Background

Ambient air pollution has become a global environmental threat [1]. The primary air pollutants mainly include particulate matter (PM), nitrogen dioxide (NO2), carbon monoxide (CO), sulfur dioxide (SO2), and ozone (O3), among which PM remains one of the most harmful forms, contributing to more than 4.2 million premature mortalities [2]. Evidence from animal and human studies suggests that exposure to air pollution can reduce fertility rates and increase the risk of miscarriage [3]. When the concentration of ambient PM is more than 40 µg/m3, an estimated 3.5 million pregnant women experience a miscarriage per year in South Asia [4]. Besides, exposure to NO2 during early pregnancy was associated with increased odds of spontaneous abortion in linear dose–response manners [5]. A case–crossover analysis reported that PM10 (particles < 10 μm), PM2.5–10 (particles between 2.5 μm and 10 μm), and PM2.5 exposure were positively associated with an increased risk of spontaneous abortion in the USA [6]. CO exposure during the first trimester of pregnancy was linked to an increased risk of spontaneous abortion in case–control research from Iran, which reported a 95 percent increase in spontaneous abortions in cases compared to controls [7]. In some cities of China such as Jiangsu, Bei**g, and Wuhan, the researchers also found that maternal exposure to air pollution was significantly associated with an increased risk of incident spontaneous pregnancy loss [8,9,10]. Although ambient air pollution has been linked with spontaneous abortion [11, 12], the relationship between recurrent spontaneous abortion (RSA) and air pollution and its underlying molecular mechanism is not well established.

RSA, one of the most common complications of pregnancy, is usually defined as three or more consecutive spontaneous abortions with the same spouse [13]. Its incidence is 2–4% among clinically recognized pregnancies [14]. RSA is a multifactorial disease condition, which has been associated with environmental pollution, chromosomal abnormalities, genital tract anatomic abnormalities, endocrine disorders, autoimmunological factors, and infectious diseases. To date, several foundational studies have employed single-cell RNA sequencing (scRNA-seq) technology to investigate the cellular composition and inter-cellular communication events at the maternal–fetal interface in patients with RSA [15,16,17]. Advancements in scRNA-seq have greatly facilitated the development of novel approaches to improve targeted therapies and precision medicines [18]. However, these studies only analyzed the pathological mechanism of RSA from the perspective of immunology and did not completely construct all cell atlas of RSA patients. A more comprehensive cell atlas is therefore needed and provides new insights and bases for studying the pathogenesis and the clinical treatment of RSA.

Accumulating evidence indicates that exposure to air pollution, especially PM2.5, causes changes in DNA methylation both in vivo and in vitro models [19, 20]. Air pollution particles have been shown to translocate into and cross the placenta, which may occur by altering placental DNA methylation patterns that lead to changes in placental function and morphology [21, 22]. DNA methylation, one of the major epigenetic modifications, plays a vital role in the regulation of gene expression, genome stabilization, X-inactivation, genomic imprinting, and chromatin modification [23]. The aberrant methylation in the placenta is more likely to contribute to the onset of diseases such as RSA, preeclampsia, gestational diabetes, and preterm birth (PTB) [24,70]. The average air pollutant exposures were calculated for each participant for each month of the 1 year prior to conception (from 1 month prior to pregnancy to 12 months prior to pregnancy) and early pregnancy was used for subsequent analysis. A logistic regression model adjusting for the age of the mother was used to estimate the odds ratio (OR) and 95% confidence intervals (CI) for increased pregnancy loss with exposure to air pollutants of 77 participants enrolled. We also performed a stratified analysis to investigate the potential health effects modified by season (warm: April to September; cool: October to March). In our study, we assigned 43 subjects including 11 patients with RSA and 32 controls to the warm season and 34 subjects including 20 patients with RSA and 14 controls to the cool season based on the date of surgery. The correlations between methylation levels of IGF2BP1 and air pollutants prior to 1 year to pregnancy were assessed using Spearman correlation coefficients. Methylation levels of IGF2BP1 at differentially methylated CpG dinucleotide (site4, site7, and site9) were ln-transformed to approximate normal distributions. Additionally, we constructed a linear regression model on each CpG dinucleotide of IGF2BP1 adjusting for the age of the mother in 77 participants to evaluate the associations between methylation and air pollution. PLS-PM was carried out to evaluate the potential mediating effect of the methylation level of the IGF2BP1 promoter on the association between air pollution and RSA in 77 participants using SmartPLS (version 3.3.3). In PLS-PM, latent variables are used to consider all relationships between explanatory and manifest variables [71]. To test the overall model fit for PLS-SEM, GoF was calculated to identify the PLS model globally [72]. The Fisher’s exact test, Spearman correlation analysis, t test, Mann–Whitney U test, and chi-square test were performed using the SPSS software (Version 25.0).