Differential dynamics of histone H3 methylation at positions K4 and K9 in the mouse zygote

Appearance of histones on paternal chromosomes

We performed the immunostaining against core histones (anti-PanHistone antibodies) in all the stages tested in combination with antibodies, recognizing the specific methylated forms of histone H3. This served as a positive control for the immunostaining procedure and allowed us to follow the dynamics of histone acquisition in the paternal pronucleus. Histones were first detected shortly after the penetration of sperm into the oocyte and the beginning of the decondensation of sperm chromatin. According to PN stages classification we could clearly detect histones on paternal pronucleus at late PN₀/early PN₁ stages (approx. 3–5 hours p.f.), exactly when the global demethylation starts [8] (Fig. 1).

Dynamic changes in H3/K4 methylation in paternal genome

Probing the mouse zygotes at different stages with antibodies specifically recognizing either mono- or tri-methylated H3/K4 revealed that these types of modifications are associated with maternal genome through all zygotic stages, including mature oocyte and seem to be rather ubiquitous (Fig. 1a,1b). As for the paternal pronucleus – we detect the appearance of H3/K4mono-methylation in the beginning of PN₁ (approx. 5 hours p.f.) stage (Fig. 1a), only slightly delayed compared to the appearance of core histones (Fig. 2). By PN₃ – PN₄ stages both paternal and maternal pronuclei show equal staining intensity. This indicates that H3/K4 specific histone methyltransferase, possibly Set9 [13], is quite active in the early zygote and methylates histone H3 after it is incorporated into the nucleosomes. In contrast to that, it has been shown recently that H3/K9 specific histone methyltransferase is inactivated immediately after the fertilization by yet unknown active mechanism, which involves de novo synthesis of some specific factors [11]. H3/K4tri-methylation becomes detectable later, starting from PN₄ stage (approx. 8–10 hours p.f.) and the difference in antibodies staining intensity between paternal and maternal pronuclei becomes indistinguishable in the last pronuclear stage PN₅ (approx 12 hours p.f.) (Fig. 1b) and in metaphase stage of first mitosis approximately 16 hours p.f. (Fig. 3a). The fact that H3/K4 first becomes mono-methylated and several hours later tri-methylated suggests progressive methylation of histone H3 at lysine 4. We also suggest that histone H3 gets incorporated into the nucleosomes being unmethylated and then undergoes methylation because we observe first the appearance of histones and then H3/K4mono-methylation. In contrast to that – acetylation of histones H3 and H4 happens before they are incorporated into the nucleosomes, and after the nucleosome assembly they can get deacetylated by histone deacetylases (HDACs) whenever required [14]. But no histone demethylase has been found so far.

H3/K9 methylation but not H3/K4 defines the genomes asymmetry in the mouse zygote

In order to compare the patterns of H3/K4 and H3/K9 methylation we performed the immunostaining of mouse zygotes using antibodies, which recognize di-methylated H3/K9. Our results are in the agreement with earlier observations that H3/K9 methylation is only attributed to the maternal genome and is completely absent from the paternal [9–11] (Fig. 1c, Fig. 3b). In normal somatic cells the absence or disruption of H3/K9 methylation leads to the chromosome instability and affects chromosomes segregation during mitosis [15]. Therefore the absence of H3/K9 methylation on paternal chromosomes is rather surprising and compromises its role in chromosomes segregation. The epigenetic asymmetry between paternal and maternal genomes is observed till 2-cell stage and is characterized by low levels of DNA methylation and H3/K9 methylation in paternal genome [8, 10, 11, 16]. In case with H3/K4 methylation – the asymmetry is observed only in the beginning of the zygotic development and is indistinguishable in the metaphase stage of the first mitotic division (Fig. 3a). Recent data from Liu et al. suggest that H3/K9 methylation does not depend on DNA methylation [11], but it is only paternal DNA which gets demethylated in the mouse zygote and at the same time it does not have detectable H3/K9 methylation. According to data published by Santos et al. [17, 18] DNA demethylation starts at PN₁ stage, i.e. at a time when we first observe the appearance of H3/K4mono-methylation (PN₁ stage, Fig. 1a), and is completed at PN₃ stage when H3/K4mono-methylation in paternal pronucleus reaches approximately the same level as in the maternal (Fig. 1a). This fact is raising the question if such a coincidence might indicate that DNA demethylation and the establishment of H3/K4 methylation are interdependent. Demethylation of paternal DNA upon the fertilization is not a universal phenomenon for mammalian species. In bovine zygote paternal DNA becomes only partially demethylated, while in sheep and rabbit zygotes the demethylation is hardly detectable [17, 18]. The analysis of chromatin modification in early zygotes of these species might help to get an answer if DNA demethylation depends on, or is directed by the specific chromatin modifications.

In vitrofertilization of mouse oocytes

As sperm and oocytes donors we used (C57BL/6 X CBA)F₁ mice. Mature oocytes were collected 14 hours post human chorionic gonadotropin injection according to standard procedures [19]. Sperm isolation and in vitro fertilization (IVF) procedures were performed as described in [20]. Briefly: the sperm was isolated from cauda epididimus of donor males and capacitated in pre-gassed HTF medium for 1,5 hours. Isolated oocytes in cumulus cell mass were placed into 100 μl drop of HTF medium with capacitated sperm and incubated in CO₂ incubator for 3, 5, or 8 hours. For longer incubation time the oocytes were incubated with sperm in HTF medium for 8 hours and then transferred into the drop of pre-gasses and pre-warmed M16 medium and incubated further for 2, 4 or 10 hours.

Immunofluorescence staining

After the removal of zona pellucida by treatment with Acidic Tyrode's solution fertilized oocytes were fixed for 20 min in 3.7% paraformaldehyde in PBS, and permeabilized with 0.2% Triton X-100 in PBS for 10 min at room temperature. The fixed zygotes were blocked overnight at 4°C in 1% BSA, 0.1% Triton X-100 in PBS. After blocking the embryos were incubated in the same solution with either anti PanHistones (mouse polyclonal, Roche), anti mono-methyl H3/K4 (rabbit polyclonal, Abcam), anti tri-methyl H3/K4 (rabbit polyclonal, Abcam) or anti di-methyl H3/K9 (rabbit polyclonal, a gift from T. Jenuwein [21] antibodies at room temperature for 1 hour, followed by several washes and incubation for 1 hour with anti-mouse secondary antibodies coupled with fluorescein (Sigma-Aldrich), and anti-rabbit secondary antibodies coupled with Rhodamine Red-X (Jackson ImmunoResearch Laboratories Inc.). After final washes the zygotes were placed on slides and mounted with a small drop of Vectashield (VectorLab) mounting medium containing 0.5 μg 4,6-diamino-2-phenylindole (DAPI). At least 20 zygotes have been analyzed for each stage of zygotic development.

Immunofluorescence microscopy

The slides were analyzed on Zeiss Axiovert 200 M inverted microscope equipped with the fluorescence module and B/W digital camera for imaging. The images were captured, pseudocoloured and merged using AxioVision software (Zeiss).