Histone modifications in embryo implantation and placentation: insights from mouse models.

The journey of new life begins with the formation of an embryo, which must be implanted in the uterus to establish a functional interaction between mother and fetus in mammals ( 1 ). Embryo implantation and placentation are key steps in the establishment of this communication and are required for a successful pregnancy. As the blastocyst acquires implantation competency, the endometrium differentiates to become receptive to the embryos under the regulation of estrogen (E2) and progesterone (P4) ( 2 ). Upon completion of implantation, the outer trophectoderm (TE) of the blastocyst begins to differentiate and forms the placenta, a transient organ that acts as a barrier between the mother and fetus ( 3 ). The placenta facilitates the exchange of nutrients and oxygen while protecting the fetus from harmful substances ( 4 ). Disturbance to implantation and placentation may lead to pregnancy-related complications, such as recurrent pregnancy loss, infertility, pre-eclampsia, fetal growth restriction, preterm birth, and stillbirth ( 5 ). Thus, a better understanding of the underlying molecular networks of embryo implantation and placentation will help to advance our understanding of the causes of pregnancy complications. Epigenetic control is the process by which gene expression is regulated by chemical modifications to DNA and its associated proteins, without altering the underlying genetic code ( 6 ). These modifications act like switches, turning genes on or off, and allowing cells to respond to changing environmental signals, and have been implicated in a wide range of biological processes, including development, aging, and disease ( 7 ). It includes a range of chemical changes, such as DNA methylation, non-coding RNA (ncRNA) expression, and histone modification we aimed at in this review. Histones can be modified in a variety of ways, including methylation, acetylation, phosphorylation, ubiquitylation and poly (ADP)-ribosylation ( 8 ). These histone modifications affect chromatin compaction and accessibility of transcription factors or cofactors, thereby regulating gene transcriptional activation or silencing. The enzymes involved in catalyzing this histone modification can be classified as "writer" and "eraser". "Writer" refers to the enzymes that add histone modifications to histones, such as histone methyltransferases, histone acetyltransferases, and so on. “Eraser” removes specific modifications, including histone demethylases and histone deacetylases ( 9 ). The study of histone modification has been a rapidly growing field in recent years, and have discovered that changes in histone modification patterns are associated with a range of diseases, including cancer, cardiovascular disease, and neurodevelopmental disorders ( 10 , 11 ). Recent studies have revealed that histone modifications play critical roles in determining the success of embryo implantation and placenta development, providing valuable insights into the underlying molecular mechanisms. In this review, we present the current findings of implantation and placentation events in various model systems and in humans, primarily focusing on the impact of histone modifications on the embryo-uterus dialogues during gestation ( Figure 1 ). A better understanding of the roles of histone modification in these processes may reveal new predictive and therapeutic targets for pregnancy-related complications. Histone modifications required for mouse embryo implantation and placentation. The upper panel illustrates the crucial writers and erasers involved in histone methylation (EZH2, Menin, and PR-Set7), histone acetylation (Hdac3, Sirt1 and P300), and histone ubiquitination (PRC1) during mouse embryo implantation. These modifications contribute to stroma decidualization by regulating critical gene expression (e.g., Wnt4 , Ccnd2 , Cola1/2 , Ptx3 , and IGFBP1 ). In the lower panel, histone modifications, including methylation and acetylation, are shown to be essential for placentation. Changes in histone methylation by writers (EED/EZH2, Suv39h1/2) or erasers (Kdm3a/4c/5b, LSD1) control gene expression required for trophoblast stemness, self-renew, invasion and differentiation. Acetylation regulators, including histone acetylases (CBP/P300) and deacetylases (Hdac1/2/6), are involved in TSC epithelial-to-mesenchymal transition (EMT) and differentiation.

As the blastomeres proceed with development, the initial determination of cell fate commences, culminating in the gradual formation of the inside cells and outside cells, ultimately giving rise to two distinct lineages, the ICM and TE ( 68 ). During this period, cells from both lineages undergo loss of totipotency and are constrained to their respective lineage differentiation trajectories ( 69 ). In this fate-determining process, covalent histone modifications play critical roles. In vitro and in vivo experiments have demonstrated that deletion of SETDB1, a H3K9 methyltransferase, in mouse embryonic stem cells (mESCs) induces their differentiation towards the TE lineage ( 70 , 71 ). SUV39H1, another H3K9 methyltransferase, also suppresses ICM-associated gene expression during TE lineage development ( 72 , 73 ). These studies indicated that the commitment of ICM versus TE is precisely regulated by the incorporation of H3K9me3 at specific chromatin domains by distinct histone methyltransferases. In addition to H3K9me3, H3K27me3, as a transcriptional repressive-associated histone modification, also plays a critical role in TE and ICM lineage differentiation ( 72 , 73 ). There is a significant difference in the global H3K27me3 level between ICM and TE. In the ICM lineage, promoter regions of TE-associated genes are enriched with H3K27me3 modification, while H3K27me3 enrichment on Cdx2 and Gata3 genes gradually decreases during TE lineage differentiation and development ( 72 , 73 ). These findings suggest the importance of H3K27me3 in cell fate determination. Interestingly, Hdac1 binds to self-renewal-related genes Oct4 , Sox2 , and Nanog in the ICM lineage, and to lineage-regulating genes Cdx2 , Elf5 , and Eomes in the TE lineage, accompanied by transcriptional activation-associated modification H3 acetylation ( 74 ). These findings suggest that although the two lineages have differences, they also share similarities, such as Hdac1 binding to target genes in different lineages to maintain their respective self-renewal abilities ( 74 ). The development of the placenta starts with contact between the blastocyst TE and the endometrial epithelium in both humans and mice. The blastocyst penetrates the endometrium via its trophoblastic cells and embeds deeply, accompanied by an abundance of proliferation and differentiation of these cells, which constitutes the primary placenta structure ( 3 ). Although both mouse and human placentas share a hemochorial nature, they exhibit distinct variations in their overall morphology ( 75 ). In humans, upon contact with the decidual, a portion of trophoblastic cells undergo fusion to form primary syncytium, which penetrates the decidual layer and embeds deeply ( 4 , 76 ). The trophoblasts with stemness undergo differentiation to form functional trophoblast subtypes, including cytotrophoblasts (CTB), syncytiotrophoblasts (STB), and extravillous trophoblasts (EVT) to fulfill the placental functions ( 77 ). These cells gradually form a villous-like structure, composed of fetal vessels and mesenchymal stroma. The CTB and STB which are located at the outer layer of the placental villi, absorb nutrients from the maternal blood and form the maternal-fetal barrier that provides protection for the fetus ( 77 , 78 ). Meanwhile, the EVT migrates and invades the decidual layer and even the myometrium to anchor the placenta firmly in the uterus and remodel the spiral arteries of the uterus to establish maternal-fetal blood circulation ( 76 – 78 ). In mice, TE continues to proliferate to form extraembryonic ectoderm (EXE) and the ectoplacental cone (EPC). The extraembryonic mesoderm appears at E6.5 and gives rise to the allantois and the extraembryonic mesodermal layers of the amnion and chorion ( 3 , 79 ). At E8.5, the fusion of the allantois with the chorion constitutes a key event in placenta maturation. The fusion facilitates the invagination of blood vessels derived from the extraembryonic mesoderm into the trophoblastic layer of the chorion, forming the key structure of the placenta-the labyrinth layer ( 79 ). As the labyrinth layer of the placenta expands, the trophoblasts differentiate into syncytiotrophoblasts (STB) and sinusoidal trophoblast giant cells (TGC) to fulfill their respective roles in placental function ( 3 ). Although mice and humans have different placental structures and key developmental genes, conditional knockout mice remain the main model for investigating placental development due to their shared classification as hemochorial placentas and the convenience of genetic manipulation ( 4 ). The mouse models that express Cre specifically in different trophoblast cell types are used to investigate the function and potential mechanisms of different genes ( Table 2 ). Mouse models expressing cre recombinase in the trophoblast. During embryonic development, cells undergo the first lineage specification event, leading to the formation of either the embryonic or extra-embryonic lineage, the latter of which will develop into the placenta. In this process, there is a widespread loss of H3K27me3 in promoter regions, followed by rapid dynamics during the morula to blastocyst transition. ( 87 , 88 ). Yang et al. reported that H3K27me3 and DNA methylation regulates key developmental genes in embryonic and extra-embryonic cells. This helps to maintain the highly regulated developmental plasticity in the embryonic cells, as well as restricting the developmental potential of the extra-embryonic cells ( 89 ). During pluripotent cell development, many genes bound by PRC2 have both the repressive H3K27me3 mark and the activating H3K4me3 mark. The H3K4me3/H3K27me3 bivalent state is absent in developmental genes until lineage differentiation initiates, and most bivalent genes are repressed ( 90 ). This regulatory mechanism allows genes to be silenced as key developmental regulators while being primed for future activation (or repression), and thus generally transition to monovalent configurations upon differentiation. In the embryonic lineage, PRC1 and poised RNAP are not recruited to the PRC2-bound genes. However, in the extra-embryonic lineage, bivalent genes are selectively targeted by Suv39h1-mediated H3K9me3 repression, which determines the initial fate ( 72 ). Maternal H3K27me3-mediated imprinting can mediate gene silencing in DNA hypomethylated regions during embryonic development ( 88 ). In trophoblast, the paternal imprinting of the Kcnq1 and IC2 domain on the distal end of mouse chromosome 7 is mainly achieved through H3K27me3 and H3K9me2 mediated by Polycomb complex, rather than relying on DNA methylation ( 91 – 93 ). Moreover, oocyte-derived H3K27me3 also plays a vital role in maintaining non-canonical gene imprinting in extraembryonic lineage cells ( 94 ). The significance of Polycomb complex in gene imprinting has been demonstrated through the whole-body knockout mouse model and triploid compensation experiments, which revealed that Eed knockout resulted in abnormal expression of the paternal imprinting gene Mash2 ( 95 ). Animal cloning can be achieved through somatic cell nuclear transfer (SCNT), but the success rate is relatively low and often results in abnormal placental development. H3K27me3-mediated gene imprinting may be a key factor ( 96 ). Additionally, aberrant upregulation of H3K27me3-mediated clustered miRNAs from Sfmbt2 has been identified as the major cause of abnormal placental hyperplasia in SCNT mice ( 97 ). During the process of placenta maturation until the end of pregnancy, there are substantial changes in the epigenome of cytotrophoblasts. Lv et al. have reported that in trophoblast, EZH2 suppresses CDX1 expression to regulate its invasion through an H3K27me3-dependent manner ( 98 ). To further investigate the impact of EZH2 on placental development, Nugent et al. used Cyp19-cre to conditionally knockout Ezh2 in the placenta. After EZH2 deletion, female fetuses became more vulnerable to prenatal stress. But the placental morphology and function changes after EZH2 knockout remain obscure ( 99 ). H3K9 methylation is a histone modification associated with transcriptional repression and is reported to regulate the function of trophoblast ( 100 , 101 ). Kdm3a is a demethylase that specifically targets H3K9 methylation. Chakraborty et al. have demonstrated that under hypoxic conditions, HIF downregulates H3K9 methylation at the Mmp12 promoter through Kdm3a, which promotes transcriptional expression of MMP9 and enhances trophoblast invasion ( 100 ). Meanwhile, methyltransferase Suv39h2 regulates H3K9 methylation and is involved in the pluripotency and differentiation of mTSC. Knockout of SUV39H2 in mTSC leads to significant changes in the landscape of H3K9 methylation and triggers differentiation ( 101 ). H3K4me3 is predominantly enriched at the transcription start site of active genes to activate transcription ( 102 ). KDM5B, as a demethylase of H3K4me3, regulates the self-renewal and H3K4 methylome in mTSC ( 103 ). Lsd1 can erase out monomethyl and dimethyl groups from H3K4me2/3 or H3K9me2/3 to activate or silence gene transcription ( 104 , 105 ). Knockout of Lsd1 leads to abnormal development of EXE and ultimately embryo death ( 106 ). It has been demonstrated that Lsd1 directly suppresses the expression of Ovol2 to maintain proper development of the EXE ( 106 ). Lsd1 also regulates mitochondrial homeostasis in mTSC through the target gene Sirt4 ( 107 ). Furthermore, Lsd1 forms a complex with Jdjm2b and Tfap2c at critical gene promoters to activate transcription and safeguard the identity of mTSC ( 108 ). The abnormal H3K4me3 in the extraembryonic lineage caused by assisted reproductive technology (ART) procedures also results in developmental abnormalities. Knocking down the modifying enzyme Kmt2e can restore the expression of the corresponding genes and promote embryonic development ( 109 ). Besides, H3K4me3 expression is also found downregulated in human preeclampsia placenta ( 110 ). A recent study reported that human iPSCs can be induced to differentiate into trophoblast-like stem cells ( 111 ), which has important implications for both basic research and potential clinical applications. Interestingly, Yu et al. found that H3K36me2 modification plays a critical role in regulating the induction of pig iPSCs into trophoblast-like stem cells. Kdm4c , as a demethylase of H3K36me3/2, activates the expression of Cdx2 to promote this trans-differentiation process ( 112 ). The establishment of a model using human trophoblast stem cells (hTSC) is of great importance for investigating the processes of trophoblast proliferation and differentiation during human placental development ( 113 ). In this context, the addition of the histone deacetylase inhibitor VPA proved to be particularly effective in promoting hTSC proliferation during the establishment phase, while other inhibitors like SAHA or TSA showed similar effects. These findings underscore the significance of histone acetylation in maintaining the pluripotency and proliferation of hTSCs, although the precise underlying mechanism remains unclear. During the differentiation of primary cytotrophoblasts, significant decreases in H3K27ac, H3K14ac, and H3ac are detected. This observation has been further validated in the Bewo, an epithelial cell line isolated from the placenta of a patient with choriocarcinoma ( 114 ). Moreover, the examination of placental tissues from pregnancy-related diseases revealed alterations in H3K8ac, H3K27ac, H3K9ac, and the histone deacetylase SIRT1 to varying extents, indicating a potential involvement of histone acetylation in the pathological mechanisms underlying placental diseases ( 110 , 115 – 118 ). Using Sirt1 -null embryos and established mTSC, it was discovered that mTSC were unable to differentiate properly following Sirt1 knockout ( 119 ). Similarly, Xiong et al. achieved trophoblast-specific knockout of Sirt1 using Elf5-cre and observed the same phenotype, indicating that Sirt1 knockout might activate key genes in trophoblasts and disrupt their developmental trajectory ( 120 ). In addition, Histone deacetylase 1 and 2 have been found critical in driving hTSC differentiation by controlling the expression of TEAD4, TP63, OVOL1 and CGB ( 114 ). Additionally, HDAC6 has been proven required for trophoblast stem cell differentiation via directly deacetylates histones on the epithelial gene promoters such as Claudin 6 and Occludin during epithelial-to-mesenchymal transition (EMT) of TS cells ( 121 ). Besides deacetylases, histone acetyltransferases have also been found important in trophoblast. Cbp is widely expressed and capable of acetylating both histone and non-histone proteins ( 122 ). Abell and colleagues discovered that the absence of Map3k4 in mTSC leads to EMT, through a mechanism in which the knockout of Map3k4 prevents the direct phosphorylation of Cbp by JNK, thereby promoting the acetylation of histone proteins H2A and H2B and facilitating the transcription of relevant genes ( 123 ). P300 is an originally identified coactivator of CBP. It plays a crucial role in acetylation of H3K9 and K3K27 associated with 11β-HSD2 expression in syncytiotrophoblasts ( 124 ). Citrullination is a post-translational modification (PTM) that is catalyzed by the peptidyl arginine deiminase (PAD) enzyme family, which includes PADI1-4 and PADI6 ( 125 ). Recent findings shed light on the significance of histone citrullination. Ballasy et al reported that Padi2 and Padi3 , the most widely expressed members of the PAD enzyme family in mTSCs, play a crucial role in regulating differentiation. Knockout of Padi2 and Padi3 resulted in decreased expression of CDX2 and SOX2 , leading to a bias towards trophoblast giant cell (TGC) differentiation. Additionally, deletion of Padi2 and Padi3 had a substantial impact on the epigenomic landscape of mTSC, resulting in a reduction in H3K9me3 and DNA methylation. Further investigation showed that decreased DNA methylation of differentiation genes, such as Gata3 , Peg3 , Socs3 , and Hand1 , was the main cause of their increased expression during differentiation ( 126 ). Furthermore, histone lactylation has also been found to play a regulatory role in trophoblast cells. In patient with preeclampsia, the placenta showed a significant increase in the H3K18la modification compared to the control group. The activation of target genes FN1 and SERPINE1 by H3K18la promotes placental fibrosis ( 127 ), indicating the functional significance of histone lactylation in placental pathology.

In recent years, emerging evidence has underscored the significance of epigenetic regulations during embryo implantation and placentation, highlighting their crucial role in these processes. Here, we focused specifically on histone modifications and summarized the key findings in this field. Despite increased knowledge on the topic, there are still many unknowns regarding the molecular basis of histone modifications and their roles in the interaction between the embryo and the uterus during gestation. Specifically, while considerable attention has been given to histone methylation and acetylation in the context of embryo implantation and placentation, further investigations are required to fully unravel the complexities of the epigenetic regulatory network involved, encompassing other types of modifications as well. In addition, despite the increasing studies of various histone modification factors at the maternal-fetal interface, comprehending the physiological and pathological functions, as well as the underlying mechanisms of these factors, remain major challenges in this field. Moreover, the existing models used in relevant research have certain limitations, primarily relying on traditional cell lines and conditional knockout mouse models. Unfortunately, many essential genes crucial for implantation and placentation cannot be thoroughly studied, as their knockout often leads to embryonic lethality or developmental defects. Hence, there is an urgent need to develop inducible Cre systems with uterus/placenta-specific gene promoters, enabling a more precise assessment of histone modifications throughout different stages of pregnancy. Additionally, considering the disparities between mice and humans, it is imperative to develop more diverse models, such as organoids and cell chips, to overcome the current limitations and effectively identify therapeutic targets for diseases such as implantation failure, preeclampsia, and fetal growth restriction.

SB wrote the manuscript, ZT summarized the mouse models and prepared the figure, SZ and DC discussed the outline and revised the manuscript. All authors contributed to the article and approved the submitted version.