Polystyrene Nanoparticles Disrupt Oxidative Phosphorylation and Impair Placental Development in Mice

Abstract

Background: Microplastics and nanoplastics, as pervasive and persistent environmental pollutants, are raising growing concerns regarding their potential risks to reproductive health, particularly pregnancy outcomes. Although the reproductive toxicity of polystyrene nanoplastics (PS-NPs) has been reported, the specific mechanisms underlying their effects on placental development and offspring health following gestational exposure remain unclear. Method: This study aimed to investigate the effects of gestational exposure to PS-NPs of different sizes (50 and 200 nm) and concentrations (1, 3, and 10 mg/mL) on placental function and embryonic development in ICR mice. An exposure model was established via tail vein injection, and samples were collected on embryonic Day 14.5 (E14.5). Results: the exposed groups tended towards increased embryo weight, embryo length, and embryo head circumference. Transcriptomic analysis revealed that PS-NP exposure significantly downregulated the expression of Ndufa5 (a subunit of mitochondrial respiratory chain complex I) and mt-CO1 (a core subunit of complex IV), but upregulated the expression of the genes Cldn1 (tight junction protein) and Erbb3 (receptor tyrosine kinase) in the placenta. Differentially expressed genes were enriched primarily in pathways related to oxidative phosphorylation, the tricarboxylic acid (TCA) cycle, and ErbB signalling. Conclusions: These changes collectively led to decreased mitochondrial ATP production, increased oxidative stress in the placenta, and potentially altered placental barrier function and trophoblast cell proliferation signalling. This study reveals a novel mechanism by which PS-NPs disrupt placental development and embryonic growth through impairment of placental energy metabolic homeostasis and key signalling pathways, thus providing crucial experimental evidence for assessing the reproductive and developmental toxicity of nanoplastics.

1. Introduction

Plastic products are extensively used in modern society for packaging, medical applications, and consumer goods, with global production continually rising. Almost 400 million tons of plastic are produced each year, a mass projected to exceed 1.1 billion tons annually by 2050 [1]. Consequently, plastic pollution has become a severe global environmental issue [2,3]. Under environmental weathering, plastic debris continuously degrades into micro- and nanoscale particles (microplastics and nanoplastics, MNPs), which are widely distributed in water, soil, air, and the food chain [4,5,6]. Epidemiological and biomonitoring studies have confirmed that MNPs can enter the human body through various routes and have been detected in diverse biological samples, including blood, placenta, and even foetal tissues [7,8,9]. Owing to their small size and large specific surface area, nanoplastics can more easily cross biological barriers and accumulate within organs, potentially amplifying their biotoxicity [10].

Based on measured levels of MNPs in human diets, the estimated annual intake via ingestion ranges from 39,000 to 52,000 particles per person, depending on age and sex. When inhalation is included, these estimates increase to between 74,000 and 121,000 particles per year. Moreover, individuals relying only on bottled water for their recommended intake may ingest an additional 90,000 microplastics annually, whereas those consuming only tap water ingest about 4000 particles [11]. Polystyrene (PS) is a common food packaging material and one of the most persistent thermoplastic polymers in terms of biodegradability. It represents a major environmental contaminant today, and its nanoparticles (PS-NPs) pose a significant risk to human exposure [12,13,14]. Existing research indicates that PS-NPs induce multiorgan toxicity, including hepatotoxicity, nephrotoxicity, pulmonary toxicity, and neurotoxicity, in mammalian models, with the disruption of the reproductive system being a particular concern [15,16,17]. Notably, studies have directly demonstrated the ability of MNPs to cross the placental barrier, as evidenced by their detection in human placenta and infant meconium [18]. Animal experiments have also revealed that maternal exposure to PS-NPs may lead to adverse pregnancy outcomes, such as foetal developmental abnormalities and miscarriage [19,20,21,22]. However, the specific molecular mechanisms by which PS-NPs affect placental development and foetal growth, particularly whether and how they interfere with the energy metabolism and cellular functions of this critical organ, are currently unclear.

The placenta is the central organ at the maternal–foetal interface and plays vital roles in maintaining normal pregnancy and foetal growth and development through functions such as substance exchange, endocrine regulation, and immune modulation [23,24]. Normal proliferation, differentiation, and metabolism of placental trophoblast cells are fundamental to appropriate placental function. Recent studies have suggested that MNP exposure may impair the fusion process of placental trophoblast cells and activate endoplasmic reticulum stress pathways, leading to placental structural abnormalities [25]. Nevertheless, the long-term effects of PS-NP exposure on the placenta, particularly on energy metabolism, and the detailed underlying mechanisms remain to be systematically elucidated.

Growing concerns over the safety of plastic exposure during pregnancy underline the urgent need to develop toxicological models that can simulate human pregnancy. Some studies have revealed the effects of microplastic exposure on reproduction using different mouse models and demonstrated that exposure to MNPs can cause reproductive and developmental toxicity in mammals [20,26,27]. Given the structural and functional similarities between mouse and human placentas, in this study, PS-NPs were used as representative nanoplastics, and a gestational exposure model was established in mice to investigate the direct effects of PS-NP exposure on placental histology and embryonic growth and development, thereby simulating the potential risks of long-term PS-NP exposure in women of reproductive age. These findings provide a new scientific perspective for understanding the mechanisms underlying the reproductive and developmental toxicity of nanoplastics.

2. Materials and Methods

2.1. PS-NP Characterization

Polystyrene nanoplastic particles (PS-NPs, 50 nm, batch: 20240401; 200 nm, batch: 20240617) were purchased from Zhongke Leiming (Beijing) Technology Co., Ltd. (Beijing, China). The nanoparticles were characterized using transmission electron microscopy (TEM, Tecnai G2 F20 S-Twin, FEI, Hillsboro, OR, USA) to confirm their spherical morphology and uniform size. PS-NPs were suspended in PBS (Changde, China) to prepare high-, medium-, and low-concentration solutions (10, 3, and 1 mg/mL, respectively) for administration. The zeta potential of the nanoparticle suspensions at these concentrations was measured to assess their dispersion stability in the administration system (Figure 1).

2.2. Experimental Animals and Administration

Specific pathogen-free (SPF) ICR mice (age: 6 weeks old) were purchased from Charles River (Foshan, China). All animal experiments were approved by the Institutional Animal Care and Use Committee of Jinan University (Approval No.: IACUC-20240527-05). After a 2-week acclimatization period in the animal facility of the Experimental Animal Management Center, the 8-week-old female mice were housed with male mice at a 2:1 ratio. The day on which a vaginal plug was detected was designated embryonic Day 0.5 (E0.5). Pregnant mice were randomly divided into the following groups: control group (tail vein injection of PBS), 50 nm PS-NP exposure groups (high, medium, and low concentrations), and 200 nm PS-NP exposure groups (high, medium, and low concentrations). Tail vein injections of 100 μL of the respective PS-NP suspensions were administered on E1.5, E2.5, E5.5, E8.5, E11.5, and E13.5. Maternal body weight, blood pressure, and blood glucose levels were monitored periodically. Euthanasia and sample collection were performed on E14.5 (29 mice of the same pregnancy period in total: 3 mice in the control group, 14 mice in the 200 nm group, 12 mice in the 50 nm group). Maternal serum, placentas, and embryos were collected. The number of embryos and the weights of the embryos and placentas were recorded (Figure 2A).

2.3. Biochemical Analysis

Collected maternal serum was sent to Guangzhou Dingguo Biotechnology Co., Ltd. (Guangzhou, China) The levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatine kinase (CK), creatine kinase-MB isoenzyme (CK-MB), uric acid (UA), and blood urea nitrogen (BUN) were measured using a BK-280 automatic biochemical analyser (BK-280, Shandong Boke Bioindustry Co., Ltd., Jinan, China) with their respective kits.

2.4. Histological Analysis

Placental tissues were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned into 5 μm slices using a rotary microtome (Leica, RM2126RT, Wetzlar, Germany). Haematoxylin and eosin (H&E) staining and periodic acid–Schiff (PAS) staining were performed. Images were captured under a light microscope (Leica DM4000B, Germany) to assess the structure of the placental layers (decidua, spongiotrophoblast, and labyrinth) and the thickness of the trophoblast.

2.5. Transcriptomic Analysis

To comprehensively assess the effects of the PS-NPs on gene expression, placental tissues from the control group and each treatment group were sent to Shanghai Majorbio Biopharm Technology Co., Ltd., (Shanghai, China) for reference-based transcriptome sequencing (NovaSeq X plus platform, Illumina, CA, USA). The screening criteria were an adjusted p value (padj) < 0.05 and an absolute fold change (|log2FC|) ≥ 1. Gene Ontology (GO) functional annotation, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment, and Reactome pathway analyses were performed on differentially expressed genes (DEGs).

2.6. Machine Learning for Screening Key Genes

Four machine learning algorithms were applied—least absolute shrinkage and selection operator (LASSO), random forest, support vector machine–recursive feature elimination (SVM-RFE), and elastic net—to analyse placental DEGs. Analyses were performed in the R environment (version 4.5.1). Genes identified by at least two algorithms were determined as final candidate key genes.

2.7. Statistical Analysis

All measurement data are presented as mean ± standard deviation. Statistical analysis was performed using GraphPad Prism software (9.3.0). Comparisons among multiple groups were conducted using one-way analysis of variance (ANOVA). The significance level was set at * p < 0.05, ** p < 0.01, and *** p < 0.001.

3. Results

3.1. Effects of PS-NP Exposure on Maternal Physiological Parameters and Embryonic Development

Throughout the treatment period, no statistically significant differences in changes in maternal body weight, blood pressure, blood glucose, or trends in liver or kidney weight were observed between any of the PS-NP exposure groups and the control group (Figure 2B–E). The total placental weight did not significantly differ among the groups (Figure 2F).

However, the analysis of embryonic development parameters revealed that compared with those in the control group, embryo weight, body length, and head circumference in both the 50 nm and the 200 nm exposure groups tended to increase, although these differences were not statistically significant. Notably, compared with the high-concentration exposure groups, the medium- and low-concentration exposure groups showed increases in embryo body weight and length, with the low-concentration group reaching statistical significance (Figure 2G–I).

3.2. Serum Biochemical Indicators and Placental PS-NP Accumulation

Maternal serum levels of ALT, AST, CK, CK-MB, UA, and BUN did not significantly differ between the exposure groups and the control group (Supplementary Figure S1), suggesting that liver, kidney, or myocardial damage was not evident under the experimental exposure conditions. Using imaging techniques of hyperspectral microscopy, distinct fluorescent signals from accumulated particles were observed in the trophoblast cells of placentas from the 200 nm PS-NP exposure group, indicating that PS-NPs can enter and accumulate in placental tissue (Supplementary Figure S2).

3.3. PS-NP-Induced Placental Histological Damage

H&E staining revealed that compared with the control group, the PS-NP-exposed group exhibited significant thinning of the trophoblast layer in the placenta (decreased trophoblast/total area ratio), a reduction in trophoblast cell number, and a disordered structural arrangement (Figure 3A1–D). These results indicate that PS-NP exposure interferes with normal placental development and structural organization.

3.4. Reference-Based Transcriptomic Analysis of Placenta

Transcriptome sequencing revealed a total of 193 significant DEGs. Analysis of the placental transcriptome sequencing results revealed significant changes in gene expression in the treatment groups compared with the control group (Figure 4A–C). In the 50 nm exposure group, 15 genes were significantly upregulated, and 10 genes were significantly downregulated. In the 200 nm exposure group, 65 genes were significantly upregulated, and 116 genes were significantly downregulated, with 13 genes being common. Thirteen genes were shared between the 200 nm exposure group and the 50 nm exposure group.

GO functional annotation and enrichment analysis revealed that these DEGs were significantly enriched in terms such as “oxidoreductase activity,” “ion binding,” and “protein binding” (Figure 5A). KEGG pathway enrichment analysis revealed that the DEGs were highly enriched in the “oxidative phosphorylation” and “chemical carcinogenesis–reactive oxygen species” pathways (Figure 4D and Figure 5B). Reactome pathway analysis further confirmed that “citric acid (TCA) cycle and respiratory electron transport”, “respiratory electron transport”, and “ATP synthesis” were the most significantly affected pathways (Figure 6B).

Notably, the expression of several genes encoding key components of the mitochondrial electron transport chain, including Ndufa5 (complex I subunit), mt-CO1 (complex IV subunits), and mt-CO2, and mt-Cytb (complex III), significantly decreased (Figure 4E). Additionally, the expression of the tight junction protein-encoding gene Cldn1 and the receptor tyrosine kinase gene Erbb3 was significantly upregulated.

3.5. Machine Learning Identifies Key Candidate Genes

Four different machine learning algorithms were comprehensively applied to further analyse placental DEGs (R, version 4.5.1): LASSO analysis screened 5 candidate key genes, random forest analysis (leave-one-out cross-validation, alpha = 1) screened 15 genes, SVM-RFE (ntree = 1000) screened 15 genes, and elastic net analysis (leave-one-out cross-validation, fixed alpha = 0.5) screened 29 genes. By analysing the results from all four algorithms, genes selected by at least two algorithms were defined as final candidate key genes, yielding a final set of 16 genes (Figure 7).

Integrating the results from the LASSO, random forest, SVM-RFE, and elastic net algorithms revealed 16 candidate key genes commonly identified by at least two algorithms. Among these genes, Bco1, mt-Nd1, mt-CO1, Ipp, Fkbp6, and others significantly overlapped with the results of the transcriptomic enrichment analysis, further confirming their central role in PS-NP-induced placental toxicity (Figure 7A–D). Protein–protein interaction (PPI) network analysis revealed that mitochondrial-related genes such as Ndufa5, mt-CO1, and mt-Nd1 were located at the network core, suggesting that they play key roles through energy metabolism modules (Figure 7E,F).

4. Discussion

In this study, the effects of PS-NPs on placental development and embryonic growth in mice were systematically evaluated using a gestational tail vein exposure model. We found that although exposure did not cause significant maternal systemic toxicity, it directly led to microstructural damage in the placenta and abnormal acceleration of embryonic growth (as evidenced by increases in embryo weight, length, and head circumference); a similar phenomenon has been observed in another study [25]. One study reported that subchronic exposure to PS-NPs increased foetal resorption and significantly raised foetal weight in neonatal mice [28]. Similarly, earlier work showed that continuous gestational exposure to PS-NPs (50 nm, 100 mg/kg/day) for 14 days induced miscarriage in mice [29]. In contrast, other studies found that maternal exposure to PS-NPs (100 nm, 10 mg/L) via drinking water from GD0 to GD17 led to foetal growth restriction [30]. Another experiment in which pregnant mice were orally administered PS-NPs (100 nm, 1, 10, and 100 mg/kg/day) from GD0.5 to GD17.5 demonstrated that PS-NPs caused placental injury and metabolic disturbances, resulting in adverse pregnancy outcomes and foetal growth restriction [31]. Meanwhile, a study observed that acute maternal pulmonary exposure to nanoplastics (intratracheal instillation on gestational day (GD) 19) decreased foetal and placental weights and increased foetal resorption frequency, but direct infusion of nanopolystyrene particles into the uterine artery did not reduce fluid flow to the foetal compartment [32]. This phenotype differs from the foetal growth restriction reported in most previous studies [33,34], suggesting that the reproductive toxicity of PS-NPs may be complex and dependent on the exposure route and dose.

Different exposure routes in a study each have distinct advantages and disadvantages. For example, oral gavage better simulates physiological exposure scenarios, while administration via drinking water is more operationally straightforward. Building on the previous findings and experience of our research, and in order to minimize the impact of individual differences in food intake, absorption, and metabolism on experimental outcomes, this study selected intravenous injection (tail vein injection) as the exposure route. This route enables precise control over the administered dose, thereby ensuring the accuracy and consistency of exposure levels. Previous research reported that the average global intake of microplastics by humans is between 0.1 and 5 g per week, equivalent to approximately 0.2–11.9 mg/kg/day [18]. Based on the body surface area (BSA) standardization method (with a mouse-to-human conversion factor of 12), the estimated microplastic exposure dose for mouse assay ranged from 2.4 to 142.8 mg/kg/day. Based on an average weight of 30 g during pregnancy, the exposure range is 0.07–4.3 mg/day [35,36,37]. Calculated based on an average weight of 30g during pregnancy, the exposure range is 0.07–4.3 mg/day, which can cover the dose range of PS-NPs administered to mice in this study (0.1–1 mg/dose).

Previous studies have shown that exposure to nanoparticles can affect reproductive toxicity, such as placental development and foetal growth, through multiple mechanisms. One study showed that exposure to PS-NPs during pregnancy induced placental oxidative damage and disrupted ATP production, contributing to placental dysplasia and barrier injury. Excessive oxidative stress, reactive oxygen species (ROS) accumulation, and lipid peroxidation are closely linked to placental dysfunction and related pathologies [38,39]. Notably, GPX4—a key inhibitor of ferroptosis—is essential for the survival of placental trophoblast cells. Downregulation of GPX4 impairs the conversion of GSH to GSSG, which in turn inhibits the clearance of ferroptotic signals and aggravates cellular ferroptosis [39,40,41,42]. One study found that PS-NP exposure significantly downregulates the expression of NAD⁺ synthases NMNAT1, NMNAT2, and NMNAT3, leading to NAD⁺ deficiency, mitochondrial dysfunction, and increased lipid peroxidation, and GPX4 expression is markedly reduced, triggering ferroptosis in the placenta [34]. In another study, PS-NPs were found to upregulate STAM2 expression in trophoblasts. This inhibited endoplasmic reticulum calcium efflux by promoting lysosomal degradation of IP3R3 while simultaneously facilitating extracellular calcium influx via TRPV6. Together, these changes promoted trophoblast proliferation, migration, and invasion [31]. It was found that ingestion of microplastics (MPs) significantly elevates ROS levels in oocytes and embryos, resulting in oxidative stress, mitochondrial dysfunction, and apoptosis [43]. Additionally, a 24 h exposure to 40 nm particles at 100 μg/mL downregulated protein expression of α5 and α1 integrin subunits, N-cadherin, matrix metalloproteinase-2, and macrophage migration inhibitory factor in trophoblast models. It also upregulated the epithelial marker E-cadherin. These alterations likely impaired the migration of HTR-8/SVneo cells and reduced the invasive potential of HTR-8/SVneo spheroids [14]. PS-NPs were further shown to impede cell migration and invasion, with the effect being exacerbated upon MDM2 knockdown or ROCK1 inhibition. Exposure to PS-NPs also impaired trophoblast proliferation, promoted apoptosis, and disrupted mitochondrial function, as evidenced by elevated ROS, reduced mitochondrial membrane potential, and altered gene expression. Increased autophagy activity and the release of inflammatory cytokines indicated elevated cellular stress [44]. Additionally, studies have shown that exposure induced inflammation of the ovaries and reduced the quality of oocytes in mice [45].

The most significant finding of this study is that PS-NP exposure specifically disrupted the energy metabolism hub of the placenta. Transcriptomic data consistently revealed impaired mitochondrial oxidative phosphorylation function. The downregulation of the key genes Ndufa5 (Complex I) and mt-CO1 (Complex IV) weakens the efficiency of ATP generation from two critical nodes—the beginning and end of the electron transport chain. This finding is logically consistent with the structural damage (trophoblast thinning) observed in the placenta, as trophoblast cell proliferation, invasion, and hormone synthesis are high-energy-demanding processes heavily reliant on mitochondrial energy supply [46,47]. Insufficient energy supply may directly lead to the attenuation of these cellular functions, thereby affecting placental villous development and maturation [48]. In the FGR placenta, the expression imbalance of Ndufa5 and other subunits of Complex I (such as NdufS3 and NdufS6) jointly disrupts the normal assembly and electron transfer functions of complex I. The defect of mitochondrial Complex I mainly impairs the function of the cytotrophoblast, which directly leads to placental dysplasia and weight loss [47,49]. Downregulation of mt-CO1, encoding the core catalytic subunit of cytochrome c oxidase (Complex IV), directly impairs the activity of the terminal enzyme in the mitochondrial respiratory chain. This decreases the mitochondrial membrane potential by sharply reducing ATP production, thus potentially increasing the generation of harmful ROS. High-energy processes in trophoblast cells (such as invading uterine spiral arteries and synthesizing hormones) are severely inhibited, leading to placental hypoplasia and functional insufficiency [50].

In addition to alterations in energy metabolism, this study revealed two other important alterations. Cldn1 encodes a tight junction protein, and its upregulation may alter the physical permeability of the placental barrier. Abnormal Cldn1 expression has been observed in placental studies of both foetal growth restriction and macrosomia [51]. We speculate that PS-NP-induced Cldn1 upregulation may be a compensatory or dysregulated response of the placenta to particle invasion or metabolic stress. However, this potential “oversealing” could affect the optimal exchange efficiency of nutrients and gases between mothers and foetuses, which might partly explain the altered embryonic growth pattern. On the other hand, the upregulation of Erbb3 revealed that PS-NPs interfered with placental developmental signalling pathways. Erbb3 is a key receptor that regulates cell proliferation, differentiation, and survival [52]. Its abnormal expression may disrupt the normal fate determination of trophoblast cells, acting in concert with mitochondrial dysfunction to exacerbate placental developmental insufficiency. Another study also indicated that maternal MNP exposure during pregnancy disrupts trophoblast syncytialisation by activating the PERK/eIF2α/ATF4 signalling pathway [28,31].

Therefore, a comprehensive action model is proposed: PS-NPs enter the maternal circulation and accumulate in the placenta directly or indirectly (e.g., by inducing oxidative stress) to inhibit the expression of key mitochondrial respiratory chain genes, leading to mitochondrial dysfunction and subsequently impeding placental energy metabolism. The resulting energy deficit weakens trophoblast cell function, causing structural developmental damage, and may also trigger cellular stress responses, leading to compensatory changes in barrier proteins (e.g., Cldn1) and growth signalling receptors (e.g., Erbb3). The synergistic disruption of energy metabolism, barrier function, and growth signalling collectively determines placental functional abnormalities and ultimately leads to aberrant embryonic development.

5. Conclusions

This study confirms that gestational exposure to PS-NPs can disrupt placental oxidative phosphorylation function by interfering with the expression of core mitochondrial energy metabolism genes (Ndufa5 and mt-CO1) while simultaneously affecting the expression of genes related to tight junctions (Cldn1) and cell proliferation signalling (Erbb3). This synergistic interference of multiple targets and pathways ultimately leads to placental structural damage and altered embryonic growth trends. This research, from a novel perspective on energy metabolism, provides an in-depth understanding of the potential molecular mechanisms underlying the reproductive toxicity of nanoplastics and offers important experimental data to support scientific health risk assessment and the development of related protective strategies.

However, this study has certain limitations. The use of engineered PS-NPs and the tail vein injection administration route in the animal model cannot fully replicate actual human exposure scenarios, differing from the complex, low-dose, long-term, multiroute mixed exposures encountered in reality. Furthermore, the specific effects of nanoparticle exposure on offspring have not been investigated and evaluated. Future research should incorporate plastic particles with different properties, employ exposure modes closer to environmental realities, and investigate their combined toxic effects with other environmental pollutants, and further evaluate the impact of nanoparticle exposure on offspring.