Downloaded from genesdev.cshlp.org on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press RESEARCH COMMUNICATION germline imprinting control regions (ICRs) is differen- Maternal-biased H3K27me3 tially established during gametogenesis and protected correlates with paternal-specific from the wave of global epigenetic reprogramming upon fertilization (Hanna and Kelsey 2014). gene expression in the human Recently, we discovered that oocyte-inherited H3K27me3 can also serve as a primary imprinting mark morula to regulate dozens of maternally imprinted genes, includ- Wenhao Zhang,1,2,3,9 Zhiyuan Chen,1,2,3,9 ing Xist, in mouse preimplantation embryos (Inoue et al. Qiangzong Yin,1,2,3 Dan Zhang,4,5,6 2017a,b). Oocyte-specific H3K27me3 domains are estab- 4,5 1,2,3,7,8 lished during oogenesis by the Polycomb-repressive Catherine Racowsky, and Yi Zhang complex 2 (PRC2) (Cao et al. 2002) and are specifically 1Howard Hughes Medical Institute, Boston Children’s Hospital, maintained on the maternal allele, which can orchestrate Boston, Massachusetts 02115, USA; 2Program in Cellular and paternal-specific gene expression in mouse early embryos Molecular Medicine, Boston Children’s Hospital, Boston, (Inoue et al. 2018). Although the majority of the putative Massachusetts 02115, USA; 3Division of Hematology/Oncology, maternal H3K27me3-regulated imprinting genes exhibit Department of Pediatrics, Boston Children’s Hospital, Boston, transiently imprinted expression in mouse early embryos, Massachusetts 02115, USA; 4Center for Infertility and at least five (i.e., Gab1, Phf17, Sfmbt2, Smoc1, and Reproductive Surgery, Brigham and Women’s Hospital, Harvard Slc38a4) maintain the imprinted expression to placenta Medical School, Boston, Massachusetts 02115, USA; and play an important role in placental development 5Department of Obstetrics, Gynecology, and Reproductive (Itoh et al. 2000; Sachs et al. 2000; Miri et al. 2013; Inoue Biology, Brigham and Women’s Hospital, Harvard Medical et al. 2017a; Matoba et al. 2018). School, Boston, Massachusetts 02115, USA; 6Key Laboratory of To determine whether asymmetric H3K27me3 dis- Reproductive Genetics (Ministry of Education) and Department tribution is also correlated to allelic gene expression in of Reproductive Endocrinology, Women’s Hospital, Zhejiang human early embryos, we profiled both allelic gene ex- University School of Medicine, Hangzhou, Zhejiang, 310006, pression and allelic H3K27me3 distribution in human P.R. China; 7Department of Genetics, Harvard Medical School, morulae. The analysis revealed that human morula Boston, Massachusetts 02115, USA; 8Harvard Stem Cell H3K27me3 is enriched at developmental gene promoters Institute, Boston, Massachusetts 02115, USA and tends to overlap with DNA partially methylated domains (PMDs). Importantly, H3K27me3 exhibits mater- Genomic imprinting is an epigenetic mechanism by nal-biased distribution and is enriched at promoters of sev- which genes are expressed in a parental origin–dependent eral paternally expressed genes (PEGs) that do not contain manner. We recently discovered that, like DNA methyla- germline differential DNA methylation. Taken together, tion, oocyte-inherited H3K27me3 can also serve as an im- our data suggest that maternal-biased H3K27me3 could printing mark in mouse preimplantation embryos. In this regulate paternal-biased gene transcription in human early study, we found H3K27me3 is strongly biased toward the embryos. maternal allele with some associated with DNA methyla- tion–independent paternally expressed genes (PEGs) in Results and Discussion human morulae. The H3K27me3 domains largely overlap with DNA partially methylated domains (PMDs) and oc- H3K27me3 is enriched at developmental gene promoters cupy developmental gene promoters. Thus, our study in the human morula not only reveals the H3K27me3 landscape but also estab- To investigate the general distribution of H3K27me3 in lishes a correlation between maternal-biased H3K27me3 the early human embryo, we profiled H3K27m3 using and PEGs in human morulae. the cleavage under targets and release using nuclease Supplemental material is available for this article. (CUT&RUN) assay (Skene et al. 2018) with morulae pooled from two separate couples. In parallel, to determine Received November 30, 2018; revised version accepted whether any allelically expressed gene is regulated by January 18, 2019. H3K27me3 in human early embryos as in mice (Inoue et al. 2017a), we also performed single morula RNA se- quencing (RNA-seq) using 15 morulae from five different Genomic imprinting is an epigenetic phenomenon that couples (Fig. 1A). To infer the allelic information from leads to parental allele-specific gene expression (Bartolo- the RNA-seq and the CUT&RUN data, maternal cumulus mei and Ferguson-Smith 2011). Since the discovery of cells from the five couples were genotyped by whole- this phenomenon, allele-specific DNA methylation has exome sequencing (WES) and/or whole-genome sequenc- been the only underlying mechanism (Bartolomei and Fer- ing (WGS) (Fig. 1A; Supplemental Table S1). guson-Smith 2011). Allelic DNA methylation at the We first evaluated reproducibility of the two H3K27me3 CUT&RUN libraries generated using morulae from two [Keywords: allele-specific gene expression; genomic imprinting; © 2019 Zhang et al. This article is distributed exclusively by Cold Spring maternal-biased H3K27me3; DNA methylation; human morulae] Harbor Laboratory Press for the first six months after the full-issue publi- 9These authors contributed equally to this work. cation date (see http://genesdev.cshlp.org/site/misc/terms.xhtml). After Corresponding author: [email protected] six months, it is available under a Creative Commons License (Attribu- Article published online ahead of print. Article and publication date are tion-NonCommercial 4.0 International), as described at http://creative- online at http://www.genesdev.org/cgi/doi/10.1101/gad.323105.118. commons.org/licenses/by-nc/4.0/. 382 GENES & DEVELOPMENT 33:382–387 Published by Cold Spring Harbor Laboratory Press; ISSN 0890-9369/19; www.genesdev.org Downloaded from genesdev.cshlp.org on October 5, 2021 - Published by Cold Spring Harbor Laboratory Press Maternal-biased H3K27me3 in the human morula (Supplemental Fig. S1D). These findings are consistent A B with the known function of Polycomb proteins in regulat- ing developmental genes and suggest that H3K27me3 at promoters may already begin to play a role in regulating gene transcription in human morulae. Global H3K27me3 tends to overlap with PMDs in the human morula In mice, although promoter H3K27me3 in oocytes is re- programmed upon fertilization and becomes equalized with the paternal allele, maternal distal H3K27me3 do- mains are largely inherited, which leads to allele-specific C D H3K27me3 distribution during the preimplantation stage (Zheng et al. 2016). In addition, because de novo DNA methylation during oogenesis is transcription dependent (Chotalia et al. 2009; Veselovska et al. 2015), transcribed regions in oocytes overlap with DNA hypermethylated do- mains, whereas nontranscribed regions are associated with PMDs and H3K27me3 peaks (Zheng et al. 2016). Sim- ilar to these observations, H3K27me3 in human morulae also tends to overlap with PMDs exemplified by genome browser views (Fig. 2A). To gain further support for this initial observation, we identified PMDs (regions with length >10 kb and average DNA methylation level <40%) (see Supplemental Material) using the human mor- ula whole-genome bisulfite sequencing (WGBS) data (Zhu Figure 1. H3K27me3 is enriched at developmental gene promoters et al. 2018) and found that, indeed, a subgroup of PMDs in the human morula. (A) A schematic summary of the data sets gen- & is enriched with H3K27me3 (Fig. 2B). To determine coloc- erated in this study. H3K27me3 CUT RUN was performed using alization of H3K27me3 within PMDs, we identified pooled morulae collected from couple 1 (C1; n = 7) and couple 2 (C2; & n = 8). Three replicates for single morula RNA-seq were performed H3K27me3 domains shared by both CUT RUN data for each couple. (B) Heat map showing gene expression levels and sets (Supplemental Fig. S2A,B; Supplemental Material) H3K27me3 enrichment at the corresponding gene bodies in human and found that the majority of H3K27me3 domains were morulae. RPKM values are Z-score normalized and log2-transform located in morula PMDs (77.3%, 11,182 out of 14,455, & normalized for CUT RUN and RNA-seq, respectively. (C) A genome vs. random regions: 49.7%, 7191 out of 14,455) (Fig. 2C). browser view showing H3K27me3 enrichment at HOXA and HOXD clusters, PAX6, and SOX3 in both morula and human embryonic Collectively, these results suggest that, similar to mouse, stem cells (hESCs). (D) Heat map showing enrichment of H3K27me3 human H3K27me3 enrichment in early embryos tends to at promoters (transcription start site ± 2.5 kb) in morulae (two repli- overlap with PMDs, consistent with the reverse relation- cates) and hESCs. RPKM was subject to Z-score normalization. ship between DNA methylation and H3K27me3 (Bogda- novićet al. 2011; Zheng et al. 2016). pairs of couples. Even though only seven to eight morulae & from each couple were available for the CUT RUN assay, H3K27me3 is biased toward the maternal allele the two H3K27me3 CUT&RUN replicates showed high in the human morula reproducibility (Supplemental Fig. S1A). Additionally, consistent with the notion that H3K27me3 is a transcrip- To determine whether H3K27me3 in human morulae ex-