Two distinct mechanisms localise cyclin B transcripts in syncytial Drosophila embryos

Drosophila embryos 0 2 JORDAN W. RAFF 0 1 2 WILLIAM G. F. WHITFIELD 0 2 DAVID M. GLOVERf 0 2 0 Cancer Research Campaign Laboratories, Cell Cycle Genetics Group, Department of Biochemistry, Medical Sciences Institute, The University , Dundee DD1 4HN, Scotland 1 Present address: Department of Biochemistry and Biophysics, University of California , San Francisco, California 94143 , USA 2 Cyclin B transcript localisation in Drosophila - We demonstrate that two independent mechanisms act on maternally derived cyclin B transcripts to concentrate the transcripts at the posterior pole of the Drosophila oocyte and at the cortex of the syncytial embryo. The cortical accumulation occurs because the cyclin B transcript is concentrated around nuclei and comigrates with them to the cortex. The perinuclear localisation of the transcript is blocked by inhibitors of microtubule polymerisation and the transcript colocalises with microtubular structures during the cell cycle, suggesting that the transcript is associated either directly or indirectly with microtubules. Neither microtubules The cyclins are a family of proteins that accumulate during interphase, and are then rapidly degraded at about the time of the metaphase-anaphase transition (for review see Hunt, 1989; Swenson et al. 1989). This unique temporal pattern suggested that the cyclins might be involved in controlling the entry into and exit from mitosis, a hypothesis that has recently received strong experimental support. First, it has been demonstrated that cyclin synthesis is required for the entry into mitosis and meiosis in a number of systems (Swenson et al. 1986; Pines and Hunt, 1987; Minshull et al. 1989; Westendorf et al. 1989). Second, in a cell-free extract from Xenopus eggs in which all endogenous mRNA has been destroyed, the addition of cyclin B mRNA reconstitutes the characteristic cycling of cyclin B protein levels, which drives the extract through multiple rounds of mitosis-like events (Murray and Kirschner, 1989). Third, if the cyclin B mRNA that is added to the extract contains a deletion of the region that encodes the first 90 amino acids, the cyclin that is synthesised is not degraded and the extract becomes blocked in mitosis, demonstrating that cyclin destruction is required for the exit from mitosis (Murray et al. 1989). The DNA sequences encoding the Drosophila nor actin filaments are required to maintain the posterior concentration of cyclin B transcripts. Instead, this seems to depend on the association of the transcripts with a component of the posterior cytoplasm. The distribution pattern of the transcript at the posterior pole throughout embryogenesis and in a variety of mutant embryos suggests that this component is associated with polar granules. homologues of cyclin A and cyclin B have recently been cloned"(Lehner and O'Farrell, 1989; Whitfield et al. 1989). Both proteins show the expected pattern of synthesis and destruction during the cell divisions that take place after cellularisation has occurred at nuclear cycle 14 (Lehner and O'Farrell, 1989, 1990; Whitfield et al. 1990). The behaviour of the cyclin proteins prior to cellularisation, however, has not been described in any detail. During this period, the nuclei in the syncytial embryo proceed through a very rapid series of nearly synchronous nuclear divisions (Zalokar and Erk, 1976; Foe and Alberts, 1983). At nuclear cycle 7-8, the majority of nuclei, which are initially located in the interior of the embryo, coordinately start to migrate to the embryo cortex. Early in cycle 9, the nuclei reach the cortex at the posterior pole, where they initiate the formation of pole buds. The nuclei in these buds undergo two further rounds of division before they pinch off from the embryo to form pole cells, the future germ cells (Mahowald, 1962; Counce, 1963; Warn et al. 1985). Early in cycle 10 the somatic nuclei reach the cortex where they proceed through three further rounds of division before cellularisation occurs. The cyclins appear to be maintained at high levels throughout these nuclear cycles, and there is no large-scale destruction of the cyclin proteins across the whole syncytium (Lehner and O'Farrell, 1989, 1990; Maldonado-Codina and Glover, unpublished data). The distribution of cyclin transcripts in the Drosophila embryo was described by Whitfield et al. (1989) and Lehner and O'Farrell (1990). Both cyclin A and cyclin B transcripts are homogeneously distributed in the early embryo, but, in addition, cyclin B transcripts are concentrated at the posterior pole at some time prior to pole bud formation. The cyclin B transcripts at the posterior pole become incorporated into the developing pole cells, where they remain at high levels throughout embryonic development. By nuclear cycle 10-11, when most of the nuclei have migrated to the embryo cortex, the majority of the cyclin B transcript in the rest of the embryo becomes tightly localised to the embryo cortex. These transcripts decrease dramatically in abundance upon cellularisation at cycle 14. Levels of the transcript then rise in somatic cells (presumably resulting from zygotic transcription) as gastrulation starts and cell divisions resume. The localisation of specific transcripts to particular regions of the developing Drosophila embryo seems to play a crucial part in embryonic development. In perhaps the best studied example, the localisation of the bicoid transcript to the anterior end of the embryo results in the formation of a morphogenic protein gradient (Driever and Nusslein-Volhard, 1988a,fo). Sequences in the 3' untranslated region of the bicoid mRNA are required for it to become correctly localised (Macdonald and Struhl, 1989). Little else is known, however, about the mechanisms that localise specific transcripts in the Drosophila embryo. In this paper, we investigate the mechanisms that are responsible for localising cyclin B transcripts both to the posterior pole and subsequently to the rest of the cortex. Materials and methods Embryo collection, injection and fixation Embryos were collected and injected as described in Raff and Glover (1988). The drugs used in this study were injected at the following concentrations: a--amanatin, 400^gml~I; colchicine, 1X10~3M; cytochalasin D, 250/itgml"1; aphidicolin, 100/igmP1. Embryos were then fixed in 1:1 mixture of 4% paraformaldehyde in PBS (PP)/heptane. Embryos that were not injected were fixed directly in this mixture. The aqueous phase was removed, and the embryos were devitellinised by the addition of an equal volume of methanol followed by vigorous shaking for 30s (Mitchison and Sedat, 1983). The embryos were rehydrated in a methanol/PP series (7:3, 1:1, 3:7 - 5min each) and then postfixed in PP for 20min. If the embryos were to be stored, they were dehydrated in an ethanol series and stored at 20C in 70% ethanol. Preparation of 35S-RNA probes All probes were made from plasmid constructs containing cDNAs inserted between (...truncated)