Introduction

Following the pioneer work of Lokha and Masui1, 2, studies using cell-free system have greatly enriched our knowledge about the events during cell cycle. Extracts derived from activated Xenopus laevis eggs able to asemble nuclei around exogeneous naked DNA, isolated chromosomes and demembranated sperm chromatin have been widely used to study the pathways and biochemistry of nuclear envelope and pore complexes assembly, lamina formation, DNA replication and nuclear cytoplasmic transportation3, 4, 5, 6. Meanwhile, extracts prepared from mataphase eggs were adopted to study spindle organization and nuclear disassembly7, 8, 9, 10. Plenty of information on cell cycle regulation concerning nuclear reassembly and mitosis has also been obtained from the studies using cell-free system9, 10. Since 1989, colleagues in our laboratory have devoted to this field and achieved a number of results11, 12, 13, 14. Recently we have successfully got the nuclei assembled in vitro around demembranated sperm chromatin in activated Xenopus egg extracts. The nuclei assembled in our system were morphologically similar to those of intact cells.

The nuclear lamina, a fibrous network located between nuclear envelope and chromatin, is composed of lamins. It has long been proposed that lamin proteins play important roles in nuclear assembly, chromatin topology and DNA replication and the interactions among membrane, lamins and chromatin bring about these processes15, 16, 17, 18. Molecular and biochemical anaylyses have revealed that the lamin proteins are highly conserved and developmentally expressed in a tissue-specific manner. Based on comparison of their primary structure, biochemical features and expression pattern, two major lamin isotypes can be distinguished, A-type lamins (mammalian laminA and C, chicken and Xenopus lamin A) and B-type lamins which seem to be constitutively expressed and can be further classified into B1 (mammalian and chicken lamin B1 and Xenopus lamin LI), B2 (mammalian and chicken B2 and Xenopus lamin LII) and B3 Xenopus lamin LIII) type lamins. When the nuclear envelope breaks down during mitotic or meiotic cell division, lamin filaments are hyperphosphorylated by the protein kinase P34cdc2 and depolymerized into oligomers. While A-type lamins are dispersed throughout the cytoplasm as soluble oligomers being not in contact with membranes, B-type lamins, though depolymerized, remain associated with the reminants of the nuclear envelope19, 20, 21, 22.

Lamin LIII is reported to be the single lamin isoform in Xenopus oocytes, eggs and early embryos and is present in embryos up to the tail bud stages, reappearing only in certain adult cells (neurons, muscle cells, and diplotene oocytes)21. Though classified as B-type lamin it is soluble during nuclear envelope break-down19, 20, 21. Lamin LI and LII appear at characteristic stages in development. Lamin LI first appears at midblastula transition and LII at the gastrula 21. Considering the importance of lamin proteins in normal cells, we were prompted to examine whether the nuclei assemnled in vitro had lamina structure and what the lamin components were. Our results showed that the reassembled nuclei did have lamina and, interestingly, lamin LII was found to be present in egg extracts and in nuclei assembled in vitro.

Materials and methods

Preparation of sperm nuclei

Sperm nuclei were prepared as described by Lohka and Masui1 with some modifications. Testes were dissected from sexually mature X. laevis which had been injected with 100 I.U. human chorionic gonadotropin (HCG) and kept for 1 h at 22°C. The testes were washed free of blood and incubated overnight at 18°C in 200 % Steinberg's solution containing antibiotics and HCG(10 I.U./ml). Sperms were released by gently squeezing the testes, harvested by centrifugation at 1,500g for 10 min and washed 3 times with nuclear isolation buffer (NIB, 15 mmol/L NaCl, 60 mmol/L KCl, 15 mmol/L Tris, pH 7.4, 1 mmol/L DTT, 0.5 mmol/L spermine, 0.15 mmol/L spermidine). Sperms were treated with NIB containing lysolecithin at 330 μg/ml and proteinase inhibitor for 10 min at 22 °C. Lysolecithin -treated sperms were washed once with ice-cold NIB/3% bovine serum albumin (BSA), 3 times with NIB/0.4% BSA and finally resuspended in NIB/30% glycerol. They were counted with a hemocytometer, and stored at −70°C.

Preparation of HeLa cell chromosome clusters

As previously described14, HeLa cells were cultured in E-MEM (Eagle's modified essential medium) and synchronized at metaphase by treatment with TdR and nocodazole. Cells were swelled in 1/ 2 Hank's solution for 2 h and then homogenized. Chromosome clusters were pelleted by centrifugation at 1,000g for 10 min, resuspended in E-MEM containing 20% newborn calf serum and 10% glycerol and stored at −70°C.

Preparation of egg extracts

Egg extracts were prepared as previously described11. Briefly, unfertilized eggs were collected from HCG-treated mature female X. laevis and dejellied with 2% cysteine. After activated by Ca2+ ionophore A23187, they were washed with buffer containing cycloheximide and cytochalasin B. Then the eggs were crushed at 10,000 g, 4°C for 15 min. The material between the lipid top layer and yolk pellet was removed and centrifuged again at 10,000 g, 4°C for 15 min. The supernatant was used as egg extracts.

Nuclear assembly in vitro

Demembranated sperm nuclei and clusters of HeLa cell chromosomes were respectively mixed with freshly prepared egg extracts at a concentration of 103 nuclei or chromosome clusters/ μl egg extracts and then the ATP regenerating system (2 mmol/L ATP, 50 μg/ml creatine kinase, 2 mmol/L phosphocreatine) was added. The mixture was incubated at 22°C.

Antibodies

All antibodies used here were antilamin monoclonal antibodies generously donated by Dr. Krohne. They were antilamin LI I/ LI I I mAb X155, antilamin LI I mAb X223 and antilamin L I/ LI I/ lamin A mAb X16726.

Immunofluorescent microscopy

Incubated mixtures were smeared onto glass slide, air-dried for 5 min and fixed for 30 min in mixture of acetone and methanol (1:1) at −20°C. After incubated in PBS containing 1% BSA for 10 min, the specimen were incubated with mAb X155 for 1 h, and washed 3 times with PBS. Then FITC labeled goat anti-mouse IgG was added and the specimen were incubated for 30 min and washed 3 times with PBS. Chromatin was stained by DAPI (DNA dye). The specimen were mounted with a coverslip and examined by fluorescence microscope.

Electron microscopy

Aliquots of the incubation mixture were fixed in ice-cold 2% glutaraldehyde and osmium tetroxide successively, dehydrated through a graded concentration series of acetone and then embedded in Epon 812. Ultrathin sections were cut, stainned with uranyl acetate and lead citrate, and finally examined under JEM-100CX transmission electron microscope.

Gel electrophoresis and immunoblotting

After diluted 10 times with NIB, the incubation mixture was centrifuged at 3,000g for 10 min. The pelleted nuclei were washed 3 times with NIB and then suspended in SDS sample buffer. Proteins were resolved on 10% SDS-PAGE and transferred to nitrocellulose filters. Filters were blocked (37°C) in 3% BSA/ TBS (200 mmol/L Tris, 500 mmol/L NaCl, pH 7.5) for 2 h and then incubated (37°C) in 1% BSA /TTBS (200 mmol/L Tris, 500 mmol/L NaCl, pH 7.5, 0.05% Tween20) containing antilamin antibody mAb X155 or mAb X223 or mAb X167 for 3.5 h. After washed with TTBS and incubated in 1% BSA /TTBS containing Biotin labeled goat anti-mouse IgG for 2 h, filters were washed with TTBS and then blocked in 3% BSA / TBS for 1.5 h. Filters were incubated in buffer 1 (100 mmol /L Tris, 1 mol/L NaCl, 2 mmol /L MgCl2, 0.05% TritonX-100, pH 7.5) containing Avidin-AP (akaline phosphotase) for 15 min and then washed successively with buffer 1 and buffer 2 (100 mmol/L Tris, 1 mol/L NaCl, 5 mmol/L MgCl2, pH 9.5). Filters were stained in buffer 3 (100 mmol/L Tris, 100 mmol/L NaCl, 5 mmol/L MgCl2, pH 9.5) containing NBT and BCIP.

Results and disscussion

Incubation of demembranated Xenopus sperm nuclei in egg extracts resulted in rapid chromatin decondensation followed by nuclear formation. The demembranated sperm chromatin was in a long, thin and condensed form which was the characteristic of the sperm nucleus and could be stained strongly with DAPI[Fig 1]. When it was added to egg extracts and incubated at certain temperature (22°C), the sperm chromatin began to elongate and swell, showing discernible decondensation, in the first 10 min of incubation[Fig 2]. After 60–90 min, nearly all of the originally sharp-contoured demembranated sperm nuclei had decondensed and changed into spherical shape like a pronucleus[Fig 3], while less than 5% of them remained in their initial form. Most of the added sperm chromatin followed the same steps in assuming morphological alteration[Fig 4]. Lysolecithin was effective in removing cell membrane and nuclear envelope of sperm. At the ultrastructral level, we observed that after it was treated with lysolecithin, sperm chromatin was devoid of double-layered nuclear envelope and was rather condensed[Fig 5]; After incubation for about 1 h, the decondensed chromatin gradually acquired a continuous nuclear envelope with pores. The newly assembled nuclei were morphologically indistinguishable from those in intact interphase cells[Fig 6].

Fig 1
figure 1

Fluorescence microscopy (DAPI stain) showing the decondensation of demembranated sperm nuclei in egg extracts. Bars: 10 μm.The highly condensed sperm chromatin before incubation.

Full size image
Fig 2
figure 2

Fluorescence microscopy (DAPI stain) showing the decondensation of demembranated sperm nuclei in egg extracts. Bars: 10 μm.Elongated and swelled sperm chromatin after incubation of 10 min.

Full size image
Fig 3
figure 3

Fluorescence microscopy (DAPI stain) showing the decondensation of demembranated sperm nuclei in egg extracts. Bars: 10 μm.Highly decondensed sperm chromatin taking a spherical shape after incubation of 60 min.

Full size image
Fig 4
figure 4

Fluorescence microscopy (DAPI stain) showing the decondensation of demembranated sperm nuclei in egg extracts. Bars: 10 μm.Most of the added demembranated sperm nuclei followed the same steps to decondense to a spherical shape after incubation of 60 min.

Full size image
Fig 5
figure 5

Ultrathin section of demembranated sperm chromatin and nuclei assembled in vitro. Bars: 1 μmDemembranated sperm chromatin devoid of nuclear envelope.

Full size image
Fig 6
figure 6

Ultrathin section of demembranated sperm chromatin and nuclei assembled in vitro. Bars: 1 μmTypical nuclei assembled in vitro with continuous nuclear envelope, pores(↑) and well-decondensed chromatin.

Full size image

Nuclei assembled in vitro were assessed for the presence of lamina by indirect immunofluorescent microscopy. When incubated with mAb X155 (specific for lamin LI I/L I I I), the demembranated sperm nuclei had no positive staining while the newly assembled nuclei exhibited positive rim-staining[Fig 7]. It was reported that only lamin LI V was found in spermatids and spermatozoa27. Due to the lack of antibody specific for lamin LIv, we could not make sure whether the demembranated sperm nuclei still had some attached lamin L I v protein, although they had no intact lamina structure after demembranation. However, even though we could not conclude from the negative result that there was not any type of lamin protein around the periphery of demembranated sperm nuclei, we surely knew that there was no lamin LI I or LI I I protein; and then from the positive result we knew that during incubation, lamin LI I or/and LI I I which were previously stored in egg extracts participated in the formation of lamina structure of the newly assembled nuclei. Our previous work had demonstrated that nuclei assembled around clusters of HeLa cell chromosomes in egg extracts also had lamina 14.

Fig 7
figure 7

Indirect immunofluorescent staining of lamina in the nucleus assembled in vitro with antilamin mAb X155. Bars: 10μm.

Full size image

Western-blotting analysis was used to detect the lamina components of nuclei assembled in vitro. Nuclei were isolated from the incubation mixture by centrifugation and proteins were separated by gel-electrophoresis and then transferred to nitrocellulose filters. The fiters were probed with antilamin LI I/ LI I I mAb X155, antilamin LI I mAb X223, or antilamin LI/LI I/lamin A mAb X167 respectively. Using mAb X155 we found that both demembranated sperm nuclei and chromosomes of HeLa cells had no egg-derived lamin, whereas nuclei assembled in vitro from both of them contained the same lamin components as those found in egg extracts[Fig 8]. This confirmed our observation by indirect immunofluorescent microscopy that the lamins in the newly formed nuclei were derived from egg extracts. It is widely believed that X. laevis oocytes, eggs and early embryos contain only one lamin isoform, the lamin LI I I, which serves as a maternal pool for the formation of lamina in the pronuclei and early cleavage nuclei. During development lamin LI I I is replaced in embryonic nuclei by lamin LI and LI I at the midblastula and gastrula stages respectively21. Considering that lamin LI I and LI I I had the same molecular weight of 68kD and mAb X155 recognized both lamin LI I and LI I I, we then used mAb X223 and mAb X167 which both reacted with lamin LI I but not with lamin LI I I26. To our interest, we found that both egg extracts and the nuclei assembled in the egg extracts reacted positively with the two antibodies. These results unequivocally suggested that besides lamin LI I I, lamin LI I was also present in egg extracts and was involved in lamina formation of the nuclei assmbled in vitro in egg extacts[Fig 9, 10]. When we are preparing this paper, it comes to our notice that the recent work of Lourim and Krohne also confirmed the existence of lamin LI I in Xenopus oocytes, eggs and embryos26.

Fig 8
figure 8

Both the nuclei assembled in vitro around sperm chromatin and chromosomes of HeLa cells had the same lamin peptides as those in egg extracts. Proteins were separated by SDS-PAGE and transferred to NC filters and then probed with mAb X155 in combination with ABC-Western-blotting technique. 1. egg extracts. 2.nuclei assembled around demembranated sperm nuclei. 3.demembranated sperm nuclei. 4. nuclei reassembled around chromosomes of HeLa cells. 5. chromosomes of HeLa cells. Molecular weight was showed at right in kD.

Full size image
Fig 9
figure 9

Lamin LI I was present in egg extracts and nuclei assembled in vitro detected by mAb X167. 1. egg extracts. 2.nuclei assembled around chromosomes of HeLa cells. 3. chromosomes of HeLa cells. 4.nuclei assembled around demembranated sperm nuclei.

Full size image
Fig 10
figure 10

Lamin LI I was present in egg extracts and nuclei assembled in vitro detected by mAb X223. 1. egg extracts. 2. nuclei assembled around demembranated sperm nuclei. 3. nuclei assembled around chromosomes of HeLa cells. 4. chromosomes of HeLa cells.

Full size image

In discussing lamina formation of mammalian cells, Gerace and Blobel proposed a different functional specialization for lamin A and B in terms of chromatin and membrane binding respectively22. Since only lamin LI I I is found in X. laevis oocytes, eggs and early embryos, it has been suggested that lamin LI I I alone can build up a lamina structure in early cleavage nuclei and carry out both functions of chromatin-binding and membrane-binding in lamina formation through the mediation of some kinds of membrane or chromatin associated receptors. It has been hypothesized that different lamins are functionally equivalent during the developmental change of lamina composition in Xenopus embryogenesis and the change may be resulted from gene regulation21. This is somewhat interesting compared with the situations in mammalian and bird embryos. Now that lamin LI I is detected in egg extracts and nuclei reassembled in vitro, questions concerning lamin function and expression turn out even more complicated than previously presumed. How do these two polypeptides function and what is their relationship? Though both classified as B-type lamin, do lamin LI I and LI I I interact during lamina formation in a manner similar to lamin A and B in mammalian cells? Moreover, the finding of lamin LI I in egg extracts and nuclei assembled in vitro will help us to explain the obvious contradiction among observations concerning the problem of whether or not the binding of lamin proteins to chromatin surface and lamina formation are required for nuclear envelope assembly in cell-free system23, 24, 25. In fact when people tried to inactivate endogeneous lamin protein in eggs by immunodepletion or microinjection with antibody, they considered only the presence of lamin LI I I but neglected lamin LI I and thus impaired the soundness of their conclusion.

Another phenomenon which interests us is that the somatic chromosomes of HeLa cells and the sperm chromatin of X. laevis can both demonstrate lamina assembly in egg extracts. Previous work has shown that egg extracts are able to induce nuclear assembly around a wide variety of DNA templates1, 3, 14. Whether they share the same lamina formation pattern and whether developmentally controlled factors in egg extracts are required for the lamina formation remain unresolved and need further study.

It has been reported that soon after their synthesis in the cytoplasm the lamin proteins, including lamin LI I and LI I I, dimerize and are transported to the nucleus. A nuclear localization signal and a CaaX box at the carboxyl terminus are necessary to the latter process15, 16, 28, 29. The CaaX motif (C: cysteine, a: an aliphatic amino acid, X: any amino acid) is a substrate for 3 successive post-translational modifications (isoprenylation of the cysteine, proteolytic removal of three carboxyl terminal acids; and carboxyl-methylation of the isoprenylated cysteine). Another kind of modification of lamin proteins is phosphorylation 10. In all our Western-blotting positive lanes there were two bands at the position of 68 kD. Because mAb X223 is specific for only lamin LI I, we regard that the doublets in Fig 10 may be resulted from different migrating abilty of two forms of lamin LI I with different modification. As for the doublets in Fig 8, there may be two possibilities: 1)Each band contains both lamin LI I and LI I I and either one of lamin LI I and LI I I is of different modification form in the two bands. Because lamin L I I and LI I I have similar molecular weight and are both B-type lamin, it is quite possible that after being modified, either in the same manner or not, they have the same mobile activity in our gel system. 2) While both of the two bands contain lamin LI I, only one band contains lamin LI I I; in other words, there is only one form of lamin L I I I which has the same migrating pattern as that of one of the two forms of lamin LI I. Since mAb X167 is specific for lamin LI I and two other lamin isoforms but not for lamin LI I I, there is one possibility that the situation in Fig 9 is just like that in Fig 10: each band represents a modification form of lamin LI I. The much slower migrating band in the three figures at the position of approximately 150 kD may be the dimer of lamin proteins, either homodimers of lamin LI I or LI I I or heterodimers of lamin LI I and LI I I.