Loss of genomic imprinting in mouse embryos with fast rates of preimplantation development in culture

Abstract

Currently, the stage of embryo development has been proposed as one of many criteria for identifying healthy embryos in infertility clinics with the fastest embryos being highlighted as the healthiest. However the validity of this as an accurate criterion with respect to genomic imprinting is unknown. Given that embryo development in culture generally requires an extra day compared to in vivo development, we hypothesized that loss of imprinting correlates with slower rates of embryonic development. To evaluate this, embryos were recovered at the 2-cell stage, separated into four groups based on morphological stage at two predetermined time points, and cultured to blastocysts. We examined cell number, embryo volume, embryo sex, imprinted Snrpn and H19 methylation, imprinted Snrpn, H19, and Cdkn1c expression, and expression of genes involved in embryo metabolism-Atp1a1, Slc2a1, and Mapk14-all within the same individual embryo. Contrary to our hypothesis, we observed that faster developing embryos exhibited greater cell numbers and embryo volumes as well as greater perturbations in genomic imprinting and metabolic marker expression. Embryos with slower rates of preimplantation development were most similar to in vivo derived embryos, displaying similar cell numbers, embryo volumes, Snrpn and H19 imprinted methylation, H19 imprinted expression, and Atp1a1 and Slc2a1 expression. We conclude that faster development rates in vitro are correlated with loss of genomic imprinting and aberrant metabolic marker expression. Importantly, we identified a subset of in vitro cultured embryos that, according to the parameters evaluated, are very similar to in vivo derived embryos and thus are likely most suitable for embryo transfer.

FIG. 1.

A) Graphical representation of the embryo separation scheme. At the first separation, embryos were divided into fast and slow groups based on whether they reached the 8-cell stage or not. At the second separation, the original fast group was split into fast/fast (FF) and fast/slow (FS) groups based on whether they had a cavity or not, and the original slow group was divided into slow/fast (SF) and slow/slow (SS) based on whether they were compacted or not. B) Individual blastocyst assay for multiple data sets. Top left: merge of bright field and Hoechst 33342 staining used to count cell numbers. Top right: Sry expression analysis used for embryo sex determination. L, ladder; F, female control; M, male control; Bl, blastocyst; -ve, negative control. Bottom left: paternal H19 methylation analysis. Filled circles, methylated CpG dinucleotides; unfilled circles, unmethylated CpGs. Each row represents one DNA strand. Bottom right: LightCycler H19 imprinted expression analysis that allows for quantitation of allelic expression. Blastocyst FS23 had 48 cells and a volume of 3.5 × 10⁵ μm³, was male, and displayed 70% hypermethylation on the H19 paternal allele and biallelic H19 expression with 19% activation from the normal silent paternal allele.

FIG. 2.

Cell numbers and embryo volumes of the four groups of cultured embryos and in vivo derived embryos. A) Representative embryos showing closest cell number and volume to mean of the group. Bar = 10 μm. B) Left: mean cell numbers in fast and slow groups based on the first separation. The fast group had significantly more cells than the in vivo derived group (*) and the slow group (**). Right: cell numbers separated by culture group. Each diamond represents one embryo, and black bars indicate mean cell number of each group. C) Left: mean embryo volume in fast and slow groups based on the first separation. Fast embryos had significantly larger volumes than in vivo derived (*) and slow embryos (**). Slow embryos displayed significantly fewer cells than in vivo derived embryos (*). Right: embryo volumes separated by culture group. Black bars indicate mean embryo volume of each group. Error bars represent standard errors of the mean.

FIG. 3.

Imprinted methylation analysis of the maternal Snrpn allele in fast, slow, fast/fast, fast/slow, slow/fast, and slow/slow groups. Mean hypermethylation levels are shown in brackets. Each group of circles represents one embryo, with the embryo name indicated in the top left. Percent hypermethylation is indicated in the top middle. Each row represents one DNA strand. Filled circles represent methylated CpG dinucleotides while unfilled circles represent unmethylated CpGs. Only data for the maternal allele are shown because no difference in the methylation status of the unmethylated paternal Snrpn allele was observed.

FIG. 4.

Graphical representation of Snrpn and H19 hypermethylation levels in culture groups. Top: maternal Snrpn hypermethylation levels. Bottom: paternal H19 hypermethylation levels. Each diamond represents one embryo, and black bars represent mean hypermethylation levels in fast, slow, fast/fast (FF), fast/slow (FS), slow/fast (SF), and slow/slow (SS) groups. The asterisk (*) in diamonds indicates two to three embryos within the group with the same hypermethylation level.

FIG. 5.

Imprinted methylation analysis of the paternal H19 allele in fast, slow, fast/fast, fast/slow, slow/fast, and slow/slow groups. Mean hypermethylation levels shown in brackets. Each group of circles represents one embryo, with the embryo name indicated in the top left. Percent hypermethylation is indicated in the top middle. Each row represents one DNA strand. Filled circles represent methylated CpG dinucleotides while unfilled circles represent unmethylated CpGs. Only the data for the paternal allele are shown because no difference in the methylation status of the unmethylated maternal H19 allele was observed.

FIG. 6.

H19 and Cdkn1c imprinted expression analysis in the four groups of cultured embryos. Red bars indicate percent expression from the maternal allele, and blue bars indicate percent expression from the paternal allele. A) H19 imprinted expression analysis. B) Cdkn1c imprinted expression analysis. Embryo names are indicated on the X axis. C) Developmental Cdkn1c imprinted expression in in vivo derived embryos. Embryo stage indicated on the X axis: 4-cell, n = 2, 12 pooled embryos each; 8-cell, n = 2, 6 pooled embryos each; Early M, early morula, n = 2, 3 pooled embryos each; Late M, late morula, n = 2, 1 embryo each; Mid-BL, midblastocyst; n = 7, 1 embryo each; Late BL, late blastocyst; n = 3, 1 embryo each.

FIG. 7.

Relative expression of genes involved in embryo metabolism in cultured and in vivo derived embryos. A) Left panel: mean relative Atp1a1 expression in fast and slow groups based on the first separation. Fast embryos displayed significantly higher Atp1a1 expression levels than slow embryos (**). Right panel: relative Atp1a1 expression in the four groups of cultured embryos and in vivo derived embryos. B) Left panel: mean relative Slc2a1 expression, corrected for cell numbers, in fast and slow groups based on the first separation. Fast and slow embryos displayed significantly lower Slc2a1 expression levels than in vivo derived embryos (*). Fast embryos displayed significantly lower Slc2a1 expression levels than slow embryos (**). Right panel: relative Slc2a1 expression corrected for cell numbers in the four groups of cultured embryos and in vivo derived embryos. C) Left panel: mean relative Mapk14 expression in fast and slow groups based on the first separation. No difference in mean relative Mapk14 expression was observed between in vivo derived, fast, and slow embryos. Right panel: relative Mapk14 expression in the four groups of cultured embryos and in vivo derived embryos. Embryo group is indicated on the X axis, each diamond represents one embryo, and black bars represent the mean relative expression in each group. Error bars represent standard errors of the mean.