Sex-influenced DNA methylation differs by placental cell type

doi:10.21203/rs.3.rs-8118227/v1

Download PDF

Research Article

Sex-influenced DNA methylation differs by placental cell type

https://doi.org/10.21203/rs.3.rs-8118227/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Sex differences in the function and morphology of the human placenta can lead to sex differences in pregnancy outcomes. X chromosome inactivation (XCI) is the primary mechanism for dosage compensation between the sexes, and is strongly associated with X-chromosome promoter DNA methylation (DNAme) in somatic cells. However, in the placenta, low X-chromosome promoter DNAme has been reported. The placenta is a complex organ consisting of cells of different developmental origins, but the sex differences in DNAme by specific cell types have not been investigated.

Methods

We examined sex-influenced DNAme from 18-19 samples each of endothelial, stromal, cytotrophoblast and Hofbauer cells, sorted from term placentas, as well as matched whole chorionic villi. We also compared these profiles with data from 65 endothelial cell samples from placental chorionic plate arteries and veins (XX=16, XY=13) and umbilical cord veins (XX=22, XY=14). All data were derived from Illumina Infinium HumanMethylation450 or EPIC DNAme arrays. Sex-stratified analyses of the X/Y and autosomal DNAme were undertaken to identify DNAme differences associated with sex chromosome complement.

Results

The DNAme distribution on the X-chromosome differed by cell type in a manner that reflected differing developmental origins. Three distinct patterns were observed in XX placental cells reflecting origins from extraembryonic mesoderm (endothelial/stromal), trophectoderm (cytotrophoblast) and epiblast (Hofbauer cells), the latter of which shared a similar distribution with blood and umbilical endothelial cells. Interestingly, the typical XCI-associated DNAme at promoter CpG islands (CGI) on the X-chromosome of XX cells was absent for endothelial/stromal cells and present only at low levels in trophoblasts, suggesting that de novo establishment of promoter-CGI DNAme on the X-chromosome may differ by the developmental origins of each cell type. Y-chromosome DNAme also varied by cell type.

Conclusion

The lack of promoter DNAme in extraembryonic mesoderm-derived cells (endothelial/stromal) suggests a distinct developmentalorigin of these populations relative to the other placental and umbilical cell types. Autosomal DNAme also showed cell-type differences consistent with a common developmental origin of endothelial/stromal cells distinct from other placental cell types. This work suggests the effects of sex chromosome complement on pregnancy outcomes may differ by placental cell type.

DNA methylation

endothelial cell

placenta

sex

X chromosome

• X and Y- chromosome DNA methylation differs by placental cell type.

• DNA methylation patterns reflect the distinct developmental origins of different placental cell types.

• DNA methylation typically present at X-linked promoters in somatic XX cells was absent in villous endothelial and stromal cells.

• Endothelial cells in the small vessels of placenta differ strikingly in DNAme from those in umbilical endothelium.

The human placenta is a complex organ comprised of diverse cells of different origins, and its function and structure exhibit sex differences associated with pregnancy outcomes. We examined the DNA methylation (DNAme) of placental cells in autosomes and sex chromosomes separately to investigate the impact of sex on the DNAme of placental cells. We used 94 placental cell samples (XX=50, XY=44) including 18-19 samples each of endothelial, stromal, Hofbauer cells, cytotrophoblasts and whole chorionic villi, and compared these with public data from 66 endothelial cell samples derived from placental plate artery and vein (XX=17, XY=13) and umbilical vein (XX=22, XY=14). The DNAme distribution of the X-chromosome differed by cell type in a manner that reflected differing developmental origins. Female (XX) placental endothelial/stromal cells showed distinct DNAme distributions from cytotrophoblast, and both differed from that of Hofbauer cells, which shared a similar DNAme distribution with blood. Interestingly, the typical DNAme associated with X chromosome inactivation was absent or low in endothelial/stromal cells and cytotrophoblasts, suggesting that the DNAme patterns may differ by cell types and their origins. Y-chromosome and autosomal DNAme also showed cell-type differences consistent with a common developmental origin of endothelial and stromal cells distinct from other placental cell types. This work provides insight into the influence of sex and cellular developmental origin on DNAme of mature human placental cell types.

The placenta is the core organ that mediates fetal development and growth during pregnancy. As it develops from the zygote, it normally has the same sex chromosome complement as the fetus. Sex differences in placental function may contribute to sex differential fetal development and growth [1, 2], and may be the result of several influences [3]. Differences in placental gene expression due to sex chromosome complement (XX or XY) lead to sex-influenced functional and physiological features even before the development of fetal gonads, and continue throughout gestation [4, 5]. In addition, differences in sex hormone exposure or indirect effects of fetal sex, such as differences in immune regulation, may impact placental gene expression and function [6–8].

In mammalian cells with two X chromosomes, X-chromosome inactivation (XCI) occurs to equalize gene dosage to the single X found in XY cells [9, 10]. XCI is the epigenetic silencing of one X leading to the transcription of only one copy of most X-linked genes; however, escape from XCI of some genes can lead to sex-dependent expression [11]. DNA methylation (DNAme) is one of multiple epigenetic marks acquired after the initiation of XCI [12, 13], and DNAme at promoter CpG Islands (CGI) on the inactive X chromosome (Xi) is typically viewed as a hallmark of XCI status [14, 15]. However, in the human placenta DNAme levels at X-chromosome promoters is generally lower than in fetal or adult somatic tissues [16–18]. We recently reported that low promoter DNAme is found at most genes on the placental Xi, regardless of their gene expression XCI status [18]. We also found that cell composition affects X-chromosome DNAme profiles in the placenta [18]. However, to our knowledge, there are no studies focused on sex-influenced DNAme in the context of diverse placental cell types and their cellular origins.

Chorionic villi (CV) are the functional units of the placenta, and are comprised of cells derived from both the trophectoderm and the inner cell mass (ICM) of the blastocyst. After implantation, trophectodermal cells differentiate into mononuclear cytotrophoblasts and their fusion product, syncytiotrophoblasts, forming primary chorionic villi [19, 20]. Secondary villi are formed by migration of extraembryonic mesenchymal cells to the villous core [19, 20]. Continuous proliferation of mesenchyme and the formation of fetal capillaries leads to the development of tertiary villi [19, 20]. In addition, a subset of cytotrophoblasts penetrate the maternal decidua from the CV and are referred to as extravillous trophoblasts [21]. The villus core thus consists of ICM-derived cell types including Hofbauer cells, placental endothelial cells and stromal (fibroblast) cells. Hofbauer cells are placental macrophages which ultimately differentiate from fetal monocytes or placental erythro-myeloid progenitors [22–24].

During early development, DNAme erasure occurs in the blastocyst, and subsequent de novo DNAme is established in the various placental cell lineages upon their differentiation, leading to lineage specific DNAme profiles [24–26]. DNAme associated with XCI in the inner cell mass (a subset of which will give rise to somatic cells) is similarly established after implantation, as cells begin to differentiate) [27]. In addition to studying epigenetic processes like XCI, DNAme studies have demonstrated placental sex differences across the autosomes [18, 28–30], However, due to the rarity of sorted placental cell DNAme profiles, our knowledge of autosomal sex differences across the different placental cell types, and their X and Y DNAme patterns, remains unexplored.

To study the influence of sex and cell type on placental DNAme, we investigated the DNAme profiles of sex-stratified autosomes, X and Y-chromosomes of four isolated human placental cell types (endothelial, stromal, Hofbauer cells, and cytotrophoblasts) with matched whole chorionic villus samples, using the 850K Illumina EPIC v1.0 DNA methylation array [24]. We observed extensive cell-type variation in X-chromosome DNAme, corresponding with differing cellular origins early in development. Our study shows that placental DNAme is complex, with unique sex and cell-influenced profiles on the X and Y-chromosomes.

Data processing

This study is based on the Illumina EPIC v1.0 methylation array data from Yuan et al. (2021) (GEO ID: GSE159526) [24]. All samples were karyotypically normal by multiplex ligation-dependent probe amplification. These data are derived from FACS-sorted cells from 19 karyotypically normal term placentas (gestational age of 36.4–40.4 weeks), including trophoblast cells, Hofbauer cells, endothelial, and stromal cells, as well as matched whole chorionic villi. Analyses were performed in R version 4.3.1. The IDAT files and phenotype data were processed following our established pipeline (Supplementary Fig. 1) [31]. X and Y-chromosome probes were used to confirm the recorded sex in sample data, and 59 SNP “rs” genotyping probes, within the array, were used to identify potentially contaminated samples and possible sample mix-ups. Ancestry probabilities of whole chorionic villi samples were estimated using the PlaNET R package [32]. Sample filtering was as described in Yuan et al. and limited to the 94 term samples including 19 (XX = 10, XY = 9) chorionic villi (CV), cytotrophoblasts (CTB), endothelial (EC), stromal cells (SC) and 18 (XX = 10, XY = 8) Hofbauer cells (Supplementary Table 1).

CpG probes with fluorescence detection p-value of > 0.01 or bead count < 3 in > 5% of samples were removed, following the protocol of Inkster et al. [33] to appropriately filter the X and Y-chromosome probes. A list of cross-hybridizing probes was obtained from Zhou et al. (2017) and removed [34]. The samples used in the original Yuan et al study were normalized by normal-exponential out-of-band (Noob) [35] and Beta-Mixture Quantile (BMIQ) normalization [36], and included the same number of samples and autosomal CpGs. On the X-chromosome, additional probes, including those in repetitive elements (n_CpGs = 1,975), cancer testis genes (n_CpGs = 622) and the X-transposed region (n_CpGs = 80) were removed [33]. After processing, 737,050 autosomal CpGs, 14,766 X-linked CpGs, and 293 Y-linked CpGs remained for downstream analyses.

Public endothelial cell data processing

Datasets of human placental villous arterial and venous endothelial cells (pAEs/pVEs) and human umbilical vein endothelial cells (uVEs) were used from Cvitic et al. (2018) (GSE106099) and Rhead et al. (2020) (GSE144804) in the form of IDAT files. Downloaded IDAT files of pAE/pVE Illumina 450K DNAme data of 30 samples were processed by following the same processing steps above. Unnecessary autosomal probes were filtered (n = 476,682), and 8,830 X-linked probes were used for analyses. PCA, sample donor checks using SNPs, hierarchical clustering, and sex checks were performed to verify the sample information. The pAE sample AEC-110 was removed as it did not cluster with its reported cell type and is likely mislabeled. There was a mismatch in AEC-103 between the reported sex and the predicted sex, and sex was accordingly reassigned as “XX”. No sex was reported for the pVE sample VEC-2d but this sample could be assigned as “XY” by performing the sex check using DNAme data, leaving a total: XX = 16, XY = 13 samples available for analysis.

Downloaded IDAT files of uVE Illumina EPIC DNAme data (n = 74) were first filtered to remove TNF-α (Tumor Necrosis Factor alpha)-treated samples (n = 37) and then were processed by following the same processing steps above, including removal of 3 samples with low fluorescence intensities to exclude 38 samples in total. Samples with GEO reported sex disagreeing with DNAme-derived sex chromosome complements (n = 6) were relabeled according to DNAme-derived sex [33]. After processing 14,861 X-chromosome probes in 36 samples (XX = 22, XY = 14) were available for analysis. Processed 450K and EPIC DNAme data were limited to the shared CpG probes, and a total 8,012 X-chromosome probes common to both arrays were used for data analyses.

Annotation

Probe annotation for UCSC CpG island, location, manifest and others for Illumina 450K and EPIC DNAme data were taken from R package IlluminaHumanMethylation450kanno.ilmn12.hg19 (v0.6.1) and IlluminaHumanMethylationEPICanno.ilm10b4.hg19 (v0.6.0) [37, 38]. The regulatory regions and UCSC annotations for the EPIC array from Bizet et al. (2022) were used for annotating CpG probes to CpG islands and regulatory regions [39].

Differentially Methylated CpGs (DMCs)

To identify differentially methylated CpGs, the R package limma (ver. 3.5.0) was used to build linear models with empirical Bayes moderation [40]. Linear regression models for the autosomes and X-chromosome were generated to compare the average DNA methylation between XX and XY. For Y-chromosomes, a linear model with a contrast matrix was built to compare the DNA methylation of one cell type to the average of all other cell types within each cell type [24]. Statistically significant DMCs were identified at a false discovery rate (FDR) of < 0.05 and mean DNAme difference > 10%. For gene set enrichment of DMCs, the R package missmethyl (v1.36.0) was used to test the enrichment of DMCs in specific gene sets [41].

Nucleotide BLAST analysis for Y-chromosome cross-hybridization

Command-line nucleotide BLAST was performed on the significant cell-DMCs of Y-chromosomes using the 50-nucleotide probe A and B sequences of DMCs. The blastn for short sequences (-short) was run against databases generated from the Human Genome build 19 (hg19) to filter out cross-hybridizing probes on the X chromosome. To detect any low chance of cross-hybridization on the X chromosome, sequences that match at nucleotide position 50 and are ≥ 40 bp and have ≥ 75% sequence identity to the X chromosome were excluded based on criteria from Chen et al. and Inkster et al [28, 42].

X-linked DNAme variation in placental cells is driven by cell type and sex.

To identify the primary variables associated with X-linked DNAme variation in the cell-type specific data from GSE159526, we first performed principal component analysis (PCA) on DNAme at 14,766 X-chromosome probes in all 94 samples. We tested for association between DNAme variation described by the principal components (PC scores) and sample variables (via linear models of the form PC ~ variables of interest) (Figure 1A, 1B). Cell type of the individual DNAme sample was strongly associated with the top 4 PCs (all nominal p-values < 0.001), which together explained > 70% of the X-chromosome DNAme variation in the dataset (PC1: 30%, PC2: 23%, PC3: 15%, PC4: 6%). Sex of the sample (XX or XY) was associated with PC1 and PC3 (p-value < 0.001). Although PC1 primarily separated samples by sex (Figure 1C), there was further subdivision by cell type within each sex along PC1, with XX stromal and endothelial cells clustering more closely to XY cells, while XX cytotrophoblasts and villi fell furthest away from XY samples along PC1. PC2 predominantly separated Hofbauer cells from all other cell types. Inferred ancestry probabilities and the technical chip variable (Sentrix ID) had only weak or non-significant associations with any of the top 10 PCs (Fig 1A, 1B).

To further visualize the relationship between sex and cell type, we performed hierarchical clustering on the same 14,766 X-chromosome probes (Figure 1D). This revealed three major clusters based on cell type: (i) cytotrophoblasts and villi, (ii) Hofbauer cells, and (iii) endothelial and stromal cells. Trophoblasts are the predominant cell type in whole chorionic villus tissue, which explains the clustering of cytotrophoblasts with villi in this analysis. Samples further subdivided by sex within each of the three main cell type clusters. Together with the PCA results, these hierarchical clustering results showed that (i) both sex and cell type are major drivers of X-chromosome DNAme in this cohort, and (ii) cell-specific patterns of DNAme are similar between cells with shared developmental origins (i.e. endothelial and stromal cells).

The distribution of X-linked DNAme of placental cells shows three distinct patterns.

To better understand what differentiates cell-type specific patterns of X-chromosome DNAme, we compared the distributions of X-chromosome DNAme across all cell types, separately in each sex (Figure 2A, 2B, Supplementary table 2). In XX cells, three general distributions of DNAme were observed. Hofbauer cells displayed a trimodal distribution, as is typical for somatic cells: a low methylated peak (0.2 ≤ β), a high methylated peak (β < 0.8), and a distinct peak of intermediate DNAme (0.2 < β ≤ 0.8). The intermediate DNAme peak reflects allele-specific DNAme associated with XCI, where promoters are fully methylated on the Xi and fully unmethylated on the Xa [43]. By contrast, cytotrophoblast and whole chorionic villi showed relatively few highly-methylated sites and lacked distinct intermediate methylation peaks. Both endothelial and stromal cells showed a distinct peak of low DNAme and a smaller peak of high DNAme, but had no clear intermediate DNAme peak, similar to the distribution observed amongst XY samples across all cell types.

DNAme of X-linked promoters show few sex differences in endothelial and stromal cells.

In somatic tissues, most X-chromosome gene promoters are roughly 50% methylated (β » 0.5) in XX cells, and unmethylated in XY cells [44]. Although whole chorionic villi show lower DNAme at X-linked promoters relative to somatic tissues (XX), genes subject to XCI in placenta do tend to show relatively higher levels of X-linked promoter DNAme in XX versus XY samples [18]. To compare the level of X promoter DNAme amongst the different cell types we calculated the sex difference in DNAme (|∆β| = XX - XY) at the 1,393 CpGs in X-linked promoter-associated CGI, based on CGI regions defined by Bizet et al. (2022) (Figure 2C). Sex differences in DNAme (i.e. DNAme |∆β| > 0.1) were observed at most X-linked CGIs in Hofbauer cells (76%) and at many CGI loci in cytotrophoblasts (45%). However, in endothelial and stromal cells, very few X-linked CGIs (3% and 7% respectively) had a sex difference in DNAme of |∆β| > 0.1, consistent with low/absent DNAme in both sexes at X-linked CGI promoters in endothelial and stromal cells. The lack of X-linked promoter DNAme in endothelial and stromal cells is also illustrated by a high correlation of X-linked DNAme between these cell types in both XX and XY samples (Supplementary Figure 2A, 2B), and our PCA showing close clustering of these cell types (Figure 1C).

In somatic cells, unlike promoters, gene bodies and intergenic regions on the X-chromosome show lower DNAme in XX compared to XY tissues [45]. We wanted to determine if the same patterns can be found in placental cells. To further evaluate DNAme by X-chromosome genomic region, we identified sex-differentially methylated CpGs (sex-DMCs) in each cell type using linear regression models with thresholds of |∆β| > |0.1| and FDR < 0.05) (Supplementary Table 3). As expected, most (45 - 75%) of the X promoter DMCs had higher XX relative to XY DNAme in Hofbauer cells, cytotrophoblasts and chorionic villi, but not in endothelial or stromal cells where few loci (4% or 7%, respectively) had sex-differential DNAme (Figure 2D). However, all cell types showed higher DNAme in XY relative to XX cells at DMCS located in X-linked enhancers, intergenic regions, and gene bodies, with cytotrophoblast showing the greatest number of sex-differentially methylated sites.

DNAme profile of placental endothelial cells differs from umbilical cord endothelial cells

To further characterize the unique X-linked DNAme patterns we observed in placental endothelial and stromal cells, we compared them to endothelial cells derived from other gestational tissues to determine whether we were detecting a specific endothelial X-linked DNAme signature. The endothelial cells evaluated in this work were FACS-isolated and were derived from placental microvessels within the chorionic villi. Blood flows between these placental microvessels and larger arteries and veins within the placental chorionic plate; these larger vessels then connect to the fetal vasculature via the umbilical cord. While, the placental villous endothelial cells derive from extraembryonic mesoderm, endothelial cells within the fetal compartment (fetal vessels) like the umbilical cord vessels, derive from embryonic precursors [46]. We thus sought to determine if low X-linked promoter DNAme was a shared property of all endothelial cells including those derived from the umbilical cord and large vessels of the placenta. To evaluate this, we compared DNAme patterns of our placental microvascular endothelial cells (henceforth called “pME”), to public Illumina HumanMethylation 450K data derived from (i) cultured human placental arterial and venous endothelial cells (pAE/pVE) obtained from the chorionic plate (GSE106099) (n_pAE = 12, XX = 7, XY = 5 / n_pVE = 17, XX = 9, XY = 8), and EPIC data derived from (ii) cultured human umbilical venous endothelial cells (uVE) (GSE144804) (n_uVE = 36, XX = 22, XY = 14) (Supplementary table 4).

PCA and hierarchical clustering on the 8,012 X-chromosome CpGs common to all endothelial cell datasets (total sample n = 65) demonstrated DNAme differences by both sex and endothelial cell sampling location (Figure 3A-3C). PC1 separated samples by sex and cell type, with the greatest separation between XX and XY cells observed in uVEs and least separation in pAEs (Figure 3B). Hierarchical clustering also showed separation by both sex and cell type, although XX pVE clustered with XY samples (Figure 3C). The X-chromosome DNAme distributions also showed similar trends (Figure 3D), with XX uVEs showing a distinct intermediate methylated peak (57% at 0.2 < ∆β ≤ 0.8) characteristic of XX somatic cells, while in pAEs and pVEs this peak was largely absent and a large portion of X-linked CpGs had low DNAme (41%, and 57% respectively at ∆β ≤ 0.2), similar to what we observed in pMEs. In contrast, X-chromosome DNAme in XY samples was similar across all endothelial cell types.

Most promoter-CGI CpGs showed a difference in DNAme by sex (∆β > |0.1|) in uVEs (76%), but not pVEs (9%), while pAEs (44%) showed an intermediate result (Figure 3E). We next tested the cell-type correlation in DNAme at all X-chromosome CpGs (Supplementary Figure 3A, 3B). The strongest correlation was observed between XX pVE and pME (r = 0.78), and uVE and pAE (r = 0.75). The weakest correlation was observed between pME and uVE samples. In other words, placental microvessels (pME) studied in our first set of analyses were most similar to placental venous endothelial cells (pVE), while the umbilical cord vein (uVE) was most similar to placental artery (pAE). These results overall suggest considerable heterogeneity in the developmental origin of endothelial cells amongst these tissues, as inter-cell correlations in X DNAme between cells of similar origin is typically over 0.90 [47].

Y-chromosome DNAme varies by cell type.

Like the X, the Y-chromosome is also under-studied in epigenome-wide DNAme analyses. As the Y contains few genes, most of which function in the testes, we did not anticipate many DNAme differences by cell type in placenta. Nonetheless, PCA and hierarchical clustering of XY samples based on Y-chromosome DNAme (n_CpGs = 293) showed three distinct clusters by cell type, parallelling those observed for the X-chromosome (Figure 4A-4C). In linear models comparing the average Y-chromosome DNAme of each cell type to the average of all other cell types, we identified many cell-influenced Y-chromosome DMCs (78, 84, 105, and 116 DMCs in endothelial, stromal, Hofbuaer cells and cytotrophoblasts, respectively) (Supplementary Figure 4). These cell-influenced Y DMCs overlapped 13 genes (see Supplementary Table 5). Five of these 13 genes (DDX3Y, EIF1AY, RPS4Y1, USP9Y, and ZFY), were previously reported to be expressed in XY term placentas [5]. These genes all possess X-linked homologs, which are also expressed in XX term placentas [5]. To confirm that Y-chromosome DNAme attributed to these X-Y homologs in our dataset was not arising from DNAme array probes cross-hybridization to their X-linked pairs, we performed a Command-line nucleotide BLAST (blastn) for short sequences on the 50-nucleotide probe sequences (probeSeqA/B) of significant Y-linked cell-DMCs of all cell types to exclude any possibility of cross-hybridization on the X chromosome due to its sequence between X and Y chromosomes. None of the Y DMC probes filtered by the selected criteria matched, suggesting that our results mostly reflect true Y-chromosome DNA methylation patterns.

Sex-influenced autosomal DNA methylation differs by cell type

We previously reported 145 CpGs on the autosomes that show sex-influenced DNAme in whole chorionic villi [28]. To determine if these sex-influenced autosomal CpGs were consistent across placental cell types we performed PCA using these 145 sex-influenced CpGs in the sorted placental cells. In scatterplots of PC1 versus PC2, samples predominantly separated by cell type and not sex; only the cytotrophoblasts/chorionic villi separated by sex (Figure 5A). These results likely reflect that the 145 CpGs were originally identified as having sex differences in data from bulk villi, and may not show a sex difference in other cell types.

Although our samples sizes were small, we wanted to evaluate the existence of sex-influenced autosomal DNAme in the individual placental cell types. By comparing DNAme in each cell type by sex using linear models, we identified multiple significant sex-influenced DMCs in endothelial cells (n_CpGs = 35 DMCs at FDR < 0.05 and Δβ > |0.1|) and whole chorionic villi (n_CpGs = 7 DMCs at FDR < 0.05 and Δβ > |0.1|) (Figure 5B, Supplementary figure 5), but none in the other cell types. Of the 35 endothelial sex-associated DMCs, the majority (89 %) had higher DNAme in XY as compared to XX cells, and were located in promoter or enhancer regions. Among the endothelial-associated sex-DMCs were CpGs in genes including LDB3, INHBB, NSD1, RAB7A, and ZNF300; of these genes LDB3 and ZNF300 had 2 or more DMCs each (n_CpGs = 2/4, respectively).

As we were likely underpowered to detect sex differences in all cell types, to evaluate whether the sex differences in DNAme identified in endothelial cells were truly cell-specific, we performed PCA on the endothelial sex-associated DMCs in all placental cell types. The scatterplot of PC1 versus PC2 (Figure 5C) showed separation by cell type on PC1, with separation by sex observed in all cell types along PC2. Not surprisingly, the sex difference in endothelial cells was greatest, however, these results suggest that some of the identified DMCs show sex differences across multiple cell types, and that we are likely underpowered and missing significance in detecting sex-associated DMCs in other sorted placental cell types.

Finally, considering sex differences in DNAme at specific genes, the ZNF300 gene was previously reported to be DM by sex in Inkster et al., (2021), and was shown to be associated with placental morphology and development [48]. In endothelial cells, we identified 4 sex-DMCs at ZNF300 (DNAme XY > XX in all cell types except Hofbauer cells), in the same promoter region reported to be sex-DMCs in Inkster et al., (2019). Among the other sex-DMCs we identified in endothelial cells, several were associated with NSD1 and LDB3: 2 DMCs were observed in NSD1 (Nuclear Receptor Binding SET Domain Protein 1) and LDB3 (LIM domain binding 3), which plays a role as histone lysine methyltransferase and generate proteins maintaining the stability of the muscle structure, with higher DNAme in XX than XY cells (both endothelial and cytotrophoblast). Figures exemplifying the sex-differential DNAme patterns at these genes are shown in Supplementary Figure 6.

In this study, we characterized patterns of X and Y-chromosome DNAme and sex-influenced autosomal DNAme in the major placental cell types (endothelial, stromal, Hofbauer, and cytotrophoblast cells), and whole chorionic villi. Specifically, we observed that placental cell types occupy one of three broad patterns of X-chromosome DNAme: (i) Endothelial/Stromal, (ii) Hofbauer, and (iii) Cytotrophoblast/Trophoblast, suggesting that patterns of X-linked DNAme may differ by the developmental origins of placental cells, as much of X chromosome DNAme patterning is established proximally to the differentiation timing of these lineages. We observed a similar three-group pattern of cell-type DNAme differences on the Y-chromosome, further supporting that these DNAme groups represent distinct developmental lineages (Fig. 6).

DNAme, including that on the X-chromosome, is largely erased in the first few cell divisions after fertilization, with de novo DNAme occurring near or after blastocyst implantation. XCI is characterized by the accumulation of epigenetic marks on one of the two X-chromosomes in XX cells, and is associated with gene silencing on the inactive X-chromosome (Xi). In early implantation period, XCI occurs as tissues differentiate from the blastocyst in humans [49, 50]. In somatic cells, CGI promoters are typically highly methylated on the Xi when the associated genes are silenced [15, 43]. In contrast, human placental chorionic villus samples show low levels of DNAme of promoters on the Xi relative to all somatic tissues studied, however allelic inactivation at the level of gene expression does still occur [51] and is somewhat correlated with DNAme levels [18].

Here we show that the distribution of DNAme on the X in cytotrophoblast cells mirrors patterns observed in whole chorionic villi with low, but not absent, promoter DNAme in XX samples. Cytotrophoblast cells are derived from the trophectoderm and together with their fusion product, the multinucleated syncytiotrophoblast, are the primary components of chorionic villi, typically accounting for > 80% of DNA content in bulk whole villi tissue [24, 31]. These cells have previously been reported to have similar autosomal DNAme profiles to whole chorionic villi, as well as syncytiotrophoblast [24]. We now extend those observations to the sex chromosomes.

Hofbauer cells are a minor cell population making up < 5% of cells in whole villi [24]. We observed that the X-chromosome DNAme distribution of Hofbauer cells resembled the pattern found in XX somatic cells, including blood and buccal cells, with large peaks of intermediate promoter DNAme reflecting methylation on the Xi and lack of DNAme on the Xa [43, 52]. The developmental origin of Hofbauer cells has been debated, and may vary with both the time of gestation and potentially the cell type isolation approach. It was proposed that mesenchymal progenitor cells give rise to first trimester Hofbauer cells [53, 54], while second and third trimester Hofbauer cells are proposed to derive from fetal monocytes [55, 56]. It has also been suggested that hypoblast-derived placental erythro-myeloid progenitors differentiate into Hofbauer cells throughout pregnancy [57]. Others have also suggested Hofbauer cells in term placentas have an epiblast origin [58, 59]. In our previous investigation of autosomal DNAme from this same dataset Hofbauer cells lacked many placenta-specific DNAme features such as partially methylated domains and placenta-specific imprinting, and in addition, hierarchical clustering of DNAme of Hofbauer cells with cord blood cell types showed Hofbauer cells clustered closest to monocytes, consistent with an origin from this cell type [24].

Although XX trophoblast cells have low mean X-linked promoter DNAme, we observed that endothelial and stromal cells had lower still X-linked promoter DNAme. During XCI, multiple epigenetic marks accumulate on the inactive X, of which DNAme is the last and is hypothesized to be important in XCI maintenance [60, 61]. In whole villi, which reflects largely trophoblast, XCI silencing appears to occur to a similar level as in somatic cells [51], despite low DNAme of X-linked promoters, suggesting that DNAme is not essential for XCI maintenance in these cells [18]. It is possible that the same is true in endothelial and stromal cells, however, largely absent promoter DNAme could indicate a greater potential for XCI escape in these cells. Allele-specific expression analyses in pure endothelial/stromal cell populations will be required to definitively answer this question. Previous work has shown that under some circumstances cultured chorionic villi, which represent mainly placental fibroblasts (stroma), can undergo reactivation of certain X-linked genes after depletion of DNAme via treatment with 5-deazacytidine [62], suggesting a potential functional importance of our results. Additional evidence toward potential instability of XCI in stroma comes from hybrids of cultured human term chorionic villi (fibroblasts) hybridized with mouse cells, which were demonstrated to reactivate a subset of X-linked genes [62].

Placental endothelial and stromal cells derive from extraembryonic mesoderm (exM) [63], the developmental origins of which (in humans) are still uncertain. While the literature commonly cites these cells as being epiblast-derived, this seems largely inferred from animal models (i.e. mouse, rats), and may not be comparable to the human cells examined here, given the lack of villus structure in rodent placentas [64, 65]. Studies in rhesus monkey and in vitro exM models have proposed a hypoblast (primitive endoderm) origin of primate extraembryonic mesoderm [66–69]. Our previous studies of autosomal DNAme also demonstrated that endothelial and stromal cells have similar DNAme profiles to each other, and showed intermediate DNAme patterns relative to trophoblast and Hofbauer cells with regard to partially methylated domains and placental-specific imprinted regions [24]. Our X-chromosome DNAme results support an origin from hypoblast or very early epiblast, as the endothelial and stromal cell sex chromosome DNAme patterns are distinct from both Hofbauer and trophoblast cells, without being intermediate between the two.

Intriguingly, in the post-implantation blastocyst, the trophoblast, hypoblast, and epiblast are reported to show different rates of de novo DNAme and different global DNAme levels [50]. Specifically, hypoblast showed an overall later timing of de novo DNAme, and lower ultimate DNAme levels as compared to the trophectoderm and epiblast, although overall X-chromosome DNAme was similar in promoter and CGI from day 6 to day 8 in pre-implantation [50]. Similarly, in the term placenta we observed the lowest level of X-linked promoter DNAme in putative hypoblast-derived endothelial and stromal cells, and the highest level of DNAme in putative epiblast-derived Hofbauer cells, aligning with this evidence from early development, and supporting the conclusion that the endothelium and stroma may be hypoblast-derived in human placental villi.

To further understand the developmental origin of placental-derived endothelial cells, we compared publicly available DNAme data derived from larger placental and umbilical vessels: pAEs, pVEs and uVEs, with our data from pMEs. The pME data derive from are endothelial cells isolated by FACS from microvessels within the terminal chorionic villus, while the pAE and pVE were collected in the placental chorionic plate, extending into the primary and intermediate chorionic villi [70]. Placental microvascular endothelial cells (pME) are derived from extraembryonic mesoderm (proposed hypoblast origin) from 18–20 days post-conception and before the connection to a fetal umbilical cord [63, 71]. In contrast, the umbilical venous endothelial cells (uVE) are macrovascular cells extracted from umbilical cord [6], which extends from the fetal heart vessels and derives from embryonic mesoderm (epiblast origin) [70, 72, 73]. Although these pAE, pME, pVE, and uVE vessels are physically connected, there is likely a transition zone with a gradient of cells from extraembryonic to embryonic origins within the placental macro-vessels (Supplementary Fig. 7). As we expected, uVEs had X-chromosome DNAme patterns consistent with their embryonic origin, and strikingly different from the low DNAme seen in pMEs, while the pAEs/pVEs showed an intermediate pattern likely reflecting the presence of mixed cells of embryonic and extraembryonic origins in pAE [74]. We cannot definitively exclude that the differences in the experimental design in addition to the cell compositions of the various datasets might have contributed to the DNAme differences observed in these analyses. Our proposed developmental scheme based on this data is shown in Fig. 6.

As sex differences in autosomal DNAme have been reproducibly observed in whole chorionic villi, we investigated how it differs by placental cell types. Some sex-DMCs were identified in endothelial cells and some of which were shared, while others differed by cell type. For example, ZNF300, showed similar sex differences across cell types except for the Hofbauer cells, as also reported by Andrews et al. (2022) using the same placental cell data. Whereas NSD1 and LDB3 had limited DNAme sex differences only in specific cell types.

We acknowledge that our results have limitations. First, our study utilized data from previously published placental cell types isolated using FACS, which may not fully represent all cells in the placenta. Second, our sample size per cell type was small, increasing variability and limiting our ability to detect subtle DNAme differences. The Illumina microarrays are also underrepresented for probes on the X and Y chromosome. This made it difficult to fully evaluate how DNAme may differ by cell type on the Y chromosome especially, although the cell-type differences observed are intriguing and support what was observed in other genomic regions. Further, we do not have matched gene expression data from these cells to determine the relationship between DNAme and X-linked gene expression, in particular whether there is more escape from XCI in endothelial/stromal cells as compared to other cell types and somatic tissues. Finally, with our current data, we cannot distinguish DNAme arising from the active and inactive X-chromosomes of XX cells and results represent an average of these distinct DNAme environments.

This study provides a comprehensive characterization of sex and cell-influenced DNAme in the placenta. Our results suggest that X-linked DNAme may reflect the cellular origins of major cell type compartments of the mature placenta. Finally, our analyses add evidence that mesenchymal cells, including endothelial and stromal cells, may originate from hypoblast derived eXM in the human placenta.

Data availability

No novel datasets were generated during the current study. The datasets used include: GSE159526; GSE106099; and GSE144804.

Ethics approval and consent to participate

Ethics approval for this study was obtained from the University of British Columbia/Children’s and Women’s Health Centre of British Columbia Research Ethics Board (H18-01695). This study used publicly available data and no additional consents were needed.

Consent for publication

Not applicable.

Availability of data and materials

Cell DNAme data and the supporting sample-specific information are available on the GEO dataset accession number (GSE159526). Placental and umbilical cord endothelial cell DNAme data and the supporting sample-specific information are available on the GEO dataset accession number (GSE106099) and (GSE144804), respectively.

Funding

This work was supported by Canadian Institutes of Health Research (CIHR) Grants to WPR (SVB-158613, GSK-171375). WPR receives an investigatorship award for salary support from the BC Children’s Hospital Research Institute; AMI received support from a CIHR Banting & Best Doctoral Fellowship.

Acknowledgements

We thank the scientific community for sharing publicly available scientific data for research purposes. We also acknowledge the Robinson lab members for helpful discussion and feedback on the data analysis and manuscript.

Author information

Authors and Affiliations

BC Children’s Hospital Research Institute, 950 W 28th Ave, Vancouver, V6H 3N1, Canada

Jiyoung Han, Amy M. Inkster, Victor Yuan, Maria S. Peñaherrera, Wendy P. Robinson

Department of Medical Genetics, University of British Columbia, 4500 Oak St, Vancouver, V6H 3N1, Canada

Jiyoung Han, Amy M. Inkster, Victor Yuan, Maria S. Peñaherrera, Wendy P. Robinson

Contributions

JH, AMI, and WPR contributed to project design, data interpretation, and writing manuscript. JH and AMI contributed to the data analysis. JH and VY contributed to data preparation and checkup. All authors read and provided critical feedback on the manuscript and approved the final version.

Competing interests

The authors declare that they have no competing interests.

Corresponding authors

Correspondence to Wendy P. Robinson.

Gabory A, Roseboom TJ, Moore T, Moore LG, Junien C. Placental contribution to the origins of sexual dimorphism in health and diseases: sex chromosomes and epigenetics. Biol Sex Differ. 2013;4(1):5.
Rosenfeld CS. Sex-Specific Placental Responses in Fetal Development. Endocrinology. 2015;156(10):3422–34.
Inkster AM, Fernández-Boyano I, Robinson WP. Sex Differences Are Here to Stay: Relevance to Prenatal Care. J Clin Med. 2021 July;5(13):3000.
Gonzalez TL, Sun T, Koeppel AF, Lee B, Wang ET, Farber CR, et al. Sex differences in the late first trimester human placenta transcriptome. Biol Sex Differ. 2018;9(1):4.
Olney KC, Plaisier SB, Phung TN, Silasi M, Perley L, O’Bryan J, et al. Sex differences in early and term placenta are conserved in adult tissues. Biol Sex Differ. 2022;13(1):74.
Cvitic S, Longtine MS, Hackl H, Wagner K, Nelson MD, Desoye G et al. The Human Placental Sexome Differs between Trophoblast Epithelium and Villous Vessel Endothelium. Colombo GI, editor. PLoS ONE. 2013;8(10):e79233.
Meakin AS, Clifton VL, Review. Understanding the role of androgens and placental AR variants: Insight into steroid-dependent fetal-placental growth and development. Placenta. 2019 Sept;84:63–8.
Meakin AS, Cuffe JSM, Darby JRT, Morrison JL, Clifton VL. Let’s Talk about Placental Sex, Baby: Understanding Mechanisms That Drive Female- and Male-Specific Fetal Growth and Developmental Outcomes. Int J Mol Sci 2021 June 15;22(12):6386.
Lyon MF. Gene Action in the X-chromosome of the Mouse (Mus musculus L). Nature. 1961;190(4773):372–3.
Ishikawa H, Rattigan Á, Fundele R, Burgoyne PS. Effects of Sex Chromosome Dosage on Placental Size in Mice1. Biol Reprod. 2003;69(2):483–8.
Balaton BP, Cotton AM, Brown CJ. Derivation of consensus inactivation status for X-linked genes from genome-wide studies. Biol Sex Differ. 2015;6(1):35.
Ehrlich M, Gama-Sosa MA, Huang LH, Midgett RM, Kuo KC, McCune RA, et al. Amount and distribution of 5-methylcytosine in human DNA from different types of tissues or cells. Nucl Acids Res. 1982;10(8):2709–21.
Blewitt ME, Gendrel AV, Pang Z, Sparrow DB, Whitelaw N, Craig JM, et al. SmcHD1, containing a structural-maintenance-of-chromosomes hinge domain, has a critical role in X inactivation. Nat Genet. 2008;40(5):663–9.
Cotton AM, Lam L, Affleck JG, Wilson IM, Peñaherrera MS, McFadden DE, et al. Chromosome-wide DNA methylation analysis predicts human tissue-specific X inactivation. Hum Genet. 2011;130(2):187–201.
Sharp AJ, Stathaki E, Migliavacca E, Brahmachary M, Montgomery SB, Dupre Y, et al. DNA methylation profiles of human active and inactive X chromosomes. Genome Res. 2011;21(10):1592–600.
Cotton AM, Avila L, Penaherrera MS, Affleck JG, Robinson WP, Brown CJ. Inactive X chromosome-specific reduction in placental DNA methylation. Hum Mol Genet. 2009;18(19):3544–52.
GTEx Consortium, Tukiainen T, Villani AC, Yen A, Rivas MA, Marshall JL, et al. Landscape of X chromosome inactivation across human tissues. Nature. 2017;550(7675):244–8.
Inkster AM, Matthews AM, Phung TN, Plaisier SB, Wilson MA, Brown CJ et al. Breaking Rules: the complex relationship between DNA methylation and X- chromosome inactivation in the human placenta. Biology Sex Differences. 2025.
Boyd JD, Hamilton WJ. The human placenta. Cambridge: Heffer; 1970. p. 365.
Turco MY, Moffett A. Development of the human placenta. Development. 2019;146(22):dev163428.
James JL, Carter AM, Chamley LW. Human placentation from nidation to 5 weeks of gestation. Part I: What do we know about formative placental development following implantation? Placenta. 2012;33(5):327–34.
De Miguel MP, Arnalich Montiel F, Lopez Iglesias P, Blazquez Martinez A, Nistal M. Epiblast-derived stem cells in embryonic and adult tissues. Int J Dev Biol. 2009;53(8–9–10):1529–40.
Thomas JR, Naidu P, Appios A, McGovern N. The Ontogeny and Function of Placental Macrophages. Front Immunol. 2021;12:771054.
Yuan V, Hui D, Yin Y, Peñaherrera MS, Beristain AG, Robinson WP. Cell-specific characterization of the placental methylome. BMC Genomics. 2021;22(1):6.
Dean W, Santos F, Reik W. Epigenetic reprogramming in early mammalian development and following somatic nuclear transfer. Semin Cell Dev Biol. 2003;14(1):93–100.
Morgan HD, Santos F, Green K, Dean W, Reik W. Epigenetic reprogramming in mammals. Hum Mol Genet. 2005;14(suppl1):R47–58.
De Moreira JC, Fernandes GR, Vibranovski MD, Pereira LV. Early X chromosome inactivation during human preimplantation development revealed by single-cell RNA-sequencing. Sci Rep 2017 Sept 7;7(1):10794.
Inkster AM, Yuan V, Konwar C, Matthews AM, Brown CJ, Robinson WP. A cross-cohort analysis of autosomal DNA methylation sex differences in the term placenta. Biol Sex Differ. 2021;12(1):38.
Andrews SV, Yang IJ, Froehlich K, Oskotsky T, Sirota M. Large-scale placenta DNA methylation integrated analysis reveals fetal sex-specific differentially methylated CpG sites and regions. Sci Rep. 2022 June 7;12(1):9396.
Bulka CM, Everson TM, Burt AA, Marsit CJ, Karagas MR, Boyle KE, et al. Sex-based differences in placental DNA methylation profiles related to gestational age: an NIH ECHO meta-analysis. Epigenetics. 2023;18(1):2179726.
Khan A, Inkster AM, Peñaherrera MS, King S, Kildea S, Oberlander TF, et al. The application of epiphenotyping approaches to DNA methylation array studies of the human placenta. Epigenetics Chromatin. 2023;16(1):37.
Yuan V, Price EM, Del Gobbo G, Mostafavi S, Cox B, Binder AM, et al. Accurate ethnicity prediction from placental DNA methylation data. Epigenetics Chromatin. 2019;12(1):51.
Inkster AM, Wong MT, Matthews AM, Brown CJ, Robinson WP. Who’s afraid of the X? Incorporating the X and Y chromosomes into the analysis of DNA methylation array data. Epigenetics Chromatin. 2023;16(1):1.
Zhou W, Laird PW, Shen H. Comprehensive characterization, annotation and innovative use of Infinium DNA methylation BeadChip probes. Nucleic Acids Res. 2016;gkw967.
Triche TJ, Weisenberger DJ, Van Den Berg D, Laird PW, Siegmund KD. Low-level processing of Illumina Infinium DNA Methylation BeadArrays. Nucleic Acids Res. 2013;41(7):e90–90.
Teschendorff AE, Marabita F, Lechner M, Bartlett T, Tegner J, Gomez-Cabrero D, et al. A beta-mixture quantile normalization method for correcting probe design bias in Illumina Infinium 450 k DNA methylation data. Bioinformatics. 2013;29(2):189–96.
Hansen KD. IlluminaHumanMethylation450kanno.ilmn12.hg19 [Internet]. Bioconductor; 2017 [cited 2025 May 25]. Available from: https://bioconductor.org/packages/IlluminaHumanMethylation450kanno.ilmn12.hg19
Kasper Daniel Hansen [Cre A. IlluminaHumanMethylationEPICanno.ilm10b4.hg19 [Internet]. Bioconductor. 2017 [cited 2025 May 25]. Available from: https://bioconductor.org/packages/IlluminaHumanMethylationEPICanno.ilm10b4.hg19
Bizet M, Defrance M, Calonne E, Bontempi G, Sotiriou C, Fuks F, et al. Improving Infinium MethylationEPIC data processing: re-annotation of enhancers and long noncoding RNA genes and benchmarking of normalization methods. Epigenetics. 2022;17(13):2434–54.
Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7):e47–47.
Phipson B, Maksimovic J, Oshlack A. missMethyl: an R package for analyzing data from Illumina’s HumanMethylation450 platform. Bioinformatics. 2016;32(2):286–8.
Chen Y, an, Lemire M, Choufani S, Butcher DT, Grafodatskaya D, Zanke BW, et al. Discovery of cross-reactive probes and polymorphic CpGs in the Illumina Infinium HumanMethylation450 microarray. Epigenetics. 2013;8(2):203–9.
Joo JE, Novakovic B, Cruickshank M, Doyle LW, Craig JM, Saffery R. Human active X-specific DNA methylation events showing stability across time and tissues. Eur J Hum Genet. 2014;22(12):1376–81.
Balaton BP, Fornes O, Wasserman WW, Brown CJ. Cross-species examination of X-chromosome inactivation highlights domains of escape from silencing. Epigenetics Chromatin. 2021;14(1):12.
Cotton AM, Price EM, Jones MJ, Balaton BP, Kobor MS, Brown CJ. Landscape of DNA methylation on the X chromosome reflects CpG density, functional chromatin state and X-chromosome inactivation. Hum Mol Genet. 2015;24(6):1528–39.
Demir R, Kayisli UA, Seval Y, Celik-Ozenci C, Korgun ET, Demir-Weusten AY, et al. Sequential Expression of VEGF and its Receptors in Human Placental Villi During Very Early Pregnancy: Differences Between Placental Vasculogenesis and Angiogenesis. Placenta. 2004 July;25(6):560–72.
Braun PR, Han S, Hing B, Nagahama Y, Gaul LN, Heinzman JT, et al. Genome-wide DNA methylation comparison between live human brain and peripheral tissues within individuals. Transl Psychiatry. 2019;9(1):47.
Ladd-Acosta C, Andrews SV, Bakulski KM, Feinberg JI, Tryggvadottir R, Yao R et al. Placenta DNA methylation at ZNF300 is associated with fetal sex and placental morphology [Internet]. 2021 [cited 2024 June 11]. Available from: http://biorxiv.org/lookup/doi/10.1101/2021.03.05.433992
Deng X, Berletch JB, Nguyen DK, Disteche CM. X chromosome regulation: diverse patterns in development, tissues and disease. Nat Rev Genet. 2014 June;15(6):367–78.
Zhou F, Wang R, Yuan P, Ren Y, Mao Y, Li R, et al. Reconstituting the transcriptome and DNA methylome landscapes of human implantation. Nature. 2019;572(7771):660–4.
Phung TN, Olney KC, Pinto BJ, Silasi M, Perley L, O’Bryan J, et al. X chromosome inactivation in the human placenta is patchy and distinct from adult tissues. Hum Genet Genomics Adv. 2022 July;3(3):100121.
Kondoh H. The Epiblast and Pluripotent Stem Cell Lines. In: Molecular Basis of Developmental and Stem Cell Regulation [Internet]. Cham: Springer International Publishing; 2024 [cited 2024 Sept 30]. pp. 3–9. (Results and Problems in Cell Differentiation; vol. 72). Available from: https://link.springer.com/10.1007/978-3-031-39027-2_1
Fox H. The incidence and significance of hofbauer cells in the mature human placenta. J Pathol. 1967;93(2):710–7.
Kaufmann P, Stark J, Stegner HE. The villous stroma of the human placenta: I. The ultrastructure of fixed connective tissue cells. Cell Tissue Res [Internet]. 1977 Feb [cited 2024 Oct 1];177(1). Available from: http://link.springer.com/10.1007/BF00221122
Moskalewski S, Ptak W, Czarnik Z. Demonstration of Cells with IgG Receptor in Human Placenta. Neonatology. 1975;26(3–4):268–73.
Selkov SA, Selutin AV, Pavlova OM, Khromov-Borisov NN, Pavlov OV. Comparative phenotypic characterization of human cord blood monocytes and placental macrophages at term. Placenta. 2013 Sept;34(9):836–9.
Thomas JR, Appios A, Calderbank EF, Yoshida N, Zhao X, Hamilton RS, et al. Primitive haematopoiesis in the human placenta gives rise to macrophages with epigenetically silenced HLA-DR. Nat Commun. 2023;14(1):1764.
True H, Blanton M, Sureshchandra S, Messaoudi I. Monocytes and macrophages in pregnancy: The good, the bad, and the ugly*. Immunol Rev. 2022 July;308(1):77–92.
Singh R, Soman-Faulkner K, Sugumar K, Embryology. Hematopoiesis. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2024 [cited 2024 Oct 2]. Available from: http://www.ncbi.nlm.nih.gov/books/NBK544245/
Lock LF, Takagi N, Martin GR. Methylation of the Hprt gene on the inactive X occurs after chromosome inactivation. Cell. 1987;48(1):39–46.
Chow J, Heard E. X inactivation and the complexities of silencing a sex chromosome. Curr Opin Cell Biol. 2009 June;21(3):359–66.
Migeon BR, Axelman J, Jeppesen P. Differential X Reactivation in Human Placental Cells: Implications for Reversal of X Inactivation. Am J Hum Genet. 2005 Sept;77(3):355–64.
Knöfler M, Haider S, Saleh L, Pollheimer J, Gamage TKJB, James J. Human placenta and trophoblast development: key molecular mechanisms and model systems. Cell Mol Life Sci. 2019 Sept;76(18):3479–96.
Snell GD, Stevens LC, Green EL. Early embryology. Biology Lab mouse. 1966;2:205–45.
Panja S, Paria BC. Development of the Mouse Placenta. Adv Anat Embryol Cell Biol. 2021;234:205–21.
Enders AC, Schlafke S, Hendrickx AG. Differentiation of the embryonic disc, amnion, and yolk sac in the rhesus monkey. Am J Anat. 1986;177(2):161–85.
Enders AC, King BF. Formation and differentiation of extraembryonic mesoderm in the rhesus monkey. Am J Anat. 1988;181(4):327–40.
Bianchi DW, Wilkins-Haug LE, Enders AC, Hay ED. Origin of extraembryonic mesoderm in experimental animals: Relevance to chorionic mosaicism in humans. Am J Med Genet. 1993 June;15(5):542–50.
Farkas K, Ferretti E. Derivation of Human Extraembryonic Mesoderm-like Cells from Primitive Endoderm. IJMS. 2023 July 12;24(14):11366.
Lang I, Schweizer A, Hiden U, Ghaffari-Tabrizi N, Hagendorfer G, Bilban M, et al. Human fetal placental endothelial cells have a mature arterial and a juvenile venous phenotype with adipogenic and osteogenic differentiation potential. Differentiation. 2008;76(10):1031–43.
Charolidi N, Host AJ, Ashton S, Tryfonos Z, Leslie K, Thilaganathan B, et al. First trimester placental endothelial cells from pregnancies with abnormal uterine artery Doppler are more sensitive to apoptotic stimuli. Lab Invest. 2019;99(3):411–20.
Mitchell B, Sharma R. The cardiovascular system. In: Embryology [Internet]. Elsevier; 2009 [cited 2024 Nov 2]. pp. 31–40. Available from: https://linkinghub.elsevier.com/retrieve/pii/B9780702032257500099
Psaltis PJ, Harbuzariu A, Delacroix S, Holroyd EW, Simari RD. Resident vascular progenitor cells–diverse origins, phenotype, and function. J Cardiovasc Transl Res. 2011;4(2):161–76.
Casanello P, Schneider D, Herrera EA, Uauy R, Krause BJ. Endothelial heterogeneity in the umbilico-placental unit: DNA methylation as an innuendo of epigenetic diversity. Front Pharmacol. 2014;5:49.

No competing interests reported.

Download PDF

Reviews received at journal
05 Jan, 2026
Reviews received at journal
29 Dec, 2025
Reviewers agreed at journal
22 Dec, 2025
Reviewers agreed at journal
22 Dec, 2025
Reviewers agreed at journal
25 Nov, 2025
Reviewers invited by journal
23 Nov, 2025
Editor assigned by journal
18 Nov, 2025
Submission checks completed at journal
17 Nov, 2025
First submitted to journal
14 Nov, 2025

You are reading this latest preprint version

Sex-influenced DNA methylation differs by placental cell type

Status:

Version 1

Abstract

Figures

Highlights

Summary

Background

Materials & methods

Data processing

Public endothelial cell data processing

Annotation

Differentially Methylated CpGs (DMCs)

Nucleotide BLAST analysis for Y-chromosome cross-hybridization

Results

Discussion

Conclusions

Declarations

Acknowledgements

Author information

Contributions

Corresponding authors

References

Additional Declarations

Supplementary Files

Status:

Version 1