Placental Abnormalities are Associated With Specific Windows of Embryo Culture in a Mouse Model

Abstract

Assisted Reproductive Technologies (ART) employ gamete/embryo handling and culture in vitro to produce offspring. ART pregnancies have an increased risk of low birth weight, abnormal placentation, pregnancy complications, and imprinting disorders. Embryo culture induces low birth weight, abnormal placental morphology, and lower levels of DNA methylation in placentas in a mouse model of ART. Whether preimplantation embryos at specific stages of development are more susceptible to these perturbations remains unresolved. Accordingly, we performed embryo culture for several discrete periods of preimplantation development and following embryo transfer, assessed fetal and placental outcomes at term. We observed a reduction in fetal:placental ratio associated with two distinct windows of preimplantation embryo development, one prior to the morula stage and the other from the morula to blastocyst stage, whereas placental morphological abnormalities and reduced imprinting control region methylation were only associated with culture prior to the morula stage. Extended culture to the blastocyst stage also induces additional placental DNA methylation changes compared to embryos transferred at the morula stage, and female concepti exhibited a higher loss of DNA methylation than males. By identifying specific developmental windows of susceptibility, this study provides a framework to optimize further culture conditions to minimize risks associated with ART pregnancies.

Keywords: assisted reproductive technologies (ART); embryo culture; imprinted gene; perinatal outcome; placenta; preimplantation embryo.

FIGURE 1

Fetal and placental outcomes in E18.5 mouse concepti after different periods of embryo culture. (A) Embryo culture experimental paradigm. White and black bars denote timing of in vivo and in vitro culture development for each group, respectively. Refer to Methods for details. (B–C) Fetal/placental ratio of morula and blastocyst transfer groups. (D) Representative images of placental cross-sections for morula and blastocyst transfer groups after Tpbpa in situ hybridization, counterstained with hematoxylin. Scale bar = 1 mm. (E–F) Junctional zone area as a percentage of the total placental area based on Tpbpa staining for morula and blastocyst transfer groups. (G–H) Parietal trophoblast giant cell counts normalized to placental length for morula and blastocyst transfer groups. All concepti were analyzed and each data point represents an individual conceptus. Concepti from a minimum of five different litters were analyzed for each experimental group (n = 15–25 per group). Black line represents the mean of each group. A significant (p < 0.05) global likelihood ratio test of differences between groups in the mixed effects model was followed by pairwise tests adjusted using a Holm-Bonferroni correction. Groups with different letters indicate significant differences between groups (adjusted p < 0.05); the same letters indicate no difference was detected.

FIGURE 2

DNA methylation status in E18.5 placentas after different periods of embryo culture during preimplantation development. Bisulfite pyrosequencing was conducted on E18.5 placentas for (A) H19/Igf2 ICR, (B) Kcnq1ot1 ICR and (C) Peg3 ICR morula transfer groups and (D) H19/Igf2 ICR, (E) Kcnq1ot1 ICR and (F) Peg3 ICR blastocyst transfer groups. All concepti were analyzed and each data point represents an individual placenta from a minimum of five different litters (n = 15–25 per group). Black line represents the mean of each group. Statistical significance of imprinted gene assays was determined by both one-way ANOVA for differences in the mean and F-ratio test for differences in variability. Groups with different letters indicate significant differences between means (p < 0.05); the same letters indicate no difference in means was detected. Groups with asterisks have significantly different variability from groups with no asterisks (p < 0.05).

FIGURE 3

Flow diagram showing probe analysis of the Infinium Mouse Methylation BeadChip. (A) Embryo culture (EC) includes both 1-cell-morula and 1-cell-blastocyst groups (n = 24) versus Natural (n = 10). (B) 1-cell-morula (n = 12) versus Natural (n = 10). (C) 1-cell-blastocyst group (n = 12) versus Natural (n = 10).

FIGURE 4

Infinium Mouse Methylation BeadChip results using Bumphunter showing the number of annotated DMRs affected by embryo culture and their genomic location. Bar graphs showing annotated DMRs by predicted genomic regions using annotatr R package: (A,E,I) 1-cell-morula versus Natural for all, for females (F) and males (M). (B,F,J) 1-cell-blastocyst versus Natural for all, for females (F) and males (M) and (D,H,L) overlap between 1-cell-morula versus Natural and 1-cell-blastocyst versus Natural for all, for females (F) and males (M) placentas. Bar graphs were done using mm10/GRCm10 mouse genome assembly. The number of identified regions were higher than our number of DMRs because some DMRs overlap with more than one genomic locus. Venn diagrams for Hypomethylated DMRs using package VennDiagram for R (this package considered multiple DMRs in a gene as one): (C,G,K) 1-cell-morula versus Natural and 1-cell-blastocyst versus Natural to find overlap between the two analyses for all, for females and males. For all analyses, 1-cell-morula (n = 12, six females and six males), 1-cell-blastocyst (n = 12, six females and six males) and Natural (n = 10, five females and five males).

FIGURE 5

Distribution of annotated DMRs from the Bumphunter results using Repetitive Elements classes. (A,D,G) Bar graph showing 1-cell-morula versus Natural for all, for females (F) and males (M). (B,E,H) Bar graph showing 1-cell-blastocyst versus Natural for all, for females (F) and males (M). (C,F,I) Bar graph of the overlap between 1-cell-morula versus Natural and 1-cell-blastocyst versus Natural for all, for females (F) and males (M). For all analyses, 1-cell-morula (n = 12, six females and six males), 1-cell-blastocyst (n = 12, six females and six males) and Natural (n = 10, five females and five males). For Bumphunter analyses, cutoff value = 0.1. For more details on the parameters used for Bumphunter, refer to Materials and Methods. Repetitive classes were found using RepeatMasker tools.

FIGURE 6

Gene ontology pathway analysis and placenta-specific gene analysis using DMRs affected by embryo culture. (A) Top gene ontology pathways affected by embryo culture using genes obtained by Bumphunter. Fold enrichment indicates the ratio of the differentially methylated gene number to the total gene number in a certain pathway. The size of each circle represents the number of genes contained in a particular cluster; a larger circle indicates a larger number of genes. The color indicates the p-value (dark orange = high, light orange = low). (B) Placental enriched specific genes affected by embryo culture; darker color blue indicates higher number of transcripts in a specific tissue. For Bumphunter analyses, cutoff value = 0.1. For more details on the parameters used for Bumphunter, refer to Materials and Methods.