INTRODUCTION

The nuclear lamina consists of a polymeric network of intermediate filament molecules, the nuclear lamins, underlying the inner nuclear membrane. The lamina is a dynamic structure, undergoing expansion during interphase of the cell cycle, and depolymerization at mitosis upon breakdown of the nuclear envelope (NE)¹ (1). Mitotic disassembly and reassembly of the lamina is regulated by reversible lamin phosphorylation and dephosphorylation (1). Interphase lamin phosphorylation has also been reported (2-6), but its significance is not fully understood.

Several lamin kinases have been identified that promote mitotic lamina solubilization or inhibit lamina assembly in vitro. They include cyclin B/p34^cdc2 (7), S6 kinase II (8), protein kinase C (PKC) (4, 9), and the cAMP-dependent protein kinase PKA (10). Down-regulation of PKA has also been shown to be essential for mitotic lamina disassembly (11). Although not a lamin kinase, Ca²⁺/calmodulin-dependent kinase II (CaM kinase II) is also involved in mitotic NE breakdown in sea urchin embryos (12). PKC has also been shown to phosphorylate chicken lamin B₂ in interphase, a process thought to regulate lamin import into the nucleus (13). These observations imply that multiple kinases regulate the dynamics of the nuclear lamina during the cell cycle.

The transformation of the sea urchin sperm nucleus into a pronucleus at fertilization provides an opportunity to investigate NE assembly/disassembly during interphase. Sea urchin eggs are fertilized in G₁ phase of the first cell cycle after completion of both meiotic divisions. At fertilization, the sperm NE vesiculates and a new NE reforms around the male pronucleus as the sperm chromatin decondenses (14). Male pronuclear formation has been duplicated in a cell-free system by incubating detergent-permeabilized sperm nuclei in fertilized egg extracts (15-19). Detergent-permeabilized sperm nuclei retain their lamina, which consists of a major 65-kDa B-type lamin (referred to as lamin B) and several minor uncharacterized lamin epitope-containing peptides (18). The first step of male pronucleus formation in vitro is the disassembly of the sperm nuclear lamina. The pronuclear lamina is reassembled only following formation of the nuclear membranes during nuclear swelling (19).

Interphase lamina disassembly requires ATP hydrolysis, consistent with the involvement of protein kinase(s) (18). One kinase activated at fertilization in the sea urchin is PKC. Fertilization stimulates phospholipase C in the egg plasma membrane, releasing diacylglycerol and inositol 1,4,5-trisphosphate from phosphoinositides. Increased inositol 1,4,5-trisphosphate triggers an intracellular release of Ca²⁺, which together with diacylglycerol activates PKC (20). Activated soluble PKC has been shown to translocate to the plasma membrane (21). Translocation of activated PKC to non-plasma membranes, such as the NE, has also been reported, as PKC moves to the nucleus of cultured mammalian cells upon mitogenic stimulation (9, 22, 23). A sea urchin PKC isoform (suPKC1) has been cloned (24) and several substrates proposed (21, 25). However, no lamin kinase activity has been attributed to fertilization-activated PKC.

We report here that phosphorylation of sperm nuclear lamin B precedes its solubilization, in an interphase egg cytosolic extract, and provide evidence that this phosphorylation is mediated by PKC. Lamin B phosphorylation and solubilization precedes decondensation of the sperm chromatin, but is not sufficient to promote chromatin decondensation.

EXPERIMENTAL PROCEDURES

1,2-Bis(2-aminophenoxy)ethane-N,N,N,N-tetraacetic acid (BAPTA), 6-dimethylaminopurine (DMAP), staurosporine, chelerythrine, and the PKA inhibitor PKI were from Sigma. The PKC pseudosubstrate inhibitor peptide ((PKC) peptide-(19-31)), p13^suc1 beads, and purified rat brain PKC were from Upstate Biotechnology (Lake Pacid, NY). Autocamtide 3 and the PKC substrate peptide ((Ser²⁵) PKC peptide-(19-36)) were from Life Technologies (Bethesda, MD). [³²P]ATP was from DuPont NEN (Brussels, Belgium). The p34^cdc2 kinase inhibitors olomoucine and roscovitine were gifts from Dr. Laurent Meijer (26). The antibody W3-1 (a gift from Dr. Jon Holy), is a chicken polyclonal antibody raised against a fusion protein encoded by a sea urchin lamin lamin B cDNA clone (27). W3-1 was previously characterized (27), and recognizes a 65 kDa B-type lamin (p65) on Western blots of Lytechinus pictus sperm and male pronuclei (18, 19). The anti-sea urchin PKC antibody, a gift from Dr. Sheldon Shen, was raised in rabbits against the NH₂-terminal domain of L. pictus PKC (suPKC1) (24).

L. pictus sperm heads were demembranated by extraction with 0.1% Triton X-100 in nuclear buffer (NB; 250 mM sucrose, 250 mM glucose; 75 mM NaCl, 0.5 mM spermidine, 0.15 mM spermine, 50 mM Hepes, pH 7.2), and demembranated nuclei were washed and resuspended in NB to 10⁸ nuclei/ml as described previously (16). Demembranated nuclei retain their lamina, including all p65 (18), as well as two lipophilic structures, which represent detergent-resistant NE specializations at each end of the nucleus, in the centriolar and acrosomal fossa regions (16).

Mature L. pictus eggs, arrested in G₁ after completion of both meiotic divisions, were fertilized and cytosolic extracts prepared 10-15 min postinsemination as described elsewhere (19). Briefly, eggs were homogenized, the lysate cleared at 10,000 × g, and the supernatant centrifuged at 150,000 × g for 3 h at 4 °C to yield a cytosolic extract. To be able to detect solubilized sperm lamin B, cytosols were immunodepleted of endogenous lamin B using the W3-1 antibody as described previously (18) prior to lamina disassembly reactions. Immunodepletions were systematically verified by immunoblotting analysis of the cytosol using the W3-1 antibody (not shown) (18). The pH of fertilized egg cytosolic extracts was 7.

Demembranated nuclei (10⁸/ml) were diluted 10-fold in NB and added to 100 µl of cytosolic extract containing an ATP-generating system (15) to a final concentration of 1000 nuclei/µl. The lamina disassembly reaction proceeded at room temperature for 30 min unless indicated otherwise, and was stopped by chilling to 4 °C. Nuclei were pelleted at 2,000 × g for 1 min a 4 °C, washed in 500 µl of NB at 4 °C, and resuspended in either NB for immunofluorescence studies, or SDS sample buffer for SDS-polyacrylamide gel electrophoresis (PAGE). The remaining cytosol was solubilized in SDS sample buffer. In some experiments, inhibitors were added to the reaction mix 15 min prior to incubating nuclei.

The chromatin of demembranated sperm nuclei incubated in cytosolic extract as described above was considered to be decondensed when the conical sperm nucleus acquired a spherical morphology, as described previously in vivo (14) and in vitro (15-19). Sperm chromatin decondensation has been shown to be accompanied by changes in sperm histone phosphorylation (28, 29).

For incorporation of radiolabeled phosphate into sperm lamin B, 0.75 µCi/µl [-³²P]ATP was added to the lamina disassembly reaction. The reaction was carried out as above, nuclei pelleted, and nuclei and cytosol solubilized in SDS sample buffer. Proteins were resolved by SDS-PAGE on 10% polyacrylamide gels, transferred to nitrocellulose, and subjected to autoradiography. Identity of lamin B was verified by immunoblotting duplicate samples using anti-lamin B antibodies (see below). In some experiments, inhibitors were included in the reaction as described above.

For in vitro phosphorylation of sperm lamin B by purified PKC, 10⁵ nuclei were incubated with 100 ng of purified rat brain PKC in 40 µl of PKC phosphorylation medium (200 mM NaCl, 50 mM Tris, pH 7.4, 10 mM MgSO₄, 100 µM CaCl₂, 40 µg/ml phosphatidylserine, 20 µM diacylglycerol, 1 mM dithiothreitol, 12 µM ATP, 0.75 µCi/µl [-³²P]ATP). [-³²P]ATP was omitted when nuclei were used for immunofluorescence. The reaction mixture was incubated for 30 min at room temperature and chilled on ice. Nuclei were pelleted at 2,000 × g for 1 min, and 5 × SDS sample buffer was added to the supernatant before boiling. Proteins were resolved by SDS-PAGE on 10% polyacrylamide gels, transferred to nitrocellulose, and revealed by autoradiography. In some experiments, inhibitors were included in the reaction as described above.

Type VII-S calf alkaline phosphatase in phosphate-buffered saline was added to nuclei or cytosol to a final concentration of 100 units/ml. Samples were incubated for 1 h at 37 °C, and the reaction was terminated by adding SDS sample buffer and boiling.

Immunofluorescence detection of lamin B was performed using the W3-1 antibody (19). DNA was labeled with 0.1 µg/ml Hoechst 33342. Images were captured, processed, and printed as described previously (19). For immunoblotting, proteins were resolved by SDS-PAGE on 10% polyacrylamide gels and transferred onto nitrocellulose. Membranes were probed with anti-lamin B antibodies (19) or anti-PKC antibodies (1:500 dilution), then with horseradish peroxidase-coupled secondary antibodies. Polaroid photographs of blots were scanned, and signals were quantified using the OptiLab/Pro software (Graftek, Mirmande, France). Immunodepletion of lamin B from the cytosol was performed as described elsewhere (18). Immunodepletion of PKC from the cytosol was done using a 1:25 dilution of the anti-PKC polyclonal antibody and protein A-Sepharose-bound anti-rabbit IgG antibodies.

Two-dimensional phosphopeptide mapping of lamin B phosphorylated in the presence of [-³²P]ATP was performed as described previously (30). Phosphorylated lamin B immobilized on nitrocellulose was incubated at 25 °C overnight in 50 mM NH₄HCO₃ containing 100 µg/ml trypsin. Efficiency of trypsinization was determined by Cerenkov counting of the supernatant and was routinely greater than 90%. Two-dimensional phosphopeptide mapping was performed on Kodak cellulose thin layer plates. Electrophoresis was carried out in pH 1.9 buffer (formic acid/glacial acetic acid/deionized water, 50:156:1794, v/v) for 18 min at 1,000 V followed by chromatography for 3 h in isobutyric acid buffer (isobutyric acid/1-butanol/pyridine/glacial acetic acid/deionized water, 65:2:5:3:29, v/v). Tryptic phosphopeptide maps were visualized by autoradiography at 70 °C.

RESULTS

A peculiarity of sea urchin eggs is that they are fertilized in interphase (G₁), after completion of both meiotic divisions. Previous studies have shown that the sea urchin sperm lamina is disassembled in G₁ phase cell-free extracts (18). To further document this interphase lamina disassembly, we examined the time course of sperm lamin B solubilization in fertilized egg cytosolic extract. Demembranated sperm nuclei that still contain a lamina (Fig. 1A, Input) were incubated in cytosol for 5 and 10 min, and examined by immunofluorescence using anti-lamin B antibodies. All detectable lamin labeling disappeared within 10 min, first from the lateral aspects of the chromatin, then from the nuclear poles (Fig. 1A). These data were confirmed by immunoblotting, thus eliminating antigen-masking artifacts. Concomitant with disassembly from nuclei, lamin B appeared progressively in the cytosol (Fig. 1B). Furthermore, the apparent molecular mass of sperm lamin B was shifted from 65 to 68 kDa prior to solubilization (Fig. 1B, upper panel), suggesting rapid phosphorylation. Alkaline phosphatase treatment of nuclei harboring the 68-kDa lamin restored its migration to a 65 kDa form (data not shown), confirming that the 68-kDa lamin (designated pp68) is a phosphorylated form of lamin B (p65).

Phosphorylation of lamin B was directly demonstrated by rapid (1 min) ³²P incorporation into sperm lamin B (Fig. 1C, upper panel), immediately followed by the release of phosphorylated lamin into the cytosol (Fig. 1C, lower panel). Identity of the 68-kDa ³²P-labeled protein and lamin B was verified by immunoprecipitation of both components from the cytosol by anti-lamin B antibodies (Fig. 1D). These results show that the only phosphorylated component of the 68-kDa protein in egg cytosol is lamin B, and indicate that sperm lamin B is phosphorylated prior to being solubilized in interphase egg cytosol.

Release of intracellular Ca²⁺ has been shown to trigger mitotic NE breakdown in the sea urchin (31). To determine whether Ca²⁺ was required for lamin solubilization in interphase fertilized egg cytosol, sperm nuclei were incubated in cytosol containing 5 mM of the Ca²⁺ chelator BAPTA (a concentration that inhibits NE growth associated with completion of male pronuclear formation in vitro) (32) and lamin B solubilization was examined by immunofluorescence. BAPTA blocked lamin solubilization and chromatin decondensation (Fig. 2A). This inhibition was reversible since incubation of BAPTA-treated nuclei in fresh cytosol restored both processes (Fig. 2A, Wash). These results were verified by immunoblotting analysis of nuclei (data not shown). BAPTA also prevented most lamin B phosphorylation and solubilization, as judged by autoradiography of cytosolic proteins (Fig. 2B), indicating that both processes are Ca²⁺-dependent.

Several lamin kinases have been identified, including p34^cdc2, PKC, and PKA (7, 9, 10, 33). In an attempt to determine which kinase(s) mediates interphase sperm lamin B phosphorylation, sperm nuclei were incubated in egg cytosol containing increasing concentrations of the following kinase inhibitors: the nonspecific kinase inhibitors DMAP and staurosporine, the p34^cdc2-specific inhibitors olomoucine and roscovitine (26), PKI, or the PKC-specific inhibitor chelerythrine (34). Nuclear proteins were immunoblotted using anti-lamin B antibodies and amounts of lamin B quantified by densitometry. DMAP and staurosporine inhibited lamin B solubilization with an IC₅₀ of 1.5 mM and 12 µM, respectively, whereas olomoucine, roscovitine, or PKI were ineffective (Fig. 3A). The most effective inhibitor of lamin B solubilization was the PKC-specific inhibitor chelerythrine (IC₅₀ 0.17 µM; Fig. 3A). As observed with Ca²⁺ chelation, these inhibitions were reversible. Immunofluorescence observations verified these results and showed a parallel between prevention of lamin solubilization by kinase inhibition and prevention of chromatin decondensation (Fig. 3B).

To determine whether the effects of these inhibitors on lamin solubilization were due to inhibition of lamin phosphorylation, sperm nuclei were added to cytosol containing [-³²P]ATP and kinase inhibitors. The presence of phosphorylated, solubilized lamin B in the cytosol was determined by autoradiography. As shown in Fig. 3C, lamin B phosphorylation was inhibited by DMAP (2 mM), staurosporine (100 µM), and chelerythrine (10 µM), but was insensitive to olomoucine, roscovitine, or PKI. These results suggest a role of PKC in sperm lamin B phosphorylation resulting in solubilization in interphase egg cytosol.

To investigate further the involvement of PKC in interphase sperm lamin B phosphorylation and solubilization, nuclei were added to cytosol preincubated with 50 µM of either a highly selective PKC inhibitor peptide or a PKC substrate peptide, in the presence of [-³²P]ATP. Lamin B phosphorylation was examined by autoradiography of cytosolic proteins, and relative amounts of phosphorylated and solubilized lamin B were determined by densitometry of duplicate autoradiograms. Both peptides prevented lamin B phosphorylation and solubilization (Fig. 4A). These inhibitor concentrations were the minimal concentrations completely abolishing lamin B phosphorylation (data not shown) (30). In contrast, preincubation of cytosol with p13^suc1-agarose beads (0.25 µg/µl cytosol), which specifically bind p34^cdc2, or 50 µM autocamtide 3, a CaM kinase II-specific substrate peptide, did not inhibit cytosolic lamin kinase activity (Fig. 4A). Together with previous data, these results suggest a role for PKC in phosphorylating and solubilizing sperm lamin B in interphase cytosol.

To determine if PKC was the only enzyme responsible for interphase sperm lamin B phosphorylation, cytosol was immunodepleted of endogenous PKC using a 1:25 dilution of a polyclonal antibody against the NH₂-terminal end of suPKC1 (24) (Fig. 4B). This antibody reacted with a 71-kDa protein on immunoblots of egg cytosol (Fig. 4B, arrow), and occasionally with 52- and 84-kDa uncharacterized proteins in some cytosol preparations (not shown) (24). Lamin B phosphorylation in PKC- or mock-depleted cytosol showed that lamin B kinase activity was abolished by ~90% by immunodepletion of PKC from the cytosol (as determined by densitometry of the autoradiogram shown in Fig. 4C). When examined by immunofluorescence using anti-lamin B antibodies, nuclei incubated in PKC-depleted cytosol exhibited peripheral lamin labeling similar to that of input nuclei, while the chromatin remained condensed (data not shown). Subsequent addition of purified rat brain PKC (100 pg/µl) to PKC-depleted cytosol restored lamin B phosphorylation and solubilization (Fig. 4C), and promoted chromatin decondensation. These results argue that sperm lamin B phosphorylation and solubilization in interphase cytosol are mediated by PKC.

Sea urchin lamin B contains several consensus PKC phosphorylation sites ((S/T)-X-(K/R); see Holy et al. (27) for Strongylocentrotus purpuratus and Lytechinus variegatus lamin sequences). To determine whether sperm lamin B is a substrate for phosphorylation by PKC, nuclei were incubated in PKC phosphorylation medium containing purified rat PKC or human _II PKC (33), and the reaction supernatant analyzed by autoradiography for the presence of phosphorylated and solubilized lamin B. Lamin B was phosphorylated and solubilized by both kinases (Fig. 5A). Phosphorylation did not occur in the absence of PKC (indicating the absence of endogenous nucleus-associated lamin kinase activity) or in the presence of the PKC pseudosubstrate inhibitor peptide (50 µM), the PKC substrate peptide (50 µM) or the PKC-specific inhibitor chelerythrine (100 µM) (Fig. 5A). This indicates that sperm lamin B is a substrate for purified PKC. Immunofluorescence analysis of nuclei incubated with rat PKC (Fig. 5B) or human _II PKC (not shown) showed that all detectable lamin B had disassembled from nuclei. However, although lamins were solubilized, the chromatin remained condensed (Fig. 5B), indicating that disassembly of the sperm nuclear lamina was not sufficient to promote chromatin decondensation in the absence of cytosol.

Final characterization of the interphase cytosolic lamin kinase was carried out by comparing the lamin B phosphorylation sites of the cytosolic kinase and of purified rat PKC. Tryptic digests of lamin B phosphorylated by both kinase preparations were subjected to two-dimensional thin layer chromatography and autoradiography. Lamin B phosphorylated by the cytosolic kinase generated 13 phosphopeptides that migrated with a pattern similar to 13 out of 14 phosphopeptides produced by purified PKC (Fig. 6, compare peptides 1-13 in left and middle panels). The identity of these 13 phosphopeptides was ascertained by their comigration when both tryptic digests were run on the same chromatogram (Fig. 6, Mix). As expected from our previous data, no phosphopeptides were detected when PKC was omitted from the phosphorylation reaction (data not shown). These results indicate that the cytosolic interphase lamin kinase accounting for the lamin phosphopeptides detected is PKC.

DISCUSSION

We report in this study that sea urchin sperm nuclear lamina disassembly in interphase egg cytosol is a result of lamin B phosphorylation mediated by PKC. The following evidence supports our conclusions: (i) sperm lamin B phosphorylation is Ca²⁺-dependent; (ii) phosphorylation is inhibited by the PKC-specific inhibitor chelerythrine, but not by inhibitors of PKA, p34^cdc2 or CaM kinase II; (iii) lamin B phosphorylation is also inhibited by highly specific PKC inhibitors of different compositions, specificities and modes of action, such as a PKC inhibitor peptide and a PKC substrate peptide; (iv) lamin B phosphorylation is abolished in cytosol immunodepleted of PKC, and restored after addition of purified PKC; (v) sperm lamin B can be phosphorylated and solubilized by purified mammalian PKC in vitro; and (vi) finally, two-dimensional phosphopeptide maps of lamin B phosphorylated by the interphase cytosolic kinase and by purified mammalian PKC are virtually identical.

The identity of the 13 phosphopeptides of lamin B phosphorylated in interphase egg cytosol and by purified PKC argues that PKC is the only kinase required for phosphorylation and solubilization of sperm lamin B. Lamin phosphorylation and solubilization by PKC alone is unprecedented, as mitotic lamin solubilization in somatic cells appears to be elicited by multiple kinases (35). Although only one sea urchin PKC isoform has been cloned (suPKC1) (24), several PKC isoforms may exist and may phosphorylate lamin B. Nonetheless, inhibition of lamin B phosphorylation in cytosol immunodepleted of PKC using an antibody against suPKC1 suggests that interphase sperm lamin B phosphorylation is elicited by a single PKC isoform.

Sea urchin eggs are fertilized at the pronuclear stage, so the female pronucleus is fully formed and remains intact as the sperm NE successively disassembles and reassembles to form the male pronuclear envelope. Thus an unresolved issue is how the integrity of the female pronucleus is maintained during sperm NE disassembly. One possibility is that the female pronucleus contains a different set of lamins that would not be a substrate for PKC (36). This is suggested by the lack of reactivity of female pronuclei by immunofluorescence and immunoblotting using several anti-lamin antibodies (27), whereas sperm nuclei are highly reactive (18). Alternatively, the female pronucleus may contain the same lamins, but with different covalent modifications or specific lamin-associated proteins that might affect phosphorylation and solubilization by PKC. A third possibility may be the lack of translocation of PKC to the female pronucleus, as a result of the cytoplasmic reorganization that follows fertilization (37). PKC may also be translocated to both nuclei, but its activation restricted to the sperm NE, perhaps because of an activator in the sperm NE. This idea is supported by the absence of phosphorylation of sea urchin embryo nuclear lamin B in interphase egg cytosolic extract,² despite the presence of several PKC phosphorylation sites (27).

The presence of a PKC activator in the nucleus has been reported (38). Nuclei have a phosphoinositide cycle distinct from that of the plasma membrane, that is responsive to extracellular stimuli (39). Thus at fertilization, nuclear PKC may be activated by the production of diacylglycerol at the NE (40), or by a lipid nuclear membrane activator similar to that identified in human leukemia cells (38). Whether similar factors exist in the sperm NE, or in specialized regions of the NE such as the lipophilic structures, is currently being investigated.

Lamin B phosphorylation and solubilization invariably precedes chromatin decondensation in vitro. Complete sperm lamin solubilization occurs within 10 min, by the time the first morphological chromatin changes are detected (15). Furthermore, all treatments that blocked lamin B phosphorylation in the present study also prevented chromatin decondensation, suggesting a role for lamin phosphorylation in the decondensation process. This is supported by the inhibition of chromatin decondensation in vivo after fertilization or microinjection of sperm nuclei into DMAP-treated eggs (41, 42). Nonetheless, if lamin B is phosphorylated and solubilized by purified PKC in the absence of cytosol, the nuclei remain condensed, suggesting that lamin solubilization is not sufficient for chromatin decondensation.

An additional step which may be necessary for chromatin decondensation is the phosphorylation of sperm histones. Sperm histones SpH1 and SpH2B are phosphorylated in vivo within 3 min of fertilization (43), as well as in vitro, albeit at a slower rate (29). Interestingly, the conversion of Sp histones to their modified form in vivo takes place in eggs treated with DMAP, indicating that the sperm histone kinase is DMAP-insensitive (41). The sperm histone kinase is thus likely to be distinct from the mitotic histone kinase p34^cdc2 (44) and from PKC, which are both inhibited by DMAP (45) (this study). Chromatin decondensation therefore appears as a multistep process involving PKC for lamin phosphorylation and lamina disassembly and an as yet unidentified DMAP-insensitive kinase for histone phosphorylation.