Cyclins E1 and E2 are required for endoreplication in placental trophoblast giant cells

Abstract

In mammalian cells, cyclin E-CDK2 complexes are activated in the late G1 phase of the cell cycle and are believed to have an essential role in promoting S-phase entry. We have targeted the murine genes CCNE1 and CCNE2, encoding cyclins E1 and E2. Whereas single knockout mice were viable, double knockout embryos died around midgestation. Strikingly, however, these embryos showed no overt defects in cell proliferation. Instead, we observed developmental phenotypes consistent with placental dysfunction. Mutant placentas had an overall normal structure, but the nuclei of trophoblast giant cells, which normally undergo endoreplication and reach elevated ploidies, showed a marked reduction in DNA content. We derived trophoblast stem cells from double knockout E3.5 blastocysts. These cells retained the ability to differentiate into giant cells in vitro, but were unable to undergo multiple rounds of DNA synthesis, demonstrating that the lack of endoreplication was a cell-autonomous defect. Thus, during embryonic development, the needs for E-type cyclins can be overcome in mitotic cycles but not in endoreplicating cells.

Fig. 1. Targeted disruption of CCNE2 in mice. (A) Schematic representation of the CCNE2 locus (top), targeting vector (middle) and targeted allele (bottom). Gray boxes represent the exons. The arrowheads represent the PCR primers used to amplify the wild-type and targeted alleles for genotyping. Restriction enzymes: A, ApaI; t, AatI; B, BamHI; h, SphI; K, KpnI; p, SpeI; S, SacI. The sites in brackets were used for cloning but lost in the process. See Materials and methods for experimental details. (B and C) PCR-based genotyping (top) and mRNA analysis (bottom) from CCNE2 and CCNE1 mutants, respectively. Genomic DNA was extracted from ear-punches and RNA from E13.5 mouse embryonic fibroblasts.

Fig. 2. Cyclin E1^–/– and E2^–/– single knockout cells are not impaired in cell cycle entry. The indicated wild-type and mutant mouse embryonic fibroblasts (from two independent preparations) were synchronized by serum starvation and contact inhibition, and subsequently released into cell cycle. DNA synthesis was monitored by flow cytometric analysis of DNA content and bromodeoxyuridine incorporation.

Fig. 3. Phenotypes of cyclin E1^–/–E2^–/– double knockout (dKO) embryos. (A) E10.5 dKO embryos are smaller than control siblings. This relative difference in size was observed in different litters. (B) The vasculature in the dKO yolk sac at E11.5 is nearly absent. Note that the characteristic pattern of yolk sac vasculature in the control (left) is missing in the dKO (right).

Fig. 4. Bromodeoxyuridine staining of E10.5 embryos shows equally abundant S phases in double knockout and control siblings.

Fig. 5. Trophoblast giant cells in double knockout (dKO) placentas have a reduced amount of DNA. Haemotoxylin and eosin (A) and Feulgen (B) staining of sections through E10.5 placentas. The giant cell layer is either magnified in the inserts (A) or indicated by the dotted lines (B). The arrowheads show the giant nuclei: note the intense staining of nuclei in the control embryos, which is proportional to DNA content and is missing in the dKO. The placental layers are indicated as follows: D, decidua; TGC, trophoblast giant cells; S, spongiotrophoblast; L, labyrinth.

Fig. 6. Cyclins E1 and E2 are required for endoreplication in trophoblast giant cells (TGCs) in vitro. (A) Flow cytometric analysis of DNA content in nuclei of differentiating trophoblast stem cells (TSCs). (B) Feulgen staining of undifferentiated TSCs (day 0) and following differentiation into TGCs (day 12).

Fig. 7. mRNA expression analysis in differentiating trophoblast stem cells (TSCs). Wild-type and double knockout TSCs were induced to differentiate for 12 days. Every 2 days, RNA was isolated and expression of specific mRNAs was quantified with reverse transcription and real-time PCR. The results, normalized to ubiquitin, are expressed as fold-induction relative to day 0 in the wild-type, and represent the average of 3–5 different experiments from different TSC clones. (A) Ubiquitin mRNA (control) and placental markers. (B) Cell cycle genes.