Ca2+-dependent Translocation of the Calcyclin-binding Protein in Neurons and Neuroblastoma NB-2a Cells*

The calcyclin-binding protein (CacyBP) binds calcyclin (S100A6) at physiological levels of [Ca2+] and is highly expressed in brain neurons. Subcellular localization of CacyBP was examined in neurons and neuroblastoma NB-2a cells at different [Ca2+] i . Immunostaining indicates that CacyBP is present in the cytoplasm of unstimulated cultured neurons in which resting [Ca2+] i is known to be ∼50 nm. When [Ca2+] i was increased to above 300 nm by KCl treatment, the immunostaining was mainly apparent as a ring around the nucleus. Such perinuclear localization of CacyBP was observed in untreated neuroblastoma NB-2a cells in which [Ca2+] i is ∼120 nm. An additional increase in [Ca2+] i to above 300 nm by thapsigargin treatment did not change CacyBP localization. However, when [Ca2+] i in NB-2a cells dropped to 70 nm, because of BAPTA/AM treatment, perinuclear localization was diminished. Ca2+-induced translocation of CacyBP was confirmed by immunogold electron microscopy and by fluorescence of NB-2a cells transfected with an EGFP-CacyBP vector. Recombinant CacyBP can be phosphorylated by protein kinase C in vitro. In untreated neuroblastoma NB-2a cells, CacyBP is phosphorylated on a serine residue(s), but exists in the dephosphorylated form in BAPTA/AM-treated cells. Thus, phosphorylation of CacyBP occurs in the same [Ca2+] i range that leads to its perinuclear translocation.

covered in the cytosolic fraction of Ehrlich ascites tumor cells, mouse brain, and spleen (14). The cDNA clone of CacyBP was isolated from mouse brain library and sequenced (15). The recombinant CacyBP interacts with calcyclin in vitro at micromolar Ca 2ϩ concentrations indicating that this interaction may indeed occur physiologically. The region of CacyBP (amino acids 178 -229), which binds calcyclin, has been identified, and the dissociation constant of the complex has been measured (16). Northern and Western blots have shown that CacyBP is expressed at the highest level in the mouse and rat brain, and immunohistochemistry performed on rat brain slices revealed that CacyBP is mainly present in neurons of the cerebellum, hippocampus, and cortex (17).
When first elucidated, the sequence of CacyBP displayed no apparent similarity to any other known protein sequence; thus, an assignment of its function was not possible. Two independent reports showed that the level of CacyBP is increased upon erythropoietin receptor activation (18,19) and that this increase parallels an increase in c-myc and dpp-1 transcriptional activity via the JAK2 pathway (19). These observations suggest that CacyBP might be involved in a signaling pathway(s) activated by erythropoietin in erythroid cells and in neuronal cells in which a high level of erythropoietin receptor was also found. Recently, a protein called SIP (siah-1 interacting protein), a component of the ␤-catenin degradation pathway, has been reported (20). The predicted human SIP protein shows 93% sequence identity with mouse CacyBP suggesting that it is the human homolog of CacyBP. Because Ca 2ϩ -dependent phenomena related to the activity of the ubiquitin/26 S proteasome complex were described (21), interaction of CacyBP with an EF-hand calcium-binding protein raises the possibility that CacyBP may function via Ca 2ϩ -dependent interactions in ubiquitination and the protein degradation process. Therefore, in this work we have studied the intracellular localization of CacyBP under different [Ca 2ϩ ] i , using primary cultures of rat cortical neurons and cultured neuroblastoma NB-2a cells. Using both immunostaining and transfection of NB-2a cells with an EGFP-CacyBP vector, we have shown that cellular localization of CacyBP is modulated by changes in [Ca 2ϩ ] i . We also found that phosphorylation of this protein on a serine residue(s) occurs within the same range of [Ca 2ϩ ] i as that which stimulates its perinuclear translocation.
Primary Culture of Cortical Neurons and KCl Treatment-Cortical neurons were prepared from newborn Harlan Sprague-Dawley rats and cultured at a density of 1500 -2000 cells/mm 2 in basal medium Eagle supplemented with 10% heat-inactivated bovine calf serum (Hyclone, Logan, UT). Cytosine arabinoside (2.5 M) was added on the second day after seeding to inhibit the proliferation of non-neuronal cells.
KCl treatment of rat cortical neurons was performed on cells cultured for 5-6 days. Neurons were incubated for 10 min with KCl, which was added to the medium to a final concentration of 30 mM. After that time cells were fixed and prepared for immunocytochemistry. In other experiments cells treated with KCl as above were incubated in the medium without KCl for 2 h and then fixed and used for immunocytochemistry.
Culture of Neuroblastoma NB-2a and PC12 Cells and Extract Preparation-Mouse neuroblastoma cells NB-2a were maintained in MEM supplemented with 10% fetal bovine serum, 25 mM bicarbonate, penicillin (100 g/ml), and streptomycin (100 g/ml). PC12 cells were grown in RPMI 1640 containing 10% horse serum, 5% fetal bovine serum, penicillin (50 units/ml), and streptomycin (50 g/ml). All cultures were maintained in the presence of 5% CO 2 at 37°C. The media were changed every 3 days and cells were passaged when confluent.
To prepare the protein extracts for Western blot analysis, cells were harvested and washed twice with PBS. The cells were homogenized mechanically using a syringe with a needle (26-gauge; 0.45 ϫ 12) 20ϫ in PBS containing 1 mM EDTA and the following protease inhibitors: leupeptin (10 mg/liter), aprotinin (5 mg/liter), soybean trypsin inhibitor (20 mg/liter), and phenylmethylsulfonyl fluoride (1 mM). The extracts were centrifuged for 45 min at 4°C at 12,000 rpm in an Eppendorf centrifuge. Protein concentration was estimated by the Bradford procedure (BioRad) with bovine serum albumin as a standard.
Construction of the EGFP-CacyBP Expression Vector-PCR was carried out with CacyBP cDNA in pBluescript vector (16) as a template with the forward primer, 5Ј-GGATCGGATCCATGGCTTCCGTTTTG-GAAGAG-3Ј and the reverse primer, 5Ј-GAGACGAATTCTCAT-CAAAATTCCGTGTCTTC-3Ј (restriction enzyme recognition sites in bold). Pfu DNA polymerase was used in 30 cycles of PCR with each cycle consisting of 1 min at 94°C; 1 min at 50°C; and 2 min at 72°C. The PCR products were digested with BamHI and EcoRI restriction enzymes, purified and ligated with pEGFP-C1 vector (CLONTECH), and linearized with BglII and EcoRI restriction enzymes. The ligation reaction mixture was used to transform Escherichia coli cells (TOP10FЈ, Invitrogen). Potential clones were screened by colony PCR, and the presence of the insert was confirmed by restriction analysis. The sequence of the insert was verified by automated DNA sequencing.
Transfection of Neuroblastoma NB-2a Cells-Two constructs were used for transfection experiments: one encoding the EGFP-CacyBP fusion protein and a second encoding only EGFP. Both genes were under the control of the cytomegalovirus promoter. Plasmids were purified from E. coli using a Qiagen Plasmid Kit MIDI according to the manufacturer's protocol. DNA transfection in NB-2a cells was performed by the calcium phosphate precipitation technique according to standard procedures (22). Cells grown on coverslips were exposed for 16 h to 2 g of plasmid DNA per coverslip.
Loading the Cells with Thapsigargin or BAPTA/AM-Neuroblastoma NB-2a cells were plated onto poly-L-lysine-coated coverslips and cultured for 2 days in MEM at 37°C in 5% CO 2 . Cells were washed twice for 5 min in PBS and incubated with 0.2 M thapsigargin dissolved in buffer containing 20 mM Hepes, pH 7.4, 137 mM NaCl, 2.7 mM KCl, 1 mM Na 2 HPO 4 , 25 mM glucose, 1 mM MgCl 2 , 1% bovine serum albumin, and 2 mM CaCl 2 for 2 min at room temperature or with 5 M BAPTA/AM and 1 mM EGTA for 30 min at 30°C. Cells treated with BAPTA/AM were then washed two times and incubated in the buffer described above containing 1 mM EGTA for 30 min at 30°C. Cells transfected with CacyBP-EGFP were treated in the same way with thapsigargin or BAPTA/AM, and EGFP fluorescence was analyzed under a Nikon Optiphot-2 microscope.
Measurement of [Ca 2ϩ ] i -The cytoplasmic level of Ca 2ϩ was examined in neuroblastoma NB-2a cells (untreated or treated with thapsigargin or BAPTA/AM) using a video imaging system (MagiCal, Applied Imaging, data processing using Tardis V8.0 (Joyce Loebl)) as described by Barań ska et al. (23). [Ca 2ϩ ] i values were calculated according to Grynkiewicz et al. (24).
Immunocytochemistry-Immunocytochemistry experiments were performed on neurons (unstimulated or stimulated with KCl) and neuroblastoma NB-2a cells (untreated or treated with thapsigargin or BAPTA/AM). Neurons were plated onto laminin-coated coverslips whereas neuroblastoma NB-2a cells were plated onto poly-L-lysinecoated coverslips. Cells were fixed with 3% paraformaldehyde in PBS (pH 7.4) for 20 min at room temperature. The coverslips were washed with PBS, incubated with 50 mM NH 4 Cl in PBS for 10 min to quench the remaining aldehyde groups (25), and permeabilized for 4 min with 0.1% Triton X-100 in PBS. The cells were washed twice with PBS, incubated with 1% bovine serum albumin in PBS for 1 h, and after that incubated with anti-CacyBP antibodies (1:800). After washing (three times for 10 min in PBS), cells were incubated with fluorescein isothiocyanateconjugated anti-rabbit antibodies (1:200) (Jackson ImmunoResearch Laboratories, Inc.) and mounted on glass slides with a mixture of glycerol and polyvinyl alcohol containing DABCO (1,4-diazobicylo-[2,2,2,]-octane). For control experiments cells were incubated with preimmune serum. Cells were analyzed under a Nikon Optiphot-2 microscope.
Quantitative Analysis of CacyBP Redistribution-Digitalized images of cells were analyzed with Quantity One software (BioRad). In each cell, three lines of 3 pixels in width were drawn across the center of the cell at a constant distance of 0.5 m from each other (see also Fig. 1, D-F), and line intensity profiles of CacyBP labeling were collected. The fluorescence intensity values were within a linear range. Next, the fluorescence intensities in the perinuclear region (I n ) and in the cytoplasm (I c ) at a constant distance from the perinuclear region (2.7 m) were determined. This distance was chosen because at this point a maximal decrease in fluorescence intensity in the cytoplasm was observed. The ratio I n /I c was calculated to express the changes in CacyBP concentration in the perinuclear region. At least 10 cells from two independent experiments were analyzed in each variant.
Electron Microscopy-Neuroblastoma NB-2a cells were fixed for 1 h at room temperature in 3% paraformaldehyde, 0.5% glutaraldehyde in PBS buffer (26). Then the cells were rinsed in PBS four times (5 min each) followed by centrifugation for 5 min (350 ϫ g). Cells were dehydrated in ethanol and propylene oxide, embedded in LR White (Polyscience) and polymerized for 72 h at 56°C.
After thin sectioning, samples were collected on the carbonformavar-coated nickel grids and incubated for 30 min in 1% bovine serum albumin, 0.1% Tween/PBS as a nonspecific blocking agent and labeled with antibodies against CacyBP (1:250) overnight at room temperature. After washing with 1% bovine serum albumin/0.1% Tween/ PBS, the cells were incubated in anti-rabbit IgG (1:20) conjugated to colloidal gold (5 nm) for 1 h at room temperature. After extensive washing (bovine serum albumin/Tween/PBS: 6ϫ 5 min, PBS: 4ϫ 5 min, H 2 O: 2ϫ 5 min), sections were stained with uranyl acetate for 30 min at room temperature. The sections were observed in a JEM 1200 EX electron microscope.
Phosphorylation of CacyBP in Vitro and in Vivo-Phosphorylation of CacyBP in vitro was performed on purified, recombinant protein expressed in E. coli (16). The reaction mixture for phosphorylation by protein kinase C (final volume, 50 l) contained 10 mM Hepes, pH 7.5, 5 mM MgCl 2 , 50 mM NaCl, 0.05 mM CaCl 2 , 0.7 g/l ␣-L-phosphatidyl-L-serine, 80 nM PMA, 0.1 mM dithiothreitol, 70 ng of kinase preparation (Calbiochem), and 0.1 mM ATP. Preincubation was carried out for 20 min at 30°C. After that the reaction was initiated by addition of the substrate (ϳ2 g of CacyBP) and [␥Ϫ 32 P]ATP at a final concentation of 0.2 mM. The reaction was carried out for 30 min at 37°C. The reaction mixture for protein kinase A (catalytic subunit from Sigma) contained 50 mM Hepes, pH 7.5, 10 mM MgCl 2 , 0.1 mM ATP, and 300 ng of kinase preparation. In this case preincubation and reaction were carried out for 30 min at 30°C. Reactions were terminated by the addition of 5 l of 4ϫ SDS sample buffer. Phosphorylated proteins were separated on 15% (w/v) SDS gels. Gels were dried and subjected to autoradiography for 18 -34 h with an Amersham Biosciences film at Ϫ70°C. Autophosphorylation of kinase preparations was performed as described above but without adding the substrate.
Phosphorylation of CacyBP in NB-2a cells was examined in untreated or BAPTA/AM-treated NB-2a cells. Cells were washed in PBS and harvested and sonicated (Branson ultrasonicator) in buffer containing 20 mM Tris-HCl, pH 7.5, 8 mM MgCl 2 , 150 mM NaCl, 0.2 mM EGTA, 1% Nonidet P-40. Extracts were centrifuged for 15 min at 4°C at 12,000 rpm in an Eppendorf centrifuge. Supernatants were used for the immunoprecipitation assay after adding the protease inhibitors (10 mg/ liter leupeptin, 5 mg/liter aprotinin, 20 mg/liter soybean trypsin inhibitor, 1 mM phenylmethylsulfonyl fluoride), and phosphatase inhibitors (0.3 mM okadaic acid, 200 M Na 3 VO 4 , 5 mM NaF). First, the solutions were incubated with protein A-Sepharose for 1 h at 4°C (preclearance). The unbound fractions were incubated with serum-containing antibodies against CacyBP for 1.5 h at 4°C and then for 1 h at 4 o with a new portion of protein A-Sepharose. The resin was washed three times in a buffer containing 20 mM Tris-HCl, pH 7.5 and 150 mM NaCl, twice in a buffer containing 20 mM Tris-HCl, pH 7.5, and 500 mM NaCl, and finally in 20 mM Tris-HCl at pH 7.5. All buffers used were supplemented with protease and phosphatase inhibitors. The resin containing bound proteins was solubilized in SDS sample buffer, boiled for 5 min at 95°C, and applied onto the SDS-polyacrylamide gel. Phosphorylation of CacyBP was analyzed using Western blot techniques with monoclonal antibodies against phosphorylated serine residues (Alexis Corp.) as described in the original protocols. For determination of the ATP effect, the supernatants were preincubated for 20 min at 37°C with PMA and CaCl 2 at final concentrations of 8 nM and 0.5 mM, respectively. Then ATP was added to a final concentration of 300 M, and the solution was incubated for 30 min at 37°C. The reaction was inhibited by adding SDS and Nonidet P-40 to a final concentration of 0.1%. Protease inhibitors were subsequently added, and the immunoprecipitation assay was performed as described above.
Other Methods-Polyclonal antibodies, Northern blots, electrophoresis, and immunoblotting were performed as described by Jastrzebska et al. (17) with the minor modifications described below. New polyclonal antibodies were prepared against recombinant CacyBP cloned in the pET30 vector (16). Control experiments performed on NB-2a cells using preimmune serum or anti-CacyBP serum saturated with CacyBP confirmed the high quality of this serum (not shown). For Northern blots total RNA was prepared from cells using an RNA purification kit (Qiagen). For Western blots antibodies against CacyBP were diluted 1:800.

Localization of CacyBP in Cultured Neurons at Different
[Ca 2ϩ ] i -To establish the effect of [Ca 2ϩ ] i on the intracellular distribution of CacyBP, we analyzed primary cultures from rat cortical neurons before and after stimulation. Unstimulated cells, KCl-activated cells, and cells in which KCl was washed out after stimulation were probed by immunocytochemistry using anti-CacyBP antibodies. In unstimulated neurons Ca-cyBP was distributed throughout the cytoplasm (Fig. 1, A and  D); this staining pattern was observed in the majority of the cells (Table I). The level of [Ca 2ϩ ] i in rat cortical neurons has been established previously to be ϳ50 nM in resting cells and above 300 nM upon KCl activation (27). We observed that activation of cultured cortical neurons by KCl induces CacyBP translocation mainly to the perinuclear region (Fig. 1, B and E). The KCl effect seems to be physiologically significant because about 85% of the cells showed increased CacyBP immunostaining in this region ( Table I). The Ca 2ϩ -induced translocation of CacyBP to the nuclear envelope was reversible. After removing KCl from the medium, the localization of CacyBP in the cytoplasm was restored in the majority of cells (Fig. 1, C and F and Table I).
We next performed a quantitative analysis of the CacyBP distribution within the cell. For this purpose, line intensity profiles of CacyBP across the cell were obtained (Fig. 1, D-F). Along these lines the fluorescence intensities in the perinuclear region (I n ) and in the cytoplasm (I c ) at a constant distance from the nuclear envelope were determined. In unstimulated neurons the ratio I n /I c was 1.09 ( Fig. 1D and Table II), indicating the lack of significant enrichment of the CacyBP in the perinuclear region. After stimulation of neurons and an increase in [Ca 2ϩ ] i the ratio I n /I c was 1.58 ( Fig. 1E and Table II), indicating the accumulation of CacyBP in the perinuclear region. After removing the stimulus, the ratio was 1.18, indicating that Ca 2ϩ -dependent translocation of CacyBP is reversible (Fig. 1F and Table II).

Localization of CacyBP in Neuroblastoma NB-2a Cells at Different [Ca 2ϩ ] i -
To learn more about the affect of Ca 2ϩ on CacyBP intracellular localization, we searched for cell lines expressing high levels of CacyBP. The level of this protein was examined in two cell lines of neuronal origin: PC12 cells and neuroblastoma NB-2a cells. Northern blotting with a fulllength cDNA probe (Fig. 2, A and B) showed that CacyBP mRNA was present at a much higher level in neuroblastoma NB-2a than in PC12 cells. Also, a much more intense immunoreactive protein band representing CacyBP was detected by Western blotting analysis from the extracts of neuroblastoma NB-2a, compared with PC12 cells (Fig. 2C). Thus, the NB-2a cells were chosen for further studies.
Neuroblastoma NB-2a cells were treated with different agents affecting [Ca 2ϩ ] i and were then fixed and stained with antibodies against CacyBP. In resting cells in which [Ca 2ϩ ] i was shown to be 120 nM (Fig. 3), immunostaining was observed mainly in the perinuclear region, similar to activated neurons (Fig. 4A). As shown in Table III more than 80% of the cells exhibited increased immunostaining in the perinuclear region. The treatment of cells with thapsigargin, which led to an increase of [Ca 2ϩ ] i to above 300 nM (Fig. 3), did not affect the staining pattern (Fig. 4B) nor did it affect the percentage of cells showing this staining pattern (Table III).
Additional evidence for changes in CacyBP distribution was obtained from experiments in which neuroblastoma NB-2a cells were transfected with an EGFP-CacyBP expression vector. In these cells the fluorescence of the EGFP-CacyBP fusion protein was visible mainly in the perinuclear region (Fig. 4D), and its distribution was not affected by thapsigargin treatment (Fig. 4E). The percentage of untreated cells containing high levels of CacyBP in the perinuclear region, detected by EGFP-CacyBP fluorescence, was about 94% and of thapsigargin-treated cells, about 86% (Table III).
Treatment of NB-2a cells with cell-permeant Ca 2ϩ -chelator BAPTA/AM lowered [Ca 2ϩ ] i to ϳ70 nM (Fig. 3), reduced immunostaining of endogenous CacyBP (Fig. 4C), and decreased   EGFP-CacyBP fluorescence in the perinuclear region (Fig. 4F). Moreover, CacyBP was no longer seen as a ring around the nuclei but rather was relocated into the cytoplasm. Only about 5% of the cells exhibited the immunostaining and EGFP-Ca-cyBP fluorescence in the perinuclear region after BAPTA/AM treatment (Table III).
To confirm these observations, quantitative analysis of Ca-cyBP distribution in NB-2a cells was performed using the same approach that was used for cortical neurons. In untreated NB-2a cells, the ratio I n /I c was 1.44 and 1.32 for cells stained with anti-CacyBP and cells expressing EGFP-CacyBP, respectively (Table IV). After treatment with BAPTA/AM the ratio I n /I c was 0.96 and 0.99 for cells stained with anti-CacyBP and cells expressing EGFP-CacyBP, respectively (Table IV). These data confirm the accumulation of CacyBP in the perinuclear region of untreated NB-2a cells and translocation of CacyBP toward the cytoplasm after decreasing [Ca 2ϩ ] i . Subcellular Localization of CacyBP by Immunogold Labeling in Electron Microscopy-Untreated and BAPTA/AM-treated neuroblastoma NB-2a cells were analyzed by the postembedding immunogold technique. Cells were fixed, probed with antibodies against CacyBP, and processed for detection using secondary antibodies conjugated to gold particles (Fig. 5). The specificity of immunolocalization of CacyBP was checked in the control experiment in which primary antibodies were omitted; no gold particles were seen (Fig. 5A). A positive reaction was observed in the cytoplasm and on both sides of the nuclear envelope of untreated neuroblastoma cells (Fig. 5, B and C). In BAPTA/AM-treated cells, CacyBP was seen in the cytoplasm but immunolocalization in the perinuclear region was significantly decreased (Fig. 5D).

Phosphorylation of CacyBP in Neuroblastoma NB-2a Cells-
One possible mechanism for Ca 2ϩ -dependent localization of CacyBP in the perinuclear region is protein phosphorylation. Analysis of the CacyBP sequence indicated several potential sites that can be phosphorylated by different kinases. To verify that CacyBP can be phosphorylated in vitro, recombinant purified CacyBP was incubated with commercially available preparations of protein kinase C and protein kinase A. As shown in Fig. 6A, CacyBP was phosphorylated by protein kinase C (lane 1) but not by protein kinase A (lane 3). Other bands seen on the autoradiogram come from autophosphorylation of kinase preparations (lanes 2 and 4).
To examine if CacyBP might exist in a phosphorylated form in vivo, the extracts of untreated (Fig. 6B, lanes 1 and 2) and BAPTA/AM-treated NB-2a cells (Fig. 6B, lanes 3 and 4) were immunoprecipitated with antibodies against CacyBP, and the precipitated proteins were analyzed by anti-phosphoserine antibodies. In the case of untreated NB-2a cells, CacyBP phosphoserine immunoreactivity was seen (Fig. 6B, lane 1). This immunoreactivity was not affected by incubation of the extracts with ATP (Fig. 6B, lane 2). These results indicate that CacyBP is phosphorylated on serine residue(s) in untreated cells. In cells treated with BAPTA/AM, which exhibit lower levels of [Ca 2ϩ ] i , the immunoprecipitated CacyBP contained no phosphoserine immunoreactivity (Fig. 6B, lane 3). However, when ATP and Ca 2ϩ were added to the extract of BAPTA/AMtreated cells prior to immunoprecipitation, the serine residue(s) of CacyBP reacted with anti-phosphoserine antibodies to a similar extent as that observed in untreated cells (Fig. 6B,  lane 4). These results show that CacyBP is not phosphorylated at low [Ca 2ϩ ] i . Thus, we find that CacyBP phosphorylation

DISCUSSION
In resting cultured cortical neurons CacyBP was distributed throughout the cytoplasm, but after stimulation that led to an increase of [Ca 2ϩ ] i , the immunoreactivity was mainly visible as a ring around the nucleus. The intracellular Ca 2ϩ level in cultured rat cortical neurons is ϳ50 nM and upon KCl stimulation increases to above 300 nM (27). Similar low levels of resting [Ca 2ϩ ] i were measured in hippocampal neurons (80 nM) (28) and hypothalamic neurons (54 or 69 nM) (29,30). Thus, an increase in [Ca 2ϩ ] i to above 300 nM upon neuron activation induces the translocation of CacyBP to the perinuclear region. The Ca 2ϩ -dependent change of CacyBP localization was also observed in neuroblastoma NB-2a cells. Transformed cells often have higher basic [Ca 2ϩ ] i than normal cells (31). We confirmed this by establishing that in untreated neuroblastoma NB-2a cells [Ca 2ϩ ] i is about 120 nM. In these cells CacyBP was mainly present in the perinuclear region. However, in cells treated with BAPTA/AM, in which [Ca 2ϩ ] i dropped to 70 nM, a case similar to the [Ca 2ϩ ] i of resting neurons, CacyBP immunoreactivity was no longer visible as a ring around the nucleus. All these data indicate that CacyBP is present throughout the cytoplasm at low [Ca 2ϩ ] i and is translocated into the perinuclear region when [Ca 2ϩ ] i is increased.
What mechanism might be responsible for Ca 2ϩ -dependent translocation of CacyBP? There is a possibility that some posttranslational modifications of CacyBP might regulate its localization within the cell as is the case with some other proteins. It has been shown, for instance, that localization of AHNAK, a giant protein originally identified in neuroblastoma, changes upon phosphorylation by protein kinase B (32). Another example is the protein TFAF2/SNX6. This protein was translocated from the cytoplasm to the nucleus by Pim-1 kinase phosphorylation (33). Also, nuclear translocation and accumulation was described for wild type p53. In this case nuclear localization was dependent on protein kinase C activity (34).
Theoretical analysis of the CacyBP sequence showed that this protein has potential phosphorylation sites for protein kinase C. In agreement with this prediction we showed that recombinant CacyBP is indeed phosphorylated by protein kinase C in vitro. Moreover, we found that CacyBP is phosphorylated on serine residue(s) in untreated neuroblastoma NB-2a cells and exists in a dephosphorylated form in the cells treated with BAPTA/AM. In other words, reversible serine phosphorylation occurs within the same [Ca 2ϩ ] i range in which CacyBP translocation to the perinuclear region takes place. Because calcyclin was found to be associated with the nuclear envelope in a calcium-dependent manner (35), it was interesting to see if calcyclin might regulate phosphorylation of CacyBP by direct interaction with this protein. It has been reported that S100B inhibits phosphorylation of p53 by direct interaction and not by influencing the kinase activity (36) and that S100C inhibits actin-activated myosin ATPase in the same way (37). Results obtained from experiments in vitro show that calcyclin has no influence on CacyBP phosphorylation by protein kinase C. 2 The results presented in this report show that CacyBP distribution within the cell is modulated by changes in [Ca 2ϩ ] i and that this 2 A. Filipek, B. Jastrzebska, and J. Kuźnicki, unpublished results.  1 and 2) or BAPTA/ AM-treated cells (lanes 3 and 4) using anti-CacyBP antibodies. Lanes 2 and 4 show how adding of ATP, PMA, and Ca 2ϩ before immunoprecipitation affected CacyBP phosphorylation. Immunoprecipitated proteins were separated on a 15% SDS gel, blotted onto nitrocellulose filter, and probed with antibodies against phosphoserine. phenomenon might be stimulated or regulated by a phosphorylation process.
Our electron microscopy studies reveal immunogold staining of CacyBP in the nucleus of NB-2a cells. In fact, Fig. 5C shows CacyBP immunogold reactivity within and on both sides of the nuclear envelope, which suggests that the CacyBP transfer to the nucleus has been trapped in this preparation. In fact, nuclear localization is not surprising because analysis of the CacyBP sequence, performed using the BLAST server software, identified a nuclear localization signal (NLS) between amino acids 144 and 160. The relationship between the putative NLS-mediated localization and Ca 2ϩ -dependent translocation of CacyBP remains an unresolved issue, as does the physiological significance of these observations. Some insight into the function of CacyBP could be obtained from a recent study showing that a human protein termed SIP, which is 93% identical to mouse CacyBP, serves as a molecular bridge between Siah-1 and Skp1 proteins, components of a ␤-catenin ubiquitin ligase system. It has been known that different regulatory complexes and subpopulations of proteasomes have a different distribution within mammalian cells (38). For instance 20 S proteasomes and their 19 S regulatory complexes were found in nuclear, cytosolic, and microsomal fractions. Our results are fully consistent with these observations. Together, these data provide evidence for the function of CacyBP in a novel ubiquitinylation pathway.