The sequences of two Drosophilu and one rabbit protein phosphatase (PP) 1 catalytic subunits were determined from their cDNA. The sequence of Drosophilu PP1al was deduced from a 2.2-kb cDNA purified from an embryonic cDNA library, while that for Drosophilu PPlP was obtained from overlapping clones isolated from both a head cDNA library and an eye imaginal disc cDNA library. The gene for Drosophilu P P l q is at 96A2-5 on chromosome 3 and encodes a protein of 327 amino acids with a calculated molecular mass of 37.3 kDa. The gene for Drosophila PPlp is localized at 9C1-2 on the X chromosome and encodes a protein of 330 amino acids with a predicted molecular mass of 37.8 kDa. PPlal shows 96% amino acid sequence identity to PP1a2 (302 amino acids), an isoform whose gene is located in the 87B6-12 region of chromsome 3 [Dombradi, V., Axton and PPlg (327 amino acids), respectively, demonstrating that the structures of both isoforms are among the most conserved proteins known throughout the evolution of the animal kingdom. The presence of characteristic structural differences between PPla and PPl P, which have been preserved from insects to mammals, implies that the a and fi isoforms may have distinct biological functions.Protein phosphatase (PP) 1 is one of the four major types of protein serinelthreonine phosphatase that have been identified in eukaryotic cells [l]. Like PP2A and PP2C, PPl is able to dephosphorylate a number of phosphoproteins in vitro, raising the important question of which protein phosphatase dephosphorylates which substrates in vivo. As one approach to address this question, we have embarked on the cloning of each of these protein phosphatases in the fruit fly Drosophilu melunoguster as the first step towards obtaining mutants defective in each of these enzymes.Recently, we isolated a cDNA encoding a protein phosphatase whose predicted amino acid sequence showed 92% identity with that of PPla from rabbit skeletal muscle over the 302 homologous amino acids [2]. This cDNA hybridised strongly to the 87B6-12 region on chromosome 3, and more weakly to three other loci, namely 96A2-5 on chromosome 3, 9C1-2 on chromosome X and 13C1-2 on chromosome X. Note. Several years ago we isolated a cDNA from a rabbit skeletal muscle library encoding a protein identical to protein phosphatase (PP) l a in nucleotide and amino acid sequence, except at the extreme N-terminus of the coding region and in the 5' untranslated region. This cDNA was initially termed P P l j [8], but as discussed in [20] we now believe it is likely to have been a cloning artefact, since the same 5' region has subsequently been found fused to a quite different gene. Consequently it is no longer referred to as PPlD. These findings suggested that several genes encoding different isoforms of PP1 might be present in Drosophilu. In support of this contention, a cDNA was isolated that comprised part of the coding region of a further form of PP1 [2, 31. Here, we have completed the amino acid sequence of this isoform, which localises to 9C1-2, and d...

SummaryMany aspects of the mitotic cycle can take place independently in syncytial Drosophila embryos. Embryos from females homozygous for the mutation gnu undergo rounds of DNA synthesis without nuclear division to produce giant nuclei, and at the same time show many cycles of centrosome replication (Freeman et al. 1986). S phase can be inhibited in wild-type Drosophila embryos by injecting aphidicolin, in which case not only do centrosomes replicate, but chromosomes continue to condense and decondense, the nuclear envelope undergoes cycles of breakdown and reformation, and cycles of budding activity continue at the cortex of the embryo (Raff and Glover, 1988). If aphidicolin is injected when nuclei are in the interior of the embryo, centrosomes dissociate from the nuclei and can migrate to the cortex. Pole cells without nuclei then form around those centrosomes that reach the posterior pole (Raff and Glover, 1989); the centrosomes presumably must interact with polar granules, the maternally-provided determinants for pole cell formation. T he pole cells form the germ-line of the developing organism, and as such may have specific requirements for mitotic cell division. This is suggested by our finding that a specific class of cyclin mRNAs, the products of the cyclin B gene, accumulate in pole cells during embryogenesis (Whitfield et al. 1989). Other genes that are essential for mitosis in early embryogenesis and in later development are discussed.

Unfertilized eggs and fertilized embryos from Drosophila mothers mutant for the plutonium (plu) gene contain giant polyploid nuclei resulting from unregulated S-phase. The PLU protein, a 19-kDa ankyrin repeat protein, is present in oocytes and early embryos but is not detectable after the completion of the initial rapid S-M cycles of the embryo. The persistence of the protein during the early embryonic divisions is consistent with a direct role in linking S-phase and M-phase. When ectopically expressed in the eye disc, PLU did not perturb the cell cycle, suggesting that PLU regulates S-phase only in early embryonic development. The pan gu (png) and giant nuclei (gnu) genes also affect the S-phase in the unfertilized egg and early embryo. We show that functional png is needed for the presence of PLU protein. By analyzing png mutations of differing severity, we find that the extent of the png mutant phenotype inversely reflects the level of PLU protein.Our data suggest that PLU protein is required at the time of egg activation and the completion of meiosis.

The plutonium (plu) gene product controls DNA replication early in Drosophila development. plu mutant females lay unfertilized eggs that have undergone extensive DNA synthesis. In fertilized embryos from plu mutant mothers, S‐phase is uncoupled from mitosis. The gene is expressed only in ovaries and embryos, null alleles are strict maternal effect mutations, and the phenotype of inappropriate DNA replication is the consequence of loss‐of‐gene function. plu therefore negatively regulates S‐phase at a time in early development when commitment to S‐phase does not depend on cyclic transcription. plu encodes a protein with two ankyrin‐like repeats, a domain for protein‐protein interaction. plu is immediately adjacent to, but distinct from, the PCNA gene.

A 1.0 kilobase cDNA coding for the complete amino acid sequence of a putative protein phosphatase (314 amino acid residues, molecular mass 36 kDa) has been isolated from a Drosophila head cDNA library. The cDNA hybridises to a single site on the right arm of the second chromosome at cytological position 55Al-3. The deduced sequence of the protein, designated protein phosphatase-Y, is homologous to the catalytic subunits of Drosophila and rabbit protein phosphatase-la: (64 and 59% identity, respectively) and rabbit protein phosphatase-2A (39% identity). These and other comparisons demonstrate that this novel enzyme is not the Drosophila counterpart of mammalian protein phosphatases 1,2A, 2B, 2C or X.Protein phosphatase; cDNA cloning; Nucleotide sequence; Amino acid sequence; (Drosophila melanogaster)

The mutagen-sensitive-101 (mus101) gene of Drosophila melanogaster was first identified 25 years ago through mutations conferring larval hypersensitivity to DNA-damaging agents. Other alleles of mus101 causing different phenotypes were later isolated: a female sterile allele results in a defect in a tissue-specific form of DNA synthesis (chorion gene amplification) and lethal alleles cause mitotic chromosome instability that can be observed genetically and cytologically. The latter phenotype presents as a striking failure of mitotic chromosomes of larval neuroblasts to undergo condensation of pericentric heterochromatic regions, as we show for a newly described mutant carrying lethal allele mus101lcd. To gain further insight into the function of the Mus101 protein we have molecularly cloned the gene using a positional cloning strategy. We report here that mus101 encodes a member of the BRCT (BRCA1 C terminus) domain superfamily of proteins implicated in DNA repair and cell cycle checkpoint control. Mus101, which contains seven BRCT domains distributed throughout its length, is most similar to human TopBP1, a protein identified through its in vitro association with DNA topoisomerase IIβ. Mus101 also shares sequence similarity with the fission yeast Rad4/Cut5 protein required for repair, replication, and checkpoint control, suggesting that the two proteins may be functional homologs.

We have cloned the three rows (thr) gene, by a combination of chromosome microdissection and P element tagging. We describe phenotypes of embryos homozygous for mutations at the thr locus. Maternal mRNA and protein appear to be sufficient to allow 14 rounds of mitosis in embryos homozygous for thr mutations. However, a small percentage of cells in syncytial blastoderm stage thr embryos sink into the interior of the embryo as if they have failed to divide properly. Following cellularisation all cells complete mitosis 14 normally. All cells become delayed at mitosis 15 with their chromosomes remaining aligned on the spindle in a metaphase-like configuration, even though both cyclins A and B have both been degraded. As cyclin B degradation occurs at the metaphase-anaphase transition, subsequent to the microtubule integrity checkpoint, the delay induced by mutations at the thr locus defines a later point in mitotic progression. Chromosomes in the cells of thr embryos do not undertake anaphase separation, but remain at the metaphase plate. Subsequently they decondense. A subset of nuclei go on to replicate their DNA but there is no further mitotic division.