Calcium-and Cell Cycle-dependent Association of Annexin 11 with the Nuclear Envelope*

,

The vertebrate annexins constitute a family of 12 distinct genes that encode proteins with the shared ability to bind calcium dependently to negatively charged phospholipids (for reviews see Refs. 1 and 2). Despite their considerable structural similarity and overlapping biophysical characteristics, research in recent years indicates that individual annexins have distinct physiological roles. Thus, although annexin function is still not understood at the mechanistic level, targeted gene disruption experiments in mice and chick DT40 cells have revealed a range of phenotypes that are generally unaccompanied by compensatory changes in expression of other annexins. These phenotypes point to roles for annexins in diverse biological processes, including vesicle trafficking (3,4), calcium signaling (5,6), and lipid raft organization (7,8).
For several annexins it is clear they do not act alone. Annexin 2 was first isolated as a component of a stable heterotetrameric complex with S100A10 (formerly p11) (9). S100 proteins form a branch of the EF-hand Ca 2ϩ -binding proteins exemplified by calmodulin and constitute a second family of calcium-binding proteins of poorly understood functions (10). The (annexin 2⅐S100A10) 2 complex is unusual in being calciumindependent. This is because the two EF-hands of S100A10 contain amino acid changes that render the canonical helixloop-helix incapable of binding calcium ions while driving structural changes that effectively lock S100A10 into a form that mimics a hypothetical Ca 2ϩ -bound structure (11). Although the complex is stable in the absence of Ca 2ϩ , the two annexin 2 monomers within the complex retain Ca 2ϩ sensitivity, rendering the quaternary protein complex potentially sensitive to changes in intracellular Ca 2ϩ concentration.
Interactions between annexins and S100 proteins are now known to extend to other members of these protein families. Annexin 1 binds to S100A11 (formerly S100C) (12,13) and annexin 11 binds to S100A6 (formerly calcyclin) (14). In contrast to the annexin 2⅐S100A10 complex, interactions between annexins 1 and 11 and their respective S100 ligands are absolutely dependent on Ca 2ϩ . For this reason, it has so far proven impossible to isolate the Ca 2ϩ -dependent annexin⅐S100 complexes from cell or tissue extracts. The discovery that annexin 11 binds to S100A6 was the result of affinity chromatography using immobilized S100A6 (15). A reverse strategy, in which the immobilized N terminus of annexin 1 was used as bait in A431 cell extracts, led to the discovery of S100A11 as a cognate binding partner (12). Given the specificity of the annexin 1-S100A11 and annexin 11-S100A6 interactions and the relatively low Ca 2ϩ requirement for their formation, it is reasonable to expect that under certain circumstances these complexes may exist in living cells. Indeed, mutation of the binding site for annexin 1 on S100A11 results in failure of S100A11 to be recruited to transferrin receptor-positive endosomes, indicating that this interaction does indeed occur in vivo (16).
Despite several studies on the interaction between annexin 11 and S100A6, including identification of the binding site to residues 49 -62 in the long N-terminal domain (17), the mutual exclusion of the two proteins from a single subcellular compartment raises the possibility that the annexin 11⅐S100A6 complex may exist only in vitro. Because annexin 11 and S100A6 are both Ca 2ϩ -binding proteins and because S100A6 protein levels oscillate in a cell cycle-dependent manner (18), we therefore examined whether Ca 2ϩ fluxes and/or cell cycle progression might regulate the co-localization of annexin 11 and S100A6. We show here that, whereas S100A6 is relatively sedentary in the nuclear envelope, annexin 11 interacts dynamically with the nuclear envelope in both a Ca 2ϩ -and cell cycle-dependent manner. These results support the idea of a functional relationship between annexin 11 and S100A6. * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Immunofluorescence Microscopy-Cells were washed in PBS, fixed/ permeabilized on ice for 30 min in 2% paraformaldehyde ϩ 0.1% Triton X-100, washed four times in PBS and blocked for 15 min with 1% BSA (Sigma). Fixed cells were then incubated in a moist chamber with primary antibody in 1% BSA for 1 h, followed by four washes of 5-10 min each in PBS before incubation with the secondary antibody in 1% BSA for another hour. After four more washes in PBS, samples were observed under a Bio-Rad Radiance 2000 AGR 3 (Q) confocal attached to a Zeiss Axiovert S100TV inverted microscope using a ϫ100 oil immersion lens. Images were acquired with LaserSharp 2000 (Bio-Rad) and processed using Metamorph 4.6r9 (Universal Imaging).
Immunoprecipitation, SDS-PAGE, and Western Blotting-A431 cell monolayers were washed in PBS, solubilized in 500 l of lysis buffer (50 mM Tris-HCl, pH 7.4, 50 mM NaCl, 30 mM sodium pyrophosphate, 50 mM sodium fluoride, 100 M sodium orthovanadate, 0.2 mM CaCl 2 , 1% Triton X-100, 0.1% BSA, 0.02% sodium azide) supplemented with 10 l of protease inhibitor mixture (Sigma) per 10 7 cells and lysed by three rounds of sonication on ice. Lysates were precleared at 16,000 g for 30 min at 4°C, followed by incubation with 30 l of 50% Protein G-Sepharose 4 fast flow slurry (Amersham Biosciences) for 30 min, and then 5 min of centrifugation at 16,000 g at 4°C. L-19 goat anti-human annexin 11 antibody (Santa Cruz Biotechnology) was immobilized on Protein G-Sepharose beads and added to the precleared lysates for 2 h at 4°C and centrifuged, and the supernatant discarded. Beads were washed four times in lysis buffer minus BSA and once in ice-cold PBS. After removal of the supernatant, immunoprecipitates were resuspended in SDS-PAGE sample buffer, boiled for 5 min, and resolved by SDS-polyacrylamide gel electrophoresis on discontinuous mini-gels. Proteins were electroblotted for 45 min at 400 mA onto Hybond-P polyvinylidene difluoride transfer membrane (Amersham Biosciences).
Membranes were treated with blocking solution (10% skimmed milk or 10% BSA in PBS-Tween (PBS ϩ 0.05% Tween 20)) for 40 min at room temperature and incubated for 16 h at 4°C with primary antibody in PBS-Tween. Following three PBS-Tween washes, membranes were incubated with secondary antibody for 90 min, also in PBS-Tween (1/ 10,000 dilution). IgG-horseradish peroxidase conjugates were used as secondary antibodies. Blots were developed after three more PBS-Tween washes using the ECL Plus Western blotting detection system (Amersham Biosciences) and visualized in a Fujifilm Intelligent Dark Box II coupled to a LAS-1000 charge coupled device-camera.
Generation of Annexin 11⅐Green Fluorescent Protein (GFP) Fusion Constructs-Full-length human annexin 11 coding sequence, annexin 11 N-terminal domain (nucleotides 1-642 in the annexin 11 coding sequence), and annexin 11 C-terminal core domain (nucleotides 624 -1515 in the annexin 11 coding sequence) were amplified by RT-PCR from A431 cDNA and cloned in-frame into pEGFP-C3 vector (Clontech) for fusion at the C-terminal end of GFP or into pEGFP-N3 vector (Clontech) for fusion at the N-terminal end of GFP.
Transient Transfection and Live Cell Imaging-Cells were transfected with the different annexin 11⅐GFP fusion constructs using LipofectAMINE 2000 reagent (Invitrogen) following the manufacturer's instructions and incubated for 24 h to allow gene expression before placing them in a 37°C preheated microscope chamber with 5% CO 2 supply. Cells were observed before and after stimulation under a Bio-Rad Radiance 2000 AGR-3 (Q) confocal attached to a Zeiss Axiovert S100TV inverted microscope using a ϫ100 oil immersion lens. Images were acquired with LaserSharp 2000 (Bio-Rad) and analyzed using Metamorph 4.6r9 (Universal Imaging).
Cell Synchronization-Cells were grown to 50% confluence and incubated overnight in medium containing 4 mM thymidine (Sigma) to achieve S phase block. Dishes were extensively washed in PBS and cultured for a further 6 h in medium containing 0.07 g/ml nocodazole (Sigma) for growth arrest at the onset of mitosis. Cells were then extensively washed and allowed to recover in normal medium for 5-60 min before immunofluorescence analysis.

RESULTS
We first examined the subcellular localization of annexin 11 and S100A6 in fixed A431 epidermoid carcinoma cells. Annexin 11 was predominantly nuclear, though excluded from the nucleoli, and was also apparent in punctate form in the cell cytoplasm (Fig. 1A). In contrast, S100A6 displayed a strong signal in the region of the nuclear envelope, together with discrete inner nuclear structures and punctate cytoplasmic staining that did not coincide with annexin 11. Thus, in unstimulated A431 cells, S100A6 and annexin 11 appear to occupy non-overlapping subcellular compartments. Similar results were obtained in HeLa and HEp2 cells (data not shown). To confirm that the S100A6-positive ring-like structure at the FIG. 1. Annexin 11 and S100A6 reside in distinct subcellular compartments in unstimulated A431 cells. A, untreated A431 cells were fixed and stained for annexin 11, S100A6, and DNA (with propidium iodide). Annexin 11 concentrates at the cell nucleoplasm and is excluded from the nucleoli and the nuclear envelope, whereas the fluorescent signal for S100A6 concentrates mainly at the nuclear envelope. B, untreated A431 cells were fixed and stained for S100A6, the inner nuclear envelope protein lamin B, and DNA. S100A6 strongly co-localizes with lamin B at the nuclear envelope of A431 cells.
nuclear periphery is the nuclear envelope, A431 cells were co-stained for lamin B, a well characterized resident of the inner nuclear membrane (Fig. 1B). The results show that S100A6 is indeed strongly enriched at the nuclear envelope and also co-localizes with lamin B at focal points within the nucleus. The constitutive association of S100A6 with the nuclear envelope in A431 cells contrasts with previous studies in vascular smooth muscle cells in which this interaction was reported to be Ca 2ϩ -dependent (19).
To investigate the Ca 2ϩ sensitivity of S100A6 and annexin 11 in our experimental model, A431 cells were exposed to ionomycin for 30 min prior to fixation and immunostaining ( Fig. 2A). Elevation of intracellular Ca 2ϩ caused the relocalization of annexin 11 to the nuclear envelope but had little effect on either S100A6 or the punctate cytoplasmic pool of annexin 11. Because annexin 11 has been reported to be a substrate for cellular protein kinases, including the calcium-activated tyrosine kinase Pyk2 (20,21), we also examined whether the response to ionomycin was accompanied by protein phosphorylation (Fig. 2B). Immunoprecipitation of annexin 11 from ionomycin-treated cells, followed by Western blotting with an anti-phosphotyrosine antibody, revealed that annexin 11 is indeed phosphorylated on tyrosine under these conditions. In contrast, neither staurosporine nor EGF treatment led to tyrosine phosphorylation of annexin 11, nor did either stimulus drive the association of annexin 11 to the nuclear envelope (not shown).
Because the lysis conditions used in these immunoprecipitation experiments favored a whole cell extraction, we next investigated the ionomycin-induced tyrosine phosphorylation of annexin 11 in cytosolic and nuclear extracts prepared in the absence and presence of 200 M Ca 2ϩ (Fig. 2C). These experiments demonstrate that preservation of tyrosine phosphorylation in lysates of ionomycin-treated A431 cells requires the presence of Ca 2ϩ in the lysis buffer. Interestingly, separation of the nuclear and cytosolic pools led to the apparent partial degradation of tyrosine-phosphorylated annexin 11 in both fractions to polypeptide species of 30 and 40 kDa, respectively. However, reprobing the blot with antisera to annexin 11 shows that the majority of the protein remains in the 55-kDa form, suggesting that tyrosine phosphorylation of annexin 11 may increase its susceptibility to proteolysis.
Because studies elsewhere have shown that annexin 11 can also be tyrosine-phosphorylated in activated VSMCs (21), we next examined whether the results obtained in A431 cells would be paralleled in VSMCs. In these experiments we performed live cell-image analysis to track the localization of an ectopically expressed annexin 11⅐GFP fusion protein in primary cultures of porcine VSMCs. A series of confocal images were obtained every 5 s for 3 min following exposure to ionomycin and clearly showed that annexin 11⅐GFP becomes tightly localized to the nuclear envelope as well as to both plasma membrane and dispersed intracellular structures within a few seconds of ionophore treatment (Fig. 3A). Treatment of VSMCs with PDGF-BB had no effect on the nuclear annexin 11⅐GFP but led to a concentration of cytoplasmic annexin 11⅐GFP partly to the plasma membrane and also to intracellular structures (Fig. 3B). The contrasting effects of ionophore and growth factor in VSMCs closely match the results obtained in A431 cells. Consistent with these findings, we observed a negligible effect of PDGF-BB on tyrosine phosphorylation of annexin 11 in stimulated VSMCs (Fig. 3C), whereas annexin 11 was clearly tyrosine-phosphorylated in ionomycin-treated cells.
Next, we investigated the structural basis for the annexin 11 response to ionomycin (Fig. 4). Two constructs were generated in which the full-length annexin 11 coding sequence was fused to GFP, one with GFP at the C terminus of annexin 11 (anx11⅐GFP) and the other with GFP at the N terminus (GFP⅐anx11). Two additional expression plasmids were constructed, one in which the four-repeat Ca 2ϩ -binding C terminus of annexin 11 was replaced with GFP (N-ter⅐GFP) and a second in which GFP replaced the long proline-, tyrosine-and glycine-rich N terminus (GFP⅐C-ter). Transient transfection of these constructs in A431 cells revealed that anx11⅐GFP and GFP⅐anx11 were expressed in both the nucleus and cytoplasm of living cells, with anx11⅐GFP tending to exhibit a greater enrichment in the nucleus. N-ter⅐GFP expression was markedly stronger in the nucleus than the cytoplasm, whereas GFP⅐C-ter was completely excluded from the nucleus. This is in agreement with previous observations that showed the N-ter-FIG. 2. Annexin 11 relocates to the nuclear envelope in response to elevation of intracellular Ca 2؉ levels. A, A431 cells were stimulated for 30 min with 2.5 M ionomycin in medium containing Ca 2ϩ and subsequently fixed and stained for annexin 11 and S100A6. Annexin 11 partially relocates to the nuclear envelope under these conditions (green arrows), where it co-localizes with S100A6 (left panels). B, A431 cells were either stimulated with 2.5 M ionomycin for 30 min, 1 M staurosporine for 3 h, 40 ng/ml EGF for 15 min, or left untreated (control). Immune complexes were resolved by SDS-PAGE followed by Western blotting with an anti-phosphotyrosine antibody, and for annexin 11 to confirm equal loadings. C, A431 cells were exposed to ionomycin or buffer alone and extracted into lysis buffer containing either 200 M CaCl 2 or 5 mM EGTA. Samples were centrifuged, and annexin 11 was subsequently immunoprecipitated from both supernatant (Cyt) and resolubilized nuclear pellet (Nuc). Following SDS-PAGE and Western transfer, the membrane was immunoblotted first with an anti-phosphotyrosine antibody and then, after stripping the blot, with an antibody to annexin 11. minal domain of annexin 11 is both necessary and sufficient for the nuclear localization of annexin 11 (22).
We then compared the responses of these expression constructs to ionomycin treatment in the presence and absence of extracellular Ca 2ϩ . Both anx11⅐GFP and GFP⅐anx11 displayed almost complete relocalization from the nucleus to the nuclear envelope, together with a change from diffuse cytoplasmic fluorescence to punctate cytoplasmic or plasma membrane-asso- Selected images from the timed series are shown, revealing a rapid and striking translocation of annexin 11 to the nuclear envelope and various other intracellular membranes. B, VSMCs were transiently transfected with annexin 11⅐GFP and imaged as in panel A before and after the addition of 50 ng/ml PDGF-BB. The images show PDGF-dependent relocalization of cytosolic annexin 11⅐GFP to the plasma membrane and cytoplasmic vesicles. C, VSMCs were exposed to either 2.5 M ionomycin or 50 ng/ml PDGF for 15 min or left untreated (control), and annexin 11 was immunoprecipitated from total cell extracts. Following SDS-PAGE and Western transfer, the blot was probed first with an antibody to phosphotyrosine and, after stripping, to an antibody to annexin 11 to confirm equal protein loadings.
FIG. 4. Live cell imaging of annexin 11⅐GFP fusions in A431 cells. A431 cells were transiently transfected with full-length human annexin 11 coding sequence fused to GFP either at the C terminus (anx11⅐GFP) or at the N terminus (GFP⅐anx11), with the annexin 11 proline-, tyrosine-, and glycine-rich (PYG) N-terminal domain fused to GFP (N-ter⅐GFP), or with the annexin 11 C-terminal core domain fused to GFP (GFP⅐C-ter). Cells expressing the fusion constructs were imaged before and after the addition of 2.5 M ionomycin, both in Ca 2ϩ -containing and Ca 2ϩ -free external medium. After stimulation with ionomycin, the full-length annexin 11 fusion proteins relocate primarily to the nuclear envelope (pink arrowheads) and can also be detected with a punctate pattern at the cytoplasm/plasma-membrane, whereas the fusion protein containing the annexin 11 N-terminal domain does not respond to ionomycin treatment. The annexin 11 C-terminal core domain containing the four conserved annexin repeats relocates to intracellular membranes but not the nuclear envelope. These changes in localization occur both in the presence and absence of external Ca 2ϩ (pink/yellow arrowheads), suggesting that mobilization of Ca 2ϩ from intracellular stores is sufficient to elicit this response. ciated fluorescence. Similar results were obtained regardless of the presence or absence of extracellular Ca 2ϩ , indicating that mobilization of Ca 2ϩ from intracellular stores alone is sufficient to elicit these responses. In contrast, N-ter⅐GFP completely failed to respond to ionomycin, and although GFP⅐C-ter exhibited sensitivity to Ca 2ϩ by relocating to intracellular membranes, there was no evidence of preferential translocation to the nuclear envelope. These results demonstrate that whereas the C-terminal core domain of annexin 11 is essential to confer Ca 2ϩ sensitivity, the N-terminal domain determines the specificity for the nuclear envelope in the Ca 2ϩ -dependent relocalization of annexin 11. Furthermore, the images obtained using anx11⅐GFP and GFP⅐anx11 demonstrate that GFP may be fused to either end of the annexin 11 molecule without interfering with normal responses to changes in intracellular Ca 2ϩ concentration.
Although these results reveal a potential mechanism whereby the annexin 11-S100A6 interaction might be regulated, we were interested to explore the possibility of a cell cycle-dependent interaction. In A431 cells at early prophase we obtained evidence in support of this idea (Fig. 5A). At this point in the cell cycle the nuclear pool of annexin 11 apparently relocates, at least in part, to the nuclear envelope. The annexin 11 remaining in the nucleus becomes distinct from the condensed DNA content, which contrasts with the picture obtained for interphase cells (Fig. 1A). We also observed, in cells further into prophase, that annexin 11 becomes associated with invaginations or folds in the nuclear envelope where it co-localizes with S100A6. The pattern of immunostaining for annexin 11 and S100A6 in these late prophase cells is similar to that reported for other nuclear envelope proteins, in particular LAP2 (23). We therefore co-stained late prophase cells for annexin 11 and LAP2 (Fig. 5B), which enabled us to confirm that annexin 11 is indeed associated with microtubule-induced folds that serve an initiating role in nuclear envelope breakdown (24). S100A6 immunostaining was virtually undetectable in A431 cells from metaphase to telophase (not shown), indicating that FIG. 5. Dynamic association of annexin 11 with the nuclear envelope during cell division. A431 cells were synchronized in mitosis, fixed, and stained for annexin 11, DNA, and S100A6 (A) or the inner nuclear membrane protein LAP2 (B) during prophase. Annexin 11 invades the nuclear envelope at discrete foci in early prophase (panel A, pink arrowheads) and concentrates at folds in the nuclear envelope where it colocalizes with S100A6 (panel A, yellow arrowheads) and with LAP2 (panel B, pink arrowheads). A431 cells were synchronized in mitosis, fixed, and stained for annexin 11, DNA, and LAP2. The images show cells engaged in nuclear envelope reassembly during late telophase and recently divided cells in early interphase (C). Annexin 11 is enriched at the nascent nuclear envelope of cells undergoing cytokinesis, together with LAP2 (top row, white arrowheads), before shuttling back into the nucleus where it resides during interphase (bottom row). S100A6 protein levels drop during mitosis. This observation is in agreement with previous studies that showed the level of expression of S100A6 is regulated in a cell cycle-dependent manner, peaking in the transition from G 0 -to S-phase in smooth muscle cells (18). We therefore immunostained A431 cells in telophase for annexin 11 and LAP2 (Fig. 5C). In late telophase, annexin 11 and LAP2 again co-localize, this time in the reforming nuclear envelope. However, while LAP2 remains clearly associated with the nuclear envelope as cells progress into early interphase, annexin 11 becomes concentrated within the nucleus. These observations show that annexin 11 displays a dynamic and biphasic interaction with the nuclear envelope, first during nuclear envelope breakdown and again during reassembly of the nuclear envelope.

DISCUSSION
Annexin 11 is unusual within the annexin family in being a predominantly nuclear protein. This is in contrast to its most extensively characterized in vitro ligand, S100A6, which is shown here and in other studies (19) to be localized to the nuclear envelope. Because the annexin literature contains several unverified reports of interactions between annexins and putative intracellular ligands, it is important to establish which, if any, of these occur in vivo because such interactions are likely to have functional consequences. A more recently identified ligand of annexin 11, namely the penta-EF-hand protein ALG-2 (apoptosis-linked gene 2) (25) is expressed in the nucleus (26), though ALG-2 also binds annexin 7, peflin, and AIP1/Alix (25,27). The binding site for ALG-2 on annexins 7 and 11 is located in the long PYG domain at the N terminus of each annexin, a domain that is also present in AIP1/Alix. Because the binding sites for ALG-2 and S100A6 on annexin 11 are therefore both located in the N terminus, these observations raise the possibility that, depending on subcellular localization and the availability of free Ca 2ϩ , S100A6 and ALG-2 may compete for and differentially modulate the properties of annexin 11. However, prior to consideration of these hypothetical possibilities it is necessary to establish a prerequisite for the annexin 11-S100A6 interaction, namely that the two proteins occupy a single subcellular compartment.
We show here that whereas annexin 11 and S100A6 are unlikely to interact in vivo in interphase cells under resting conditions, they are both present at the nuclear envelope upon elevation of intracellular calcium. Interestingly, despite the fact that annexin 11 and S100A6 are both Ca 2ϩ -binding proteins, annexin 11 alone manifested its Ca 2ϩ sensitivity as a change in localization. Moreover, the N-terminal domain of annexin 11, known to contain the S100A6-binding site, was necessary for the relocalization of annexin 11 to the nuclear envelope in ionomycin-treated A431 cells, reinforcing the idea that the association of annexin 11 with the nuclear envelope in the presence of calcium may occur via a direct interaction with S100A6. The finding that ionomycin treatment leads not only to the relocalization of annexin 11 to the nuclear envelope but also to its tyrosine phosphorylation suggests that a Ca 2ϩ -activated tyrosine kinase may be involved in the specific association of annexin 11 with the nuclear envelope. Thus, in two distinct cell types we have shown that ionophore-mediated translocation of annexin 11 to the nuclear envelope is accompanied by annexin 11 tyrosine phosphorylation, whereas growth factor stimulation has no effect on either parameter. The failure of EGF and PDGF-BB to induce tyrosine phosphorylation of annexin 11 is consistent with a previous report in which it was shown that stimulation of VSMCs with PDGF-BB induced tyrosine phosphorylation of annexin 11 only when added in conjunction with peroxovanadate (21). Nevertheless, these observations raise the question why EGF and PDGF fail to reproduce the responses observed with ionomycin, given that both growth factors also mobilize Ca 2ϩ . The most likely explanation for this apparent discrepancy is that the nucleus is 'buffered' from growth factor-mediated changes in cytosolic Ca 2ϩ but not from the effects of ionophore (28, 29).
The role of annexin 11 at the nuclear envelope during nuclear envelope breakdown and nuclear envelope reassembly is not known, but our findings illustrate a physiological situation in which annexin 11 and S100A6 might interact in vivo. The enrichment of annexin 11 at the nuclear envelope invaginations created by microtubules during nuclear envelope breakdown (24) is of particular interest because it could indicate a role for annexin 11 as a link between the nuclear envelope and the microtubules at these sites. Furthermore, several reports have established that calcium is required for nuclear envelope breakdown in eukaryotic cells (30,31) and that spontaneous cell cycle-dependent Ca 2ϩ transients are generated by pulsatile fluxes in inositol trisphosphate (32). Calcium is also required for lamin B degradation (33), indicating a role for Ca 2ϩ -binding proteins in these events. As a cell cycle-regulated nuclear Ca 2ϩbinding protein, it is clearly possible that via its interaction with S100A6 annexin 11 has a role in membrane deconstruction during nuclear envelope breakdown. In contrast, the reappearance of annexin 11 during nuclear envelope reassembly may simply reflect passive transit of the protein to its notionally inactive nucleoplasmic location. If the putative annexin 11⅐S100A6 complex possesses such destructive activity, it would make sense that S100A6 protein levels are almost undetectable at the point in the cell cycle of nuclear envelope reassembly, thereby eliminating any possible and untimely deconstructive activity of the annexin 11⅐S100A6 complex.
In conclusion, we have demonstrated that elevation of intracellular Ca 2ϩ concentration and cell cycle-dependent events are independently capable of driving the translocation of annexin 11 from the nucleoplasm to the nuclear envelope. Our observations support the idea that annexin 11 and S100A6 interact in vivo and provide evidence for a role for annexin 11 in the regulation of nuclear envelope dynamics.