Early Expression of Tet1 and Tet2 in Mouse Zygotes Altered DNA Methylation Status and Affected Embryonic Development

Abstract

Ten-eleven translocation (Tet) dioxygenases can induce DNA demethylation by catalyzing 5-methylcytosine(5mC) to 5-hydroxymethylcytosine(5hmC), and play important roles during mammalian development. In mouse, Tet1 and Tet2 are not expressed in pronucleus-staged embryos and are not involved in the genomic demethylation of early zygotes. Here, we investigated the influence of Tet1 and Tet2 on methylation of parental genomes by ectopically expressing Tet1 and Tet2 in zygotes. Immunofluorescence staining showed a marked 5hmC increase in the maternal pronucleus after injection of Tet1 or Tet2 mRNA into zygotes. Whole-genome bisulfite sequencing further revealed that Tet2 greatly enhanced the global demethylation of both parental genomes, while Tet1 only promoted the paternal demethylation. Tet1 and Tet2 overexpression altered the DNA methylation across genomes, including various genic elements and germline-specific differently methylated regions. Tet2 exhibited overall stronger demethylation activity than Tet1. Either Tet1 or Tet2 overexpression impaired preimplantation embryonic development. These results demonstrated that early expression of Tet1 and Tet2 could substantially alter the zygotic methylation landscape and damage embryonic development. These findings provide new insights into understanding the function of Tet dioxygenases and the mechanism of DNA methylation in relation to embryogenesis.

Keywords: DNA methylation; Tet1; Tet2; embryonic development; mice; zygotes.

Figure 1

Methylation patterns of mouse zygotes microinjected with Tet mRNA. (A) Immunofluorescence staining for 5mC and 5hmC in zygotes injected with GFP (n = 39), Tet1 (n = 29), or Tet2 (n = 25) mRNAs. The non-injected zygotes were used as the control (n = 27). (B) Immunofluorescence staining for H3K9me2 and Stella in zygotes. “♀”, female pronucleus, “♂”, male pronucleus. Scale bar = 20 μm.

Figure 2

Methylation levels of parental genomes in gametes and zygotes. (A) A global view of DNA methylation in each sample. The proportions of 100 bp tiles at five methylation levels are shown. (B) Changes in the methylation levels of representative regions obtained using the IGV browser. (C) Mean DNA methylation levels of various functional regions and transposon regions. (D,E) The paternal and maternal genome methylation levels in subcategories of LTR repeats, respectively.

Figure 3

Analysis of Tet1- and Tet2-acted sites. (A,B) The numbers of methylation changed-sites from Tet1 and Tet2 injection compared with GFP injection. The DMRs overlapping region ≥ 1 bp and q-value < 0.01 were considered methylation overlapped sites. The term “hypomethylated sites” was defined by more than 20% methylation reduction than the GFP injection control, while “hypermethylated sites” were the opposite. (C–J) The propensity of Tet1 and Tet2 to act on sequences flanking CG. The observed/expected (obs/exp) values are presented as the frequency of occurrence of individual bases. The sum of the (obs/exp − 1)2 values of A, C, G, and T at CG positions context (−10 to +10), considered relative variation, were examined separately on the parental genome. The analysis of hypomethylated sites (C,D) and hypermethylated sites (E,F) is shown (hypomethylation and hypermethylation here represent a single CG position situation, FDR < 0.05). (G–J) Detailed base frequency for different positions of CG. “n” indicates the numbers of sequences analyzed.

Figure 4

Effects of Tet1 and Tet2 on germline-specific DMRs. Germline-specific DMRs were defined as DMRs with ≥75% methylation levels in one parent and ≤ 25% in the other, and were examined for their presentation on the parental genome of zygotes. (A,B) Heatmaps for the methylation distribution of sperm-specific DMRs (n = 16,583) (A) and oocyte-specific DMRs (n = 100) (B). (C,D) Bar graphs for the mean methylation (±SEM) of the sperm-specific DMRs (C) and oocyte-specific DMRs (D) were examined in all samples. (E) Heatmap for the methylation levels of 25 maternal and 3 paternal known gICRs. The mean methylation of these gICRs from various groups is shown in the box graph above the heatmap. (F) Bar graphs for the mean methylation levels of 5 representative gICRs. (G–J) Graphical representation of methylation in paternally imprinted genes (H19 and Rasgrf1) and maternal imprinted genes (Snurf/Snrpn and Kcnq1ot1) at a locus in the gametes and zygotes. Different colors of short lines highlight the tracked methylated CpGs. “*”, p < 0.05; “**”, p < 0.01; “***”, p < 0.001; “****”, p < 0.0001.

Figure 5

Genes with promoter methylation altered by the action of Tet1 and Tet2. (A,B) GO-term enrichment analysis of functions of genes with hypomethylated promoters caused by Tet1 (A) or Tet2 (B). The genes were enriched with criteria at −lg (p−value), p−value < 0.01.

Figure 6

The effect of overexpression of Tet1 and Tet2 on embryonic development. (A) The blastocysts developed from zygotes injected with GFP, Tet1, or Tet2 mRNA. (B) Preimplantation development rates of the injected zygotes. (C) Representatives of blastocytes stained with DAPI for labeling the cells. (D) Statistical analysis of total cell number in blastocysts. The number of stained blastocysts in each group was more than 25. Scale bar = 100 μm. “**”, p < 0.001; “***”, p < 0.0001.