Distribution of PCNA in Drosophila embryo during nuclear division cycles

The proliferating cell nuclear antigen (PCNA) is a nuclear protein of about M_r 29 000, which functions as a co-factor of DNA polymerase <5 (reviewed by Fairman, 1990). The PCNA is necessary for the synthesis of the leading and lagging strands of simian virus 40 DNA (Tsurimoto et al. 1990) and has functional similarity with the protein coded by the gene 45 of T4 phage (Tsurimoto and Stillman, 1990). The PCNA is also necessary for cell cycle progression and cellular proliferation (Jaskulski et al. 1988; Liu et al. 1989). The cDNAs and genes for PCNA from mammals (Almendral et al. 1987; Matsumoto et al. 1987; Travali et al. 1989; Yamaguchi et al. 1991), Drosophila (Yamaguchi eí al. 1990), plants (Suzuka et al. 1991) and yeast (Bauer and Burgers, 1990) have been cloned and completely sequenced. Expression of the human PCNA gene in relation to cell cycle progression is under the control of complex regulatory mechanisms (reviewed by Baserga, 1991).

During embryogenesis in Drosophila, the first 13 nuclear divisions occur rapidly in a syncytial cytoplasm (Zalokar and Erk, 1976; Foe and Alberts, 1983). The first 7 rounds take place within the interior of the embryo, then the majority of nuclei migrate to the cortex during cycles 8 and 9, leaving behind a small number of yolk nuclei. The nuclei at the posterior pole cellularize by the end of cycle 10. These are the germ cell progenitors called pole cells. The remainder of the surface nuclei divide four more times, and then are cellularized during interphase of cycle 14. Each of the 13 nuclear division cycles consists of S phase and M phase without any intervening gap phases such as G ₁ or G₂. The 13 cycles also lack feedback regulation to monitor the completion of S phase (reviewed by Glover, 1991). In cycle 14, G₂ phase initially appears and the timing of cells to enter mitosis comes under the control of string, a cdc25 homologue (O’Farrell et al. 1989), whose transcription is activated within the so-called mitotic domains (Foe, 1989).

Although it is known that the PCNA mRNA and its protein are maternally stored and their levels are highest in 0-3 h embryos (Ng et al. 1990; Yamaguchi et al. 1990), spatial distribution of the PCNA in the embryo and its dynamics during development have not yet been studied. The present paper describes the results of immunocytochemical studies on the spatial distribution of PCNÁ in the Drosophila melanogaster embryo during the first 13 nuclear division cycles.

A full length Drosophila PCNA cDNA (1.0 kb) was isolated from the plasmid pUC19DcPCNA01 (Yamaguchi et al. 1990) by digestion with EcoRI and inserted into the unique AcoRI site of the pTDT7 (Date et al. 1990), a plasmid/phage chimeric vector containing the T7 phage late gene promoter. After isolating a single-stranded recombinant DNA, loop-out mutagenesis (Date et al. 1990) was carried out using a synthetic oligonucleotide 5′-GTGCCTCGAACATAGCTGTTTCCT-3′. The 5′ portion from the T residue (the 13th nucleotide) contains the complementary sequence to the cDNA and the 3’ portion from the A residue (the 14th nucleotide) contains the complementary sequence to the vector. The plasmid pTDT7dPCNA obtained contains a Shine-Dalgarno sequence at the appropriate distance from the translation initiation codon of the inserted cDNA and can encode the whole Drosophila PCNA. E. coli strain BL21 carrying a single copy of the gene for T7 RNA polymerase in the chromosome under lacUV5 promoter (Studier and Moffat, 1986) was used as a host for the expression plasmid.

Embryos of Drosophila melanogaster (Canton S) 0-3 h old were collected, dechorionated using 10% sodium hypochlorite and stored at -80°C. Frozen embryos were thawed and homogenized in a buffer containing 50 mM Tris-HCl, pH 8.0, 0.5 mM magnesium acetate, 0.05 mM EDTA, 5 mM KC1, 0.35 M sucrose, 1 mM dithiothreitol and ImM phenylmethylsulfonyl fluoride. The homogenate was centrifuged at 100 000 g for 1 h at 4°C. Bacterial lysate was prepared as described previously (Date et al. 1988).

Protein solutions were appropriately diluted in a solution containing 0.0625 M TTíB-HCI, pH 6.8, 2% sodium dodecyl sulfate (SDS), 10% glycerol, 20 mM dithiothreitol and bromophenol blue. Proteins were denatured by heating for 3 min at 100 °C. The polypeptides were fractionated by electrophoresis in a 10% polyacrylamide gel containing 1% SDS and were stained with Coomassie brilliant blue or were transferred to a nitrocellulose membrane using a Bio-Rad semi-dry transfer apparatus under conditions recommended by the supplier.

The blotted membrane was blocked with TBS solution (50 mM Tris-HCl, pH 8.3, and 150 mM NaCl) containing 20% goat serum for 30 min at room temperature and then incubated with either a rabbit anti-Drosophila PCNA (preabsorbed with E. coli acetone powder and used at a dilution of 1:1000) (Ng et al. 1990), or a mouse anti-rabbit PCNA (used at a dilution of 1:500) (Boehringer) for 16h at 4 °C. After extensive washing with TBS, the membrane was incubated with either an alkaline phosphatase-conjugated goat anti-rabbit IgG (preabsorbed with E. coli acetone powder and fixed Drosophila embryos, and used at a dilution of 1:4000) (Promega Biotech) or an alkaline phosphatase-conjugated goat anti-mouse IgG (preabsorbed with fixed Drosophila embryos and used at a dilution of 1:4000) (Promega Biotech) for 2h at room temperature. After extensive washing with TBS containing 0.3% Triton X-100, color was developed in a solution containing 100 mM Tris-HCl, pH9.5, lOOmM NaCl, 5mM MgCl₂, 0.34mgml⁻¹ nitroblue tétrazolium salt, and 0.175 mg ml ¹ 5-bromo-4-chloro-3-indolyl phosphate toluidinium salt.

Canton S embryos 0-3 h old were collected, dechorionated, fixed and devitellinized as described by Ashbumer (1989). Embryos were blocked with TBS containing 10% goat serum and 0.15% Triton X-100 (TBT). Incubation with the primary antibody was carried out in the same solution for 16 h at 4 °C. The embryos were washed extensively in TBS containing 0.3% Triton X-100, reblocked with TBT and incubated with the preabsorbed alkaline phosphatase-conjugated secondary antibody for 16h at 4°C. After extensive washing with TBS containing 0.3% Triton X-100, color was developed as described above. Finally, embryos were stained with DAPI at a final concentration of 1 μg ml⁻¹ for 5 min, and then extensively washed with TBT. Stained embryos were mounted in a solution containing glycerol. A monoclonal mouse anti-rabbit PCNA IgG and an alkaline phosphatase-conjugated goat antimouse IgG were mainly used as primary and secondary antibodies, respectively.

A polyclonal rabbit anti-Drosophila PCNA IgG in combination with an alkaline phosphatase-corjugated goat anti-rabbit IgG showed essentially the same staining patterns.

Specificity of the polyclonal anti-Drosophila PCNA (kindly given by P. A. Fisher) and the monoclonal anti-rabbit PCNA was examined by Western immunoblotting. The anti-Drosophila PCNA reacted with a protein which had an apparent molecular weight of 36 000 in a crude extract of 0-3 h Drosophila embryos (Fig. 1, lane f). It is known that the Drosophila PCNA consists of a single polypeptide of 260 amino acids with a molecular mass of 28 830 (Yamaguchi et al. 1990), but it migrated at a position corresponding to M_r 36000 in SDS-polyacrylamide gel electrophoresis (Ng et al. 1990). The antibody also reacted with the M_r 36 000 protein in a whole lysate of isopropyl /3-thiogalactoside (IPTG)-induced E. coli BL21 carrying the pTDT7dPCNA, an expression plasmid for Drosophila PCNA (Fig. 1, lane e). This band was not detected in the lysate of the control BL21 that did not carry the expression plasmid (Fig. 1. lane d). The cross-reactive M_r 50000 polypeptide in crude embryo extracts that has been reported with this antibody (Ng et al. 1990) was not detected under our experimental conditions.

Monoclonal anti-rabbit PCNA also reacted with the M_r 36 000 polypeptide exclusively in the crude embryo extract (Fig. 1, lane i). About 70% of amino acid residues of Drosophila PCNA are identical to those of mammalian homologues (Yamaguchi et al. 1990). Possibly, the antibody reacts with a conserved region of the PCNA protein. The results indicate that both anti-Drosophila PCNA and anti-rabbit PCNA antibodies react specifically with the Drosophila PCNA protein.

An immunocytochemical method was employed to detect PCNA in embryos during the first 13 nuclear division cycles. Although data with the monoclonal anti-rabbit PCNA antibody are presented, the polyclonal antiDrosophila PCNA antibody gave essentially the same results with slightly higher background staining (data not shown).

Staining results in embryos at interphase of various nuclear division cycles are shown in Fig. 2. Embryos in early nuclear cycles (until cycle 5) were viewed at an internal focal plane (Fig. 2, panels A to C) and those in later cycles (from cycle 7) were viewed at a surface focal plane (Fig. 2, panels D to K). Strong nuclear staining was observed throughout 13 cycles. Staining is more uniform in nuclei at early cycles (until cycle 10) than in those at late cycles (Fig. 3). Furthermore, strong staining of the cytoplasm of unfertilized eggs (Fig. 2, panel A) confirmed the previous study by a Western blotting analysis that indicated maternal storage of the PCNA protein (Ng et al. 1990).

A high level of synchrony in mitotic stages was observed in some embryos (Fig. 2 and Fig. 4). However, the frequently observed asynchrony in an individual embryo allowed us to determine the precise timing of disappearance and reappearance of the antibody-staining material in chromosomes/nuclei in each nuclear division cycle. For instance, one of the cycle 10 embryos had metaphase chromosomes in the center of the embryo, anaphase chromosomes in the anterior region, and telophase nuclei near the anterior end, as judged from the shapes of 4,6-diamidino-2-phenylindole (DAPI)-stained materials and their positions in the embryo (Fig. 5, panel C). The appearance of weak DAPI-staining in interphase nuclei, except at its peripheral region, is due to quenching by immunochemical staining. No antibody-staining was observed in metaphase chromosomes (Fig. 5, panels A and B). The antibody-staining of chromosomes reappeared at anaphase as seen in Fig. 5, panel B. The staining of anaphase chromosomes observed with other cycle 10 embryo is shown in Fig. 4, panel H. In another case, one of the cycle 11 embryos (Fig. 6, panels C and E) had interphase nuclei in the center of the embryo, prophase nuclei in the anterior and posterior regions and metaphase chromosomes near the anterior and posterior ends. The staining distinctly disappeared in the area containing metaphase chromosomes, while it was detected in the adjacent area containing prophase nuclei (Fig. 6, panels B and D).

The results from a number of immunostained embryos can be summarized as followings. Metaphase chromosomes were not stained by anti-PCNA antibody throughout 13 cycles. Antibody staining of chromosomes reappeared at anaphase in cycles earlier than cycle 10 (Fig. 4, panels A to H). In later nuclear division cycles, the timing of reappearance of the staining was delayed progressively. In nuclear division cycle 11, anaphase chromosomes were unstained in some embryos, weakly stained in others (Fig. 4, panel I and K). Strong staining was detected in nuclei at telophase and interphase stages of all cycle 11 embryos (Fig. 2, panel G). In cycle 12, antibody-staining in anaphase chromosomes was hardly detectable (Fig. 7, panels A and B) and even that in telophase nuclei was very faint (Fig. 7, panels A, B, D, E and G). Strong nuclear staining reappeared probably at the beginning of the next S phase (Fig. 7, panels D, E and G). These results indicate that PCNA dissociates from chromosomes at the beginning of metaphase and reassociates with them during later mitotic phases. The timing of reassociation delays progressively in later nuclear division cycles.

The anti-PCNA antibody also stained the cytoplasm of embryos throughout the 13 cycles. Images at internal focal planes are shown in Fig. 8. Staining in the cytoplasm was relatively uniform throughout all mitotic phases until nuclear division cycle 9 (Fig. 8, panel A to C). In cycle 10, cytoplasmic staining was relatively uniform at both interphase (Fig. 8, panel D) and metaphase (Fig. 8, panel E). However, at anaphase and telophase, strong staining in both the central yolk mass and the layer of cortical cytoplasm underneath nuclei at the surface of the embryo was observed (Fig. 8, panel F). In later cycles, strong staining in the central yolk mass and the cortical cytoplasm was detected during metaphase and telophase (Fig. 8, panels H, I, K, L, and N; a high-magnification photograph of an embryo at cycle 11 metaphase is also shown in Fig. 9, panel B). During interphase of these cycles, staining in the yolk mass became weak and that in the cortical cytoplasm underneath surface nuclei completely faded away (Fig. 8, panels G, J, M, and O; a high-magnification photograph of an embryo at cycle 11 interphase is also shown in Fig. 9, panel A). These results suggest that the PCNA synthesized in the syncytial cytoplasm during metaphase and telophase stages is transported into the surface nuclei at the time preceding the onset of S phase.

In cultured mammalian cells, PCNA was detected as a punctate pattern of immunofluorescence in the S phase nucleus, which was similar to the pattern obtained with the antibody to bromodeoxyuridine used for pulse-labeling of nascent DNA (Bravo and Macdonald-Bravo, 1985, 1987; Celis and Celis, 1985). Although this pattern of PCNA distribution varied depending on methods of fixation, it was interpreted as representing a localization of the PCNA at the sites of ongoing DNA replication (Bravo and MacDonald-Bravo, 1987).

By immunostaining analysis with the specific anti-PCNA antibody, we demonstrated that staining patterns in interphase nuclei of Drosophila embryos were rather uniform and similar to the fluorescence patterns obtained with DAPI, especially during early nuclear division cycles. It has been estimated that in order to complete DNA replication within an S phase of 3 to 4 min, 40000 to 100 000 synchronous initiations must occur in each nucleus, based on the fork movement rate of 2.6 kb min⁻¹ fork⁻¹ (Blumenthal et al. 1973). Thus it may be unlikely that individual sites of replication in the relatively small nucleus of Drosophila embryo can be distinguished with a light microscope. It is noted that nuclear division cycles slow progressively during cycles 10 (7.6 min), 11 (9.7 min), 12 (12.6 min) and 13 (16 min) (Foe, 1989). Although it is not known whether all mitotic phases are equally retarded or perhaps only the interphase is retarded, slight deviation from uniformity in the staining pattern during late nuclear division cycles (Fig. 3, panel C) might represent the decrease of initiations or clustering of replicon domains that could result in lengthening of S phase.

After dissociating from chromosomes at metaphase, the reassociation of PCNA with them is delayed in later nuclear cycles (cycles 10 to 13). Control mechanisms for dissociation and reassociation of the PCNA with chromosomes (nuclei) are not known yet. The fibrous elements that are supposed to mediate the rapid saltatory movement of particles between the yolk and the exterior of the embryo (Foe and Alberts, 1983) might also function to transport the PCNA synthesized in syncytial cytoplasm into nuclei in cycles 10-13. This saltatory movement of particles can first be detected during cycle 10, and increases progressively during the interphases of cycles 10-13, slowing slightly during mitosis 10-12 (Foe and Alberts, 1983).

In nuclear cycles 11-13, strong staining in both the central yolk mass and the cortical cytoplasm was observed during metaphase and telophase. Staining in these regions disappeared during interphase of these cycles. In the Xenopus system, it is reported that completion of DNA replication inactivates the DNA replication complex (Hutchison and Kill, 1989). If this is also the case with Drosophila embryo, degradation of the DNA replication complex containing the PCNA might occur after the completion of each round of DNA replication and the PCNA protein might be newly synthesized in the cytoplasm to prepare for the next S phase. The PCNA mRNA that is maternally stored in a relatively large amount (Yamaguchi et al. 1990) is likely to be translated to protein during these mitotic phases. The fact that the injection of cycloheximide arrests progression of the next S phase in Drosophila syncytial blastoderm embryos (Zalokar and Erk, 1976) is also suggestive. Alternatively, it may also be possible that the PCNA protein simply becomes thoroughly diffused at metaphase without degradation and reassociates with the chromosomes at the following mitotic phases.

Cyclic changes in distribution patterns during nuclear division cycles are reported with a number of proteins (Dequin et al. 1984; Frasch et al. 1986; Lin and Wolfner, 1991). Some proteins distribute in the cytoplasm during mitosis, similar to the behavior of PCNA, although the exact timing of dissociation and reassociation with the chromosome is different among proteins. Other proteins migrate onto the chromosomes after metaphase. Of course, there are a number of proteins that continuously associate with the chromosome, the spindle region or the centrosome throughout the division cycle. Periodic changes in distribution patterns of PCNA presented in this study, and those reported with other proteins, may reflect a control mechanism for the very rapid nuclear division cycles in Drosophila early embryos, where putative mitotic triggers such as the two cdc 2-like genes, the two cyclin genes, and the string gene are likely to be not yet rate-limiting (Glover, 1991).

Finally, the pTDT7dPCNA plasmid constructed in this study would be useful for future studies on structurefunction relationships of the PCNA protein, since an efficient site-directed loop-out mutagenesis method has been established with the pTDT7-based plasmids (Date et al. 1990).

We thank Y. Nishida for many helpful discussions and P. A. Fisher for the gift of the anti-Drosophila PCNA antibody. This work was supported in part by Grants-in-Aid from the Ministry of Education, Science, and Culture, Japan.