Placental chorionic plate DNA methylation patterns correlate with DNA methylation at SOCS3 in newborn human peripheral blood cells

doi:10.21203/rs.3.rs-4553640/v1

Download PDF

Research Article

Placental chorionic plate DNA methylation patterns correlate with DNA methylation at SOCS3 in newborn human peripheral blood cells

https://doi.org/10.21203/rs.3.rs-4553640/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Regulation of placental function is fundamental for fetal development. Various in utero environments, including pregnancy complications, interfere with changes in DNA methylation and influence placental functions and child development following birth. However, data on the association between genome-wide DNA methylation patterns in the placenta and changes in DNA methylation in the postnatal peripheral blood cells of the same individuals remain limited. Herein, we aimed to reveal the association between epigenetic changes in fetal appendages at birth and longitudinal epigenetic changes in the tissues of neonates.

Methods

Using a DNA methylation array, we examined the epigenetics of placental chorionic plates from 136 participants who were born between 22 and 42 weeks of gestation. We then examined DNA methylation levels of 62 pairs of umbilical cord blood and postnatal peripheral blood cells to investigate their association with the epigenetics of chorionic plates in identical newborns.

Results

Unsupervised classification of chorionic plates by the most variable DNA methylation levels between samples revealed contrasting methylation patterns in the genes involved in blood vessel formation. The epigenetic classification of the chorionic plate was significantly associated with intrauterine inflammation, neonatal respiratory diseases, and DNA methylation levels of cg18181703 within the suppressor of the cytokine signaling 3 (SOCS3) gene in neonatal peripheral blood cells. A significant association between DNA methylation levels at cg18181703 in cord blood and the classification were nullified when gestational age at birth was considered as a covariate. Meanwhile, longitudinal methylation levels at cg18181703 were confirmed in the peripheral blood cells of neonates of specific groups classified by chorionic plate epigenetics, independent of gestational age.

Conclusions

DNA methylation patterns in chorionic plates during intrauterine inflammation were associated with DNA methylation levels of cg18181703 in neonates. Methylation of cg1818170 is known to have a causal effect on child height. Our study suggests that changes in chorionic plate function with DNA methylation changes may program infant growth via the DNA methylation levels of cg18181703 in blood cells.

DNA methylation

chorionic plates

newborn

peripheral blood cells

The placenta supplies nutrients to the fetus, exchanges gases in the umbilical cord blood, receives waste products from the fetus, and separates the maternal body from the fetus. Placental and somatic tissues develop from the trophectoderm (TE) and the inner cell mass (ICM), respectively. During human development, the TE and ICM are formed in blastocysts 4–5 days post-fertilization. Blastocysts are implanted into the maternal endometrium 6–7 days post-fertilization [1]. The polar TE differentiates into primary syncytiotrophoblasts and proliferative cytotrophoblasts during blastocyst implantation [2]. Further proliferation and differentiation of cytotrophoblasts cover the villous core, which is of extraembryonic mesodermal origin, presumably from the ICM, resulting in a villous structure [1]. The fetal vasculature is formed in the villous core. Thus, the chorionic villi develop from the TE and extraembryonic mesoderm.

Compared with other tissues, mammalian placentas are hypomethylated [3]. Genome-wide DNA methylation in the placental chorionic villi is distinctly bimodal, with peaks in partially methylated (approximately 45% methylation) and highly methylated (approximately 80% methylation) regions [4, 5]. The transcript levels of genes, the bodies of which belong to highly methylated regions, are higher than those in partially methylated regions in the chorionic villi [5]. In embryonic stem cells derived from ICM, early developmental regulatory genes are unmethylated in large genomic regions [6], while other genomic regions are highly methylated [7]. Therefore, placental gene expression is unique and linked to DNA methylation with tissue-specific regulation.

Various pregnancy-associated complications or the perinatal environment are associated with fetal and neonatal mortality and morbidity as well as adverse health outcomes in offspring later in life. These factors induce changes in placental DNA methylation, too. Diabetes during pregnancy alters DNA methylation, with concomitant protein expression levels of some genes on the fetal side of the placenta [8]. Preeclampsia alters DNA methylation in the chorionic villi [9, 10] and fetal [11, 12] or maternal [13] side of the placenta. Differentially methylated loci in chorionic villi from early onset preeclampsia negatively correlate with gene expression [9]. In cases of acute chorioamnionitis (CAM), the chorionic villi in the placenta (but not the chorion or amnion) show significantly altered DNA methylation levels [14]. In addition to pregnancy complications, maternal pre-pregnancy body mass index (BMI) is associated with placental DNA methylation status [15]. Furthermore, maternal smoking during pregnancy is associated with DNA methylation on the fetal side of the placenta, which has been linked to birth outcomes and gene expression [16]. These studies suggest that perinatal factors may affect the health of newborns later in life by altering placental function coordinated by DNA methylation.

The prenatal in utero environment is also longitudinally associated with DNA methylation in the blood [17, 18]. Therefore, we determined whether aberrant epigenetic regulation in the placenta influenced neonatal health and DNA methylation after birth. In addition, DNA methylation analyses of the chorionic villi have been reported in previous studies [4, 5, 9, 10, 14]; however, those on chorionic plates are limited. In this study, we clustered individuals based on their DNA methylation patterns in the placental chorionic plate using unsupervised clustering. We chose the chorionic plate as it is closest to the fetus in the placenta. Additionally, the chorionic plate develops from the extraembryonic mesenchyme and mesoderm [19], suggesting that its epigenetic regulatory mechanisms will be more similar to those of somatic tissues than those of the chorionic villi. Our results show that genome-wide DNA methylation levels of chorionic plates are intermediate between those of chorionic villi and cord blood cells. Subsequently, we analyzed the DNA methylation in chorionic plates, cord blood cells, and postnatal peripheral blood (PB) cells a few months after birth to determine the relationship between the in utero environment and neonatal characteristics; additionally, we assessed neonatal morbidity in 62 individuals from the same person. The findings of the present study could reveal an association between perinatal environments and longitudinal epigenetic changes in neonates.

Study population

Between October 2014 and July 2016, we recruited 147 mother-infant pairs from Tokyo, Japan, for a prospective cohort study. Informed consent was obtained from all the 144 participants. Three refused to participate. The participants had live births between 22 and 42 weeks of gestation. Umbilical cord blood samples and placental chorionic plates were obtained at delivery. Samples from participants shown to have congenital diseases (n = 4) and a sample obtained from an egg-donor offspring (n = 1) were excluded from further analysis. Postnatal PB samples were obtained from 70 participants via venipuncture at least 2 weeks following birth, near the due date (36–44 weeks postmenstrual age).

Chorionic plate separation and DNA extraction

After separating the amnion from the placenta, chorionic plate tissue samples (one cm square) were obtained from five different locations. The blood vessels were trimmed away. Subsequently, pieces of chorionic plate were prepared by removing the chorionic villi using a razor under the microscope and washing with 1× phosphate buffered saline; the samples were stored at − 80°C until genomic DNA extraction. The chorionic plate was lysed overnight in lysis buffer containing 100 mM Tris-HCl, 5 mM EDTA, 200 mM NaCl, 0.2% SDS, and 200 ng/mL proteinase K at 55°C. Genomic DNA was extracted by adding phenol-chloroform-isoamyl alcohol (24:25:1, pH 7.9) (Nacalai Tesque) to the lysed solution and stored at − 80°C for the microarray experiments. Data from three individuals were collected to compare the DNA methylation status between intra-individual chorionic villi and chorionic plates.

Mononuclear cell separation

From the blood samples, cord blood mononuclear cells or peripheral blood mononuclear cells were collected as previously described [20]. Briefly, blood cells were separated via gradient centrifugation using Ficoll-Hypaque within 24 h of sample collection.

DNA methylation microarray analysis and data processing

The extracted DNA was bisulfite-converted using an EpiTect Plus DNA bisulfite kit (Qiagen, Hilden, Germany). DNA methylation from the chorionic plate was analyzed using an Infinium Methylation EPIC BeadChip array (Illumina, San Diego, CA, USA). DNA methylation in the mononuclear cells of the blood was analyzed using the Infinium Methylation 450k BeadChip array (Illumina) as previously described [20] and the data was deposited in the Gene Expression Omnibus database under the accession number GSE110829. Methylation data were acquired using the iScan system (Illumina) as IDAT files and processed using the Minfi and ChAMP packages (https://bioconductor.org/biocLite.R) in R-3.6.0. The background was corrected using the NOOB method in the Minfi package (v.1.32.0). NOOB-corrected data were normalized using BMIQ in the ChAMP package (v.2.16.2). The manifest file was annotated using “IlluminaHumanMethylationEPICanno.ilm10b4.hg19”. Probes with detection p-values > 0.01 in at least one sample, bead count < 3 in at least 5% samples, and non-CpG targeting probes were removed. Single-nucleotide polymorphism-related probes, multi-hit probes defined by Nordlund et al. [21], and probes located on either the X or Y chromosome were also filtered. This yielded 754,101 autosomal probes from 136 chorionic plate assays. The beta value (which is the ratio of the methylated probe intensity to the overall intensity, sum of the methylated and unmethylated probe intensities) was calculated for each probe.

Cell fraction estimation

Reference-based deconvolution of whole-blood cells was performed using the R package FlowSorted.CordBlood.450k (v.1.24.0) according to Andrews and Bakulski [22], and FlowSorted.Blood.450k (v.1.14.0) according to Jaffe AE [23]. The placental cell fraction was estimated using the R package planet (v.1.1.0), as proposed by Yuan et al. [24].

Gene Ontology analysis

The annotation of differentially methylated 712 CpG sites between groups A-E was performed using Genomic Regions Enrichment of Annotations Tool Version 4.0.4 [25], with the association rule setting of “single nearest gene” within 1,000 kb.

Preparing placental tissue blocks

After weighing and examining gross morphology, whole placentas were fixed using 20% formaldehyde. Five paraffin blocks were systematically obtained from each placenta for pathological examination by systematic random sampling. Chorionic plates were included in three out of five blocks collected at the umbilical cord attachment area, placental limbus, and placenta centrale, respectively. Briefly, 5-mm-wide linear parallel slices of placental tissue were cut at an interval of approximately 3 cm perpendicular to the greatest dimension of the placental axis. All linear slices were vertically cut into small pieces at an interval of 3 cm. Each block was made vertically from the maternal side to the fetal side. Each block was cut into 3-µm-thick sections and subjected to HE staining.

Statistical analysis

Unsupervised clustering was performed based on the 1,000 most variable beta values between 136 chorionic plates. Hierarchical clustering was performed using the Euclidean linkage method with Ward.D distance of the R package gplots. The DNA methylation beta values of blood cells were transformed into M values. Linear mixed model regression was performed using M values adjusted for covariates to compare associations between placental groups and unsupervised clustering classified by beta values. The risk ratio was calculated using the R package epitab (v. 0.5.10.1) with a Wald confidence interval and Fisher’s exact test. A p value < 0.05 was considered significant. The differences in the genome-wide DNA methylation beta values between paired chorionic plate and chorionic villi of three individuals were analyzed using the Wilcoxon signed rank test. Welch’s t-test was performed to identify the differences in genome-wide DNA methylation beta values between chorionic plates and cord blood cells.

Association of prenatal conditions with DNA methylation patterns in the chorionic plates

We examined the histology of each placenta to determine the clinical features of the five groups of 136 chorionic plates classified according to their DNA methylation patterns. Pathological diagnosis revealed that 15 of the 16 samples (94%) from groups A and B had CAM. The incidence of CAM in groups C–E was 6/15 (40%), 1/13 (8%), and 7/92 (8%), respectively. Additionally, CAM in groups A and B was classified at a high frequency as stage III using Blanc’s classification, which classifies CAM by the location of neutrophil infiltration in the chorion and amnion [26]. Of the CAMs in the 136 chorionic plates, 7/9 (78%), 4/6 (67%), 1/6 (17%), 0/1 (0%), and 0/7 (0%) were classified as stage III in groups A–E, respectively (Fig. 1A). These results suggested that the classification of chorionic plates based on DNA methylation patterns was related to in utero inflammation.

Maternal neutrophil infiltration into the chorionic membrane is a well-known pathological condition in stage III CAM [27-29]. To determine whether the DNA methylation patterns of the 712 probes in 136 chorionic plates were influenced by the infiltration of inflammatory cells into the chorionic plates, we predicted cell composition of each placenta sample from the DNA methylation data. Deconvolution of placental cells by the “Planet” [24] (which allows estimation of cell composition directly from placental DNA methylation data) indicated the enrichment of nucleated red blood cells (nRBCs) especially in group A (Fig. 1A). Hematoxylin-eosin (HE)-staining of the pathological sections of the chorionic plates showed cell infiltration into chorionic plates with CAM in group A, shown as in “CP_1” in Fig. 1B. However, the staining pattern of the infiltrated cells differed from that of the nRBCs in the umbilical vein, the cytoplasmic staining of which was brighter than those of other cells (Fig. 1B), suggesting that the marker probes for “nRBCs” of “Planet” cross-reacted with other cells in chorionic plates. Therefore, we determined whether DNA methylation marker probes for estimating blood cell types could indicate the infiltration of blood cell types into the chorionic plate. The cells were first estimated using the deconvolution method for adult PB cells using “FlowSorted.Blood.450K” [23] owing to the suggested infiltration of maternal neutrophils. The results showed enrichment of granulocytes (Gran) in group A compared with that in group E; a certain percentage of “Gran” was detected in group E. Most placentas in group E were not diagnosed with CAM. In fact, infiltrated cells were absent in the chorionic membranes of the samples in group E, and these were not diagnosed as CAM, as confirmed using HE staining (Fig. S2). Bakulski et al. indicated that “FlowSorted.Blood.450K” overestimated the granulocytes [22]. Meanwhile, “FlowSorted.CordBlood.450K” showed “Gran” in limited chorionic plate samples, especially in group A (Fig. 1A). The chorionic plate (CP_2) in group A was diagnosed with villitis but not CAM. HE-staining of CP_2 did not show any infiltration of neutrophils into the chorionic plate (Fig. 1B). Consistent with this result, estimation of cell types of CP_2 using “FlowSorted.CordBlood.450K” resulted in few “Gran” components. These results suggested that neutrophil infiltration in the chorionic plate can be estimated by applying the DNA methylation pattern of the chorionic plate to the “FlowSorted.CordBlood.450K” algorithm. Additionally, the DNA methylation pattern at the 712 probes in the chorionic plates was found not to be associated with the “Gran” fraction as confirmed in the samples of groups A and B based on the “Gran” fraction estimated by “FlowSorted.CordBlood.450K”.

Genes neighboring the 712 probes were significantly enriched in the “blood vessel” function

To determine the differences in the functions of the chorionic plate between groups, the 712 probes were annotated to genes using GREAT (version 4.0.4) [25], with the rule of a single nearest gene and maximum 1 Mb extension. Of the 712 probes, 707 were annotated with 590 genes, while 5 were located in genomic regions where genes were not present in the neighborhood of their probes (Table S1). Gene ontology (GO) analysis of the 590 genes revealed 20 enriched ontology terms in GO biological processes at a significant level. The top 10 GO terms are listed in Table 1. Six of the 10 terms related to blood vessels were hit by nearly identical genes. Among the 712 probes, only four were annotated to genes involved in GO terms associated with neutrophil function, which were not significantly enriched. These results indicated that the contrasting DNA methylation patterns of the 712 probes between groups A and E were associated with functional changes in the chorionic plates (especially angiogenesis) but not with cell contamination.

Association between DNA methylation patterns of chorionic plates and postnatal newborn outcome

To determine the relationship between DNA methylation patterns in chorionic plates and neonatal features (including their epigenetics), we examined postnatal PB cells from identical newborns at least 2 weeks following birth, around their due date [20]. We filtered out mother-newborn pairs, chorionic plates, cord blood cells, and PB cells of whom were collected and neonatal medical records of whom were accessible. In total, 62 pairs were included in the analyses (Fig. S3A, S3B). In particular, 6, 4, 8, 7, and 37 individuals remained in groups A, B, C, D, and E, respectively (Fig. 1A). Finally, all neonates in groups A and B were born to mothers diagnosed with stage III CAM, according to Blanc’s classification. Furthermore, 2 and 3 of the 37 neonates in group E were born to mothers diagnosed with stage I and II CAM, respectively. We initially investigated whether groups classified based on DNA methylation patterns in the chorionic plate were associated with neonatal morbidity. CAM is associated with the development of BPD. A meta-analysis showed that CAM exposure was significantly associated with BPD28 and BPD36, defined as supplemental oxygen requirements on postnatal day 28 and at the postmenstrual age of 36 weeks, respectively, but not with RDS [30]. Therefore, we examined the risk ratio of these three symptoms in groups A–D compared with that of group E. Those in every CAM category were also examined and compared with those without CAM. Newborns in group A showed the highest incidence of all three symptoms (Fig. 2). The risk ratios of group A for BPD28, BPD36, and RDS were significantly higher than those for group E (P = 0.00049, 0.0025, and 0.0064, respectively). The risk ratios for BPD28 and BPD36 (but not for RDS) of group B were significantly higher than those of group E (P = 0.0049, 0.0028, and 0.29, respectively). The confidence intervals for the risk ratios of group A for these three symptoms were all higher than those of stage III CAM. Notably, the newborn diagnosed with stage III CAM but whose DNA methylation pattern in chorionic plates was classified in group C developed none of these three symptoms (Fig. S3A). The results suggested that abnormal DNA methylation pattern of the chorionic plate was one of the factors involved in the development of pulmonary disease in newborns.

Association between DNA methylation of chorionic plates and postnatal DNA methylation in blood cells

Subsequently, we compared the DNA methylation of postnatal PB cells [20] between the groups to investigate the association between abnormal chorionic plate DNA methylation and neonatal epigenetic features in blood cells. Regression analysis adjusted for covariates (sex, estimated cell fraction, and batch) showed that cg18181703 and cg24200501 were significantly associated with group A compared with group E, with Bonferroni-adjusted P-values of 0.0047 and 0.0155, respectively (Fig. S3A). These two CpG sites were the only two significant CpG sites with a false discovery rate (FDR) q < 0.05. When analyzing the association of groups A and B with PB DNA methylation, DNA methylation at cg18181703 was the only significant CpG with a Bonferroni-adjusted P-value < 0.05 and q < 0.05. Differences in the mean beta value of cg18181703 in PB were -0.09 and -0.08, in groups A and B, respectively, versus that of group E (Fig. 3A). cg18181703 was designed to detect methylation at the CpG of chr17:76,354,621 (hg19) in exon 2, with two exons of the suppressor of cytokine signaling 3 (SOCS3) gene located 1,127 bp, 1,537 bp, and 2,385 bp downstream of the transcription start site of the SOCS3 mRNA (NM_001378932.1, NM_003955.5, and NM_001378933.1, respectively). We investigated whether DNA methylation at cg18181703 in PB was associated with CAM since all individuals in groups A and B were diagnosed with CAM. Regression analysis showed that DNA methylation at cg18181703 in PB was insignificantly associated with CAM (Bonferroni-adjusted P-value: 1), whereas the P-value of cg18181703 was the second-lowest (P = 2.78 e-06, q = 0.572). Only one CpG (cg15959270) was significantly associated with CAM in PB, with a Bonferroni-adjusted P = 0.035 and q = 0.035 (Table S2). These results suggested that the change in DNA methylation of cg18181703 in neonatal PB was associated with DNA methylation changes in chorionic plates rather than CAM, consistent with the finding that the risk ratio for BPD28 and BPD36 was higher in groups A and B than in CAM.

BPD28 was significantly associated with DNA methylation in PB at two CpG sites (cg11376147 and cg20374917) but not at cg18181703 (Table S2). Neither RDS nor BPD36 were associated with CpGs in the PB. The DNA methylation changes in cg18181703 in neonatal PB cells may be related to the function of the chorionic plate.

Longitudinal association between DNA methylation of chorionic plates and that in blood cells

We examined the DNA methylation patterns of the cord blood cells of the 62 pairs to determine whether DNA methylation changes in cg18181703 in PB cells had already occurred at birth. The cell type fractions of groups A and B contained significantly more “Gran” populations than the other groups (Fig. S3B). In contrast, postnatal PB cells showed similar distribution between the samples (Fig. S3A and S3C). Regression analysis with adjusted covariates (gestational age at birth, estimated cell fraction, batch difference, and sex of the newborn) did not show any significant difference in DNA methylation among the 408,789 probes examined with Bonferroni-adjusted P-value or q < 0.05. When gestational age was removed from the covariates in the analysis, 19 and 873 CpGs were significantly associated with group A compared with that in group E, with a Bonferroni-adjusted P-value and q < 0.05, respectively. Those of cg18181703 were 1 and 0.029, respectively. A simple t-test comparison of the DNA methylation levels of cg18181703 in cord blood cells without covariates or multiple testing adjustments showed significant differences (P < 0.001; Fig. 3B). Gestational age at birth in group A was significantly lower (mean 24.4 weeks) than that in group E (mean 32.2 weeks) (P-value = 3.4e-07). The DNA methylation levels at cg18181703 were significantly associated with gestational age at birth (Pearson’s correlation coefficient = 0.659, P-value = 5.8e-09) (Fig. 3C). Therefore, the analysis adjusted for gestational age as a covariate nullified the significance of the difference in DNA methylation in cg18181703 in the cord blood between groups A and E. The ratio of DNA methylation beta value divided by postmenstrual age did not significantly change between groups in cord blood; however, a significantly lower ratio was observed in the PB of groups A and B than that in group E (Fig. 3D). Therefore, we could not distinguish the effects of DNA methylation patterns in chorionic plates from those of gestational age at birth on the DNA methylation of cg18181703 in newborn cord blood cells. These results confirmed the longitudinal association between DNA methylation at cg18181703 in the blood cells and the identified groups. At the same time, intra-individual changes in DNA methylation at cg18181703 during the postnatal period were detected in some individuals (Fig. S3B). These results showed that DNA methylation at cg18181703 is plastic; however, longitudinal changes can be detected based on the DNA methylation pattern in chorionic plates, at least for a few months following birth.

In this study, we aimed to reveal the association between epigenetic changes in fetal appendages at birth and longitudinal epigenetic changes in the tissues of neonates. We found that the chorionic plates showed contrasting DNA methylation patterns at several CpG sites. Hierarchical clustering of 136 placental chorionic plates according to the most variable DNA methylation levels resulted in a minor characteristic group, the DNA methylation pattern of which contrasted that of a major group. Most individuals in this group were diagnosed with stage II or III histological CAM according to Blanc’s classification, which classifies CAM based on the location of neutrophil infiltration into the chorion and amnion. One individual in the minor group was not diagnosed with histological CAM; however, villitis was confirmed pathologically. These results indicated that contrasting DNA methylation at chorionic plates was associated with “intrauterine inflammation or infection or both” (Triple I) [31]. Konwar et al. reported that no CpG sites were differentially methylated in the amnion or chorion in cases of acute CAM [14] using the same array as used in this study. The authors reported that 66 CpG sites were differentially methylated in the chorionic villi of CAM with FDR <0.15 and mean group difference in DNA methylation >0.05 [14]. The number of differentially methylated CpG sites in the chorionic villi in the study by Konwar [14] was considerably lower than in the chorionic plates in this study. This suggests that functional changes in the chorionic plate reflect the effects of in utero infections more than those in the chorionic villi. Intrauterine inflammation may be related to “blood vessel morphogenesis” in the chorionic plates, as indicated by GO analysis of the 712 CpG sites that showed contrasting DNA methylation patterns between the minor and major groups. The density of perforating chorionic vessels in preterm placentas highly correlated with the incidence and severity of neonatal BPD [32]. “Blood vessel morphogenesis” in the chorionic plate may be involved in the development of BPD, since we confirmed that the risk ratio of BPD was higher in the minor groups (Fig. 2).

Further DNA methylation analysis of postnatal PB cells revealed that the methylation levels of cg18181703 were associated with the DNA methylation pattern of the chorionic plates. Hypomethylation of cg18181703 in PB was restricted to minor groups A and B (Fig. 2). Methylation at cg18181703 correlated with child height in different socioeconomic cohorts and exerted causal effects on height [33]. Extrauterine growth restriction (particularly in terms of height during hospitalization) may be associated with cg18181703 hypomethylation, which may originate from chorionic plate function. Additionally, cg18181703 hypomethylation in blood cells is reported in other studies investigating the effects of BMI [34-37], obesity [38,39], type 2 diabetes [36,40], and smoking [41-43] in adults. Obesity and type 2 diabetes mellitus are associated with systemic inflammation. Smoking causes lung inflammation; therefore, cg18181703 hypomethylation is a potential biomarker of inflammation. The term “fetal inflammatory response syndrome” (FIRS) was coined to describe the presence of systemic inflammation in the fetus owing to intra-amniotic infection [44]. FIRS also occurs upon exposure to non-infection-related stimuli (such as danger signals or alarmins). An elevated concentration of fetal plasma interleukin-6 (IL-6) defines FIRS. FIRS is believed to exacerbate the effects of immaturity and neonatal inflammation, thereby increasing the risk of neonatal complications and long-term morbidity [45]. IL-6 activates Janus tyrosine kinases (JAKs) to induce several downstream signaling cascades, including the activation of signal transduction and activator of transcription 3 (STAT3)-dependent gene expression via the phosphorylation of tyrosine 705 of STAT3 [46]. STAT3-mediated expression of SOCS3 results in a potent negative regulation. SOCS3 binds to the IL-6 receptor/JAK complex and blocks further STAT3 phosphorylation [47]. Therefore, cg18181703 methylation in SOCS3 can be regulated by IL-6 signaling. However, high levels of IL-6 in cord blood did not necessarily correlate with DNA hypomethylation of cg18181703 in the cord or postnatal blood from a limited number of individuals, sera of whom was collected [48] (Fig.S4). Single-cell analysis showed that SOCS3 was expressed in B cells, monocytes, natural killer cells, and T cells. In particular, SOCS3 expression was highest in monocytes [49]. Groups A and B showed a specific cell distribution in the cord blood but not in postnatal PB, as estimated from the DNA methylation data (Fig. S2C). However, cg18181703 hypomethylation was frequently only observed in groups A and B in cord and postnatal blood. Further analysis is required to determine whether the regulation of DNA methylation at cg18181703 in the perinatal blood is cell-specific.

We detected neutrophil infiltration in chorionic plates using genome-wide DNA methylation analysis of chorionic plates. “FlowSorted.CordBlood.450K” is more superior than “FlowSorted.Blood.450K” for estimating granulocytes in chorionic plates. Estimation using “FlowSorted.CordBlood.450K” agreed with the results of pathological staining. “FlowSorted.Blood.450K” overestimates granulocytes [22].

A limitation of this study is that the estimation of the "Gran" fraction of the chorionic plate was performed using a cord blood estimation algorithm. In the future, the cells that make up the chorionic plate should be accurately identified and an estimation algorithm should be developed based on the DNA methylation characteristics of each cell of the chorionic plate. Another limitation is that the sample number of the longitudinal analysis of this study was insufficient to distinguish between the effects of very preterm birth and CAM. Further analysis of follow-up studies on very preterm infants is required.

In this study, we found that the chorionic plate was highly sensitive to infection or inflammation in utero. The results suggested that aberrant DNA methylation at the genes involved in “blood vessel morphogenesis” within the chorionic plates is associated with the risk of developing postnatal pulmonary disease in newborns. The epigenetic diversity of the chorionic plate was associated with DNA methylation levels of SOCS3 in neonatal blood cells, as detected by the cg18181703 probe. The DNA methylation changes in cg18181703 during the postnatal period were longitudinal from birth. Further studies, such as single-cell-level analyses, are warranted to investigate the effects of cg18181703 hypomethylation in blood cells under systemic conditions.

BMI: body mass index

CAM: acute chorioamnionitis

FIRS: fetal inflammatory response syndrome

GO: Gene Ontology

HE: Hematoxylin-eosin

ICM: inner cell mass

IL-6: interleukin-6

nRBCs: nucleated red blood cells

PB: peripheral blood

SOCS3: suppressor of cytokine signaling 3

JAKs: Janus tyrosine kinases

STAT3: signal transduction and activator of transcription 3

TE: trophectoderm

Ethics approval and consent to participate

This study strictly adhered to the guidelines of the Declaration of Helsinki. All methods were performed in accordance with the following ethical guidelines in Japan: Ethics Guidelines for Human Genome/Gene Analysis Research, Ethical Guidelines for Medical and Health Research Involving Human Subjects, and Ethical Guidelines for Epidemiological Research. Ethical approval was obtained from the ethical committees of Human Genome Ethics Committee of the University of Tokyo Hospital, Tokyo, Japan (approval ID: G10036), Ethics Committee of the National Center for Child Health and Development, Tokyo, Japan (approval ID: 234), and Ethics and Personal Information Protection Committee of Tokyo Metropolitan Bokutoh Hospital, Tokyo, Japan (approval ID: 38). The study was explained to each participant and informed consent was obtained prior to participation. All the participants provided written informed consent for themselves and their newborns.

Consent for publication

Not applicable.

Availability of data and materials

All DNA methylation array data are available in the Gene Expression Omnibus (GEO) at GSE110829 (blood) and GSE265760 (chorionic plate).

Competing Interests

The authors declare that they have no competing interests.

Funding

JSPS KAKENHI : Tomoko Kawai JP22K11793, Kenichiro Hata JP21H02887 and JP21K19584.

Japan Agency for Medical Research and Development, AMED: Naoto Takahashi JP16gk0110007.

National Center for Child Health and Development (NCCHD) Valid : Tomoko Kawai 2020B-21, Kenichiro Hata 2022A-3.

Authors’ contributions

T.K. and K.H designed the study; K.K., T.K., M.I., and H.K. performed the experiments; S.A. and K.O. contributed the analytical tools; T.K., T.S., and K.K. analyzed the data. T.K. wrote the paper; K.K., T.K., N.T., and K.H. acquired funding; K.K., T.I., T.N., U.K., and H.H. contributed resources; T.K., K.K., K.M., and K.H. reviewed and edited the paper.

Acknowledgments

English language editing services were provided by Editage.

Papuchova H, Latos PA. Transcription factor networks in trophoblast development. Cell Mol Life Sci. 2022;79:337.
Siriwardena D, Boroviak TE. Evolutionary divergence of embryo implantation in primates. Philosoph Transac R Soc B Biol Sci. 2022;377:20210256.
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 of cells. Nucleic Acids Res. 1982;10:2709-21.
Lister R, Pelizzola M, Dowen RH, Hawkins RD, Hon G, Tonti-Filippini J, et al. Human DNA methylomes at base resolution show widespread epigenomic differences. Nature. 2009;462:315-22.
Schroeder DI, Blair JD, Lott P, Yu HOK, Hong D, Crary F, et al. The human placenta methylome. Proc Natl Acad Sci U S A. 2013;110:6037-42.
Xie W, Schultz MD, Lister R, Hou Z, Rajagopal N, Ray P, et al. Epigenomic analysis of multilineage differentiation of human embryonic stem cells. Cell. 2013;153:1134-48.
Loyfer N, Magenheim J, Peretz A, Cann G, Bredno J, Klochendler A, et al. A DNA methylation atlas of normal human cell types. Nature. 2023;613:355-64.
Alexander J, Teague AM, Chen J, Aston CE, Leung YK, Chernausek S, et al. Offspring sex impacts DNA methylation and gene expression in placentae from women with diabetes during pregnancy. PLOS ONE. 2018;13:e0190698.
Blair JD, Yuen RKC, Lim BK, McFadden DE, von Dadelszen P, Robinson WP. Widespread DNA hypomethylation at gene enhancer regions in placentas associated with early-onset pre-eclampsia. Mol Hum Reprod. 2013;19:697-708.
Wilson SL, Leavey K, Cox BJ, Robinson WP. Mining DNA methylation alterations towards a classification of placental pathologies. Hum Mol Genet. 2018;27:135-46.
Yeung KR, Chiu CL, Pidsley R, Makris A, Hennessy A, Lind JM. DNA methylation profiles in preeclampsia and healthy control placentas. Am J Physiol Heart Circ Physiol. 2016;310:H1295-303.
Lim JH, Kang YJ, Bak HJ, Kim MS, Lee HJ, Kwak DW, et al. Epigenome-wide DNA methylation profiling of preeclamptic placenta according to severe features. Clin Epigenetics. 2020;12:128.
Leseva MN, Binder AM, Ponsonby AL, Vuillermin P, Saffery R, Michels KB, et al. Differential gene expression and limited epigenetic dysregulation at the materno-fetal interface in preeclampsia. Hum Mol Genet. 2020;29:335-50.
Konwar C, Price EM, Wang LQ, Wilson SL, Terry J, Robinson WP. DNA methylation profiling of acute chorioamnionitis-associated placentas and fetal membranes: insights into epigenetic variation in spontaneous preterm births. Epigenetics Chromatin. 2018;11:63.
Fernandez-Jimenez N, Fore R, Cilleros-Portet A, Lepeule J, Perron P, Kvist T, et al. A meta-analysis of pre-pregnancy maternal body mass index and placental DNA methylation identifies 27 CpG sites with implications for mother-child health. Commun Biol. 2022;5:1313.
Everson TM, Vives-Usano M, Seyve E, Cardenas A, Lacasaña M, Craig JM, et al. Placental DNA methylation signatures of maternal smoking during pregnancy and potential impacts on fetal growth. Nat Commun. 2021;12:5095.
Taschereau A, Thibeault K, Allard C, Juvinao-Quintero D, Perron P, Lutz SM, et al. Maternal glycemia in pregnancy is longitudinally associated with blood DNAm variation at the FSD1L gene from birth to 5 years of age. Clin Epigenetics. 2023;15:107.
Richmond RC, Simpkin AJ, Woodward G, Gaunt TR, Lyttleton O, McArdle WL, et al. Prenatal exposure to maternal smoking and offspring DNA methylation across the lifecourse: findings from the Avon Longitudinal Study of Parents and Children (ALSPAC). Hum Mol Genet. 2015;24:2201-17.
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;76:3479-96.
Kashima K, Kawai T, Nishimura R, Shiwa Y, Urayama KY, Kamura H, et al. Identification of epigenetic memory candidates associated with gestational age at birth through analysis of methylome and transcriptional data. Sci Rep. 2021;11:3381.
Nordlund J, Bäcklin CL, Wahlberg P, Busche S, Berglund EC, Eloranta ML, et al. Genome-wide signatures of differential DNA methylation in pediatric acute lymphoblastic leukemia. Genome Biol. 2013;14:r105.
Bakulski KM, Feinberg JI, Andrews SV, Yang J, Brown S, L McKenney S, et al. DNA methylation of cord blood cell types: applications for mixed cell birth studies. Epigenetics. 2016;11:354-62.
Jaffe AE, Irizarry RA. Accounting for cellular heterogeneity is critical in epigenome-wide association studies. Genome Biol. 2014;15:R31.
Yuan V, Hui D, Yin Y, Peñaherrera MS, Beristain AG, Robinson WP. Cell-specific characterization of the placental methylome. BMC Genomics. 2021;22:6.
McLean CY, Bristor D, Hiller M, Clarke SL, Schaar BT, Lowe CB, et al. GREAT improves functional interpretation of cis-regulatory regions. Nat Biotechnol. 2010;28:495-501.
Blanc WA. Pathology of the placenta, membranes, and umbilical cord in bacterial, fungal, and viral infections in man. Monogr Pathol. 1981;(22):67-132.
Redline RW, Faye-Petersen O, Heller D, Qureshi F, Savell V, Vogler C, et al. Amniotic infection syndrome: nosology and reproducibility of placental reaction patterns. Pediatr Dev Pathol. 2003;6:435-48.
Motomura K, Romero R, Galaz J, Tarca AL, Done B, Xu Y, et al. RNA sequencing reveals distinct immune responses in the chorioamniotic membranes of women with preterm labor and microbial or sterile intra-amniotic inflammation. Infect Immun. 2021;89:e00819-20.
Steel JH, O'Donoghue K, Kennea NL, Sullivan MH, Edwards AD. Maternal origin of inflammatory leukocytes in preterm fetal membranes, shown by fluorescence in situ hybridisation. Placenta. 2005;26:672-7.
Villamor-Martinez E, Álvarez-Fuente M, Ghazi AMT, Degraeuwe P, Zimmermann LJI, Kramer BW, et al. Association of chorioamnionitis with bronchopulmonary dysplasia among preterm infants: a systematic review, meta-analysis, and metaregression. JAMA Netw Open. 2019;2:e1914611.
Higgins RD, Saade G, Polin RA, Grobman WA, Buhimschi IA, Watterberg K, et al. Evaluation and management of women and newborns with a maternal diagnosis of chorioamnionitis: summary of a workshop. Obstet Gynecol. 2016;127:426-36.
Shapiro S, Trail-Burns E, Slader MG, Laptook A, De Paepe ME. Correlation between chorionic plate vascularization and risk of bronchopulmonary dysplasia in extremely preterm infants. Placenta. 2020;101:154-8.
Issarapu P, Arumalla M, Elliott HR, Nongmaithem SS, Sankareswaran A, Betts M, et al. DNA methylation at the suppressor of cytokine signaling 3 (SOCS3) gene influences height in childhood. Nat Commun. 2023;14:5200.
Wahl S, Drong A, Lehne B, Loh M, Scott WR, Kunze S, et al. Epigenome-wide association study of body mass index, and the adverse outcomes of adiposity. Nature. 2017;541:81-6.
Mendelson MM, Marioni RE, Joehanes R, Liu C, Hedman ÅK, Aslibekyan S, et al. Association of body mass index with dna methylation and gene expression in blood cells and relations to cardiometabolic disease: a Mendelian randomization approach. PLOS Med. 2017;14:e1002215.
Geurts YM, Dugué PA, Joo JE, Makalic E, Jung CH, Guan W, et al. Novel associations between blood DNA methylation and body mass index in middle-aged and older adults. Int J Obes (Lond). 2018;42:887-96.
Li S, Wong EM, Bui M, Nguyen TL, Joo JE, Stone J, et al. Inference about causation between body mass index and DNA methylation in blood from a twin family study. Int J Obes (Lond). 2019;43:243-52.
Ali O, Cerjak D, Kent JW, James R, Blangero J, Carless MA, et al. Methylation of SOCS3 is inversely associated with metabolic syndrome in an epigenome-wide association study of obesity. Epigenetics. 2016;11:699-707.
Xu K, Zhang X, Wang Z, Hu Y, Sinha R. Epigenome-wide association analysis revealed that SOCS3 methylation influences the effect of cumulative stress on obesity. Biol Psychol. 2018;131:63-71.
Fraszczyk E, Thio CHL, Wackers P, Dollé MET, Bloks VW, Hodemaekers H, et al. DNA methylation trajectories and accelerated epigenetic aging in incident type 2 diabetes. GeroScience. 2022;44:2671-84.
Joehanes R, Just AC, Marioni RE, Pilling LC, Reynolds LM, Mandaviya PR, et al. Epigenetic signatures of cigarette smoking. Circ Cardiovasc Genet. 2016;9:436-47.
Sikdar S, Joehanes R, Joubert BR, Xu CJ, Vives-Usano M, Rezwan FI, et al. Comparison of smoking-related DNA methylation between newborns from prenatal exposure and adults from personal smoking. Epigenomics. 2019;11:1487-500.
Dugué PA, Jung CH, Joo JE, Wang X, Wong EM, Makalic E, et al. Smoking and blood DNA methylation: an epigenome-wide association study and assessment of reversibility. Epigenetics. 2020;15:358-68.
Gomez R, Romero R, Ghezzi F, Yoon BH, Mazor M, Berry SM. The fetal inflammatory response syndrome. Am J Obstet Gynecol. 1998;179:194-202.
Jung E, Romero R, Yeo L, Diaz-Primera R, Marin-Concha J, Para R, et al. The fetal inflammatory response syndrome: the origins of a concept, pathophysiology, diagnosis, and obstetrical implications. Semin Fetal Neonatal Med. 2020;25:101146.
Martino N, Ramos RB, Lu S, Leyden K, Tomaszek L, Sadhu S, et al. Endothelial SOCS3 maintains homeostasis and promotes survival in endotoxemic mice. JCI Insight. 2021;6.
Kershaw NJ, Murphy JM, Liau NPD, Varghese LN, Laktyushin A, Whitlock EL, et al. SOCS3 binds specific receptor-JAK complexes to control cytokine signaling by direct kinase inhibition. Nat Struct Mol Biol. 2013;20:469-76.
Takahashi N, Takahashi K, Kobayashi M, Yada Y, Koike Y, Kono Y. Constitutively high-level expression of TGFβ isoforms in cord blood and its relationship to perinatal findings. Cytokine. 2015;73:101-7.
Zhang W, Cai Z, Liang D, Han J, Wu P, Shan J, et al. Immune cell-related genes in juvenile idiopathic arthritis identified using transcriptomic and single-cell sequencing data. Int J Mol Sci. 2023;24:19619.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Placental chorionic plate DNA methylation patterns correlate with DNA methylation at SOCS3 in newborn human peripheral blood cells

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Background

Methods

Study population

Chorionic plate separation and DNA extraction

Mononuclear cell separation

DNA methylation microarray analysis and data processing

Cell fraction estimation

Results

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1