DNA Methylation in the Placenta and Maternal Socioeconomic Status: The SPAH Study

DNA Methylation in the Placenta and Maternal Socioeconomic Status: The SPAH Study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article DNA Methylation in the Placenta and Maternal Socioeconomic Status: The SPAH Study Ella O. Beraldo, Ann E. Borders, Amy M Inkster, Linda M Ernst, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7546517/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 24 Apr, 2026 Read the published version in Clinical Epigenetics → Version 1 posted 16 You are reading this latest preprint version Abstract Background: Disparities in socioeconomic status have been associated with adverse pregnancy outcomes, including preterm birth and fetal growth restriction. As the barrier between maternal exposures and the fetus, the placenta has been proposed to play a role in the mechanisms leading to poor health outcomes seen with socioeconomic disadvantage. We hypothesized that exposure to lower SES during pregnancy may lead to altered placental DNA methylation (DNAme) that is in turn associated with other pregnancy outcomes. Methods: Placental samples from the Stress, Pregnancy, and Health Study (SPAH) study (n=493) were processed for DNAme analysis using the Illumina Infinium MethylationEPIC BeadChip array. Linear modelling was used to assess whether placental DNAme was associated with Socioeconomic Position, Financial Resources, and/or Disadvantage. Results: At FDR 0.05, we observed only 2 CpGs associated with Socioeconomic Position after correcting for gestational age and ancestry, while at a less stringent |∆β| >0.02 threshold there were 77 and 22 CpG associations with Socioeconomic Position and Disadvantage respectively. However, these changes seemed to be explained by genetic variation influencing DNAme in combination with population stratification by socioeconomic status. We did observe associations between socioeconomic status and DNAme-inferred cell composition and epigenetic age acceleration, with intrinsic (p=0.047) and extrinsic (p=0.050) age acceleration being slightly accelerated with lower levels of Socioeconomic Position. Financial Resources and Disadvantage SES trended in the same direction as Social Position, with lower socioeconomic status seen with higher levels of age acceleration, though not reaching significance. No meaningful associations in sex stratified analyses were identified, although XX placentas showed higher cytotrophoblast:syncytiotrophoblast ratio than XY placentas (p=0.00013). Conclusions: Our results emphasize the importance in placental studies involving diverse cohorts to account for genetic variation in order to avoid false findings. This study also demonstrates the challenges with elucidating mechanisms underlying socioeconomic associated outcomes, given the complex nature of correlated variables. Further investigation is required to elucidate whether epigenetic age acceleration is an adverse effect of exposure to prenatal maternal stress associated with socioeconomic disparities, or alternatively, if placental epigenetic aging is a potential healthy adaption to pregnancy complications that increase the risk of preterm delivery. DNA methylation placenta socioeconomic status pregnancy epigenetic age Figures Figure 1 Figure 2 Figure 3 Figure 4 Background Birth outcomes vary by socioeconomic status (SES), with low SES linked to higher rates of fetal growth restriction (FGR), preterm birth (PTB), and preeclampsia (1–5). These disparities in adverse outcomes continue post birth, as children born into low SES households have increased rates of mental health conditions and morbidity and mortality from chronic disease (6–9). What remains unclear is how SES influences biological processes that are proximally involved in brain maturation, health problems, and other developmental outcomes that are patterned by SES. It has been postulated that socioeconomic disadvantage becomes biologically embedded during sensitive periods of gestation, specifically through modulation of epigenetic processes in the placenta (10). The placenta is thought to be implicated in this process as it functions as a barrier protecting the fetus from maternal exposures and facilitates maternal-fetal exchange during gestation (10–12). Individuals living in low-SES conditions are disproportionately exposed to a variety of psychological stressors (e.g., material hardship, stigmatization, mistreatment, job instability, neighborhood violence) and environmental pollutants (e.g., fine particulate matter, volatile organic compounds, environmental tobacco smoke, microplastics and other harmful chemicals such as phthalates) (13–18). Such individuals exposed both early and later in life also have a higher risk for chronic diseases such as hypertension and diabetes (19). In both rodent and primate models approximating human socioeconomic disadvantage during pregnancy, animals show dysregulation of biological functions such as psychological stress response, nutrient imbalance, and glucocorticoid excess (20–22). In humans, socioeconomic adversity has been associated with differential DNAme or transcription of genes involved in placental function, cortisol hormone signalling, immune activation and fetal maturation (10,23,24). However, the association between SES, or specific aspects of SES, and placental functioning remains unclear, as SES is a multidimensional construct. Conventional indicators of SES such as household income, education, and occupational prestige are inherently inter-related, but can diverge substantially for some individuals. For example, some jobs that command a high salary, like trades, do not require a university education; others that do involve high levels of education, such as academic careers in the humanities, do not necessarily command high salaries. Different elements of SES may also affect health through distinct pathways, for example, income provides an indication of the material resources individuals have at their disposal, whereas education and occupation are windows into social status, health literacy and social networks (14). Genome-wide DNAme studies are commonly employed for exploring links between environmental exposures and health outcomes, and altered placental DNAme has been reported in association with some societal and environmental exposures, such as air pollution and other environmental chemicals like bisphenol A (BPA) (25–28). Furthermore, epigenetic age acceleration, the difference between age predicted by DNAme clocks and chronological age has been negatively associated with SES in adult tissues (29,30). However, the influence of SES on the placental DNA methylome and epigenetic age has not yet been well-characterized. Studying the placenta, however, presents some challenges as it is a heterogeneous tissue composed of multiple cell types of varying developmental origin, each with a distinct DNAme profile (31). Furthermore, genetic variation contributes to 20–70% of DNAme variation (32–35) and thus DNAme can vary between populations of different ancestry due to differences in allele frequencies (36–38). Thus, accounting for cellular and genetic variation is necessary before it can be concluded that changes to DNAme are attributable to the exposure of interest (39,40). In this study, we analyzed Illumina Infinium MethylationEPIC v1.0 DNAme array data obtained from 493 placentas derived from a sociodemographically diverse cohort of pregnancies recruited from Chicago, Illinois. We aimed to identify patterns of DNAme associated with SES, and to evaluate whether the nature and strength of these associations varied depending on how SES is assessed and/or fetal sex. We further examined “epivariables” inferred from the DNAme data directly, including estimated cell composition (31) and epigenetic age acceleration (41). As stress effects on gestational biology can be sex-specific (20), we hypothesized that pregnancy in individuals exposed to lower SES may lead to altered placental DNAme that is associated with other SES-related variables such as PTB and/or FGR, and that these associations may differ by fetal sex. Methods & Materials Cohort The data presented here come from the Stress, Pregnancy, and Health Study (SPAH) study. The study protocol was approved by the Institutional Review Boards of Northwestern University (STU00206269) and Endeavor Health (EH17-006). Written informed consent was obtained from each participant. This prospective observational cohort study was designed to evaluate multiple measures of socioeconomic status in relation to pregnancy disparities. SPAH enrolled 605 pregnant individuals mid-pregnancy, recruited from four clinical sites across the Chicago metropolitan area (Erie Family Health Center, The Center for Maternal and Fetal Health at Evanston Hospital, The NorthShore Community Health Center at Evanston Hospital, and NorthShore Lincolnwood Medical Group Office) between March 2018 and September 2022. Eligible participants were recruited at prenatal clinical sites before 25 weeks gestation to complete a series of questionnaires during second and third trimesters. Pregnant individuals were eligible for the study if they were 18 years or older, carrying a singleton pregnancy, and English speaking. Individuals were excluded from the study if there were fetal congenital anomalies or known chromosomal abnormalities in the pregnancy. Of the 605 pregnant women recruited for SPAH, biopsies for placental DNAme were obtained from 509 of these pregnancies. Samples from 16 placentas were removed from the DNAme cohort during data processing for failing quality checks (as described further below in section 2.5.) yielding 493 cases for analysis. Measures of SES Disparities SES and demographic data were collected between 20–26 weeks of gestation via interview and survey. The SES interview collected information on the pregnant individual and their financial resources, such as household income and savings, educational attainment, occupation, governmental assistance, and demographics. Three composites were created to assess various dimensions of the participants’ SES (42). To capture aspects of SES related to socioeconomic position (i.e., standing in society based on prestige and power), we calculated a composite based on standardized levels of highest household educational attainment and occupational prestige, using methods from The National Statistics Socio-Economic Classification (NS-SEC) (2010) (43) to code for occupational prestige. This composite’s emphasis on educational attainment and occupational prestige mirrors the Hollingshead Index, the most widely used indicator of SES in the social sciences, though we exclude that measure’s consideration of participant sex and marital status, which are less relevant in contemporary society. To capture financial resources , a separate composite was calculated as a continuous variable including standardized levels of household income-to-poverty ratio (IPR), total savings and assets, and household savings relative to cost of living (i.e., “if you lost all your current source(s) of household income (your paycheck, public assistance, or other forms of income), how long could you continue to live at your current address and standard of living?”). A socioeconomic disadvantage composite was computed to summarize overall disadvantage, incorporating select features of both socioeconomic position (i.e., highest household educational attainment) and financial resources (i.e., income below federal poverty threshold and whether savings was less than 2 months of living expenses). In addition, the socioeconomic disadvantage composite included other features less commonly included in SES research, but which are still notable indicators of greater disadvantage (or lower SES) (i.e., recipient of TANF, WIC, SNAP, CHIP, SSI, or Medicaid). The composite index here as a count score also captures a different distribution compared to our continuous measures. The composite was computed as a count score calculated on a 0–5 scale, with one point for each of the following: household income less than twice the federal poverty threshold (IPR < 2.0), savings less than 2 months of living expenses, highest education in the household less than a two-year college degree, receipt of government assistance (including TANF, WIC, SNAP, CHIP, SSI, or Medicaid), and self or partner currently unemployed. This socioeconomic disadvantage composite has been utilized in previous biological studies of SES (10,44). Placental Collection Placentas were sampled from 509 of the 605 participants. The vast majority of missing samples were from complicated deliveries, where staff could not obtain specimens within that timeframe. At the time of delivery, research staff or obstetric providers obtained 0.4 cm 3 chorionic villous biopsies from three separate cotyledons from the fetal-facing side of each placenta to minimize spatial variability in DNA methylation. BiopsiesAll biopsies were collected from 509 obtained within 6 hours of the 605 participants delivery, and 59% were obtained within 1 hour of delivery (average time of 1.76 hours from delivery). Samples from an additional 16 placentas were removed for failing quality checks (as described further below in section 2.5.) yielding 493 cases for analysis. The biopsies were stored at -80°C until the end of the study, at which time high-quality DNA was extracted from using the PerkinElmer Chemagic 360 System at the Northwestern University NUSeq Core Facility. The extraction process is based on the PerkinElmer Chemagen M-PVA Magnetic Bead technology (PerkinElmer, Waltham, Massachusetts). Briefly, the collected placental tissue samples were first lysed in the presence of protease using gentleMACS (Miltenyi-Biotec, Bergisch Gladbach, Germany). Then the chemagen M-PVA magnetic beads were added to bind DNA from the lysates, followed by several rounds of washing before elution of DNA from the beads. The extracted DNA was then checked for quality using NanoDrop, and quantified using Qubit. The purified DNA samples were stored at -20 o C. DNA Methylation Arrays DNAme was assayed at the NUSeq Core Facility using the Infinium Human MethylationEPIC Beadchip v1.0 array (Illumina, Inc. CA, USA), which targets over 850,000 CpG sites. Samples were randomly plated on each chip. A 500 ng DNA sample was used to perform bisulfite conversion followed by Illumina’s protocol for DNA methylation profiling. BeadChips were scanned with an Illumina iScan instrument. DNA Methylation Processing and Quality Control DNAme data (IDAT files) were read into R v 4.2.2 for processing according to a previously published pipeline relying on the minfi, ewastools, and conumee R packages (45). In brief, epiphenotyping variables for genetic ancestry, gestational age (GA), and cell composition were estimated from the DNAme data itself using the PlaNET R package, as previously described (45). The raw data was normalized for analysis using the dasen noob combined normalization method (46,47). After normalization, we excluded poor-quality probes (bead count 0.01 in > 5% of samples, n = 12,737), as well as previously identified cross-hybridizing probes (n = 103,376) from our dataset (48). Of the 509 unique placental samples run on the array, 16 samples failed checks and were removed Two samples were removed for failing probe quality checks (> 1% of array probes, failed detection P/bead count). Another 8 samples failed multiple measures: they were identified to have much lower inter-sample correlation values in the whole cohort, were flagged for probable maternal contamination identified using the ewastools R package (49), and separated from the rest of the dataset on PC1 and PC2 during principal component analysis. In 5, cases clinical reported fetal sex did not match the sex chromosome complement of the DNAme sample, inferred using X and Y chromosome probes as previously described (45). Finally, one sample appeared to have a mosaic trisomy 7 based on aneuploidy detection using the R package conumee 2.0 (50) and was excluded, as trisomy can have profound effects on DNAme. After extensive data processing and quality control steps, a total of 748,484 probes (n = 732,102 autosomal; n = 16,382 chrX; n = 272 chrY) in 493 samples remained for analysis. Epivariable estimation Genetic Ancestry Ancestry of the placenta reflects both the maternal and paternal genetic contribution and is measured on a continuum relative to multiple reference populations. PlaNET is a tool that estimates ancestry probabilities from the DNAme itself and are provided as continuous variables along three major axes of population variation relative to African, European, and East Asian populations. Although other major populations such as South Asian and Amerindian ancestry cannot be captured with this DNAme based measure, correcting for PlaNET ancestry variables can improve reproducibility in EWAS studies (45,51). Although there was a relationship between maternal self-reported race and placental ancestry estimates in the SPAH cohort, these are distinct phenomena and there is considerable variation in placental ancestry within maternal racial groups (Fig. S1 C), as well as within both Hispanic and non-Hispanic maternal ethnicity (Fig. S1 D). Cell Composition Cell composition is a major driver of placental DNAme (31), as different cell types can have markedly different epigenetic profiles and can also vary greatly between datasets due to systematic sampling techniques. To gain a better understanding of the major cell types influencing DNAme in the SPAH cohort, we used PlaNET to estimate the composition of 6 major placental cell types across all sample in the SPAH cohort (Fig. S1 B). The overall cell composition of the SPAH cohort showed high levels of predicted syncytiotrophoblast (average > 0.75%) (Fig. S1 B), with several samples being estimated as 100% syncytiotrophoblast, likely reflecting sampling from the tips of floating villi, avoiding any vessels. Gestational Age and Epigenetic Age Acceleration Estimates Gestational age (GA) can be estimated using several placental epigenetic clocks. We chose the control placental placental epigenetic clock (CPC) published by Lee et al. (2019) (41) as it was trained on normative pregnancies without known pathologies, and the effect of SES on placental epigenetic age acceleration remains unknown (41). Clinical GA correlated well (R = 0.6, p < 0.001) with epigenetic age estimates (Fig. S1 A). Extrinsic epigenetic age acceleration was calculated by taking the residuals of a linear regression model with CPC predicted epigenetic age as the dependent variable and chronological GA as the independent variable. Intrinsic age acceleration was calculated in the same manner but also included PlaNET-estimated cell proportions for 6 cell typesas covariates to account for cell composition. The association between epigenetic age acceleration and the Socioeconomic Position and Financial Resources composites was evaluated using linear models, both in whole cohort and sex-stratified analyses. Wilcoxon Rank Sum tests were used to evaluate whether epigenetic age acceleration varied by Disadvantage score, again both in whole cohort and sex-stratified analyses. Linear Modelling DNAme data at all filtered autosomal CpGs (n = 732,102) were converted to M values prior to linear modeling, to test for DNAme differences across the (i) Socioeconomic Position, (ii) Financial Resources and (iii) Disadvantage Composite scores. Three separate linear models were run with the three SES scores as the primary variables of interest in each. Models were run using the limma R package (52,53), and a False Discovery Rate (FDR) of 0.05 was used to establish statistical significance, while a biological cutoff |Δβ| ≥ 0.05 was used to reduce false positive results.. We also report results also using a less stringent effect size cut-off of |Δβ| ≥ 0.02 which was established as the likely limit of technical detection in this data based on the maximum standard error across all CpGs for all samples (37). In the 493 placentas analyzed missing values for maternal BMI (n = 23) were imputed to the median (31.06), while missing data for mode of conception (n = 14), and marijuana use during pregnancy (n = 13) were imputed to: 06kg/m 2 for BMI, “non-assisted” for mode of conception and “no use during pregnancy” for marijuana”, respectively, as both aligned with the overwhelming majority of the cohort. Results Cohort characteristics Table 1 displays maternal demographic and clinical characteristics of the 493 placentas analyzed. Among the participants, 21.7% had a household education level of high school diploma or less. 20.5% were low income or poor according to the income to poverty ratio. In terms of racial identity, the majority of participants self-identified as ‘white’ (n=318, 64.5%), followed by ‘Black’ (n=87, 17.6%) and ‘Asian’ (n=53, 10.8%), with mixed race (more than one chosen category) and other races comprising the remainder (n=56, 11.3%). In terms of ethnicity, 123 participants self-identified as ‘Hispanic’ (24.9%). The mean GA at delivery was 38.6 weeks, with a range of 24.0 to 41.7 weeks; and 11.2% (n=55) delivered preterm (<37 weeks gestation). An excess of male fetuses (XY placentas) (n=276, 56.0%) as compared to female fetuses (XX placentas (n=217) was observed (p<0.05). Table 1. Maternal demographics and clinical characteristics of studied placentas (n=493). Cohort Feature Count (% or mean as indicated) Maternal Age (mean, (SD, range)) 33.4 (5.6, 18.4-51.7) Race (n, %) American Indian or Alaskan Native Asian Black Hawaiian Native or Pacific Islander white Mixed race Other race 1 (0.2) 48 (9.7) 76 (15.4) 4 (0.8) 303 (61.5) 18 (3.7) 43 (8.7) Ethnicity (n, %) Hispanic Non-Hispanic 123 (24.9) 370 (75.1) SES measurements (mean (SD, range)) Financial resources composite Socioeconomic position composite Disadvantage composite 0.0 (0.73, -1.19-5.21) 0.02 (0.88, -2.19-0.94) 1.06 (1.37, 0-5) Pregestational Diabetes (n, %) Type 1 Type 2 No pregestational diabetes 15 (3.1) 30 (6.1) 403 (90.8) Gestational Diabetes (n, %) Gestational Diabetes, Insulin Required Gestational Diabetes, No Insulin Required No gestational diabetes 27 (5.5) 27 (5.5) 437 (89.0) Complications (n, %) Preeclampsia Gestational Hypertension 49 (10.0) 37 (7.6) Substance Use During Pregnancy (n, %) Alcohol Smoking Marijuana 24 (5.0) 5 (1.0) 23 (4.8) Gestational Age (Weeks) (mean (SD, range)) 38.6 (2.0, 24.0-41.7) Fetal Sex (male:female) 276 : 217 What are the relationships between clinical, sociodemographic, and epivariables in the SPAH cohort? Before engaging in an epigenome-wide association study (EWAS), we sought to understand how clinical and sociodemographic and cell composition variables related to one another to inform downstream analyses and to identify possible collinearities. Of the demographic and DNAme-derived variable relationships, we found first that expected variables (i.e. cell composition estimates, PlaNET derived ancestry estimates/maternal self-reported race categories, diabetes coded as pregestational or gestational, mode of delivery, and gestational age dependent variables) were strongly correlated with each other (R 2 > 0.5) (Fig. S2). There were some significant associations with our primary exposure variables (Socioeconomic Position, Financial Resources, and Disadvantage scores) (Fig. S3). Considering maternal demographic and health variables, white race was associated with more advantageous SES, while Black or “other” race and Hispanic ethnicity were all associated with more adverse SES across all three measures. In addition, high BMI was associated Disadvantage, whereas as both pregestational and gestational diabetes was associated with measures reflecting higher SES. At the level of the placenta, higher European ancestry probability was associated with SES advantage while higher African ancestry probability was associated with measures of lower SES. What are drivers of DNAme variation in the SPAH cohort? By performing principal components analysis (PCA) on the overall data we can determine how DNAme variation is distributed amongst the samples, and how the PCs relate to clinical, demographic, epiphenotype and technical factors (45). After data processing and normalization, technical variables (the plate the sample was run on, the chip the sample was run on, the row location on the chip) were not significantly associated with either PC1 or PC2 (Fig. S4). PC1 accounted for 16.7% of the variance and was strongly associated with cell composition. PC2 accounted for 4.3% of variance was most strongly associated with gestational age, sex, PlaNET European estimated ancestry and maternal self-reported white race (p < 0.01). There were also lesser but significant (p < 0.01) associations with PlaNET African estimated ancestry and maternal self-reported Hispanic ethnicity. The Socioeconomic Position and Disadvantage composites were associated with PC2 (p-value < 0.01) while Resources and Disadvantage composite measures were weakly associated with PC3 (p < 0.05). As we found no correlation in cell composition estimates with the SES measures (Fig. S3) and as altered cell composition can be an interesting aspect of changes to placental function, we chose not to correct for cell composition in our models, but to test these effects separately. In contrast, it is estimated that 20-70% of DNAme variation is influenced by genetic variation (32,35,54,55) and genetic variation associated with placental ancestry was also correlated with SES measures. Therefore, to detect DNAme changes due to SES, we needed to account for ancestry influenced DNAme. Is autosomal DNAme associated with socioeconomic disadvantage? Table 2. The number of differentially methylated CpGs (FDR 0.02 1123 69 149 > 0.05 29 9 0 Model B) DNAme ~ SES + Ancestry + GA > 0.02 77 0 22 > 0.05 2 0 0 Model C) DNAme ~ SES + Ancestry + GA + Ethnicity > 0.02 0 0 0 > 0.05 0 0 0 Model D) DNAme ~ SES + All Correlated Covariates > 0.02 0 0 0 > 0.05 0 0 0 We used linear modelling to assess differential DNAme across autosomal CpGs to evaluate whether placental DNAme was associated with Socioeconomic Position, Financial Resources, and/or Disadvantage (Table 2). As a number of variables were associated with SES that could also affect DNAme, we took a sequential approach to evaluate how different variables might influence results. We first ran a crude model (Model A) correcting for no covariates to have a baseline of differential methylation in the SPAH cohort. Under this model and at FDR 0.05 in models testing Socioeconomic Position, Financial Resources, and Disadvantage, respectively (Table 2). However, 1123, 69 and 149 CpGs reached a minimal biological threshold of |∆β| > 0.02, with those same comparisons (Table 2). As the variables most strongly associated with PC2 were gestational age at delivery and PlaNET ancestry estimates, and these are well established to drive placental DNAme variation (56,57), we repeated the linear modeling adding in these covariates (Model B). This correction eliminated most of the initial hits. Only 2 CpGs were associated with Socioeconomic Position meeting a biological threshold of |∆β| > 0.05 (Fig. 1A), both of which showed high variance and differed by genetic ancestry (Fig. S5, Fig. 2). The CpG in ATP2C2 with a trimodal distribution (assumed to represent three genotypes) as is typically seen for methylation strongly influenced by one or more genetic variants. The CpG in C1orf141 may be influenced by multiple factors, but the unusually high variance in all genetic ancestries, suggests that DNAme at this site is unlikely to be biologically meaningful in the context of SES exposure. This suggests that these two sites are influenced by genetic variation, despite correcting for ancestry in our models. Using a relaxed threshold of (|∆β| > 0.02) (Table 2), 77 CpGs were associated with Socioeconomic Position and 22 were associated with Disadvantage. We compared these CpGs to a comprehensive list of Illumina CpG sites known to be influenced by genetic variation in adult blood (33) and found that most of our candidate CpGs were associated with variation in nearby SNPs. To further understand what variables might be driving DNAme at these 77 loci, we performed PCA on their DNAme beta values. Interestingly, we found a strong separation across PC1 and PC2 with Hispanic ethnicity (Fig. S6). It is important to note that Hispanic ethnicity is commonly associated with a variable mix of African, European, and Native American ancestries (58), the latter of which was not accounted for in our ancestry epivariables. We thus decided to next correct for Hispanic ethnicity (Model C), which resulted in no significant CpGs in any of the SES models that met either significance threshold. As maternal health or exposure related variables could also potentially obscure psychosocial associations of SES with DNAme we lastly ran a model (Model D) adjusting for any maternal health covariates that were significantly (p 0.15, with n > 10 for case vs. control of the variable in question) with the three SES variables. These additional covariates included mode of conception (natural or assisted), parity, maternal gestational and pregestational diabetes (yes/no), maternal BMI, marijuana use during pregnancy (yes/no), maternal age at delivery, pre-term delivery (yes/no), maternal self-reported Hispanic ethnicity (yes/no), and maternal self-reported race (Black, white, other.). Are there sex associated effects of socioeconomic disadvantage on placental DNAme? Sex differences have been reported in human development and some pregnancy outcomes show a sex bias (59,60). Feto-placental sex may modify the effect of prenatal maternal stress on child health outcomes (61). Because sex-differential effects with stress exposure can be obscured when males and females are analyzed together, we sex-stratified data and reran Model B (correcting for gestational age at delivery and PlaNET ancestry estimates) for each of the SES composites (Fig. S7). One CpG met significance (FDR 0.05) with Socioeconomic Position in XY samples only (Fig. S7D); no other sex-stratified autosomal models returned any significantly differentially methylated CpGs. Upon investigation the CpGs associated with Socioeconomic Position score in XY placentas again appeared to be influenced by genetic variation based on the distribution of DNAme values and an association with ancestry (Fig. S5C). Most EWAS studies do not include analysis of the sex chromosomes because of analytical complexity associated with sex chromosome dosage differences between XX and XY individuals and the impact of X-chromosome inactivation on the epigenome (62). However, there are many functionally important genes on the X chromosome that can be meaningful in the context of placental as well as neurological development (63). To fully explore if the effects of SES were influenced by sex, we thus analyzed X and Y chromosome data from each of the sexes separately (i.e. XX X-chromosome, XY X-chromosome, XY Y-chromosome) (Fig. S8). There was no evidence for altered DNAme on the X or Y chromosome in any of our models (Model B run for each of the SES composites scores) in either XX or XY samples at FDR <0.05. Are there differences in cytotrophoblast:syncytiotrophoblast ratio associated with SES scores? Evaluation of epivariables, inferred from DNAme data such as cell composition, can be a statistically powered way to detect placental changes associated with pathology (64). Although we found no overall correlation between cell composition estimates and any of the SES measures (Fig. S3), we were interested to evaluate the cytotrophoblast:syncytiotrophoblast (cyt:syn) ratio, which is increased in preeclampsia (51,65), and decreased with gestational age and male sex (45,66). In this cohort, the cyt:syn ratio was negatively correlated with gestational age, (p<0.001) (Fig. 3A) (45) and XX placentas had a higher cyt:syn ratio than XY placentas (p=0.00013) (Fig. S9A), consistent with previous reports (45,66,67). This sex difference in ratio was significant across gestational ages (Fig. S9B). The cyto:syn ratio was not associated with predicted genetic ancestry groups nor with Hispanic ethnicity, nor was it associated with any of the 3 SES composites (Fig. 3B-D) Because of the effect of sex on the cyto:syn ratio we also performed a sex-stratified analysis on this ratio. We observed that the cyto:syn ratio was significantly and negatively associated with Socioeconomic Position (Fig. S9C) and Resources (Fig. S9D) in XY, but not XX, samples and. XY placentas also showed decreased cyt:syn ratio with the lowest level of Disadvantage, though the trend continued across Disadvantage levels (Fig. S9E). However, to determine if these effects were sex specific we evaluated the interaction term (sex*SES composite score in question) (68), and observed that it was not significant for either the Socioeconomic Position or Disadvantage analyses. The interaction term of sex and Resources was significant, suggesting that there may be an effect of Resources SES on the cyt:syn that is limited to XY samples. Are there differences in epigenetic age acceleration in association with SES scores? In adults, low SES has been linked to epigenetic age acceleration (i.e. a predicted age from DNAme that is greater than chronological age) (29,30). Epigenetic age acceleration has also been linked to several adverse health outcomes including cancer, cardiovascular disease, diabetes, dementia, and overall mortality risk in several different tissues (30,69–71). Epigenetic clocks have been developed specifically for placenta but have not been widely explored for how these are affected by exposure conditions. We therefore evaluated epigenetic age acceleration in the SPAH cohort both without accounting for cell composition (extrinsic) for an overall picture of epigenetic aging and after adjustment for placental cell types (intrinsic) to determine if one specific cell population was driving changes. Certain demographic variables were associated with both measures of epigenetic age acceleration in the SPAH cohort (Fig. S2). As expected, extrinsic, but not intrinsic, epigenetic age acceleration, was associated with all cell estimates (p<0.05). Also as expected, both epigenetic age acceleration measures were associated with PlaNET-predicted GA (p<0.01). Both extrinsic and intrinsic age acceleration was also associated with fetal sex (p<0.05), BMI (p<0.05), Hispanic ethnicity (p<0.01), maternal White Race (p<0.05), and maternal Hawaiian Native Pacific Islander Race (p<0.01). Epigenetic age acceleration was not associated with the PlaNET-estimated ancestry variables, nor with the Resources and Disadvantage SES composite scores (p<0.05). When investigated further, we found a borderline (p=0.05) association between placental epigenetic age acceleration (both extrinsic and intrinsic) and decreasing Socioeconomic Position (Fig. 4A,B). Though the trend was not significant between epigenetic age acceleration and either Resources or Disadvantage, the trend was in the same direction as with Socioeconomic Position, with increased epigenetic age acceleration associated with decreased Resources and Disadvantage scores (Fig. 4C-F). When sex-stratified, there were no significant interactions with sex and placental epigenetic aging across any of the 3 SES scores (Fig. S10). Discussion Low SES shows established links to increased rates of adverse pregnancy outcomes in diverse populations world-wide. This may be due to multiple factors including differences in nutritional availability, maternal health, or exposures such as pollutants, pathogens, or stress. Many of these factors have been associated with changes to DNAme in blood (26) or placenta (27). In this study, we examined whether placental DNAme was associated with differing components of SES. At the global DNAme level, the only large-scale changes we observed seemed to be explained by population stratification by SES, or differences in genetic variation influencing DNAme. This implies that large-scale DNAme changes in the placenta are unlikely to explain the adverse health outcomes associated with maternal SES. However, we did observe some associations between SES and DNAme-inferred cell composition and epigenetic age acceleration, which are worth confirming in independent cohorts and exploring more to understand. A challenge to interpreting epigenome-wide association studies is that a significant portion of DNAme variation is influenced by genetic variation, and the associated allele frequencies can differ between populations. Specifically, it is estimated that in blood, ~ 10–30% of Illumina 450k/850K CpGs are differentially methylated by ancestry between African and European populations (72) and ~ 22% between East Asian and European populations (73). As the SES composites were associated with race and ethnicity, which can in turn be associated with differences in genetic ancestry, it was important to account for genetic ancestry in our models. While there is a loose relationship between race and ancestry, it should be noted that race is a social construct, with no genetic basis, and ancestry estimates vary by race across the United States and are dynamic and changing (58). Genetic ancestry is measured on a continuum, typically on multiple axes relative to major ancestral population groups (e.g. European, African, East Asian, Amerindian, and South Asian). While imperfect, epiphenotyping methods allowed us to roughly estimate the portion of variance associated with European, East Asian, and African ancestry when only DNAme data (and not genetic) is available for a set of samples (56). After adjusting for inferred ancestry probabilities in our epigenome-wide association models, few associations were found, and most remaining were known to be influenced by genetic variation and also were strongly associated with Hispanic ethnicity in our study. Because our ancestry estimates cannot account for Native American ancestry, which can be an important component of ancestry in Hispanic populations (74), we subsequently corrected for Hispanic ethnicity in the models. After doing so, we did not detect any remaining significant associations with SES, indicating that there were no large-scale DNAme associations with SES in this study population. Social aspects of race and ethnicity can be an important component of SES and should still have been preserved to some extent in our models. However, we cannot exclude the possibility that confounding between race/ethnicity and ancestry may have prevented us from detecting important associations. It is also important to note that genes which are differentially expressed between ancestry groups tend to be enriched for genes that interact with the environment and ancestry-specific disease effects (75). Given that SES-race associations have been associated with increased risk of health complications (76), further study is needed to investigate how genetic variation-DNAme-environment interactions may affect SES-related outcomes. As most reports of altered placental DNAme in association with environmental exposures have been of small effect size and not consistently reproduced, other mechanisms should also be considered. Adverse pregnancy outcomes could be due to changes to gene transcription that are independent of DNAme (such as small non-coding RNAs, histone modifications) or direct effects of molecules in maternal blood that cross the placenta into fetal circulation (e.g. some viruses or chemicals). Furthermore, the mediators of low SES on pregnancy outcomes are likely diverse, making it more challenging to detect associations at a population level. We observed significant differences in the trophoblast cell ratio between XX and XY samples, with XX placentas having higher cyt:syn ratio than XY placentas. We observed the same sex difference in XX and XY cyt:syn ratio in a previous cohort profiling placental DNAme in relation to environmental flooding induced maternal stress (cite QF2011 paper when published) and this was also reported in an independent study of IVF and control placentas (66). Thus, this appears to be a biological sex difference in placentas independent of stress exposure. Subsequent studies should explore if differences in cell composition contribute to differences in XX and XY vulnerability to in utero exposures and prevalence of placental complications (20,63,77). It is unclear how SES related prenatal maternal stress is associated with this ratio difference, as the interaction with sex was significant with only with Resources SES. The specific aspects of Resources SES that influence this potential sex difference should be verified and then further explored. In the present study, we found that both intrinsic and extrinsic age acceleration was slightly accelerated with lower levels of Socioeconomic Position composite scores. Although the association between epigenetic age acceleration with Financial Resources and Disadvantage SES composite scores did not reach significance, they did trend in the same direction with lower levels of SES associated with higher levels of age acceleration. Further, our results mirror reports of SES and epigenetic aging in adults, as exposure to socioeconomic disadvantage in early life (< 5 years of age), but not current socioeconomic disadvantage exposure in adulthood (ages 15 to 55 years of age), was associated with accelerated epigenetic aging of monocytes from adult peripheral blood, indicating significance of exposure to SES during critical developmental periods (29). Our results are also in line with other results of SES effects in the placenta; in utero exposure to maternal SES adversity was associated with differential DNAme of genes with important roles in gene transcription and placental function, indicating that exposure to SES may impact placental DNAme in a way that may be related to biological pathways relevant to fetal and pregnancy outcomes (23). However, such changes may also be the consequence of healthy adaption to stress, as a recent placental study showed that epigenetic age acceleration was associated with shorter NICU stays (78), or may be a secondary response to other changes. Our study is not without its limitations. The epiphenotyping tool PlaNET is only able to estimate a limited number of placental cell types and does not distinguish amongst subtypes of trophoblast, such as those that have been identified through single cell sequencing (79–81). Furthermore, DNAme-based genetic ancestry estimates are less robust than those computed using genome-wide SNP genotypes; further, the placental ancestry estimation tool to date was not developed for ancestry groups such as Native American or South Asian populations, and may not accurately detect genetic ancestry variation along these axes (31,56). Additionally, the SPAH cohort, though diverse compared to other placental datasets, still did recruit a majority of women of white self-reported race (64.5%) and distributions of SES composite scores skewed in the direction of higher SES. Thus, efforts should be made to confirm our results in other diverse cohorts with greater representation of low SES individuals, as well as to account for genetic variation using more robust approaches. The major strength in our study lies in the careful and deliberate analyses of the effects of SES on placental DNAme variation. Many studies make no attempt to correct for genetic ancestry, cell type composition, or gestational age when studying effects of maternal exposures on epigenetics, which may result in false positives and limit the ability to detect reproducible and robust associations. In addition, we also carefully investigated the effect of sex via not only sex-stratified analyses, but also in direct analysis of the sex chromosomes, which are very often discarded in EWAS studies. Conclusions In conclusion, this study aimed to identify patterns of DNAme associated with exposure to low SES, and whether the nature and strength of these associations varied depending on how SES was assessed (i.e. Socioeconomic Position, Financial Resources, or Socioeconomic Disadvantage). Overall, no meaningful associations of prenatal maternal stress on the placental DNA methylome in this study were observed after accounting for genetic ancestry variation, emphasizing that placental EWAS studies involving diverse cohorts should account for genetic variation in their analyses to avoid false findings. This study also underscores the challenges with elucidating mechanisms underlying SES linked outcomes, given the complex nature of correlated variables. It remains a question for further investigation as to whether epigenetic age acceleration is an adverse effect of exposure to prenatal maternal stress associated with SES, given its association with pregnancy complications (73–75), or alternatively, if placental epigenetic aging is a healthy adaption to complications that increase the risk of preterm delivery. Lastly, its important to note that DNAme is a relatively stable mark strongly associated with cell composition and genetic variation, but gene expression levels are influenced by many factors in addition to DNAme. Thus, this work does not exclude that there might be changes to other aspects of gene regulation (e.g. histone marks or small non-coding RNAs) in association with exposure to maternal stress and SES. Declarations Ethics approval and consent to participate The study protocol was approved by the Institutional Review Boards of Northwestern University (STU00206269) and NorthShore University Health System (EH17-006). Consent for publication Not applicable. Availability of data and materials The datasets generated and/or analyzed during the current study are available in the Gene Expression Omnibus under accession number GSE307289. Competing interests The authors declare that they have no competing interests. Acknowledgements We gratefully acknowledge the study participants who donated their placentas to the SPAH study. The research team would like to thank: Britney P. Smart, Janedelie Romero, Katherine Vause, Claire Fisher, Renee M. Odom, Chuhan Wu, Jane Drage, Megan Choi, Jasmin Flowers, Hee Moon, Veronica Passarelli, Adam Leigh, Lauren Hoffer, Shanti Gallivan, Tao Jiang, Lavisha Singh, the NorthShore University Department of Pathology and Laboratory Medicine, and Northwestern University’s Foundations of Health Research Center. We also thank the Northwestern University NUSeq Core Facilities for running of the Illumina EPIC arrays. Thank you to Hannah Illing for assistance in array processing/quality control, preparation of the Gene Expression Omnibus dataset and feedback on the manuscript. We gratefully acknowledge the work and staff of Endeavor Health and the NorthShore University HealthSystem for their assistance with patient recruitment and support of the study. Authors’ contributions GEM and AEB established the overall study design, project funding, cohort recruitment, and oversaw sample collection for the SPAH cohort. LME, AAF, and LSKD participated in the collection and management of the clinical data. MSP contributed project data management. EOB conducted all data processing/analysis and drafted the manuscript. WPR and AMI participated in data analysis. GEM, WPR and AMI participated in manuscript editing. All authors were involved in aspects of study design and read, revised, and approved the final manuscript. Funding This work was supported in part by the National Institutes of Health, National Institute on Minority Health and Health Disparities (R01MD011749). The funders had no role in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication. EOB received salary support from Canadian Health Research Grants GSK-171375 and PJT-169131); AMI received salary support from a CIHR CGS-D award and CIHR grant GSK-171375. WPR receives salary support through an investigatorship award from the BC Children’s Hospital Research Institute. References Blumenshine P, Egerter S, Barclay CJ, Cubbin C, Braveman PA. Socioeconomic disparities in adverse birth outcomes: a systematic review. Am J Prev Med. 2010 Sep;39(3):263–72. Nicholls-Dempsey L, Badeghiesh A, Baghlaf H, Dahan MH. How does high socioeconomic status affect maternal and neonatal pregnancy outcomes? A population-based study among American women. Eur J Obstet Gynecol Reprod Biol X. 2023 Oct 12;20:100248. Bushnik T, Yang S, Kaufman JS, Kramer MS, Wilkins R. Socioeconomic disparities in small-for-gestational-age birth and preterm birth. Health Rep. 2017 Nov 15;28(11):3–10. Joseph KS, Liston RM, Dodds L, Dahlgren L, Allen AC. Socioeconomic status and perinatal outcomes in a setting with universal access to essential health care services. CMAJ Can Med Assoc J. 2007 Sep 11;177(6):583–90. Thomson K, Moffat M, Arisa O, Jesurasa A, Richmond C, Odeniyi A, et al. Socioeconomic inequalities and adverse pregnancy outcomes in the UK and Republic of Ireland: a systematic review and meta-analysis. BMJ Open. 2021 Mar 1;11(3):e042753. Hertzman C, Boyce T. How experience gets under the skin to create gradients in developmental health. Annu Rev Public Health. 2010;31:329-347 3p following 347. Barch DM, Donohue MR, Elsayed NM, Gilbert K, Harms MP, Hennefield L, et al. Early Childhood Socioeconomic Status and Cognitive and Adaptive Outcomes at the Transition to Adulthood: The Mediating Role of Gray Matter Development Across Five Scan Waves. Biol Psychiatry Cogn Neurosci Neuroimaging. 2022 Jan 1;7(1):34–44. Schultz WM, Kelli HM, Lisko JC, Varghese T, Shen J, Sandesara P, et al. Socioeconomic Status and Cardiovascular Outcomes. Circulation. 2018 May 15;137(20):2166–78. Miller GE, Chen E, Parker KJ. Psychological Stress in Childhood and Susceptibility to the Chronic Diseases of Aging: Moving Towards a Model of Behavioral and Biological Mechanisms. Psychol Bull. 2011 Nov;137(6):959–97. Miller GE, Borders AE, Crockett AH, Ross KM, Qadir S, Keenan-Devlin L, et al. MATERNAL SOCIOECONOMIC DISADVANTAGE IS ASSOCIATED WITH TRANSCRIPTIONAL INDICATIONS OF GREATER IMMUNE ACTIVATION AND SLOWER TISSUE MATURATION IN PLACENTAL BIOPSIES AND NEWBORN CORD BLOOD. Brain Behav Immun. 2017 Aug;64:276–84. King S, Kildea S, Austin MP, Brunet A, Cobham VE, Dawson PA, et al. QF2011: a protocol to study the effects of the Queensland flood on pregnant women, their pregnancies, and their children’s early development. BMC Pregnancy Childbirth. 2015 May 6;15(1):109. Avishai-Eliner S, Brunson KL, Sandman CA, Baram TZ. Stressed-out, or in (utero)? Trends Neurosci. 2002 Oct;25(10):518–24. Matthews KA, Gallo LC. Psychological perspectives on pathways linking socioeconomic status and physical health. Annu Rev Psychol. 2011;62:501–30. Chen E, Miller GE. Socioeconomic Status and Health: Mediating and Moderating Factors. Annu Rev Clin Psychol. 2013 Mar 28;9(Volume 9, 2013):723–49. Knobel P, Hwang I, Castro E, Sheffield P, Holaday L, Shi L, et al. Socioeconomic and racial disparities in source-apportioned PM2.5 levels across urban areas in the contiguous US, 2010. Atmos Environ. 2023 Jun 15;303:119753. Burnett R, Chen H, Szyszkowicz M, Fann N, Hubbell B, Pope CA, et al. Global estimates of mortality associated with long-term exposure to outdoor fine particulate matter. Proc Natl Acad Sci U S A. 2018 Sep 18;115(38):9592–7. Adamkiewicz G, Zota AR, Fabian MP, Chahine T, Julien R, Spengler JD, et al. Moving Environmental Justice Indoors: Understanding Structural Influences on Residential Exposure Patterns in Low-Income Communities. Am J Public Health. 2011 Dec;101(Suppl 1):S238–45. Husted G, Cassels S, Ackert E, Sweeney S. Socio-demographic patterning of urinary mono-ethyl phthalate levels among children and adults in the U.S. (1999–2018). Environ Res Commun [Internet]. 2024 Aug 27;6(8). Available from: https://iopscience.iop.org/article/10.1088/2515-7620/ad6403 Stringhini S, Carmeli C, Jokela M, Avendaño M, Muennig P, Guida F, et al. Socioeconomic status and the 25 × 25 risk factors as determinants of premature mortality: a multicohort study and meta-analysis of 1·7 million men and women. Lancet Lond Engl. 2017 Mar 25;389(10075):1229–37. Bronson SL, Bale TL. The Placenta as a Mediator of Stress Effects on Neurodevelopmental Reprogramming. Neuropsychopharmacology. 2016 Jan;41(1):207–18. Hanson MA, Gluckman PD. Early developmental conditioning of later health and disease: physiology or pathophysiology? Physiol Rev. 2014 Oct;94(4):1027–76. Coe CL, Lubach GR. Vital and vulnerable functions of the primate placenta critical for infant health and brain development. Front Neuroendocrinol. 2014 Oct;35(4):439–46. Santos Jr HP, Bhattacharya A, Martin EM, Addo K, Psioda M, Smeester L, et al. Epigenome-wide DNA methylation in placentas from preterm infants: association with maternal socioeconomic status. Epigenetics. 2019 Aug 3;14(8):751–65. Appleton AA, Armstrong DA, Lesseur C, Lee J, Padbury JF, Lester BM, et al. Patterning in Placental 11-B Hydroxysteroid Dehydrogenase Methylation According to Prenatal Socioeconomic Adversity. PLOS ONE. 2013 Sep 5;8(9):e74691. Ghazi T, Naidoo P, Naidoo RN, Chuturgoon AA. Prenatal Air Pollution Exposure and Placental DNA Methylation Changes: Implications on Fetal Development and Future Disease Susceptibility. Cells. 2021 Nov 5;10(11):3025. Martin EM, Fry RC. Environmental Influences on the Epigenome: Exposure- Associated DNA Methylation in Human Populations. Annu Rev Public Health. 2024 Apr 1;39(1):309–33. Mortillo M, Marsit CJ. Select early-life environmental exposures and DNA methylation in the placenta. Curr Environ Health Rep. 2023 Mar;10(1):22–34. Bozack AK, Trasande L. Prenatal chemical exposures and the methylome: current evidence and opportunities for environmental epigenetics. Epigenomics. 16(23–24):1443–51. Austin MK, Chen E, Ross KM, McEwen LM, Maclsaac JL, Kobor MS, et al. Early-life socioeconomic disadvantage, not current, predicts accelerated epigenetic aging of monocytes. Psychoneuroendocrinology. 2018 Nov 1;97:131–4. Oblak L, van der Zaag J, Higgins-Chen AT, Levine ME, Boks MP. A systematic review of biological, social and environmental factors associated with epigenetic clock acceleration. Ageing Res Rev. 2021 Aug;69:101348. Yuan V, Hui D, Yin Y, Peñaherrera MS, Beristain AG, Robinson WP. Cell-specific characterization of the placental methylome. BMC Genomics. 2021 Jan 6;22(1):6. Bell CG, Gao F, Yuan W, Roos L, Acton RJ, Xia Y, et al. Obligatory and facilitative allelic variation in the DNA methylome within common disease-associated loci. Nat Commun. 2018 Jan 2;9:8. Villicaña S, Castillo-Fernandez J, Hannon E, Christiansen C, Tsai PC, Maddock J, et al. Genetic impacts on DNA methylation help elucidate regulatory genomic processes. Genome Biol. 2023 Jul 31;24:176. Do C, Lang CF, Lin J, Darbary H, Krupska I, Gaba A, et al. Mechanisms and Disease Associations of Haplotype-Dependent Allele-Specific DNA Methylation. Am J Hum Genet. 2016 May 5;98(5):934–55. Do C, Shearer A, Suzuki M, Terry MB, Gelernter J, Greally JM, et al. Genetic–epigenetic interactions in cis: a major focus in the post-GWAS era. Genome Biol. 2017 Jun 19;18:120. Fraser HB, Lam LL, Neumann SM, Kobor MS. Population-specificity of human DNA methylation. Genome Biol. 2012 Feb 9;13(2):R8. Patin E, Hasan M, Bergstedt J, Rouilly V, Libri V, Urrutia A, et al. Natural variation in the parameters of innate immune cells is preferentially driven by genetic factors. Nat Immunol. 2018 Mar;19(3):302–14. Galanter JM, Gignoux CR, Oh SS, Torgerson D, Pino-Yanes M, Thakur N, et al. Differential methylation between ethnic sub-groups reflects the effect of genetic ancestry and environmental exposures. eLife. 6:e20532. Henikoff S, Greally JM. Epigenetics, cellular memory and gene regulation. Curr Biol. 2016 Jul 25;26(14):R644–8. Lappalainen T, Greally JM. Associating cellular epigenetic models with human phenotypes. Nat Rev Genet. 2017 Jul;18(7):441–51. Lee Y, Choufani S, Weksberg R, Wilson SL, Yuan V, Burt A, et al. Placental epigenetic clocks: estimating gestational age using placental DNA methylation levels. Aging. 2019 Jun 24;11(12):4238–53. Lauren S. Keenan-Devlin, Alexa Freedman, Renee M. Odom-Konja, Michelle A. Chen, Tao Jiang, Linda M. Ernst, et al. Associations of objective, subjective, childhood, and neighborhood measures of socioeconomic status with adverse pregnancy outcomes. The National Statistics Socio-economic classification (NS-SEC) - Office for National Statistics [Internet]. [cited 2024 Aug 8]. Available from: https://www.ons.gov.uk/methodology/classificationsandstandards/otherclassifications/thenationalstatisticssocioeconomicclassificationnssecrebasedonsoc2010#using-the-derivation-tables Miller GE, Brody GH, Yu T, Chen E. A family-oriented psychosocial intervention reduces inflammation in low-SES African American youth. Proc Natl Acad Sci U S A. 2014 Aug 5;111(31):11287–92. 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 Oct 4;16(1):37. Pidsley R, Y Wong CC, Volta M, Lunnon K, Mill J, Schalkwyk LC. A data-driven approach to preprocessing Illumina 450K methylation array data. BMC Genomics. 2013 May 1;14(1):293. 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 Apr;41(7):e90. Zhou W, Laird PW, Shen H. Comprehensive characterization, annotation and innovative use of Infinium DNA methylation BeadChip probes. Nucleic Acids Res. 2017 Feb 28;45(4):e22. Murat K, Grüning B, Poterlowicz PW, Westgate G, Tobin DJ, Poterlowicz K. Ewastools: Infinium Human Methylation BeadChip pipeline for population epigenetics integrated into Galaxy. GigaScience. 2020 May 1;9(5):giaa049. Daenekas B, Pérez E, Boniolo F, Stefan S, Benfatto S, Sill M, et al. Conumee 2.0: enhanced copy-number variation analysis from DNA methylation arrays for humans and mice. Bioinformatics. 2024 Feb 1;40(2):btae029. Inkster AM, Illing HJ, Robinson WP. Some lessons learned from genomic and epigenomic studies of the placenta. Placenta [Internet]. 2025 Apr 17 [cited 2025 May 18]; Available from: https://www.sciencedirect.com/science/article/pii/S0143400425001237 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 Apr 20;43(7):e47. StoreyLab/biobroom [Internet]. Storey Lab; 2023 [cited 2024 Jan 10]. Available from: https://github.com/StoreyLab/biobroom Cheung WA, Shao X, Morin A, Siroux V, Kwan T, Ge B, et al. Functional variation in allelic methylomes underscores a strong genetic contribution and reveals novel epigenetic alterations in the human epigenome. Genome Biol. 2017 Mar 10;18:50. Bell JT, Pai AA, Pickrell JK, Gaffney DJ, Pique-Regi R, Degner JF, et al. DNA methylation patterns associate with genetic and gene expression variation in HapMap cell lines. Genome Biol. 2011;12(1):R10. 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 Aug 9;12(1):51. Konwar C, Del Gobbo G, Yuan V, Robinson WP. Considerations when processing and interpreting genomics data of the placenta. Placenta. 2019 Sep 1;84:57–62. Bryc K, Durand EY, Macpherson JM, Reich D, Mountain JL. The Genetic Ancestry of African Americans, Latinos, and European Americans across the United States. Am J Hum Genet. 2015 Jan 8;96(1):37–53. Inkster AM, Fernández-Boyano I, Robinson WP. Sex Differences Are Here to Stay: Relevance to Prenatal Care. J Clin Med. 2021 Jul 5;10(13):3000. Braun AE, Mitchel OR, Gonzalez TL, Sun T, Flowers AE, Pisarska MD, et al. Sex at the interface: the origin and impact of sex differences in the developing human placenta. Biol Sex Differ. 2022 Sep 16;13(1):50. Sutherland S, Brunwasser SM. Sex Differences in Vulnerability to Prenatal Stress: a Review of the Recent Literature. Curr Psychiatry Rep. 2018 Sep 1;20(11):102. 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 Jan 7;16(1):1. Bale TL. The placenta and neurodevelopment: sex differences in prenatal vulnerability. Dialogues Clin Neurosci. 2016 Dec;18(4):459–64. Campbell KA, Colacino JA, Puttabyatappa M, Dou JF, Elkin ER, Hammoud SS, et al. Placental cell type deconvolution reveals that cell proportions drive preeclampsia gene expression differences. Commun Biol. 2023 Mar 13;6(1):1–15. Fernández-Boyano I, Inkster AM, Yuan V, Robinson WP. eoPred: predicting the placental phenotype of early-onset preeclampsia using public DNA methylation data. Front Genet. 2023;14:1248088. Lemaire M, Smith KW, Wilson SL. Examining Fetal Sex-Specific Placental DNA Methylation Intensities and Variability Post In Vitro Fertilization [Internet]. bioRxiv; 2024 [cited 2024 Sep 19]. p. 2024.08.08.604307. Available from: https://www.biorxiv.org/content/10.1101/2024.08.08.604307v1 Inkster AM, Konwar C, Peñaherrera MS, Brain U, Khan A, Price EM, et al. Profiling placental DNA methylation associated with maternal SSRI treatment during pregnancy. Sci Rep. 2022 Dec 30;12(1):22576. Christians JK, Chow NA. Are there sex differences in fetal growth strategies and in the long-term effects of pregnancy complications on cognitive functioning? J Dev Orig Health Dis. 2022 Dec;13(6):766–78. Jones MJ, Goodman SJ, Kobor MS. DNA methylation and healthy human aging. Aging Cell. 2015;14(6):924–32. Horvath S. DNA methylation age of human tissues and cell types. Genome Biol. 2013 Dec 10;14(10):3156. Horvath S, Raj K. DNA methylation-based biomarkers and the epigenetic clock theory of ageing. Nat Rev Genet. 2018 Jun;19(6):371–84. Husquin LT, Rotival M, Fagny M, Quach H, Zidane N, McEwen LM, et al. Exploring the genetic basis of human population differences in DNA methylation and their causal impact on immune gene regulation. Genome Biol. 2018 Dec 18;19(1):222. Hatton AA, Cheng FF, Lin T, Shen RJ, Chen J, Zheng Z, et al. Genetic control of DNA methylation is largely shared across European and East Asian populations. Nat Commun. 2024 Mar 28;15(1):2713. Banda Y, Kvale MN, Hoffmann TJ, Hesselson SE, Ranatunga D, Tang H, et al. Characterizing Race/Ethnicity and Genetic Ancestry for 100,000 Subjects in the Genetic Epidemiology Research on Adult Health and Aging (GERA) Cohort. Genetics. 2015 Aug;200(4):1285–95. Wang J, Zhang Z, Lu Z, Mancuso N, Gazal S. Genes with differential expression across ancestries are enriched in ancestry-specific disease effects likely due to gene-by-environment interactions. Am J Hum Genet. 2024 Oct 3;111(10):2117–28. Sonderlund AL, Charifson M, Schoenthaler A, Carson T, Williams NJ. Racialized economic segregation and health outcomes: A systematic review of studies that use the Index of Concentration at the Extremes for race, income, and their interaction. PLOS ONE. 2022 Jan 28;17(1):e0262962. Carpenter T, Grecian SM, Reynolds RM. Sex differences in early-life programming of the hypothalamic–pituitary–adrenal axis in humans suggest increased vulnerability in females: a systematic review. J Dev Orig Health Dis. 2017 Apr;8(2):244–55. Clark J, Bulka CM, Martin CL, Roell K, Santos HP, O’Shea TM, et al. Placental epigenetic gestational aging in relation to maternal sociodemographic factors and smoking among infants born extremely preterm: a descriptive study. Epigenetics. 2022 Dec;17(13):2389–403. Van Buren E, Azzara D, Rangel-Moreno J, Garcia-Hernandez M de la L, Murphy SP, Cohen ED, et al. Single-cell RNA sequencing reveals placental response under environmental stress. Nat Commun. 2024 Aug 2;15(1):6549. Wang Q, Li J, Wang S, Deng Q, An Y, Xing Y, et al. Single-cell transcriptional profiling reveals cellular and molecular divergence in human maternal-fetal interface. Sci Rep. 2022 Jun 28;12(1):10892. Shannon MJ, Baltayeva J, Castellana B, Wächter J, McNeill GL, Yoon JS, et al. Cell trajectory modeling identifies a primitive trophoblast state defined by BCAM enrichment. Dev Camb Engl. 2022 Jan 1;149(1):dev199840. Additional Declarations No competing interests reported. Supplementary Files AdditionalFile1.docx Cite Share Download PDF Status: Published Journal Publication published 24 Apr, 2026 Read the published version in Clinical Epigenetics → Version 1 posted Editorial decision: Revision requested 16 Feb, 2026 Reviews received at journal 06 Feb, 2026 Reviews received at journal 02 Feb, 2026 Reviews received at journal 29 Jan, 2026 Reviews received at journal 28 Jan, 2026 Reviewers agreed at journal 21 Jan, 2026 Reviewers agreed at journal 20 Jan, 2026 Reviewers agreed at journal 20 Jan, 2026 Reviewers agreed at journal 19 Jan, 2026 Reviewers agreed at journal 19 Jan, 2026 Reviews received at journal 29 Nov, 2025 Reviewers agreed at journal 19 Nov, 2025 Reviewers invited by journal 18 Nov, 2025 Editor assigned by journal 09 Sep, 2025 Submission checks completed at journal 08 Sep, 2025 First submitted to journal 05 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7546517","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":515175941,"identity":"24e8d081-c0d3-4c4e-9668-c9d5a151f987","order_by":0,"name":"Ella O. Beraldo","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Ella","middleName":"O.","lastName":"Beraldo","suffix":""},{"id":515175942,"identity":"ee1302eb-c5ae-4be6-8097-c7f17f15f4e5","order_by":1,"name":"Ann E. Borders","email":"","orcid":"","institution":"University of Chicago","correspondingAuthor":false,"prefix":"","firstName":"Ann","middleName":"E.","lastName":"Borders","suffix":""},{"id":515175943,"identity":"2e1a19d9-2ca3-4be7-8aac-3015d9f5d149","order_by":2,"name":"Amy M Inkster","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Amy","middleName":"M","lastName":"Inkster","suffix":""},{"id":515175944,"identity":"d423c795-62d7-4936-9836-b7c8cc27dad4","order_by":3,"name":"Linda M Ernst","email":"","orcid":"","institution":"University of Chicago","correspondingAuthor":false,"prefix":"","firstName":"Linda","middleName":"M","lastName":"Ernst","suffix":""},{"id":515175946,"identity":"82b2cc6c-4d0a-4d35-9c47-98d02c4cb3b6","order_by":4,"name":"Alexa A. Freedman","email":"","orcid":"","institution":"Northwestern University","correspondingAuthor":false,"prefix":"","firstName":"Alexa","middleName":"A.","lastName":"Freedman","suffix":""},{"id":515175948,"identity":"0e56c6ae-ca5c-4185-ac81-a987f3580f6b","order_by":5,"name":"Jungwon Kim","email":"","orcid":"","institution":"Northwestern University","correspondingAuthor":false,"prefix":"","firstName":"Jungwon","middleName":"","lastName":"Kim","suffix":""},{"id":515175950,"identity":"326103c6-4986-4924-afe5-5c63ca99ca07","order_by":6,"name":"Lauren S. Keenan-Devlin","email":"","orcid":"","institution":"University of Chicago","correspondingAuthor":false,"prefix":"","firstName":"Lauren","middleName":"S.","lastName":"Keenan-Devlin","suffix":""},{"id":515175952,"identity":"b9cc3621-8821-4840-866b-8d2fa3611503","order_by":7,"name":"Maria S. Peñaherrera","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"S.","lastName":"Peñaherrera","suffix":""},{"id":515175954,"identity":"f784f89d-e081-48c1-9141-7662e808879f","order_by":8,"name":"Wendy Robinson","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAvUlEQVRIiWNgGAWjYHACNoYEEMXew8DA2ECSFp4zpGgBA4kcIrUY3D587MHDHXVy5pJvD374ueMeA3/7AQJazqWlGySeOWxsOTsvWbL3TDGDxJkE/FrMzvCYSSS2HUjccDvHjJmxLYHBgIE4LXWJG26egWrhf0CUFubEDTd4oFokCNhif4YN6Je2w8YGZ0B+aUvgkbhBwBbJHuZjD3+21ckZHD8LDLG2BDn+fgK2YAAeEtWPglEwCkbBKMAGAKJMQhGK0YSDAAAAAElFTkSuQmCC","orcid":"","institution":"University of British Columbia","correspondingAuthor":true,"prefix":"","firstName":"Wendy","middleName":"","lastName":"Robinson","suffix":""},{"id":515175955,"identity":"12cafaa6-74cc-4694-b247-17701b05f2d5","order_by":9,"name":"Gregory E. Miller","email":"","orcid":"","institution":"University of Chicago","correspondingAuthor":false,"prefix":"","firstName":"Gregory","middleName":"E.","lastName":"Miller","suffix":""}],"badges":[],"createdAt":"2025-09-05 18:23:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7546517/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7546517/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s13148-026-02136-7","type":"published","date":"2026-04-24T15:58:48+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":96242318,"identity":"cae05584-60d6-4439-8ad4-b04126de8acd","added_by":"auto","created_at":"2025-11-19 07:12:35","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1378079,"visible":true,"origin":"","legend":"","description":"","filename":"SPAHManuscriptSept2025..docx","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/f0c6079a26eff24f4e1872c4.docx"},{"id":96242217,"identity":"0d6011c6-e62d-4afb-b9db-2df134f15078","added_by":"auto","created_at":"2025-11-19 07:12:21","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12809,"visible":true,"origin":"","legend":"","description":"","filename":"28dfbc9eb73647febe6522c817a249d8.json","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/2137bcb15979701cd1cbcd0f.json"},{"id":96243060,"identity":"650663f8-859d-4f74-839e-897acc36177a","added_by":"auto","created_at":"2025-11-19 07:15:22","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2163406,"visible":true,"origin":"","legend":"","description":"","filename":"AdditionalFile1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/fc1e77dcba32a3de5d924aa4.docx"},{"id":95885756,"identity":"8b9d4321-caa9-4dca-8b94-874f0b206aea","added_by":"auto","created_at":"2025-11-14 04:41:37","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":123303,"visible":true,"origin":"","legend":"","description":"","filename":"28dfbc9eb73647febe6522c817a249d81enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/37e999f0eb049086892a8092.xml"},{"id":95885747,"identity":"05bae1cf-b07c-4d14-a447-d426922aef5a","added_by":"auto","created_at":"2025-11-14 04:41:37","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":226536,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/60974e33407119f7a9ea88ff.png"},{"id":95885758,"identity":"8ae77e5c-e750-4814-ab48-968d79906abb","added_by":"auto","created_at":"2025-11-14 04:41:37","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":534303,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/6c588f3a36bf7ff15d100a68.png"},{"id":96241781,"identity":"4d6bff34-27b9-410a-b8fe-c30153f2ab13","added_by":"auto","created_at":"2025-11-19 07:11:22","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":156814,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/8b510e6a94e99eebfbbde3e4.png"},{"id":95885751,"identity":"5f677b70-2415-410d-b4b4-79983bf17c7c","added_by":"auto","created_at":"2025-11-14 04:41:37","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":232241,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/cbff4950a925e85a5fd1cfa8.png"},{"id":96241582,"identity":"1c089164-b927-4d15-9010-a193a711684a","added_by":"auto","created_at":"2025-11-19 07:11:02","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39763,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/f13c4f1f2f330b0093044e3c.png"},{"id":96242305,"identity":"b40cdc59-0816-492a-889a-07669100552c","added_by":"auto","created_at":"2025-11-19 07:12:35","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":196207,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/600becf29105ae616034b6c3.png"},{"id":96242046,"identity":"6aa954ce-c4b5-4e18-8017-a2651c0f4598","added_by":"auto","created_at":"2025-11-19 07:11:53","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":153085,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/2ed0f151976fb66fae23f8d8.png"},{"id":96242401,"identity":"724ee5d0-b048-4da7-80df-3d635a27f937","added_by":"auto","created_at":"2025-11-19 07:12:54","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":226009,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/f3bdc2bd8724d74becca7ce4.png"},{"id":96241639,"identity":"cb1b3575-b0db-4ba8-953a-92b6f0d1826c","added_by":"auto","created_at":"2025-11-19 07:11:10","extension":"xml","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":120619,"visible":true,"origin":"","legend":"","description":"","filename":"28dfbc9eb73647febe6522c817a249d81structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/cf5290bd15fb93c5aa17cb5a.xml"},{"id":95885760,"identity":"a157cf2e-cdf6-4452-8ecb-b5a1d8e9cf94","added_by":"auto","created_at":"2025-11-14 04:41:37","extension":"html","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":131645,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/b1a1a45cdb3ceb927ebdb119.html"},{"id":95885742,"identity":"1a3adc11-be9f-4a7e-8ece-6c5fdd0d610e","added_by":"auto","created_at":"2025-11-14 04:41:37","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":122990,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVolcano plots for differential DNAme in association with SES exposure.\u003c/strong\u003e Llinear models were corrected for gestational age and PlaNET estimated ancestry (Model B). False discovery rate (FDR) is depicted along the Y axis. More significant (lower FDR) values are shown at the top of the plot. Vertical dashed lines outlines |Δβ| = 0.02 (inner) and 0.05 (outer), and horizontal dashed line indicates FDR = 0.05. Gene names label the CpGs that pass FDR \u0026lt; 0.05, |∆β| \u0026gt; 0.05 cutoff. (A) Volcano plot for Model B run with Social Position composite score as the SES exposure variable. (B) Volcano plot for Model B run with Resources composite score as the SES exposure. (C) Volcano plot for Model B run with Disadvantage composite score as the SES exposure.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/122024f27bdd976cf65d5679.png"},{"id":95885743,"identity":"673bedb9-b6b8-4e73-be09-19387f735cdc","added_by":"auto","created_at":"2025-11-14 04:41:37","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":59227,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDNA methylation with Social Position SES.\u003c/strong\u003e Associations of methylation at the significant CpGs sites (FDR \u0026lt; 0.05, |∆β| \u0026gt; 0.05) identified by linear modelling in association with Social Position SES in whole cohort autosome model. Data distribution is shown on the right of the box plots, with jittered raw data depicted on the left. Beta values are depicted along the y axis, PlaNET predicted ancestry groups are displayed along the x axis. Significant relationships (p value \u0026lt; 0.05) are shown where present.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/88fbe999c0ed9fbe1ff7a551.png"},{"id":95885744,"identity":"ae3e0d8d-4e8a-4349-ad08-dd513cfaa6e6","added_by":"auto","created_at":"2025-11-14 04:41:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":79099,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTrophoblast cell composition with SES variables.\u003c/strong\u003e Association of predicted cytotrophoblast to syncytiotrophoblast (cyt:syn) ratio with (A) gestational age at delivery, (B) Socioeconomic Position, (C) Resources, and (D) Disadvantage composite scores. Spearman’s correlation and significance values (A-C) and statistically significant comparisons (Wilcoxon Rank Sum Test p\u0026lt;0.05) (D) are indicated if present.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/0ba4a25ce4e99917ab4b5f3f.png"},{"id":95885745,"identity":"6384732d-7023-4264-bd1c-e7050a46ceaf","added_by":"auto","created_at":"2025-11-14 04:41:37","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":154085,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePlacental epigenetic age acceleration with SES exposure.\u003c/strong\u003e Relationships between PlaNET-estimated extrinsic (A,C,E) and intrinsic (B,D,F) epigenetic age acceleration and Socioeconomic Position (A,B), Resources (C,D), and Disadvantage (E,F) composite scores. Spearman’s correlation and significance values (A-D) and Wilcoxon Rank Sum Test (p\u0026lt;0.05) values (E-F) are shown above respective plots if significant.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/ab8af32e2a09bb0d580d38d0.png"},{"id":107929230,"identity":"dd0bf4d4-8d6d-46f6-9481-688a39f7cb70","added_by":"auto","created_at":"2026-04-27 16:14:14","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":660122,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/36f45451-4b16-4b4a-9882-45eea52c65d8.pdf"},{"id":96242666,"identity":"7401f846-532a-4909-a961-a9dec5bbae73","added_by":"auto","created_at":"2025-11-19 07:13:55","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":2163406,"visible":true,"origin":"","legend":"","description":"","filename":"AdditionalFile1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7546517/v1/f44a097acfe44cb70ea0072b.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"DNA Methylation in the Placenta and Maternal Socioeconomic Status: The SPAH Study","fulltext":[{"header":"Background","content":"\u003cp\u003eBirth outcomes vary by socioeconomic status (SES), with low SES linked to higher rates of fetal growth restriction (FGR), preterm birth (PTB), and preeclampsia (1\u0026ndash;5). These disparities in adverse outcomes continue post birth, as children born into low SES households have increased rates of mental health conditions and morbidity and mortality from chronic disease (6\u0026ndash;9). What remains unclear is how SES influences biological processes that are proximally involved in brain maturation, health problems, and other developmental outcomes that are patterned by SES.\u003c/p\u003e\u003cp\u003eIt has been postulated that socioeconomic disadvantage becomes biologically embedded during sensitive periods of gestation, specifically through modulation of epigenetic processes in the placenta (10). The placenta is thought to be implicated in this process as it functions as a barrier protecting the fetus from maternal exposures and facilitates maternal-fetal exchange during gestation (10\u0026ndash;12). Individuals living in low-SES conditions are disproportionately exposed to a variety of psychological stressors (e.g., material hardship, stigmatization, mistreatment, job instability, neighborhood violence) and environmental pollutants (e.g., fine particulate matter, volatile organic compounds, environmental tobacco smoke, microplastics and other harmful chemicals such as phthalates) (13\u0026ndash;18). Such individuals exposed both early and later in life also have a higher risk for chronic diseases such as hypertension and diabetes (19).\u003c/p\u003e\u003cp\u003eIn both rodent and primate models approximating human socioeconomic disadvantage during pregnancy, animals show dysregulation of biological functions such as psychological stress response, nutrient imbalance, and glucocorticoid excess (20\u0026ndash;22). In humans, socioeconomic adversity has been associated with differential DNAme or transcription of genes involved in placental function, cortisol hormone signalling, immune activation and fetal maturation (10,23,24). However, the association between SES, or specific aspects of SES, and placental functioning remains unclear, as SES is a multidimensional construct. Conventional indicators of SES such as household income, education, and occupational prestige are inherently inter-related, but can diverge substantially for some individuals. For example, some jobs that command a high salary, like trades, do not require a university education; others that do involve high levels of education, such as academic careers in the humanities, do not necessarily command high salaries. Different elements of SES may also affect health through distinct pathways, for example, income provides an indication of the material resources individuals have at their disposal, whereas education and occupation are windows into social status, health literacy and social networks (14).\u003c/p\u003e\u003cp\u003eGenome-wide DNAme studies are commonly employed for exploring links between environmental exposures and health outcomes, and altered placental DNAme has been reported in association with some societal and environmental exposures, such as air pollution and other environmental chemicals like bisphenol A (BPA) (25\u0026ndash;28). Furthermore, epigenetic age acceleration, the difference between age predicted by DNAme clocks and chronological age has been negatively associated with SES in adult tissues (29,30). However, the influence of SES on the placental DNA methylome and epigenetic age has not yet been well-characterized. Studying the placenta, however, presents some challenges as it is a heterogeneous tissue composed of multiple cell types of varying developmental origin, each with a distinct DNAme profile (31). Furthermore, genetic variation contributes to 20\u0026ndash;70% of DNAme variation (32\u0026ndash;35) and thus DNAme can vary between populations of different ancestry due to differences in allele frequencies (36\u0026ndash;38). Thus, accounting for cellular and genetic variation is necessary before it can be concluded that changes to DNAme are attributable to the exposure of interest (39,40).\u003c/p\u003e\u003cp\u003eIn this study, we analyzed Illumina Infinium MethylationEPIC v1.0 DNAme array data obtained from 493 placentas derived from a sociodemographically diverse cohort of pregnancies recruited from Chicago, Illinois. We aimed to identify patterns of DNAme associated with SES, and to evaluate whether the nature and strength of these associations varied depending on how SES is assessed and/or fetal sex. We further examined \u0026ldquo;epivariables\u0026rdquo; inferred from the DNAme data directly, including estimated cell composition (31) and epigenetic age acceleration (41). As stress effects on gestational biology can be sex-specific (20), we hypothesized that pregnancy in individuals exposed to lower SES may lead to altered placental DNAme that is associated with other SES-related variables such as PTB and/or FGR, and that these associations may differ by fetal sex.\u003c/p\u003e"},{"header":"Methods \u0026 Materials","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eCohort\u003c/h2\u003e\u003cp\u003eThe data presented here come from the Stress, Pregnancy, and Health Study (SPAH) study. The study protocol was approved by the Institutional Review Boards of Northwestern University (STU00206269) and Endeavor Health (EH17-006). Written informed consent was obtained from each participant. This prospective observational cohort study was designed to evaluate multiple measures of socioeconomic status in relation to pregnancy disparities. SPAH enrolled 605 pregnant individuals mid-pregnancy, recruited from four clinical sites across the Chicago metropolitan area (Erie Family Health Center, The Center for Maternal and Fetal Health at Evanston Hospital, The NorthShore Community Health Center at Evanston Hospital, and NorthShore Lincolnwood Medical Group Office) between March 2018 and September 2022. Eligible participants were recruited at prenatal clinical sites before 25 weeks gestation to complete a series of questionnaires during second and third trimesters. Pregnant individuals were eligible for the study if they were 18 years or older, carrying a singleton pregnancy, and English speaking. Individuals were excluded from the study if there were fetal congenital anomalies or known chromosomal abnormalities in the pregnancy. Of the 605 pregnant women recruited for SPAH, biopsies for placental DNAme were obtained from 509 of these pregnancies. Samples from 16 placentas were removed from the DNAme cohort during data processing for failing quality checks (as described further below in section 2.5.) yielding 493 cases for analysis.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eMeasures of SES Disparities\u003c/h3\u003e\n\u003cp\u003eSES and demographic data were collected between 20\u0026ndash;26 weeks of gestation via interview and survey. The SES interview collected information on the pregnant individual and their financial resources, such as household income and savings, educational attainment, occupation, governmental assistance, and demographics.\u003c/p\u003e\u003cp\u003eThree composites were created to assess various dimensions of the participants\u0026rsquo; SES (42). To capture aspects of SES related to \u003cb\u003esocioeconomic position\u003c/b\u003e (i.e., standing in society based on prestige and power), we calculated a composite based on standardized levels of highest household educational attainment and occupational prestige, using methods from The National Statistics Socio-Economic Classification (NS-SEC) (2010) (43) to code for occupational prestige. This composite\u0026rsquo;s emphasis on educational attainment and occupational prestige mirrors the Hollingshead Index, the most widely used indicator of SES in the social sciences, though we exclude that measure\u0026rsquo;s consideration of participant sex and marital status, which are less relevant in contemporary society.\u003c/p\u003e\u003cp\u003eTo capture \u003cb\u003efinancial resources\u003c/b\u003e, a separate composite was calculated as a continuous variable including standardized levels of household income-to-poverty ratio (IPR), total savings and assets, and household savings relative to cost of living (i.e., \u0026ldquo;if you lost all your current source(s) of household income (your paycheck, public assistance, or other forms of income), how long could you continue to live at your current address and standard of living?\u0026rdquo;). A \u003cb\u003esocioeconomic disadvantage composite\u003c/b\u003e was computed to summarize overall disadvantage, incorporating select features of both socioeconomic position (i.e., highest household educational attainment) and financial resources (i.e., income below federal poverty threshold and whether savings was less than 2 months of living expenses). In addition, the socioeconomic disadvantage composite included other features less commonly included in SES research, but which are still notable indicators of greater disadvantage (or lower SES) (i.e., recipient of TANF, WIC, SNAP, CHIP, SSI, or Medicaid). The composite index here as a count score also captures a different distribution compared to our continuous measures. The composite was computed as a count score calculated on a 0\u0026ndash;5 scale, with one point for each of the following: household income less than twice the federal poverty threshold (IPR\u0026thinsp;\u0026lt;\u0026thinsp;2.0), savings less than 2 months of living expenses, highest education in the household less than a two-year college degree, receipt of government assistance (including TANF, WIC, SNAP, CHIP, SSI, or Medicaid), and self or partner currently unemployed. This socioeconomic disadvantage composite has been utilized in previous biological studies of SES (10,44).\u003c/p\u003e\n\u003ch3\u003ePlacental Collection\u003c/h3\u003e\n\u003cp\u003ePlacentas were sampled from 509 of the 605 participants. The vast majority of missing samples were from complicated deliveries, where staff could not obtain specimens within that timeframe. At the time of delivery, research staff or obstetric providers obtained 0.4 cm\u003csup\u003e3\u003c/sup\u003e chorionic villous biopsies from three separate cotyledons from the fetal-facing side of each placenta to minimize spatial variability in DNA methylation. BiopsiesAll biopsies were collected from 509 obtained within 6 hours of the 605 participants delivery, and 59% were obtained within 1 hour of delivery (average time of 1.76 hours from delivery). Samples from an additional 16 placentas were removed for failing quality checks (as described further below in section 2.5.) yielding 493 cases for analysis.\u003c/p\u003e\u003cp\u003eThe biopsies were stored at -80\u0026deg;C until the end of the study, at which time high-quality DNA was extracted from using the PerkinElmer Chemagic 360 System at the Northwestern University NUSeq Core Facility. The extraction process is based on the PerkinElmer Chemagen M-PVA Magnetic Bead technology (PerkinElmer, Waltham, Massachusetts). Briefly, the collected placental tissue samples were first lysed in the presence of protease using gentleMACS (Miltenyi-Biotec, Bergisch Gladbach, Germany). Then the chemagen M-PVA magnetic beads were added to bind DNA from the lysates, followed by several rounds of washing before elution of DNA from the beads. The extracted DNA was then checked for quality using NanoDrop, and quantified using Qubit. The purified DNA samples were stored at -20\u003csup\u003eo\u003c/sup\u003eC.\u003c/p\u003e\n\u003ch3\u003eDNA Methylation Arrays\u003c/h3\u003e\n\u003cp\u003eDNAme was assayed at the NUSeq Core Facility using the Infinium Human MethylationEPIC Beadchip v1.0 array (Illumina, Inc. CA, USA), which targets over 850,000 CpG sites. Samples were randomly plated on each chip. A 500 ng DNA sample was used to perform bisulfite conversion followed by Illumina\u0026rsquo;s protocol for DNA methylation profiling. BeadChips were scanned with an Illumina iScan instrument.\u003c/p\u003e\n\u003ch3\u003eDNA Methylation Processing and Quality Control\u003c/h3\u003e\n\u003cp\u003eDNAme data (IDAT files) were read into R v 4.2.2 for processing according to a previously published pipeline relying on the minfi, ewastools, and conumee R packages (45). In brief, epiphenotyping variables for genetic ancestry, gestational age (GA), and cell composition were estimated from the DNAme data itself using the PlaNET R package, as previously described (45). The raw data was normalized for analysis using the \u003cem\u003edasen noob\u003c/em\u003e combined normalization method (46,47). After normalization, we excluded poor-quality probes (bead count\u0026thinsp;\u0026lt;\u0026thinsp;3 or detection P value\u0026thinsp;\u0026gt;\u0026thinsp;0.01 in \u0026gt;\u0026thinsp;5% of samples, n\u0026thinsp;=\u0026thinsp;12,737), as well as previously identified cross-hybridizing probes (n\u0026thinsp;=\u0026thinsp;103,376) from our dataset (48). Of the 509 unique placental samples run on the array, 16 samples failed checks and were removed Two samples were removed for failing probe quality checks (\u0026gt;\u0026thinsp;1% of array probes, failed detection P/bead count). Another 8 samples failed multiple measures: they were identified to have much lower inter-sample correlation values in the whole cohort, were flagged for probable maternal contamination identified using the \u003cspan type=\"ItalicUnderline\" class=\"ItalicUnderline\" name=\"Emphasis\"\u003eewastools\u003c/span\u003e R package (49), and separated from the rest of the dataset on PC1 and PC2 during principal component analysis. In 5, cases clinical reported fetal sex did not match the sex chromosome complement of the DNAme sample, inferred using X and Y chromosome probes as previously described (45). Finally, one sample appeared to have a mosaic trisomy 7 based on aneuploidy detection using the R package \u003cem\u003econumee 2.0\u003c/em\u003e (50) and was excluded, as trisomy can have profound effects on DNAme. After extensive data processing and quality control steps, a total of 748,484 probes (n\u0026thinsp;=\u0026thinsp;732,102 autosomal; n\u0026thinsp;=\u0026thinsp;16,382 chrX; n\u0026thinsp;=\u0026thinsp;272 chrY) in 493 samples remained for analysis.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eEpivariable estimation\u003c/h2\u003e\u003cdiv id=\"Sec9\" class=\"Section3\"\u003e\u003ch2\u003eGenetic Ancestry\u003c/h2\u003e\u003cp\u003eAncestry of the placenta reflects both the maternal and paternal genetic contribution and is measured on a continuum relative to multiple reference populations. PlaNET is a tool that estimates ancestry probabilities from the DNAme itself and are provided as continuous variables along three major axes of population variation relative to African, European, and East Asian populations. Although other major populations such as South Asian and Amerindian ancestry cannot be captured with this DNAme based measure, correcting for PlaNET ancestry variables can improve reproducibility in EWAS studies (45,51). Although there was a relationship between maternal self-reported race and placental ancestry estimates in the SPAH cohort, these are distinct phenomena and there is considerable variation in placental ancestry within maternal racial groups (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eC), as well as within both Hispanic and non-Hispanic maternal ethnicity (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eD).\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\n\u003ch3\u003eCell Composition\u003c/h3\u003e\n\u003cp\u003eCell composition is a major driver of placental DNAme (31), as different cell types can have markedly different epigenetic profiles and can also vary greatly between datasets due to systematic sampling techniques. To gain a better understanding of the major cell types influencing DNAme in the SPAH cohort, we used PlaNET to estimate the composition of 6 major placental cell types across all sample in the SPAH cohort (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eB). The overall cell composition of the SPAH cohort showed high levels of predicted syncytiotrophoblast (average\u0026thinsp;\u0026gt;\u0026thinsp;0.75%) (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eB), with several samples being estimated as 100% syncytiotrophoblast, likely reflecting sampling from the tips of floating villi, avoiding any vessels.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eGestational Age and Epigenetic Age Acceleration Estimates\u003c/h2\u003e\u003cp\u003eGestational age (GA) can be estimated using several placental epigenetic clocks. We chose the control placental placental epigenetic clock (CPC) published by Lee et al. (2019) (41) as it was trained on normative pregnancies without known pathologies, and the effect of SES on placental epigenetic age acceleration remains unknown (41). Clinical GA correlated well (R\u0026thinsp;=\u0026thinsp;0.6, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) with epigenetic age estimates (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eA).\u003c/p\u003e\u003cp\u003eExtrinsic epigenetic age acceleration was calculated by taking the residuals of a linear regression model with CPC predicted epigenetic age as the dependent variable and chronological GA as the independent variable. Intrinsic age acceleration was calculated in the same manner but also included PlaNET-estimated cell proportions for 6 cell typesas covariates to account for cell composition. The association between epigenetic age acceleration and the Socioeconomic Position and Financial Resources composites was evaluated using linear models, both in whole cohort and sex-stratified analyses. Wilcoxon Rank Sum tests were used to evaluate whether epigenetic age acceleration varied by Disadvantage score, again both in whole cohort and sex-stratified analyses.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eLinear Modelling\u003c/h2\u003e\u003cp\u003eDNAme data at all filtered autosomal CpGs (n\u0026thinsp;=\u0026thinsp;732,102) were converted to M values prior to linear modeling, to test for DNAme differences across the (i) Socioeconomic Position, (ii) Financial Resources and (iii) Disadvantage Composite scores. Three separate linear models were run with the three SES scores as the primary variables of interest in each. Models were run using the limma R package (52,53), and a False Discovery Rate (FDR) of 0.05 was used to establish statistical significance, while a biological cutoff |Δβ| \u0026ge; 0.05 was used to reduce false positive results.. We also report results also using a less stringent effect size cut-off of |Δβ| \u0026ge; 0.02 which was established as the likely limit of technical detection in this data based on the maximum standard error across all CpGs for all samples (37). In the 493 placentas analyzed missing values for maternal BMI (n\u0026thinsp;=\u0026thinsp;23) were imputed to the median (31.06), while missing data for mode of conception (n\u0026thinsp;=\u0026thinsp;14), and marijuana use during pregnancy (n\u0026thinsp;=\u0026thinsp;13) were imputed to: 06kg/m\u003csup\u003e2\u003c/sup\u003e for BMI, \u0026ldquo;non-assisted\u0026rdquo; for mode of conception and \u0026ldquo;no use during pregnancy\u0026rdquo; for marijuana\u0026rdquo;, respectively, as both aligned with the overwhelming majority of the cohort.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eCohort characteristics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTable 1 displays maternal demographic and clinical characteristics of the 493 placentas analyzed. Among the participants, 21.7% had a household education level of high school diploma or less. 20.5% were low income or poor according to the income to poverty ratio. In terms of racial identity, the majority of participants self-identified as \u0026lsquo;white\u0026rsquo; (n=318, 64.5%), followed by \u0026lsquo;Black\u0026rsquo; (n=87, 17.6%) and \u0026lsquo;Asian\u0026rsquo; (n=53, 10.8%), with mixed race (more than one chosen category) and other races comprising the remainder (n=56, 11.3%). In terms of ethnicity, 123 participants self-identified as \u0026lsquo;Hispanic\u0026rsquo; (24.9%). The mean GA at delivery was 38.6 weeks, with a range of 24.0 to 41.7 weeks; and 11.2% (n=55) delivered preterm (\u0026lt;37 weeks gestation). An excess of male fetuses (XY placentas) (n=276, 56.0%) as compared to female fetuses (XX placentas (n=217) was observed (p\u0026lt;0.05).\u003c/p\u003e\n\u003cp\u003eTable 1. Maternal demographics and clinical characteristics of studied placentas (n=493).\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 55.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCohort Feature\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCount (% or mean as indicated)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 55.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMaternal Age (mean, (SD, range))\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44.5%;\"\u003e\n \u003cp\u003e33.4 (5.6, 18.4-51.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 55.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRace (n, %)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eAmerican Indian or Alaskan Native\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eAsian\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eBlack\u003c/p\u003e\n \u003cp\u003eHawaiian Native or Pacific Islander\u0026nbsp;\u003c/p\u003e\n \u003cp\u003ewhite\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eMixed race\u003c/p\u003e\n \u003cp\u003eOther race\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44.5%;\"\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n \u003cp\u003e1 (0.2)\u003c/p\u003e\n \u003cp\u003e48 (9.7)\u003c/p\u003e\n \u003cp\u003e76 (15.4)\u003c/p\u003e\n \u003cp\u003e4 (0.8)\u003c/p\u003e\n \u003cp\u003e303 (61.5)\u003c/p\u003e\n \u003cp\u003e18 (3.7)\u003c/p\u003e\n \u003cp\u003e43 (8.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 55.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEthnicity (n, %)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eHispanic\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eNon-Hispanic\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44.5%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e123 (24.9)\u003c/p\u003e\n \u003cp\u003e370 (75.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 55.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSES measurements\u0026nbsp;\u003c/strong\u003e(mean (SD, range))\u003c/p\u003e\n \u003cp\u003eFinancial resources composite\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eSocioeconomic position composite\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eDisadvantage composite\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44.5%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e0.0 (0.73, -1.19-5.21)\u003c/p\u003e\n \u003cp\u003e0.02 (0.88, -2.19-0.94)\u003c/p\u003e\n \u003cp\u003e1.06 (1.37, 0-5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 55.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePregestational Diabetes (n, %)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eType 1\u003c/p\u003e\n \u003cp\u003eType 2\u003c/p\u003e\n \u003cp\u003eNo pregestational diabetes\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44.5%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e15 (3.1)\u003c/p\u003e\n \u003cp\u003e30 (6.1)\u003c/p\u003e\n \u003cp\u003e403 (90.8)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 55.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGestational Diabetes (n, %)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eGestational Diabetes, Insulin Required\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eGestational Diabetes, No Insulin Required\u003c/p\u003e\n \u003cp\u003eNo gestational diabetes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44.5%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e27 (5.5)\u003c/p\u003e\n \u003cp\u003e27 (5.5)\u003c/p\u003e\n \u003cp\u003e437 (89.0)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 55.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eComplications (n, %)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003ePreeclampsia\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eGestational Hypertension\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44.5%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e49 (10.0)\u003c/p\u003e\n \u003cp\u003e37 (7.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 55.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSubstance Use During Pregnancy (n, %)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eAlcohol\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eSmoking\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eMarijuana\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44.5%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e24 (5.0)\u003c/p\u003e\n \u003cp\u003e5 (1.0)\u003c/p\u003e\n \u003cp\u003e23 (4.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 55.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGestational Age (Weeks) (mean (SD, range))\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44.5%;\"\u003e\n \u003cp\u003e38.6 (2.0, 24.0-41.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 55.5%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFetal Sex (male:female)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 44.5%;\"\u003e\n \u003cp\u003e276 : 217\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eWhat are the relationships between clinical, sociodemographic, and epivariables in the SPAH cohort?\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBefore engaging in an epigenome-wide association study (EWAS), we sought to understand how clinical and sociodemographic and cell composition variables related to one another to inform downstream analyses and to identify possible collinearities. Of the demographic and DNAme-derived variable relationships, we found first that expected variables (i.e. cell composition estimates, PlaNET derived ancestry estimates/maternal self-reported race categories, diabetes coded as pregestational or gestational, mode of delivery, and gestational age dependent variables) were strongly correlated with each other (R\u003csup\u003e2\u003c/sup\u003e \u0026gt; 0.5) (Fig. S2). There were some significant associations with our primary exposure variables (Socioeconomic Position, Financial Resources, and Disadvantage scores) (Fig. S3). Considering maternal demographic and health variables, white race was associated with more advantageous SES, while Black or \u0026ldquo;other\u0026rdquo; race and Hispanic ethnicity were all associated with more adverse SES across all three measures. In addition, high BMI was associated Disadvantage, whereas as both pregestational and gestational diabetes was associated with measures reflecting higher SES. At the level of the placenta, higher European ancestry probability was associated with SES advantage while higher African ancestry probability was associated with measures of lower SES. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWhat are drivers of DNAme variation in the SPAH cohort?\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBy performing principal components analysis (PCA) on the overall data we can determine how DNAme variation is distributed amongst the samples, and how the PCs relate to clinical, demographic, epiphenotype and technical factors (45). After data processing and normalization, technical variables (the plate the sample was run on, the chip the sample was run on, the row location on the chip) were not significantly associated with either PC1 or PC2 (Fig. S4). PC1 accounted for 16.7% of the variance and was strongly associated with cell composition. PC2 accounted for 4.3% of variance was most strongly associated with gestational age, sex, PlaNET European estimated ancestry and maternal self-reported white race (p \u0026lt; 0.01). There were also lesser but significant (p \u0026lt; 0.01) associations with PlaNET African estimated ancestry and maternal self-reported Hispanic ethnicity. The Socioeconomic Position and Disadvantage composites were associated with PC2 (p-value \u0026lt; 0.01) while Resources and Disadvantage composite measures were weakly associated with PC3 (p \u0026lt; 0.05). As we found no correlation in cell composition estimates with the SES measures (Fig. S3) and as altered cell composition can be an interesting aspect of changes to placental function, we chose not to correct for cell composition in our models, but to test these effects separately. \u0026nbsp;In contrast, it is estimated that 20-70% of DNAme variation is influenced by genetic variation (32,35,54,55) and genetic variation associated with placental ancestry was also correlated with SES measures. Therefore, to detect DNAme changes due to SES, we needed to account for ancestry influenced DNAme.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIs autosomal DNAme associated with socioeconomic disadvantage?\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTable 2. The number of differentially methylated CpGs (FDR\u0026lt;0.05) in each whole cohort linear model run for each of the 3 SES composite scores.\u003c/p\u003e\n\u003cdiv align=\"Left\"\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"643\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 42.5466%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eModel Run\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.85093%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e∆\u0026beta;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22.6708%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSocioeconomic Position\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 11.0248%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResources\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14.9068%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDisadvantage\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 42.5466%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eModel A) DNAme ~ SES\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.85093%;\"\u003e\n \u003cp\u003e\u0026gt; 0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22.6708%;\"\u003e\n \u003cp\u003e1123\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 11.0248%;\"\u003e\n \u003cp\u003e69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14.9068%;\"\u003e\n \u003cp\u003e149\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 42.5466%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.85093%;\"\u003e\n \u003cp\u003e\u0026gt; 0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22.6708%;\"\u003e\n \u003cp\u003e29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 11.0248%;\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14.9068%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 42.5466%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eModel B) DNAme ~ SES + Ancestry + GA\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.85093%;\"\u003e\n \u003cp\u003e\u0026gt; 0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22.6708%;\"\u003e\n \u003cp\u003e77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 11.0248%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14.9068%;\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 42.5466%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.85093%;\"\u003e\n \u003cp\u003e\u0026gt; 0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22.6708%;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 11.0248%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14.9068%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 42.5466%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eModel C) DNAme ~ SES + Ancestry + GA + Ethnicity\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.85093%;\"\u003e\n \u003cp\u003e\u0026gt; 0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22.6708%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 11.0248%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14.9068%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 42.5466%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.85093%;\"\u003e\n \u003cp\u003e\u0026gt; 0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22.6708%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 11.0248%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14.9068%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 42.5466%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eModel D) DNAme ~ SES + All Correlated Covariates\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.85093%;\"\u003e\n \u003cp\u003e\u0026gt; 0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22.6708%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 11.0248%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14.9068%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 42.5466%;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 8.85093%;\"\u003e\n \u003cp\u003e\u0026gt; 0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 22.6708%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 11.0248%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 14.9068%;\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eWe used linear modelling to assess differential DNAme across autosomal CpGs to evaluate whether placental DNAme was associated with Socioeconomic Position, Financial Resources, and/or Disadvantage (Table 2). As a number of variables were associated with SES that could also affect DNAme, we took a sequential approach to evaluate how different variables might influence results. We first ran a crude model (Model A) correcting for no covariates to have a baseline of differential methylation in the SPAH cohort. Under this model and at FDR\u0026lt;0.05, only 29, 9, and 0 CpGs passed the standard biological threshold of |∆\u0026beta;| \u0026gt; 0.05 in models testing Socioeconomic Position, Financial Resources, and Disadvantage, respectively (Table 2). However, 1123, 69 and 149 CpGs reached a minimal biological threshold of |∆\u0026beta;| \u0026gt; 0.02, with those same comparisons (Table 2).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAs the variables most strongly associated with PC2 were gestational age at delivery and PlaNET ancestry estimates, and these are well established to drive placental DNAme variation (56,57), we repeated the linear modeling adding in these covariates (Model B). This correction eliminated most of the initial hits. \u0026nbsp;Only 2 CpGs were associated with Socioeconomic Position meeting a biological threshold of |∆\u0026beta;| \u0026gt; 0.05 (Fig. 1A), both of which showed high variance and differed by genetic ancestry (Fig. S5, Fig. 2). The CpG in \u003cem\u003eATP2C2\u003c/em\u003e with a trimodal distribution (assumed to represent three genotypes) as is typically seen for methylation strongly influenced by one or more genetic variants. The CpG in \u003cem\u003eC1orf141\u003c/em\u003e may be influenced by multiple factors, but the unusually high variance in all genetic ancestries, suggests that DNAme at this site is unlikely to be biologically meaningful in the context of SES exposure. This suggests that these two sites are influenced by genetic variation, despite correcting for ancestry in our models.\u003c/p\u003e\n\u003cp\u003eUsing a relaxed threshold of (|∆\u0026beta;| \u0026gt; 0.02) (Table 2), 77 CpGs were associated with Socioeconomic Position and 22 were associated with Disadvantage. We compared these CpGs to a comprehensive list of Illumina CpG sites known to be influenced by genetic variation in adult blood (33) and found that most of our candidate CpGs were associated with variation in nearby SNPs. To further understand what variables might be driving DNAme at these 77 loci, we performed PCA on their DNAme beta values. Interestingly, we found a strong separation across PC1 and PC2 with Hispanic ethnicity (Fig. S6). It is important to note that Hispanic ethnicity is commonly associated with a variable mix of African, European, and Native American ancestries (58), the latter of which was not accounted for in our ancestry epivariables. We thus decided to next correct for Hispanic ethnicity (Model C), which resulted in no significant CpGs in any of the SES models that met either significance threshold.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAs maternal health or exposure related variables could also potentially obscure psychosocial associations of SES with DNAme we lastly ran a model (Model D) adjusting for any maternal health covariates that were significantly (p \u0026lt; 0.05) correlated (R \u0026gt; 0.15, with n \u0026gt; 10 for case vs. control of the variable in question) with the three SES variables. These additional covariates included mode of conception (natural or assisted), parity, maternal gestational and pregestational diabetes (yes/no), maternal BMI, marijuana use during pregnancy (yes/no), maternal age at delivery, pre-term delivery (yes/no), maternal self-reported Hispanic ethnicity (yes/no), and maternal self-reported race (Black, white, other.).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAre there sex associated effects of socioeconomic disadvantage on placental DNAme?\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSex differences have been reported in human development and some pregnancy outcomes show a sex bias (59,60). Feto-placental sex may modify the effect of prenatal maternal stress on child health outcomes (61). Because sex-differential effects with stress exposure can be obscured when males and females are analyzed together, we sex-stratified data and reran Model B (correcting for gestational age at delivery and PlaNET ancestry estimates) for each of the SES composites (Fig. S7). One CpG met significance (FDR \u0026lt; 0.05, |∆\u0026beta;| \u0026gt; 0.05) with Socioeconomic Position in XY samples only (Fig. S7D); no other sex-stratified autosomal models returned any significantly differentially methylated CpGs. Upon investigation the CpGs associated with Socioeconomic Position score in XY placentas again appeared to be influenced by genetic variation based on the distribution of DNAme values and an association with ancestry (Fig. S5C).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMost EWAS studies do not include analysis of the sex chromosomes because of analytical complexity associated with sex chromosome dosage differences between XX and XY individuals and the impact of X-chromosome inactivation on the epigenome (62). However, there are many functionally important genes on the X chromosome that can be meaningful in the context of placental as well as neurological development (63). To fully explore if the effects of SES were influenced by sex, we thus analyzed X and Y chromosome data from each of the sexes separately (i.e. XX X-chromosome, XY X-chromosome, XY Y-chromosome) (Fig. S8). There was no evidence for altered DNAme on the X or Y chromosome in any of our models (Model B run for each of the SES composites scores) in either XX or XY samples at FDR \u0026lt;0.05.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAre there differences in cytotrophoblast:syncytiotrophoblast ratio associated with SES scores?\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEvaluation of epivariables, inferred from DNAme data such as cell composition, can be a statistically powered way to detect placental changes associated with pathology (64). Although we found no overall correlation between cell composition estimates and any of the SES measures (Fig. S3), we were interested to evaluate the cytotrophoblast:syncytiotrophoblast (cyt:syn) ratio, which is increased in preeclampsia (51,65), and decreased with gestational age and male sex (45,66).\u003c/p\u003e\n\u003cp\u003eIn this cohort, the cyt:syn ratio was negatively correlated with gestational age, (p\u0026lt;0.001) (Fig. 3A) (45) and XX placentas had a higher cyt:syn ratio than XY placentas (p=0.00013) (Fig. S9A), consistent with previous reports (45,66,67). This sex difference in ratio was significant across gestational ages (Fig. S9B). The cyto:syn ratio was not associated with predicted genetic ancestry groups nor with Hispanic ethnicity, nor was it associated with any of the 3 SES composites (Fig. 3B-D)\u003c/p\u003e\n\u003cp\u003eBecause of the effect of sex on the cyto:syn ratio we also performed a sex-stratified analysis on this ratio. We observed that the cyto:syn ratio was significantly and negatively associated with Socioeconomic Position (Fig. S9C) and Resources (Fig. S9D) in XY, but not XX, samples and. XY placentas also showed decreased cyt:syn ratio with the lowest level of Disadvantage, though the trend continued across Disadvantage levels (Fig. S9E). However, to determine if these effects were sex specific we evaluated the interaction term (sex*SES composite score in question) (68), and observed that it was not significant for either the Socioeconomic Position or Disadvantage analyses. The interaction term of sex and Resources was significant, suggesting that there may be an effect of Resources SES on the cyt:syn that is limited to XY samples.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAre there differences in epigenetic age acceleration in association with SES scores?\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn adults, low SES has been linked to epigenetic age acceleration (i.e. a predicted age from DNAme that is greater than chronological age) (29,30). Epigenetic age acceleration \u0026nbsp;has also been linked to several adverse health outcomes including cancer, cardiovascular disease, diabetes, dementia, and overall mortality risk in several different tissues (30,69\u0026ndash;71). Epigenetic clocks have been developed specifically for placenta but have not been widely explored for how these are affected by exposure conditions. We therefore evaluated epigenetic age acceleration in the SPAH cohort both without accounting for cell composition (extrinsic) for an overall picture of epigenetic aging and after adjustment for placental cell types (intrinsic) to determine if one specific cell population was driving changes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCertain demographic variables were associated with both measures of epigenetic age acceleration in the SPAH cohort (Fig. S2). As expected, extrinsic, but not intrinsic, epigenetic age acceleration, was associated with all cell estimates (p\u0026lt;0.05). Also as expected, both epigenetic age acceleration measures were associated with PlaNET-predicted GA (p\u0026lt;0.01). Both extrinsic and intrinsic age acceleration was also associated with fetal sex (p\u0026lt;0.05), BMI (p\u0026lt;0.05), Hispanic ethnicity (p\u0026lt;0.01), maternal White Race (p\u0026lt;0.05), and maternal Hawaiian Native Pacific Islander Race (p\u0026lt;0.01). Epigenetic age acceleration was not associated with the PlaNET-estimated ancestry variables, nor with the Resources and Disadvantage SES composite scores (p\u0026lt;0.05).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhen investigated further, we found a borderline (p=0.05) association between placental epigenetic age acceleration (both extrinsic and intrinsic) and decreasing Socioeconomic Position (Fig. 4A,B). Though the trend was not significant between epigenetic age acceleration and either Resources or Disadvantage, the trend was in the same direction as with Socioeconomic Position, with increased epigenetic age acceleration associated with decreased Resources and Disadvantage scores (Fig. 4C-F). When sex-stratified, there were no significant interactions with sex and placental epigenetic aging across any of the 3 SES scores (Fig. S10).\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eLow SES shows established links to increased rates of adverse pregnancy outcomes in diverse populations world-wide. This may be due to multiple factors including differences in nutritional availability, maternal health, or exposures such as pollutants, pathogens, or stress. Many of these factors have been associated with changes to DNAme in blood (26) or placenta (27). In this study, we examined whether placental DNAme was associated with differing components of SES. At the global DNAme level, the only large-scale changes we observed seemed to be explained by population stratification by SES, or differences in genetic variation influencing DNAme. This implies that large-scale DNAme changes in the placenta are unlikely to explain the adverse health outcomes associated with maternal SES. However, we did observe some associations between SES and DNAme-inferred cell composition and epigenetic age acceleration, which are worth confirming in independent cohorts and exploring more to understand.\u003c/p\u003e\u003cp\u003eA challenge to interpreting epigenome-wide association studies is that a significant portion of DNAme variation is influenced by genetic variation, and the associated allele frequencies can differ between populations. Specifically, it is estimated that in blood, ~\u0026thinsp;10\u0026ndash;30% of Illumina 450k/850K CpGs are differentially methylated by ancestry between African and European populations (72) and ~\u0026thinsp;22% between East Asian and European populations (73). As the SES composites were associated with race and ethnicity, which can in turn be associated with differences in genetic ancestry, it was important to account for genetic ancestry in our models. While there is a loose relationship between race and ancestry, it should be noted that race is a social construct, with no genetic basis, and ancestry estimates vary by race across the United States and are dynamic and changing (58). Genetic ancestry is measured on a continuum, typically on multiple axes relative to major ancestral population groups (e.g. European, African, East Asian, Amerindian, and South Asian). While imperfect, epiphenotyping methods allowed us to roughly estimate the portion of variance associated with European, East Asian, and African ancestry when only DNAme data (and not genetic) is available for a set of samples (56). After adjusting for inferred ancestry probabilities in our epigenome-wide association models, few associations were found, and most remaining were known to be influenced by genetic variation and also were strongly associated with Hispanic ethnicity in our study. Because our ancestry estimates cannot account for Native American ancestry, which can be an important component of ancestry in Hispanic populations (74), we subsequently corrected for Hispanic ethnicity in the models. After doing so, we did not detect any remaining significant associations with SES, indicating that there were no large-scale DNAme associations with SES in this study population.\u003c/p\u003e\u003cp\u003eSocial aspects of race and ethnicity can be an important component of SES and should still have been preserved to some extent in our models. However, we cannot exclude the possibility that confounding between race/ethnicity and ancestry may have prevented us from detecting important associations. It is also important to note that genes which are differentially expressed between ancestry groups tend to be enriched for genes that interact with the environment and ancestry-specific disease effects (75). Given that SES-race associations have been associated with increased risk of health complications (76), further study is needed to investigate how genetic variation-DNAme-environment interactions may affect SES-related outcomes. As most reports of altered placental DNAme in association with environmental exposures have been of small effect size and not consistently reproduced, other mechanisms should also be considered. Adverse pregnancy outcomes could be due to changes to gene transcription that are independent of DNAme (such as small non-coding RNAs, histone modifications) or direct effects of molecules in maternal blood that cross the placenta into fetal circulation (e.g. some viruses or chemicals). Furthermore, the mediators of low SES on pregnancy outcomes are likely diverse, making it more challenging to detect associations at a population level.\u003c/p\u003e\u003cp\u003eWe observed significant differences in the trophoblast cell ratio between XX and XY samples, with XX placentas having higher cyt:syn ratio than XY placentas. We observed the same sex difference in XX and XY cyt:syn ratio in a previous cohort profiling placental DNAme in relation to environmental flooding induced maternal stress (cite QF2011 paper when published) and this was also reported in an independent study of IVF and control placentas (66). Thus, this appears to be a biological sex difference in placentas independent of stress exposure. Subsequent studies should explore if differences in cell composition contribute to differences in XX and XY vulnerability to \u003cem\u003ein utero\u003c/em\u003e exposures and prevalence of placental complications (20,63,77). It is unclear how SES related prenatal maternal stress is associated with this ratio difference, as the interaction with sex was significant with only with Resources SES. The specific aspects of Resources SES that influence this potential sex difference should be verified and then further explored.\u003c/p\u003e\u003cp\u003eIn the present study, we found that both intrinsic and extrinsic age acceleration was slightly accelerated with lower levels of Socioeconomic Position composite scores. Although the association between epigenetic age acceleration with Financial Resources and Disadvantage SES composite scores did not reach significance, they did trend in the same direction with lower levels of SES associated with higher levels of age acceleration. Further, our results mirror reports of SES and epigenetic aging in adults, as exposure to socioeconomic disadvantage in early life (\u0026lt;\u0026thinsp;5 years of age), but not current socioeconomic disadvantage exposure in adulthood (ages 15 to 55 years of age), was associated with accelerated epigenetic aging of monocytes from adult peripheral blood, indicating significance of exposure to SES during critical developmental periods (29). Our results are also in line with other results of SES effects in the placenta; \u003cem\u003ein utero\u003c/em\u003e exposure to maternal SES adversity was associated with differential DNAme of genes with important roles in gene transcription and placental function, indicating that exposure to SES may impact placental DNAme in a way that may be related to biological pathways relevant to fetal and pregnancy outcomes (23). However, such changes may also be the consequence of healthy adaption to stress, as a recent placental study showed that epigenetic age acceleration was associated with shorter NICU stays (78), or may be a secondary response to other changes.\u003c/p\u003e\u003cp\u003eOur study is not without its limitations. The epiphenotyping tool PlaNET is only able to estimate a limited number of placental cell types and does not distinguish amongst subtypes of trophoblast, such as those that have been identified through single cell sequencing (79\u0026ndash;81). Furthermore, DNAme-based genetic ancestry estimates are less robust than those computed using genome-wide SNP genotypes; further, the placental ancestry estimation tool to date was not developed for ancestry groups such as Native American or South Asian populations, and may not accurately detect genetic ancestry variation along these axes (31,56). Additionally, the SPAH cohort, though diverse compared to other placental datasets, still did recruit a majority of women of white self-reported race (64.5%) and distributions of SES composite scores skewed in the direction of higher SES. Thus, efforts should be made to confirm our results in other diverse cohorts with greater representation of low SES individuals, as well as to account for genetic variation using more robust approaches.\u003c/p\u003e\u003cp\u003eThe major strength in our study lies in the careful and deliberate analyses of the effects of SES on placental DNAme variation. Many studies make no attempt to correct for genetic ancestry, cell type composition, or gestational age when studying effects of maternal exposures on epigenetics, which may result in false positives and limit the ability to detect reproducible and robust associations. In addition, we also carefully investigated the effect of sex via not only sex-stratified analyses, but also in direct analysis of the sex chromosomes, which are very often discarded in EWAS studies.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eIn conclusion, this study aimed to identify patterns of DNAme associated with exposure to low SES, and whether the nature and strength of these associations varied depending on how SES was assessed (i.e. Socioeconomic Position, Financial Resources, or Socioeconomic Disadvantage). Overall, no meaningful associations of prenatal maternal stress on the placental DNA methylome in this study were observed after accounting for genetic ancestry variation, emphasizing that placental EWAS studies involving diverse cohorts should account for genetic variation in their analyses to avoid false findings. This study also underscores the challenges with elucidating mechanisms underlying SES linked outcomes, given the complex nature of correlated variables. It remains a question for further investigation as to whether epigenetic age acceleration is an adverse effect of exposure to prenatal maternal stress associated with SES, given its association with pregnancy complications (73\u0026ndash;75), or alternatively, if placental epigenetic aging is a healthy adaption to complications that increase the risk of preterm delivery. Lastly, its important to note that DNAme is a relatively stable mark strongly associated with cell composition and genetic variation, but gene expression levels are influenced by many factors in addition to DNAme. Thus, this work does not exclude that there might be changes to other aspects of gene regulation (e.g. histone marks or small non-coding RNAs) in association with exposure to maternal stress and SES.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eThe study protocol was approved by the Institutional Review Boards of Northwestern University (STU00206269) and NorthShore University Health System (EH17-006).\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNot applicable.\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated and/or analyzed during the current study are available in the Gene Expression Omnibus under accession number\u0026nbsp;GSE307289.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe gratefully acknowledge the study participants who donated their placentas to the SPAH study. The research team would like to thank: Britney P. Smart, Janedelie Romero, Katherine Vause, Claire Fisher, Renee M. Odom, Chuhan Wu, Jane Drage, Megan Choi, Jasmin Flowers, Hee Moon, Veronica Passarelli, Adam Leigh, Lauren Hoffer, Shanti Gallivan, Tao Jiang, Lavisha Singh, the NorthShore University Department of Pathology and Laboratory Medicine, and Northwestern University’s Foundations of Health Research Center.\u0026nbsp;We also thank the Northwestern University NUSeq Core Facilities for running of the Illumina EPIC arrays. Thank you to Hannah Illing for assistance in array processing/quality control, preparation of the Gene Expression Omnibus dataset and feedback on the manuscript. We gratefully acknowledge the work and staff of Endeavor Health and the NorthShore University HealthSystem for their assistance with patient recruitment and support of the study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors’ contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGEM and AEB established the overall study design, project funding, cohort recruitment, and oversaw sample collection for the SPAH cohort. LME, AAF, and LSKD participated in the collection and management of the clinical data. MSP contributed project data management. EOB conducted all data processing/analysis and drafted the manuscript. WPR and AMI participated in data analysis. GEM, WPR and AMI participated in manuscript editing. All authors were involved in aspects of study design and read, revised, and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported in part by the National Institutes of Health, National Institute on Minority Health and Health Disparities (R01MD011749). The funders had no role in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication. EOB received salary support from Canadian Health Research Grants GSK-171375 and PJT-169131); AMI received salary support from a CIHR CGS-D award and CIHR grant GSK-171375. WPR receives salary support through an investigatorship award from the BC Children’s Hospital Research Institute.\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBlumenshine P, Egerter S, Barclay CJ, Cubbin C, Braveman PA. Socioeconomic disparities in adverse birth outcomes: a systematic review. Am J Prev Med. 2010 Sep;39(3):263\u0026ndash;72. \u003c/li\u003e\n\u003cli\u003eNicholls-Dempsey L, Badeghiesh A, Baghlaf H, Dahan MH. How does high socioeconomic status affect maternal and neonatal pregnancy outcomes? A population-based study among American women. Eur J Obstet Gynecol Reprod Biol X. 2023 Oct 12;20:100248. \u003c/li\u003e\n\u003cli\u003eBushnik T, Yang S, Kaufman JS, Kramer MS, Wilkins R. Socioeconomic disparities in small-for-gestational-age birth and preterm birth. Health Rep. 2017 Nov 15;28(11):3\u0026ndash;10. \u003c/li\u003e\n\u003cli\u003eJoseph KS, Liston RM, Dodds L, Dahlgren L, Allen AC. Socioeconomic status and perinatal outcomes in a setting with universal access to essential health care services. CMAJ Can Med Assoc J. 2007 Sep 11;177(6):583\u0026ndash;90. \u003c/li\u003e\n\u003cli\u003eThomson K, Moffat M, Arisa O, Jesurasa A, Richmond C, Odeniyi A, et al. Socioeconomic inequalities and adverse pregnancy outcomes in the UK and Republic of Ireland: a systematic review and meta-analysis. BMJ Open. 2021 Mar 1;11(3):e042753. \u003c/li\u003e\n\u003cli\u003eHertzman C, Boyce T. How experience gets under the skin to create gradients in developmental health. Annu Rev Public Health. 2010;31:329-347 3p following 347. \u003c/li\u003e\n\u003cli\u003eBarch DM, Donohue MR, Elsayed NM, Gilbert K, Harms MP, Hennefield L, et al. Early Childhood Socioeconomic Status and Cognitive and Adaptive Outcomes at the Transition to Adulthood: The Mediating Role of Gray Matter Development Across Five Scan Waves. Biol Psychiatry Cogn Neurosci Neuroimaging. 2022 Jan 1;7(1):34\u0026ndash;44. \u003c/li\u003e\n\u003cli\u003eSchultz WM, Kelli HM, Lisko JC, Varghese T, Shen J, Sandesara P, et al. Socioeconomic Status and Cardiovascular Outcomes. Circulation. 2018 May 15;137(20):2166\u0026ndash;78. \u003c/li\u003e\n\u003cli\u003eMiller GE, Chen E, Parker KJ. Psychological Stress in Childhood and Susceptibility to the Chronic Diseases of Aging: Moving Towards a Model of Behavioral and Biological Mechanisms. Psychol Bull. 2011 Nov;137(6):959\u0026ndash;97. \u003c/li\u003e\n\u003cli\u003eMiller GE, Borders AE, Crockett AH, Ross KM, Qadir S, Keenan-Devlin L, et al. MATERNAL SOCIOECONOMIC DISADVANTAGE IS ASSOCIATED WITH TRANSCRIPTIONAL INDICATIONS OF GREATER IMMUNE ACTIVATION AND SLOWER TISSUE MATURATION IN PLACENTAL BIOPSIES AND NEWBORN CORD BLOOD. Brain Behav Immun. 2017 Aug;64:276\u0026ndash;84. \u003c/li\u003e\n\u003cli\u003eKing S, Kildea S, Austin MP, Brunet A, Cobham VE, Dawson PA, et al. QF2011: a protocol to study the effects of the Queensland flood on pregnant women, their pregnancies, and their children\u0026rsquo;s early development. BMC Pregnancy Childbirth. 2015 May 6;15(1):109. \u003c/li\u003e\n\u003cli\u003eAvishai-Eliner S, Brunson KL, Sandman CA, Baram TZ. Stressed-out, or in (utero)? Trends Neurosci. 2002 Oct;25(10):518\u0026ndash;24. \u003c/li\u003e\n\u003cli\u003eMatthews KA, Gallo LC. Psychological perspectives on pathways linking socioeconomic status and physical health. Annu Rev Psychol. 2011;62:501\u0026ndash;30. \u003c/li\u003e\n\u003cli\u003eChen E, Miller GE. Socioeconomic Status and Health: Mediating and Moderating Factors. Annu Rev Clin Psychol. 2013 Mar 28;9(Volume 9, 2013):723\u0026ndash;49. \u003c/li\u003e\n\u003cli\u003eKnobel P, Hwang I, Castro E, Sheffield P, Holaday L, Shi L, et al. Socioeconomic and racial disparities in source-apportioned PM2.5 levels across urban areas in the contiguous US, 2010. Atmos Environ. 2023 Jun 15;303:119753. \u003c/li\u003e\n\u003cli\u003eBurnett R, Chen H, Szyszkowicz M, Fann N, Hubbell B, Pope CA, et al. Global estimates of mortality associated with long-term exposure to outdoor fine particulate matter. Proc Natl Acad Sci U S A. 2018 Sep 18;115(38):9592\u0026ndash;7. \u003c/li\u003e\n\u003cli\u003eAdamkiewicz G, Zota AR, Fabian MP, Chahine T, Julien R, Spengler JD, et al. Moving Environmental Justice Indoors: Understanding Structural Influences on Residential Exposure Patterns in Low-Income Communities. Am J Public Health. 2011 Dec;101(Suppl 1):S238\u0026ndash;45. \u003c/li\u003e\n\u003cli\u003eHusted G, Cassels S, Ackert E, Sweeney S. Socio-demographic patterning of urinary mono-ethyl phthalate levels among children and adults in the U.S. (1999\u0026ndash;2018). Environ Res Commun [Internet]. 2024 Aug 27;6(8). Available from: https://iopscience.iop.org/article/10.1088/2515-7620/ad6403\u003c/li\u003e\n\u003cli\u003eStringhini S, Carmeli C, Jokela M, Avenda\u0026ntilde;o M, Muennig P, Guida F, et al. Socioeconomic status and the 25 \u0026times; 25 risk factors as determinants of premature mortality: a multicohort study and meta-analysis of 1\u0026middot;7 million men and women. Lancet Lond Engl. 2017 Mar 25;389(10075):1229\u0026ndash;37. \u003c/li\u003e\n\u003cli\u003eBronson SL, Bale TL. The Placenta as a Mediator of Stress Effects on Neurodevelopmental Reprogramming. Neuropsychopharmacology. 2016 Jan;41(1):207\u0026ndash;18. \u003c/li\u003e\n\u003cli\u003eHanson MA, Gluckman PD. Early developmental conditioning of later health and disease: physiology or pathophysiology? Physiol Rev. 2014 Oct;94(4):1027\u0026ndash;76. \u003c/li\u003e\n\u003cli\u003eCoe CL, Lubach GR. Vital and vulnerable functions of the primate placenta critical for infant health and brain development. Front Neuroendocrinol. 2014 Oct;35(4):439\u0026ndash;46. \u003c/li\u003e\n\u003cli\u003eSantos Jr HP, Bhattacharya A, Martin EM, Addo K, Psioda M, Smeester L, et al. Epigenome-wide DNA methylation in placentas from preterm infants: association with maternal socioeconomic status. Epigenetics. 2019 Aug 3;14(8):751\u0026ndash;65. \u003c/li\u003e\n\u003cli\u003eAppleton AA, Armstrong DA, Lesseur C, Lee J, Padbury JF, Lester BM, et al. Patterning in Placental 11-B Hydroxysteroid Dehydrogenase Methylation According to Prenatal Socioeconomic Adversity. PLOS ONE. 2013 Sep 5;8(9):e74691. \u003c/li\u003e\n\u003cli\u003eGhazi T, Naidoo P, Naidoo RN, Chuturgoon AA. Prenatal Air Pollution Exposure and Placental DNA Methylation Changes: Implications on Fetal Development and Future Disease Susceptibility. Cells. 2021 Nov 5;10(11):3025. \u003c/li\u003e\n\u003cli\u003eMartin EM, Fry RC. Environmental Influences on the Epigenome: Exposure- Associated DNA Methylation in Human Populations. Annu Rev Public Health. 2024 Apr 1;39(1):309\u0026ndash;33. \u003c/li\u003e\n\u003cli\u003eMortillo M, Marsit CJ. Select early-life environmental exposures and DNA methylation in the placenta. Curr Environ Health Rep. 2023 Mar;10(1):22\u0026ndash;34. \u003c/li\u003e\n\u003cli\u003eBozack AK, Trasande L. Prenatal chemical exposures and the methylome: current evidence and opportunities for environmental epigenetics. Epigenomics. 16(23\u0026ndash;24):1443\u0026ndash;51. \u003c/li\u003e\n\u003cli\u003eAustin MK, Chen E, Ross KM, McEwen LM, Maclsaac JL, Kobor MS, et al. Early-life socioeconomic disadvantage, not current, predicts accelerated epigenetic aging of monocytes. Psychoneuroendocrinology. 2018 Nov 1;97:131\u0026ndash;4. \u003c/li\u003e\n\u003cli\u003eOblak L, van der Zaag J, Higgins-Chen AT, Levine ME, Boks MP. A systematic review of biological, social and environmental factors associated with epigenetic clock acceleration. Ageing Res Rev. 2021 Aug;69:101348. \u003c/li\u003e\n\u003cli\u003eYuan V, Hui D, Yin Y, Pe\u0026ntilde;aherrera MS, Beristain AG, Robinson WP. Cell-specific characterization of the placental methylome. BMC Genomics. 2021 Jan 6;22(1):6. \u003c/li\u003e\n\u003cli\u003eBell CG, Gao F, Yuan W, Roos L, Acton RJ, Xia Y, et al. Obligatory and facilitative allelic variation in the DNA methylome within common disease-associated loci. Nat Commun. 2018 Jan 2;9:8. \u003c/li\u003e\n\u003cli\u003eVillica\u0026ntilde;a S, Castillo-Fernandez J, Hannon E, Christiansen C, Tsai PC, Maddock J, et al. Genetic impacts on DNA methylation help elucidate regulatory genomic processes. Genome Biol. 2023 Jul 31;24:176. \u003c/li\u003e\n\u003cli\u003eDo C, Lang CF, Lin J, Darbary H, Krupska I, Gaba A, et al. Mechanisms and Disease Associations of Haplotype-Dependent Allele-Specific DNA Methylation. Am J Hum Genet. 2016 May 5;98(5):934\u0026ndash;55. \u003c/li\u003e\n\u003cli\u003eDo C, Shearer A, Suzuki M, Terry MB, Gelernter J, Greally JM, et al. Genetic\u0026ndash;epigenetic interactions in cis: a major focus in the post-GWAS era. Genome Biol. 2017 Jun 19;18:120. \u003c/li\u003e\n\u003cli\u003eFraser HB, Lam LL, Neumann SM, Kobor MS. Population-specificity of human DNA methylation. Genome Biol. 2012 Feb 9;13(2):R8. \u003c/li\u003e\n\u003cli\u003ePatin E, Hasan M, Bergstedt J, Rouilly V, Libri V, Urrutia A, et al. Natural variation in the parameters of innate immune cells is preferentially driven by genetic factors. Nat Immunol. 2018 Mar;19(3):302\u0026ndash;14. \u003c/li\u003e\n\u003cli\u003eGalanter JM, Gignoux CR, Oh SS, Torgerson D, Pino-Yanes M, Thakur N, et al. Differential methylation between ethnic sub-groups reflects the effect of genetic ancestry and environmental exposures. eLife. 6:e20532. \u003c/li\u003e\n\u003cli\u003eHenikoff S, Greally JM. Epigenetics, cellular memory and gene regulation. Curr Biol. 2016 Jul 25;26(14):R644\u0026ndash;8. \u003c/li\u003e\n\u003cli\u003eLappalainen T, Greally JM. Associating cellular epigenetic models with human phenotypes. Nat Rev Genet. 2017 Jul;18(7):441\u0026ndash;51. \u003c/li\u003e\n\u003cli\u003eLee Y, Choufani S, Weksberg R, Wilson SL, Yuan V, Burt A, et al. Placental epigenetic clocks: estimating gestational age using placental DNA methylation levels. Aging. 2019 Jun 24;11(12):4238\u0026ndash;53. \u003c/li\u003e\n\u003cli\u003eLauren S. Keenan-Devlin, Alexa Freedman, Renee M. Odom-Konja, Michelle A. Chen, Tao Jiang, Linda M. Ernst, et al. Associations of objective, subjective, childhood, and neighborhood measures of socioeconomic status with adverse pregnancy outcomes. \u003c/li\u003e\n\u003cli\u003eThe National Statistics Socio-economic classification (NS-SEC) - Office for National Statistics [Internet]. [cited 2024 Aug 8]. Available from: https://www.ons.gov.uk/methodology/classificationsandstandards/otherclassifications/thenationalstatisticssocioeconomicclassificationnssecrebasedonsoc2010#using-the-derivation-tables\u003c/li\u003e\n\u003cli\u003eMiller GE, Brody GH, Yu T, Chen E. A family-oriented psychosocial intervention reduces inflammation in low-SES African American youth. Proc Natl Acad Sci U S A. 2014 Aug 5;111(31):11287\u0026ndash;92. \u003c/li\u003e\n\u003cli\u003eKhan A, Inkster AM, Pe\u0026ntilde;aherrera 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 Oct 4;16(1):37. \u003c/li\u003e\n\u003cli\u003ePidsley R, Y Wong CC, Volta M, Lunnon K, Mill J, Schalkwyk LC. A data-driven approach to preprocessing Illumina 450K methylation array data. BMC Genomics. 2013 May 1;14(1):293. \u003c/li\u003e\n\u003cli\u003eTriche TJ, Weisenberger DJ, Van Den Berg D, Laird PW, Siegmund KD. Low-level processing of Illumina Infinium DNA Methylation BeadArrays. Nucleic Acids Res. 2013 Apr;41(7):e90. \u003c/li\u003e\n\u003cli\u003eZhou W, Laird PW, Shen H. Comprehensive characterization, annotation and innovative use of Infinium DNA methylation BeadChip probes. Nucleic Acids Res. 2017 Feb 28;45(4):e22. \u003c/li\u003e\n\u003cli\u003eMurat K, Gr\u0026uuml;ning B, Poterlowicz PW, Westgate G, Tobin DJ, Poterlowicz K. Ewastools: Infinium Human Methylation BeadChip pipeline for population epigenetics integrated into Galaxy. GigaScience. 2020 May 1;9(5):giaa049. \u003c/li\u003e\n\u003cli\u003eDaenekas B, P\u0026eacute;rez E, Boniolo F, Stefan S, Benfatto S, Sill M, et al. Conumee 2.0: enhanced copy-number variation analysis from DNA methylation arrays for humans and mice. Bioinformatics. 2024 Feb 1;40(2):btae029. \u003c/li\u003e\n\u003cli\u003eInkster AM, Illing HJ, Robinson WP. Some lessons learned from genomic and epigenomic studies of the placenta. Placenta [Internet]. 2025 Apr 17 [cited 2025 May 18]; Available from: https://www.sciencedirect.com/science/article/pii/S0143400425001237\u003c/li\u003e\n\u003cli\u003eRitchie 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 Apr 20;43(7):e47. \u003c/li\u003e\n\u003cli\u003eStoreyLab/biobroom [Internet]. Storey Lab; 2023 [cited 2024 Jan 10]. Available from: https://github.com/StoreyLab/biobroom\u003c/li\u003e\n\u003cli\u003eCheung WA, Shao X, Morin A, Siroux V, Kwan T, Ge B, et al. Functional variation in allelic methylomes underscores a strong genetic contribution and reveals novel epigenetic alterations in the human epigenome. Genome Biol. 2017 Mar 10;18:50. \u003c/li\u003e\n\u003cli\u003eBell JT, Pai AA, Pickrell JK, Gaffney DJ, Pique-Regi R, Degner JF, et al. DNA methylation patterns associate with genetic and gene expression variation in HapMap cell lines. Genome Biol. 2011;12(1):R10. \u003c/li\u003e\n\u003cli\u003eYuan 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 Aug 9;12(1):51. \u003c/li\u003e\n\u003cli\u003eKonwar C, Del Gobbo G, Yuan V, Robinson WP. Considerations when processing and interpreting genomics data of the placenta. Placenta. 2019 Sep 1;84:57\u0026ndash;62. \u003c/li\u003e\n\u003cli\u003eBryc K, Durand EY, Macpherson JM, Reich D, Mountain JL. The Genetic Ancestry of African Americans, Latinos, and European Americans across the United States. Am J Hum Genet. 2015 Jan 8;96(1):37\u0026ndash;53. \u003c/li\u003e\n\u003cli\u003eInkster AM, Fern\u0026aacute;ndez-Boyano I, Robinson WP. Sex Differences Are Here to Stay: Relevance to Prenatal Care. J Clin Med. 2021 Jul 5;10(13):3000. \u003c/li\u003e\n\u003cli\u003eBraun AE, Mitchel OR, Gonzalez TL, Sun T, Flowers AE, Pisarska MD, et al. Sex at the interface: the origin and impact of sex differences in the developing human placenta. Biol Sex Differ. 2022 Sep 16;13(1):50. \u003c/li\u003e\n\u003cli\u003eSutherland S, Brunwasser SM. Sex Differences in Vulnerability to Prenatal Stress: a Review of the Recent Literature. Curr Psychiatry Rep. 2018 Sep 1;20(11):102. \u003c/li\u003e\n\u003cli\u003eInkster AM, Wong MT, Matthews AM, Brown CJ, Robinson WP. Who\u0026rsquo;s afraid of the X? Incorporating the X and Y chromosomes into the analysis of DNA methylation array data. Epigenetics Chromatin. 2023 Jan 7;16(1):1. \u003c/li\u003e\n\u003cli\u003eBale TL. The placenta and neurodevelopment: sex differences in prenatal vulnerability. Dialogues Clin Neurosci. 2016 Dec;18(4):459\u0026ndash;64. \u003c/li\u003e\n\u003cli\u003eCampbell KA, Colacino JA, Puttabyatappa M, Dou JF, Elkin ER, Hammoud SS, et al. Placental cell type deconvolution reveals that cell proportions drive preeclampsia gene expression differences. Commun Biol. 2023 Mar 13;6(1):1\u0026ndash;15. \u003c/li\u003e\n\u003cli\u003eFern\u0026aacute;ndez-Boyano I, Inkster AM, Yuan V, Robinson WP. eoPred: predicting the placental phenotype of early-onset preeclampsia using public DNA methylation data. Front Genet. 2023;14:1248088. \u003c/li\u003e\n\u003cli\u003eLemaire M, Smith KW, Wilson SL. Examining Fetal Sex-Specific Placental DNA Methylation Intensities and Variability Post In Vitro Fertilization [Internet]. bioRxiv; 2024 [cited 2024 Sep 19]. p. 2024.08.08.604307. Available from: https://www.biorxiv.org/content/10.1101/2024.08.08.604307v1\u003c/li\u003e\n\u003cli\u003eInkster AM, Konwar C, Pe\u0026ntilde;aherrera MS, Brain U, Khan A, Price EM, et al. Profiling placental DNA methylation associated with maternal SSRI treatment during pregnancy. Sci Rep. 2022 Dec 30;12(1):22576. \u003c/li\u003e\n\u003cli\u003eChristians JK, Chow NA. Are there sex differences in fetal growth strategies and in the long-term effects of pregnancy complications on cognitive functioning? J Dev Orig Health Dis. 2022 Dec;13(6):766\u0026ndash;78. \u003c/li\u003e\n\u003cli\u003eJones MJ, Goodman SJ, Kobor MS. DNA methylation and healthy human aging. Aging Cell. 2015;14(6):924\u0026ndash;32. \u003c/li\u003e\n\u003cli\u003eHorvath S. DNA methylation age of human tissues and cell types. Genome Biol. 2013 Dec 10;14(10):3156. \u003c/li\u003e\n\u003cli\u003eHorvath S, Raj K. DNA methylation-based biomarkers and the epigenetic clock theory of ageing. Nat Rev Genet. 2018 Jun;19(6):371\u0026ndash;84. \u003c/li\u003e\n\u003cli\u003eHusquin LT, Rotival M, Fagny M, Quach H, Zidane N, McEwen LM, et al. Exploring the genetic basis of human population differences in DNA methylation and their causal impact on immune gene regulation. Genome Biol. 2018 Dec 18;19(1):222. \u003c/li\u003e\n\u003cli\u003eHatton AA, Cheng FF, Lin T, Shen RJ, Chen J, Zheng Z, et al. Genetic control of DNA methylation is largely shared across European and East Asian populations. Nat Commun. 2024 Mar 28;15(1):2713. \u003c/li\u003e\n\u003cli\u003eBanda Y, Kvale MN, Hoffmann TJ, Hesselson SE, Ranatunga D, Tang H, et al. Characterizing Race/Ethnicity and Genetic Ancestry for 100,000 Subjects in the Genetic Epidemiology Research on Adult Health and Aging (GERA) Cohort. Genetics. 2015 Aug;200(4):1285\u0026ndash;95. \u003c/li\u003e\n\u003cli\u003eWang J, Zhang Z, Lu Z, Mancuso N, Gazal S. Genes with differential expression across ancestries are enriched in ancestry-specific disease effects likely due to gene-by-environment interactions. Am J Hum Genet. 2024 Oct 3;111(10):2117\u0026ndash;28. \u003c/li\u003e\n\u003cli\u003eSonderlund AL, Charifson M, Schoenthaler A, Carson T, Williams NJ. Racialized economic segregation and health outcomes: A systematic review of studies that use the Index of Concentration at the Extremes for race, income, and their interaction. PLOS ONE. 2022 Jan 28;17(1):e0262962. \u003c/li\u003e\n\u003cli\u003eCarpenter T, Grecian SM, Reynolds RM. Sex differences in early-life programming of the hypothalamic\u0026ndash;pituitary\u0026ndash;adrenal axis in humans suggest increased vulnerability in females: a systematic review. J Dev Orig Health Dis. 2017 Apr;8(2):244\u0026ndash;55. \u003c/li\u003e\n\u003cli\u003eClark J, Bulka CM, Martin CL, Roell K, Santos HP, O\u0026rsquo;Shea TM, et al. Placental epigenetic gestational aging in relation to maternal sociodemographic factors and smoking among infants born extremely preterm: a descriptive study. Epigenetics. 2022 Dec;17(13):2389\u0026ndash;403. \u003c/li\u003e\n\u003cli\u003eVan Buren E, Azzara D, Rangel-Moreno J, Garcia-Hernandez M de la L, Murphy SP, Cohen ED, et al. Single-cell RNA sequencing reveals placental response under environmental stress. Nat Commun. 2024 Aug 2;15(1):6549. \u003c/li\u003e\n\u003cli\u003eWang Q, Li J, Wang S, Deng Q, An Y, Xing Y, et al. Single-cell transcriptional profiling reveals cellular and molecular divergence in human maternal-fetal interface. Sci Rep. 2022 Jun 28;12(1):10892. \u003c/li\u003e\n\u003cli\u003eShannon MJ, Baltayeva J, Castellana B, W\u0026auml;chter J, McNeill GL, Yoon JS, et al. Cell trajectory modeling identifies a primitive trophoblast state defined by BCAM enrichment. Dev Camb Engl. 2022 Jan 1;149(1):dev199840. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"clinical-epigenetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"clep","sideBox":"Learn more about [Clinical Epigenetics](http://clinicalepigeneticsjournal.biomedcentral.com/)","snPcode":"13148","submissionUrl":"https://submission.nature.com/new-submission/13148/3","title":"Clinical Epigenetics","twitterHandle":"@OAgenetics","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"DNA methylation, placenta, socioeconomic status, pregnancy, epigenetic age","lastPublishedDoi":"10.21203/rs.3.rs-7546517/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7546517/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eDisparities in socioeconomic status have been associated with adverse pregnancy outcomes, including preterm birth and fetal growth restriction. As the barrier between maternal exposures and the fetus, the placenta has been proposed to play a role in the mechanisms leading to poor health outcomes seen with socioeconomic disadvantage. We hypothesized that exposure to lower SES during pregnancy may lead to altered placental DNA methylation (DNAme) that is in turn associated with other pregnancy outcomes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003ePlacental samples from the Stress, Pregnancy, and Health Study (SPAH) study (n=493) were processed for DNAme analysis using the Illumina Infinium MethylationEPIC BeadChip array.\u003cstrong\u003e \u003c/strong\u003eLinear modelling was used to assess whether placental DNAme was associated with Socioeconomic Position, Financial Resources, and/or Disadvantage.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eAt FDR \u0026lt;0.05 and |∆β| \u0026gt;0.05, we observed only 2 CpGs associated with Socioeconomic Position after correcting for gestational age and ancestry, while at a less stringent |∆β| \u0026gt;0.02 threshold there were 77 and 22 CpG associations with Socioeconomic Position and Disadvantage respectively. However, these changes seemed to be explained by genetic variation influencing DNAme in combination with population stratification by socioeconomic status. We did observe associations between socioeconomic status and DNAme-inferred cell composition and epigenetic age acceleration, with intrinsic (p=0.047) and extrinsic (p=0.050) age acceleration being slightly accelerated with lower levels of Socioeconomic Position. Financial Resources and Disadvantage SES trended in the same direction as Social Position, with lower socioeconomic status seen with higher levels of age acceleration, though not reaching significance. No meaningful associations in sex stratified analyses were identified, although XX placentas showed higher cytotrophoblast:syncytiotrophoblast ratio than XY placentas (p=0.00013).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eOur results emphasize the importance in placental studies involving diverse cohorts to account for genetic variation in order to avoid false findings. This study also demonstrates the challenges with elucidating mechanisms underlying socioeconomic associated outcomes, given the complex nature of correlated variables. Further investigation is required to elucidate whether epigenetic age acceleration is an adverse effect of exposure to prenatal maternal stress associated with socioeconomic disparities, or alternatively, if placental epigenetic aging is a potential healthy adaption to pregnancy complications that increase the risk of preterm delivery.\u003c/p\u003e","manuscriptTitle":"DNA Methylation in the Placenta and Maternal Socioeconomic Status: The SPAH Study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-14 04:41:33","doi":"10.21203/rs.3.rs-7546517/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-16T08:44:14+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-06T21:50:08+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-02T22:08:22+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-30T02:49:41+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-28T20:13:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"333672026523935418138111383650607199594","date":"2026-01-21T14:24:23+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"40506495575466191584679195133434734589","date":"2026-01-21T02:01:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"340074934659855188137908987225802350782","date":"2026-01-20T13:30:16+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"177999177587096869190396037401111111169","date":"2026-01-19T18:30:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"269054031321128440619969782991592779484","date":"2026-01-19T14:16:29+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-29T15:51:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"197650689261051378709086022533474783362","date":"2025-11-19T21:49:13+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-18T14:49:14+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-09T06:05:58+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-08T05:55:27+00:00","index":"","fulltext":""},{"type":"submitted","content":"Clinical Epigenetics","date":"2025-09-05T18:10:00+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"clinical-epigenetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"clep","sideBox":"Learn more about [Clinical Epigenetics](http://clinicalepigeneticsjournal.biomedcentral.com/)","snPcode":"13148","submissionUrl":"https://submission.nature.com/new-submission/13148/3","title":"Clinical Epigenetics","twitterHandle":"@OAgenetics","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9807c5da-6723-4771-bbaf-d12713a81adc","owner":[],"postedDate":"November 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-04-27T16:12:16+00:00","versionOfRecord":{"articleIdentity":"rs-7546517","link":"https://doi.org/10.1186/s13148-026-02136-7","journal":{"identity":"clinical-epigenetics","isVorOnly":false,"title":"Clinical Epigenetics"},"publishedOn":"2026-04-24 15:58:48","publishedOnDateReadable":"April 24th, 2026"},"versionCreatedAt":"2025-11-14 04:41:33","video":"","vorDoi":"10.1186/s13148-026-02136-7","vorDoiUrl":"https://doi.org/10.1186/s13148-026-02136-7","workflowStages":[]},"version":"v1","identity":"rs-7546517","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7546517","identity":"rs-7546517","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}