Spatial and temporal distribution of a Drosophila melanogaster embryonic variable nuclear antigen

Spatial and temporal distribution of a Drosophila melanogaster embryonic variable nuclear antigen

Biochimica et BiophysicaActa, 1049 (1990) 249-254 Elsevier 249 BBAEXP 92075 Spatial and temporal distribution of a Drosophila melanogaster embryoni...

1MB Sizes 0 Downloads 35 Views

Biochimica et BiophysicaActa, 1049 (1990) 249-254 Elsevier

249

BBAEXP 92075

Spatial and temporal distribution of a Drosophila melanogaster embryonic variable nuclear antigen N i c h o l a s J. G a y

Departmentof Biochemistry, Universityof Cambridge, Cambridge(U.K.) (Received 15 January 1990)

Key words: Nuclearantigen; Antigendistribution; Embryogenesis;Cyclin; String gene; (Drosophila) A polyclonal antibody has been prepared that specifically recognises a nuclear protein antigen in Drosophila embryos. During development, the antigen appears initially to be uniformly distributed but by nuclear division cycle 10 is seen to accumulate in nuclei in a manner suggesting that it is destroyed or becomes modified upon transition from S- to M phase of the nuclear division cycle. This conclusion is supported by the observed disappearance of the antigen from the postblastoderm nuclei in a manner that reflects the pattern of the first asynchronous postblastoderm cell division and persistence in the polyploid nuclei of the amnioserosa that do not undergo further cell or nuclear division. In Western blot experiments, the antibody detects specifically a 105 kDa nuclear protein that probably corresponds to the antigen detected in embryos by immunocytochemical means. Introduction Embryonic development of Drosophila melanogaster commences with 13 very rapid metachronous divisions of the zygotic nucleus, a process that depends largely upon protein products derived from maternal RNA. [1,2]. After completion of these nuclear divisions, plasma membranes enclose the syncitial nuclei to form the cellular blastoderm and a large increase in zygotic transcription is observed [3]. The morphogenetic cellular movements of gastrulation follow immediately and during this period the cells becomes committed to division in a precisely regulated asynchronous pattern which depends on zygotic transcription and reflects the newly specified indentities of the cells [1]. An understanding of how nuclear and cellular division is regulated during embryogenesis and how precise patterns of cells division are established in terminally differentiated cells is an important challenge of molecular biology and Drosophila embryos are a good model system for the study of this problem, due to the available genetic and a molecular techniques (for a review see Ref. 4). Such methods have been used to isolate three genes, cyclin A, cyclin B and string, the protein products of which are important for cell-cycle regulation during embryogenesis [5,7]. Cyclin A is thought to be a component of a conserved mitotic regulator, maturation promotion fac-

Correspondence: N.J. Gay, Department of Biochemistry,University of Cambridge, Tennis Court Road, Cambridge, CB2 1QW, U.K.

tor (MPF) that functions to trigger mitosis and meiosis [8]. In Drosphila, cyclin A is found uniformly distributed the syncitial embryo, but in cellular embryos it accumulates in interphase cytoplasm, localises to the nucleus at prophase and dissipates ~ metaphase [5]; while the transcript for cyclin B accumulates during early development in the pole cells [7]. Mutations at the string locus cause cell cycle arrest at the G2-mitosis transition, suggesting that zygotic transcription of the gene is required for initiation of the 14th mitosis. The product of the string gene is a 60 kDa protein related in sequence to cdc 25, a mitotic regulator of Schizosaccharomyces pornbe [6]. In this paper, I report the serendipitous detection using immunocytochemical methods of a nuclear antigen in Drosphila embryos that dissipates from syncitial and cellular nuclei upon the G2-mitosis transition. The antibody preparation detects a single protein species of 105 kDa in soluble nuclear protein extracts which probably corresponds to the antigen detected in situ in embryos. Thus, the antigen is unlikely to be Drosophila cyclin A or string protein, b,at may be another gene product under similar regulation during nuclear and cellular division. Materials and Methods

Antibody A rabbit was inoculated with 0.25 mg of fl-galactosidase in Freund's complete adjuvant, boosted at 4 weeks (incomplete Freund's adjuvant) and bled at 6

0167-4781/90/$03.50 © 1990 ElsevierScience Publishers B.V. (BiomedicalDivision)

250 weeks Serum was precipitated by the addition of 60% ammonium sulphate. The precipitate was collected by centrifugation redissolved in and dialysed against PBS (3 mM NaH2PO 4, 7 mM Na2HPO 4 and 0.15 M NaC1). Most of the anti-fl galactoside antibodies in the serum were removed by passing it over a fl-galactosidaseSepharose 4B affinity column once [9]. Anti-fl-galactosidase specificites purified by a second passage of the serum over the affinity column were used in this experiments described below.

Immunocytochernical staining of Drosophila embryos Embryos were collected from apple juice agar plates, dechorionated, devitellenised and fixed as described [10]. The embryos were rehydrated in PBS +0.1% Triton X-100 + 0.2% BSA (PBT), blocked with PBT + 5% normal goat serum (NGS) (2 h), incubated with antibody in the same buffer overnight (4°C) at a dilution of 1/30, washed (PBT, 5 × 10 rain), incubated with biotinylated goat anti-rabbit secondary antibody (ICN, 1/500 dilution) in PBT + NGS (2 h, 25 ° C) and washed in PT(PBS + 0.1% Tween 20) (5 × 5 min). The embryos were incubated in Vectastain ABC reagent (Vector Laboratories) (1/50 dilution in PT, 2 h), washed (5 × 5 min, PT) and developed with 2 5 0 / ~ g / m diaminobenzidine 0.06% H202 in FT. The embryos were washed 2 × 2 min, PBS) and stained with DAPI (4'6'-diamidino-2-phenylindole) at l ~ g / m l in PBS. The embryos were then washed (2 × 2 mins, PBS), dehydrated; cleared in xylene and mounted in De Pex (from BDH). Western blot analysis Western blot analysis was performed as described in Ref. 11 except that the secondary antibody (125I-labeled goat anti-rabbit IgG, whole antibody, Amersham International) was pre-incubated with a blot to which whole embryo extract had been bound. Preparation of embryo protein extracts For whole extracts, embryos were collected from apple juice agar plates, dechorionated in 50% bleach and homogenised into 25 mM Tris(pH 8.0), 150 mM NaC1, 0.2% SDS, 0.5% NP40, 0.5% sodium deoxycholate and 1.0 mM phenyl methyl sulphonyl fluoride. Insoluble material was removed by centrifugation. Saltextracted nuclear proteio of embryos were prepared as described [12]. Results

Immunocytochemical detection of a specific nuclear antigen in early Drosophila embryos. A rabbit polyclonal antibody purified as described above was used to stain Drosophila embryos and was found fortuitously to detect a specific nuclear antigen (Fig. 1).

In early cleavage-stage embryos, this antigen is distributed throughout the embryo (Fig. la) and no compartmentalisation is evident. However, at nuclear division cycle 11 or 12, the syncitial blastoderm stage, the antigen becomes specifically localised in the syncitial nuclei (Fig. lb). It is difficult to be sure at precisely which nuclear division cycle the antigen first localises to nuclei, but it seems to be absent from nuclei at cycle 9-10. This uniform pattern of nuclear localisation persists through to nuclear cycle 14, the cellular blastoderm stage (Fig. lc) but when the morphogenetic movements of gastrulation commence, the pattern differentiates. Lines of strongly staining cells are frequently observed along the lips of the ventral and cephalic furrows (Fig.1 d and e) and strong staining is seen in the polyploid nuclei of the cells which form the dorsal most structures of embryo, the anmioserosa, After gastrulation, the overall pattern of nuclear staining decays but the strong staining in the amnioserosal cells persists in both the germ-band-extended (Fig. lf) and the germ-band shortened embryo (Fig. lg). Control experiments in which the secondary antibody alone and different affinity purified anti /3galactosidase antibodies were used, did not reveal such staining patterns.

The nuclear antigen has cyclin like properties Although many of the syncitial blastoderm embryos observed have a uniform nuclear distribution of the antigen, some have patches of stained nuclei emanating from the anterior and posterior embryonic poles (Fig. 2c) whilst others have patches of stained nuclei located in the equator of the embryo (Fig. 2a). These patterns can be seen most clearly in DAPI fluorescence quenching (Fig. 2b and d). In this technique, the embryos are stained both histochemically and with the DNA-specific fluorescent dye DAPI. In nuclei that contain the histochemical stain, the fluorescence of DAPI is quenched and thus unstained nuclei appear bright in comparison to stained ones. The polar pattern of nuclear staining is reminiscent of the mitotic waves observed in the metachronous 10th-13 th nuclear division cycles [2]. However, mitotic waves invariably originate at the embryonic poles and cannot per se explain the existence of equatorial nuclear staining. Furthermore, in some embryos in which all the nuclei are in anaphase, no nuclear staining can be detected (Fig. 2g, h). The only explanation of these patterns in that the antigen is present in the nuclei during S phase of the nuclear division cycle but is destroyed, or otherwise dissipates from the nuclei, upon S-mitosis transition. Examination of embryos that have completed the 13th nuclear division reveals some with nearly uniform staining (Fig. lc) and others with irregular patterns (Fig. 2e). This may reflect either some degree of asynchrony in the completion of the 13th mitosis or

251

(b)

Fig. 1. Spatiotemporal about nuclear division

distribution of the antigen during embryogenesis of Drosophila melanogaster. (a) Cleavage stage embryo. (b) Embryo at 12. (c) Cellularising blastoderm embryo. (d) Gastrulating embryo ventral view. (e) Gastrulating embryo, dorsal view. (f) Germ band extended embryo. as = amnioserosa (g) Germ and shortened embryo.

variations within the embryo in resynthesis or nuclear reaccumulation of the antigen after completion of the 13th nuclear division cycle. The differentiated nuclear staining patterns observed in gastrulating embryos correlate well with the known temporal pattern of the first asynchronous post-blastoderm mitotic divisions [ll]. Thus, in Fig. 2f, the antigen is seen to be absent from two dorso-ventrally symmetric groups of cells that correspond to mitotic activity centres Ecla and Eclb. In addition, in Fig. Id, there are strongly

staining cells on the ventral furrow (the presumptive ventral neurogenic region) while dorsally the nuclei are unstained. This corresponds to an invasive wave of mitotic activity proceeding from a dorsal anterior-posterior belt that corresponds to the dorsal epidermal anlage. The persistence of the antigen in the nuclei of the amnioserosa is to be expected as these cells do not undergo any postblastoderm mitotic divisions. Thus the antigen appears to have properties during nuclear division cycles 12-14 analogous to those of a group of

Fig. 2. Accumulation and dissipation nuclear antigen during the mitotic cycle. (a), (c) and (g) Bright field images of histochemically images. (a) and (b) Syncitial blastodem embryo with patch of equatorial embroys. (b), (d), (e), (f) and (h) are DAPI UV fluorescence staining. (c) and (d) Syncitial blastoderm embryo with polar nuclear staining. (d) Cellularising blastoderm embryo with irregular pattern of fluorescence. (f) Gastrulating embryo with nuclear staining lost from the 1st and 2nd groups of asynchronous post blastoderm mitotic cells, 1 and 2. (g) and (h) Blastoderm embryo in which all nuclei are at anaphase.

embryonic proteins called cyclins. Cyclins are produced from maternal mRNA in the cleavage-stage embryos of clams, frogs and Drosophila and their levels rise during S phase, but drop rapidly at the G2-mitosis transition.

stained nuclear nuclear marked

A possible alternative explanation for these observations is that the epitope(s) recognised by the antibody become masked, for example, by phosphorylation during mitosis.

253

The antibody detects a single nuclear protein species Western blot analysis using this antibody preparation detects a single protein species, with an estimated molecular mass of 105 kDa, in both detergent-soluble extracts of whole embryos and highly purified nuclear protein (Fig. 3a and b). Given the specificity displayed by the antibody and the purification of the protein in soluble nuclear extracts, it seems likely that this protein species corresponds to the antigen detected in the immunocytochemical staining experiments. Furthermore, it is unlikely that this protein corresponds to Drosophila cyclin A or string protein, both of which are about 60 kDa.

Discussion

The nuclear antigen described in this paper accumulates in the nuclei of Drosophila embryos during G1 and S phases of the cell or nuclear division cycle but is destroyed or becomes modified at the G2-M transition. The conclusion that the antigen is destroyed rather than dissipating from nuclei in some other manner is based on the observation that it is not seen to redistribute cytoplasmically during mitosis. By contrast, other nuclear proteins (e.g., bicoid [13] and engrailed [14]) are clearly seen to redistribute within the cytoplasm in both the syncitial and cellular embryo. Thus, the antigen has properties analogous to a group of embryonic proteins called cyclins, and its seems likely that its function or the modification of its function is important for regulation of nuclear and cellular division in Drosophila embryos. The antibody preparation used detects a single protein species of 105 kDa in embryonic nuclear extracts which probably corresponds to the antigen detected in embryos by immunocytochemical means and this finding indicates that the protein is distinct from either Drosophila cyclin A or string, two proteins known from genetic and biochemical evidence to be important for cell division cycle control. Using this antibody preparation, it should be possible to isolate a molecular clone corresponding to this antigen by immunological probing of a library of embryonic cDNA cloned in the expression vector h g t l l [15]. The availability of such a c D N A clone would enable the primary structure of the protein to be determined, information that might well enable an identification to be made on the basis of amino-acid sequence homology with other proteins of known biochemical function and would allow the preparation of further antibodies which could be used to determine whether the protein is destroyed in a similar fashion to cyclins or becomes modified in response to S or G2-M transition. Acknowledgements

N.J.G. is Senior Research Fellow in Biochemistry, Christ's College, Cambridge. This work is supported by a project grant from the Wellcome Trust. ......

~

4S

Fig. 3. Western blot analysis of soluble embryo protein extracts (a) Soluble protein extract of whole embryo (200/~g) ( 0 - 4 h). (b) Soluble nuclear protein extract (75 #g) (2-12 h). The samples were separated by SDS-polyacrylamide gel electrophoresis (7%) and the blot was autoradiographed at - 70 ° C with a screen for 6 h. A 1 : 30 dilution of the antibody [16] preparation was used. The positions to which molecular mass markers migrated is indicated.

References 1 Campos-Ortega, J.A. and Hartenstein, V. (1985) in The Embryonic Development of Drosophila melanogaster, Springer, New York. 2 Foe, V.E. and Alberts, B. (1983) J. Cell Sci., 61, 31-70. 3 Edgar, B.A. and Schubiger, G. (1986) Cell 44, 871-877. 4 Glover, D.M. (1989) J. Cell Sci. 92, 137-146. 5 Lehner C.F. and O'Farrell, P.H. (1989) Cell 56, 957-968. 6 Edgar, B.A. and O'Farrell, P.H. (1989) Cell 57, 177-197. 7 Whitefield, W.G.F., Gonzales, C., Sanchez-Herrero, E. and GIover, D.M. (1989) Nature 338, 337-340.

254 8 9 10 11

Minshull, J., Blow, J.J. and Hunt, T. (1989) Cell 56 (947-956. Axen, R., Porath, J. and Ernback, S. (1967) Nature 214, 1302-1304. Karr, T.L. and Alberts, B.M. (1986) J. Cell Biol. 102, 1494-1504. Towbin, G., Staehlin, T. and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76, 4350-4354. 12 Soeller, W., Poole, S. and Kornberg, T. (1988) Genes Devel. 2, 68-81.

13 Driever, W. and Nusslein-Volhard, C. (1988) Cell 54, 83-93. 14 Di Nardo, S. et al. (1985) Cell 53, 59-69. 15 Patel, N.H., Snow, P.M. and Goodman, C.S. (1987) Cell 48, 975-988. 16 Laemlli, U.K. (1970) Nature 227, 680-685.