Although an essential component of assisted reproductive technologies, ovarian stimulation, or superovulation, may interfere with the epigenetic reprogramming machinery during early embryogenesis and gametogenesis. To investigate the possible impact of superovulation particularly on the methylation reprogramming process directly after fertilization, we performed immunofluorescence staining of pronuclear (PN) stage embryos with antibodies against 5mC and 5hmC. PN stage embryos obtained by superovulation displayed an increased incidence of abnormal methylation and hydroxymethylation patterns in both maternal and paternal pronuclear DNA. Subsequent single-cell RT-qPCR analyses of the Tet1, Tet2, and Tet3 genes revealed no significant expression differences between PN stage embryos from spontaneously and superovulated matings that could be causative for the abnormal methylation and hydroxymethylation patterns. To analyze the possible contribution of TET-independent replication-associated demethylation mechanisms, we then determined the 5mC and 5hmC levels of PN stage mouse embryos using immunofluorescence analyses after inhibition of DNA replication with aphidicolin. Inhibition of DNA replication had no effect on abnormal methylation and hydroxymethylation patterns that still persisted in the superovulated group. Interestingly, the onset of DNA replication, which was also analyzed in these experiments, was remarkably delayed in the superovulated group. Our findings imply an impact of superovulation on both replication-dependent and -independent or yet unknown demethylation mechanisms in PN stage mouse embryos. In addition, they reveal for the first time a negative effect of superovulation on the initiation of DNA replication in PN stage mouse embryos.

Over the last few decades, the use of assisted reproductive technologies (ART) has dramatically increased among couples with fertility problems. Especially, it has become a standard medical therapy in technologically advanced countries. Superovulation with gonadotropins is an essential ART procedure that increases the number of oocytes to achieve high pregnancy rates. It consists of the oral or injectable administration of gonadotropins to females and is widely used for human subfertility treatment and increasing the number of offspring from animals, especially from research animals. Numerous studies have strongly indicated that superovulation can lead to unhealthy oocyte maturation, impaired embryo development, decreased implantation rate, and increased postimplantation loss [Fossum et al., 1989; Ertzeid and Storeng, 1992, 2001; Van der Auwera and D'Hooghe, 2001].

In early embryogenesis, reprogramming of DNA methylation in the zygote was originally believed to begin with active demethylation of the paternal genome involving enzymatic oxidation of 5mC to 5hmC followed by passive demethylation of the maternal genome in a replication-dependent mechanism after the 2-cell stage [Smith et al., 2012]. The enzymatic oxidation of 5mC to 5hmC was shown to be catalyzed by the members of the ten-eleven translocation (TET) protein family (TET1, TET2 and TET3) [Tahiliani et al., 2009; Ito et al., 2010]. Subsequent studies demonstrated that TET proteins can further oxidize 5hmC to 5-formylcytosine (5fC) and 5-carboxylcytosine (5caC) [Ito et al., 2010, 2011; He et al., 2011]. In particular, TET3 is highly expressed in oocytes and zygotes, and zygotes injected with TET3 siRNA displayed reduced 5hmC and increased 5mC levels in the paternal pronucleus [Ito et al., 2010; Szwagierczak et al., 2010; Gu et al., 2011; Iqbal et al., 2011; Wossidlo et al., 2011].

In 2014, the original model of active demethylation of the paternal genome and passive demethylation of the maternal genome could no longer be maintained due to similar findings of 3 different studies. They stated that replication-dependent dilution is the major contributor to paternal DNA demethylation [Guo et al., 2014; Shen et al., 2014] and Tet3-dependent DNA demethylation also occurs on the maternal genome in zygotes before the first mitotic division [Shen et al., 2014; Wang et al., 2014]. Very recently, the understanding of zygote reprogramming experienced a further spectacular twist. Using an ultrasensitive liquid chromatography-mass spectrometry (LC-MS) method, Amouroux et al. [2016] demonstrated that the initial loss of global 5mC in the paternal genome occurs independently of the completion of S phase and starts before Tet3-mediated 5hmC appears. By transplantation of Dnmt-triple-negative embryonic stem cells into oocytes after spindle removal, they further showed that the accumulation of paternal 5hmC is dependent on maternally inherited DNA methyltransferases (Dnmt3a and Dnmt1). Moreover, they provided evidence that Tet3-mediated hydroxylation targets de novo methylation activities in the zygote after an initial active Tet3-independent demethylation, possibly involving the base excision repair (BER) pathway.

Studies in humans associated ART, and in particular superovulation, with epigenetic disturbances by documenting an increased incidence of imprinting disorders [DeBaun et al., 2003; Gicquel et al., 2003; Maher et al., 2003; Ørstavik et al., 2003; Halliday et al., 2004; Sutcliffe et al., 2006; Lim et al., 2009; Vermeiden and Bernardus, 2013]. In mouse, adverse effects of superovulation on imprinted gene methylation and/or expression in oocytes, embryos, fetuses, and placentas have been described [Borghol et al., 2006; Fauque et al., 2007; Sato et al., 2007; Fortier et al., 2008; Khoueiry et al., 2008; Market-Velker et al., 2010; El Hajj et al., 2011; de Waal et al., 2012; Shi et al., 2014]. In addition, our group reported a negative impact of superovulation on the expression of mRNAs encoding the BER proteins APEX1 and POLB as well as the 5-methyl-CpG-binding domain protein MBD3 in single early mouse morula embryos. Studies at the mouse zygote stage detected a higher incidence of abnormal global methylation patterns and reduced developmental potential of mouse preimplantation embryos from superovulated matings [Shi and Haaf, 2002]. More recently, Huffman et al. [2015] further linked superovulation to a reduction of global DNA methylation as well as increased histone H3 lysine 9 (H3K9) and H3K14 acetylation in maternal pronuclei of mouse zygotes.

In the present study, we performed various molecular analyses to investigate the possible effects of superovulation on Tet3- and DNA replication-mediated genome-wide methylation reprogramming in early stage mouse embryos. We carried out immunofluorescence analyses to determine global 5mC and 5hmC levels in 5 different pronuclear (PN) stages of mouse embryos from superovulated and spontaneously ovulated matings. We used single-cell RT-qPCR analysis to measure the expression levels of Tet1, Tet2, and Tet3 genes in single PN stage embryos from spontaneously and superovulated matings. In addition, we determined the initiation kinetics of DNA replication as well as global 5mC and 5hmC levels after inhibition of DNA replication with aphidicolin to analyze the possible contribution of TET-independent demethylation mechanisms such as replication-dependent passive processes.

Animals

All zygotes used in this study were obtained via in vitro fertilization (IVF) in the laboratories of the Translational Animal Research Center (TARC) at the Johannes Gutenberg University Mainz. For all IVF experiments, FVB/N mice were used and maintained under specific-pathogen free conditions in the facilities of the TARC.

Collection of Spermatozoa and Oocytes

Spermatozoa were collected from the cauda epididymidis of 3-12-months-old male mice. The sperm suspension was incubated in a dish containing 100 µL of FERTIUP® Mouse Sperm Preincubation Medium (KYUDO CO., LTD., Saga, Japan) for 1 h to allow for capacitation at 37°C with 5% CO2 in air. Oocytes were collected from 2-3-months-old female mice. Superovulated female mice (n = 74) and spontaneously ovulated female mice (n = 89) were used to obtain all PN stage embryos (online suppl. Table 1; see www.karger.com/doi/10.1159/000493779 for all online suppl. material). For superovulation, females were intraperitoneally injected with 5 IU of pregnant mare's serum gonadotropin (PMSG; Intergonan®, MSD Animal Health, Unterschleißheim, Germany) and 48-52 h later with 5 IU human chorionic gonadotropin (hCG; Ovogest®, MSD Animal Health). Spontaneously ovulated females were not injected with hormones and were kept in the same cage with superovulated females to synchronize the ovulation cycle. Females were dissected and oviducts were prepared and cleaned 14-15 h after the second injection. The cumulus-oocyte-complexes were then released with a dissecting needle and dragged into a drop of 200 µL CARD MEDIUM (KYUDO CO). The fertilization dishes were incubated at 37°C with 5% CO2 for 30-60 min before insemination.

Insemination and Embryo Culture

After separate preincubation of sperm and oocyte samples, 3 µL sperm suspension was added to the 200 µL drop of CARD MEDIUM containing oocytes covered with mineral oil in the fertilization dishes. Then, the fertilization dishes were placed into an incubator (37°C, 5% CO2). After 3-6 h, the inseminated oocytes were washed 3× in 100 µL drops of EmbryoMax® Human Tubal Fluid media (HTF; Merck Millipore, Schwalbach, Germany) covered with mineral oil and cultured at 37°C with 5% CO2 [Nakagata et al., 2013]. After fertilization, PN stages of embryos were identified as described [Santos et al., 2002; Wossidlo et al., 2011]. After morphological inspection, ‘high quality' embryos were used for further analysis.

Whole-Mount Immunofluorescence Staining Analysis of Early Mouse Embryos

The zona pellucida was first removed with acidic Tyrode's solution followed by fixation for 30 min with 2.5% paraformaldehyde in PBS and permeabilization for 1 h with 0.5% Triton X-100 in PBS at room temperature. Cellular DNA was then denatured with 2 N HCl for 30 min followed by neutralization with 0.1 M borate buffer (pH 8.5) for 10 min. Subsequently, the embryos were blocked overnight at 4°C in 0.1% Triton X-100/1% BSA in PBS. In the next step, the embryos were incubated with a mouse primary antibody against 5mC (1:400 dilution; Diagenode, Ougrée, Belgium) and a rabbit primary antibody against 5hmC (1:400 dilution; Active Motif, Carlsbad, CA, USA) at room temperature for 2.5 h. After incubation, the embryos were washed several times with PBS and incubated for 1 h with Alexa Fluor® 488 Goat Anti-Mouse IgG (H+L) (Thermo Fisher Scientific, Waltham, MA, USA) and Alexa Fluor® 594 Goat Anti-Rabbit IgG (H+L) (Thermo Fisher Scientific) secondary antibodies with dilutions of 1:500. Nuclear staining was performed in parallel to secondary antibody incubation using Hoechst 33342 (Thermo Fisher Scientific). After further washing steps with PBS, the embryos were mounted in a small drop of Aqua-Poly/Mount mounting medium (Polysciences, Eppelheim, Germany) on slides and were analyzed using a confocal laser scanning microscope (Carl Zeiss, Thornwood, NY, USA). Cell fluorescence was measured using ImageJ (1.48v) as described by McCloy et al. [2014]. Dot plots were generated and statistical analyses (unpaired 2-tailed Student t tests) were performed using GraphPad Prism 6 (GraphPad Software, San Diego, CA, USA).

Inhibition of DNA Replication in PN Stage Embryos

Aphidicolin is a tetracyclic diterpene antibiotic and specifically inhibits DNA polymerase α, the primary polymerase in charge of nuclear DNA replication. It inhibits zygotic DNA replication but it does not influence zygotic PN maturation [Arand et al., 2015]. PN3 and PN4 stage embryos were used for inhibition experiments because it is known that zygotic DNA replication starts between the late PN3 and the early PN5 stage in mouse embryos [Wossidlo et al., 2010].

Zygotes obtained from spontaneously ovulated and superovulated females were transferred into HTF medium containing either dimethyl sulfoxide (DMSO) or 3 µg/mL aphidicolin (Sigma-Aldrich, Munich, Germany) at 3 h post fertilization (hpf) to inhibit PN DNA replication. While the aphidicolin treatment continued, the zygotes were labelled with 500 µM bromodeoxyuridine (BrdU, Thermo Fisher Scientific) for 2 h to verify inhibition of DNA replication. After the treatment, the zygotes from both groups were briefly washed with PBS. Then, 5mC and 5hmC stainings were performed as described above in parallel to BrdU stainings for which a rat anti-BrdU primary antibody (Abcam, Cambridge, UK) and an Alexa Fluor® 555 Goat Anti-Rat IgG (H+L) secondary antibody (Thermo Fisher Scientific) were used.

Lysis and Reverse Transcription of Single-Cell PN Stage Embryos

Embryos from 5 different PN stages (PN1-PN5) were subjected to single-cell mRNA expression analysis. The CelluLyser™ Micro Lysis and cDNA Synthesis Kit (TATAA Biocenter, Sweden) manual was followed for direct cDNA synthesis from single cells.

Isolation and Reverse Transcription of Total RNA from Mouse Brain Tissue

Total RNA from mouse brain tissue was extracted using RNeasy® Plus Micro Kit (Qiagen, Hilden, Germany) according to the manufacturer's specifications. For reverse transcription, the protocol described in the PrimeScript™ RT Master Mix Perfect Real Time Kit (TAKARA, Japan) manual was followed.

Single-Cell mRNA Expression Analysis by RT-qPCR

Each measurement for detection of Tet1, Tet2, and Tet3 mRNA expression levels was made with a cDNA sample of a single-cell embryo from a spontaneously ovulated or a superovulated female. Additionally, a cDNA sample from mouse brain tissue (2 ng/µL) was always used as a positive control to make an interplate calibration for compensating technical variations between the RT-qPCR runs. The fluorescence dye SYBR Green [SYBR® Premix Ex Taq; Tli RNaseH Plus (2×) Kit; TAKARA, Japan] was utilized for detection of PCR amplification. The reactions were run on an ABI StepOnePlus™ Real-Time PCR System (Life Technologies, Karlsruhe, Germany) and the cycling conditions were as follows: 95°C for 30 s; 45 cycles of 95°C for 5 s, 60°C for 30 s and 72°C for 30 s. Primer sets for Tet1, Tet2, and Tet3 genes were the same as described by Blaschke et al. [2013]. The single-cell RT-qPCR data was analyzed according to Ståhlberg et al. [2013]. In this study, GenEx (ver. 6, MultiD) software was used for interplate calibration, assay efficiency correction, and conversion to log-scale of single-cell RT-qPCR data.

DNA Methylation and Hydroxymethylation Analysis of PN Stage Mouse Embryos

We performed quantitative whole-mount immunofluorescence staining to determine genome-wide 5mC and 5hmC levels in embryos from spontaneously ovulated and superovulated females at 5 different PN stages (PN1-PN5) (Fig. 1A).

Fig. 1

Comparison of 5mC and 5hmC levels of PN1-PN5 pronuclear stage embryos from spontaneously ovulated and superovulated matings. A Whole-mount 5mC and 5hmC labelling in early mouse embryos. 5hmC preferentially appears in the paternal pronucleus of early mouse preimplantation embryos. Shown are representative images of PN stage embryos (PN1-PN5) stained with 5mC and 5hmC antibodies. mC, green; hmC, red; ♂, male pronucleus; ♀, female pronucleus. Scale bars, 10 µm. B Quantification of 5mC intensities in paternal pronuclei at PN1-PN5. C Quantification of 5mC intensities in maternal pronuclei at PN1-PN5. D Quantification of 5hmC intensities in paternal pronuclei at PN1-PN5. E Quantification of 5hmC intensities in maternal pronuclei at PN1-PN5. Number of zygotes analyzed for 5mC and 5hmC in each parental pronucleus subgroup: spontaneous ovulation PN1 = 21; superovulation PN1 = 16; spontaneous ovulation PN2 = 12; superovulation PN2 = 20; spontaneous ovovulation PN3 = 10; superovulation PN3 = 22; spontaneous ovulation PN4 = 14; superovulation PN4 = 35; spontaneous ovulation PN5 = 34; superovulation PN5 = 26. Quantifications are represented as the mean of intensities. Dot plots for PN1-PN5 stage embryos with the intensity of each individual embryo and the p values are given in the online suppl. Fig. 1, 2, 3, 4, and 5. Asterisks indicate significant differences, *p < 0.05; **p < 0.01; ***p < 0.001. Statistical analyses were carried out using an unpaired 2-tailed Student t test. Error bars indicate the standard deviation. CTCF, corrected total cell fluorescence.

Fig. 1

Comparison of 5mC and 5hmC levels of PN1-PN5 pronuclear stage embryos from spontaneously ovulated and superovulated matings. A Whole-mount 5mC and 5hmC labelling in early mouse embryos. 5hmC preferentially appears in the paternal pronucleus of early mouse preimplantation embryos. Shown are representative images of PN stage embryos (PN1-PN5) stained with 5mC and 5hmC antibodies. mC, green; hmC, red; ♂, male pronucleus; ♀, female pronucleus. Scale bars, 10 µm. B Quantification of 5mC intensities in paternal pronuclei at PN1-PN5. C Quantification of 5mC intensities in maternal pronuclei at PN1-PN5. D Quantification of 5hmC intensities in paternal pronuclei at PN1-PN5. E Quantification of 5hmC intensities in maternal pronuclei at PN1-PN5. Number of zygotes analyzed for 5mC and 5hmC in each parental pronucleus subgroup: spontaneous ovulation PN1 = 21; superovulation PN1 = 16; spontaneous ovulation PN2 = 12; superovulation PN2 = 20; spontaneous ovovulation PN3 = 10; superovulation PN3 = 22; spontaneous ovulation PN4 = 14; superovulation PN4 = 35; spontaneous ovulation PN5 = 34; superovulation PN5 = 26. Quantifications are represented as the mean of intensities. Dot plots for PN1-PN5 stage embryos with the intensity of each individual embryo and the p values are given in the online suppl. Fig. 1, 2, 3, 4, and 5. Asterisks indicate significant differences, *p < 0.05; **p < 0.01; ***p < 0.001. Statistical analyses were carried out using an unpaired 2-tailed Student t test. Error bars indicate the standard deviation. CTCF, corrected total cell fluorescence.

Close modal

Expectedly, the 5mC intensity on the paternal pronuclei of both groups was markedly reduced after the PN2 stage, but the reduction was less marked in the superovulated group compared to the spontaneously ovulated group (Fig. 1B, online suppl. Figs. 2A, 3A, 4A). Furthermore, the 5mC levels of the maternal pronuclei were significantly lower in the superovulated group compared to the spontaneously ovulated group during the majority of phases of PN stage embryo development (Fig. 1C, online suppl. Fig. 2B, 3B, 4B). In addition, significantly decreased 5hmC levels were measured in the paternal pronuclei of superovulated PN3 and PN4 stage embryos compared to the spontaneously ovulated counterparts (Fig. 1D, online suppl. Figs. 3C, 4C). 5hmC staining of the maternal pronuclei was generally less intense with significantly lower values in superovulated embryos detected only at the PN3 stage (Fig. 1E, online suppl. Fig. 3D).

Single-Cell mRNA Expression Analysis of Tet Genes

To investigate the involvement of Tet genes in the disturbance of epigenetic reprogramming in superovulated PN stage embryos, we analyzed single-cell mRNA expression of Tet1, Tet2, and Tet3 genes at 5 different PN stages (PN1-PN5). In total, we investigated 62 PN stage embryos from each the spontaneously and the superovulated group which were almost equally distributed among the different PN stages (Table 1).

Table 1

Numbers of single mouse zygotes analyzed for Tet1, Tet2, and Tet3 mRNA expression at each PN stage

Numbers of single mouse zygotes analyzed for Tet1, Tet2, and Tet3 mRNA expression at each PN stage
Numbers of single mouse zygotes analyzed for Tet1, Tet2, and Tet3 mRNA expression at each PN stage

Relative expression of the Tet3 mRNA was significantly higher than that of Tet1 and Tet2 mRNAs at each PN stage (Fig. 2). However, no significant difference in Tet1, Tet2, and Tet3 mRNA expression levels between embryos from the spontaneously ovulated and the superovulated group were observed for all analyzed PN stages. In addition, the mRNA expression levels of each gene were similar at the different PN stages in the same group.

Fig. 2

Tet1, Tet2, and Tet3 mRNA expression levels at the 5 different PN stages after spontaneous ovulation and superovulation. No significant expression differences between spontaneously ovulated and superovulated PN stage embryos were observed. Box plots show the log2-fold relative mRNA expression of Tet1, Tet2, and Tet3in embryos from spontaneously ovulated (blue) and superovulated matings (green). The median is represented by thick horizontal lines. The bottom of the box indicates the 25th percentile, the top the 75th percentile. Outliers are displayed as open circles, extreme outliers as stars.

Fig. 2

Tet1, Tet2, and Tet3 mRNA expression levels at the 5 different PN stages after spontaneous ovulation and superovulation. No significant expression differences between spontaneously ovulated and superovulated PN stage embryos were observed. Box plots show the log2-fold relative mRNA expression of Tet1, Tet2, and Tet3in embryos from spontaneously ovulated (blue) and superovulated matings (green). The median is represented by thick horizontal lines. The bottom of the box indicates the 25th percentile, the top the 75th percentile. Outliers are displayed as open circles, extreme outliers as stars.

Close modal

Inhibition and Onset of DNA Replication in PN Stage Mouse Embryos

To investigate the effect of superovulation on DNA replication possibly followed by an impairment of demethylation, we inhibited DNA replication in PN3 and PN4 stage embryos using aphidicolin. Aphidicolin-treated zygotes [aphidicolin (+)] and control zygotes [aphidicolin (-)] from both the superovulation and the spontaneously ovulated group were labeled with BrdU to validate inhibition of DNA replication by immunofluorescence staining using an anti-BrdU antibody. As a proof for successful inhibition of DNA replication, no BrdU incorporation was detected in pronuclei of aphidicolin-treated zygotes, while strong BrdU intensity was observed in control zygotes indicating ongoing DNA replication (Fig. 3A).

Fig. 3

Inhibition and initiation of DNA Replication in PN stage mouse embryos. A Validation of inhibition of DNA replication with BrdU. PN3 and PN4 stage embryos were incubated in medium containing BrdU in the presence (+) or absence (-) of aphidicolin and fixed at 6 and 8 hpf. Shown are representative images of embryos from the spontaneously ovulated group. B BrdU labelling in PN3 and PN4 stage embryos after spontaneous ovulation and superovulation in the absence (-) of aphidicolin. The experiments for both groups were performed in parallel and the same BrdU-containing medium was used. Shown are representative images of embryos from the spontaneously ovulated and the superovulated group. Red signals indicate the BrdU staining in the pronuclei. ♂, male pronucleus; ♀, female pronucleus. Scale bars, 10 µm. C Superovulation influences the onset of DNA replication in paternal and maternal pronuclei of PN stage embryos. PN3 and PN4 stage embryos were incubated with BrdU-containing medium and BrdU intensities of pronuclei were measured with ImageJ. Number of zygotes analyzed for BrdU signal in each parental pronucleus subgroup: spontaneous ovulation PN3 = 7; spontaneous ovulation PN4 = 4; superovulation PN3 = 13; superovulation PN4 = 6. Double asterisks indicate significant differences (p < 0.01), p = 0.0039 (for paternal pronuclei on the left ) and p = 0.0013 (for maternal pronuclei on the right). Statistical analyses were carried out using an unpaired 2-tailed Student t test. The long red horizontal lines represent the median, the short red horizontal lines the standard deviation. CTCF, corrected total cell fluorescence.

Fig. 3

Inhibition and initiation of DNA Replication in PN stage mouse embryos. A Validation of inhibition of DNA replication with BrdU. PN3 and PN4 stage embryos were incubated in medium containing BrdU in the presence (+) or absence (-) of aphidicolin and fixed at 6 and 8 hpf. Shown are representative images of embryos from the spontaneously ovulated group. B BrdU labelling in PN3 and PN4 stage embryos after spontaneous ovulation and superovulation in the absence (-) of aphidicolin. The experiments for both groups were performed in parallel and the same BrdU-containing medium was used. Shown are representative images of embryos from the spontaneously ovulated and the superovulated group. Red signals indicate the BrdU staining in the pronuclei. ♂, male pronucleus; ♀, female pronucleus. Scale bars, 10 µm. C Superovulation influences the onset of DNA replication in paternal and maternal pronuclei of PN stage embryos. PN3 and PN4 stage embryos were incubated with BrdU-containing medium and BrdU intensities of pronuclei were measured with ImageJ. Number of zygotes analyzed for BrdU signal in each parental pronucleus subgroup: spontaneous ovulation PN3 = 7; spontaneous ovulation PN4 = 4; superovulation PN3 = 13; superovulation PN4 = 6. Double asterisks indicate significant differences (p < 0.01), p = 0.0039 (for paternal pronuclei on the left ) and p = 0.0013 (for maternal pronuclei on the right). Statistical analyses were carried out using an unpaired 2-tailed Student t test. The long red horizontal lines represent the median, the short red horizontal lines the standard deviation. CTCF, corrected total cell fluorescence.

Close modal

Surprisingly, BrdU intensity was much lower in untreated PN3 stage embryos from the superovulated group than in the untreated PN3 stage embryos from the spontaneously ovulated group (Fig. 3B). However, the BrdU intensity became stronger in PN4 stage embryos from the superovulated group which indicates a delayed initiation of DNA replication in the superovulated group (Fig. 3B).

Quantification of BrdU intensities revealed that the majority of the superovulated PN3 stage embryos did not show any BrdU staining or only weak BrdU staining (Fig. 3C). As expected, BrdU intensities of both maternal and paternal pronuclei of PN3 stage embryos were significantly different between the spontaneously ovulated and the superovulated group. However, this significant difference disappeared in the PN4 stage embryos (Fig. 3C).

5mC and the 5hmC Levels after Inhibition of DNA Replication

Double immunofluorescence staining was performed to quantify the 5mC and the 5hmC levels in the same aphidicolin-treated and untreated zygotes for which BrdU incorporation was measured. Significantly decreased 5mC levels were observed in the paternal pronuclei of PN3 stage embryos from superovulated females compared to those of spontaneously ovulated females independent of aphidicolin treatment (Fig. 4A). However, in the paternal pronuclei of PN4 stage embryos, 5mC levels displayed no significant difference between the superovulated and the spontaneously ovulated group.

Fig. 4

Inhibition of PN DNA replication does not rescue abnormal methylation and hydroxymethylation patterns in the superovulated group. Whole-mount 5mC and 5hmC labelling in early mouse embryos after incubation in the presence (+) and absence (-) of aphidicolin. A 5mC intensities in the paternal pronuclei of PN3 and PN4 stage embryos with or without aphidicolin treatment. p = 0.0425 (for aphidicolin-treated PN3 embryos on the left) and p = 0.0046 (for untreated PN3 embryos on the right). B 5mC intensities in the maternal pronuclei. p < 0.0001. C 5hmC intensities in the paternal pronuclei. p values from left to right: p = 0.0014, p < 0.0001, p < 0.0001. D 5hmC intensities in the maternal pronuclei. p = 0.0001. For 5hmC quantification with the confocal laser scanning microscope, a different filter (red) was used than for 5mC quantification (green). 5mC and 5hmC signal intensities are represented as CTCF (corrected total cell fluorescence.). Number of zygotes analyzed for 5mC and 5hmC in each parental pronucleus subgroup in the presence (+) of aphidicolin: spontaneous ovulation PN3 = 13; spontaneous ovulation PN4 = 15; superovulation PN3 = 17; superovulation PN4 = 19. Number of zygotes analyzed for 5mC and 5hmC in each parental pronucleus subgroup in the absence (-) of aphidicolin: spontaneous ovulation PN3 = 22; spontaneous ovulation PN4 = 23; superovulation PN3 = 28; superovulation PN4 = 16. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Statistical analyses were carried out using an unpaired 2-tailed Student t test. The long red horizontal lines represent the median, the short red horizontal lines the standard deviation.

Fig. 4

Inhibition of PN DNA replication does not rescue abnormal methylation and hydroxymethylation patterns in the superovulated group. Whole-mount 5mC and 5hmC labelling in early mouse embryos after incubation in the presence (+) and absence (-) of aphidicolin. A 5mC intensities in the paternal pronuclei of PN3 and PN4 stage embryos with or without aphidicolin treatment. p = 0.0425 (for aphidicolin-treated PN3 embryos on the left) and p = 0.0046 (for untreated PN3 embryos on the right). B 5mC intensities in the maternal pronuclei. p < 0.0001. C 5hmC intensities in the paternal pronuclei. p values from left to right: p = 0.0014, p < 0.0001, p < 0.0001. D 5hmC intensities in the maternal pronuclei. p = 0.0001. For 5hmC quantification with the confocal laser scanning microscope, a different filter (red) was used than for 5mC quantification (green). 5mC and 5hmC signal intensities are represented as CTCF (corrected total cell fluorescence.). Number of zygotes analyzed for 5mC and 5hmC in each parental pronucleus subgroup in the presence (+) of aphidicolin: spontaneous ovulation PN3 = 13; spontaneous ovulation PN4 = 15; superovulation PN3 = 17; superovulation PN4 = 19. Number of zygotes analyzed for 5mC and 5hmC in each parental pronucleus subgroup in the absence (-) of aphidicolin: spontaneous ovulation PN3 = 22; spontaneous ovulation PN4 = 23; superovulation PN3 = 28; superovulation PN4 = 16. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Statistical analyses were carried out using an unpaired 2-tailed Student t test. The long red horizontal lines represent the median, the short red horizontal lines the standard deviation.

Close modal

Similarly, the 5mC levels of maternal pronuclei did not differ significantly between PN3 and PN4 stage embryos from the superovulated and the spontaneously ovulated group. Interestingly, 5mC levels dramatically increased in maternal pronuclei of superovulated PN4 embryos not treated with aphidicolin compared to maternal pronuclei of untreated PN4 embryos from the spontaneously ovulated group (Fig. 4B).

Paternal accumulation of 5hmC also showed a significant reduction after superovulation in aphidicolin-treated PN3 and PN4 stage embryos as well as nontreated PN4 stage embryos (Fig. 4C). A significant difference of maternal 5hmC accumulation in the maternal pronuclei was only detected in the aphidicolin-treated PN4 zygotes that showed decreased 5hmC levels after superovulation (Fig. 4D).

In the present study, we further investigated the possible negative effects of superovulation as an essential ART procedure by focusing on the embryonic methylation and hydroxymethylation machinery, the specific expression of genes involved in the demethylation process (Tet1, Tet2, and Tet3), and the onset of DNA replication in the PN stage mouse embryos.

The results of our whole-mount immunofluorescence stainings indicated a negative impact of superovulation on the early embryonic methylation and hydroxymethylation machinery. Lower methylation and hydroxymethylation were detected in both maternal and paternal pronuclei of superovulation-derived zygotes. Particularly, superovulation-derived PN3 stage embryos displayed reduced methylation and hydroxymethylation levels. The PN3 stage is specifically known to be critical for epigenetic reprogramming since it represents the starting point of de novo methylation, hydroxymethylation, and DNA replication after fertilization [Amouroux et al., 2016].

Previous studies also used immunofluorescence staining with antibodies against 5mC to document abnormal genome-wide methylation reprogramming in superovulated mouse zygotes and 2-cell embryos [Shi and Haaf, 2002; Huffman et al., 2015]. Shi and Haaf [2002] used mouse 2-cell embryos from superovulated females, nonsuperovulated matings, and IVF for 5mC staining and in vitro development analysis up to the blastocyst stage. They further investigated the effects of different culture medium conditions and different genetic backgrounds on the embryos. In agreement with our observation, they detected aberrant methylation patterns in the superovulated group that also depended on the culture conditions and the genetic background. In addition, their results indicated a higher failure (14%) to develop to the blastocyst stage for embryos obtained from superovulated matings compared to those from spontaneously ovulated matings (5%). Also similar to our findings, the recently published study of Huffman et al. [2015] indicated a hypomethylation of the maternal pronucleus of zygotes from superovulated females. However, this study only analyzed maternal pronuclei of PN3 and PN4 embryos and thus lacks a detailed analysis of paternal pronuclei as well as all PN stages. In addition, it only investigated DNA methylation and not DNA hydroxymethylation. Thus, our study is the first study that showed a negative impact of superovulation on postzygotic reprogramming with a detailed methylation and hydroxymethylation analysis in maternal and paternal pronuclei of embryos from 5 different PN stages. Nevertheless, additional studies are needed to analyze genome-wide 5mC and 5hmC levels together with the levels of the other 5mC oxidative derivatives 5fC and 5caC in further cleavage stages of mouse embryos. Furthermore, these analyses should be extended to postimplantation mouse embryos and fetuses as well as neonates to determine whether these epigenetic abnormalities persist during development.

It was tempting to speculate that our finding of diminished 5hmC levels in PN stage embryos from superovulated matings may be due to superovulation-induced alterations of zygotic Tet1, Tet2, and/or Tet3 mRNA amounts. Because of the unique characteristics, temporal dynamics and stochastic variations observed among individual cells, single-cell analysis is the best way to understand cell heterogeneity and responses to stimuli. However, our single-cell mRNA expression analysis of Tet1, Tet2, and Tet3 revealed no clear expression differences between embryos of all 5 PN stages derived from spontaneously ovulated and superovulated matings. These results suggest that the detected aberrant methylation and hydroxymethylation patterns are not caused by differences in TET-mediated active demethylation mechanisms, but rather by replication-dependent passive processes or other up to now unknown mechanisms. Consistent with these results of single-cell targeted gene expression analysis, the 3 Tet genes were also detected as nonregulated genes in our unpublished data of an RNA-Seq-based comparative transcriptome analysis of each 2 different pools of PN5 stage mouse embryos from superovulated and spontaneously ovulated matings. However, it should be considered that there might be still a direct or indirect effect of superovulation on enzyme activity or posttranscriptional regulation of TET1, TET2, and/or TET3. Additional studies focusing on these issues have to be performed.

We further addressed whether superovulation affects DNA replication and thus leads to impaired demethylation in zygotes by inhibiting DNA replication with aphidicolin in PN3 and PN4 stage embryos from spontaneously ovulated and superovulated females. Surprisingly, we observed a significantly delayed but not completely abrogated onset of DNA replication in embryos from the superovulated group. This finding indicates that superovulation has a negative impact on replication timing but not on the entire replication mechanism. Nevertheless, further experiments are necessary to elucidate the potential effects of superovulation on DNA replication at the molecular level. Of note, there is already convincing evidence pointing out such effects in the literature. It is known that DNA replication timing may cause changes in gene expression, alterations in epigenetic modifications, and an increase of structural malformations [Donley and Thayer, 2013]. Additionally, studies in yeast indicated that late-replicating regions of the genome have higher rates of spontaneous mutagenesis than early-replicating regions [Lang and Murray, 2011]. Accordingly, the observed delayed onset of DNA replication in superovulated mouse embryos may contribute to the development of ART-related disorders.

Even though we observed an impact of superovulation on the onset of DNA replication, its inhibition of it did not rescue or modify the disturbances of methylation and hydroxymethylation in the superovulated group. Thus, this is the first study to describe a superovulation-induced impairment of DNA replication which is not causally linked to the already reported superovulation-induced failures of methylation reprogramming.

When we compared the methylation and hydroxymethylation levels of nontreated aphidicolin (-) control embryos from superovulated and spontaneously ovulated matings (Fig. 4), we could replicate the superovulation-associated lower methylation and hydroxymethylation levels observed in our first experiment (Fig. 1) only in some, but not all of the analyzed PN stages and parental pronucleus groups. The only difference between the 5mC and 5hmC immunofluorescence staining experiments is the DMSO contained in the medium used for the nontreated aphidicolin (-) zygotes in the second experiment. DMSO was included in the medium since it was used as a solvent for aphidicolin in the medium of the treated aphidicolin (+) counterparts. In this context, it has to be considered that DMSO has been reported to effect increased Dnmt3a expression and DNA methylation changes in mouse embryoid bodies [Iwatani et al., 2006]. Thus, the partially discrepant findings of the 5mC and 5hmC immunofluorescence staining experiments may have resulted from the DMSO in the culture medium.

For a considerable time, there was the widely accepted hypothesis, that the paternal genome is actively demethylated in zygotes by Tet3-dependent oxidation of 5mC [Gu et al., 2011; Iqbal et al., 2011; Wossidlo et al., 2011], while the maternal genome undergoes a passive 5mC dilution due to DNA replication [Rougier et al., 1998]. However, recently published studies stated that DNA replication is the main contributor to DNA demethylation of the paternal genome [Guo et al., 2014; Shen et al., 2014], and Tet3-dependent DNA demethylation also occurs on the maternal genome in zygotes before the first mitotic division [Shen et al., 2014; Wang et al., 2014]. Overall, these 3 studies demonstrated that 5mC can be removed from the zygotic genome by 3 processes: Tet-mediated active demethylation, replication-dependent demethylation, or replication-dependent removal of 5hmC after oxidation [Gkountela and Clark, 2014]. On the other hand, a new study published in 2016 concluded that Tet3 and the formation of 5hmC is not needed for the initial loss of 5mC in the paternal genome [Amouroux et al., 2016]. This study further stated that global DNA demethylation is mainly performed by replication-independent mechanisms. They showed that the inhibition of zygotic DNA replication by aphidicolin at 10 hpf caused a nonsignificant accumulation of 5mC. However, the experiment was performed for only 1 PN stage and thus their conclusion of an only small contribution of DNA replication to demethylation in zygotes is drawn on an only very limited experimental basis. Nevertheless, the data from our study also confirmed that DNA replication is not the only mechanism involved in postzygotic DNA demethylation. Similar to Amouroux et al. [2016], we thus speculate that BER repair activity is required for Tet3- and replication-independent demethylation processes. We further conjecture that superovulation impairs proper function of the BER pathway and thus leads to the observed disturbances of zygotic methylation reprogramming. Interestingly, this assumption is corroborated by our previous finding of a downregulation of mRNAs encoding the BER proteins APEX1 and POLB in single early mouse morula embryos from superovulated matings [Linke et al., 2013].

In conclusion, the results obtained in this study demonstrate for the first time that the exogenous administration of gonadotrophins disrupts hydroxymethylation and methylation in both the maternal and paternal pronucleus of zygotes and, most interestingly, delays the onset of zygotic DNA replication. Thus, the findings of this study add to the growing evidence that superovulation may create a negative impact on epigenetic regulation and gene expression. Overall, our findings give important new insights in epigenetic changes during early embryogenesis associated with ART and help to improve the quality and developmental potential of ART-derived embryos.

This work was supported by a fellowship of the International PhD Programme (IPP) coordinated by the Institute of Molecular Biology, Mainz, Germany. We thank Marisa Alles and Anna Hörner for their assistance with the IVF experiments.

Mouse experiments were approved by the committee for animal experimentation at Mainz University Medical Center and were performed in accordance with the European Union normative for care and use of experimental animals.

The authors have no conflicts of interest to declare.

1.
Amouroux R, Nashun B, Shirane K, Nakagawa S, Hill PWS, et al: De novo DNA methylation drives 5hmC accumulation in mouse zygotes. Nat Cell Biol 18:225-233 (2016).
[PubMed]
2.
Arand J, Wossidlo M, Lepikhov K, Peat JR, Reik, W, et al: Selective impairment of methylation maintenance is the major cause of DNA methylation reprogramming in the early embryo. Epigenetics Chromatin 8:1 (2015).
[PubMed]
3.
Blaschke K, Ebata KT, Karimi MM, Zepeda-Martínez J, Goyal P, et al: Vitamin C induces Tet-dependent DNA demethylation and a blastocyst-like state in ES cells. Nature 500:222-226 (2013).
[PubMed]
4.
Borghol N, Lornage J, Blachère T, Sophie Garret A, Lefèvre A: Epigenetic status of the H19 locus in human oocytes following in vitro maturation. Genomics 87:417-426 (2006).
[PubMed]
5.
DeBaun MR, Niemitz EL, Feinberg AP: Association of in vitro fertilization with Beckwith-Wiedemann syndrome and epigenetic alterations of LIT1 and H19. Am J Hum Genet 72:156-160 (2003).
[PubMed]
6.
de Waal E, Yamazaki Y, Ingale P, Bartolomei MS, Yanagimachi R, et al: Gonadotropin stimulation contributes to an increased incidence of epimutations in ICSI-derived mice. Hum Mol Genet 21:4460-4472 (2012).
[PubMed]
7.
Donley N, Thayer MJ: DNA replication timing, genome stability and cancer: late and/or delayed DNA replication timing is associated with increased genomic instability. Semin Cancer Biol 23:80-89 (2013).
[PubMed]
8.
El Hajj N, Trapphoff T, Linke M, May A, Hansmann T, et al: Limiting dilution bisulfite (pyro)sequencing reveals parent-specific methylation patterns in single early mouse embryos and bovine oocytes. Epigenetics 6:1176-1188 (2011).
[PubMed]
9.
Ertzeid G, Storeng R: Adverse effects of gonadotrophin treatment on pre- and postimplantation development in mice. J Reprod Fertil 96:649-655 (1992).
[PubMed]
10.
Ertzeid G, Storeng R: The impact of ovarian stimulation on implantation and fetal development in mice. Hum Reprod 16:221-225 (2001).
[PubMed]
11.
Fauque P, Jouannet P, Lesaffre C, Ripoche MA, Dandolo L, et al: Assisted Reproductive Technology affects developmental kinetics, H19 Imprinting Control Region methylation and H19 gene expression in individual mouse embryos. BMC Dev Biol 7:116 (2007).
12.
Fortier AL, Lopes FL, Darricarrere N, Martel J, Trasler JM: Superovulation alters the expression of imprinted genes in the midgestation mouse placenta. Hum Mol Genet 17:1653-1665 (2008).
[PubMed]
13.
Fossum GT, Davidson A, Paulson RJ: Ovarian hyperstimulation inhibits embryo implantation in the mouse. J In Vitro Fert Embryo Transf 6:7-10 (1989).
[PubMed]
14.
Gicquel C, Gaston V, Mandelbaum J, Siffroi JP, Flahault A, et al: In vitro fertilization may increase the risk of Beckwith-Wiedemann syndrome related to the abnormal imprinting of the KCN1OT gene. Am J Hum Genet 72:1338-1341 (2003).
[PubMed]
15.
Gkountela S, Clark AT: A big surprise in the little zygote: the curious business of losing methylated cytosines. Cell Stem Cell 15:393-394 (2014).
[PubMed]
16.
Gu TP, Guo F, Yang H, Wu HP, Xu GF, et al: The role of Tet3 DNA dioxygenase in epigenetic reprogramming by oocytes. Nature 477:606-610 (2011).
[PubMed]
17.
Guo F, Li X, Liang D, Li T, Zhu P, et al: Active and passive demethylation of male and female pronuclear DNA in the mammalian zygote. Cell Stem Cell 15:447-458 (2014).
[PubMed]
18.
Halliday J, Oke K, Breheny S, Algar E, J Amor D, et al: Beckwith-Wiedemann syndrome and IVF: a case-control study. Am J Hum Genet 75:526-528 (2004).
[PubMed]
19.
He YF, Li BZ, Li Z, Liu P, Wang Y, et al: Tet-mediated formation of 5-carboxylcytosine and its excision by TDG in mammalian DNA. Science 333:1303-1307 (2011).
[PubMed]
20.
Huffman SR, Pak Y, Rivera RM: Superovulation induces alterations in the epigenome of zygotes, and results in differences in gene expression at the blastocyst stage in mice. Mol Reprod Dev 82:207-217 (2015).
[PubMed]
21.
Iqbal K, Jin SG, Pfeifer GP, Szabo PE: Reprogramming of the paternal genome upon fertilization involves genome-wide oxidation of 5-methylcytosine. Proc Natl Acad Sci USA 108:3642-3647 (2011).
[PubMed]
22.
Ito S, D'Alessio AC, Taranova OV, Hong K, Sowers LC, et al: Role of Tet proteins in 5mC to 5hmC conversion, ES-cell self-renewal and inner cell mass specification. Nature 466:1129-1133 (2010).
[PubMed]
23.
Ito S, Shen L, Dai Q, Wu SC, Collins LB, et al: Tet proteins can convert 5-methylcytosine to 5-formylcytosine and 5-carboxylcytosine. Science 333:1300-1303 (2011).
[PubMed]
24.
Iwatani M, Ikegami K, Kremenska Y, Hattori N, Tanaka S, et al: Dimethyl sulfoxide has an impact on epigenetic profile in mouse embryoid body. Stem Cells 24:2549-2556 (2006).
[PubMed]
25.
Khoueiry R, Khoureiry R, Ibala-Rhomdane S, Méry L, Blachère T, et al: Dynamic CpG methylation of the KCNQ1OT1 gene during maturation of human oocytes. J Med Genet 45:583-588 (2008).
[PubMed]
26.
Lang GI, Murray AW: Mutation rates across budding yeast chromosome VI are correlated with replication timing. Genome Biol Evol 3:799-811 (2011).
[PubMed]
27.
Lim D, Bowdin SC, Tee L, Kirby GA, Blair E, et al: Clinical and molecular genetic features of Beckwith-Wiedemann syndrome associated with assisted reproductive technologies. Hum Reprod 24:741-747 (2009).
[PubMed]
28.
Linke M, May A, Reifenberg K, Haaf T, Zechner U: The impact of ovarian stimulation on the expression of candidate reprogramming genes in mouse preimplantation embryos. Cytogenet Genome Res 139:71-79 (2013).
[PubMed]
29.
Maher ER, Brueton LA, Bowdin SC, Luharia A, Cooper W, et al: Beckwith-Wiedemann syndrome and assisted reproduction technology (ART). J Med Genet 40:62-64 (2003).
[PubMed]
30.
Market-Velker BA, Zhang L, Magri LS, Bonvissuto AC, Mann MRW: Dual effects of superovulation: loss of maternal and paternal imprinted methylation in a dose-dependent manner. Hum Mol Genet 19:36-51 (2010).
[PubMed]
31.
McCloy RA, Rogers S, Caldon CE, Lorca T, Castro A, et al: Partial inhibition of Cdk1 in G2 phase overrides the SAC and decouples mitotic events. Cell Cycle 13:1400-1412 (2014).
[PubMed]
32.
Nakagata N, Takeo T, Fukumoto K, Kondo T, Haruguchi Y, et al: Cryobiology applications of cryopreserved unfertilized mouse oocytes for in vitro. Cryobiology 67:188-192 (2013).
[PubMed]
33.
Ørstavik KH, Eiklid K, van der Hagen CB, Spetalen S, Kierulf K, et al: Another case of imprinting defect in a girl with Angelman syndrome who was conceived by intracytoplasmic semen injection. Am J Hum Genet 72:218-219 (2003).
[PubMed]
34.
Rougier N, Bourc'his D, Gomes DM, Niveleau A, Plachot M, et al: Chromosome methylation patterns during mammalian preimplantation development. Genes Dev 12:2108-2113 (1998).
[PubMed]
35.
Santos F, Hendrich B, Reik W, Dean W: Dynamic reprogramming of DNA methylation in the early mouse embryo. Dev Biol 241:172-182 (2002).
[PubMed]
36.
Sato A, Otsu E, Negishi H, Utsunomiya T, Arima T: Aberrant DNA methylation of imprinted loci in superovulated oocytes. Hum Reprod 22:26-35 (2007).
[PubMed]
37.
Shen L, Inoue A, He J, Liu Y, Lu F, et al: Tet3 and DNA replication mediate demethylation of both the maternal and paternal genomes in mouse zygotes. Cell Stem Cell 15:459-470 (2014).
[PubMed]
38.
Shi W, Haaf T: Aberrant methylation patterns at the two-cell stage as an indicator of early developmental failure. Mol Reprod Dev 63:329-334 (2002).
[PubMed]
39.
Shi X, Chen S, Zheng H, Wang L, Wu Y: Abnormal DNA methylation of imprinted loci in human preimplantation embryos. Reprod Sci 21:978-983 (2014).
[PubMed]
40.
Smith ZD, Chan MM, Mikkelsen TS, Gu H, Gnirke A, et al: A unique regulatory phase of DNA methylation in the early mammalian embryo. Nature 484:339-344 (2012).
[PubMed]
41.
Ståhlberg A, Rusnakova V, Forootan A, Anderova M, Kubista M: RT-qPCR work-flow for single-cell data analysis. Methods 59:80-88 (2013).
[PubMed]
42.
Sutcliffe AG, Peters CJ, Bowdin S, Temple K, Reardon W, et al: Assisted reproductive therapies and imprinting disorders-a preliminary British survey. Hum Reprod 21:1009-1011 (2006).
[PubMed]
43.
Szwagierczak A, Bultmann S, Schmidt CS, Spada F, Leonhardt H: Sensitive enzymatic quantification of 5-hydroxymethylcytosine in genomic DNA. Nucleic Acids Res 38:e181 (2010).
[PubMed]
44.
Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, et al: Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science 324:930-935 (2009).
[PubMed]
45.
Van der Auwera I, D'Hooghe T: Superovulation of female mice delays embryonic and fetal development. Hum Reprod 16:1237-1243 (2001).
[PubMed]
46.
Vermeiden JPW, Bernardus RE: Are imprinting disorders more prevalent after human in vitro fertilization or intracytoplasmic sperm injection? Fertil Steril 99:642-651 (2013).
[PubMed]
47.
Wang L, Zhang J, Duan J, Gao X, Zhu W, et al: Programming and inheritance of parental DNA methylomes in mammals. Cell 157:979-991 (2014).
[PubMed]
48.
Wossidlo M, Arand J, Sebastiano V, Lepikhov K, Boiani M, et al: Dynamic link of DNA demethylation, DNA strand breaks and repair in mouse zygotes. EMBO J 29:1877-1888 (2010).
[PubMed]
49.
Wossidlo M, Nakamura T, Lepikhov K, Marques CJ, Zakhartchenko V, et al: 5-Hydroxymethylcytosine in the mammalian zygote is linked with epigenetic reprogramming. Nat Commun 2:241 (2011).
[PubMed]