Dynamic expression of TET1, TET2, and TET3 dioxygenases in mouse and human placentas throughout gestation

Placenta 59 (2017) 46e56

Contents lists available at ScienceDirect

Placenta journal homepage: www.elsevier.com/locate/placenta

Dynamic expression of TET1, TET2, and TET3 dioxygenases in mouse and human placentas throughout gestation Joanna Rakoczy a, b, Nisha Padmanabhan a, b, 1, Ada M. Krzak c, Jens Kieckbusch b, c, Tereza Cindrova-Davies a, b, Erica D. Watson a, b, * a b c

Department of Physiology, Development and Neuroscience, University of Cambridge, Cambridge CB2 3EG, UK Centre for Trophoblast Research, University of Cambridge, Cambridge CB2 3EG, UK Department of Obstetrics and Gynaecology, University of Cambridge School of Clinical Medicine, Cambridge CB2 0SP, UK

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 June 2017 Received in revised form 7 September 2017 Accepted 18 September 2017

Introduction: Throughout pregnancy, the placenta dynamically changes as trophoblast progenitors differentiate into mature trophoblast cell subtypes. This process is in part controlled by epigenetic regulation of DNA methylation leading to the inactivation of ‘progenitor cell’ genes and the activation of ‘differentiation’ genes. TET methylcytosine dioxygenases convert 5-methylcytosine (5-mC) to 5hydroxymethylcytosine (5-hmC) during DNA demethylation events. Here, we determine the spatiotemporal expression of TET1, TET2, and TET3 in specific trophoblast cell populations of mouse and human placentas throughout gestation, and consider their role in trophoblast cell differentiation and function. Methods: In situ hybridization analysis was conducted to localize Tet1, Tet2, and Tet3 mRNA at key stages of mouse placental development. The distribution of 5-mC and 5-hmC in these samples was also evaluated. In comparison, expression patterns of TET1, TET2, and TET3 protein in human placentas were determined in first trimester and term pregnancies. Results: In mouse, Tet1-3 mRNA was widely expressed in trophoblast cell populations from embryonic (E) day 8.5 to E12.5 including in progenitor and differentiated cells. However, expression became restricted to specific trophoblast giant cell subtypes by late gestation (E14.5 to E18.5). This coincided with cellular changes in 5-mC and 5-hmC levels. In human, cell columns, extravillous trophoblast and syncytiotrophoblast expressed TET1-3 whereas only TET3 was expressed in villus cytotrophoblast cells in first trimester and term placentas. Discussion: Altogether, our data suggest that TET enzymes may play a dynamic role in the regulation of transcriptional activity of trophoblast progenitors and differentiated cell subtypes in mouse and human placentas. © 2017 Elsevier Ltd. All rights reserved.

Keywords: 5-Methylcytosine 5-Hydroxymethylcytosine DNA methylation Trophoblast Placenta development

1. Introduction

Abbreviations: 5-caC, 5-carboxylcytosine; 5-fC, 5-formylcytosine; 5-hmC, 5hydroxymethylcytosine; 5-mC, 5-methylcytosine; DNMT, DNA methyltransferase; E, embryonic day; EPC, ectoplacental cone; ES cells, embryonic stem cells; EVT, extravillous cytotrophoblast; GlyT cells, glycogen trophoblast cells; ICM, inner cell mass; SpT, spongtiotrophoblast; SynT, syncytiotrophoblast; TET, Ten eleven translocation methylcytosine dioxygenase; TGCs, trophoblast giant cells. * Corresponding author. Department of Physiology, Development and Neuroscience, Physiology Building, Downing Street, Cambridge CB2 3EG, UK. E-mail address: [email protected] (E.D. Watson). 1 Present address: Department of Cancer & Stem Cell Biology, Duke-NUS Medical School, Singapore. https://doi.org/10.1016/j.placenta.2017.09.008 0143-4004/© 2017 Elsevier Ltd. All rights reserved.

The transition from progenitor to differentiated cell requires dynamic epigenetic regulation. DNA methylation is a stable epigenetic mark that is generally associated with transcriptional silencing. Occurring mainly at CpG dinucleotides, DNA methylation is catalysed by DNA methyltransferases (DNMTs). Conversely, the Ten eleven translocation (TET) family of methylcytosine dioxygenases actively demethylate cytosines by targeting 5-methylcytosine (5-mC) for oxidation to generate 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5-fC), and 5-carboxylcytosine (5-caC) [1,2]. In general, 5-hmC is transient in the epigenome and found at a lower levels than 5-mC [3]. However, some cells contain high levels

J. Rakoczy et al. / Placenta 59 (2017) 46e56

of 5-hmC, such as embryonic stem cells (ESCs) [4] and Purkinje neurons [5]. While the role of 5-mC in transcriptional regulation is well studied, the function of 5-hmC is less clear [6,7]. Nonetheless, DNA modified with 5-hmC is generally present in euchromatin suggesting that it promotes transcriptional activity [8,9]. Trophoblast cells have hypomethylated DNA relative to the embryo [10], which is necessary for normal placental development and therefore, a successful pregnancy. Gene knockout studies indicate that TET enzymes are critical for differentiation of a wide range of cell populations including ESCs [11], blood [12], and skeletal cells [13]. However, the role of TET enzymes in trophoblast differentiation and function is unclear. While architecturally different, both humans and mice have hemochorial placentas that comprises of three main layers: an outer maternal layer containing decidual cells and vasculature that brings maternal blood to and from the conceptus; a ‘junctional’ region that attaches the placenta to the uterus and consists of trophoblast cells with invasive and/or endocrine functions; and an inner layer composed of highly branched placental villi consisting of outer trophoblast cells that are bathed in maternal blood and surround an inner core of stroma and fetal capillaries [14]. Placenta villi are where maternal and fetal blood come in close proximity for nutrient and waste exchange. During human placental development, extravillous cytotrophoblast (EVT) cells differentiate from progenitors in cell columns and invade into the uterine spiral arteries [15]. Villous formation results when the chorionic plate gives rise to stem villi, which branch to form terminal villi with a continuous layer of multinucleated syncytiotrophoblast cells that directly contact maternal blood [15]. Beneath these cells lie a progenitor population of cytotrophoblast that fuse to form syncytiotrophoblast cells [15]. In mice, primary trophoblast giant cells (TGCs), which provide a barrier between the maternal decidua and the conceptus, the ectoplacental cone (EPC), and the chorion form by embryonic (E) day 8.0 [16]. The EPC contains progenitor cells of the junctional zone including spongiotrophoblast (SpT) cells, glycogen trophoblast (GlyT) cells and some TGC subtypes [17,18]. Alternatively, the chorion contains syncytiotrophoblast progenitor cells [19]. Upon chorioallantoic attachment at E8.5, branching morphogenesis is initiated and villi form the labyrinth layer [19]. The labyrinth is functional by E10.5 when maternal and fetal blood circulations are established and separated by a trilaminar layer of trophoblast cells including the sinuisoidal-TGCs that contact maternal blood, and a bilayer of syncytiotrophoblast cells that surround fetal capillaries [14,20]. Defects in mouse and human placenta development have deleterious effects on fetal development [14]. Though some differences exist, mechanisms of epigenetic regulation are largely thought to be analogous in mouse and human placentas. Both models contain imprinted genes that are regulated by DNA methylation and are critical for placental development (e.g. Igf2, H19, Mest, Cdkn1c, etc.) [21]. As in the mouse, 5-mC, 5-hmC, and histone methylation marks are present in human trophoblast cells during gestation [22]. In the early mouse embryo, patterns of 5-mC and 5-hmC differ between progenitor and differentiated trophoblast cells [23] suggesting differential regulation of methylation as cells become specified. In ESCs [11] and the embryo proper [24], TET enzymes promote differentiation, though similar evidence is lacking in the trophoblast lineage. Placental development is divided into the ‘development phase’ when trophoblast cell proliferation and differentiation occurs, and the ‘mature phase’ when the placenta increases in size and blood volume due to cell morphogenesis and dilation of blood vessels [25]. Indeed, the transition from the ‘development’ to ‘mature’ phase is marked by a transcriptional change of ~4000 genes including cell cycle, cell growth, and morphogenesis genes [26]. It is

47

hypothesized that DNA methylation machinery may be implicated in this process, although the extent to which TET enzymes are involved is unknown. Loss of Tet expression in the Tet1 knockout [27] and the Tet1/2 double knockout [28] mice results in postnatal and prenatal growth restriction, respectively, which suggest a role in placental development and function. To better understand epigenetic dynamics in trophoblast cells between early and late pregnancy, we investigate the spatiotemporal expression of TET1, TET2, and TET3 in correlation with nuclear 5-mC and 5-hmC levels in specific trophoblast cell subtypes throughout mouse and human placental development. 2. Methods 2.1. Mouse dissections This research was regulated under the Animals (Scientific Procedures) Act 1986 Amendment Regulations 2012 following ethical review by the University of Cambridge Animal Welfare and Ethical Review Body (AWERB). C57Bl/6 mice (http://jaxmice.jax.org) were mated to generate conceptuses at E8.5, E10.5, E12.5, E14.5, E16.5, and E18.5. Noon of the day that the copulatory plug was detected was designated E0.5. At least three placentas per time point were assessed. C57Bl/6 adult male brains were dissected for use as a positive control. Tissue was fixed overnight in 4% paraformaldehyde (PFA)/1x phosphate buffered saline (PBS) at 4
C. Samples were processed for cryosectioning (10 mm) or paraffin sectioning (7 mm). 2.2. Human placentas First trimester placentas (N ¼ 4) and decidua basalis samples (N ¼ 3) were collected between 7 and 11 weeks gestation with informed written consent and the approval of the Joint UCL/UCLH Committees on the Ethics of Human Research (05/Q0505/82) from patients undergoing surgical termination of pregnancy under general anesthesia for psycho-social reasons. Gestational age was confirmed by ultrasound measurement of the crown rump length of the embryo. Term placentas (N ¼ 3) were collected from normal term singleton pregnancies delivered by elective caesarean section at 39-weeks gestation with informed written consent of the patients and permission of the Cambridgeshire 2 Research Ethics Committee (03/360). All tissue samples were fixed in 4% PFA/1x PBS and embedded for paraffin sectioning. 2.3. In situ hybridization Synthesis of digoxigenin-labelled sense and anti-sense probes was performed by PCR amplification of cDNA templates using DIG RNA Labelling Mix (Roche, Cat. No. 11277073910) and gene-specific primers (Supplementary Table 1). In situ hybridization was performed as previously described [29]. Briefly, for probes against Tet1, Tet2, and Tet3, cryosections were rehydrated in 1x PBS under RNase-free conditions, treated with 30 mg/ml proteinase K (Roche) in 50 mM Tris/2.5 mM EDTA for 10 min at room temperature, acetylated with 0.25% acetic anhydride (Sigma) in 0.1M triethanolamine (Sigma) buffer for 10 min, and hybridized with ~50 ng of DIG-labelled riboprobe per section in 200 ml hybridization buffer (Amresco) overnight at 55
C. Sense probes were generated as negative controls, and were used in conditions similar to above. For probes against Ctsq and Tpbpa, the same protocol was performed with the following modifications. Dewaxed and rehydrated paraffin sections were treated with 30 mg/ml proteinase K for 20 min at room temperature and probe hybridization occurred at 65
C overnight. All posthybridization washes, antibody detection of DIG-

48

J. Rakoczy et al. / Placenta 59 (2017) 46e56

labelled probes, and colour development were performed as previously described in detail [29]. Sections were counterstained with Nuclear Fast Red (Sigma-Aldrich). A NanoZoomer 2.0-RS (Hamamatsu) and NDP.view2 viewing software program (Hamamatsu) were used to obtain images. 2.4. Immunohistochemistry Tissue sections were boiled in a pressure cooker in 0.01M sodium citrate, 0.05% Tween 20 (pH 6) or 0.01M Tris-EDTA buffer, 0.05% Tween 20 (pH 9), and blocked in host serum for 1 h at room temperature. Primary and secondary antibody dilutions were made in 5% host serum, 1% bovine serum albumin (Sigma-Aldrich) in 1x Tris-buffered saline with 0.1% Tween-20 (Sigma). Primary antibodies and dilutions included: anti-5-hmC (1:300; 39769, Active Motif), anti-5-mC (1:100; BI-MECY-0100, Eurogentec), anti-TET1 (1:100, NBP1-78966, Novus Biologicals), anti-TET2 (1:100; EB09642, Everest Biotech), anti-TET3 (1:50, sc-139186, Santa Cruz), anti-F4/80 (1:100, MCA497R, BioRad), anti-cytokeratin 7 (1:100, M7018, Dako), and biotinylated DBA lectin (6.6 mg/ml, B-1035, Vector Laboratories). Secondary antibodies and dilutions included: horseradish peroxidase (HRP)-conjugated donkey anti-rabbit IgG (1:300; ab6802, Abcam), HRP-conjugated donkey anti-goat IgG (1:300; ab6885, Abcam), biotin-conjugated goat anti-mouse IgG (1:300; 115-065-205, Jackson ImmunoResearch), and ImmPRESS™ reagent anti-rat IgG kit (MP-7444, Vector Laboratories). DNA was counterstained with haematoxylin. For negative controls, placenta tissue sections were treated as above but incubated with the secondary antibody only. 2.5. qRT-PCR Total RNA was extracted from whole placentas using TRIzol reagent (Sigma-Aldrich). Reverse-transcription was performed with a RevertAid H Minus First Strand cDNA Synthesis Kit (Fermentas) using 2 mg RNA/20 ml reaction. PCR amplification was conducted using MESA SYBR® Green qPCR MasterMix Plus (Eurogentec) on a DNA Engine Opticon® 2 system (MJ Research, Inc.). Transcript levels were normalised to Hprt1 and Polr2a RNA levels. Fold change was quantified using the DDCt method. Experiments were conducted in triplicate with five biological replicates. For primer sequences, refer to Supplemental Table 1. 2.6. Statistics Relative mRNA expression levels were calculated using a oneway ANOVA with a Tukey post-hoc test using GraphPad Prism 7 software. p < 0.05 was considered significant. 3. Results 3.1. Tet mRNA is widely expressed in mouse trophoblast progenitors, yolk sac, and maternal decidua at E8.5 To explore Tet1-3 mRNA expression in mouse placenta throughout gestation, in situ hybridization was performed on histological sections. Sense riboprobes against Tet1-3 were used on placental tissue as negative controls (Figs. 1e4; Supplementary Fig. S1). To ensure riboprobe specificity, adult mouse brain tissue, where Tet gene expression is known [59,60], was used as a positive control (Supplementary Fig. S2). At E8.5, Tet1, Tet2, and Tet3 mRNA expression was apparent throughout the chorion and EPC (Fig. 1). The nuclei of chorionic trophoblast cells ubiquitously immunostained for 5-mC (Fig. 1AB). In contrast, heterogeneous levels of 5hmC were apparent in chorionic cells: regions of low 5-hmC levels

associated with branch-point initiation sites (Fig. 1Y) where the cells have begun differentiating into syncytiotrophoblast [30]. In contrast, EPC cells displayed strong 5-hmC and weak 5-mC levels (Fig. 1Z,AC). Parietal-TGCs were also positive for Tet mRNA (Fig. 1G,O,W), exhibited strong 5-hmC staining (Fig. 1AA), and lower levels of 5-mC (Fig. 1AD). Cytoplasmic staining for 5-mC was also present in parietal-TGCs at E8.5 (Fig. 1AD), though this is likely background staining as indicated by the negative controls (Supplementary Fig. S1D). The yolk sac, important for nutrient transfer and haematopoiesis in early gestation [31], also expressed Tet1-3 mRNA at E8.5. Tet1 and Tet3 mRNA was detected throughout the visceral yolk sac (Fig. 1F,V), though Tet2 mRNA was limited to its mesothelial layer (Fig. 1N). Tet expression was undetectable in the parietal yolk sac (Fig. 1F,N,V). This data suggests a role for TETs in extraembryonic tissue at E8.5 beyond trophoblast cells. Additionally, maternal decidua also expressed Tet genes at E8.5. Although Tet2 and Tet3 mRNA was widely expressed throughout the decidua (Fig. 1P, X), Tet1 mRNA expression was restricted to a specific mesometrial cell population (Fig. 1H; Supplementary Fig. S3). To identify the Tet1þ cells, we immunostained for markers specific for two main leukocyte subsets in the mouse decidua including tissue-resident uterine natural killer cells (DBA lectinþ) and macrophages (F4/80þ) on serial sections adjacent to those stained for Tet1. Neither F4/80 nor DBA lectin was expressed in Tet1þ cells (Fig. 1H; Supplementary Fig. S3) indicating that Tet1þ cells located in the maternal decidua are yet-to-be identified. 5-hmC and 5-mC staining within the decidua was widespread (Supplementary Fig. S3) and most similar to Tet2 and Tet3 mRNA expression. 3.2. Tet mRNA expression becomes spatially restricted in the mouse placenta by late gestation To determine whether Tet genes are expressed in mid-to late gestation, we performed qRT-PCR analysis on whole mouse placentas at E10.5, E14.5, and E18.5. Tet1-3 mRNA expression was consistently highest at E10.5 with levels decreasing by E18.5 (Fig. 2AeC). Next, the spatial localization of Tet gene expression was determined in mouse placentas at E10.5 (Fig. 2), E14.5 (Fig. 3), and E18.5 (Fig. 4) by in situ hybridization. Tet1-3 mRNA was detected in the labyrinth at E10.5-E12.5 including syncytiotrophoblast cells and sinusoidal-TGCs (Fig. 2E,J,O; Supplementary Fig. S4). By E14.5E18.5, a gradual restriction of Tet mRNA to the sinusoidal-TGCs was apparent (Fig. 3B,G,L, 4B,I,P; Supplementary Fig. S4), which might indicate that DNA methylation is dynamic in sinusoidal-TGCs after differentiation is complete. To show specificity of Tet mRNA to sinusoidal-TGCs, serial sections of placentas at E18.5 were stained for Tet1-3 and the sinusoidal-TGC marker Ctsq (Supplementary Fig. S2). Immunostaining for 5-mC and 5-hmC revealed that nuclear 5mC levels were maintained in all labyrinth cells from E10.5-E18.5 (Figs. 2U, 3R and 4X). While syncytiotrophoblast nuclei had high 5mC levels throughout gestation, sinusoidal-TGC nuclei revealed patchy 5-mC immunostaining. Alternatively, high 5-hmC levels were present in all labyrinth trophoblast cells at E10.5 (Fig. 2S) but levels gradually decreased by E18.5 (Figs. 3P and 4V) when sinusoidal-TGC nuclei revealed irregular 5-hmC immunostaining and syncytiotrophoblast nuclei did not display appreciable levels of 5-hmC (Fig. 4V). Altogether, these data are consistent with Tet expression in the labyrinth and suggest restricted activity of TET enzymes in the late placenta. Beyond the labyrinth, all Tets were expressed in the EPC/SpT cells at E10.5-E18.5 (Fig. 2F,K,P, 3C,H,M, 4C,J,Q; Supplementary Fig. S4). In GlyT cells, which are only apparent between E14.5E16.5 and identifiable by morphology, Tet1-3 were strongly

J. Rakoczy et al. / Placenta 59 (2017) 46e56

49

Fig. 1. The mRNA of all Tets is widely expressed throughout the mouse conceptus at E8.5. (A-X) Spatial localization of (A-H) Tet1, (I-P) Tet2, and (Q-X) Tet3 mRNA expression (purple) in representative histological sections of whole mouse conceptuses at E8.5 as determined by in situ hybridization. Sections are counterstained with Nuclear Fast Red (pink). (A, I, Q) A whole conceptus at E8.5. Dotted line and yellow arrows indicate ectoplacental cone. White arrowheads indicate parietal-TGCs. Higher magnification images are shown to the right to depict Tet mRNA expression in the (B-C, J-K, R-S) chorion, (B, D, J, L, R, T) ectoplacental cone (dotted line), (F, N, V) mesothelium (black arrow) and endoderm (black arrowheads) of the visceral yolk sac, (G, O, W) parietal-TGCs (white arrowheads), and (H, P, X) maternal decidua. Boxes in B and R indicates region of higher magnification in G and W, respectively. (H) Serial sections of maternal decidua showing Tet1 mRNA expression (purple) and either the macrophage marker F4/80 (brown) or the uterine natural killer cell marker DBA-lectin (brown). Matching coloured arrows indicate the same cell in serial sections. (E, M, U) Sense negative controls of ectoplacental cone and chorion. (Y-AD) Immunostaining of (Y-AA) 5-hmC (brown) and (AB-AD) 5-mC (brown) in the (Y, AB) chorion, (Z, AC) ectoplacental cone, and (AA, AD) parietal-TGCs of mouse placentas at E8.5. DNA stained with hematoxylin (blue). Red arrowheads indicate chorion branchpoint initiation sites. Ch, chorionic trophoblast layer; Dec, maternal decidua; Em, embryo; EPC, ectoplacental cone; Me, chorion mesothelial layer; P-TGC, parietal trophoblast giant cell. Scale bars: A, H, O ¼ 1 mm; B, I, P ¼ 100 mm; C-G, J-N, Q-U, V-AA ¼ 25 mm.

expressed (Fig. 3C,H,M; Supplementary Fig. S4). Serial sections of placentas were stained for Tet1-3 and the spongiotrophoblast layer specific marker Tpbpa to emphasize cell-specificity (Supplementary Fig. S2). Immunostaining of 5-mC in SpT and GlyT cells was consistent from E10.5-E18.5 (Figs. 2V, 3S and 4Y; Supplementary Fig. S4). Most SpT and GlyT nuclei displayed 5-hmC throughout gestation (Figs. 2T, 3Q and 4W). However, SpT cells at E18.5 showed heterogeneous immunostaining for 5-hmC with some cells showing robust levels and others showing weak levels (Fig. 4W). 3.3. Tet mRNA is expressed in all mouse trophoblast giant cell subtypes We sought to determine whether other TGC subtypes expressed

Tet mRNA. Regardless of their placental location, TGCs line maternal blood sinuses [32,33] and produce hormones and cytokines that regulate maternal adaptations to pregnancy [34]. While parietalTGCs, which separate the spongiotrophoblast layer from the maternal decidua and are identified by their relatively large nuclei, expressed Tet2-3 mRNA in a heterogeneous manner at E10.5 (Fig. 5BeC), Tet1 expression was not appreciable (Fig. 5A). By E14.5E18.5, all parietal-TGCs expressed Tet1-3 mRNA (Fig. 5FeG,K-M) with Tet3 expression displaying a considerable reduction by E18.5 (Fig. 5M). Reciprocal patterns of 5-hmC and 5-mC in parietal-TGCs were observed with 5-hmC levels gradually decreasing (Fig. 5D,I,N) and 5-mC levels steadily increasing (Fig. 5E,J,O) by E18.5 reflecting Tet3 expression in particular. At E14.5-E18.5, Tet1-3 mRNA was detectable in all canal-TGCs,

50

J. Rakoczy et al. / Placenta 59 (2017) 46e56

Fig. 2. Relatively high expression of Tet1-3 mRNA in mouse placenta at E10.5. (A-C) qRT-PCR analysis of (A) Tet1, (B) Tet2, and (C) Tet3 mRNA expression levels in whole mouse placentas at E10.5, E14.5, and E18.5. Data is depicted as mean ± standard error. Expression levels were normalised to both Hprt1 and Polr2a mRNA levels and depicted as an average. N ¼ 5 placentas per time point. Statistical significance was calculated by one-way ANOVA with Tukey posthoc test. xP ¼ 0.05, *P < 0.05, **P < 0.01, ****P < 0.0001. (D-P) Spatial localization of (D-F) Tet1, (I-K) Tet2, and (N-P) Tet3 mRNA expression in representative histological sections of mouse placentas at E10.5 as determined by in situ hybridization (purple). Sections were counterstained with Nuclear Fast Red (pink). (G-H, L-M, Q-R) Negative controls showing sense probe expression in the (G, L, Q) labyrinth and (H, M, R) spongiotrophoblast. (D, I, N) Whole placentas at E10.5. Boxes indicate regions of higher magnification in panels to the right depicting the (E, J, O) labyrinth layer and (F, K, P) spongiotrophoblast layer (dotted line). (S-V) Immunostaining of (S-T) 5-hmC and (U-V) 5-mC in placentas at E10.5. Representative regions of the (S, U) labyrinth layer and (T, V) spongiotrophoblast layer are shown. Arrows, parietal-TGCs; arrowheads, sinusoidal-TGCs; Dec, maternal decidua; Lab, labyrinth; SpT, spongtiotrophoblast layer; f, fetal capillary; m, maternal blood sinusoid. Scale bars: D, I, N ¼ 1 mm; E-F, J-K, O-P ¼ 25 mm, G-H, L-M, Q-R ¼ 100 mm.

which surround the spaces that carry oxygenated maternal blood to the base of the labyrinth (Fig. 5PeR,U-W), and both 5-hmC and 5-mC levels were steady (Fig. 5SeT,X-Y). In contrast, spiral artery-TGCs that surround the spiral arteries in the maternal decidua expressed all three Tets at E18.5 and not E14.5 (Fig. 5ZeAB,AE-AG). Correspondingly, 5-mC levels were strong in spiral artery-TGCs at E14.5 (Fig. 5AD) whereas 5-hmC levels were low (Fig. 5AC).

3.4. TET proteins are expressed in the human placenta throughout gestation To assess the spatiotemporal pattern of TET expression in human placenta, histological sections from first-trimester placenta and decidua basalis (7e11 weeks gestation) and term (39-weeks gestation) pregnancies were immunostained for TET1-3 proteins. Histological sections immunostained with the secondary antibody only were used as negative controls (Supplementary Fig. S5). All

J. Rakoczy et al. / Placenta 59 (2017) 46e56

51

Fig. 3. Tet mRNA is expressed in mouse placenta at E14.5. (A-O) Spatial localization of (A-C) Tet1, (F-H) Tet2, and (K-M) Tet3 mRNA expression in histological sections of mouse placentas at E14.5 as determined by in situ hybridization (purple). Sections were counterstained with Nuclear Fast Red (pink). (A, F, K) Whole placentas at E14.5. Boxes indicate regions of higher magnification in panels to the right depicting the (B, G, L) labyrinth layer and (C, H, M) spongiotrophoblast layer. Negative controls showing sense probe expression in (D, I, N) labyrinth and (E, J, O) spongiotrophoblast. (P-S) Immunostaining of (P-Q) 5-hmC and (R-S) 5-mC in the (P, R) labyrinth layer and (Q, S) spongiotrophoblast layer at E14.5. Arrowheads indicate sinusoidal-TGCs. The dotted lines in A, F, and K indicate the border between the labyrinth and the spongiotrophoblast layers. The dotted lines in C, E, H, J, M and O indicate the border between spongiotrophoblast cells and glycogen trophoblast cells. Dec, maternal decidua; f, fetal capillaries; GlyT, glycogen trophoblast cells; Lab, labyrinth; m, maternal blood sinusoids; SpT, spongiotrophoblast. Scale bars: A, F, K ¼ 1 mm; B-C, G-H, L-M, P-S ¼ 25 mm, D-E, I-J, N-O ¼ 100 mm.

three TETs were expressed in trophoblast cell columns of firsttrimester placentas (Fig. 6AeB,I-J,Q-R). Cell columns include the progenitors of EVT cells nearest the villi that gradually differentiate as the cells move towards the decidua [35]. Based on cell location, TET1 expression appeared less robust in column trophoblast progenitors than in differentiated column cells (Fig. 6AeB). Maternal decidua from first-trimester pregnancies expressed all three TET proteins in a subset of EVT cells (Fig. 6CeD,K-L,S-T) as confirmed by immunostaining serial sections with the pan-trophoblast cell marker cytokeratin 7 (Supplementary Fig. S5AeF). In placental villi, TET1-3 proteins were detected in syncytiotrophoblast cells at all pregnancy stages assessed (Fig. 6FeH,M-P,U-X) with the exception of villi at 8-weeks when TET1 was undetectable in all villus cells (Fig. 6E). Villus cytotrophoblast cells, which fuse to form the syncytiotrophoblast layer upon differentiation, expressed TET3 protein

from 10-weeks gestation until term (Fig. 6UeX). Otherwise, TET1 and TET2 were undetectable in villus cytotrophoblast cells at all stages assessed (Fig. 6EeH,M-P). Interestingly, both cytoplasmic and nuclear TET immunostaining was present (Fig. 6). Altogether, this data suggests that TET1-2 enzymes are largely restricted to differentiated villus trophoblast (i.e., syncytiotrophoblast) in first trimester and term human placentas whereas TET3 expression is present in syncytiotrophoblast cells and their progenitors from early to late gestation. TET expression patterns in term placentas observed in our study were consistent with those observed in the Human Protein Atlas (http://www.proteinatlas.org). 4. Discussion Investigation into the regulation of DNA methylation during

52

J. Rakoczy et al. / Placenta 59 (2017) 46e56

Fig. 4. Tet mRNA is no longer detected in syncytiotrophoblast cells in mouse placenta at E18.5. Spatial localization of (A-D) Tet1, (H-K) Tet2, and (O-R) Tet3 mRNA expression in histological sections of mouse placentas at E18.5 as determined by in situ hybridization (purple). DNA was counterstained with Nuclear Fast Red (pink). (A, H, O) Histological sections of whole placentas at E18.5. Labelled boxes indicate regions of higher magnification in panels to the right depicting the (B, I, S) labyrinth layer, (C, J, Q) spongiotrophoblast layer, and (D, K, R) chorionic plate. (E-G, L-N, S-U) Similar regions are depicted for negative controls showing sense probe expression. (V-Y) Immunostaining of (V-W) 5-hmC and (XY) 5-mC in the (V, X) labyrinth layer and (W, Y) spongiotrophoblast layer at E18.5. DNA was stained with hematoxylin (blue). Arrowheads indicate sinusoidal-TGCs. The dotted lines in A, H, and O indicate the border between the labyrinth and the spongiotrophoblast layers. Dec, maternal decidua; Lab, labyrinth; GlyT, glycogen trophoblast cells; SpT, spongiotrophoblast. Scale bars: A, H, O ¼ 1 mm, B-D, I-K, P-R ¼ 25 mm, E-G, L-N, S-U ¼ 100 mm.

development has revealed important roles for TET dioxygenases in ESCs [11] and embryonic development [24,28]. Despite this, information regarding TET activity in the placenta is scarce. Here, we demonstrate that Tets are expressed across gestation in mouse trophoblast progenitors, a subset of differentiated trophoblast cells, and other extraembryonic cells. All three mouse Tets showed similar mRNA expression patterns that correlated with 5-mC and 5hmC levels in specific trophoblast subtypes. Similarly, TET1-3 proteins were expressed in a subset of EVTs and their progenitors in the first trimester human placenta. Despite this, differential TET expression was apparent in human placental villi. In early and late gestation, TET1 and TET2 were detected in syncytiotrophoblast cells whereas TET3 was expressed in both syncytiotrophoblast and

their cytotrophoblast progenitors suggesting a possible unique role for TET3 cytotrophoblast differentiation and function. In support of our findings, others have shown that the ratio of 5-hmC:5-mC levels is much higher in syncytiotrophoblast cells than villus cytotrophoblast across gestation [22]. The reverse is true of villus cytotrophoblast where 5-mC was far more robust than 5 hmC [22]. While TET expression in progenitor cells might be expected, these data suggest that terminally differentiated trophoblast cells also require TET enzymatic activity. This might be to maintain an epigenetically dynamic transcriptional state throughout pregnancy. Cell differentiation requires changes to DNA and histone methylation marks to repress ‘progenitor genes’ and to upregulate ‘differentiation genes’ [36]. Therefore, TET regulation of DNA

J. Rakoczy et al. / Placenta 59 (2017) 46e56

53

Fig. 5. Differential Tet mRNA expression in trophoblast giant cell subtypes in mouse placentas from E10.5-E18.5. mRNA expression of Tet1, Tet2, and Tet3 in specific trophoblast giant cell (TGC) subtypes as determined by in situ hybridization including (A-C, F-H, K-M) parietal-TGCs (P-TGCs; black arrowheads), (P-R, U-W) canal-TGCs (C-TGCs; white arrowheads), and (Z-AB, AE-AG) spiral artery-TGCs (SpA-TGCs; arrows) in mouse placentas from E10.5-E18.5. Immunostaining of (D, I, N, S, X, AC, AH) 5-hmC (brown) and (E, J, O, T, Y, AD, AI) 5-mC (brown) in specific TGC subtypes at E10.5-E18.5. In situ hybridization stained sections were counterstained with Nuclear Fast Red (pink). Immunostained sections were counterstained with hematoxylin (blue). M, maternal blood sinusoid. Scale bars: 25 mm.

methylation during the developmental phase of the placenta (E8.5E13.5) would help to promote the transcription of cell cycle and differentiation genes. Alternatively, during the mature placenta phase (E14.5-E18.5), Tet mRNA was restricted to specific terminallydifferentiated trophoblast subtypes including invasive spiral artery-TGCs, sinusoidal-TGCs of the labyrinth, and SpT and GlyT cells of the junctional zone. Importantly, these Tet-expressing cells, similar to human syncytiotrophoblast cells where TET1-3 proteins were also expressed, have an endocrine function whereby they produce hormones and cytokines required to modulate maternal physiology during pregnancy, parturition, lactation, and offspring rearing [34,37e39]. Genes encoding hormones must be responsive throughout pregnancy, and TET enzymes might help to regulate this process by promoting the demethylation of DNA and transcriptional activity. Examples of such secreted factors in mouse may include the following. Rodent TGCs, SpT and GlyT cells secrete prolactin-like proteins at mid-to late gestation [26] into the maternal circulation to regulate maternal adaptation to pregnancy

[29,40,41]. SpT cells limit uterine endothelium growth by expressing anti-angiogenic proteins FLT1 and proliferin [42,43]. While control of these genes by DNA methylation has not yet been demonstrated in the placenta, it is possible that TET activity is required to ensure normal function of differentiated trophoblast. Due to gastrulation defects leading to the early lethality of Tet1/ 2/3 triple knockout mice [24], it is not possible to investigate the collective role of TETs in the regulation of DNA methylation during trophoblast differentiation and function in these mutants. However, Tet1 single knockout and Tet1/2 double knockout in mice result in growth restriction at birth or at midgestation, respectively [27,28]. Fetal growth restriction correlates to poor placental development and function [44e48]. Since Tet1 and Tet2 are widely expressed in trophoblast progenitors in mice and differentiated trophoblast cells in mice and humans, it is possible that TET1-2 play a yet-to-be determined role in trophoblast differentiation and function. Further exploration is required in the context of the knockout models. Similarly, loss of DNMTs results in DNA

54

J. Rakoczy et al. / Placenta 59 (2017) 46e56

Fig. 6. Differential TET expression in trophoblast cells of first and third trimester human placentas. Immunostaining of TET1, TET2, and TET3 (brown) in first-trimester and term placental histological sections. Sections were counterstained with hematoxylin (blue). Specific placental structures shown include: (A-B, I-J, Q-R) cell columns, (C-D, K-L, S-T) maternal decidua basalis, and (E-H, M-P, U-X) placental villi. Gestational stage is indicated in each panel. Boxed regions in (E-H, M-P, Q-T) are shown at a higher magnification in the insets of each panel. White arrowheads indicate villus cytotrophoblast cells. Black arrowheads indicate syncytiotrophoblast cells. v, villi; c, cell column; IVS, intervillus space. Scale bars: A-D, I-L, Q-T ¼ 20 mm, E-H, M-P, U-X ¼ 100 mm (insets ¼ 25 mm).

demethylation and phenotypes, such as growth restriction, developmental delay, placental phenotypes (e.g., chorion defects), and embryonic lethality. Mouse Dnmt3L mRNA is expressed predominately in the chorion trophoblast, which small in Dnmt3L/ conceptuses leading to placental insufficiency that is associated with embryonic growth restriction [61]. Dnmt1 knockout and Dnmt3a/3b double knockout mice result in developmental delay, growth restriction, and/or embryonic lethality by E10.5 [49e51]. Embryonic lethality at E10.5 is often associated with a placental phenotype [14]. Furthermore, maternal loss of Dnmts causes a chorioallantoic attachment defect [52], a process that is required for placental development [14]. Together, these models indicate that regulation of DNA methylation is likely required for trophoblast differentiation and/or function. In support of this hypothesis, the Tet, 5-mC, and 5-

hmC expression data in our study show the presence of these epigenetic marks in trophoblast cells throughout gestation. Singlecell methylome and transcriptome techniques will be required to determine the epigenetic effects of Tet deficiency in specific trophoblast cell subtypes. The best known function for TET enzymes is the oxidation of 5mC to form 5-hmC, 5-fC, and 5-caC during the demethylation of CpG sites [53]. Therefore, we made an association between Tet mRNA expression and 5-hmC levels to indicate a role for TETs during trophoblast differentiation and function. The cells in which TETs were expressed also displayed high 5-hmC and lower 5-mC [this study; 33]. Apart from DNA demethylation, TET enzymes also regulate gene transcription through other mechanisms by directly recruiting O-GlcNAc transferase (OGT) to transcriptional start sites

J. Rakoczy et al. / Placenta 59 (2017) 46e56

to promote GlcNAcylation of histones and transcription [54,55]. Separately, TET proteins associate with active [56] and repressive [57] histone marks to epigenetically regulate gene expression during differentiation. To fully understand the role of TETs in the placenta, these alternative functions should also be considered. Interestingly, we observed cytoplasmic localization of TET immunostaining in the human placenta, which may indicate a new cytoplasmic function for TET or that TETs are sequestered in the cytoplasm to regulate their nuclear function as was previously shown in other cell types [58]. Continued expression of TET enzymes in trophoblast cells after differentiation likely indicates a dynamic epigenome and transcriptome in the placenta until birth. Defects in epigenetic remodeling, particularly during the transition between trophoblast progenitor and differentiated cells, may prevent a healthy placenta and thus a successful pregnancy. Further exploration of the role of TETs in placenta development and function is required. Conflict of interest

[12]

[13]

[14] [15]

[16] [17]

[18]

[19] [20]

None. [21]

Acknowledgements [22]

We thank Dr B. Musial for technical support. The following support was given: Newton International Fellowship to J.R., Centre for Trophoblast Research (CTR) graduate studentship to N.P., a CTR Next Generation Fellowship to J.K, and CTR funding to T.C.D. This work was funded by the Lister Institute for Preventative Medicine and an Isaac Newton Trust/Wellcome Trust/ISSF/University of Cambridge joint grant to E.D.W. Appendix A. Supplementary data

[23]

[24]

[25]

[26]

Supplementary data related to this article can be found at https://doi.org/10.1016/j.placenta.2017.09.008.

[27]

References

[28]

[1] S. Ito, L. Shen, Q. Dai, S.C. Wu, L.B. Collins, J.A. Swenberg, C. He, Y. Zhang, Tet proteins can convert 5-methylcytosine to 5-formylcytosine and 5carboxylcytosine, Science 333 (6047) (2011) 1300e1303. [2] M. Tahiliani, K.P. Koh, Y. Shen, W.A. Pastor, H. Bandukwala, Y. Brudno, S. Agarwal, L.M. Iyer, D.R. Liu, L. Aravind, A. Rao, Conversion of 5methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1, Science 324 (5929) (2009) 930e935. [3] M.A. Hahn, P.E. Szabo, G.P. Pfeifer, 5-Hydroxymethylcytosine: a stable or transient DNA modification? Genomics 104 (5) (2014) 314e323. [4] I. Choi, R. Kim, H.W. Lim, K.H. Kaestner, K.J. Won, 5-hydroxymethylcytosine represses the activity of enhancers in embryonic stem cells: a new epigenetic signature for gene regulation, BMC Genomics 15 (2014) 670. [5] S. Kriaucionis, N. Heintz, The nuclear DNA base 5-hydroxymethylcytosine is present in Purkinje neurons and the brain, Science 324 (5929) (2009) 929e930. [6] D. Globisch, M. Munzel, M. Muller, S. Michalakis, M. Wagner, S. Koch, T. Bruckl, M. Biel, T. Carell, Tissue distribution of 5-hydroxymethylcytosine and search for active demethylation intermediates, PLoS One 5 (12) (2010) e15367. [7] C.X. Song, K.E. Szulwach, Y. Fu, Q. Dai, C. Yi, X. Li, Y. Li, C.H. Chen, W. Zhang, X. Jian, J. Wang, L. Zhang, T.J. Looney, B. Zhang, L.A. Godley, L.M. Hicks, B.T. Lahn, P. Jin, C. He, Selective chemical labeling reveals the genome-wide distribution of 5-hydroxymethylcytosine, Nat. Biotechnol. 29 (1) (2011) 68e72. [8] K.E. Szulwach, X. Li, Y. Li, C.X. Song, J.W. Han, S. Kim, S. Namburi, K. Hermetz, J.J. Kim, M.K. Rudd, Y.S. Yoon, B. Ren, C. He, P. Jin, Integrating 5hydroxymethylcytosine into the epigenomic landscape of human embryonic stem cells, PLoS Genet. 7 (6) (2011) e1002154. [9] G. Ficz, M.R. Branco, S. Seisenberger, F. Santos, F. Krueger, T.A. Hore, C.J. Marques, S. Andrews, W. Reik, Dynamic regulation of 5hydroxymethylcytosine in mouse ES cells and during differentiation, Nature 473 (7347) (2011) 398e402. [10] F. Santos, B. Hendrich, W. Reik, W. Dean, Dynamic reprogramming of DNA methylation in the early mouse embryo, Dev. Biol. 241 (1) (2002) 172e182. [11] M.M. Dawlaty, A. Breiling, T. Le, M.I. Barrasa, G. Raddatz, Q. Gao, B.E. Powell,

[29]

[30]

[31] [32]

[33]

[34] [35] [36]

[37] [38]

[39] [40]

[41]

55

A.W. Cheng, K.F. Faull, F. Lyko, R. Jaenisch, Loss of Tet enzymes compromises proper differentiation of embryonic stem cells, Dev. Cell 29 (1) (2014) 102e111. H. Yan, Y. Wang, X. Qu, J. Li, J. Hale, Y. Huang, C. An, J. Papoin, X. Guo, L. Chen, Q. Kang, W. Li, V.P. Schulz, P.G. Gallagher, C.D. Hillyer, N. Mohandas, X. An, Distinct roles for TET family proteins in regulating human erythropoiesis, Blood 129 (14) (2017) 2002e2012. X. Zhong, Q.Q. Wang, J.W. Li, Y.M. Zhang, X.R. An, J. Hou, Ten-eleven Translocation-2 (Tet2) is involved in myogenic differentiation of skeletal myoblast cells in vitro, Sci. Rep. 7 (2017) 43539. E.D. Watson, J.C. Cross, Development of structures and transport functions in the mouse placenta, Physiol. (Bethesda) 20 (2005) 180e193. E. Maltepe, A.I. Bakardjiev, S.J. Fisher, The placenta: transcriptional, epigenetic, and physiological integration during development, J. Clin. Invest. 120 (4) (2010) 1016e1025. J.C. Cross, Z. Werb, S.J. Fisher, Implantation and the placenta: key pieces of the development puzzle, Science 266 (5190) (1994) 1508e1518. M. Gasperowicz, A. Rai, J.C. Cross, Spatiotemporal expression of Notch receptors and ligands in developing mouse placenta, Gene Expr. Patterns 13 (7) (2013) 249e254. D.G. Simmons, A.L. Fortier, J.C. Cross, Diverse subtypes and developmental origins of trophoblast giant cells in the mouse placenta, Dev. Biol. 304 (2) (2007) 567e578. J.C. Cross, D.G. Simmons, E.D. Watson, Chorioallantoic morphogenesis and formation of the placental villous tree, Ann. N. Y. Acad. Sci. 995 (2003) 84e93. P.M. Coan, A.C. Ferguson-Smith, G.J. Burton, Ultrastructural changes in the interhaemal membrane and junctional zone of the murine chorioallantoic placenta across gestation, J. Anat. 207 (6) (2005) 783e796. E.C. Nelissen, A.P. van Montfoort, J.C. Dumoulin, J.L. Evers, Epigenetics and the placenta, Hum. Reprod. Update 17 (3) (2011) 397e417. N.M. Fogarty, G.J. Burton, A.C. Ferguson-Smith, Different epigenetic states define syncytiotrophoblast and cytotrophoblast nuclei in the trophoblast of the human placenta, Placenta 36 (8) (2015) 796e802. Y. Li, M.K. Seah, C. O'Neill, Mapping global changes in nuclear cytosine base modifications in the early mouse embryo, Reproduction 151 (2) (2016) 83e95. H.Q. Dai, B.A. Wang, L. Yang, J.J. Chen, G.C. Zhu, M.L. Sun, H. Ge, R. Wang, D.L. Chapman, F. Tang, X. Sun, G.L. Xu, TET-mediated DNA demethylation controls gastrulation by regulating Lefty-Nodal signalling, Nature 538 (7626) (2016) 528e532. M.Y. Rennie, J. Detmar, K.J. Whiteley, A. Jurisicova, S.L. Adamson, J.G. Sled, Expansion of the fetoplacental vasculature in late gestation is strain dependent in mice, Am. J. Physiol. Heart Circ. Physiol. 302 (6) (2012) H1261eH1273. K. Knox, J.C. Baker, Genomic evolution of the placenta using co-option and duplication and divergence, Genome Res. 18 (5) (2008) 695e705. M.M. Dawlaty, K. Ganz, B.E. Powell, Y.C. Hu, S. Markoulaki, A.W. Cheng, Q. Gao, J. Kim, S.W. Choi, D.C. Page, R. Jaenisch, Tet1 is dispensable for maintaining pluripotency and its loss is compatible with embryonic and postnatal development, Cell Stem Cell 9 (2) (2011) 166e175. M.M. Dawlaty, A. Breiling, T. Le, G. Raddatz, M.I. Barrasa, A.W. Cheng, Q. Gao, B.E. Powell, Z. Li, M. Xu, K.F. Faull, F. Lyko, R. Jaenisch, Combined deficiency of Tet1 and Tet2 causes epigenetic abnormalities but is compatible with postnatal development, Dev. Cell 24 (3) (2013) 310e323. D.G. Simmons, S. Rawn, A. Davies, M. Hughes, J.C. Cross, Spatial and temporal expression of the 23 murine Prolactin/Placental Lactogen-related genes is not associated with their position in the locus, BMC Genomics 9 (2008) 352. J.C. Cross, H. Nakano, D.R. Natale, D.G. Simmons, E.D. Watson, Branching morphogenesis during development of placental villi, Differentiation 74 (7) (2006) 393e401. W.P. Jollie, Development, morphology, and function of the yolk-sac placenta of laboratory rodents, Teratology 41 (4) (1990) 361e381. M. Hemberger, T. Nozaki, M. Masutani, J.C. Cross, Differential expression of angiogenic and vasodilatory factors by invasive trophoblast giant cells depending on depth of invasion, Dev. Dyn. 227 (2) (2003) 185e191. S.L. Adamson, Y. Lu, K.J. Whiteley, D. Holmyard, M. Hemberger, C. Pfarrer, J.C. Cross, Interactions between trophoblast cells and the maternal and fetal circulation in the mouse placenta, Dev. Biol. 250 (2) (2002) 358e373. D. Hu, J.C. Cross, Development and function of trophoblast giant cells in the rodent placenta, Int. J. Dev. Biol. 54 (2e3) (2010) 341e354. G.J. Burton, C.P. Sibley, E. Jauniaux, Placental Anatomy and Physiology, seventh ed., Elsevier, 2012. G.L. Sen, J.A. Reuter, D.E. Webster, L. Zhu, P.A. Khavari, DNMT1 maintains progenitor function in self-renewing somatic tissue, Nature 463 (7280) (2010) 563e567. L. Ogren, F. Talamantes, Prolactins of pregnancy and their cellular source, Int. Rev. Cytol. 112 (1988) 1e65. A. Malassine, J.L. Frendo, D. Evain-Brion, A comparison of placental development and endocrine functions between the human and mouse model, Hum. Reprod. Update 9 (6) (2003) 531e539. M.J. Soares, T. Konno, S.M. Alam, The prolactin family: effectors of pregnancydependent adaptations, Trends Endocrinol. Metab. 18 (3) (2007) 114e121. M.J. Soares, The prolactin and growth hormone families: pregnancy-specific hormones/cytokines at the maternal-fetal interface, Reprod. Biol. Endocrinol. 2 (2004) 51. R. Ain, L.N. Canham, M.J. Soares, Gestation stage-dependent intrauterine

56

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49] [50]

[51]

[52]

J. Rakoczy et al. / Placenta 59 (2017) 46e56 trophoblast cell invasion in the rat and mouse: novel endocrine phenotype and regulation, Dev. Biol. 260 (1) (2003) 176e190. D. Jackson, O.V. Volpert, N. Bouck, D.I. Linzer, Stimulation and inhibition of angiogenesis by placental proliferin and proliferin-related protein, Science 266 (5190) (1994) 1581e1584. Y. He, S.K. Smith, K.A. Day, D.E. Clark, D.R. Licence, D.S. Charnock-Jones, Alternative splicing of vascular endothelial growth factor (VEGF)-R1 (FLT-1) pre-mRNA is important for the regulation of VEGF activity, Mol. Endocrinol. 13 (4) (1999) 537e545. R. Oh-McGinnis, A.B. Bogutz, L. Lefebvre, Partial loss of Ascl2 function affects all three layers of the mature placenta and causes intrauterine growth restriction, Dev. Biol. 351 (2) (2011) 277e286. C. Hitz, D. Vogt-Weisenhorn, P. Ruiz, W. Wurst, T. Floss, Progressive loss of the spongiotrophoblast layer of Birc6/Bruce mutants results in embryonic lethality, Genesis 42 (2) (2005) 91e103. Q. Zheng-Fischhofer, M. Kibschull, M. Schnichels, M. Kretz, E. Petrasch-Parwez, J. Strotmann, H. Reucher, B.D. Lynn, J.I. Nagy, S.J. Lye, E. Winterhager, K. Willecke, Characterization of connexin31.1-deficient mice reveals impaired placental development, Dev. Biol. 312 (1) (2007) 258e271. S.L. Withington, A.N. Scott, D.N. Saunders, K. Lopes Floro, J.I. Preis, J. Michalicek, K. Maclean, D.B. Sparrow, J.P. Barbera, S.L. Dunwoodie, Loss of Cited2 affects trophoblast formation and vascularization of the mouse placenta, Dev. Biol. 294 (1) (2006) 67e82. S.J. Tunster, M. Van De Pette, R.M. John, Isolating the role of elevated Phlda2 in asymmetric late fetal growth restriction in mice, Dis. Model Mech. 7 (10) (2014) 1185e1191. E. Li, T.H. Bestor, R. Jaenisch, Targeted mutation of the DNA methyltransferase gene results in embryonic lethality, Cell 69 (6) (1992) 915e926. H. Lei, S.P. Oh, M. Okano, R. Juttermann, K.A. Goss, R. Jaenisch, E. Li, De novo DNA cytosine methyltransferase activities in mouse embryonic stem cells, Development 122 (10) (1996) 3195e3205. M. Okano, D.W. Bell, D.A. Haber, E. Li, DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development, Cell 99 (3) (1999) 247e257. M.R. Branco, M. King, V. Perez-Garcia, A.B. Bogutz, M. Caley, E. Fineberg,

[53] [54] [55]

[56]

[57]

[58]

[59]

[60]

[61]

L. Lefebvre, S.J. Cook, W. Dean, M. Hemberger, W. Reik, Maternal DNA methylation regulates early trophoblast development, Dev. Cell 36 (2) (2016) 152e163. R.M. Kohli, Y. Zhang, TET enzymes, TDG and the dynamics of DNA demethylation, Nature 502 (7472) (2013) 472e479. Q. Chen, Y. Chen, C. Bian, R. Fujiki, X. Yu, TET2 promotes histone O-GlcNAcylation during gene transcription, Nature 493 (7433) (2013) 561e564. Y. Kizuka, S. Kitazume, K. Okahara, A. Villagra, E.M. Sotomayor, N. Taniguchi, Epigenetic regulation of a brain-specific glycosyltransferase N-acetylglucosaminyltransferase-IX (GnT-IX) by specific chromatin modifiers, J. Biol. Chem. 289 (16) (2014) 11253e11261. T.S. Mikkelsen, M. Ku, D.B. Jaffe, B. Issac, E. Lieberman, G. Giannoukos, P. Alvarez, W. Brockman, T.K. Kim, R.P. Koche, W. Lee, E. Mendenhall, A. O'Donovan, A. Presser, C. Russ, X. Xie, A. Meissner, M. Wernig, R. Jaenisch, C. Nusbaum, E.S. Lander, B.E. Bernstein, Genome-wide maps of chromatin state in pluripotent and lineage-committed cells, Nature 448 (7153) (2007) 553e560. F. Neri, D. Incarnato, A. Krepelova, S. Rapelli, A. Pagnani, R. Zecchina, C. Parlato, S. Oliviero, Genome-wide analysis identifies a functional association of Tet1 and Polycomb repressive complex 2 in mouse embryonic stem cells, Genome Biol. 14 (8) (2013) R91. Y. Arioka, A. Watanabe, K. Saito, Y. Yamada, Activation-induced cytidine deaminase alters the subcellular localization of Tet family proteins, PLoS One 7 (9) (2012) e45031. G.A. Kaas, C. Zhong, D.E. Eason, D.L. Ross, R.V. Vachhani, G.L. Ming, J.R. King, H. Song, J.D. Sweatt, TET1 controls CNS 5-methylcytosine hydroxylation, active DNA demethylation, gene transcription, and memory formation, Neuron 79 (2013) 1086e1093. A. Szwagierczak, S. Bultmann, C.S. Schmidt, F. Spada, H. Leonhardt, Sensitive enzymatic quantification of 5-hydroxymethylcytosine in genomic DNA, Nucleic Acid. Res. 38 (19) (2010) e181. K. Hata, M. Okana, H. Lei, E. Li, Dnmt3L cooperates with the Dnmt3 family of de novo DNA methyltransferases to establish maternal imprints in mice 129 (8) (2002) 1983e1993.