Date Published: February 2, 2017
Publisher: Public Library of Science
Author(s): Xin Pan, Maria Bowman, Rodney J. Scott, John Fitter, Roger Smith, Tamas Zakar, Monica Uddin.
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.
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 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).