Phosphorylation-related accumulation of the 125k nuclear matrix protein mitotin in human mitotic cells

Considerable experimental evidence shows that in eukaryotes the G₂ phase events preparing the cell for mitosis involve the phosphorylation of several nuclear proteins (Adlakha et al. 1985; Halleck et al. 1987; Lee and Nurse, 1988; Dunphy and Newport, 1988), including histones, Hl in particular (Gurley et al. 1978; Meijer and Pondaven, 1988), nuclear envelope lamins (Gerace and Blobel, 1980; Ottaviano and Gerace, 1985) and other nuclear matrix proteins (Henry and Hodge, 1983). The extensive protein phosphorylation is related to the selective activation of several protein kinases in premitotic and mitotic cells (Langan, 1978; Matthews and Huebner, 1985; Dessevet al. 1988). Recently, a p34 protein kinase, homologous to the protein kinase encoded by the yeast cdc2.⁺/CDC28 gene, has been identified in human cells (Lee and Nurse, 1987; Draetta et al. 1987) and shown to be involved in a cascade of protein phosphorylations obviously required for mitosis (Draetta and Beach, 1988; Riabowol et al. 1989).

In previous studies, using a specific monoclonal antibody, we have identified in human cells a 125K/6.5 (K=10³M_r; 6.5 is pl) protein associated with the intranuclear matrix that is present in proliferating, but not in quiescent cells (Philipova et al. 1987; Yankulov et al. 1989). The 125K/6.5 protein displays a speckled nucleo-plasmic distribution throughout interphase, but is markedly increased during late G₂ and M phases of the cell cycle and was designated as ‘mitotin’ (Todorov et al. 1988a). This antigen is also increased in meiotic cells during rat spermatogenesis (Hadjiolova et al. 1989). In the present work we studied the synthesis and turnover of mitotin in cell cycle synchronized human WISH cells. We studied further the in vivo phosphorylation of mitotin during S and G₂ phases of the cell cycle. These results add evidence on the possible involvement of this nuclear matrix protein in the complex events preparing the cell for mitosis.

WISH cells (ATCC CCL25) are grown on MEM containing 5% fetal bovine serum. Cells are synchronized by a double thymidine block (see Adams, 1980). Cells in S, S/G₂ and G₂/M phases are studied at different times after release from the block (Todorov et al. 1988a). Synchronized cells are labeled with [³⁵S]methionine (l mCi ml⁻¹, Amersham, Bucks) for 2h at 37°C in methionine-depleted medium. After the labeling, some of the cells are washed and treated with cycloheximide (5 μg ml⁻¹) for an additional 2h. Equal numbers of cells are lysed and the total proteins of each sample are analyzed by twodimensional acrylamide gel electrophoresis.

To study protein phosphorylation, synchronized cells are labeled for a long period of time (10 h) with 10 μCiml⁻¹ of [³²P]orthophosphate. Alternatively, for short-term labeling (2h) 100 μCiml⁻¹ of [³²P] orthophosphate is added at 2 or 8h after release from the block.

For immunoprecipitation (see Laliberte et al, 1984) the cells are lysed in lysis buffer (0.15M-NaCl, 20mM-Tris-HCl, pH 7.6, 10 mM-sodium phosphate, 0.5% sodium deoxycholate, 1% Triton X-100, 0.05% sodium dodecyl sulfate and 1 mM-phenylmethanesulfonyl fluoride (PMSF)) and disrupted with an ultrasonic disintegrator. The material is clarified by centrifugation at 13 000g for 10 min and immunoprecipitated by overnight incubation at 4°C with 50 μg ml⁻¹ of the purified monoclonal anti-mitotin antibody (Philipova et al. 1987), followed by addition of a suspension of IgG-Sorb (The Enzyme Center Inc., Malden, Massachusetts) and further incubation for 1 h at 22°C and 1 h at 4°C. The IgG-Sorb is then washed several times with lysis buffer, containing 0.5 M-NaCl, followed by lysis buffer, and finally with TBS (10 mM-Tris-HCl, pH 7.6, 0.15mM-NaCl, 1 mM-PMSF). The bound proteins are eluted either with 9.5 M-urea, 2% Triton X-100, 50mM-dithiothreitol and 2% Ampholines (Todorov et al. 1988a) or by boiling the IgG-Sorb in acrylamide-SDS gel electrophoresis sample buffer (Laemmli, 1970).

The samples are analyzed by one-dimensional electrophoresis in 12.5% polyacrylamide gels (Laemmli, 1970). The two-dimensional gel electrophoresis separation is performed using the system of O’Farrell (1975). For the first dimension (isoelectric focusing, IEF) we used 4% polyacrylamide gels with a 4: 1 mixture of pH 3–10 and pH 5–8 Ampholines (Pharmacia, Uppsala). The second dimension was in 12.5% polyacryl-amide-SDS gels. ³⁵S- and ³²P-labeled proteins are detected by autoradiography.

For immunofluorescence analyses synchronized cells grown on coverslips are fixed in cold methanol, followed by acetone. The intracellular localization of mitotin is investigated after treatment with 10 μg ml⁻¹ of the purified anti-mitotin monoclonal antibody (Philipova el al. 1987) and a FITC-conjugated rabbit anti-mouse IgG (DAKOPATTS, Denmark).

The immunocytochemical distribution of mitotin at different stages after release from the double thymidine block is shown in Fig. 1. In S phase cells the antigen displays a speckled nucleoplasmic distribution (Fig. 1 A). At the beginning and during mitosis it is markedly increased, showing a distinct extrachromosomal distribution (Fig. 1C,E) (see also Todorov et al. 1988a).

In a preliminary set of experiments we tried to quantitate the differences in the amounts of p125 in lysates isolated from interphase and mitotic cells fractionated as described earlier (Todorov et al. 1988a). Lysates from equal numbers of cells were fractionated by twodimensional gel electrophoresis, stained with Coomassie Brilliant Blue R250, and the absorbance of the eluted pl25 (mitotin) spots was determined according to Ball (1986). The results show that the amount of pl25 in mitotic cells is about sevenfold higher than in interphase cells. Unexpectedly, short-term labeling with [³⁵S]meth-ionine did not show such dramatic changes, the labeling of pl25 in mitotic cells being only about twofold more intense than in interphase cells. In order to clarify the reason for this discrepancy we carried out experiments with WISH cells synchronized by a double thymidine block (Adams, 1980). The results obtained upon [³⁵S]methionine labeling are outlined in Fig. 2. The results show that mitotin is synthesized throughout the cell cycle, the labeling in G₂ and G₂/M phases being noticeably higher than in S phase (Fig. 2A,C,E). These results are in line with experiments with a cDNA probe showing that the amount of mitotin mRNA is about twofold higher in G₂, as compared to S phase WISH or Raji cells (Todorov et al. 1988b).

We studied the fate of mitotin upon cycloheximide block of protein synthesis. As can be seen in Fig, 1, immunofluorescence of synchronized WISH cells, treated for 2h with cycloheximide, reveals a distinct decrease in the antigen (Fig. IB and D). However, this decrease is not as pronounced in the case of mitotic cells (Fig. IF). Further, we studied the turnover of mitotin, short-term labeled (2 h) with [³⁵S]methionine, upon a 2 h cycloheximide block of protein synthesis. As shown in Fig. 2, the turnover of mitotin in S and S/G₂ phase cells is very rapid and the protein labeled for 2h is barely detectable after 2h cycloheximide. In contrast, mitotin labeled during G₂ phase decreases only slightly upon cycloheximide treatment (Fig. 2Bf). These results demonstrate that the mitotin synthesized during G₂ and G₂/M is metabolically stabilized, its continuing synthesis resulting in the recorded marked increase in the amount of mitotin in mitotic cells. Observation of the twodimensional gel electrophoretograms reveals further that mitotin, labeled with [³⁵S]methionine during G₂ and G₂/M phase, produced reproducibly two or three spots instead of the single spot observed in S phase (see Fig. 2B). The additional spots of mitotin labeled during G₂ phase are located at slightly more acidic pH, suggesting the modification of this protein in pre-mitotic and mitotic cells. Further, the two more acidic spots appear to be metabolically stabilized in cycloheximide-treated cells (Fig. 2B,e,f), thus suggesting the possibility of mitotin phosphorylation during G₂ phase. Therefore, we carried out experiments to study the phosphorylation state of mitotin at different stages of the cell cycle. The experiments were carried out again with WISH cells synchronized by a double thymidine block (Adams, 1980). After release from the block the cells were incubated for 10 h in the presence of [³²P] phosphate. At this stage about 60% of the cells are in mitosis, the remaining being in G₂ and G 2/M. The cells were harvested, lysed and two-dimensional acrylamide gel electrophoretic fractionation was carried out. The results shown in Fig. 3 reveal the net accumulation of mitotin (three spots) in mitotic cells (Fig. 3A) and the presence of a distinct ³²P-labeled spot comigrating with pl25 (Fig. 3B).

To obtain direct evidence for the phosphorylation of mitotin in mitotic cells we carried out acrylamide gel electrophoresis analyses of the ³²P-labeled proteins immunoprecipitated with the anti-mitotin monoclonal anti body (Philipova et al. 1987). The results (Fig. 4) show that after 2 h labeling in S phase the antibody precipitates a [^3sS]methionine-labeled protein (Fig. 4A), but does not precipitate a detectable phosphorylated product (Fig. 4D). On the other hand, in cells labeled with [³⁵S]methionine in G₂ and G₂/M, the antibody precipitates a single major band of about 125×10³A/_r (Fig. 4B) that is phosphorylated (Fig. 4F). Upon two-dimensional acrylamide gel electrophoresis (Fig. 5) it can be seen that the anti-mitotin monoclonal antibody immunoprecipitates three [³⁵S]methionine-labeled polypeptide products with identical M_r values of 125×10³ and similar pl values (Fig. 5A), two of which are phosphorylated (Fig. 5B). It is noteworthy that the more-acidic pl25 phosphorylated spot appears to be more intense although its [³⁵S]methionine labeling is markedly lower (see also Fig. 2B, E and F).

The results obtained in this work demonstrate that in cultured human cells: (1) the nuclear matrix-associated protein mitotin accumulates during late G₂ and G₂/M phases of the cell cycle; (2) two phosphorylated forms of mitotin are present in mitotic cells, but absent in S phase cells; and (3) the phosphorylated forms of mitotin are metabolically stabilized, thus resulting in the marked accumulation of this protein in premitotic and mitotic cells. An increased metabolic stability possibly related to phosphorylation was observed recently in the case of the ets-2 proto-oncogene nuclear protein (Fujiwara et al. 1988). Our results implicate more closely the participation of the nuclear matrix-associated protein mitotin (pl25/6.5) in the complex events preparing the cell for mitosis. It should be noted that previous studies from other laboratories have reported the phosphorylation of nuclear proteins in the 110–130 (×10³) M_r range in G₂ phase or mitotic cells (Henry and Hodge, 1983; Turner et al. 1985; Davis and Rao, 1987). However, their relationship with the nuclear matrix protein mitotin studied here remains to be ascertained. The observed considerable (5-to 10-fold) phosphorylation-related accumulation of mitotin in premitotic nuclei and mitotic cells strongly suggests that this protein may be more directly involved in the complex cascade of nuclear events preparing the cell for mitosis (Lee & Nurse, 1987; Draetta & Beach, 1988). When considering such a role for mitotin, it should be borne in mind that this protein is clearly extrachromosomal in both late G₂ and during mitosis (Todorov et al. 1988a) as well as during meiosis (Hadjiolova et al. 1989). Further studies are needed to clarify whether the accumulation of mitotin is needed for a possibly crucial interaction with some nuclear or cytoplasmic structures during mitosis and meiosis.

The authors are greatly indebted to Mrs S. Nikolova for help in preparation of the manuscript. This work was supported by grant no. 363/1987 from the State Committee on Science.