Sample ascertainment 

Pregnancies were prospectively ascertained through poster recruitment (hospital, midwives and doctors' offices) or through referral to the EMMA (Evaluating Maternal Markers of Acquired Risk of Preeclampsia) Clinic, BC Women's Hospital. Women are seen in this clinic if they have at least one of: past history of preeclampsia (severe, early onset and/or associated with perinatal loss), pre-existing hypertension, unexplained low first trimester PAPP-A (<0.60 multiples of the median [MoM]), unexplained elevated second trimester alpha-fetoprotein (AFP; >2.5MoM) or human chorionic gonadotrophin (HCG;>3.0MoM).

Ethics approval for this study was obtained through the University of British Columbia and the BC Children's & Women's Hospital ethics boards. Consent was obtained during pregnancy. Placentas were obtained at birth and assigned to control (N=22, mean gestational age (GA)=39.0 weeks), intrauterine growth restricted only (IUGR, N=13, mean GA=35.4 weeks), preeclampsia (PET, N=17, mean GA=35.9 weeks), and PET with IUGR (P+I, N=21, mean GA=32.5 weeks) outcomes. Intrauterine growth restriction was defined as either (1) birth weight <3rd percentile for gender and gestational age using Canadian charts [24], or (2) birth weight <10th percentile with either: (a) persistent uterine artery notching at 22+0 to 24+6 weeks gestation, (b) absent or reversed end diastolic velocity on umbilical artery Doppler, and/or (c) oligohydramnios (amniotic fluid index <50mm). Preeclampsia was defined as: (1) at least two of the following: hypertension (sBP ≥140mmHg and/or dBP ≥90mmHg, twice, >4h apart) after 20 weeks, and proteinuria defined as ≥0.3g/d or ≥2 + dipstick proteinuria after 20 weeks [25], (2) non-hypertensive and non-proteinuric HELLP syndrome, using Sibai's criteria [26], or (3) an isolated eclamptic seizure without preceding hypertension or proteinuria, using the British Eclampsia Survey Team (BEST) criteria to define eclampsia [27] (see Supplementary Table 1 for additional clinical information).

Two distinct placental samples of ∼1cm3 were biopsied (one near the cord insertion and one near the placental periphery) from the fetal side of each placenta for DNA extraction. From each site, the surface layers of amnion and chorion were removed before DNA and RNA extraction. Extraction of DNA and RNA was performed using standard techniques. For methylation studies, approximately 300ng of DNA was treated with sodium bisulfite (EZ DNA Methylation-Gold™, Zymo Research, Orange, CA, USA).

All placentas were evaluated for the presence of placental trisomy using comparative genomic hybridization [28] and any identified trisomy was confirmed using microsatellite markers. Trisomy was observed in four placentas (Control PM65: 47,XXX; IUGR PM41: 47,XX,+7 and PM72: 46,XX/47,XX,+13; and PET+IUGR PM60: 47,XX,+2) [53]. Trisomy was also present in the amnion for case PM65 but not for cases PM41, PM72 or PM60, indicating that in the latter three cases the trisomy was likely confined to the placenta.

2.2. Whole genome methylation arrays 

GoldenGate Methylation Cancer Panel 1 arrays (Illumina Inc., San Diego, CA) were used to identify candidate imprinted genes that merit further study. Two independent villous samples were analyzed from each placenta (control, N=5 placentas; IUGR, N=5; and PET, N=4) for a total of 28 analyzed samples. Parameters for differential methylation analysis were as follows: normalization=average; reference group=control placentas; error model=t-test. The BeadStudio software uses diffscores to represent expression differences between groups. Samples with a diffscore of greater than ±13 correspond to a nominal (uncorrected) significance p<0.05 (diffscore = |10log pval|). Negative diffscores reflect hypomethylation compared to the control group and positive diffscores reflect hypermethylation compared to the control group. The Benjamini-Hochberg correction was used to correct for multiple comparisons and results were then further analyzed using the Significance Analysis of Microarrays (SAM) software [29].

2.3. Methylation-sensitive single nucleotide primer extension (Ms-SNuPE) for ICR1 

Bisulfite converted DNA was PCR amplified using primers F6005 and R6326 and PCR as previously reported [30]. Each 10μL PCR contained: 1X Rose Taq buffer (including 2mM MgCl2), 0.125mM dNTP, 3pmol of each primer, 0.1U Rose Taq (Rose Scientific, Edmonton, AB, Canada), and 2μL bisulfite converted DNA, with an initial denaturation at 94°C, 10min; 30 cycles of 94°C for 45s, 61°C for 45s, 72°C for 1min; a final extension at 72°C for 10min. The first PCR was followed by a 20μL semi-nested PCR of 35 cycles using primers F6115 and R6326 using 1μL of PCR product from the first reaction. PCR products were cleaned using DNA Clean & Concentrator™-5 (Zymo Research, Orange, CA, USA).

Methylation-sensitive Single Nucleotide Primer Extension (Ms-SNuPE) was used to assess methylation at two CpGs (C10 and C12) within the 6th CTCF binding site of ICR1 that were previously identified as being differentially methylated by parental origin and representative of the region [31], [32]. SNaPshot Multiplex Kit (Applied Biosystems, Foster City, CA, USA) was used to according to manufacturer's directions. The reaction was terminated by dephosphorylation using 1U of calf intestinal phosphatase (Invitrogen, Carlsbad, CA, USA) and incubation at 37°C for 1h followed by deactivation of the enzyme at 72°C for 15min. Products were sized and quantified on an ABI Prism 310 Genetic Analyzer. Using this method, we previously observed a strong correlation between estimated methylation values for the independent PCR assays of C10 and C12 in blood samples (r=0.95, p<0.0001, N=87) and for repeat estimates from distinct bisulfite conversions (r=0.8, p<0.0001, N=93) [33] and have applied this assay to diagnosis of hypomethylation in Silver–Russell Syndrome patients. Assessment of methylation at ICR1 by pyrosequencing was not performed as a published assay for this region [34] was found to span the rs2107425 polymorphism, and further assays we attempted showed an amplification bias in some individuals based on an SNP within the amplified region (rs10732516). No amplification bias or association of methylation with local sequence variation was detected with the Ms-SNuPE assay (data not shown).

2.4. Pyrosequencing for ICR2, candidate gene and LINE-1 methylation 

Pyrosequencing was used to assess methylation at seven CpGs within ICR2 including the differentially methylated NotI site that is often altered in BWS [19], [35] and is used in diagnostic testing for BWS [17]. Bisulfite converted DNA was amplified by PCR (see Supplementary Table 2 for primer sequences). Each 25μL PCR contained: 1X HotStarTaq buffer (including 1.5mM MgCl2), 0.2mM dNTP, 5pmol of each primer, 1.0U HotStarTaq DNA Polymerase (QIAGEN Inc., Mississauga, ON, Canada), and 2μL bisulfite converted DNA. Thermocycling conditions included: an initial denaturation at 95°C for 10min; 40 cycles of 95°C for 40s, 55°C for 40s, 72°C for 40s; a final extension at 72°C for 7min. Sequencing of PCR products (10μL) using PyroMark™ MD (Biotage AB, Uppsala, Sweden) was performed according to manufacturer's directions Pyro Q-CpG software (v1.0.9, 2006, Biotage AB) was used to analyze results.

Methylation at 13 CpG sites within the CDKN1C promoter was assessed using a pyrosequencing assay available from the Biotage PyroMark™ Assay Database. Pyrosequencing assays for the following candidate genes were also developed: H19 promoter, PEG10 promoter, PLAGL1 promoter, SNRPN promoter, MEST exon 1 (see Supplementary Table 2 for primer sequences). The PCR and thermocycling conditions were identical to those described above. Methylation at 7 CpG sites from a consensus sequence found within LINE-1 elements was also preformed according to manufacturer's directions (PyroMark™ LINE-1 Kit, Biotage AB).

2.5. Whole genome expression arrays 

HumanRef-8 v2 BeadChip gene expression arrays (Illumina Inc., San Diego, CA) were used to correlate methylation patterns with gene expression. Total mRNA was extracted from a subset of the placentas using an RNeasy® kit (QIAGEN). Two independent villous samples were analyzed from each placenta (control, N=5 placentas; IUGR, N=5; and PET, N=4) for a total of 28 analyzed samples. The samples were chosen because RNA was available and the placentas were processed soon after delivery. The BeadChip array was processed in the Centre for Molecular Medicine and Therapeutics (CMMT) BioAnalyzer Core Facility (Vancouver, BC, Canada). Output was analyzed using Illumina's BeadStudio software (v3.2.7, 2007). Parameters for differential expression analysis were as follows: normalization=average; reference group=control placentas; error model=t-test. Samples with a diffscore of greater than ±13 correspond to a nominal (uncorrected) significance p<0.05 (diffscore = |10log pval|). Negative diffscores reflect underexpression compared to the control group and positive diffscores reflect overexpression compared to the control group. The Benjamini-Hochberg correction was used to correct for multiple comparisions and results were then further analyzed using the SAM software [29].

Back to Article Outline

3. Results 

3.1. Illumina methylation results 

Although 1505 CpG sites are assayed on the GoldenGate Methylation Cancer Panel, only 70 are located within the promoter region of imprinted genes (Supplementary Table 3). None of the differentially methylated ICRs are included on this array. After analysis with SAM and correction for multiple comparisons, no CpGs in either the IUGR or PET groups were found to be significantly altered.

3.2. ICR1 methylation by MS-SNuPE 

Methylation at ICR1 (associated with IGF2 expression) was quantified at two CpG sites (C10 and C12) from each of two sampling sites within each placenta. There was a strong correlation between C10 and C12 methylation levels from a single placental sample (r=0.82, p<0.0001). However, this correlation was weaker than the C10-C12 correlation observed in blood samples analyzed in the same laboratory by this method (r=0.95, p<0.0001N=87) [33] which suggests that methylation of these sites is more variable in the placenta than in peripheral blood. The within-placenta between–site correlation was r=0.47 for C10, r=0.48 for C12 and was r=0.56 when comparing the average C10 and C12 methylation for each site (p<0.0001 for each correlation) (Fig. 1). To obtain a methylation value representative of the whole placenta, it is thus important to average data from multiple sites. For subsequent comparisons, methylation values were averaged across the two CpGs and two sampling sites to obtain a single methylation value for each placenta.

  • View full-size image.
  • Fig. 1 

    Intra-placental correlation for percent methylation at the H19/IGF2 ICR. The methylation values as measured by SNuPE from two separate sampling sites from one placenta were correlated (r=0.56, p<0.0001). Percent methylation at C10 and C12 were averaged to obtain a single methylation value for each sample.

The mean ICR1 methylation values for each clinical group were: controls (N=22), 36.7%; IUGR (N=13), 30.8%; PET (N=17), 38.3%; PET+IUGR (N=21), 37.6% (Fig. 2, Table 1). Methylation at ICR1 showed significant between group differences (p<0.001, one-way ANOVA). This effect was due to a reduction of methylation in the IUGR group compared to all other groups (p<0.0001, compared to controls) and 7 of the 13 placentas in the IUGR group had methylation values at least 2 SD below the mean of the control group. These seven placentas were: PM30, PM35, PM41, PM47, PM123, PM128, PM120. There was no difference between mean methylation in PET, with or without IUGR, and the control group. There were also no significant differences in ICR1 methylation when considering early onset (N=6) and late onset (N=11) preeclampsia separately (not shown). There was no correlation between methylation and sex, gestational age, time to placental sampling after birth, mode of delivery, oligohydramnios, symmetrical vs. asymmetrical IUGR, maternal gestational diabetes mellitus or the presence of placental trisomy. There was a significant correlation between methylation and gestational age corrected birth weight (measured in SD relative to the mean) (r=0.29, p=0.015), which was more pronounced when the cases with preeclampsia were excluded from the analysis (r=0.62, p<0.0001). Nonetheless, a similar correlation between methylation and GA-corrected birth weight was present within the preeclampsia (with or without IUGR) group analyzed separately (r=0.29, p=0.038). It appears that a greater average methylation in the preeclampsia group overall is confounding the association with IUGR in comparisons with control placentas.

  • View full-size image.
  • Fig. 2 

    Inter-group comparisons for percent methylation at the H19/IGF2 ICR. There is a significant reduction in mean methylation (average of two sites measured by SNuPE) in placental villi associated with normotensive IUGR compared to controls (p<0.01), PET (p<0.01), and PET+IUGR (p<0.01) pregnancies.

Table 1. Summary of mean methylation values obtained using MS-SNuPE (ICR1) and pyrosequencing (ICR2, H19, CDKN1C, PEG10, PLAGL1, SNRPN, MEST, LINE-1). ICR1 methylation was significantly different between groups (P<0.0001) due to reduced methylation in villi from IUGR placentas compared to each of the other groups. No other groups were significantly altered.
ControlIUGRPETPET+IUGR
VilliN=22N=13N=17N=21
ICR136.7±3.0% (S.E. 0.6%)30.8±3.2% (S.E. 1.0%)∗38.3±4.0% (S.E. 1.0%)37.6±3.8% (S.E. 0.8%)
ICR265.4±3.4% (S.E. 0.7%)64.8±3.2% (S.E. 0.9%)65.2±3.0% (S.E. 0.7%)64.7±3.3% (S.E. 0.7%)
H19 promoter49.5±4.4% (S.E. 0.9%)48.9±3.3% (S.E. 0.9%)49.3±4.2% (S.E. 1.0%)48.5±6.8% (S.E. 1.5%)
CDKN1C promoter5.2±2.8% (S.E. 0.7%)4.5±0.6% (S.E. 0.6%)7.0±5.5% (S.E. 1.7%)6.9±3.3% (S.E. 0.9%)
PEG10 promoter57.1±5.8% (S.E. 1.2%)56.6±11.9% (S.E. 3.3%)57.9±8.0% (S.E. 1.9%)57.9±6.4% (S.E. 1.5%)
PLAGL1 promoter52.6±2.6% (S.E. 0.5%)53.0±2.1% (S.E. 0.9%)54.2±6.6% (S.E. 1.8%)53.7±3.9% (S.E. 0.8%)
SNRPN promoter46.3±3.0% (S.E. 0.6%)50.7±4.8% (S.E. 1.3%)47.8±4.1% (S.E. 1.0%)51.4±5.8% (S.E. 1.3%)
MEST exon 158.9±9.5% (S.E. 2.0%)60.1±8.4% (S.E. 2.3%)59.3±11.0% (S.E. 2.7%)59.7±8.8% (S.E. 2.0%)
LINE-149.6±2.0% (S.E. 0.6%)50.0±2.0%, (S.E. 0.7%)48.5±2.4% (S.E. 0.7%)51.0±4.6% (S.E. 1.3%)

AmnionN=5N=5N=5N=4
ICR135.8±2.9% (S.E. 1.3%)36.4±3.4%, (S.E. 1.7%)34.5±3.3% (S.E. 1.5%)37.2±1.4% (S.E. 0.7%)

∗p<0.001 compared to control (t-test).

To determine if reduced methylation was restricted to chorionic villi, ICR1 methylation was assessed in a subset of amnion samples. There was no significant difference between methylation at ICR1 within the amnion in any of the clinical groups and specifically no reduction in methylation in the IUGR group as compared to the control group (Table 1). In addition, chorionic villi from five control and five IUGR group placentas were separated into trophoblast and mesenchyme by enzymatic digestion. Mean ICR1 methylation was similar between cell-types in both the IUGR and control groups and there was not a significant correlation between methylation level of trophoblast and mesenchyme from a single site (r=0.14), though sample size was likely too small to evaluate such an effect.

3.3. ICR2, candidate gene, and LINE1 methylation by pyrosequencing 

Methylation at ICR2 was quantified at seven adjacent CpG sites, including those in the BWS diagnostic NotI site. There was a correlation in methylation levels between different CpGs from the same sample and for average methylation between placental sites (r=0.38, p=0.001). As for ICR1, methylation across the CpGs and the two placental sampling sites was averaged to give one measurement per placenta. No significant differences in ICR2 methylation were detected between any of the clinical groups or between early and late onset preeclampsia (Table 1).

Methylation level at 13 CpG sites within the CDKN1C promoter was very low and no significant differences were detected between any of the groups. Whole blood and saliva were also tested with this method and also showed very little methylation; however, promoter regions associated with CpG islands are frequently unmethylated [36]. None of the other candidate genes were significantly altered between the groups, including H19 (Table 1). As an indirect marker of genome-wide methylation, methylation status was also evaluated at a consensus sequence within LINE-1 elements, which make up about 15% of the genome. The mean methylation values for each clinical group were all approximately 50% (Table 1). It thus appears that the altered methylation at ICR1 does not likely stem from a general hypomethylation of the genome.

3.4. Illumina expression results 

Although over 22,000 transcripts are assayed on the Illumina HumanRef-8 v2 BeadChip, for the present study we only considered the expression level of 44 reported imprinted genes (59 transcripts) (Supplementary Table 4). Among these, only IGF2 (transcripts NR_003512.1, NM_001007139.4, NM_00612.4) was significantly underexpressed in placental villi associated with IUGR compared to controls (average expression level of 16,700 vs 35,000, p<0.0001). Other imprinted genes had decreased expression (e.g. CDKN1C in PET) or increased expression (MEST in IUGR and PET, SNRPN in IUGR); however, these were not significant when analyzed by SAM. In addition, only IGF2 was significantly altered after using the Benjamini-Hochberg correction. This result was not supported by quantitative real time PCR using either β-actin or β-2-microglobulin as an endogenous control; however, other groups have reported both increases [37] and decreases [18] in IGF2 expression in IUGR placentas. Unlike DNA methylation, which is relatively stable, placental RNA degrades quite rapidly after birth. Most of our placentas were >6h post-birth at the time of sampling, and thus the mRNA data must be interpreted cautiously.

Back to Article Outline

4. Discussion 

The placenta is a remarkably adaptive organ that mediates the exchange of nutrients between two genetically distinct individuals, the mother and the fetus, which may have conflicting needs [38]. Many imprinted genes studied in the mouse appear to regulate fetal growth in a manner that maintains a balance between maternal nutrient supply and fetal growth [8], [38], [39]. Altered expression of both IGF2 and PHLDA2 have been reported in human placentas associated with SGA or IUGR births[18], [40], [41]. However, assessment of DNA methylation provides an independent tool for the assessment of the placental genome. DNA methylation at critical sites can reflect the availability of a gene for transcription, which may lead to altered expression depending on other regulatory factors present. DNA methylation is largely programmed during cell differentiation and early development, and may be more resistant to the transient changes in gene expression associated with labor and delivery [42], [43], [44], as well as the effects of sample processing, as has been shown to influence RNA quality [45].

Although both IUGR and PET associated placentas show similar deficiencies in trophoblast invasion, only normotensive IUGR associated placentas showed reduced methylation at ICR1 in this study as compared to controls. We found no effect of gestational age, mode of delivery, or placental processing time on methylation at ICR1. The reduced methylation associated with IUGR may reflect an adaptive process serving to adjust placental and fetal growth in response to poor placental perfusion and prevent maternal preeclampsia. Consistent with this possibility, average methylation in preeclampsia tended to be higher than controls, particularly in the absence of IUGR. This supports the idea that failure to limit fetal growth in the presence of poor placental perfusion could in turn contribute to the development of maternal preeclampsia. The presence of high levels of trisomy in two of the IUGR associated placentas, which is likely the initial cause of placental dysfunction, is consistent with the hypothesis that reduced methylation at ICR1 may be a consequence of other placental abnormalities rather than a spontaneous defect. Methylation at ICR1 has been shown in a number of studies to be particularly responsive to environmental influences such as culture media [10], [11], environmental toxins (e.g. TCDD) [46] and prenatal ethanol exposure [47].

Previous studies have reported a reduction in IGF2 expression in placentas from pregnancies associated with IUGR or SGA [18], [40] and complete loss of placental Igf2 expression is associated with fetal growth restriction in mice [7]. Furthermore, selective deletion of the placental specific form of Igf2 from murine placentas also leads to a significant decrease in fetal weight, with pups being 69% of normal weight at birth [48]. Reduced placental Igf2 expression leads to a reduction in size of all placental layers and alters the diffusional exchange characteristics of the placenta [49]. In human pregnancies, reduced exchange surface area, and likely reduced transfer capacity of the placenta, has been noted in IUGR [50]. Altered placental transfer to the fetus may also be a mechanism involved in the pathogenesis of IUGR, as the developing fetus will not be able to receive adequate nutrition to allow for normal growth.

Although a reduction in IGF2 expression was observed in SGA placentas in a recent report, loss of methylation at ICR1 was not observed in the same placentas [18]. The differences in methylation values between the two studies may reflect differences in patient ascertainment or sampling procedures. An increase in average methylation of ICR1 in preeclamptic placentas may confound the relationship of decreased methylation with IUGR, if these placentas are not analyzed separately. While there is much overlap between cases diagnosed as SGA or IUGR, these are different diagnostic criteria, and in our study cases with low birth weight were required to show other prenatal indicators of poor placental function to be classified as IUGR. Furthermore, we removed the amnion and chorion from the villous sample prior to DNA extraction. Including amnion could dilute the methylation effect, as we observed normal methylation in amnion even when reduced in the placental villi. Another study found biallelic expression and loss of imprinting at H19 in placentas from preeclamptic women [51]; however, our results do not support these findings.

In the present study we did not find evidence for altered methylation at ICR2 (KvDMR1), nor specifically at CDKN1C, a gene associated with altered growth in BWS. Very little methylation (∼5%) was detected at CDKN1C despite being reported to be differentially methylated in murine placenta [52], and some imprinted genes, including those within the ICR2 cluster, may not be imprinted in the human term placenta [12]. The mouse knockout of CDKN1C displays some phenotypes of preeclampsia and BWS [21], and a modest reduction in CDKN1C expression was observed in both the IUGR and PET group.

DNA methylation at ICR1 is well established to affect IGF2 expression and is routinely used in the clinical diagnosis of post-natal fetal growth disorders [17], [20], [33]. It may also provide a useful diagnostic tool for assessing the etiology of IUGR. Further studies will be necessary to determine if reduced methylation at ICR1 associated with normotensive IUGR is an early or late event in IUGR and, thus, could provide any prognostic value to the pregnancy. If altered methylation leads to a reduction of IGF2 expression as a compensatory response to other factors such as poor placental perfusion, it may be a beneficial to the mother. A full understanding of the genetic and or environmental conditions leading to reduced IGF2 in some pregnancies with abnormal trophoblast invasion and not in others (i.e. those with preeclampsia) will be important before any therapies attempting to improve fetal growth are initiated.

Back to Article Outline

Acknowledgements 

We would like to thank Ryan Yuen, Ruby Jiang, Courtney Hanna and the other members of the Robinson Lab for their assistance and Dr. Angela Devlin for use of the Biotage PyroMark™ MD. We would also like to thank our participants, as our work would not be possible without them. This work has been funded by the Canadian Institutes for Health Research (to WPR) and by graduate student scholarships from the Canadian Institutes for Health Research and Michael Smith Foundation for Health Research (to DKB).

Back to Article Outline

Appendix. Supplementary data 

play application

download application

Back to Article Outline

References 

  1. Sibai B, Dekker G, Kupferminc M. Pre-eclampsia. Lancet. 2005;365:785–799
  2. Kaufmann P, Black S, Huppertz B. Endovascular trophoblast invasion: implications for the pathogenesis of intrauterine growth retardation and preeclampsia. Biol Reprod. 2003;69:1–7
  3. Granger JP, Alexander BT, Llinas MT, Bennett WA, Khalil RA. Pathophysiology of hypertension during preeclampsia linking placental ischemia with endothelial dysfunction. Hypertension. 2001;38:718–722
  4. Goswami D, Tannetta DS, Magee LA, Fuchisawa A, Redman CW, Sargent IL, et al. Excess syncytiotrophoblast microparticle shedding is a feature of early-onset pre-eclampsia, but not normotensive intrauterine growth restriction. Placenta. 2006;27:56–61
  5. Kalousek DK, Vekemans M. Confined placental mosaicism. J Med Genet. 1996;33:529–533
  6. Yong PJ, Langlois S, von Dadelszen P, Robinson WP. The association between preeclampsia and placental trisomy 16 mosaicism. Prenat Diagn. 2006;26:956–961
  7. Fowden AL, Sibley C, Reik W, Constancia M. Imprinted genes, placental development and fetal growth. Horm Res. 2006;65:50–58
  8. Angiolini E, Fowden A, Coan P, Sandovici I, Smith P, Dean W, et al. Regulation of placental efficiency for nutrient transport by imprinted genes. Placenta. 2006;27:S98–S102
  9. Morgan HD, Santos F, Green K, Dean W, Reik W. Epigenetic reprogramming in mammals. Hum Mol Genet. 2005;14:R47–R58
  10. Doherty AS, Mann MR, Tremblay KD, Bartolomei MS, Schultz RM. Differential effects of culture on imprinted H19 expression in the preimplantation mouse embryo. Biol Reprod. 2000;62:1526–1535
  11. Mann MRW, Lee SS, Doherty AS, Verona RI, Nolen LD, Schultz RM, et al. Selective loss of imprinting in the placenta following preimplantation development in culture. Development. 2004;131:3727–3735
  12. Monk D, Arnaud P, Apostolidou S, Hills FA, Kelsey G, Stanier P, et al. Limited evolutionary conservation of imprinting in the human placenta. Proc Natl Acad Sci U S A. 2006;103:6623–6628
  13. Reik W, Walter J. Genomic imprinting: parental influence on the genome. Nat Rev Genet. 2001;2:21–32
  14. Engel N, West AG, Felsenfeld G, Bartolomei MS. Antagonism between DNA hypermethylation and enhancer-blocking activity at the H19 DMD is uncovered by CpG mutations. Nat Genet. 2004;36:883–888
  15. Gicquel C, Rossignol S, Cabrol S, Houang M, Steunou V, Barbu V, et al. Epimutation of the telomeric imprinting center region on chromosome 11p15 in Silver–Russell syndrome. Nat Genet. 2005;37:1003–1007
  16. Weksberg R, Smith AC, Squire J, Sadowski PD. Beckwith–Wiedemann syndrome demonstrates a role for epigenetic control of normal development. Hum Mol Genet. 2003;12:R61–R68
  17. Weksberg R, Shuman C, Smith AC. Beckwith–Wiedemann syndrome. Am J Med Genet C Semin Med Genet. 2005;137:12–23
  18. Guo L, Choufani S, Ferreira J, Smith A, Chitayat D, Shuman C, et al. Altered gene expression and methylation of the human chromosome 11 imprinted region in small for gestational age (SGA) placentae. Dev Biol. 2008;320:79–91
  19. Smilinich NJ, Day CD, Fitzpatrick GV, Caldwell GM, Lossie AC, Cooper PR, et al. A maternally methylated CpG island in KvLQT1 is associated with an antisense paternal transcript and loss of imprinting in Beckwith–Wiedemann syndrome. Proc Natl Acad Sci U S A. 1999;96:8064–8069
  20. Schonherr N, Meyer E, Roos A, Schmidt A, Wollmann HA, Eggermann T. The centromeric 11p15 imprinting centre is also involved in Silver–Russell syndrome. J Med Genet. 2007;44:59–63
  21. Kanayama N, Takahashi K, Matsuura T, Sugimura M, Kobayashi T, Moniwa N, et al. Deficiency in p57Kip2 expression induces preeclampsia-like symptoms in mice. Mol Hum Reprod. 2002;8:1129–1135
  22. Maher ER, Reik W. Beckwith–Wiedemann syndrome: imprinting in clusters revisited. J Clin Invest. 2000;105:247–252
  23. Wagschal A, Feil R. Genomic imprinting in the placenta. Cytogenet Genome Res. 2006;113:90–98
  24. Kramer MS, Platt RW, Wen SW, Joseph KS, Allen A, Abrahamowicz M, et al. A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics. 2001;108:e35
  25. Magee LA, Helewa M, Moutquin JM, von Dadelszen P. Diagnosis, evaluation, and management of the hypertensive disorders of pregnancy. J Obstet Gynaecol Can. 2008;30:S1–S48
  26. Audibert F, Friedman SA, Frangieh AY, Sibai BM. Clinical utility of strict diagnostic criteria for the HELLP (hemolysis, elevated liver enzymes, and low platelets) syndrome. Am J Obstet Gynecol. 1996;175:460–464
  27. Douglas KA, Redman CWG. Eclampsia in the United Kingdom. BMJ. 1994;309:1395–1400
  28. Lestou VS, Lomax BL, Barrett IJ, Kalousek DK. Screening of human placentas for chromosomal mosaicism using comparative genomic hybridization. Teratology. 1999;59:325–330
  29. Tusher VG, Tibshirani R, Chu G. Significance analysis of microarrays applied to the ionizing radiation response. Proc Natl Acad Sci U S A. 2001;98:5116–5121
  30. Kerjean A, Dupont J-, Vasseur C, Le Tessier D, Cuisset L, Paldi A, et al. Establishment of the paternal methylation imprint of the human H19 and MEST/PEG1 genes during spermatogenesis. Hum Mol Genet. 2000;9:2183–2187
  31. Sievers S, Alemazkour K, Zahn S, Perlman EJ, Gillis AJ, Looijenga LH, et al. IGF2/H19 imprinting analysis of human germ cell tumors (GCTs) using the methylation-sensitive single-nucleotide primer extension method reflects the origin of GCTs in different stages of primordial germ cell development. Genes Chromosomes Cancer. 2005;44:256–264
  32. Takai D, Gonzales FA, Tsai YC, Thayer MJ, Jones PA. Large scale mapping of methylcytosines in CTCF-binding sites in the human H19 promoter and aberrant hypomethylation in human bladder cancer. Hum Mol Genet. 2001;10:2619–2626
  33. Penaherrera MS, Weindler S, Van Allen M, Yong SL, Metzger D, McGillivray B, et al. Methylation profiling in individuals with Silver–Russell Syndrome. American Journal of Medical Genetics, in press.
  34. Byun HM, Wong HL, Birnstein EA, Wolff EM, Liang G, Yang AS. Examination of IGF2 and H19 loss of imprinting in bladder cancer. Cancer Res. 2007;67:10753–10758
  35. Weksberg R, Nishikawa J, Caluseriu O, Fei Y-, Shuman C, Wei C, et al. Tumor development in the Beckwith–Wiedemann syndrome is associated with a variety of constitutional molecular 11p15 alterations including imprinting defects of KCNQ1OT1. Hum Mol Genet. 2001;10:2989–3000
  36. Jones PA, Takai D. The role of DNA methylation in mammalian epigenetics. Science. 2001;293:1068–1070
  37. Abu-Amero SN, Ali Z, Bennett P, Vaughan JI, Moore GE. Expression of the insulin-like growth factors and their receptors in term placentas: a comparison between normal and IUGR births. Mol Reprod Dev. 1998;49:229–235
  38. Reik W, Constância M, Fowden A, Anderson N, Dean W, Ferguson-Smith A, et al. Regulation of supply and demand for maternal nutrients in mammals by imprinted genes. J Physiol. 2003;547:35–44
  39. Constancia M, Kelsey G, Reik W. Resourceful imprinting. Nature. 2004;432:53–57
  40. McMinn J, Wei M, Schupf N, Cusmai J, Johnson EB, Smith AC, et al. Unbalanced placental expression of imprinted genes in human intrauterine growth restriction. Placenta. 2006;27:540–549
  41. Apostolidou S, Abu-Amero S, O'Donoghue K, Frost J, Olafsdottir O, Chavele KM, et al. Elevated placental expression of the imprinted PHLDA2 gene is associated with low birth weight. J Mol Med. 2007;85:379–387
  42. Torricelli M, Giovannelli A, Leucci E, De Falco G, Reis FM, Imperatore A, et al. Labor (term and preterm) is associated with changes in the placental mRNA expression of corticotrophin-releasing factor. Reprod Sci. 2007;14:241–245
  43. Torricelli M, Giovannelli A, Leucci E, Florio P, De Falco G, Torres PB, et al. Placental neurokinin B mRNA expression increases at preterm labor. Placenta. 2007;28:1020–1023
  44. Torricelli M, Galleri L, Voltolini C, Biliotti G, Florio P, De Bonis M, et al. Changes of placental Kiss-1 mRNA expression and maternal/cord kisspeptin levels at preterm delivery. Reprod Sci. 2008;15:779–784
  45. Fajardy I, Moitrot E, Vambergue A, Vandersippe-Millot M, Deruelle P, Rousseaux J. Time course analysis of RNA stability in human placenta. BMC Mol Biol. 2009;10:21
  46. Wu Q, Ohsako S, Ishimura R, Suzuki JS, Tohyama C. Exposure of mouse preimplantation embryos to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) alters the methylation status of imprinted genes H19 and Igf2. Biol Reprod. 2004;70:1790–1797
  47. Haycock PC, Ramsay M. Exposure of mouse embryos to ethanol during preimplantation development: effect on DNA methylation in the h19 imprinting control region. Biol Reprod. 2009;81:618–627
  48. Constancia M, Hemberger M, Hughes J, Dean W, Ferguson-Smith A, Fundele R, et al. Placental-specific IGF-II is a major modulator of placental and fetal growth. Nature. 2002;417:945–948
  49. Sibley CP, Coan PM, Ferguson-Smith AC, Dean W, Hughes J, Smith P, et al. Placental-specific insulin-like growth factor 2 (Igf2) regulates the diffusional exchange characteristics of the mouse placenta. Proc Natl Acad Sci U S A. 2004;101:8204–8208
  50. Mayhew TM, Ohadike C, Baker PN, Crocker IP, Mitchell C, Ong SS. Stereological investigation of placental morphology in pregnancies complicated by pre-eclampsia with and without intrauterine growth restriction. Placenta. 2003;24:219–226
  51. Yu L, Chen M, Zhao D, Yi P, Lu L, Han J, et al. The H19 gene imprinting in normal pregnancy and pre-eclampsia. Placenta. 2009;30:443–447
  52. Lewis A, Mitsuya K, Umlauf D, Smith P, Dean W, Walter J, et al. Imprinting on distal chromosome 7 in the placenta involves repressive histone methylation independent of DNA methylation. Nat Genet. 2004;36:1291–1295
  53. Robinson WP, Peñaherrera MS, Jiang R, Avila L, Sloan J, McFadden DE, et al. Assessing the role of placental trisomy in preeclampsia and intrauterine growth restriction. Prenat Diagn. 2010;30:1–8

PII: S0143-4004(09)00393-2

doi:10.1016/j.placenta.2009.12.003

Placenta
Volume 31, Issue 3 , Pages 197-202, March 2010

Article Tools

* Image Usage & Resolution