Impact of mono(2-ethylhexyl) phthalate (MEHP) on the development of mouse embryo in vitro

Abstract

Mono(2-ethylhexyl) phthalate (MEHP) is the most studied metabolite of di(2-ethylhexyl) phthalate (DEHP), a phthalate found in cosmetics, flooring, paints, and plastics products, including toys and medical tubing. Humans are frequently exposed to this compound due to its ubiquitous presence in our environment. DEHP and MEHP are known to be endocrine-disrupting chemicals and exposure levels have been associated to decreased reproductive success. However, few studies have focused on the direct effects of MEHP on embryos. The present study investigated effects of MEHP (0.1, 1, 10, 100 and 1000 μM) on mice preimplantation embryonic development, evaluating percentage of blastocyst formation, hatching from zona pellucida, methylation-related genes, cell lineage commitment, micronucleation, and adherens junction marker at different stages of development during in vitro culture for 6 days. We show MEHP negatively impacts embryo competence by reducing blastocyst formation and hatching at 100 and 1000 μM. In addition, 100 μM MEHP increases the expression of Tet3 gene in blastocysts, which is related to a reduction of DNA methylation, an important mechanism regulating gene expression. Exposed embryos that completed the hatching process in groups 0.1, 1 and 10 μM MEHP had similar number of inner cell mass and trophectoderm cells compared to the control, while micronucleation occurrence and E-cadherin expression was not affected in exposed morulae by MEHP at 10 or 100 μM. Our results showed that high concentrations of MEHP can negatively impact embryo development. New studies unveiling the mechanism of toxicity involved and encompassing further developmental stages are warranted for further understanding.

1. Introduction

The chemical industry produces a variety of compounds to facilitate and improve modern-day life. However, many of them have not been evaluated properly and completely for their toxicological risks before being released onto the market. One example is di(2-ethylhexyl) phthalate (DEHP), which has large-scale use to provide flexibility to plastics and is commonly found in food packaging, toys, paints, flooring, personal care products, and medical supplies, such as medical tubing and blood and dialysis bags (Mariana et al., 2016; Benjamin et al., 2017). DEHP can leach from the plastic matrix of these materials and enter the human body via ingestion, inhalation, and dermal absorption (Hannon et al., 2016; Benjamin et al., 2017). Mono(2-ethylhexyl) phthalate (MEHP) is the first active metabolite generated after DEHP intake. For this reason, it is the most common metabolite of DEHP used for in vitro experiments (Koch et al., 2005; Wang et al., 2012; Benjamin et al., 2017).

Humans are continually exposed to DEHP, and consequently to MEHP. Of concern, some studies identified women having higher exposure levels than men, especially professionals working in beauty salons, due to occupational exposure to cosmetic products (Kolena et al., 2017; Boyle et al., 2021). Average daily exposure to DEHP is estimated to range from 3 to 30 μg/kg/day, while occupational exposure can reach 200 μg/kg/day (Hannon et al., 2016). Urinary levels of MEHP for the average adult population ranged from 1.1 – 3.6 ng/mL (0.004 – 0.013 μM) in studies conducted in the United States and Germany (Koch et al., 2017; Li et al., 2020). However, higher values have been detected in the human population, for example, a maximum of 114 ng/mL (0.41 μM) and 325 ng/mL (1.17 μM) were detected in urine for women and men undergoing in vitro fertilization, and 50 μM MEHP in plasma has been reported in patients receiving blood transfusions (Sjöberg et al., 1985; Begum et al., 2021).

MEHP is a known endocrine-disrupting chemical and its high concentration in bodily fluids has been associated with several negative reproductive parameters, such as a smaller number of oocytes retrieved, fewer live births in women undergoing in vitro fertilization, and the occurrence of endometriosis (Cobellis et al., 2003; Hauser et al., 2016). Moreover, additional studies analyzing human cohorts have examined the effects of MEHP and DEHP on the female reproductive system, revealing a clear negative effect on overall successful reproduction (Latini et al., 2003; Meeker et al., 2009; Zhang et al., 2009; Toft et al., 2011). In animal studies using rodents, MEHP has been shown to inhibit ovarian follicle growth, to reduce ovarian steroid hormone levels, and to increase reactive oxygen species in vitro, however few studies focused on early embryo development (Lovekamp-Swan and Davis, 2003; Fan et al., 2010; Wang et al., 2012). Investigating the effects of MEHP on oxidative stress and maternal-to-embryonic transition in preimplantation mouse embryos, Chu et al. (2013) showed that 100 μM MEHP reduced the number of blastocysts formed and 1000 μM MEHP blocked embryos at 2-cell stage by disrupting the embryonic genome activation, which could not be rescued by antioxidant enzymes. Analyses such as cell counts, expression of adhesion proteins and methylation genes, which are the focus of the present paper, were not included in this previous work.

Embryo development encompasses a complex series of events dependent upon strict regulation at different levels, from gene expression to cellular differentiation. Timing of murine embryonic development can vary depending on the strain. In general, the zygote becomes a 2-cell embryo 24 hours after fertilization. This embryo needs to rapidly pass through several cell divisions while simultaneously initiating genomic activation, during which the embryonic DNA begins to be transcribed and translated into functional proteins, which usually occurs at 2-cell stage in mice (Niakan et al., 2012; Saiz and Plusa, 2013). Three days after fertilization, the cells undergo compaction forming a morula, which then, after cavitation, progresses to blastocyst stage on day 4 and begin to go through the first cell fate decision. Cell lineage specification is achieved by differential expression of transcription factors, for example, caudal-type homeobox-2 (CDX-2) is expressed solely by trophectoderm (TE) cells, while octamer-binding protein 4 (OCT-4), also known as POU5F1, is expressed by inner cell mass (ICM) at the blastocyst stage (Strumpf et al. 2005; Saiz and Plusa, 2013). The correct differentiation of cells is a sign of embryo quality and a decrease in OCT-4 expression has been correlated with a lower quality embryo (Van Thuan et al., 2006). These morphological changes occur inside the zona pellucida, a protective layer composed of mucopolysaccharides and glycoproteins, which is removed by the time of implantation in the uterus (Wolpert et al., 2015). The blastocyst cultured in vitro escapes from the zona pellucida by day 5, through the process of hatching, which is promoted by contractions and by signaling molecules such as prostaglandin and prostacyclin, but this process is still not fully understood (Niimura, 2003; Huang et al., 2004; Kirkegaard et al., 2013; Zhan et al., 2018). During development, occurrence of micronucleation is not uncommon, when small parts of a chromosome are left behind during mitotic division. This leads to a piece of genetic material being left outside the nucleus when the cell returns to interphase and its incidence is negatively correlated to embryo quality (Meriano et al., 2004; Hintzsche et al., 2017).

Gene expression can also be altered by epigenetic mechanisms, independent from DNA sequence changes. Several mechanisms for epigenetic control were described, including methylation, which is crucial for successful embryo development (Dean et al., 2003; Hallberg et al., 2022). The zygote needs to erase markers from maternal and paternal gametes and establish its own pattern of DNA methylation (Reik and Dean, 2001). To that end, after fertilization, global DNA methylation level decreases until embryo reaches the morula stage, in mice. In blastocyst stage, the level increases again, although differentially between cell lineages, being more elevated in the ICM (Dean et al., 2003). Those waves of de-methylation and re-methylation occur through the action of DNA methyltransferase enzymes (DNMTs) and ten-eleven translocation proteins (TET), responsible for adding or removing methyl groups to DNA nucleotides and making this location inaccessible, or accessible, for transcription (Tan and Shi, 2012; Uysal et al., 2015). Disruption of the methylation process may be detrimental to development in many ways. MEHP and other endocrine-disrupting chemicals, such as bisphenol A and per- and polyfluoroalkyl substances (PFASs), have been reported to alter methylation during embryogenesis. For example, bovine blastocysts derived from oocytes matured in the presence of MEHP presented an increased in Dnmt3b gene expression, while mice exposed in utero to bisphenol A shown a decreased in methylation of the homeobox gene Hoxa10 in the uterus, which results in increased sensitivity to estrogen signaling (Bromer et al., 2010; Kalo and Roth, 2017). Bovine cumulus-oocyte complexes (COCs) exposed to two different types of PFASs for only 22 hours during in vitro maturation resulted in blastocyst with altered methylation landscape (Hallberg et al., 2021; Hallberg et al., 2022). These observations suggest that DNA methylation should also be considered as a target for MEHP action during mice embryo development.

The majority of studies reporting that phthalates have negative effects on embryo development used animal models with indirect exposure and measures, such as the number of pups born, number of pups born alive versus dead, and the number of implantation sites in dams that had been exposed to a particular phthalate (Schilling et al., 2001; Grande et al., 2006; Lyche et al., 2009). Thus, the aim of the present study was to investigate the effect of direct exposure of a wide range of concentrations of MEHP on embryo development, examining parameters such as blastocyst formation, hatching ability, cell differentiation, and methylation markers, contributing to the safety evaluation of this chemical.

2. Material and methods

2.1. Chemicals

Mono(2-ethylhexyl) phthalate (MEHP) was purchased from AccuStandard (#ALR-138N, CAS# 4376–20-9, 100% purity). DMSO was from MilliporeSigma (#D2650, ≥99.7% purity). Pregnant mare serum gonadotropin (PMSG) was purchased from Prospec (#HOR-272), human chorionic gonadotropin (HCG) from MilliporeSigma (#230734), and hyaluronidase from bovine testes from MilliporeSigma (#H4272). We initially used medium specific for embryo culture KSOMaa Evolve from Zenith Biotech (#ZEKS-050), in the first experimental trials. This product was discontinued, and we then switched to Global culture medium from Life Global (#LGGG-100) for subsequent experiments. The chemical composition of both media is believed to be the same and no difference was observed in embryo development with the different reagents (Supplemental Figure 1).

2.2. Embryo culture and development

Female CD-1 mice aged 35 days (n = 4–5) were obtained from Charles River Laboratories and housed at the University of Illinois at Urbana-Champaign at the Carl R. Woese Institute for Genomic Biology Animal Facility. Animals were provided food and water ad libitum and housed in a controlled animal room environment, maintained at a temperature of 22 ± 1°C and light-dark cycles of 12 hours. They were allowed to acclimate for at least one week before undergoing the superovulation procedure.

To obtain the embryos, six to eight-week-old females were submitted to a superovulation protocol receiving 6 IU of PMSG at 4:00 PM and 6 IU of HCG 45 hours later, at 1:00 PM. Immediately after HCG injection, females were placed with B6D2F1 males of proven fertility until the next morning. At 10:00 AM, females were euthanized by cervical dislocation and their oviducts were collected in holding medium (MOPS buffer supplemented with 5% FBS). Using a stereomicroscope warmed at 37°C, the ampullas of the oviducts were perforated with a needle, and presumptive embryos (at zygote stage) surrounded by cumulus cells were transferred to a dish containing fresh holding medium. All subsequent steps were done at 37°C, with the aid of a plate warmer. Hyaluronidase (500 μg/mL) was added to the dish to dissociate the embryos from the cumulus cells for approximately one minute. Presumptive zygotes were washed twice in holding medium and once in the culture medium (KSOMaa or Global) in a dish covered with a layer of mineral oil. Presumptive zygotes from different females were mixed in the last dish and then randomly distributed to the final wash containing the respective treatments. Finally, presumptive zygotes were transferred to culture droplets of the corresponding treatments with a glass mouth pipette.

Embryos were cultured in 20 μL drops of either control medium (KSOMaa or Global), DMSO 0.05% (vehicle), or the following concentrations of MEHP, 0.1, 1, 10, 100, and 1000 μM. Each dish (60 mm diameter; Falcon), containing several 20 μL drops from the same treatment group covered with 9 mL of mineral oil (Up&Up), was prepared 24 hours before the collection of the embryos and allowed to equilibrate and adjust for pH inside the incubator at 37°C and 5% CO₂. The pH was checked after one and six days inside the incubator and ranged between 7.23 and 7.39. Five to fifteen embryos were placed inside each drop and observed after four and six days in culture for identification of blastocysts and to monitor hatching from the zona pellucida, which should be completed by the last day of observations.

After six days in culture, pictures were taken under a contrast phase microscope (Olympus IX70), diameter was measured using the cellSens Standard software (Olympus), and the embryos were fixed in 4% paraformaldehyde (Electron Microscopy Sciences, #19200). For each embryo, two measurements of diameter were taken to obtain an average. Data presented in this study were obtained and pooled from 8 independent experiments, with the number of embryos per treatment group ranging from 1 to 33 for each experiment.

For analyses conducted with morulae and blastocysts (non-hatched), embryos were cultured for only 3 or 4 days, respectively. All the analyses were conducted by researchers not blind to the treatment of subjects.

2.3. Immunofluorescence staining

To investigate the effects of MEHP on early cell lineage commitment, the numbers of inner cell mass cells and trophectoderm cells in hatched blastocysts were counted after immunofluorescence staining with antibodies to detect OCT-4 (specific for inner cell mass cells) and CDX-2 (specific for trophectoderm) transcription factors and visualized by confocal microscopy.

After six days in culture, hatched blastocysts were fixed in 4% paraformaldehyde for 20 minutes and then stored in phosphate-buffered saline (PBS) with 0.5% BSA covered with mineral oil at 4°C until staining. Hatched blastocysts were washed with wash buffer (PBS, 0.1% Triton X-100, 0.1% PVP) and permeabilized with 1% Triton X-100 in PBS for 30 minutes. After another wash, samples were blocked for 2 hours at room temperature with 10% donkey serum in blocking buffer (PBS, 0.1% Triton X-100, 1% BSA, 0.1 M Glycine) followed by incubation with primary antibodies rabbit anti-OCT-4 (Cell Signaling, #D6C8T, 1:1600 dilution) and mouse anti-CDX-2 (Biogenex, #MU392A, 1:600 dilution) overnight at 4°C. The next day, samples were washed three times with wash buffer and then incubated with fluorescent-labeled secondary antibodies CY3 donkey anti-rabbit 555 nm (Jackson ImmunoResearch, #711–165-152, dilution 1:500) and CY5 donkey anti-mouse 690 nm (Jackson ImmunoResearch, #715–175-151, dilution 1:400) for 2 hours in the dark at room temperature. Antibodies were diluted in antibody buffer (PBS, 0.1% Triton X-100, 1% BSA), except anti-CDX-2 which was diluted in diluent provided by the manufacturer. Incubation with the secondary antibody was followed by two washes with antibody buffer (20 minutes each) and one wash with wash buffer (1 hour) followed by incubation in mounting medium containing DAPI (Vector Laboratories, #H-1200) for 30 minutes. Embryos were mounted in the center of a 35 mm glass-bottom dish with a 14 mm micro-well (Cellvis, #D35–14-1.5-N) surrounded by mineral oil, coverslipped, and kept in the dark at 4°C until imaging.

For analysis of E-cadherin expression in morulae, immunofluorescence staining was conducted with a goat anti-h/m E-Cadherin primary antibody (R&D Systems, #AF748, dilution 1:100) and bovine Alexa Fluor 488 anti-goat 488 nm secondary antibody (Jackson ImmunoResearch, #805–545–180, dilution 1:300). The same protocol for immunostaining described above was used for this protein.

All immunofluorescence-stained hatched blastocysts and morulae were imaged with a confocal microscope (Carl Zeiss LSM 880 Airyscan System, Carl Zeiss, Oberkochen, Germany) at the Microscopy Core Facility at the Carl R. Woese Institute for Genomic Biology at the University of Illinois at Urbana-Champaign. The samples were excited with 405, 488, 561, and 633 nm laser wavelengths and images acquired at appropriate band pass emission filters for each channel. The acquired images were pseudocolored according to their emission wavelengths and displayed in an 8-bit relative intensity scale. To obtain all the expression pattern in a multicellular embryo, multiple 2D images were taken along the entire 3D volume of the embryos (Z stacks). Optimal pinhole diameters and the section thickness were selected automatically by the software for all the channels within a given embryo depending on the wavelengths and objective. Images were acquired as soon as possible to avoid diffusion, especially, the samples stained for E-cadherin detection which were imaged on the same day to reduce variation.

2.4. Inner cell mass and trophectoderm cell counting

Inner cell mass cells express OCT-4 transcription factor in nuclei which were pseudocolored green, while trophectoderm cells expressing CDX-2 transcription factor in nuclei were pseudocolored red. DAPI counterstaining of all nuclei in the embryo were pseudocolored blue. Only hatched blastocysts were used in this analysis to avoid comparison between different blastocyst stages. From the Z stack 2D images, 3D reconstruction was performed by doing a maximum intensity projection algorithm in the 3D visualization program Imaris (Bitplane, Zurich, Switzerland). The number of cells expressing each marker was counted with a combination of the orthogonal slicer and counting tools in the Imaris program. All image intensities were adjusted, if necessary, across all treatments for optimal display purposes.

2.5. Quantification of E-cadherin expression

Images of morulae stained for E-cadherin were acquired using a Plan-Apochromat 20x/0.8 NA objective with a 488 nm excitation, appropriate laser power and master gain settings, which were kept constant for all the samples. The intensity of fluorescence was measured in morulae from two independent experiments using the profile tool in ZEN 2.3 Lite software. A line was hand-drawn across the cell membranes encompassing the entire area of the morula. The fluorescence intensity profile of this line was calculated, and the values of the peaks were recorded and averaged. For each morula, three different focal planes along the Z-axis were used for analysis starting at the middle of the embryo and 10 slices above and below of that center focal plane.

2.6. Micronuclei assessment in morulae

Visualization and quantification of micronucleation were performed using the same Imaris software as mentioned above, after 3D reconstruction and maximum intensity projections, which allowed for the rotation of the embryo and a more clear overall visualization and quantitative assessment of the cell nucleus and micronuclei stained by DAPI.

2.7. RNA extraction and quantitative RT-PCR

The two higher concentrations of MEHP, excluding 1000 μM that did not produce enough sample, were chosen for analyses of expression of specific genes. Embryos retrieved at the zygote stage were cultured in either control medium, DMSO 0.05%, 10 μM, or 100 μM MEHP. After four days in culture, five blastocysts (non-hatched) per group were transferred to tubes containing 25 μL extraction buffer (from Arcturus PicoPure RNA isolation kit, Applied Biosystems, #12204–01), frozen in liquid nitrogen, and stored at −80°C until subsequent extraction of RNA. Blastocysts from 3 independent experiments were analyzed. RNA extractions and cDNA syntheses were performed on the same day for all the samples to minimize variation.

RNA was isolated using the Arcturus PicoPure RNA extraction kit following the manufacturer’s instructions. Briefly, tubes containing the blastocyst samples were thawed on ice, an additional 25 μL extraction buffer was added to a total volume of 50 μL, and then incubated at 42°C for 30 minutes. Following centrifugation, the supernatant containing the extracted RNA was transferred to a new tube, an equal volume of 70% ethanol was added and mixed well. The mixture was then transferred to a pre-conditioned purification column and centrifuged for 2 minutes at 100 g to bind the RNA to the column, followed by centrifugation at 16000 g for 30 seconds to remove flowthrough. Wash buffer 1 was added to the column and centrifuged, followed by DNase digestion (RNase-Free DNase Set, Qiagen, #79254) for 15 minutes at room temperature. The column was then washed with wash buffer 1 again, followed by two washes with wash buffer 2. The column was transferred to a new 0.5 mL collection tube and 11 μL of warm elution buffer were applied directly to the membrane and incubated for 5 minutes at room temperature. RNA was eluted by centrifugation at 1000 g for 90 seconds immediately followed by centrifugation at 16000 g for 90 seconds. Isolated RNA samples were stored at −80°C.

RNA was reverse transcribed to cDNA with the Transcriptor First Strand cDNA Synthesis kit (Roche, #04379012001). Following the manufacturer’s instructions, the sample RNA (5 μL), random hexamers (60 μM), protector RNA inhibitor (20 U), deoxynucleotide mix (1 mM), transcriptor reverse transcriptase (10 U), reaction buffer, and RNA-free water were mixed and incubated in a thermocycler for 10 minutes at 25°C, followed by 60 minutes at 50°C, and 5 minutes at 85°C for enzyme inactivation. Synthesis of cDNA was done twice for each RNA sample to provide sufficient cDNA to analyze all the genes of interest. A negative control containing water, and no template, was included as an internal control.

The cDNA was diluted 3.3 or 5 times for quantitative polymerase RT-PCR depending on the number of genes analyzed in each plate. A primers mix containing 0.3 μL forward primer (200 nM), 0.3 μL reverse primer (200 nM), 7.5 μL SYBR Green (PowerSYBR Green PCR Master Mix, Applied Biosystems, #4367659), and 1.9 μL water (HyPure Molecular Biology grade, GE, #SH30538.02) per well was used. Primer mix (10 μL) and diluted cDNA (5 μL) were combined into a final reaction volume of 15 μL in each well in a 384-well plate (MicroAmp Optical 384-well Reaction Plate, Applied Biosystems, #4309843). The amplification reactions were run in triplicates for 40 cycles of 95°C for 15 seconds and 60°C for 1 minute. The quantitative RT-PCR amplification and quantification were carried out using the QuantStudio^™ 7 Flex Real-Time PCR System 384-well at the Roy J. Carver Biotechnology Center at the University of Illinois at Urbana-Champaign. Relative fold change of expression was calculated using the Pffafl method and H2A histone family member Z (H2afz) was used as the housekeeping gene. All primer pairs had their efficiency analyzed and these ranged from 96.91 to 115.60%. Table 1 lists the primers’ specifications.

Table 1:

List of primers, conditions, and efficiencies.

Gene	Gene Name	Gene ID	Left primer	Right primer	Amplicon Product (nt)	Efficiency
H2afz	H2A histone family, member Z (H2afz), transcript variant 1	NM_016 750.3	TTCCAGTGGACT GTATCTCTGTG	CGAATGCAGAA ATTTGGTTG	109	108.16%
Dnmt1	DNA methyltransferase (cytosine-5) 1 (Dnmt1), transcript variant 1	NM_001 199431.1	CAGAGACTCCC GAGGACAGA	TTTACGTGTCG TTTTTCGTCTC	65	109.64%
Dnmt3 a	DNA methyltransferase 3A (Dnmt3a), transcript variant 3	NM_001 271753.1	AAACGGAAACG GGATGAGT	ACTGCAATTAC CTTGGCTTTCT	71	99.50%
Dnmt3 b	DNA methyltransferase 3B (Dnmt3b), transcript variant 9	NM_001 271747.1	ATGATCGATGC CATCAAGGT	GGGAAGCCGAA GATCCTG	89	112.61%
Tet1	tet methylcytosine dioxygenase 1 (Tet1), transcript variant 1	NM_001 253857.2	TGCCATTATGGG ATAGAATCG	GAGCGGACAGA TGAATGGAC	60	115.60%
Tet3	tet methylcytosine dioxygenase 3 (Tet3), transcript variant 2	NM_183 138.2	AAGACGCCACG AAAGTTCC	TGAAAGCTATT CCGGAGCAC	71	96.91%
Ptgs1	prostaglandin-endoperoxide synthase 1	NM_008 969.4	AGGAGTCTCTC GCTCTGGTT	CAGGGATTGAC TGGTGAGGG	104	107.97%
Ptgs2	prostaglandin-endoperoxide synthase 2	NM_011 198.4	CTGACCCCCAA GGCTCAAAT	ATTTAAGTCCA CTCCATGGCCC	128	98.39%
Ptgis	prostaglandin I2 (prostacyclin) synthase	NM_008 968.4	ATGCCATCAAC AGCATCAAA	GCCATATCTGC TGAGGTCAAA	110	97.71%

2.8. Statistical analyses

Model selection approach as used to test the effect of MEHP concentration and three other covariates on the probability of reaching the blastocyst or hatched blastocyst stages. All models were logistic regressions with each embryo that successfully became a blastocyst (or, separately, a hatched blastocyst) recorded as a survival event (Y = 1), and each embryo initially present in the plate that failed to reach the stage in question recorded as a failure event (Y = 0). In addition to the concentration of MEHP as the fixed-effect predictor of interest, were considered as fixed-effect covariates: the number of embryos in the plate and the culture medium used in the experiment (because the original medium KSOMaa used in the first 4 replicate experiments became unavailable and was switched to Global medium for the remaining replicates). The replicate experiment number was considered a random-effect covariate. Using AIC, all 16 possible models were compared, with each of these four covariates present or absent (including the model with no predictors, based solely on the overall survival probability across all treatments and replicates). For these analyses, DMSO 0.05% (vehicle) were used to represent a concentration of MEHP equal to zero. Confidence intervals for parameter estimates were obtained by bootstrapping (1000 resamplings of 75% of the data). For ease of visualization, predicted probability curves were plotted as a function of MEHP concentration on logarithmic scale (therefore omitting from the plot the zero MEHP concentration, which was nonetheless considered when fitting the models). Similar model ranks were obtained when using the log10 instead of absolute MEHP concentration as a predictor (which requires using only non-zero concentrations and therefore excluding the controls). Code implementing the analyses in R v.3.6.1 is provided as supplementary information. Mixed-effect model estimation was done with R package lme4.

The overall percentages of blastocysts formed, blastocysts hatched, ratio of hatched to total blastocysts, and incidence of micronuclei were analyzed by chi-square and multiple comparisons between groups were done using the Fisher’s exact test (two-sided) considering Bonferroni adjustment. Continuous variables, presented as mean ± standard error of the mean (SEM), were tested for normality of residuals using the Shapiro-Wilk test and compared between groups using one-way ANOVA and Tukey’s post-test when data were normal; or Kruskal-Wallis test and Dunn’s post-test when data were not normally distributed. Data for Tet3 expression were log-transformed to achieve normality of residuals before analysis by ANOVA and Tukey’s post-test. Statistical analyses were performed using GraphPad Prism 9.0.1 software and a p-value < 0.05 was considered significant.

3. Results

3.1. MEHP reduces blastocyst formation and hatching

After six days of culture in vitro with different concentrations of MEHP, the overall percentage of embryos that reach the blastocyst stage and the overall percentage of blastocysts that hatched from the zona pellucida were calculated. Model selection strongly supports a negative effect of MEHP on the probability of reaching both the blastocyst and hatched blastocyst stages (Figure 1). For both stages, all models that included the MEHP concentration as a predictor outperformed all models that did not (whether or not we log-transformed the MEHP concentration before including it as a predictor) (Supplemental Tables 1 and 2). According to the best model for each stage, increasing the MEHP concentration by 1 μM decreases the log-odds ratio of becoming a blastocyst by 0.0029 (95% bootstrap CI 0.0025–0.0039) and the log-odds ratio of becoming a hatched blastocyst by 0.024 (95% CI 0.019–0.035). Model selection also suggests that the probability of reaching the blastocyst and hatched blastocyst stages varied across experimental replicates independently of the MEHP concentration (all models that included a random effect of the experimental replicate outperformed all models that did not; Supplemental Tables 1 and 2). Model selection also suggests that increasing the total number of embryos initially present in the plate decreases the probability of reaching blastocyst stage (the two best models include this effect, whereas the best model without it has a ΔAIC of 4. The best model for hatched blastocysts does not include an effect of the number of embryos in the plate, but the second-best model includes it and performs nearly identically). The best models did not include an effect of the culture medium, but model selection does not strongly reject that possibility (Supplemental Tables 1 and 2). Taken together, these results strongly suggest MEHP decreases the probability of reaching the blastocyst and hatched blastocyst stages, even after controlling for potential confounders.

Figure 1:

Estimated probability of reaching the blastocyst and hatched blastocyst stages as a function of the concentration of MEHP in μM. X-axis was log-transformed to improve visualization, and therefore estimates for an MEHP concentration of zero are not shown (yet that concentration was used for model fitting). The estimated curves are based on the predictors included in the best model for the blastocyst stage (MEHP concentration and the number of embryos in the plate as fixed-effect predictors, and the replicate experiment as a random-effect predictor). Although these predictors correspond to the second-best model for hatched blastocysts (the best model does not include the effect of the number of embryos), they perform nearly as well as the best model (ΔAIC = 1.4), and so, for consistency, plots based on the same predictors for both stages are shown.

MEHP at 100 and 1000 μM significantly reduced the percentage of blastocyst formation, with 1000 μM appearing to be very toxic to the embryos. In the control group, 43.8% of the embryos developed to the blastocyst stage, while only 20.9% of embryos treated with 100 μM MEHP, and only 5.1% of the ones treated with 1000 μM reached the blastocyst stage (Table 2). MEHP at 100 and 1000 μM also had a significant negative impact on the percentage of blastocysts that hatched successfully. Of the total embryos exposed to 1000 μM MEHP, none were able to hatch from the zona pellucida, while within the embryos exposed to 100 μM MEHP, only 4.7% completed hatching, compared to 30.0% in the control group (Table 2). Representative pictures of the embryos after 6 days in culture are shown in Figure 2. Becoming a blastocyst is not sufficient for the embryo to successfully hatch from the zona pellucida. Therefore, we also determined the effect of MEHP exposure on hatching ability among the embryos that did reach the blastocyst stage, which is shown as the ratio of hatched blastocysts/total blastocysts in Table 2. Data showed that 68.6% of the blastocysts formed in the control group were able to hatch from the zona pellucida by day 6, while for the blastocysts in the 100 μM MEHP group, that percentage was only 22.2%. These results indicate MEHP at 100 μM, or higher concentrations, compromise hatching ability independently of blastocyst formation. The lower concentrations of MEHP (0.1, 1, and 10 μM) did not significantly affect embryo development to the blastocyst stage or the hatching process.

Table 2:

Embryonic development parameters of embryos cultured in the presence of different concentrations of MEHP.

	Total number of embryos	Blastocysts (%)	Hatched (%)	Hatched blastocysts/total blastocysts	Diameter of hatched blastocyst (μm)
Control	80	35 (43.8%)	24 (30.0%)	68.6%	128.0 ± 5.60
DMSO 0.05%	81	47 (58.0%)	30 (37.0%)	63.8%	113.3 ± 4.37
MEHP 0.1 μM	82	44 (53.7%)	27 (32.9%)	61.4%	130.4 ± 4.88
MEHP 1μM	89	31 (34.8%)	21 (23.6%)	67.7%	127.3 ± 6.34
MEHP 10 μM	77	27 (35.1%)	18 (23.4%)	66.7%	152.5 ± 17.46
MEHP 100 μM	86	18 (20.9%) *	4 (4.7%) *	22.2% *	103.3 ± 15.29
MEHP 1000 μM	98	5 (5.1%) *	0 (0.0%) *	0.0% *	NA

^*Percentages were analyzed by chi-square and multiple comparisons by Fisher’s exact test considering Bonferroni adjustment. Diameter (mean ± SEM) was analyzed by Kruskal-Wallis and multiple comparisons by Dunn’s test. Number of samples for diameter analysis were: Control n=20; DMSO 0.05% n=30; MEHP 0.1 μM n=25; MEHP 1 μM n=19; MEHP 10 μM n=16; MEHP 100 μM n=4. Analyses were performed in GraphPad Prism 9, p-value < 0.05 was considered as significant and marked with an. NA: not applicable.

Figure 2:

Representative images of embryos from different treatment groups after 6 days in culture. Pictures were taken under a phase-contrast microscope. Examples of hatched blastocysts (black arrowheads), hatching blastocyst (black cross), non-hatched blastocyst surrounded by the zona pellucida (blue arrowhead), degraded embryos (black asterisks), and empty zona pellucidas (red arrows) are shown. The insert on MEHP 1000 μM group shows 1 of the 5 blastocysts that developed in this group. Scale bars: 100 μm.

The diameter of the hatched blastocysts was assessed to investigate whether MEHP influences the size of the embryos. MEHP at 1000 μM concentration was not evaluated since no blastocyst hatched in this group. The average diameter of hatched blastocysts after six days in culture was not affected by MEHP (0.1 – 100 μM). The diameter ranged from 103.3 ± 15.29 μm in the MEHP 100 μM treatment group to 152.5 ± 17.46 in the MEHP 10 μM treatment group and was not statistically different between the groups (Table 2).

An important factor to consider is whether the embryos are experiencing developmental delays. Therefore, we reorganized the data by observational day (days 4 and 6) and determined the percentages of embryos from each treatment group that were non-hatched blastocysts (pink bars), had initiated hatching (called hatching blastocyst; blue bars), or had completed hatching (gray bars) (shown in Figure 3). On day 4, the majority of blastocysts were still enclosed in the zona pellucida and only a few had started the hatching process in all the treatment groups. The negative effect of 1000 μM MEHP on blastocyst formation was already evident and was statistically significant compared to the control group. By day 6, the majority of the blastocysts had already completely hatched in all treatment groups, except for the 100 and 1000 μM MEHP treatment groups, where the percentage of hatched blastocysts was statistically reduced compared to the control. We also observed that some embryos developed to the blastocyst stage after day 4, which is considered late, but this was not related to any treatment group. Embryos exposed to 100 and 1000 μM MEHP were not delayed but were unable to develop into blastocysts.

Figure 3:

Effect of MEHP on blastocyst hatching at days 4 and 6 of culture. The percentages of blastocysts enclosed by the zona pellucida (pink), blastocysts initiating hatching (blue), and blastocysts that completed hatching (gray) are shown for each treatment group at day 4 and day 6. On day 4, the percentage of non-hatched blastocysts for 1000 μM MEHP was statistically different from control. On day 6, the percentage of hatched blastocysts for 100 and 1000 μM MEHP treatments were statistically different from control. Control n=80; DMSO 0.05% n=81; MEHP 0.1 μM n=82; MEHP 1 μM n=89; MEHP 10 μM n=77; MEHP 100 μM n=86; MEHP 1000 μM n=98. Percentages were analyzed by chi-square and multiple comparisons by Fisher’s exact test considering Bonferroni adjustment. Analyses were performed in GraphPad Prism 9, p-value < 0.05 was considered as significant and marked with an *.

3.2. MEHP at concentrations up to 10 μM does not alter the number of inner cell mass and trophectoderm cells

Hatched blastocysts were fixed after six days in culture and stained by immunofluorescence for detection of inner cell mass (ICM) and trophectoderm (TE) cells, as shown in Figure 4. The ICM can be identified by expression of the OCT-4 transcription factor, while cells from the TE express the CDX-2 transcription factor. The number of cells from ICM, TE, the total number of cells, and the ratio between those cell types are presented in Table 3. Treatment groups 100 and 1000 μM MEHP were not included in this analysis due to limited number of hatched blastocysts formed in these groups.

Figure 4:

Immunofluorescence imaging of hatched blastocyst for detection of inner cell mass (ICM) and trophectoderm (TE) cells. A representative blastocyst from each treatment is presented using an orthogonal projection standard deviation algorithm (Zeiss Zen version 3.5, Carl Zeiss Obercohen, Germany), with OCT-4 (green) marking ICM, CDX-2 (red) marking TE, nuclei (blue), and their merged counterparts. Scale bars: 20 μm.

Table 3:

Number of cells from the inner cell mass and trophectoderm, total number of cells, and ratio between the two cell types in hatched blastocysts. N = the number of embryos analyzed for each endpoint.

	N	Inner cell mass (ICM)	Trophectoderm (TE)	Total	Ratio TE:ICM
Control	11	23.45 ± 1.88	88.09 ± 4.17	111.5 ± 5.74	3.90 ± 0.23
DMSO 0.05%	15	25.40 ± 3.97	83.33 ± 7.58	108.7 ± 10.64	4.61 ± 0.79
MEHP 0.1 μM	13	27.23 ± 1.97	100.80 ± 6.35	128.0 ± 7.94	3.80 ± 0.21
MEHP 1 μM	12	26.l5 ± 4.21	88.92 ± 6.62	115.7 ± 9.46	3.92 ± 0.41
MEHP 10 μM	10	22.10 ± 2.11	84.50 ± 4.60	106.6 ± 5.07	4.16 ± 0.47

Values are mean ± SEM. ICM and ratio of TE:ICM were analyzed by Kruskal-Wallis and multiple comparisons by Dunn’s test. TE and total number of cells were analyzed by ANOVA and Tukey’s post-test. Analyses were performed in GraphPad Prism 9, p-value < 0.05 was considered as significant and marked with an *.

The number of ICM cells and TE cells per blastocyst was similar for all analyzed treatment groups. The total number of cells per hatched blastocyst was similar between treatment groups, ranging from 106.6 ± 5.07 to 128.0 ± 7.94 in the 10 μM and 0.1 μM MEHP treatments, respectively. The ratio of ICM to TE cells was also not statistically different between the treatment groups, ranging from 3.80 ± 0.21 to 4.61 ± 0.47 for the treatment groups. In summary, MEHP at 0.1, 1, and 10 μM concentrations did not affect the initial cell differentiation events in blastocyst development.

3.3. MEHP at 100 μM increases the expression of the Tet3 gene in blastocysts

Since MEHP at 100 μM reduced the blastocyst formation, we focused on that concentration for further analyzes. Expression of several genes that control the methylation process was assessed by quantitative RT-PCR of blastocysts cultured in vitro for 4 days. In addition to 100 μM MEHP, control, DMSO 0.05% and 10 μM MEHP treatment groups were included in this analysis.

DNA methyltransferases are enzymes known to add methyl groups to the DNA, coded by DNA methyltransferase genes (Dnmt), while tet methylcytosine dioxygenase enzymes act to remove methyl groups from DNA strands and are encoded by tet methylcytosine dioxygenase genes (Tet). We measured the expression of several different subtypes of these genes, Dnmt1, Dnmt3a, Dnmt3b, Tet1, and Tet3 (Figure 5). MEHP at 100 μM significantly increased the expression of Tet3 (2.07-fold change), which suggests that MEHP can alter the methylation status of the blastocyst, probably leading to decreased methylation of DNA.

Figure 5:

Effect of MEHP exposure on gene expression in blastocysts. MEHP at 100 μM increased expression of Tet3 (p=0.0213), but not of any of the other genes for enzymes related to the methylation process, or to prostacyclin synthesis. Data are presented as the mean ± SEM from three independent experiments. For each sample, five blastocysts from the same treatment group, cultured for 4 days, were pooled and frozen for RNA extraction. H2afz was used as the housekeeping gene. Gene expression for each gene was compared between the treatments separately. Dnmt1, Dnmt3a, Tet1, Tet3, and Ptgis were analyzed by ANOVA and Tukey’s post-test. Dnmt3b was analyzed by Kruskal-Wallis and Dunn’s post-test. Analyses were done in GraphPad Prism 9, p-value < 0.05 was considered as significant and marked with an *.

MEHP interfered with the hatching process, leading to a reduced percentage of blastocysts that completely hatched from the zona pellucida, and prostacyclin has been reported to play a role in blastocyst hatching. Hence, expression of prostaglandin I2 (prostacyclin) synthase (Ptgis), prostaglandin-endoperoxide synthase 1 (Ptgs1) (coding for COX-1), and prostaglandin-endoperoxide synthase 2 (Ptgs2) (coding for COX-2) were also investigated. Ptgis expression was similar between all the treatment groups analyzed, while Ptgs1 and Ptgs2 expression were below the detection threshold. The negative effect of MEHP on blastocyst hatching through prostacyclin mediation could not be determined.

3.4. MEHP at 10 μM and 100 μM does not alter the expression of E-cadherin in morulae

Next, we decided to analyze embryos at the morula stage, the stage immediately before the embryos become blastocysts, to investigate possible reasons for the impaired development progression. Experimental groups control, DMSO 0.05%, 10 μM and 100 μM MEHP were included in the subsequent analyzes. MEHP at 1000 μM treatment group did not produce enough samples to analyze.

Structural cohesion is important to hold the embryo’s shape, as fluid flows in to form the blastocoel. Adherens junctions, mediated by E-cadherin, are important to provide a tight connection between the blastomeres and keep them adhered together against the pressure of the accumulating blastocoel fluid. Alterations in E-cadherin expression could be one of the reasons for failure to progress further in the cavitation process. However, E-cadherin expression, assessed by fluorescence intensity, was not different between the treatment groups analyzed. Figure 6 shows 3D reconstructions (A) and the middle plane (B) of morulae stained for E-cadherin detection from different treatment groups, as well as quantification of E-cadherin expression (C). Representative pictures were chosen including samples with the lower and the higher fluorescence intensity.

Figure 6:

Immunofluorescence staining and quantification of E-cadherin expression in morulae. A: 3D projections of E-cadherin staining of morulae from different experimental groups are shown. B: Representative images of a single median focal plane of the same morulae shown in A. Scale bars: 10 μm. C: Expression of E-cadherin, measured by fluorescence intensity, was not affected by MEHP exposure. Data are presented as mean ± SEM and are from two independent experiments. Control n=4; DMSO 0.05% n=7; MEHP 10 μM n=4; MEHP 100 μM n=8. Data were analyzed by ANOVA and Tukey’s post-test, performed in GraphPad Prism 9, p-value < 0.05 was considered as significant.

3.5. MEHP at 10 μM and 100 μM does not alter the number of cells in morulae

The number of blastomeres in the morulae was counted based on DAPI nuclear staining. The number of cells in the compacted morulae ranged from 18 to 37 in the control group, from 13 to 33 in the DMSO 0.05% group, from 12 to 32 in the 10 μM MEHP, and from 12 to 33 in the 100 μM MEHP group. However, the average number of blastomeres was not different between the treatment groups (Figure 7).

Figure 7:

The effect of MEHP exposure on cell number in compacted morulae. Cell number was determined by counting nuclei stained with DAPI and was not affected by MEHP exposure. Data are presented as mean ± SEM from two independent experiments. Control n=7; DMSO 0.05% n=13; MEHP 10 μM n=6; MEHP 100 μM n=13. Data were analyzed by Kruskal-Wallis and Dunn’s post-test, performed in GraphPad Prism 9, p-value < 0.05 was considered as significant.

3.6. MEHP at 10 μM and 100 μM does not impact the incidence of micronucleation or the number of micronuclei in morulae

Micronuclei (MN) can be easily identified with DAPI staining which stains the DNA. Examples of micronuclei detected in morulae from each treatment group are shown in Figure 8A. The presence of micronucleation is an indication of nuclear fragmentation, a negative feature for embryo development, resulting in increased damage that decreases embryo quality.

Figure 8:

Analyses of micronucleation in morulae treated with MEHP. A: Representative images of morulae with micronuclei (pointed by red arrows). Nuclei stained with DAPI are shown in white while the E-cadherin distribution is displayed in green. Scale bars: 10 μm. B: Incidence of micronucleation in morulae was not altered by exposure to MEHP. Percentages were analyzed by chi-square and multiple comparisons by Fisher’s exact test considering Bonferroni adjustment. Control n=7; DMSO 0.05% n=13; MEHP 10 μM n=6; MEHP 100 μM n=13. C: The average number of micronuclei in affected morulae was also not different between the experimental groups. Averages were compared by Kruskal-Wallis and Dunn’s post-test. Control n=5; DMSO 0.05% n=8; MEHP 10 μM n=4; MEHP 100 μM n=11. Analyses were performed in GraphPad Prism 9, p-value < 0.05 was considered as significant.

The incidence of micronucleation was calculated by determining the percentage of embryos that had at least one micronucleus detected (Figure 8B). All the treatment groups had more than 60% incidence of micronucleation, and that percentage was not statistically different between the treatment groups analyzed. The average number of micronuclei present in affected morulae was also calculated (Figure 8C). The number of micronuclei present per embryo was not influenced by MEHP exposure. Most of the cells with micronuclei had only one, but sometimes there were two or more micronuclei in the same cell. In addition, the percentage of cells exhibiting micronucleation within the morulae ranged from 6.41 ± 1.64% to 12.54 ± 3.80% in 10 μM MEHP and DMSO 0.05% groups, respectively, and were not different between the treatment groups analyzed (data not shown).

4. Discussion

The goal of our study was to investigate the impacts of direct exposure to MEHP on early embryo development, using mice as the animal model. Our results demonstrated that MEHP exposure at higher concentrations harms embryo development, reducing the percentage of blastocysts formed as well as their hatching capacity, and also potentially altering the methylation process by increasing the expression of Tet3.

MEHP has been reported previously to impair the maturation of murine, equine, and bovine oocytes (Anas et al., 2003; Dalman et al., 2008; Ambruosi et al., 2011). MEHP added to the medium for in vitro maturation of mouse oocytes reduced the number of oocytes able to pass the germinal vesicle stage and impaired the development of embryos produced from the MEHP-exposed oocytes. Interestingly, none of the embryos derived from oocytes cultured in 50, 100, 200, and 400 μM MEHP reached the morula stage (Dalman et al., 2008). Exposure of cumulus-oocyte complexes (COCs) to 50 μM MEHP during oocyte maturation had a carryover effect reducing the number of blastocysts at days 7–8 in the bovine (Grossman et al., 2012). Moreover, another study, comparing the effects of MEHP on bovine COCs and denuded oocytes maturing in vitro, showed that denuded oocytes were more susceptible to MEHP, as none of the denuded oocytes cultured with 50, 75, and 100 μM MEHP reach metaphase II during resumption of meiosis in the maturation process (Anas et al., 2003). These studies point to a detrimental effect of MEHP early on during oocyte development, but this negative impact can have long-lasting consequences and reduce the number of embryos produced.

Using an experimental design similar to the present study, where preimplantation embryos were exposed to MEHP, Chu et al. (2013) showed that MEHP can also be harmful to preimplantation embryo development. In agreement with our findings, they found that MEHP at 100 μM significantly reduced the number of blastocysts formed, as well as the number of blastocysts that hatched (although the latter was not statistically significant in their study). In addition, they demonstrated that 1000 μM MEHP blocks embryos at the 2-cell stage by impairing proper initiation of embryonic genome activation (EGA). They further reported that none of the embryos cultured in 1000 μM MEHP passed the 2-cell stage, which was not the case in our present study, where 5 embryos (5.10%) reached the blastocyst stage, albeit looking disorganized and unhealthy (insert on Figure 1). The mechanism of embryotoxicity for the 100 μM concentration cannot be explained by EGA inactivation, since the embryos were not blocked at the 2-cell stage, indicating MEHP can act through different mechanisms. Taken our results together with the previously reported in literature, they suggest that MEHP has a negative impact at different phases of embryo production, affecting oocyte maturation, initiation of EGA, formation of blastocysts, and hatching competence.

The process of hatching and its molecular players are not completely understood. It is known that COX2 and prostacyclin play a role in the hatching process. The use of a COX-2 inhibitor in culture decreased the number of hatched blastocysts, while the concomitant addition of iloprost, a prostacyclin analog, reestablished the normal hatching percentage (Huang et al., 2004). In turn, MEHP has been implicated in prostaglandin signaling by several groups (Wang et al., 2010; Tetz et al., 2013; Tetz et al., 2015). Therefore, we evaluated the expression of Ptgis, Ptgs1, and Ptgs2 genes in blastocysts exposed to MEHP. The role of prostacyclin pathway in MEHP-induced hatching failure was not conclusive, since the expression of Ptgis was not altered and Ptgs1 and Ptgs2 were not detected in those samples.

An important event during embryo development is the differentiation of cells committing to the inner cell mass (ICM) or trophectoderm (TE) lineage, the first step in cell fate specification. The development of these two cell types in the blastocyst is imperative for embryo viability. MEHP at 50 μM decreased gene expression of Pou5f1 (also known as Oct-4) in matured oocytes and 2-cell embryos, as well as reduced the total number of cells in bovine blastocysts (Grossman et al., 2012). In contrast, 100 and 1000 μM MEHP increased OCT-4 protein levels at the 2- to 4-cell stage in mouse embryos (Chu et al., 2013). In the present study, OCT-4 and CDX-2 were used as markers to count the number of ICM and TE cells in the hatched blastocysts, but their respective expression was not quantified. No difference was detected in the numbers of these two cell types, in the ratio between them, or in the total number of cells in the blastocysts between the treatment groups up to 10 μM MEHP.

DNA methylation is an important mechanism for epigenetic regulation of gene expression. The methylation pattern is established during early embryo development and therefore, this period is a vulnerable window of exposure to chemicals that could interfere with the methylation process. Some endocrine-disrupting chemicals have been shown to impact DNA methylation. Maternal exposure to DEHP during pregnancy increased the levels of Dnmt1, Dnmt3a, and Dnmt3b in fetal and neonatal mouse testes (Wu et al., 2010). Furthermore, bovine oocytes treated with MEHP had increased Dnmt3b expression later when in the blastocyst stage, possibly explaining their poor development (Kalo and Roth, 2017). Alterations in the germ line cells are especially important because their effects can be passed down to the next generations (Hallberg et al., 2021; Hallberg et al., 2022). In the present study, mouse blastocysts exposed to 100 μM MEHP showed increased expression of Tet3, a gene that functions in the removal of methyl groups from DNA. This is an important finding because changes in methylation patterns can lead to the incorrect spatiotemporal expression of diverse genes, which can have an immediate effect on embryo development or have consequences later in life and, possibly, in the future progeny (if primordial germ cells are affected). Expression of Dnmt1, Dnmt3a, Dnmt3b, and Tet1 was not altered by MEHP exposure.

Since the number of embryos reaching the blastocyst stage was reduced by the higher concentrations of MEHP, we examined embryos at the preceding morula stage for structural impairment. During morula compaction, the outer cells form adherens junctions mediated by E-cadherin, and embryos lacking E-cadherin cannot form the blastocoel (Stephenson et al., 2010). Interestingly, Sobarzo et al. (2015) reported that MEHP disrupts the expression of epithelial adhesion proteins such as adherens junctions, tight junctions, and gap junctions in Sertoli cells. Expression of this essential protein, E-cadherin, was quantified in morulae treated with 10 and 100 μM MEHP for 3 days and appeared similar in all the treatment groups. The average number of cells in morulae was also not affected by MEHP exposure at these concentrations.

Micronucleation is a fairly common event resulting from chromosome fragments or whole chromosomes not being incorporated into the cell nucleus during mitotic division (Meriano et al., 2004; Hintzsche et al., 2017). This can lead to loss or gain of genetic material and even aneuploidy in the daughter cells, and its occurrence is widely used for genotoxicity screening (Kort et al., 2016; Hintzsche et al., 2017). In human embryos, the presence of micronucleation and severity of this feature are correlated with aneuploidy, lower implantation rate, and lower live birth in in vitro fertilized embryos, and can be a marker for detecting defective embryos or ones with a poor prognosis (Royen et al., 2003; Desch et al., 2017). In the present study, exposure to MEHP at 10 and 100 μM did not increase the incidence or the number of micronuclei in morulae.

In conclusion, our findings showed that MEHP exposure during the early stages of embryo development causes a reduction in blastocyst formation and embryo competence, as well as alterations in one of the enzymes that regulate methylation. Several morphological endpoints important for embryo development were assessed, but the specific mechanisms damaging embryos could not be confirmed. The potential for successful implantation, gestation, and live birth of embryos exposed to MEHP during the early days of development needs to be investigated. In addition, in-depth analysis of changes in the methylation status of key genes involved in development will lead to better insights into the mechanisms of MEHP exposure.

Highlights.

• MEHP at 100 and 1000 μM reduced blastocyst formation and hatching ability.

• Tet3 expression was increased in blastocysts cultured for 4 days with 100 μM MEHP.

• Number of ICM and TE cells was not altered by MEHP at 0.1, 1 or 10 μM.

• MN incidence or E-cadherin expression was similar in morulae at 10 and 100 μM MEHP.

1. List of abbreviations:

CDX-2

caudal-type homeobox-2

DAPI

4′,6-diamidino-2-phenylindole

DEHP

di(2-ethylhexyl) phthalate

Dnmt

DNA methyltransferase

EDC

endocrine-disrupting chemical

EGA

embryonic genome activation

H2afz

H2A histone family member Z

ICM

inner cell mass

MEHP

mono(2-ethylhexyl) phthalate

MN

micronuclei

OCT-4

octamer-binding protein 4

PTGIS

prostaglandin I2 (prostacyclin) synthase

Ptgs1

prostaglandin-endoperoxide synthase 1

Ptgs2

prostaglandin-endoperoxide synthase 2

PVP

poly-vinyl-pyrrolidone

TE

trophectoderm

TET

ten-eleven translocation/methylcytosine dioxygenase