transcription/translation machinery is accompanied by the appearance of diacetylated histone H4, colocalizing with RNA polymerase II at the periphery of blastomere nuclei

In the mouse, new embryonic transcription appears after fertilization during the S phase of the first cell cycle in the male pronucleus, while the female pronucleus is largely repressed (Bouniol et al., 1995; Christians et al., 1995; Majumder et al., 1993; Ram and Schultz, 1993). Both pronuclei then become equally repressed during the subsequent G2 phase, concomitant with de novo synthesis of histones H2A, H2B and H1 (Wiekowski et al., 1997) and extensive histone H4 deacetylation (Adenot et al., 1997). The first intense burst of translation-coupled zygotic transcription (zygotic genome activation, ZGA) appears at the early G1 phase of the two-cell stage with the activation of a few genes (Conover et al., 1991; Latham et al., 1991) that include, among others, the heat shock gene hsp70.1 (Bensaude et al., 1983; Christians et al., 1995). This transcriptional wave is marked by a number of features, as replacement of the oocyte-derived largest subunit of RNA polymerase II with an embryo-specific form (Bellier et al., 1997) and synthesis of proteins relevant to RNA synthesis, maturation and translation (Davis et al., 1996; Latham et al., 1995; Worrad et al., 1994). These result in a substantial change in the embryo’s transcriptional and translational controls with respect to those acting at the one-cell stage. The increase in the overall efficiency of embryo’s transcription/translation machinery is accompanied by the appearance of diacetylated histone H4, colocalizing with RNA polymerase II at the periphery of blastomere nuclei (Worrad et al., 1995). This suggests that at the time of ZGA, the mouse zygotic genome is largely assembled into compact chromatin and that only a few chromosomal domains, preferentially located near the nuclear membrane, are engaged in transcription. These features imply that a locally acting mechanism(s) of chromatin relief newly appears at ZGA. Information on this issue is so far only indirect and mostly based on evidence derived from transient expression experiments, in which enhancer-less or enhancer-provided DNA constructs were microinjected into either one-cell or two-cell mouse embryos (Majumder et al., 1993). It appeared that, whilst mid one-cell embryos efficiently express both types of DNA constructs, two-cell embryos can only transcribe the enhancer-provided ones, unless they are treated with a histone deacetylase inhibitor (Majumder et al., 1993). Enhancers are thus dispensable for transcription of injected constructs at the one-cell stage, but are required at the early two-cell stage to relieve chromatin-dependent repression. This model of enhancer function at ZGA has recently been 1541 Development 127, 1541-1551 (2000) Printed in Great Britain © The Company of Biologists Limited 2000 DEV4259