Oxygen-induced alterations in the expression of chromatin modifying enzymes and the transcriptional regulation of imprinted genes

Abstract

Embryo culture and assisted reproductive technologies have been associated with a disproportionately high number of epigenetic abnormalities in the resulting offspring. However, the mechanisms by which these techniques influence the epigenome remain poorly defined. In this study, we evaluated the capacity of oxygen concentration to influence the transcriptional control of a selection of key enzymes regulating chromatin structure. In mouse embryonic stem cells, oxygen concentrations modulated the transcriptional regulation of the TET family of enzymes, as well as the de novo methyltransferase Dnmt3a. These transcriptional changes were associated with alterations in the control of multiple imprinted genes, including H19, Igf2, Igf2r, and Peg3. Similarly, exposure of in vitro produced bovine embryos to atmospheric oxygen concentrations was associated with disruptions in the transcriptional regulation of TET1, TET3, and DNMT3a, along with the DNA methyltransferase co-factor HELLS. In addition, exposure to high oxygen was associated with alterations in the abundance of transcripts encoding members of the Polycomb repressor complex (EED and EZH2), the histone methyltransferase SETDB1 and multiple histone demethylases (KDM1A, KDM4B, and KDM4C). These disruptions were accompanied by a reduction in embryo viability and suppression of the pluripotency genes NANOG and SOX2. These experiments demonstrate that oxygen has the capacity to modulate the transcriptional control of chromatin modifying genes involved in the establishment and maintenance of both pluripotency and genomic imprinting.

Keywords: Assisted reproductive technologies; DNA methylation; DNMT; Developmental programming; Epigenetics; Genomic imprinting; Histone demethylase; Oxidative stress; TET.

Fig. 1.

The impact of oxygen concentrations on bovine preimplantation development and the transcriptional regulation of select genes involved in the oxidative stress response and the control of pluripotency at the blastocyst stage. A. Cleavage and blastocyst rates of bovine embryos cultured under low (5%) and high (20% - atmospheric) oxygen concentrations (5% oxygen n = 682, 20% oxygen n = 890. from 4 independent replicates). Here, development rates represent the total number of blastocysts on day 7 divided by the total number of presumptive zygotes placed into culture. B. Measurement of transcripts encoding genes involved in oxygen sensing and the oxidative stress response using reverse transcriptase quantitative polymerase chain reaction (RT-qPCR). C. RT-qPCR measurement of transcripts encoding key bovine pluripotency genes. For experiments employing RT-qPCR, measurements were normalized to the geometric mean of GAPDH, SDHA and YWHAZ. n = 6. (P-value: * ≤ 0.05; ** ≤ 0.01; *** ≤ 0.001).

Fig. 2.

Oxygen-induced changes in the transcriptional regulation of genes controlling chromatin structure at the blastocyst stage of bovine development. A. RT-qPCR measurements of transcripts encoding the bovine ten-eleven translocation (TET) gene family. B. Transcript levels of the bovine DNA methyltransferase genes and an associated recruiter, HELLS. C. Measurement of transcripts encoding the bovine histone lysine demethylase family of genes. D. Gene transcript levels of the histone modifying complexes PRC1, PRC2, and histone methyltransferases. RT-qPCR measures were normalized to the geometric mean of GAPDH, SDHA, and YWHAZ. n = 6. (P-value: * ≤ 0.05; ** ≤ 0.01; *** ≤ 0.001).

Fig. 3.

Oxygen-induced changes in a mouse embryonic stem cells. A. Schematic representation of the experimental workflow. Cells split from a common flask (Passage 0) were separated into high and low treatment groups. Low (5%) O₂ cells were cultured in a “hypoxia chamber”. B. Representative image of murine embryonic stem cells at mid-confluency during experimental treatments. C. Surface area (as calculated using ImageJ software) between the treatment groups. D-E. Measurement of transcripts encoding genes involved in the oxidative stress response at cellular passage 1 and 8. RT-qPCR measures were normalized to the geometric mean of Hprt, Mrpl, Ppia, and Ywhaz. n = 6 (P-value: * ≤ 0.05; ** ≤ 0.01; *** ≤ 0.001).

Fig. 4.

Oxygen-induced changes in the transcriptional regulation of genes controlling DNA methylation in mouse embryonic stem cells. A-B. Transcript levels of the ten-eleven translocation (TET) family of genes at passage 1 and 8. C-D. Transcript levels of the DNA methyltransferase [DNMT] family of genes at passage 1 and 8. RT-qPCR measures were normalized to the geometric mean of Hprt, Mrpl, Ppia, and Ywhaz. n = 6 (P-value: * ≤ 0.05).

Fig. 5.

Oxygen-induced changes in the transcriptional regulation of imprinted genes within mouse embryonic stem cells. Analysis of imprinted gene transcript levels at passage 1 and 8, as determined by RT-qPCR. Measurements were normalized to the geometric mean of the reference genes Gapdh, Hprt, Mrpl and Sdha, and samples graphed relative to Passage 0 (20% oxygen). n = 5 (P-value: * ≤ 0.05; ** ≤ 0.01; *** ≤ 0.001).

Fig. 6.

Analysis of oxygen-induced changes in parent-of-origin expression patterns for select imprinted loci. A. Visual representation of allelic inheritance from the distinct genetic backgrounds used in this study. This strategy allows tracking of the parental alleles by using single nucleotide polymorphisms. B. Visual representation of balanced, bi-allelically expressed transcripts as well as imbalanced, imprinted genes. C. Allele-specific transcript analysis of a select panel of known imprinted genes, as determined using deep RNA-sequencing. (P-value: **** ≤ 0.0001).