Placental CRH production increases with advancing pregnancy in women and its course predicts gestational length. We hypothesized that CRH gene expression in the placenta is epigenetically controlled setting gestational trajectories characteristic of normal and pathological pregnancies. Here we determined histone modification and DNA methylation levels and DNA methylation patterns at the CRH promoter in primary trophoblast cultures by chromatin immunoprecipitation combined with clonal bisulfite sequencing and identified the transcriptionally active epialleles that associate with particular histone modifications and transcription factors during syncytialisation and cAMP-stimulation. CRH gene expression increased during syncytial differentiation and cAMP stimulation, which was associated with increased activating and decreased repressive histone modification levels at the promoter. DNA methylation levels remained unchanged. The nine CpGs of the CRH proximal promoter were partially and allele-independently methylated displaying many (>100) epialleles. RNA-polymerase-II (Pol-II) bound only to three particular epialleles in cAMP-stimulated cells, while phospho-cAMP response element-binding protein (pCREB) bound to only one epiallele, which was different from those selected by Pol-II. Binding of TATA-binding protein increased during syncytial differentiation preferentially at epialleles compatible with Pol-II and pCREB binding. Histone-3 acetylation was detected only at epialleles targeted by Pol-II and pCREB, while gene activating histone-4 acetylation and histone-3-lysine-4 trimethylation occurred at CRH epialleles not associated with Pol-II or pCREB. The suppressive histone-3-lysine-27 trimethyl and–lysine-9 trimethyl modifications showed little or no epiallele preference. The epiallele selectivity of activating histone modifications and transcription factor binding demonstrates the epigenetic and functional diversity of the CRH gene in trophoblasts, which is controlled predominantly by the patterns, not the overall extent, of promoter methylation. We propose that conditions impacting on epiallele distribution influence the number of transcriptionally active CRH gene copies in the trophoblast cell population determining the gestational trajectory of placental CRH production in normal and pathological pregnancies.
Citation: Pan X, Bowman M, Scott RJ, Fitter J, Smith R, Zakar T (2017) Promoter Methylation Pattern Controls Corticotropin Releasing Hormone Gene Activity in Human Trophoblasts. PLoS ONE 12(2): e0170671. doi:10.1371/journal.pone.0170671
Editor: Monica Uddin, University of Illinois at Urbana-Champaign, UNITED STATES
Received: August 25, 2016; Accepted: January 9, 2017; Published: February 2, 2017
Copyright: © 2017 Pan et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: X.P. received support from the China Scholarship Council (CRC) and the School of Medicine and Public Health Research Training Scheme (RTS), The University of Newcastle, Australia. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Corticotropin releasing hormone (CRH) concentration increases exponentially in the maternal plasma during the third trimester of pregnancy. Spontaneous preterm birth is associated with an accelerated rise, while post-term pregnancies are characterised by a significantly retarded increase of CRH level in the maternal circulation . The trajectory of the change is predictive of gestational length suggesting that the process controlling maternal CRH levels may be linked to the mechanism that determines the time of birth.
The source of plasma CRH in pregnancy is the syncytial layer of the placenta . It has been proposed that CRH output is accelerated with advancing gestation because of the increasing syncytial content of the placenta and the feed-forward action of glucocorticoids, which upregulates CRH gene expression in trophoblasts . Furthermore, there is strong evidence that cAMP is a powerful stimulant of CRH gene activity in cultured trophoblast cells supporting the possibility that trophoblast-derived hormonal factors acting via cAMP, such as hCG, may also drive CRH expression in an autocrine feed-forward manner [4, 5]. Studies using transfected CRH promoter-reporter constructs have indicated that CRH gene regulation in trophoblast cells depends on complex molecular interactions centered around a canonical cAMP response element (CRE) in the proximal promoter . A second, placenta-specific cAMP-response element and an NF-kappa-B response sequence have also been implicated in CRH expression control under various basal, glucocorticoid- and cAMP-stimulated conditions [7–9].
Recent studies revealed an additional level of CRH gene regulation involving epigenetic chromatin modifications in the promoter region. In the central nervous system of rats and mice, CRH gene activity is inhibited by DNA methylation and histone deacetylation in the CRH promoter region, while in human trophoblasts, dynamic histone-3, lysine-9 acetylation at the CRH promoter has been shown to underpin the glucocorticoid stimulation of CRH expression [10–12]. A regulatory role of DNA methylation at CpG dinucleotides is supported by the comprehensive analysis of the placental methylome, which has placed the CRH gene in a highly methylated domain of the trophoblast genome containing genes with tissue specific expression . Furthermore, the CRH proximal promoter is located within a CpG island shore region (relative to the intragenic CpG island), which is typically associated with tissue-specific differential methylation and gene expression . The proximal promoter of the human CRH gene contains 9 CpG dinucleotides, which can be methylated. These CpGs are conserved in the mouse, rat and the human including their positions within the promoter sequence . Notably, one of the CpGs in the CRH proximal promoter is located within the CRE. Methylation of the CpG within the CRE has been shown to have major effects on the transcription factor binding properties of the sequence, diminishing its affinity to pCREB (phosphorylated cAMP response element binding protein) and enhancing its activity to bind CEBPA (CCAAT/Enhancer binding protein, alpha) with profound functional consequences on gene expression control . Based on this information, we hypothesized that CpG methylation in the promoter regulates CRH gene activity in human trophoblasts by influencing transcription factor binding and the deposition of histone modifications that control chromatin structure coordinately with DNA methylation. We used primary human trophoblast cells in the experiments, which differentiate to syncytia in culture and respond to cAMP stimulation with robust increase of CRH expression . Combining chromatin imunoprecipitation (ChIP) with clonal bisulfite sequencing allowed us to determine CpG methylation with single base resolution in individual CRH promoters in subpopulations of CRH gene copies bound to transcription factors or marked with various histone modifications. The results uncovered the epigenetic and functional heterogeneity of the CRH promoter in the trophoblasts. The results have also suggested that only a minority of CRH gene copies is configured epigenetically to be transcriptionally active in response to 8-Br-cAMP. The size of this subpopulation, potentially influenced by the uterine environment, may define the gestational trajectory of CRH expression in normal and pathological pregnancies.
Materials and Methods
Primary human trophoblast cell isolation and culture
The study was approved by the Hunter New England Health Human Ethics Committee and the University of Newcastle Human Ethics Committee (Approval No.: 03/02/12/3.15). All participants provided written informed consent to participate in the study. Placentae were obtained from uncomplicated singleton pregnancies after spontaneous labour and delivery at the John Hunter Hospital, Newcastle NSW, Australia. Participants delivered at term (37–41 weeks of pregnancy) as confirmed by early gestation ultrasound dating. Labour was uninduced and unaugmented, and no maternal or fetal morbidities were encountered during pregnancy and labour. Cytotrophoblasts were isolated, purified and cultured as described by Kliman et al.,  with minor modifications detailed in the Supplementary Methods (S1 File). Trophoblast cultures were established from six placentae, referred to as Placentae No. 1 –No. 6 in chronological order of collection. CRH mRNA, CRH peptide and CRH promoter methylation were determined in cultures from Placentae No. 1, No. 2 and No. 3. Chromatin isolated from one primary trophoblast culture was sufficient to perform 5 combined chromatin immunoprecipitation (ChIP)–bisulfite-sequencing procedures. For this reason, cultures from Placentae No. 1, No. 2 and no. 3 were used for pCREB, TBP (TATA-binding protein, after vehicle and 8-Br-cAMP treatment) and acetylated-histone 3 (acH3) ChIP-bisulfite sequencing, and Placentae No. 4, No. 5 and No. 6 were used for acetylated histone 4 (acH4), trimethyl histone 3-lysine 4 (H3K4me3), trimethyl histone 3-lysine 27 (H3K27me3) and trimethyl histone 3-lysine 9 (H3K9me3) ChIP-bisulfite sequencing. Cultures from all 6 placentae were processed for RNA polymerase II (Pol-II) ChIP-bisulfite sequencing analysis. The placenta numbers are indicated in the respective figures.
RNA extraction, cDNA synthesis and real-time PCR
The procedures of extracting and purifying RNA from cultured trophoblasts, removing contaminating DNA, reverse transcription and measuring CRH mRNA relative abundance has been described in detail  including primers  (S3 Table), reference RNA choice (Alien) and quantitation method (delta-delta-Ct) .
Measurement of CRH peptide concentration
Clonal bisulfite sequencing of the CRH proximal promoter
The methods of DNA extraction, bisulfite conversion, amplification and isolation of the CRH proximal promoter fragment, cloning and sequencing have been described in detail  and outlined in Supplementary Methods (S1 File).
Chromatin immunoprecipitation (ChIP)
Bisulfite sequencing of ChIP-selected DNA
Purified DNA recovered after ChIP was bisulfite-converted using the Cells-to-CpG Bisulfite Conversion Kit (Applied Biosystems) with thermal cycle conditions following the manufacturer’s protocol for small input DNA (between 50 pg and 100 ng). The CRH proximal promoter sequence was amplified by nested PCR, gel purified, and processed for clonal bisulfite sequencing as described in Ref.  and Supplementary Methods (S1 File).
CRH mRNA relative abundance and peptide production.
The CRH mRNA relative abundance and peptide production values were logarithmically transformed to approach normal distribution and analysed by ANOVA (analysis of variance) with time (0, 24, 48 and 72 hours), treatments (± 8-Br-cAMP), time-treatments interaction and placenta ID as independent variables. Time, treatments and their interaction were entered in the model as repeated measures variables. Variance on placenta ID was significant. The ANOVA was followed by postestimation where contrasts were tested between time and treatment at each experimental time point by F-tests. The P-values were corrected for multiple testing according to Sidak.
CRH promoter methylation.
Methylation was considered a binary dependent variable at each CpG site being either unmethylated or methylated. Data were entered in a generalized estimating equations (GEE) model using the logit link function as recommended for repeated measures analysis for binary outcomes . Placenta number, representing independent experimental repeats, was entered as the panel variable. Postestimation Wald tests were used to determine the overall significance of the effects associated with treatment and CpG sites. The association between treatments and epiallele frequencies was tested using Pearson’s chi-square test.
ChIP data, expressed as per cent recovery of input, were transformed to achieve normal distribution and analysed by repeated measures ANOVA with the following independent variables: CRH promoter region (Sites 1–5, Fig 1A), time_treatment (0h, 72 h vehicle, 72 h 8-Br-cAMP), interaction between time_treatment and Sites. Placenta number was entered in the model representing experimental repeats. Postestimation F-tests were performed to explore time_treatment effects at each promoter site. The P-values were adjusted for multiple testing according to Sidak.
(A) Scheme of the CRH promoter. The arrows flag the two principal transcription initiation sites. CpG sites are marked by the numbered lollypop symbols, and red lines denote regulatory sequences. Promoter regions amplified by real-time PCR after chromatin immunoprecipitation (CRH1–5) are also indicated. The first transcribed nucleotide from the proximal initiation site is +1. (B) CRH mRNA relative abundance in trophoblast cultures. Cells were treated with vehicle or 8-Br-cAMP for the indicated times and mRNA relative abundance was determined by real-time RT-PCR. Bars show the average and SEM of 3 independent experiments as fold-increase over 0 h. *, significant pairwise differences between the indicated time points; #, significant stimulation by 8-Br-cAMP at each time point (P < 0.05, ANOVA with repeated measures). (C) Heat map demonstrating CpG methylation levels in the CRH proximal promoter. Methylation levels are illustrated by the color scale on the right. The vertical axis shows the numbers of the CpG sites and the horizontal axis indicates the treatments, which included freshly isolated cells (Fresh), adherent cells after 16 h in culture (0 h), vehicle treatment (Veh) or 8-Br-cAMP treatment (cAMP) for the following 72 h. Methylation levels were significantly different among CpG sites (P<0.001, but not among treatments, GEE-Generalized estimating equations).
Other statistical procedures.
Methylation frequencies between ChIP-selected and unselected promoter epialleles (shown in S2 Table) were compared by Fisher’s exact test. The incidences of epialleles bearing methylation patterns compatible with Pol-II or pCREB binding (Table 1) were compared using the Exact Unconditional Homogeneity/Independence Test  (accessed at www.stat.ncsu.edu/exact). Data obtained with different placentae were pooled for these tests. Correlations were established using Spearman’s non-parametric method. All statistical calculations (except for the Berger test) were performed using the STATA software package (College Station, TX USA). P-values < 0.05 were considered significant.
Time course of CRH mRNA induction and peptide production in trophoblast primary cultures and the effect of 8-Br-cAMP
Cytotrophoblast cells spontaneously syncytialised during 72 h in culture as reported previously [4, 25]. Syncytial differentiation was accompanied by significant 2.2–3.5- fold increase of CRH mRNA abundance between 24 h and 72 h of culture (Fig 1B). Treatment with 8-Br-cAMP (2.5 x 10−4 moles/L) stimulated CRH mRNA levels relative to vehicle at all time points, which was expected [5, 7, 22, 26–28] with a maximum of 15-fold increase at 72 h (P < 0.001). Thus, CRH mRNA abundance was 52-fold higher after 72 h of 8-Br-cAMP treatment than in untreated cells at 0 h showing robust time dependent upregulation of CRH gene activity (Fig 1B). CRH peptide release followed a similar trend. CRH peptide was detectable in the culture medium at 24 h, and its concentration increased significantly, 3.3-fold, between 24 h and 48 h and further 2.9-fold between 48 h and 72 h. 8-Br-cAMP significantly increased CRH peptide production compared to vehicle at each time point (S1 Fig). In line with these observations, CRH mRNA and peptide values measured under identical treatment conditions were highly correlated (Spearman’s rho = 0.86, P < 0.001; 21 observations).
Methylation of CpG dinucleotides in the CRH proximal promoter
The CRH proximal promoter contains 9 CpG dinucleotides as illustrated in Fig 1A. Cytosine methylation at these sites was determined by clonal bisulfite sequencing. Ten randomly selected clones were sequenced at each treatment condition with three biological replicate cultures. The heat map in Fig 1C shows % methylation at each CpG in freshly isolated cytotrophoblasts (Fresh), after 16 h of culture (0 h treatment) and after treatment with vehicle (Veh) or 8-Br-cAMP (cAMP) for additional 72 h. We chose the 72 h treatment period because 8-Br-cAMP-induced gene activation was the strongest at this time. Overall methylation frequency varied between 47.8% and 51.1% exhibiting no significant difference with treatments or incubation times (9 CpGs x 10 clones x 3 placentae = 270 CpGs at each treatment). Methylation frequencies varied significantly, however, among the individual CpG sites (10 clones x 3 placentae x 4 treatments = 120 CpGs per site) with CpG4 being the lowest (25%) and CpG7 being the highest (64.2%) (Fig 1C). These results suggested that methylation of CpGs in the CRH promoter was partial and differential. Furthermore, methylation was stable under our experimental conditions with frequencies exhibiting no significant change during culture, with syncytial differentiation, or in response to 8-Br-cAMP.
The methylation patterns of the cloned individual copies of the promoter are presented in Fig 2A–2D. Overall, 106 distinct methylation patterns (epialleles) were found in the four groups, and there was no significant association between treatments and epiallele frequencies (S1 Table). The methylation patterns did not indicate allele-specific methylation despite the overall methylation levels being close to 50%.
Patterns of CpG methylation in the CRH proximal promoter were determined by clonal bisulfite sequencing. Ten randomly selected clones representing individual copies of the promoter were sequenced under each treatment condition from cultures derived from 3 placentae. (A) Freshly isolated cells, (B) cells cultured for 16 h without treatment, (C) cells treated with vehicle or (D) with 8-Br-cAMP for the subsequent 72 h period. The horizontal lines indicate individual promoters, and open and closed circles denote unmethylated and methylated CpGs, respectively. CpG sites are numbered as in Fig 1A. The arrows (←) mark the two promoter copies with methylation patterns identical to a pattern found in the Pol-II-bound promoter fraction isolated by ChIP, shown in Fig 6B.
Histone modifications at the CRH promoter
Posttranslational histone modifications alter chromatin structure and influence transcription factor binding to chromatin. We have used chromatin immunoprecipitation (ChIP) to determine the level of key histone modifications at the CRH promoter and explore their involvement in the epigenetic control of CRH expression. For the ChIP analyses, four PCR primer pairs (CRH2-5) were used to scan the proximal promoter, while CRH1 served as the upstream control. The positions of the amplicons are shown in Fig 1A. Accordingly, the CRE was covered by CRH2 and-3, the first and the second TATA-box (TATA I and TATA II) were covered by CRH5 and CRH3, respectively, and the alternative cAMP-response element comprising CDXARE (caudal-type homeobox protein response element) , was included in CRH4.
The results of ChIP analysis using a (pan-)acetyl-histone-3 (acH3) antibody are shown in Fig 3A. Histone-3 acetylation is typically associated with transcriptionally active genes  and it has been detected in the CRH promoter area including the upstream control, CHR1. The level of acH3 increased significantly during 72h of culture in the region covered by CRH2 –CRH5. There was no additional increase in 8-Br-cAMP-treated cells. Histone-4 acetylation (acH4, measured by a (pan-)acetyl-histone-4 antibody) and histone-3, lysine-4 trimethylation (H3K4me3) are also associated in general with transcriptionally active genes [29, 30]. ChIP results in Fig 4A and 4C demonstrate that both modifications are present at the CRH promoter, increase significantly during culture and are enhanced further by 8-Br-cAMP treatment. We have also determined whether the repressive histone modifications histone-3, lysine 27-trimethylation (H3K27me3) and histone-3, lysine 9-trimethylation (H3K9me3) [29, 30] are present at the CRH proximal promoter. Both modifications were found throughout the ChIP-scanned region (including the upstream control, S2 Fig Panel A and S3 Fig Panel A, respectively). The levels of these inhibitory modifications dropped significantly during culture and even further in the presence of 8-Br-cAMP (with the exception of H3K9me3 at CRH5, showing increase (S3 Fig Panel A).
(A) acH3 at the CRH proximal promoter was determined by ChIP using a (pan-)acetylated histone-3 antibody. Significant treatment effects at each promoter region (Fig 1A) are denoted by asterisk (*, P < 0.05, ANOVA with repeated measures, N = 3 placentae). (B) Methylation patterns of promoter copies associated with acH3 in 8-Br-cAMP-treated cultures. Arrows (←) mark promoter epialleles compatible with Pol-II binding, while (x) marks epialleles that can bind pCREB shown in Fig 6B and 6D, respectively. Placenta numbers are on the left. Filled circles indicate methylated CpGs and open circles indicate unmethylated CpGs. The positions of CpG Sites are as in Fig 1A.
(A) H3K4me3 and (C) acH4 at the CRH proximal promoter were measured by ChIP. Significant treatment effects on modification levels are indicated by asterisks at each promoter region (Fig 1A; *, p < 0.05, ANOVA with repeated measures, N = 3 placentae). (B) and (D) Methylation patterns of promoter copies marked by H3K4me3 and acH4, respectively, were determined in 8-Br-cAMP-treated cultures. Placenta numbers are on the left; filled and open circles indicate methylated and unmethylated CpGs, respectively. The positions of CpG Sites are as shown in Fig 1A. Notably, none of the H3K4me3- or acH4-associated epialleles had methylation patterns compatible with Pol-II or pCREB binding.
Transcription factor binding to the CRH promoter
The CRH proximal promoter contains two TATA box sequence elements upstream of the two major transcription initiation sites, respectively (Fig 1A). In transcriptionally active promoters these elements are expected to associate with the TATA-binding protein TBP, a general transcription factor that is part of the preinitiation complex . ChIP results in Fig 5A demonstrate that TBP-binding to the promoter was low in untreated cells and increased significantly over both TATA sites in cultures treated with vehicle for 72h. Stimulation with 8-Br-cAMP during the 72h culture period did not result in further increase in TBP binding, just a partial decrease of the ChIP signal over the region covered by CRH5. (The reason for this decrease is uncertain, but may be technical such as epitope masking .)
(A) Binding of TBP to the CRH proximal promoter was determined by ChIP. Significant treatment effects at each promoter region (shown in Fig 1A) are indicated by asterisks (*, P < 0.05, ANOVA with repeated measures, N = 3 placentae; n.d, undetectable ChIP signal). (B) and (C), Methylation patterns of promoter copies bound by TBP in vehicle treated and 8-Br-cAMP-treated cultures, respectively. Arrows (←) on the right indicate patterns compatible with Pol-II binding, while (x) denotes methylation pattern compatible with binding by pCREB as shown in Fig 6B and 6D, respectively. Placenta numbers are on the left; filled circles indicate methylated CpGs and open circles indicate unmethylated CpGs. The positions of CpG Sites are as shown in Fig 1A.
Binding of the transcription factor pCREB to the CRE in the CRH promoter is considered critical for CRH gene regulation in trophoblast cells . To determine pCREB binding to the CRH promoter in our culture system we performed ChIP analysis using an antibody specific for this transcription factor. Fig 6C shows that pCREB binding to the region scanned by CRH2–4 was detected in untreated cells (0h). Binding of pCREB to this region increased significantly during culture for 72h. Treatment with 8-Br-cAMP significantly increased pCREB binding to the CRH3 region and concomitantly decreased pCREB binding at the CRH4 region enhancing the ChIP- signal at the CRE.
(A) Pol-II binding and (C) pCREB binding to the CRH promoter were determined by ChIP. PCR-amplified promoter regions are shown in Fig 1A. Bars represent means ± SEM (N = 6 and 3 placentae for Pol-II and pCREB, respectively). Asterisks indicate significant pairwise differences between treatments at each region (*, P < 0.05, ANOVA with repeated measures). Binding to the CRH1 region was undetectable (n.d.). (B) and (D), methylation patterns of CRH promoter copies bound by Pol-II and pCREB, respectively, were determined by clonal bisulfite sequencing of DNA isolated by ChIP. Ten randomly selected clones generated from 8-Br-cAMP treated cultures from each placenta were sequenced. Placenta numbers are on the left; filled and open circles indicate methylated and unmethylated CpGs, respectively. CpG Sites numbers are as shown in Fig 1A.
To assess the transcriptional activity of the CRH gene, we have performed chromatin immunoprecipitation (ChIP) using an antibody against RNA-polymerase-II (Pol-II; Fig 6A). Small, but significant binding of Pol-II was detected to the CRH4-5 region after 72 h of culture compared to 0 h in agreement with the increase of CRH mRNA level during spontaneous syncytialisation (Fig 1B). Treatment with 8-Br-cAMP resulted in a strong increase of Pol-II binding to the promoter region scanned by CRH3-5, which was in agreement with the robust stimulation of mRNA expression by the cyclic nucleotide.
The relationship of CpG methylation with transcription factor binding and histone modifications at the CRH promoter
The diverse patterns of methylation (Fig 2, S1 Table) indicated that a variety of CRH promoter epialleles was present in the trophoblast cell population. An important question to address is whether the various epialleles are equivalent functionally or show differences of ability to harbor histone modifications and drive gene expression during syncytial differentiation and cAMP-stimulation. Pol-II ChIP with 8-Br-cAMP-treated cells yielded sufficient amounts of DNA to perform bisulfite sequencing to assess methylation levels and epiallele variety. Data in S2 Table show that the methylation level of CpGs in Pol-II-associated promoters was 92.3%, significantly higher than in the non-immunoprecipitated control (47.4%, p<0.001). Considering that CpG methylation levels did not change under our experimental conditions (Fig 1B), this indicated that the CRH gene copies bound by Pol-II had relatively high levels of promoter methylation. Furthermore, examination of the arrangement of methylated CpGs in individual cloned copies (Fig 6B) revealed that the methylation pattern of Pol-II-associated promoters was uniform in each placenta (N = 6) and followed 3 distinct types: all 9 CpGs methylated (3 placentae), all but CpG4 methylated (2 placentae) and all but CpGs4 and 6 methylated. Notably, the CRE (containing CpG2) was methylated in all copies associated with Pol-II suggesting that its pCREB-binding activity essential for mediating 8-Br-cAMP-stimulation  was suppressed.
Bisulfite sequencing of promoter copies immunoprecipitated with pCREB antibody indicated that he pattern of CpG methylation was uniform within and among placentae (N = 3 placentae with 10 promoter copies each) with CpGs 1, 2, 3, and 4 unmethylated and CpGs 5, 6, 7, 8, and 9 methylated (Fig 6D). Thus, CpG2 in the CRE was unmethylated as expected in agreement with pCREB binding .
CRH promoter DNA immunoprecipitated by the TBP antibody from vehicle treated and 8-Br-cAMP treated cultures was bisulfite sequenced as well. The TBP-bound CRH promoters had significantly higher methylation levels than promoter sequences in control (not ChIP-selected) DNA in both vehicle treated and 8-Br-cAMP treated cells (S2 Table). In addition, promoter copies associated with TBP were significantly enriched in methylation patterns compatible with Pol-II and pCREB binding (Fig 5B and 5C and Table 1).
We have performed ChIP-bisulfite analyses to explore whether histone modifications associate differentially with CRH promoter epialleles. Bisulfite sequencing of the DNA precipitated with the acH3 antibody indicated significantly higher methylation level of the CRH promoters than in the non-immunoprecipitated control (S2 Table). The methylation patterns of all sequenced promoter copies were compatible with either Pol-II or pCREB binding (Fig 3B and Table 1). Thus, histone-3 acetylation not only increased during syncytial differentiation, but it appeared selective at CRH gene copies targeted by Pol-II and pCREB in 8-Br-cAMP-treated cells.
The methylation level of CRH promoter copies associated with acH4 and H3K4me3 modifications, determined by ChIP-bisulfite sequencing using a (pan-)acH4 and a H3K4me3 antibody with 8-Br-cAMP-treated cells, was significantly lower than the non-immunoprecipitated control (S2 Table). Furthermore, the methylation patterns of the cloned promoter copies indicated that these histone modifications were not associated with promoter epialleles that bound either Pol-II, pCREB, or were acetylated at H3 (Table 1 and Fig 4B and 4D vs. Figs 6B, 6D and 3B).
The overall methylation level of the H3K27me3- associated promoters was not different from the non-immunoprecipitated control (S2 Table), but this group of promoter copies was enriched significantly in Pol-II and pCREB-binding epialleles (S2 Fig Panel A and Table 1). The H3K9me3-associated promoter copies were less methylated than the non-immunoprecipitated control and were not enriched in epialleles binding Pol-II or pCREB (S3 Fig Panel B, S2 Table and Table 1).
The results of this study show that CRH expression is controlled in trophoblasts by epigenetic mechanisms that include histone modifications and DNA methylation. Syncytial differentiation in primary cultures was accompanied by increasing levels of activating (acH3, acH4, H3K4me3) and decreasing levels of suppressive (H3K27me3, H3K9me3) histone modifications at the CRH proximal promoter. Treatment with 8-Br-cAMP robustly stimulated CRH gene expression and augmented the changes in histone modifications with the exception of acH3, which appeared unresponsive to cAMP (Fig 3A). This feature suggests that H3-acetylation may be a cAMP-independent process poising the CRH promoter for increased transcription factor access during differentiation. In agreement with this, enhanced recruitment of TBP and pCREB was observed to the promoter during syncytialisation in the absence of 8-Br-cAMP (Figs 5A and 6C). Treatment with the cyclic nucleotide caused a marked increase of Pol-II recruitment to the poised promoter in agreement with the large increase of CRH gene activity (Fig 6A).
The analysis of CpG methylation radically reshaped the picture of epigenetic CRH gene regulation based solely on ChIP data. Bisulfite sequencing has demonstrated that the CRH proximal promoter, comprising 9 CpG dinucleotides, is partially methylated in trophoblast cells. Methylation level overall was close to 50% exhibiting no significant change during culture, or after exposure to 8-Br-cAMP. Methylation levels of the particular CpG sites were significantly different, however, indicating context specific susceptibility to methylation, which remained unchanged under the culture conditions employed. The patterns of methylation determined by clonal bisulfite sequencing did not indicate allele dependency (such as a mixture of either fully methylated or unmethylated promoter copies); rather, they revealed that a variety of epialleles is present in the trophoblast cell population. These attributes suggest that the CRH proximal promoter represents an “intermediate methylation region” (IM) in the genome as determined by recent comparative epigenomic studies involving a number of human tissues and cell types . The IM regions are generally associated with tissue specific regulatory functions  in line with the observation that promoter methylation controls hypothalamic CRH expression in rodents [10, 36].
The varying methylation patterns of individual promoter copies raised the question whether the CRH promoter epialleles were functionally different. We addressed this possibility using ChIP combined with bisulfite sequencing of the immunoprecipitated DNA , which is an novel approach to explore cross-talk between transcription factor binding, histone modifications and DNA methylation at both the genomic and the single gene levels [38, 39]. Surprisingly, Pol-II-associated CRH promoter copies were significantly more methylated than the promoter population overall, and only 3 specific methylation patterns were detected in the Pol-II-bound fraction: (1) all 9 CpGs methylated, (2) only CpG4 unmethylated and (3) CpGs 4 and 6 unmethylated. During the study we have encountered 117 various CRH promoter methylation patterns in the primary trophoblasts (106 in Fig 2, listed in S1 Table, and additional 11 in the other experimental settings). Finding only three of them in the Pol-II ChIP-isolated fractions indicates high epiallele selectivity of Pol-II binding. A further striking observation was that CpG2 in the CRE was uniformly methylated in the Pol-II-bound fraction. This argues against the involvement of the pCREB-CRE interaction mediating the stimulation of CRH gene activity by 8-Br-cAMP, since this interaction is disrupted by CpG methylation within the CRE . Nevertheless, pCREB binding increased to the CRE region during syncytialisation and cAMP-stimulation, and the pCREB-bound promoters were unmethylated at CpG2, as expected. The methylation pattern of the pCREB-bound promoter copies was different from the patterns bound by Pol-II, however, demonstrating that Pol-II and pCREB bound to separate sub-populations of the CRH promoter characterized by distinct epialleles.
ChIP-bisulfite analysis of TBP binding showed significant enrichment of promoter epialleles compatible with Pol-II and pCREB binding, which suggested that chromatin remodeling that allowed promoter access for TBP in syncytia occurred preferentially in a subpopulation of CRH epialleles that were capable of binding Pol-II and pCREB.
We have extended the ChIP-bisulfite analysis to key histone modifications that contribute to the epigenetic landscape, influence chromatin structure and correlate with gene activity . Remarkably, the promoter fraction associated with acetylated histone-3 consisted entirely of copies possessing methylation patterns compatible with either Pol-II or pCREB binding, which highlights the importance of this histone modification for transcription factor access to the region. Dynamic acetylation of H3K9 has recently been found critical for the gestational and glucocorticoid dependent upregulation of CRH expression in trophoblasts and for transcription factor binding to CRH gene regulatory elements . We used a pan-K-acetylated H3 antibody in our ChIP assays, but the results by Di Stefano et al.  suggest that the immunoprecipitated chromatin fraction we have analysed for DNA methylation contained acetyl-H3K9. It remains to be established whether a dynamic turnover of this modification occurs when CRH expression is stimulated by 8-Br-cAMP.
Promoter copies associated with the gene activating acH4 and H3K4me3 modifications did not carry CpG methylation patterns compatible with Pol-II and pCREB binding. This suggests that these modifications affected preferentially a subpopulation of CRH promoters not involved in the 8-Br-cAMP-evoked response. Enrichment of Pol-II- and pCREB-binding methylation patterns was significant in the H3K27me3-marked epialleles relative to the non-immunoprecipitated input suggesting that loss of this mark might have contributed to the epigenetic activation of CRH in the trophoblast cultures. Promoter copies marked by H3K9me3 did not show significant enrichment of Pol-II- and pCREB-binding epialleles arguing against the possibility that loss of this modification contributed appreciably to the changes in CRH gene expression under the conditions employed.
An important outcome of our ChIP-bisulfite analyses is that it has revealed the epigenetic, functional and structural heterogeneity of the CRH promoter population in term trophoblasts. Based on the epiallele selective associations between the CRH promoter, Pol-II and pCREB, we can infer with reasonable confidence that the CRE-pCREB interaction is not involved in the cAMP-dependent regulation of the gene, at least in our regulatory model using 8-Br-cAMP. Interestingly, this interaction was shown to be crucial for regulating CRH promoter activity in assays using transfected promoter-reporter constructs . Our results suggest that 8-Br-cAMP induces the chromatin-integrated endogenous gene via a different pathway possibly involving the alternative cAMP-response element, which does not bind pCREB  (Fig 1A). The activating epigenetic marks H3K4me3 and acH4 are not required for the stimulation, but acH3 appears critical. The molecular apparatus that forms this unique signaling pathway remains to be determined.
A published quantitative analysis of Pol-II localisation in placental tissue has shown that 20–40% of syncytiotrophoblast nuclei are transcriptionally inactive . Our ChIP-bisulfite analyses indicated that less than 5% of CRH promoter copies were engaged in transcription in 8-Br-cAMP-stimulated primary trophoblasts as only 2 of 120 sequenced CRH promoter copies showed methylation patterns compatible with Pol-II binding in DNA not subjected to ChIP (Fig 2). Epipolymorphism is widespread in the placenta , but the epiallele selectivity of CRH gene activation is intriguing even if more transcriptionally active CRH epialleles could be found in addition to the 3 we have identified by examining a larger number of placentae. It is unclear whether specific methylation patterns are involved functionally in the recruitment of particular CRH promoter copies for transcription, or just mark the selected group of epialleles. A related important question is whether other stimulants of CRH expression, such as glucocorticoids, target the same set of epialleles as 8-Br-cAMP. This is likely, since these epialleles are preferentially acetylated on histone-3, which is a hallmark of glucocorticoid action on CRH as reported . In our hands glucocorticoid was a weak stimulant (eliciting less than 2-fold increase in CRH expression), not providing enough DNA for bisulfite sequencing after Pol-II ChIP. The function of the CRH promoter sub-population activated by H3K4me3 and acH4 modifications, but not involved in the 8-Br-cAMP-response, also remains to be determined. Finally, it will be critical to explore how is the epiallele distribution of CRH promoters established in the trophoblast cell population. Methylated CpG levels were unchanging with no indication of significant turnover in our culture system, which contained non-dividing term trophoblasts (Fig 1C). The trophoblast cells were isolated from term placentae after spontaneous labor, and the methylation levels and patterns could have been influenced by intrauterine conditions during pregnancy and in labor resulting in changes in the number of transcriptionally competent CRH epialleles. Data describing the epiallele distribution of the CRH promoter population in placentae at different times of gestation and after labor may uncover factors that impact on methylation levels and patterns and determine the gestational trajectory of CRH production in normal and pathological pregnancies.
S1 Fig. CRH peptide production by primary trophoblast cell cultures.
Cell cultures were treated with vehicle or 8-Br-cAMP (250 μM) and CRH peptide concentration in the culture medium was determined by radioimmunoassay. Bars show the average and SEM of 3 independent experiments. *, significant pairwise differences between time points as indicated by the arrows; #, significant stimulation by 8-Br-cAMP at each time point (p<0.05).
S2 Fig. Methylation patterns of CRH promoter copies marked by histone-3 lysine-27 trimethylation (H3K27me3).
(A) H3K27me3 marking at the CRH proximal promoter was determined by chromatin immunoprecipitation (ChIP). Significant treatment effects at each promoter region (Fig 1A) are denoted by asterisk (*, p<0.05, ANOVA with repeated measures, N = 3 placentae). (B) Methylation patterns of CRH promoter copies marked by H2K27me3 were determined by bisulfite sequencing of ChIP-isolated DNA from 8-Br-cAMP-treated cultures. Arrows (←) mark promoter epialleles compatible with Pol-II binding, while (x) indicates epialleles that can bind pCREB. Placenta numbers are on the left; filled and open circles indicate methylated and unmethylated CpGs, respectively. The positions of CpG Sites are as shown in Fig 1A.
S3 Fig. Methylation patterns of CRH promoter copies marked by histone-3 lysine-9 trimethylation (H3K9me3).
(A) H3K9me3 marking of the promoter was determined by chromatin immunoprecipitation (ChIP). Significant treatment effects at each promoter region (Fig 1A) are denoted by asterisk (*, p<0.05, ANOVA with repeated measures, N = 3 placentae). (B) Methylation patterns of CRH promoter copies in chromatin marked by H3K9me3 were determined by bisulfite sequencing of DNA isolated by ChIP from 8-Br-cAMP-treated cultures. The arrow (←) marks a promoter epiallele capable of Pol-II binding. Placenta numbers are on the left. Filled and open circles indicate methylated and unmethylated CpGs, respectively. The positions of CpG Sites are as shown in Fig 1A.
S1 Table. Frequency of methylation patterns of the CRH promoter.
S2 Table. Methylation levels in CRH promoter copies associated with transcription factors and modified histones.
S3 Table. Primer Sequences.
S1 File. Supplementary Methods.
This work was supported by the China Scholarship Council (CRC) and the School of Medicine and Public Health Research Training Scheme (RTS), The University of Newcastle, Australia. Help by Ms. Robyn Miller with data analysis is gratefully acknowledged.
- Conceptualization: TZ RS.
- Formal analysis: TZ.
- Funding acquisition: XP RS TZ.
- Investigation: XP MB.
- Methodology: XP TZ JF.
- Project administration: TZ RS JF.
- Resources: RS RJS.
- Supervision: TZ RS RJS JF.
- Validation: TZ XP.
- Visualization: XP TZ.
- Writing – original draft: TZ.
- Writing – review & editing: RS RJS JF.
- 1. McLean M, Bisits A, Davies J, Woods R, Lowry P, Smith R. A placental clock controlling the length of human pregnancy. Nature medicine. 1995;1(5):460–3. pmid:7585095
- 2. Riley SC, Walton JC, Herlick JM, Challis JR. The localization and distribution of corticotropin-releasing hormone in the human placenta and fetal membranes throughout gestation. J Clin Endocrinol Metab. 1991;72(5):1001–7. doi: 10.1210/jcem-72-5-1001. pmid:2022703
- 3. Smith R, Paul J, Maiti K, Tolosa J, Madsen G. Recent advances in understanding the endocrinology of human birth. Trends in Endocrinology & Metabolism. 2012;23(10):516–23.
- 4. Kliman HJ, Nestler JE, Sermasi E, Sanger JM, Strauss JF 3rd. Purification, characterization, and in vitro differentiation of cytotrophoblasts from human term placentae. Endocrinology. 1986;118(4):1567–82. doi: 10.1210/endo-118-4-1567. pmid:3512258
- 5. Cheng YH, Nicholson RC, King B, Chan EC, Fitter JT, Smith R. Corticotropin-releasing hormone gene expression in primary placental cells is modulated by cyclic adenosine 3 ',5 '-monophosphate. J Clin Endocr Metab. 2000;85(3):1239–44. doi: 10.1210/jcem.85.3.6420. pmid:10720069
- 6. King BR, Smith R, Nicholson RC. The regulation of human corticotrophin-releasing hormone gene expression in the placenta. Peptides. 2001;22(5):795–801. pmid:11337093
- 7. Scatena CD, Adler S. Characterization of a human-specific regulator of placental corticotropin-releasing hormone. Molecular endocrinology. 1998;12(8):1228–40. doi: 10.1210/mend.12.8.0150. pmid:9717848
- 8. Wang B, Palomares K, Parobchak N, Cece J, Rosen M, Nguyen A, et al. Glucocorticoid Receptor Signaling Contributes to Constitutive Activation of the Noncanonical NF-kappaB Pathway in Term Human Placenta. Molecular endocrinology. 2012.
- 9. Wang B, Parobchak N, Rosen T. RelB/NF-kappaB2 regulates corticotropin-releasing hormone in the human placenta. Molecular endocrinology. 2012;26(8):1356–69. doi: 10.1210/me.2012-1035. pmid:22734038
- 10. Elliott E, Ezra-Nevo G, Regev L, Neufeld-Cohen A, Chen A. Resilience to social stress coincides with functional DNA methylation of the Crf gene in adult mice. Nature neuroscience. 2010;13(11):1351–3. doi: 10.1038/nn.2642. pmid:20890295
- 11. Wang A, Nie W, Li H, Hou Y, Yu Z, Fan Q, et al. Epigenetic upregulation of corticotrophin-releasing hormone mediates postnatal maternal separation-induced memory deficiency. PloS one. 2014;9(4):e94394. doi: 10.1371/journal.pone.0094394. pmid:24718660
- 12. Di Stefano V, Wang B, Parobchak N, Roche N, Rosen T. RelB/p52-mediated NF-kappaB signaling alters histone acetylation to increase the abundance of corticotropin-releasing hormone in human placenta. Sci Signal. 2015;8(391):ra85. doi: 10.1126/scisignal.aaa9806. pmid:26307012
- 13. Schroeder DI, Blair JD, Lott P, Yu HO, Hong D, Crary F, et al. The human placenta methylome. Proc Natl Acad Sci U S A. 2013;110(15):6037–42. doi: 10.1073/pnas.1215145110. pmid:23530188
- 14. Irizarry RA, Ladd-Acosta C, Wen B, Wu Z, Montano C, Onyango P, et al. The human colon cancer methylome shows similar hypo- and hypermethylation at conserved tissue-specific CpG island shores. Nat Genet. 2009;41(2):178–86. doi: 10.1038/ng.298. pmid:19151715
- 15. Rishi V, Bhattacharya P, Chatterjee R, Rozenberg J, Zhao J, Glass K, et al. CpG methylation of half-CRE sequences creates C/EBPalpha binding sites that activate some tissue-specific genes. Proc Natl Acad Sci U S A. 2010;107(47):20311–6. doi: 10.1073/pnas.1008688107. pmid:21059933
- 16. Pan X, Bowman M, Scott RJ, Fitter J, Nicholson RC, Smith R, et al. Methylation of the Corticotropin Releasing Hormone Gene Promoter in BeWo Cells: Relationship to Gene Activity. International Journal of Endocrinology. 2015;2015:1–8.
- 17. Sehringer B, Zahradnik HP, Deppert WR, Simon M, Noethling C, Schaefer WR. Evaluation of different strategies for real-time RT-PCR expression analysis of corticotropin-releasing hormone and related proteins in human gestational tissues. Analytical and bioanalytical chemistry. 2005;383(5):768–75. doi: 10.1007/s00216-005-0067-9. pmid:16189679
- 18. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25(4):402–8. doi: 10.1006/meth.2001.1262. pmid:11846609
- 19. Smith R, Wickings EJ, Bowman ME, Belleoud A, Dubreuil G, Davies JJ, et al. Corticotropin-releasing hormone in chimpanzee and gorilla pregnancies. J Clin Endocrinol Metab. 1999;84(8):2820–5. doi: 10.1210/jcem.84.8.5906. pmid:10443686
- 20. Mitchell C, Johnson R, Giles WB, Zakar T. Prostaglandin H synthase-2 gene regulation in the amnion at labour: histone acetylation and nuclear factor kappa B binding to the promoter in vivo. Mol Hum Reprod. 2008;14(1):53–9. doi: 10.1093/molehr/gam086. pmid:18204072
- 21. Mitchell C, Johnson R, Bisits A, Hirst J, Zakar T. PTGS2 (Prostaglandin Endoperoxide Synthase-2) Expression in Term Human Amnion in Vivo Involves Rapid mRNA Turnover, Polymerase-II 5'-Pausing, and Glucocorticoid Transrepression. Endocrinology. 2011;152(5):2113–22. doi: 10.1210/en.2010-1327. pmid:21385935
- 22. Adler GK, Smas CM, Fiandaca M, Frim DM, Majzoub JA. Regulated expression of the human corticotropin releasing hormone gene by cyclic AMP. Mol Cell Endocrinol. 1990;70(2):165–74. pmid:2161364
- 23. Williamson DS, Bangdiwala SI, Marshall SW, Waller AE. Repeated measures analysis of binary outcomes: applications to injury research. Accid Anal Prev. 1996;28(5):571–9. pmid:8899038
- 24. Berger RL. More Powerful Tests from Confidence Interval p Values. The American Statistician. 1996;50(4):314–8.
- 25. Coutifaris C, Babalola GO, Abisogun AO, Kao LC, Chin U, Vadillo-Ortega F, et al. In vitro systems for the study of human trophoblast implantation. Ann N Y Acad Sci. 1991;622:191–201. pmid:2064180
- 26. Spengler D, Rupprecht R, Van LP, Holsboer F. Identification and characterization of a 3',5'-cyclic adenosine monophosphate-responsive element in the human corticotropin-releasing hormone gene promoter. Molecular endocrinology. 1992;6(11):1931–41. doi: 10.1210/mend.6.11.1480179. pmid:1480179
- 27. Majzoub JA, Muglia LJ, Martinez C, Jacobson L. Molecular and transgenic studies of the corticotropin-releasing hormone gene. Ann N Y Acad Sci. 1995;771:293–300. pmid:8597407
- 28. King BR, Smith R, Nicholson RC. Novel glucocorticoid and cAMP interactions on the CRH gene promoter. Mol Cell Endocrinol. 2002;194(1–2):19–28. pmid:12242024
- 29. Smolle M, Workman JL. Transcription-associated histone modifications and cryptic transcription. Biochimica et biophysica acta. 2013;1829(1):84–97. doi: 10.1016/j.bbagrm.2012.08.008. pmid:22982198
- 30. Wang Z, Schones DE, Zhao K. Characterization of human epigenomes. Current opinion in genetics & development. 2009;19(2):127–34.
- 31. Lee TI, Young RA. Transcription of eukaryotic protein-coding genes. Annu Rev Genet. 2000;34:77–137. doi: 10.1146/annurev.genet.34.1.77. pmid:11092823
- 32. Aparicio O, Geisberg JV, Sekinger E, Yang A, Moqtaderi Z, Struhl K. Chromatin Immunoprecipitation for Determining the Association of Proteins with Specific Genomic Sequences In Vivo. In: Harkins EW, editor. Current Protocols in Molecular Biology. Unit 21.3 (Supplement 69). Hoboken, NJ: John Wiley and Sons, Inc.; 2005. p. 21.3.1–.3.33.
- 33. Nicholson RC, King BR. Regulation of CRH gene expression in the placenta. Frontiers of hormone research. 2001;27:246–57. pmid:11450430
- 34. Iguchi-Ariga SM, Schaffner W. CpG methylation of the cAMP-responsive enhancer/promoter sequence TGACGTCA abolishes specific factor binding as well as transcriptional activation. Genes & development. 1989;3(5):612–9.
- 35. Elliott G, Hong C, Xing X, Zhou X, Li D, Coarfa C, et al. Intermediate DNA methylation is a conserved signature of genome regulation. Nat Commun. 2015;6:6363. doi: 10.1038/ncomms7363. pmid:25691127
- 36. McGill BE, Bundle SF, Yaylaoglu MB, Carson JP, Thaller C, Zoghbi HY. Enhanced anxiety and stress-induced corticosterone release are associated with increased Crh expression in a mouse model of Rett syndrome. Proc Natl Acad Sci U S A. 2006;103(48):18267–72. doi: 10.1073/pnas.0608702103. pmid:17108082
- 37. Li Y, Tollefsbol TO. Combined chromatin immunoprecipitation and bisulfite methylation sequencing analysis. Methods Mol Biol. 2011;791:239–51. doi: 10.1007/978-1-61779-316-5_18. pmid:21913084
- 38. Statham AL, Robinson MD, Song JZ, Coolen MW, Stirzaker C, Clark SJ. Bisulfite sequencing of chromatin immunoprecipitated DNA (BisChIP-seq) directly informs methylation status of histone-modified DNA. Genome research. 2012;22(6):1120–7. doi: 10.1101/gr.132076.111. pmid:22466171
- 39. Zinn RL, Pruitt K, Eguchi S, Baylin SB, Herman JG. hTERT is expressed in cancer cell lines despite promoter DNA methylation by preservation of unmethylated DNA and active chromatin around the transcription start site. Cancer Res. 2007;67(1):194–201. doi: 10.1158/0008-5472.CAN-06-3396. pmid:17210699
- 40. Roadmap Epigenomics C, Kundaje A, Meuleman W, Ernst J, Bilenky M, Yen A, et al. Integrative analysis of 111 reference human epigenomes. Nature. 2015;518(7539):317–30. doi: 10.1038/nature14248. pmid:25693563
- 41. Nicholson RC, King BR, Smith R. Complex regulatory interactions control CRH gene expression. Frontiers in bioscience: a journal and virtual library. 2004;9:32–9.
- 42. Fogarty NM, Mayhew TM, Ferguson-Smith AC, Burton GJ. A quantitative analysis of transcriptionally active syncytiotrophoblast nuclei across human gestation. J Anat. 2011;219(5):601–10. doi: 10.1111/j.1469-7580.2011.01417.x. pmid:21883201
- 43. Landan G, Cohen NM, Mukamel Z, Bar A, Molchadsky A, Brosh R, et al. Epigenetic polymorphism and the stochastic formation of differentially methylated regions in normal and cancerous tissues. Nat Genet. 2012;44(11):1207–14. doi: 10.1038/ng.2442. pmid:23064413