Functional and molecular reorganization of the nucleolar apparatus in maturing mouse oocytes.

In mammalian preovulatory oocytes, rRNA synthesis is down-regulated until egg fertilization and zygotic genome reactivation, but the underlying regulatory mechanisms of this phenomenon are poorly characterized. We examined the molecular organization of the rRNA synthesis and processing machineries in fully grown mouse oocytes in relation to ongoing rDNA transcription and oocyte progression throughout meiosis. We show that, at the germinal vesicle stage, the two RNA polymerase I (RNA pol I) subunits, RPA116 and PAF53/RPA53, and the nucleolar upstream binding factor (UBF) remain present irrespective of ongoing rDNA transcription and colocalize in stoichiometric amounts within discrete foci at the periphery of the nucleolus-like bodies. These foci are spatially associated with the early pre-rRNA processing protein fibrillarin and in part with the pre-ribosome assembly factor B23/nucleophosmin. After germinal vesicle breakdown, the RNA pol I complex disassembles in a step-wise manner from chromosomes, while UBF remains associated with chromosomes until late prometaphase I. Dislodging of UBF, but not of RNA pol I, is impaired by the phosphatase inhibitor okadaic acid, thus strengthening the idea of a relationship between UBF dynamics and protein phosphorylation. Since neither RNA pol I, UBF, fibrillarin, nor B23 is detected at metaphase II, i.e., the normal stage of fertilization, we conclude that these nucleolar proteins are not transported to fertilized eggs by maternal chromosomes. Together, these data demonstrate an essential difference in the dynamics of the major nucleolar proteins during mitosis and meiosis.

Induction of early transcription in one-cell mouse embryos by microinjection of the nonhistone chromosomal protein HMG-I.

In the mouse embryo, the onset of zygotic transcription occurs at the end of the first cell cycle, upon completion of DNA replication. We show that the nonhistone chromosomal protein HMG-I, whose translocation into the pronuclei of one-cell embryos is linked to this first round of DNA synthesis, plays a critical role in the activation of zygotic transcription. Indeed, microinjection of purified HMG-I results in a higher nuclear accumulation of the protein and triggers an earlier activation of zygotic transcription, an effect which is abolished by the preincubation of the protein with a specific antibody directed against its AT-hook DNA-binding motifs. Significantly, microinjection of this antibody also prevents the normal onset of transcription in the embryo, suggesting that endogenous HMG-I is similarly involved in this process. Finally, microinjection of the exogenous protein modifies chromatin structure as measured by in situ accessibility to DNase I. We propose that general chromosomal architectural factors such as HMG-I can modulate the accessibility of chromatin to specialized regulatory factors, thereby promoting a transcriptionally competent state.

Differential transcriptional activity associated with chromatin configuration in fully grown mouse germinal vesicle oocytes.

It was previously shown that fully grown ovarian germinal vesicle (GV) oocytes of adult mice exhibit several nuclear configurations that differ essentially by the presence or absence of a ring of condensed chromatin around the nucleolus. These configurations have been termed, respectively, SN (surrounded nucleolus) and NSN (nonsurrounded nucleolus). Work from our and other laboratories has revealed ultrastructural and functional differences between these two configurations. The aims of the present study were 1) to analyze the equilibrium between the SN and the NSN population as a function of the age of the mice and the time after hCG-induced ovulation and 2) to study the polymerase I (pol I)- and polymerase II (pol II)-dependent transcription in both types of oocytes through the detection of bromouridine incorporated into nascent RNA. We show 1) that ovarian GV oocytes exhibiting the SN-type configuration can be found as soon as 17 days after birth in the C57/CBA mouse strain and 2) that the SN:NSN ratio of ovarian GV oocytes is very low just after hCG-induced ovulation and then increases progressively with the time after ovulation. Furthermore, we demonstrate that the SN configuration correlates strictly with the arrest of both pol I- and pol II-dependent transcription in mice at any age. Finally, we show that ribosomal genes are located at the outer periphery of the nucleolus in the NSN configuration and that pol I-dependent perinucleolar transcription sites correspond to specific ultrastructural features of the nucleolus. Altogether, these results provide clear-cut criteria delineating transcriptionally active GV oocytes from those that are inactive, and confirm that the SN-type configuration is mostly present in preovulatory oocytes.

Dynamic organization of DNA replication in one-cell mouse embryos: relationship to transcriptional activation.

We have analyzed the spatial and temporal relationship between transcription and replication sites during the first cell cycle in mouse embryos. Embryos were microinjected with both 5-bromouridine-5'-triphosphate and digoxygenin-11-deoxyuridine-5'-triphosphate to visualize transcription and replication sites respectively. We detected six different phases of replication during S phase and dated the onset of zygotic transcription at the end of the S phase. Using confocal microscopy, we showed that there is essentially no colocalization of replication and transcription sites at this stage of development. Moreover, studies on aphidicolin-treated embryos demonstrated that inhibition of DNA replication does not hinder transcriptional activation at the 1-cell stage.