Effect of PM2.5 Exposure On Gestational Hypertension and Fetal Size in Preeclampsia-Like Rats

Studies have reported that gestational PM 2.5 exposure is associated with preeclampsia (PE) and fetal growth restriction (FGR). However, whether maternal exposure to PM 2.5 causes adverse pregnancy outcomes is still largely unknown. Pregnant Sprague-Dawley rats were exposed to either ltered (FA) or PM 2.5 air during the whole pregnant period. A PE-like rat model was established by intraperitoneal injection of L-NAME (300 mg/kg) from GD12 to until GD20. Systolic blood pressure (SBP), weight gain, pup weight and placental weight were measured. The percentages of rat Treg/Th17 cells and Th17-related cytokines were examined by ow cytometry. Gene expression proles were analyzed by microarray, and the expression of differentially expressed genes were validated by qRT-PCR. The results showed that maternal PM 2.5 exposure had no effect on SBP but was associated with LBW and a higher labyrinth/basal zone ratio. The percentages of splenic Th17 cells from the PM 2.5 group in PE-like rats were higher than those from the FA or PM 2.5 groups in healthy controls. A signicantly decreased Treg/Th17 cell ratio was found in the PM 2.5 group in PE-like rats. The mRNA expression of Foxp3 was downregulated, while the mRNA expression of ROR α and ROR γτ was upregulated after PM 2.5 exposure. Furthermore, we observed that both the mRNA and protein expression of TNF-a, CCL2, CCL3 and CCR1 increased in the PM 2.5 groups. Our study suggested that systemic inammation may contribute to the development of FGR associated with PM 2.5 exposure throughout pregnancy. the dark with Fixation/Permeabilization Diluent (eBioscience, San Diego, CA, USA). The cells were stained with anti-mouse/rat Foxp3-PE (eBioscience, San Diego, CA, USA) or anti-mouse/rat IL-17A-APC (eBioscience, San Diego, CA, USA) antibodies. Finally, the cells were resuspended in 500 µl of PBS for subsequent ow cytometric analysis. All data were acquired on a FACScalibur (BD Biosciences, San Jose, CA) and processed using BD FACS Diva software (BD Biosciences, San Jose, CA). Serum levels of IL-17A, IL-6 and IL-22 were measured by ow cytometry using a multi-LegendPlex kit (Biolegend) according to the instructions of the manufacturer.


Introduction
Air pollution is a major global public health issue and is estimated to cause 4.2 million premature deaths globally (Cohen et al. 2017). The composition of air pollutants is complex and changeable, and particles with an aerodynamic diameter of 2.5 µm or less (PM 2.5 ) are the main pollutants affecting urban air quality worldwide. Because of its smaller size, larger surface area, easier enrichment of toxic chemicals and various microorganisms, the harm of PM 2.5 to public health has become a social hot point.
There is increasing epidemiological evidence that maternal exposure to PM 2.5 is associated with high risks of pregnancy complications and adverse pregnancy outcomes, including preeclampsia (PE) ( (Liu et al. 2016), which does not represent real-world exposure and may overestimate the hazards of exposure. Hence, the relationship between maternal exposure to PM 2.5 and adverse pregnancy outcomes and its underlying mechanism need to be further explored.
Several studies have suggested that maternal PM 2.5 exposure may elicit systemic in ammation and promote the release of in ammatory mediators, such as proin ammatory cytokines and acute phase proteins, to the blood stream, which interfere with maternal and fetal growth (Wang et al. 2021; Xia et al. 2019). Previous studies reported that PM 2.5 exposure impair differentiation of Treg cells, promote differentiation of Th17 cells, and aggravate the in ammatory response (Cong et al. 2020; L Sun et al. 2020). Meanwhile, the balance between Treg and Th17 cell immunity is particularly important for normal placental development and pregnancy maintenance. It has been shown that Treg/Th17 imbalance is involved in PE and FGR pathogenesis. However, fewer studies have explored the role of PM 2.5 exposure and Treg/Th17 cells in PE and FGR.
In the present study, we evaluated whether maternal PM 2.5 exposure changed the proportion and differentiation of Treg/Th17 cells in preeclampsia-like rats. Therefore, pregnant rats were exposed to either airborne PM 2.5 or ltered air (FA) throughout pregnancy in a whole-body exposure chamber. The effects of PM 2.5 exposure on preeclampsia and fetal outcome were investigated. We evaluated systematic and local in ammation after PM 2.5 exposure. The analyses of global gene expression pro les in placentas of PM 2.5 or FA exposed rats were performed using microarray.

Animals
Animal care and use were carried out in accordance with guidelines approved by Capital Medical University, Beijing, China. All protocols were approved by the Institutional Animal Care and Use Committee (IACUC) at Beijing Luhe Hospital, Capital Medical University. Sprague-Dawley rats purchased from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China) were used in this study.

Preparation of PE-like rat model
Female Sprague-Dawley (SD) rats weighing 200-250 g were raised in a light-and humidity-controlled room with free access to food and water. The day on which pregnancy was con rmed by the presence of vaginal spermatozoa was designated GD0. Pregnant rats were randomly divided into four groups as follows: ltered-air+NaCl (FN), PM 2.5 -air+NaCl (PN), ltered-air+L-NAME (FL), and PM 2.5 -air+L-NAME (PL) groups (n=5 each group). Rats in FL and PL group were injected with 300 mg/kg L-NAME from GD12 to GD20 to establish preeclamptic models. PM 2.5 exposure Pregnant rats were exposed to either ambient PM 2.5 or FA from GD0 to GD20 (September-October 2016) in a "real-world" airborne PM 2.5 exposure system for 24 h per day and 7 days per week as described in previous studies (Guan et al. 2019;Liu et al. 2020), which is located in the Institute of Neuroscience of Beijing Luhe Hospital. For PM 2.5 , a cyclone was used in the PM 2.5 -exposed group to remove particles larger than 2.5 μm. For FA, a high-e ciency particulate air lter (HEPA) was positioned in the inlet valve to the exposure system to remove all of the particles from the air stream. Real-time PM 2.5 concentration were measured continuously using monitoring equipment (JK-PDR1500, Jinan Qianya Puri cation Equipment Co., Ltd.) installed inside and outside the exposure facility.

Blood pressure measurement
The blood pressure of rats was measured by the noninvasive blood pressure methodology (Zhongshi Scienti c Technology Co., Ltd., Beijing, China), which consists of placing a cuff on the animal's tail to occlude blood ow. All rats were weighed and had their blood pressure measured every 3 days until delivery. Three blood pressure values were adopted within the variation <5 mmHg, and the average was taken for the blood pressure values.

Sample collection and histological analysis
On GD21, the animals were euthanized, and the blood, spleen and placentas were rapidly removed for further analyses immediately after delivery. The following parameters of pregnancy outcomes were evaluated: fetal weight and placental weight. Sections of rat placentas at 5 μm thickness were stained with hematoxylin and eosin (H&E) according to the standard H&E protocol. All sections were from the middle of the respective placentas. The areas of the labyrinth and basal zone were determined using sections with the maximum parts for the layer of the whole placenta in ImageJ (NIH, USA). A single observer blinded to the experimental conditions at the time of the assessments performed all ImageJ assessments and quanti cation. The assessments were repeated a second time with a second blinded observer to con rm all results. The mean area per group was calculated from ve individuals.

Flow cytometry
The rat spleen tissues were gently teased to release cells in PBS. The splenic mononuclear cells were isolated by density centrifugation. Isolated splenic mononuclear cells were stimulated and incubated with Cell Stimulation Cocktail (plus Protein Transport Inhibitor) (eBioscience, San Diego, CA, USA) for 6 h. Then, the cells were xed and permeabilized for 30 min at room temperature in the dark with Fixation/Permeabilization Diluent (eBioscience, San Diego, CA, USA). The cells were stained with antimouse/rat Foxp3-PE (eBioscience, San Diego, CA, USA) or anti-mouse/rat IL-17A-APC (eBioscience, San Diego, CA, USA) antibodies. Finally, the cells were resuspended in 500 µl of PBS for subsequent ow cytometric analysis. All data were acquired on a FACScalibur (BD Biosciences, San Jose, CA) and processed using BD FACS Diva software (BD Biosciences, San Jose, CA). Serum levels of IL-17A, IL-6 and IL-22 were measured by ow cytometry using a multi-LegendPlex kit (Biolegend) according to the instructions of the manufacturer.

Microarray analysis and series test of cluster (STC) analysis
For microarray analysis, placental samples from the four groups were used to extract total RNA. Total RNA was quanti ed by a NanoDrop ND-2000 (Thermo Scienti c), and RNA integrity was assessed using an Agilent Bioanalyzer 2100 (Agilent Technologies). The expression pro les of the placental mRNAs were detected using an Agilent SurePrint G3 Rat GE V2.0 Microarray (OE Biotech, Shanghai, China). The threshold for up-and downregulated genes was a fold-change (FC) value of 2.0 or greater.
Gene Ontology and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were applied to determine the roles of these differentially expressed mRNAs that exhibited a signi cance value of P<0.05. The series test of cluster (STC) algorithm of gene expression was performed to enrich the expression tendency of the genes and narrow the target genes with greatly signi cant differential expression among the four groups.

RNA isolation and real-time RT-PCR
Total RNA was isolated from the placenta using RNeasy kits (Qiagen) according to the manufacturer's protocol, and 1 µg of total RNA was reverse transcribed with TransScript II All-in-One First-Strand cDNA Synthesis SuperMix for qPCR (Transgene, Beijing, China). cDNA was quanti ed using SYBR Green PCR Master Mix (Applied Biosystems, USA). The sequences of the primers used for real-time PCR are shown in Table 1. The quantitative fold changes in mRNA expression were determined relative to glyceraldehyde 3phosphate dehydrogenase (GAPDH) mRNA levels in each corresponding group.

ELISA
Placental tissues were weighed, placed in lysis buffer, homogenized and centrifuged at 10 000 ×g for 5 min. The supernatants were collected, and the ELISA assays were conducted by quantitative sandwich enzyme immunoassays using commercial ELISA kits according to the manufacturer's protocol (Cloud-Clone Corp, Mbbiology, Wuhan, China). For each sample, the assay was conducted in triplicate, with duplicate standard curves.

Statistics
Data are presented as the mean ± standard error of the mean. Statistical analysis was performed using Statistical Package for Social Science (SPSS) for Windows (version 22.0 software, SPSS Inc., Chicago, IL, USA) and GraphPad Prism software (version 7.0, GraphPad). The differences were analyzed by ANOVA with Student-Newman. A p value less than 0.05 was considered signi cant.  Figure 1B). PM 2.5 exposure during pregnancy has no effect on maternal SBP or gestational weight gain As shown in Figure 2A, there was no difference in the baseline SBP among the four groups. The FN group had a relatively consistent SBP throughout the entire gestation period (SBP of 107.00±7.04 mmHg on GD8 and of 109.90±4.82 mmHg on GD17) ( Table 2). Among the four groups, three of the groups (PN, FL, and PL) demonstrated similar maternal systolic blood pressure changes during pregnancy, with an increase in blood pressure from GD8 to GD20. The SBP of the PN group slightly increased from 111.20±3.96 mmHg on GD8 to 126.60±6.34 mmHg on GD17; however, there was no signi cant difference between the PN and FN groups. There was a signi cant increase in SBP in both the FL and PL groups (from 114.3±6.79 mmHg on GD8 to 132.10±3.15 mmHg on GD17 and from 104.2±2.83 mmHg on GD8 to 131.50±8.15 mmHg on GD17, respectively) compared with that in the FN group, which suggests that administration of L-NAME successfully induced a PE-like animal model ( Figure 2). However, there was no signi cant difference between the PL and FL groups. These ndings indicate that maternal PM 2.5 exposure during pregnancy did not increase the blood pressure of either pregnant or preeclamptic rats.
As shown in Figure 2B, we also observed less net weight gain in the other three groups than in the FN group, but the differences were not signi cant ( Table 2). The results indicated that maternal PM 2.5 exposure has no effect on gestational weight gain.
Maternal PM 2.5 exposure results in decreased pup weight and an altered labyrinth-to-basal zone ratio To evaluate the effect of PM 2.5 exposure on the fetus, we evaluated the development of the fetus and placenta on GD21. Pups delivered from the PL group were characterized by fetal growth restriction and malformation ( Figure 3A). The PL group had a signi cantly lower pup weight than both the FN group (p<0.01) and PN group (p<0.01) ( Figure 3B), whereas PM 2.5 exposure only throughout the gestational period did not signi cantly affect pup weight. Additionally, we did not observe any signi cant difference in placental weight. These data suggested that ambient PM 2.5 exposure during pregnancy might affect birth weight.
Histology of the placenta was performed to compare the morphology of the labyrinth and basal zones in all groups. The relative area of the labyrinth zone showed no signi cant change in the other three groups compared to that in the FN group. The relative area of the basal zone decreased from the FN to the PL group ( Figure 3C). The ratio between the labyrinth and the basal zone was increased in the FL and PL groups compared to that in the FN group ( Figure 3D). Additionally, the labyrinth zones of the placentas from the PM 2.5 exposure groups had fewer red blood cells than those from the FN group, suggesting the potential for less e cient oxygen and nutrient exchange ( Figure 3E). The basal zones of the placentas from the PM 2.5 exposure groups had larger and more dispersed trophoblast giant cells than those from the FN group, which also displayed more intense neutrophil in ltration ( Figure 3E). Together, these results suggest that maternal PM 2.5 exposure could lead to changes in placental structure and subsequently decrease pup weight.
Maternal PM 2.5 exposure results in a reduced Treg/Th17 ratio in spleen tissue in PE-like rats To assess the effect of PM 2.5 exposure on Th17 cell expansion, spleen samples were collected for ow cytometry analysis. Th17 cells were de ned as CD4 + IL-17A + . Gating strategies for Th17 cells are shown in Figure 4A. The prevalence of Th17 cells refers to the ratio of CD4 + IL-17A + cells to the total amount of CD4 + T lymphocytes. As shown in Figure 4B, the prevalence of splenic Th17 cells showed an increase in the other three groups compared with that in the FN group, but the differences were not signi cant. We further determined the Treg/Th17 cell ratio and found a signi cant decrease in the Treg/Th17 cell ratio in the PL group compared with that in the FL group. These results revealed that maternal PM 2.5 exposure resulted in a reduced Treg/Th17 ratio in PE-like rats.
In the present study, the serum levels of IL-17A, IL-6 and IL-22 were measured in four groups using multifactor ow cytometry. Levels of systemic in ammatory reactions are increased in rats exposed to PM 2.5 , especially after L-NAME treatment Fine particulate matter exposure may promote systemic in ammation reactions. To explore changes in mRNA expression after PM 2.5 exposure, we conducted microarray analyses on the placentas from 4 groups and examined the most differentially expressed genes (DEGs) using bioinformatic analyses. Figure 5A shows the heatmap representation of the top 1000 DEGs among the 4 groups. Red represents upregulated genes, and blue represents downregulated genes. A total of 597 genes showed upregulation, and 473 genes showed downregulation between the PN and FN groups, whereas, as shown in Figure 5B, 1168 genes were upregulated and 1057 were downregulated between the PL and FL groups. The results of GO analysis showed that the upregulated DEGs were signi cantly enriched in the cellular response to interferon-gamma, neutrophil chemotaxis, monocyte chemotaxis and immune response ( Figure 5C). STC analysis results of the screened DEGs are presented in Supplemental Figure 1. A total of 12 signi cant pro les were identi ed. Our STC analysis showed that DEGs in Pro le 40 were persistently upregulated from the FN to PL group ( Figure 6A), which included some classic in ammatory genes, such as TNFa, CCL2, CCL3 and CCR1. Figure 6B shows the heatmap representation of 13 DEGs from Pro le 40.
To examine whether the expression of the characteristic transcription factors of Treg and Th17 cells underlies the reduced Treg/Th17 cell ratio in preeclamptic rats after PM 2.5 exposure, we further determined Foxp3, ROR α and ROR γτ mRNA expression using real-time qRT-PCR. As shown in Figure 6C, Foxp3 mRNA expression was the lowest in the placentas of the PL group, while ROR α and ROR γτ mRNA expression showed increased expression from the FN to the PL group. Furthermore, we observed that both the mRNA and protein expression of TNF-a, CCL2, CCL3 and CCR1 increased from the FN to the PL group.

Discussion
In this study, we explored the effect of PM 2.5 on the blood pressure of PE-like rats and their perinatal outcomes. PM 2.5 is a heterogeneous mixture of ne particulate matter and is more harmful to health than other air pollutants. Recently, epidemiologic studies revealed positive associations between maternal exposure to PM 2.5 during pregnancy and adverse pregnancy outcomes, especially gestational In the present study, we observed that the BP of the PN group was slightly higher than that of the FN group in late pregnancy, while the difference was not signi cant.
Furthermore, there was no difference between the PL and FL groups, which suggests that maternal exposure to PM 2.5 does not increase the risk of preeclampsia. Differences in exposure protocols may in part explain some of these differing ndings. The major methods of PM 2.5 exposure in previous in vivo experiments are intratracheal inhalation and intratracheal instillation. Both methods have some common weaknesses, such as high dosage exposure, which cannot simulate real-world human exposure.
One of the most important ndings is that we observed that PE-like rats exposed to PM 2.5 have a signi cantly reduced fetal weight and alterations in placental histology compared to those of pregnant rats exposed to PM 2.5 . An increasing number of studies have demonstrated that maternal exposure to PM 2.5 has an adverse effect on fetal development and is closely related to adverse birth outcomes, such as preterm birth, still birth, term low birth weight and small for gestational age (Basu et  Previous studies demonstrated that the balance between Treg and Th17 cells plays a critical role in the pathogenesis of PE and FGR (Cotechini et al. 2014). Prins JR reported that the expression of IL-6, which stimulates the differentiation of Th17 cells and inhibits the generation of FOXP3 + Treg cells, increased in chorionic villous tissues of women who develop preeclampsia associated with FGR. Another study using the reduced uterine perfusion pressure (RUPP) rat model of preeclampsia found that adoptive transfer of RUPP Th17 cells induced intrauterine growth restriction and increased proin ammatory cytokines in normal pregnant rats. The systemic in ammatory is also considered to be the most important mechanism underlying PM 2.5 -induced adverse health problems. In this study, we investigated the distribution of T cell subtypes in a preeclamptic rat model after PM 2.5 exposure and found that Th17 cells were increased in both the PN and PL groups. The Treg/Th17 ratio was signi cantly decreased in the PL group compared with that in the FL group. Furthermore, our study demonstrated a signi cant increase in the levels of proin ammatory cytokines, including IL-17A, IL-6 and IL-22, after PM 2.5 exposure. As important proin ammatory cytokines, IL-17A and IL-22 are secreted by Th17 cells, and IL-6 is involved in inducing the development of Th17 cells and inhibiting Treg differentiation. The collective pattern of changes in serum cytokines may indicate that systemic in ammation increased in PE-like rats exposed to PM 2.5 due to exacerbating the imbalance of Treg and Th17 cells. Therefore, PM 2.5 exposure exacerbated Treg and Th17-mediated immunological dysfunction.
Our placental microarray data revealed that the differentially regulated genes were involved in more than 20 biological processes that can be clustered into the in ammatory response and neutrophil chemotaxis. The transcription factors Foxp3 and ROR γτ are the lineage-speci c markers for the differentiation of Treg and Th17 cells, respectively. Our data showed that Foxp3 mRNA decreased, while ROR α and ROR γτ mRNA signi cantly increased in placental tissues from the PL group. Furthermore, we observed that exposure to

Conclusion
In summary, this study demonstrates that maternal exposure to PM 2.5 may decrease pup weight in PE-like rats. The characteristic patterns of changes in cytokines in the PE-like rats exposed to PM 2.5 reported in our study suggested that gestational PM 2.5 exposure may promote systemic and local in ammation in PE-like rats and result in placental insu ciency by exacerbating the imbalance of Treg and Th17 cells. These ndings shed light on the potential mechanisms by which PM 2.5 may cause or exacerbate FGR.

Declarations
Ethics approval and consent to participate

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analyzed during this study are included in this published article and its supplementary information les.
2 . Sun M, Yan W, Fang K, Chen D, Liu J, Chen Y et al (2020) The correlation between pm2.5 exposure and hypertensive disorders in pregnancy: A meta-analysis. Daily concentrations of PM2.5 measured in polluted chambers during the experimental period. (A) Experimental design: rats were exposed to ltered-air or PM2.5 air during the whole pregnancy. Blood pressure was measured every three days. From GD12 to GD20, 300 mg/kg body weight L-NAME was subcutaneously injected every day. At GD21, rats were euthanized for spleen harvesting. (B) The mean concentrations of PM2.5 chambers were signi cantly higher than those of ltered-air chambers. ***p<0.001.

Figure 2
The variation tendency of SBP (A) and net weight gain (B) during pregnancy in the four groups. * p<0.05 # p<0.05 vs the FN group, ** p<0.01 ##p<0.01 vs the FN group. Values are expressed as the mean ±SEM. Groups: FN, rats treated with ltered air throughout pregnancy; PN, rats treated with PM2.5air throughout pregnancy; FL, rats treated with L-NAME (300 mg/kg body weight/day) starting from GD10 in a ltered-air chamber throughout pregnancy; PL, rats treated with L-NAME (300 mg/kg body weight/day) starting from GD10 in a PM2.5-air chamber throughout pregnancy. GD, gestational day; L-NAME, N-nitro-L-arginine methyl ester; SBP, systolic blood pressure. (B). The net weight gain of each rat was observed throughout pregnancy (the difference between GD20 body weight and GD0body weight) and after GD12 (the difference between GD20 body weight and GD12 body weight). The ratio of labyrinth to basal zone areas. (E) Hematoxylin and eosin (H&E) staining of placentas was subsequently performed in the FN, PN, FL and PL groups. Top row, lower magni cation views. Bottom two rows, higher magni cation views of the labyrinth (middle row) and decidua (bottom row). Data are presented as the mean ± SEM. * p<0.05, **p<0.01, ***p<0.001.