Impaired mitotic progression and preimplantation lethality in mice lacking OMCG1, a new evolutionarily conserved nuclear protein

Abstract

While highly conserved through evolution, the cell cycle has been extensively modified to adapt to new developmental programs. Recently, analyses of mouse mutants revealed that several important cell cycle regulators are either dispensable for development or have a tissue- or cell-type-specific function, indicating that many aspects of cell cycle regulation during mammalian embryo development remain to be elucidated. Here, we report on the characterization of a new gene, Omcg1, which codes for a nuclear zinc finger protein. Embryos lacking Omcg1 die by the end of preimplantation development. In vitro cultured Omcg1-null blastocysts exhibit a dramatic reduction in the total cell number, a high mitotic index, and the presence of abnormal mitotic figures. Importantly, we found that Omcg1 disruption results in the lengthening of M phase rather than in a mitotic block. We show that the mitotic delay in Omcg1-/- embryos is associated with neither a dysfunction of the spindle checkpoint nor abnormal global histone modifications. Taken together, these results suggest that Omcg1 is an important regulator of the cell cycle in the preimplantation embryo.

FIG. 1.

Omcg1 transcripts and protein structure. (A) Nucleotide and predicted amino acid sequence of mouse Omcg1 cDNA. Nucleotide sequence corresponds to Omcg1 mRNA RefSeq NM_025884; sequences 3′ of nt 2441 (empty arrowhead) have been derived from a DDK oocyte cDNA (dashed underlining). Sequence data for this additional sequence are available from GenBank/EMBL/DDBJ under accession no. AY786540. The black arrowheads point to the 3′ end of Omcg1 shorter transcripts: AK017640 (1,287 nt) and AK008475 (1,578 nt). The deadenylation signal (attta), oocyte-specific cytoplasmic polyadenylation element (ttttat), and polyadenylation signal (aataaa) are indicated in bold italic. The coiled-coil and zinc finger domains are shadowed in gray and black, respectively. Putative bipartite nuclear localization signals (NLS) are indicated by solid underlining. The short stretch of negatively charged residues is boxed. (B) Schematic representation of the 3′ UTR of the three Omcg1 transcripts. (C) Schematic representation of OMCG1 protein.

FIG. 2.

Comparison between OMCG1 and its orthologous proteins. (A) Sequence comparison of OMCG1 and related proteins. Homo sapiens (human), accession no. BAB70992; Rattus norvegicus (rat; protein sequence was determined from conceptual translation of the rat genomic locus, identified by sequence alignment of the mouse cDNA with rat genomic sequence, clone CH230-105H4), accession no. AC128344; Danio rerio (zebra fish), accession no. AAH65604; C. elegans (nematode), SwissProt accession no. Q22139; and D. melanogaster (drosophila), accession no. AAF56392. Identical amino acid residues are shaded black, similar residues are shaded gray, and different residues are not shaded. Intermediate part of OMCG1 (underlined) is poorly conserved. (B) Percentage of sequence identity between related proteins.

FIG. 3.

Expression of Omcg1 in adult and preimplantation embryo. (A) Northern blot analysis of poly(A)⁺ mRNA of E7.5 to E17.5 embryos and of various adult tissues probed with an Omcg1-specific probe (upper panels) or a β-actin probe (lower panels). (B) OMCG1 immunodetection in metaphase II oocytes and early embryos. Bar, 30 μm.

FIG. 4.

Targeted disruption of Omcg1 gene. (A) Schematic diagram showing the Omcg1 wild-type locus, the targeting vector, and the targeted allele. Black boxes represent Omcg1 ORF, while white boxes represent the 5′ and 3′ UTRs. The hatched boxes represent the βgeo cassette, and the gray box represents the Pgk-DTA cassette used for negative selection. Arrows 1 to 3 indicate the positions and the orientations of PCR primers used for genotype analysis in panel C. External and internal probes used in panel B are shown as solid bars beneath the targeted allele. Restriction enzymes: A, AflII; Bs, Bsp120I. (B) Southern blot analysis of genomic DNA obtained from wild-type and heterozygous ES cells. (C) Genotyping analysis of E4.5 embryos recovered from Omcg1^+/− heterozygous intercrosses. A 470-bp band is amplified from wild-type allele by primers 1 and 2, whereas a 430-bp is amplified from targeted allele by primers 1 to 3. (D) Omcg1-targeting constructs were electroporated into CK35 Omcg1^+/+ cells or Omcg1^+/− ES cells. After selection, resistant clones were analyzed by Southern blotting and/or by PCR to determine their genotypes.

FIG. 5.

Aberrant blastocyst outgrowths of Omcg1^−/− embryos. E3.5 blastocysts from Omcg1 intercrosses were cultured in vitro and photographed every 24 h using phase-contrast microscopy. At the end of the culture, each embryo was recovered and genotyped by nested PCR. Wild-type (A) and heterozygous blastocysts hatched from their zona pellucida after 1 day and attached to the culture dish. After 3 days in culture, outgrowths were composed of the inner cell mass surrounded by a single layer of trophectoderm giant cells (TGC). Omcg1-deficient embryos either failed to hatch (B) or, when they did, failed to proliferate (C). After 1 day in culture, some were composed of larger trophectoderm cells (white arrow) and a small globular ICM (asterisk in panel B). (D and E) Two examples of the culture of chemically dezoned null blastocysts. The first one (D) attached but failed to spread on the dish and to proliferate; the second one (E) also attached to the dish but, after 2 days in culture, most cells were degenerating (empty arrowheads), while a few plurinucleated cells could still be seen (black arrows).

FIG. 6.

OMCG1, OCT-3/4, and TROMA-1 distribution in Omcg1^+/− and Omcg1^−/− blastocysts. (A) Immunodetection of OMCG1 protein in E3.5 and E4.5 Omcg1^+/− and Omcg1^−/− blastocysts. OMCG1 nuclear signal is present in E3.5 embryos but disappears in E4.5 null embryos. Nuclei were counterstained by Hoechst 33342. Embryos were observed under a fluorescence microscope. Insets show higher magnifications of a single cell. (B and C) Immunodetection of two markers of cell lineages in E4.5 embryos from Omcg1^+/− intercrosses: OCT-3/4 and TROMA-1 are expressed in ICM (B) and TE (C), respectively. Embryos were observed by confocal microscopy. In panel B, images show a medial orientation of a three-dimensional reconstruction. After analysis, embryos were individually recovered and genotyped. Bar, 20 μm.

FIG. 7.

Cell parameters of E4.5 Omcg1^+/− intercross embryos. (A) Nuclei from E4.5 intercross blastocysts were stained with DAPI and counted under a fluorescent microscope. (B) TUNEL assay was performed to detect apoptotic cells in E4.5 embryos. (C) Determination of mitotic cell numbers was done by immunofluorescence with an anti-HH3ser10P antibody. In each experiment, embryos were recovered after microscopic examination and genotyped by nested PCR. Statistical analyses were performed by ANOVA. An asterisk indicates significant difference. n, number of embryos analyzed in each genotype.

FIG. 8.

Analysis of the cell cycle of E4.5 Omcg1^+/− intercross embryos. (A) E4.5 intercross blastocysts were cultured for 3 h in the presence of BrdU. The number of BrdU-positive cells was counted using an anti-BrdU antibody in embryos with three or fewer mitotic cells (white box) and with seven or more mitotic cells (black box). (B) The number of mitotic cells was established, using an anti-HH3ser10P antibody, in E4.5 embryos from Omcg1^+/− intercrosses cultured for different times and under different conditions: E4.5 embryos (white boxes), E4.5 embryos cultured for 7 h in absence of drugs (blue boxes), E4.5 embryos cultured for 7 h in the presence of mitomycin C (orange boxes), and E4.5 embryos cultured for 3 h in the presence of nocodazole (green boxes). n, number of embryos analyzed. (C) HH3ser10P immunofluorescence of embryos analyzed in panel B. Bar, 40 μm. (D) MAD2 immunofluorescence on mitotic cells from E4.5 intercross blastocysts with or without nocodazole treatment. Bar, 4 μm.

FIG. 9.

Mitotic spindle defects in deficient embryos. E4.5 Omcg1^−/− blastocysts were stained with Hoechst 33342 (blue), anti-α-tubulin (red), and anti-HH3ser10P or anti-γ-tubulin antibodies (green). (A) Normal organization of the spindle and the metaphasic plate in an Omcg1^−/− cell. Some Omcg1^−/− cells displayed abnormal spindles, for example, a tripolar spindle (B) or misalignment between spindle and chromosomes accompanied by isolated chromosomes (C, asterisk). (D) Normal organization of the mitotic spindle was often associated with two centrosomes. (E and F) Some Omcg1-null cells exhibited abnormal numbers of centrosomes. In nonmitotic cells, centrosomes were stained green, while in mitotic cells they were stained yellow because of the colocalization of α- and γ-tubulin to the poles of the spindle. Bar, 2 μm.