Mutations at the pebble locus of Drosophila melanogaster result in embryonic lethality. Examination of homo-zygous mutant embryos at the end of embryogenesis revealed the presence of fewer and larger cells which contained enlarged nuclei. Characterization of the embryonic cell cycles using DAPI, propidium iodide, anti-tubulin and anti-spectrin staining showed that the first thirteen rapid syncytial nuclear divisions proceeded normally in pebble mutant embryos. Following cellular-ization, the postblastoderm nuclear divisions occurred (mitoses 14, 15 and 16), but cytokinesis was never observed. Multinucleate cells and duplicate mitotic figures were seen within single cells at the time of the cycle 15 mitoses. We conclude that zygotic expression of the pebble gene is required for cytokinesis following cellularization during Drosophila embryogenesis. We postulate that developmental regulation of zygotic transcription of the pebble gene is a consequence of the transition from syncytial to cellular mitoses during cycle 14 of embryogenesis.

Recent dramatic advances in the molecular genetic analysis of the eukaryotic cell cycle have revealed highly conserved mechanisms that regulate progress through the cell cycle (for reviews, see Nurse, 1990; Reed, 1991). Much less is known about the usage of cell cycle control points in the developmental regulation of cell proliferation. Developmental regulation of the cell cycle in eukaryotes has been documented in a number of cases. In the budding yeast, Saccharomyces cerevi-siae, cells arrest at the G1 phase of the cell cycle in response to the mating pheromone, a-factor, (for a review, see Thomer, 1981). During embryogenesis in Drosophila melanogaster, a developmentally controlled period of cell cycle arrest occurs in the G2 phase of cycle 14. In this case, cell cycle arrest has been shown to be the result of the regulated expression of string mRNA (Edgar and O’Farrell, 1989). string is a Drosophila homologue of the cdc25 mitotic initiator of Schizosaccharomyces pombe, which encodes a product required for the activation, by dephosphorylation, of the p34cdc2 protein kinase (Gould and Nurse, 1989). Maturation of Xenopus laevis oocytes also involves a period of cell cycle arrest at the G2-M transition. The arrested oocytes provided an assay for a factor, termed maturation or M-phase promoting factor (MPF) that is found in mitotic cells. When added to arrested oocytes this factor triggered entry into mitosis (Masui and Markert, 1971; Smith and Ecker, 1971). This factor was subsequently shown to contain the p34cdc2 protein (Dunphy et al., 1988; Gautier et al., 1988) which is required for entry into mitosis in all eukaryotes examined to date. The relationship between growth arrest in mammalian tissue culture cells and cell cycle regulation is poorly understood, but the recent demon-stration that activation of macrophages by the CSF1 growth factor results in the expression of cyclin-like molecules (Matsushime et al., 1991) may be the first clue to such regulation. These newly identified cyclins may be members of the G1 cyclin group required for progression past “start” in the Saccharomyces cerevisiae cell cycle (Richardson et al., 1989).

Development of Drosophila melanogaster provides an excellent system for the study of the developmental regulation of cell proliferation. The cellular basis of Drosophila embryogenesis has been well characterized and molecular genetic studies have led to the identifi-cation of regulatory genes that generate the complex patterns of cell and tissue types during organogenesis (Ingham, 1988). The embryonic cell cycles of Dros-ophila have been characterized in detail. With the exception of the vitellophages and germ cells, which are not discussed in this paper, the first thirteen mitoses are parasynchonous and rapid, occurring at approximately 10 minute intervals (Rabinowitz, 1941; Foe and Alberts, 1983). During cycles 7 to 10, a series of nuclear movements towards the periphery of the embryo results in the formation of a syncytial blastoderm.

Following the 13 parasynchronous mitoses, cell proliferation ceases for at least 70 minutes (Foe, 1989). During this period, cellularization proceeds by the growth of membranes from the periphery of the embryo to enclose the nuclei, resulting in the formation of the cellular blastoderm.

The 14th mitoses are radically different from all preceding mitoses. They are the first cellular mitoses and they occur asynchronously, groups of cells dividing in a complex spatio-temporal pattern. Each group of synchronously dividing cells, termed a mitotic domain, has been accurately mapped with respect to its position in the embryo and the time at which it undergoes mitosis (Foe, 1989). The domains frequently coincide with organ primordia, so this pattern of mitosis is one of the earliest manifestations of organ formation.

As noted above, zygotic transcription of one gene, string, has been shown to regulate the timing and spatial occurrence of the 14th mitoses (Edgar and O’Farrell, 1989, 1990). Here we describe a second gene, pebble (pbl) the transcription of which is required for cytokin-esis during the postblastoderm mitoses.

Strains, media and growth conditions

Fly stocks were grown on standard media at 25°C or room temperature. Egg lays were performed at 25°C on apple juice agar plates (Wieschaus and Ntisslein-Volhard, 1986). Egg lays were usually for one hour with the eggs then allowed to age for the desired time at 25°C before removal of the chorion membranes with 4% hypochlorite.

The EMS alleles used in this study were pbl5D, pbl7O, pbl1ID and pbl5B (Jürgens et al., 1984). Df(3L)pbrR and Df(3L)pblx deficiency strains were generated during this study and are described in the text, w; ve TE 217/TM3Sb was obtained from G. Ising. The wild-type strain used was Canton-S.

Recombination mapping and X-ray mutagenesis

Recombination percentages were calculated as the percentage of recombinant progeny scored for the markers indicated in Fig. 1. X-ray mutagenesis was performed by irradiating batches of 100 w; ve TE 217/TM3Sb males with 4800 rad and mating them to 100 w; TM3Sb/TM6b or w; rusteca females. Putative deficiencies were scored by the reversion of TE 217 to w-.

Fig. 1.

(A) Location of the pbl locus, pbl has been localized to subdivision 66B of chromosome 3. Recombination percentages are indicated above arrows connecting the appropriate markers. The extent of deficiencies, Df(3L)pblX1, 65F3;66B9 and Df(3L)pblNK, 66B1;B2, are indicated by hatched boxes. Abbreviations used are Hn; Henna, [Pw+]; P[w]30 (G.Rubin, personal communication), h; hairy. (B and C) Cytological preparations of Df(3L)pblxl/ + and Df(3L)pblNR/+ respectively. Arrows indicate the loopout in the wild-type chromosome.

Fig. 1.

(A) Location of the pbl locus, pbl has been localized to subdivision 66B of chromosome 3. Recombination percentages are indicated above arrows connecting the appropriate markers. The extent of deficiencies, Df(3L)pblX1, 65F3;66B9 and Df(3L)pblNK, 66B1;B2, are indicated by hatched boxes. Abbreviations used are Hn; Henna, [Pw+]; P[w]30 (G.Rubin, personal communication), h; hairy. (B and C) Cytological preparations of Df(3L)pblxl/ + and Df(3L)pblNR/+ respectively. Arrows indicate the loopout in the wild-type chromosome.

Polytene chromosome spreads

Polytene spreads were performed using the method of Pardue (1986).

Immunohistochemistry

Dechorionated embryos were fixed using the method of Karr and Alberts (1986) and antibody staining was performed as described by Foe (1989) except that the secondary antibody incubations were performed at 37°C for 2 hours in the presence of 1 mg/ml RNAaseA. Nuclei were observed by the addition to the mounting media of 10 μg/ml propidium iodide. RNAase treatment of embryos was used to avoid propidium iodide staining of cytoplasmic RNA, allowing nuclei to be clearly observed, p-phenylenediamine was added to the mounting medium (90% glycerol/PBS) to prevent fluorescence quenching (Johnson and Nogueira Araujo, 1981). Alternatively, nuclei were visualized by incubating fixed embryos in 10 μg/ml DAPI for 3 minutes. Antibodies used were MC10-2, an anti-a--spectrin monoclonal antibody (Pesacreta et al., 1989) and YL1/2, a rat anti-tyrosinated a- tubulin antibody (Kilmartin et al., 1982). MC10-2 was used at 1:1; YL1/2 (SeraLabs) at 1:5 and 22C10 at 1:100. Secondary and tertiary antibodies used were 1:30 biotinylated anti-mouse (Amersham); 1:100 streptavidin Texas Red (Amer-sham) and 1:30 FITC-anti-mouse (Silenus). In all exper-iments, the progeny of flies heterozygous for particular pebble mutations were examined. Approximately one-quarter of the embryos showed the mutant phenotype in all cases. Although we could not discriminate between the pbl/+ and +/+ embryos, we refer to this class of phenotypically wild-type embryos as “wild-type” throughout this paper.

Microscopy

Fluorescence microscopy was performed on a Zeiss Axioplan microscope equipped with objectives Plan-Neofluar 20× /0.5 and 100× /1.3 oil immersion and Planapochromat 40× /1.0 oil immersion. DAPI staining was observed using a Zeiss No. 2 filter. Confocal microscopy was performed with a Bio-Rad MRC 600 Confocal Imaging System fitted to an Olympus BH2-RFCA microscope, using objective SPlan 100× /1.25 oil immersion. Images were recorded by photographing the screen of a Mitsubishi FA3435KE/V colour monitor.

Genetic characterization of the pebble region

Recombination mapping of pbl placed it between the genes Henna and hairy on chromosome 3 (Table 1, Fig. 1A.). Deficiencies covering this region were not available, so attempts were made to generate them using a transposable element insertion, TE217, pre-viously localized to 66B1-8 (G. Ising, personal com-munication). Recombination between pbl and TE 217 showed these to be less than 0.5 map units apart. TE elements are very large transposable elements, several hundred kilobases in size, carrying a region of the X-chromosome including the genes roughest and white (Ising and Block, 1984). This large size may result in the suppression of recombination, and therefore underesti-mation of the genetic distance between the point of insertion and the pbl locus. Despite this uncertainty, TE217 was used to detect potential deficiencies by the loss of the w+ phenotype. The screening of 16797 flies recovered 19 revertants to w~, from which 13 lines were established. Two deficiencies were generated. X-irradiation generated a deficiency, termed Df(3L)pblXI, which failed to complement any pbl mutant alleles. Df(3L)pblXI was mapped cytologically as a deficiency from 65F3 to 66B9 of the Drosophila salivary gland polytene chromosome map described by Bridges (1941) (Fig. IB). A spontaneous revenant to w- was also found among the TE217 stock and shown to be the result of a deficiency. This deficiency, Df(3L)pblNR, was found to delete bands 66B1-2, (Fig. 1C) and failed to complement any pbl mutant allele. We have therefore generated deficiencies used in the further analysis of the pbl phenotype described below and assigned pbl to the cytological interval, 66B1 – 2.

Table 1.

Recombination between markers in subdivision 66

Recombination between markers in subdivision 66
Recombination between markers in subdivision 66

Mutations at the pbl locus result in an abnormal nuclear phenotype

pbl mutations result in recessive embryonic lethality with a characteristic mutant cuticle phenotype (Jürgens et al., 1984). Approximately one quarter of offspring from heterozygous parents exhibited a mutant pheno-type as visualized by DAPI staining of 15– 20 hours AED (after egg deposition) embryos. Mutant embryos derived from parents carrying the EMS-induced pbl70 allele showed nuclei that were reduced in number and exhibited gross morphological alterations (Fig. 2B) relative to normal embryos in which nuclei were organized into clearly definable organ structures (Fig. 2A). The pbl5D, pbl11D and pbf5B alleles yielded an identical mutant phenotype to that of pbl70 (results not shown), as did embryos tram-heterozygous for the deficiencies Df(3L)pblNR and Df(3L)pblxl (Fig. 2C). Apart from those nuclei undergoing mitosis, nuclei of wild-type 5-7 hours AED embryos were evenly distrib-uted and of a constant size (Fie. 2D). Homozygous pbl70 (Fig. 2E) and Df(3L)pbNR (Fig. 2F) embryos displayed nuclei of uneven distribution and varying size. Large diffusely stained structures were present together with smaller brightly stained nuclei in clusters of two and four. From this preliminary characterization, we conclude that the pbl mutation results in disruption of cell proliferation during Drosophila embryogenesis.

Fig. 2.

Nuclear morphology as revealed by DAPI staining and fluorescence. (A)-(C) 16– 20 hours AED embryos. (A) wild-type (B) pbl70 homozygote (C) Df(3L)pbtNR/DfiSLipbl* Bar=200 microns. (D-F) 5– 7 hours AED embryos. (D) wild-type (E) pbl70 homozygote (F) Df(3L)pblNR homozygote. Bar=400 microns.

Fig. 2.

Nuclear morphology as revealed by DAPI staining and fluorescence. (A)-(C) 16– 20 hours AED embryos. (A) wild-type (B) pbl70 homozygote (C) Df(3L)pbtNR/DfiSLipbl* Bar=200 microns. (D-F) 5– 7 hours AED embryos. (D) wild-type (E) pbl70 homozygote (F) Df(3L)pblNR homozygote. Bar=400 microns.

Cells of pbl mutant embryos are multinucleate after the postblastoderm mitoses

Examination of mutant embryos, using DAPI staining of nuclei and Differential Interference Contrast mi-croscopy, failed to reveal any disruption either of the first 13 rapid syncytial divisions or of cellularization (results not shown). The appearance of abnormal nuclei was first noted after the 14th mitosis, the first postblastoderm mitosis. This is also the first mitosis after cellularization of the peripheral nuclei. To determine if the nuclear clustering, observed in the DAPI-stained embryos, was the result of multinucleate cells, 4– 5 hours AED embryos were double stained with an antibody directed against cr-spectrin, which is distributed on the inner surface of plasma membranes (Pesacreta et al., 1989), and with propidium iodide. This allowed co-visualization of nuclei and the sur-rounding cell membrane. Embryos were viewed by laser scanning confocal microscopy, thereby generating an optical slice through the whole-mount embryos and showing the relationship of nuclei to plasma mem-branes. In the photomicrographs shown in Fig. 3 and described below, the occurrence of apparently anucleate cells is due to the nucleus of that cell not being in the plane of the image.

Fig. 3.

Nuclear phenotype of pbl embryos. Cell membranes were stained with anti-a-spectrin Ab (red) and nuclei with propidium iodide, (yellow-green) (A) wild-type (B-D) pbl70 homozygotes showing multinucleate cells and multiple mitotic figures. (E) pbl70 homozygote showing large diffuse nuclei. Bar=5 microns.

Fig. 3.

Nuclear phenotype of pbl embryos. Cell membranes were stained with anti-a-spectrin Ab (red) and nuclei with propidium iodide, (yellow-green) (A) wild-type (B-D) pbl70 homozygotes showing multinucleate cells and multiple mitotic figures. (E) pbl70 homozygote showing large diffuse nuclei. Bar=5 microns.

Cells in stained wild-type embryos contained single mitotic figures or single interphase nuclei of uniform size. The example shown is at the germ band retraction stage with all nuclei in interphase (Fig. 3A). The cells of homozygous pbl70 embryos were much larger in size and were variable in shape. Cells containing four nuclei were observed (Figs 3B, 3D) with some undergoing mitosis (Fig. 3C). The cell at the top of Figure3B shows a binucleate cell in telophase of cycle 14 whilst the cell in the centre is tetranucleate, having just undergone mitosis 15. After cessation of the postblastoderm mitoses, most nuclei were observed as large diffuse interphase structures (Fig. 3E).

The presence of mitotic figures at the appropriate postblastoderm stages and of enlarged multinucleate cells subsequent to this suggest that cells of pbl embryos proceed through the nuclear divisions of the postblasto-derm mitoses but do not undergo cytokinesis to produce two daughter cells with individual nuclei.

Cells in pbl mutant embryos can have more than one mitotic spindle

As described above, the cycle 14 nuclear mitoses occur normally in pbl mutant embryos, but apparently in the absence of cytokinesis. To determine whether spindle formation was normal in the mutant mitoses, 5– 7 hours AED embryos were stained with YL1/2 antiserum directed against tyrosinated a-tubulin (Kilmartin et al., 1982). Wild-type and heterozygous embryos exhibited a wild-type pattern of mitotic spindles arranged in domains as described by Foe (1989) (Fig. 4A,C). Observation of a large number of embryos, one quarter of which must have been homozygous for the pbl70 mutation showed that the mutant embryos were indistinguishable from wild-type or heterozygous em-bryos during the cycle 14 mitoses prior to cytokinesis (results not shown). Mitoses at the time expected for the 15th mitoses, however, were observed to have duplicate mitotic apparatuses. Figs 4B and 4D-4F show several examples of the presence of two mitotic spindles in cells undergoing the 15th mitosis. Fig. 4F is from an embryo rrans-heterozygous for the deficiencies Df(3L)pblNR and Df(3L)pblXI and shows a group of four overlapping spindles. These spindles were ob-served to be disordered, presumably because of the interference of the establishment or stability of each spindle by the presence of the other. The presence of an apparent tripolar spindle in Fig. 4D is due to a cell containing two orthogonal spindles in which one spindle pole was out of the plane of the image.

Fig. 4.

Mitotic spindles of (A) wild-type and (B) pbl70 embryos revealed by staining with a rat anti-tyrosinated-o’-tubulin Ab. Bar=25 microns. C and D are enlargements of A and B. D and E are examples of pbl70 homozygotes showing cells with multiple spindles. (F) Df(3L)pblNR/DftSLjpbl*1 showing four overlapping spindles within a single cell. Bar= 5 microns.

Fig. 4.

Mitotic spindles of (A) wild-type and (B) pbl70 embryos revealed by staining with a rat anti-tyrosinated-o’-tubulin Ab. Bar=25 microns. C and D are enlargements of A and B. D and E are examples of pbl70 homozygotes showing cells with multiple spindles. (F) Df(3L)pblNR/DftSLjpbl*1 showing four overlapping spindles within a single cell. Bar= 5 microns.

The appearance of duplicate mitotic figures within single cells following the 14th mitosis is further evidence that the mutant phenotype is due to normal postblasto-derm nuclear divisions in the absence of cytokinesis, resulting in polyploid cells.

Developmental effects of the pbl mutation

Although normal cell proliferation was seen to be disrupted at the end of cycle 14 in pbl mutant embryos, postblastoderm morphogenetic movements and tissue differentiation still occurred, pbl embryos gastrulated and germ-band extension occurred to varying degrees. A few embryos extended the germ band fully and formed some segmental structures. For example, the embryo shown in Fig. 5A contained an almost complete cephalopharyngeal skeleton and four rudimentary den-ticle bands. Thus cells of pbl mutant embryos were seen to differentiate. Differentiation of other tissue types was demonstrated. mAb 22C10, a monoclonal antibody directed against differentiating cells of the peripheral nervous system (Zipursky et al., 1984), stained subsets of cells that had the general distribution expected of the developing PNS but were disorganized in comparison to the wild-type pattern (result not shown). Thus the mitotic disruption caused by the pbl mutation does not prevent the cells from receiving and interpreting, at least to some degree, the positional information that directs pattern formation and cell differentiation in normal embryos. A similar observation has been made in cells mutant for the string gene which blocks all mitotic activity beyond that of the G2 phase of cycle 14 (Hartenstein and Posakony, 1990).

Fig. 5.

Differentiated structures present in pbl embryos. (A) pbl70 homozygote that has completed germ band retraction and has partially segmented, exhibiting rudimentary denticle bands (small arrowhead) and cephalopharyngeal skeleton present (large arrowhead). (B) DAPI stain of the same embryo shown in A demonstrating pbl phenotype. Bar=200 microns.

Fig. 5.

Differentiated structures present in pbl embryos. (A) pbl70 homozygote that has completed germ band retraction and has partially segmented, exhibiting rudimentary denticle bands (small arrowhead) and cephalopharyngeal skeleton present (large arrowhead). (B) DAPI stain of the same embryo shown in A demonstrating pbl phenotype. Bar=200 microns.

We have used a variety of cell biological techniques to characterize the phenotype of pbl mutant embryos. The key observations were (1) the normal development of embryos up to the cellular blastoderm stage, (2) the appearance of mitotic figures with normal mitotic spindles and with normal chromosome segregation at the times expected for the cycle 14 mitoses, (3) the presence of multinucleate cells following these mitoses, (4) the appearance at later stages of multiple mitotic figures within single cells and (5) the presence of fewer cells with enlarged nuclei at the end of embryogenesis. There are several possible explanations for these observations. One is that the pbl gene encodes a factor required for cytokinesis so that nuclear mitoses proceed in the absence of cytokinesis in pbl mutants. A second is that cell fusion follows the normal occurrence of the postblastoderm mitoses. A third is that a novel cell cycle, such as the “endo-cell cycle”, proposed by Smith and Orr-Weaver (1991) for the generation of polytene cells later in development, is prematurely triggered in pbl mutant embryos. The cell fusion explanation is unlikely as we never see a quanta! decrease or increase in cell size at any stage during embryogenesis, as would be expected if cytokinesis occurred normally and was followed by fusion of the cells. Premature activation of the endo-cell cycle is also unlikely to account for the mutant phenotype, as these cycles involve endo-reduplication of the DNA rather than nuclear mitoses followed by nuclear fusion. A role for pebble in cytokinesis accounts for all the observations described in this paper and is the most likely explanation for the pebble mutant phenotype.

While the phenotype of pbl mutant embryos clearly implies a role for the pbl gene product in cytokinesis, a more subtle phenotype, that of the irregularity of cell shape, may indicate an additional role for the pbl gene product. Although the irregular cell shape most probably reflects physical forces acting on the large multinucleate cells during morphogenesis, it could result from a role for the pbl product in cytoskeletal interactions.

The large polyploid diffuse interphase nuclei seen late in development in most cells is likely to be an indirect effect of the disruption of cytokinesis. They could result either from a failure of chromosome segregation (disordered and intersecting spindles were frequently observed in cells with duplicate mitotic figures) or nuclear fusion during development.

Little is known about the process of cytokinesis, although a number of mutants affecting cytokinesis have been described. These include the Drosophila mutants spaghetti-squash, a mutation in the non-muscle myosin fight chain (Karess et al., 1991), zipper (also called mhc-c), a mutation in the non-muscle myosin heavy chain that effects cell shape change, morphogen-esis and possibly cytokinesis (D. Kiehart, personal communication), four wheel drive, a mutation causing failure of cytokinesis during male meiosis, resulting in early spermatids with four nuclei (N. Wolf and M. Fuller, manuscript in preparation) and abnormal spindle, a mutation in a non-tubulin component of the meiotic spindle (Casai et al., 1990). The C. elegans spe-ll mutant (Hill et al., 1989) and mutations at five Tetrahymena loci (Frankel et al., 1977) have also been shown to result in cytokinetic defects. One of the Tetrahymena loci encodes a protein that is concentrated at the cleavage furrow (Ohba et al., 1986).

We do not yet know whether a relationship exists between the pbl gene and any of these genes. Furthermore, we cannot as yet assign a specific role for pebble in cytokinesis. An actinomyosin motor generates the force necessary for furrowing (Otto and Schroeder, 1990), but the positions of actin and myosin genes of Drosophila do not correspond to the pbl gene (Lindsley and Zimm, 1985, 1990). Candidates for the pbl gene product include accessory proteins to myosin and actin, such as the barbed end-capping protein radixin (Sato et al., 1991), some of which must be involyed in the initiation and constriction of the contractile ring, or the association of the contractile ring with the plasma membrane. The INCENP proteins (Cooke et al., 1987; Earnshaw and Cooke, 1991) also behave in a fashion that suggests a possible role in the process of cytokinesis.

A striking feature of the pbl gene that sets it apart from other Drosophila genes implicated in cytokinesis is the requirement for zygotic expression of this gene prior to cytokinesis at the 14th mitosis. It is clear from genetic analysis of cell cycle genes in Drosophila (Gatti and Baker, 1989) that the majority of genes required for cell cycle progression are provided as maternal products in sufficient quantities to allow the embryo to proceed through all 16 embryonic mitoses. Thus, cell cycle mutants are most frequently mutants that disrupt imaginai disc and neural growth, resulting in late larval or pupal lethality, or they are maternal lethal mutants (Glover, 1989). The only other gene characterized to date for which zygotic expression is required for progression through the 14th cell cycle is the string gene. Even in this case, a large amount of maternal mRNA is present initially and is actively degraded during cycle 14, establishing the need for zygotic transcription of the gene (Edgar and O’Farrell, 1989). Why should the pbl mutant specifically affect the cycle 14 cytokineses and why should there be a requirement for zygotic pbl transcription early in embryogenesis? The answer to the first question is likely to be that the 14th mitosis is the first cellular mitosis and therefore the first mitosis in which there is a requirement for cellular cytokinesis. Thus, the pbl product may not be required during the syncytial mitoses prior to cycle 14.

The reason for the requirement for zygotic transcrip-tion is less clear. One possibility is that there is no maternal pbl mRNA provided, so that the mutant phenotype arises at the first cellular mitosis at which time the pbl gene product is first required. An alternative is that the maternal products are present during the first 13 mitoses but are actively degraded during cycle 14, as is the case for the stg mRNA. Regardless of the reason, maternal pbl products are not present when the gene product is required at the 14th mitosis. Near-saturating mutageneses of Drosophila have identified most of the genes whose zygotic transcription is required for normal cuticle formation during embryogenesis, pbl is the only such zygotic gene to have been shown to be required for cytokinesis, so we presume all other necessary factors are present as maternal products. The requirement for zygotic ex-pression may result from developmental events that convert the syncytial blastoderm to a cellular blasto-derm immediately prior to the 14th mitosis. Cellulariz-ation proceeds by the inward migration of furrow canals from the periphery of the embryo in a process that can be considered a modified cytokinesis (Warn et al., 1990). We postulate that pebble plays a key role in the process of mitotic cytokinesis and that the presence of the pebble product earlier than the 14th mitosis may interfere with the modified cytokinesis that occurs during cellularization. Transcriptional regulation of the pebble gene would ensure that the product is absent during cellularization, while expression during cycle 14 would generate the product in time for the first cellular division. This is a particularly exciting possibility, as it suggests that the pbl gene product may play a key role in the regulation of postblastoderm cytokineses. We are currently attempting to test this postulate by the isolation and further characterization of this gene.

We would like to thank Dr. C. Nüsslein-Volhard for bringing the pebble mutation to our attention and for providing fly strains, Dr W. Francis and Dr D. Turner for assistance with X-irradiation of flies, Dr T.J. Lockett, Dr B. Smith and Professor M. Vadas for assistance and use of the confocal microscope, Dr G. Ising and Dr G. Rubin for fly strains, Dr D. Branton and Dr N. Brink for providing antibodies, Dr M. Green for valuable suggestions on the work and to the other members of our laboratory for helpful discussions. This work was supported by the National Health and Medical Research Council of Australia (Grant No. 890191 to R.S.) and by the CSIRO-University of Adelaide Collaborative Research Fund. G.H. was supported by an Australian Postgraduate Research Award.

Bridges
,
P. N.
(
1941
).
A revised map of the left limb of the third chromosome of Drosophila melanogaster
.
J. Hered
.
32
,
64
65
.
Casal
,
J.
,
Gonzalez
,
C.
,
Wandosell
,
F.
,
Avila
,
J.
and
Ripoll
,
P.
(
1990
).
Abnormal meiotic spindles cause a cascade of defects during spermatogenesis in asp males of Drosophila
.
Development
108
,
251
260
.
Cooke
,
C. A.
,
Heck
,
M. M. S.
and
Earnshaw
,
W. C.
(
1987
).
The inner centromere protein (INCENP) antigens: Movement from inner centromere to midbody during mitosis
.
J. Cell Biol. 105,2053-2067
.
Dunphy
,
W. G.
,
Brizuela
,
L.
,
Beach
,
D.
and
Newport
,
J.
(
1988
).
The Xenopus cdc2 protein is a component of MPF, a cytoplasmic regulator of mitosis
.
Cell
54
,
423
431
.
Earnshaw
,
W. C.
and
Cooke
,
C. A.
(
1991
).
Analysis of the distribution of INCENPs throughout mitosis reveals the existence of a pathway of structural changes in the chromosomes during metaphase and early events in cleavage furrow formation
.
J. Cell Sci
.
98
,
443
461
.
Edgar
,
B. A.
and
O’farrell
,
P. H.
(
1989
).
Genetic control of cell division patterns in the Drosophila embryo
.
Cell
57
,
177
187
.
Edgar
,
B. A.
and
O’farrell
,
P. H.
(
1990
).
The 3 Postblastoderm Cell Cycles of Drosophila Embryogenesis Are Regulated in G2 by String
.
Cell
62
,
469
480
.
Foe
,
V. A.
and
Alberts
,
B. M.
(
1983
).
Studies of nuclear and cytoplasmic behaviour during the five mitotic cycles that precede gastrulation in Drosophila embryogenesis
.
J. Cell Sci
.
61
,
31
70
.
Foe
,
V. E.
(
1989
).
Mitotic domains reveal early commitment of cells in Drosophila embryos
.
Development
107
,
1
22
.
Frankel
,
J.
,
Nelsen
,
E. M.
and
Jenkins
,
L. M.
(
1977
).
Mutations affecting cell division in Tetrahymena pyriformis, Syngen 1. II. Phenotypes of single and double homozygotes
.
Devi Biol
.
58
,
255
275
.
Gatti
,
M.
and
Baker
,
B. S.
(
1989
).
Genes controlling essential cell-cycle functions in Drosophila melanogaster
.
Genes Dev
.
3
,
438
453
.
Gautier
,
J.
,
Norbury
,
C.
,
Lohka
,
M.
,
Nurse
,
P.
and
Mailer
,
J.
(
1988
).
Purified maturation-promoting factor contains the product of a Xenopus homolog of the fission yeast cell cycle control gene cdc2+
.
Cell
54
,
433
439
.
Glover
,
D. M.
(
1989
).
Mitosis in Drosophila
.
J. Cell Sci. 92,137-146
.
Gould
,
K. L.
and
Nurse
,
P.
(
1989
).
Tyrosine phosphorylation of the fission yeast cdc2+ protein kinase regulates entry into mitosis
.
Nature
342
,
39
45
.
Hartensteln
,
V.
and
Posakony
,
J. W.
(
1990
).
Sensillum development in the absence of cell division: the sensillum phenotype of the Drosophila mutant string
.
Devi Biol
.
138
,
147
158
.
Hill
,
D. P.
,
Shakes
,
D. C.
,
Ward
,
S.
and
Strome
,
S.
(
1989
).
A sperm-supplied product essential for initiation of normal embryogenesis in Caenorhabditis elegans is encoded by the paternal-effect embryonic-lethal gene, spe-11
.
Devi Biol
.
136
,
154
166
.
Ingham
,
P. W.
(
1988
).
The molecular genetics of embryonic pattern formation in Drosophila
.
Nature
335
,
25
34
.
Ising
,
G.
and
Block
,
K.
(
1984
).
A transposon as a cytogenetic marker in Drosophila melanogaster
.
Molec. gen. Genet
.
196
,
6
16
.
Johnson
,
G. D.
and
Nogueira Araujo
,
G. M. C.
(
1981
).
A simple method of reducing the fading of immunofluorescence during microscopy
.
J. Immunol. Methods
43
,
349
350
.
Jürgens
,
J.
,
Wieschaus
,
G.
,
Nüsslein-volhard
,
E.
and
Kinding
,
H.
(
1984
).
Mutations affecting the pattern of the larval cuticle in Drosophila melanogaster. II. Zygotic loci on the third chromosome
.
Wilhelm Roux’s Arch, devl Biol
.
193
,
283
295
.
Karess
,
R. E.
,
Chang
,
X.
,
Edwards
,
K. A.
,
Kulkaml
,
S.
,
Aguilera
,
I.
and
Kiehart
,
D. P.
(
1991
).
The regulatory light chain of nonmuscle myosin is encoded by spaghettii-squash, a gene required for cytokinesis in Drosophila
.
Cell
65
,
1177
1189
.
Karr
,
T. L.
and
Alberts
,
B. M.
(
1986
).
Organization of the cytoskeleton in early Drosophila embryos
.
J. Cell Biol
.
102
,
1494
1509
.
Kilmartin
,
J. V.
,
Wright
,
B.
and
Milstein
,
C.
(
1982
).
Rat monoclonal antitubulin antibodies derived by using a new nonsecreting rat cell line
.
J. Cell Biol
.
93
,
576
582
.
Lindsley
,
D. L.
and
Zimm
,
G.
(
1985
).
The genome of Drosophila melanogaster, Part 1: Genes A-K
.
Drosophila Information Service
62
,
5
169
.
Lindsley
,
D. L.
and
Zimm
,
G.
(
1990
).
The genome of Drosophila melanogaster, Part 4: Genes L-Z, Balancers, Transposable Elements
.
Drosophila Information Service
68
,
1
368
.
Masui
,
Y.
and
Markert
,
C. L.
(
1971
).
Cytoplasmic control of nuclear behavior during meiotic maturation of frog oocytes
.
J. exp. Zool
.
177
,
129
146
.
Matsushime
,
H.
,
Roussel
,
M. F.
,
Ashmun
,
R. A.
and
Sherr
,
C. J.
(
1991
).
Colony-stimulating factor 1 regulates novel cyclins during the G1 phase of the cell cycle
.
Cell
65
,
701
713
.
Nurse
,
P.
(
1990
).
Universal control mechanism regulating onset of M-phase
.
Nature
344
,
503
508
.
Ohba
,
H.
,
Ohmori
,
I.
,
Numa Ta
,
O.
and
Watanabe
,
Y.
(
1986
).
Purification and immunofluorescence localization of the mutant gene product of a Tetrahymena cdaAl mutant affecting cell division
.
J. Biochem (Tokyo)
.
100
,
797
808
.
Otto
,
J. J.
and
Schroeder
,
T. E.
(
1990
).
Association of actin and myosin in the contractile ring
.
Ann. New York Acad. Sci
.
582
,
179
184
.
Pardue
,
M. L.
(
1986
).
In situ hybridization to DNA of chromosomes and nuclei
. In
Drosophila: a Practical Approach
(ed.
D. B.
Roberts
) pp.
111
-
137
. Oxford: IRL Press, Ltd.
Pesacreta
,
T. C.
,
Byers
,
T. J.
,
Dubreuil
,
R.
,
Kiehart
,
D.
and
Branton
,
D.
(
1989
).
Drosophila spectrin: the membrane skeleton during embryogenesis
.
J. Cell Biol
.
108
,
1697
1709
.
Rabinowitz
,
M.
(
1941
).
Studies on the cytology and early embryology of the egg of Drosophila melanogaster
.
J. Morph
.
69
,
1
49
.
Reed
,
S. I.
(
1991
).
Gl-specific cyclins: in search of an S-phase promoting factor
.
Trends in Genetics
.
7
,
95
99
.
Richardson
,
H. E.
,
Wittenberg
,
C.
,
Cross
,
F.
and
Reed
,
S. I.
(
1989
).
An essential G1 function for cyclin-like proteins in yeast
.
Cell
59
,
1127
1133
.
Sato
,
N.
,
Yonemura
,
S.
,
Obinata
,
T.
,
Tsukita
,
S.
and
Tsnkfta
,
S.
(
1991
).
Radixin, a barbed end-capping actin-modulating protein, is concentrated at the cleavage furrow during cytokinesis
.
J. Cell Biol
.
113
,
321
330
.
Smith
,
A. V.
and
Orr-Weaver
,
T. L.
(
1991
).
The regulation of the cell cycle during Drosophila embryogenesis: the transition to polyteny
.
Development
112
,
997
1008
.
Smith
,
L. D.
and
Ecker
,
R. E.
(
1971
).
The interaction of steroids with Rana pipiens oocytes in the induction of maturation
.
Devi Biol
.
25
,
233
247
.
Thorner
,
J.
(
1981
).
Pheromonal regulation of development in Saccharomyces cerevisiae
. In
The Molecular Biology of the Yeast Saccharomyces, Life Cycle and Inheritance
, (ed.
J. N.
Strathem
,
E. W.
Jones
and
J. R.
Broach
) pp.
143
180
.
Cold Spring Harbor
:
Cold Spring Harbor Laboratory
.
Warn
,
R. M.
,
Warn
,
A.
,
Planques
,
V.
and
Robert-Nicoud
,
M.
(
1990
).
Cytokinesis in the early Drosophila embryo
.
Ann. New York Acad. Sci
.
582
,
222
232
.
Wieschaus
,
E.
and
Nüsslein-volhard
,
C.
(
1986
).
Looking at embryos
. In
Drosophila: a Practical Approach
, (ed.
D. B.
Roberts
) pp.
199
227
.
Oxford
:
IRL Press, Ltd
.
Zipursky
,
S. L.
,
Venkatesh
,
T. R.
,
Teplow
,
D. B.
and
Benzer
,
S.
(
1984
).
Neuronal development in the Drosophila retina: monoclonal antibodies as molecular probes
.
Cell
36
,
15
26
.