EXPERIMENTAL
CELL
RESEARCH
200,460-466
(1992)
The 180-kDa lsoform of Topoisomerase II Is Localized in the Nucleolus and Belongs to the Structural Elements of the Nucleolar Remnant NICOLETTA *Z&it&o
ZINI,*
ALBERTO M. MARTELLI,t PATRIZIA SABATELLI,* SPARTACO SANTI,$ CLAUDIA GIULIA C. B. ASTALDI RICOTTI,~ AND NADIR M. MARALDI**’
di Citomorfologia Normale e Patologica, Consiglio Nazionale delle Richerche, c/o I.O.R., Normale, Universitd di Bologna, Bologna, Italy; *Lab. Biologia Cellulare e Microscopia and $Istituto di Genetica Biochimica ed Evoluzionistica de1 Consiglio Nazionale
Monoclonal antibodies raised against two isoforms (170 and 150/180 kDa) of DNA topoisomerase II showed distinct fluorescence patterns in HeLa cells in different moments of the cell cycle (C. Negri et al., 1992, Exp. Cell Res. 200, 452-469). The ultrastructural distribution of the 160/180-kDa isoform, which in immunotluorescence showed a localization into the nucleolar region, has been analyzed by electron microscopy with a gold-conjugated secondary antibody in HeLa and KS62 cells. The results indicate that this isoform of the enzyme is exclusively localized in the nucleolus, mainly in the dense fibrillar component, while the nucleoplasm of interphase cells and the chromosomes of mitotic cells are completely negative. The antibody also reacts with the nucleolus of isolated nuclei and with the nucleolar remnant of purified nuclear matrices. A quantitative evaluation of the label distribution indicates that the percentage of label in the nucleolar remnant of isolated matrix is almost identical to that of the nucleolus in whole cells. The interaction with the insoluble proteins of the isolated nuclear matrix is also demonstrated by quantitative immunoblotting in which the M6Ab specifically stains a unique band corresponding to the 150/180-kDa isoform of topoisomerase II. The localization of the 150/180-kDa isoform of topoisomerase II in the nucleolar remnant strongly suggests that it represents a structural element for the spatial organization and for the regulation of transcription of the ribosomal genes. 0 1992 Academic PresB,
Inc.
INTRODUCTION Two types of DNA topoisomerase are present in eukaryotic cells; while topoisomerase I is a quite ubiquitous and constant soluble enzyme, topoisomerase II is a structural, largely insoluble protein that quantitatively
1 To whom
reprint
requests
should
be addressed.
0014-4827/92 $5.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
460
NEGRI,~
Bologna, Italy; tlsituto di Anatomia Elettronica, I.O.R., Bologna, Italy; delle Richerche, Pavia, Ztaly
Umana
varies from cell type to cell type and during the cell cycle [l]. Topoisomerase II is one of the major components of the nuclear matrix, conceivably involved in the maintenance of the chromatin loop architecture during interphase and in chromosome condensation [Z, 31. The localization of topoisomerase II at the attachment sites of chromatin loops to the matrix is suggested by numerous evidences: topoisomerase II is intimately associated with newly replicated DNA molecules [4]; by immunofluorescence topoisomerase II occupies a limited subdomain within the chromosome [l] and by ultrastructural immunocytochemistry the localization is defined at the base of the chromatin loops in isolated chromosomes [5]. Therefore, while the role of topoisomerase II in DNA replication is widely demonstrated, its involvement in transcription is uncertain. In fact, topoisomerase II is not concentrated at puffs on polytene chromosomes [6]; however, several findings indicate that enhancer sequences are linked to the nuclear matrix through topoisomerase II binding [7, 81. Two isoforms of topoisomerase II have been recently identified and characterized, a 170-kDa isoform almost undetectable in nonproliferating cells and a 180-kDa isoform prevailingly present in interphase cells; the hypothesis has been advanced that the 180-kDa isoform could be involved in ribosomal RNA transcription [9]. We produced several monoclonal antibodies specific to 170- and 180-kDa topoisomerase II isoforms. The anti-180-kDa MoAb has been referred to as anti-1501 180-kDa, since the 180-kDa topoisomerase II degrades to 150-kDa after its isolation [28]. In this work the 150/180-kDa isoform of topoisomerase II has been identified at the ultrastructural level in the nucleolus of interphase cells, while it is undetectable in mitotic chromosomes; moreover, it remains detectable as an insoluble structural element in the nucleolar remnant of isolated nuclear matrix. This suggests that this isoform of topoisomerase II might play a struc-
NUCLEOLAR
tural-enzymatic tion.
function
MATERIALS
in ribosomal
AND
LOCALIZATION
RNA transcrip-
METHODS
Cell cultures. The K562 human myeloid cell line was maintained in RPM1 1640 (GIBCO, Grand Island, NY), buffered with 25 n&f Hepes, supplemented with 10% fetal calf serum (FCS; GIBCO), 4 mM glutamine, 100 U/ml penicillin, and 0.1 mg/ml streptomycin. The cells were subcultured in fresh medium three times per week. HeLa cells, grown as monolayers, were cultured in F12 medium (GIBCO) supplemented with 10% FCS, 4 n&f glutamine, 100 U/ml penicillin, and 0.1 mg/ml streptomycin, trypsinized when confluent (normally every 3-4 days), and splitted 1:lO in fresh medium. Mitotic-arrested cells were obtained by adding colcemid (GIBCO, 0.2 pg/ml final concentration) to exponentially growing cultures and incubating for 24 h at 37°C in 5% CO, atmosphere. Exponentially growing K562 cells were washed ZsoZution of nucki. twice with Dulbecco’s phosphate-buffered saline (PBS), pH 7.4, and then resuspended to a concentration of lo7 cell/ml in 10 mM Tris-Cl, pH 7.4,2 mM MgCl*, for 5 min at 0°C. Triton X-100 was then added to 0.3% (w/v). Cells were sheared through a 22-gauge needle. Nuclei were separated by centrifugation at SOOg for 8 min. Isolated nuclei were washed twice with 10 mM Tris-Cl, pH 7.4, 2 mM MgCl, and finally resuspended to 2 mg DNA/ml in 0.25 M sucrose in 10 mM Tris-Cl, pH 7.4, 5 n-J4 MgCl,. All buffers for the isolation of nuclei and of nuclear matrices contained the following protease inhibitors: 1 mM phenylmethylsulfonyl fluoride, 0.5 mikf N-ethylmaleimide, 1 pgl ml leupeptin, 0.2 @g/ml aprotinin (Sigma, St. Louis, MO). Nuclei isolated from K562 cells Preparation of nuclear matrices. were incubated in the resuspending buffer for 45 min at 37°C and then digested with 30 U DNase I (EC 3.1.21.1) (Sigma) per milligram DNA for 45 min at 0°C. At the end of the digestion an equal volume of 4 M NaCl in 10 m&f Tris-Cl, pH 7.4,0.2 mM MgClz was added, followed by 8 vol of 2 M NaCl, 10 m&f Tris-Cl, pH 7.4, 0.2 n-&f MgCl, (high salt buffer), and centrifuged at 1OOOg for 10 min. The pellet was washed with 10 mM Tris-Cl, pH 7.4,0.2 n&f MgCl, (low salt buffer), and centrifuged at 1OOOg for 10 min. Antibodies. MoAbs to topoisomerase II from HeLa cells have been prepared and tested as described in the accompanying paper [28]. The screened hybridoma supernatants have been used undiluted or diluted up 1:5 for electron microscope immunocytochemistry. Electron microscopy immunocytochemistry. The samples were fixed in 1% glutaraldehyde in 0.1 M phosphate buffer, pH 7.2, dehydrated up to 70% ethanol, and embedded in London Resin White (LR White; Polyscience, Warrington, PA). The polymerization was carried out at 0°C. Thin sections were treated with normal goat serum (Sigma) for 30 min at room temperature, incubated for 16 h at 4’C with MoAbs anti-150/180-kDa topoisomerase II (diluted up to 1:5), and then incubated with a goat anti-mouse IgG conjugated with lo-nm colloidal gold particles (Bio Cell, Cardiff, UK) diluted 1:20 in 0.02 M Tris-Cl, pH 8.2, containing 0.1% bovine serum albumin (BSA) (Sigma), for 1 h at room temperature. The sections were stained with aqueous uranyl acetate and lead citrate. Controls consisted of sections not incubated with primary antibody. Observations were carried out with Zeiss EM 109 or Philips 400 electron microscope. Protein was Zmmunoblots of nuclear and nuclear matrix proteins. quantitated according to Lowry et al. [lo]. For immunoblotting analysis, nuclear proteins (350 pg), nuclear matrix proteins recovered from an equivalent number of nuclei (122 pg, since the recovery of nuclear proteins in the matrix fraction was 35%), and proteins extracted by high and low salt buffers [228 pg, obtained by trichloroacetic acid (TCA) precipitation of the extracts] were separated on a 8% polyacrylamide-0.1% sodium dodecyl sulfate
OF
180-kDa
TOPOISOMERASE
II
461
gel and then electrophoretically transferred to 2.5-cm wide X 8.5-cm long pieces of nitrocellulose paper in 0.192 M glycine, 0.025 M Tris, 20% methanol, pH 8.3, at 4°C for 24 h at 50 V (constant). Equal strips were then cut from the nitrocellulose pieces and saturated overnight at 4°C in PBS, pH 7.4, containing 3% BSA. The strips were then reacted for 6 h at room temperature in PBS-O.l% BSA containing a 1:lO dilution of the supernatant of anti-150/180-kDa topoisomerase II MoAb. After four washes (5 min each) in PBS, 0.1% Tween 20, the strips were incubated for 2 h at room temperature in PBS, 0.1% BSA, 0.1% Tween 20 containing a 1:500 dilution of an alkaline phosphatase-conjugated goat anti-mouse IgG (Sigma). After five washes in PBS-Tween 20 as above, antibody binding was detected in 100 n-&f Tris-Cl, pH 9.5, 4 n&f MgCl,, 0.1 mg/ml p-nitrotetrazolium blue, 0.05 mg/ml5-bromo-4-chloro-3-indolyl phosphate (Sigma). Quantitative evaluations were performed by densitometric scanning of the strips using a 1312 ISCO gel scanner.
RESULTS
Electron Microscope Immunocytochemistry
The immunocytochemical method that has been selected is based on a postembedding technique in which the antibody reacts against the antigen exposed on the surface of the thin sections. In order to maintain the antigenic specificity, the fixation has been done with a low concentration of glutaraldehyde and the dehydration with organic solvent has been limited up to 70% ethanol. The embedding has been performed in the hydrophilic LR White which also contributes to the preservation of the reactivity of the antigen sites. With these treatments, the cell morphology is quite preserved at the cytoplasmic level and almost perfect at the nuclear level. After the incubation with the MoAbs against the l70and the 150/180-kDa topoisomerase II, and with a secondary antibody conjugated with colloidal gold, the background is negligible and control sections, incubated only with the secondary antibody, are completely unlabeled. The observations were carried out in HeLa and in K562 cell lines either in exponentially growing cultures arrested in mitosis by colcemide or in quiescent cultures in which interphase cells prevail. In agreement with immunofluorescence findings [28], the group of MoAbs against the 170-kDa isoform label the nucleoplasm while the nucleolus is negative (data not reported). By using the group of MoAbs against the 150/180kDa isoform, in interphase HeLa (Fig. 1A) and K562 cells (Fig. 1B) the gold particles appear localized in the nucleolus, mainly over the dense fibrillar component (DFC) or at its border, while the heterochromatin associated with the nucleolus, the interchromatin granules, and the peripheral clumps of heterochromatin are almost unlabeled. In K562 mitotic cells arrested in metaphase, neither the chromosomes nor the cytoplasm are labeled (Fig. 1C). In order to determine whether this topoisomerase II
FIG. 1. MoAbs to the 150/180-kDa topoisomerase II isoform. Goat anti-mouse IgG conjugated with 10 nm colloidal gold. (A) HeLa cells. The labeling is present in the nucleolus (N) and mainly in the DFC. (B, C) K562 cells. In interphase cells, the nucleolus (N) is specifically labeled (B), while, in mitotic cells, the chromosomes (CH) are almost unlabeled (C). Bars, 0.5 @cm.
NUCLEOLAR
EPIG.
2.
LOCALIZATION
OF
180-kDa
TOPOISOMERASE
II
Nuclei and nuclear matrices isolated from K562 cells. MoAbs to the X0/180-kDa topoisomerase II isoform. Goat anti-mouse with 10 nm colloidal gold. (A) In isolated nuclei, the labeling is present mainly in the nucleolar area (N). In the nucleoplasm, the chromatin (CH) and the interchromatin granules (IG) are almost unlabeled. (B) In the purified nuclear matrix the label is present nut :leolar remnant (NR), while the inner matrix (IM) and the nuclear lamina (NL) are almost negative. Bars, 0.5 pm. con dugated
463
IgG 110th in the
464
ZINI ET AL. DEF
150kDao
whole nucleus. This figure is maintained also in purified nuclear matrices in which the label in the nucleolar remnant represents about 97% of that present in the whole matrix. This indicates that the HO-kDa topoisomerase II isoform is resistant to the treatments with high salt and could be considered as a structural element of the nuclear matrix and mainly, as indicated by the quantitative evaluation of the label distribution, of the nucleolar remnant. Quantitation of Immunoblots of Nuclear and Nuclear Matrix Proteins
FIG. 3. Immunoblotting analysis of the reactivity of MoAbs to the 150/180-kDa topoisomerase II isoform in nuclear and nuclear matrix fractions. Nuclear proteins, nuclear matrix proteins from an equivalent number of nuclei, and proteins extracted by high and low salt buffers were electrophoretically transferred to nitrocellulose paper and then probed with the MoAb to 150/180-kDa topoisomerase II (see Materials and Methods). (Lanes A, B) Total nuclear proteins. (Lanes C, D) Nuclear matrix proteins. (Lanes E, F) Nuclear proteins soluble in high and low salt buffers and precipitated by TCA. In B, D, and F the strips were reacted with the secondary antibody only. The arrow corresponds to exactly 150-kDa.
isoform could be considered a nuclear structural component, as the 170-kDa isoform detected in the chromosome scaffold [11, its immunolocalization has been done in nuclear and subnuclear fractions. After the isolation procedure for obtaining purified nuclei from K562 cells, the nuclear fraction appears homogeneous and free from cytoplasmic contaminations. The nuclear components are well preserved and the nuclear envelope is completely removed by the detergent treatment. The gold particles are present only on the nucleolus. Also in this case the DFC is the site of more intense labeling (Fig. 2A). Isolated nuclei were then treated with nucleases and with high and low salt extractions in order to obtain the residual insoluble nuclear matrix fraction. The morphology of purified nuclear matrices is well preserved, displaying its main constituents, that is, the nuclear lamina, the inner nuclear matrix network, and the nucleolar remnant; the immunogold reaction is almost exclusively present on the nucleolar remnant (Fig. 2B). Quantitative
Evaluation of the Label Distribution
The specificity of the immunolabeling for the nucleolar components, both in intact cells as well as in isolated nuclei and nuclear matrices, has been confirmed by a quantitative evaluation of the label distribution on the nucleolar area versus the nucleoplasmic area. In intact cells and in isolated nuclei, the label in the nucleolar area represents about 98% of the label present in the
In order to determine whether the extracting procedures for obtaining the nuclear matrix affect the recovery of the HO-kDa protein, we performed Western blot assays on nuclear and nuclear matrix proteins, in which the MoAbs to 150/180-kDa protein appear to be specific for a band of 150-kDa molecular weight (Fig. 3). A quantitative evaluation by densitometric scanning indicates that about 95% of the immunoreaction detectable in the total nuclear proteins (Fig. 3, lane A) is present in the nuclear matrix proteins (Fig. 3, lane Cl. No reaction is detectable in the proteins extracted by high and low salt buffers obtained by TCA precipitation of the extracts (Fig. 3, lane El. DISCUSSION The enzymatic functions of topoisomerase II have been mainly investigated in vitro and the mechanism of cleavage of both the DNA strands in the presence of ATP for relaxing superhelical twist in replicating DNA molecules has been widely clarified [ll]. More undefined are the roles of this enzyme in vivo, because other enzymes share the DNA relaxation properties of topoisomerase II. Moreover, the variations of topoisomerase II level and activity during the cell cycle have not been completely clarified. In fact, while the decrease of topoisomerase II content in the resting cell population could reflect its instability accompanying the programmed chromosome decondensation [5], the possible role of topoisomerase II during interphase in relationship with transcription is not well defined [6-81. On the other hand, the process of transcription seemsto require, at the chromatin loop level, deep structural transitions, involving the relaxation of the superhelical twist. Topoisomerase II plays a dual catalytic/structural of the nuclear role; in fact it is the main constituent matrix and of the chromosome scaffold, and the existence of matrix-associated regions consensus sequences interdispersed in the genome suggests that each chromatin
loop
is bound
to the nucleoskeleton
through
to-
poisomerase II itself [8]. Moreover, the actual localization of topoisomerase II at the central scaffold of the
NUCLEOLAR
LOCALIZATION
chromosomes has been documented by immunocytochemical methods in light and electron microscopy [5], demonstrating that this enzyme is involved also in the structural modifications of the matrix that occur during the cell cycle. Recently, the identification of two isoforms of topoisomerase II, characterized at the genetic and immunocytochemical levels [9, 121, provides new insight for the interpretation of the functional activity of the enzyme. In fact, the 170- and 180-kDa forms of topoisomerase II, both ATP-dependent and capable of relaxing DNA superhelical twist, require different KC1 concentrations for their catalytic activity and exhibit different heat denaturation and a selective sensitivity to inhibitors [9]. More interestingly, the variations of activity and concentration of the two isoforms are cell-cycle specific; in fact, the 170-kDa level decreases within resting cells, while the 180-kDa level diminishes in proliferating cells and seems not to be required for chromosome condensation [9]. It has been therefore hypothesized that the two isoforms play different functions in the cell. In fact, the 170-kDa isoform is inhibited by A-T rich oligomers, while the 180-kDa isoform is not; this suggests that the 170-kDa form is particularly localized in the A-T rich sequences present in the replication origins at the base of the chromatin loops [9]. On the contrary, G-C rich sequences have been found in promoter regions of genes of houskeeping proteins [ 131 and in nucleolar DNA [ 141. Therefore it has been hypothesized that 180-kDa topoisomerase II could be involved in transcription, especially of ribosomal RNA genes [9]. The reported results indicate that the MoAbs against the 150/180-kDa isoform do not react at all with chromosomes and with the nucleoplasm of interphase cells, but only with the nucleolus. These findings strongly support the hypothesis that the 180-kDa isoform of topoisomerase II, detectable only in interphase cells, is not a component of the chromosome scaffold involved in DNA replication, but an integral element of the nucleolus. In fact, the specific nucleolar immunolabeling with the anti-150/180-kDa MoAbs is retained also in isolated nuclei and, more significantly, in purified nuclear matrices. The results obtained with quantitative immunoblotting analysis indicate that almost all the 150/180-kDa topoisomerase II isoform is recovered in the nuclear matrix fraction. In fact, when proteins from equivalent amounts of nuclei and nuclear matrices are reacted with the MoAbs against the 150/180-kDa topoisomerase II, the immunoblotted bands obtained exhibit almost the same density. On the contrary, no immunoreaction is detectable in the soluble proteins precipitated by TCA from the proteins extracted by high and low salt buffers. It is worth noting that about 35% of total nuclear protein is retained in the final matrix fraction.
OF
180-kDa
TOPOISOMERASE
465
II
The quantitative immunoelectronmicroscopy analysis, moreover, indicates that almost all the 150/180kDa topoisomerase II isoform present in the nuclear matrix is detectable in the nucleolar remnant and not in the other matrix components. The nuclear matrix, aside from the structural role in arranging the chromatin loops, is considered to be involved in DNA replication and transcription [15]. The more striking evidences are represented by the enrichment of the matrix residual DNA in newly replicated DNA [16-181 and by the capability of isolated nuclear matrix to carry out in vitro DNA synthesis on the replication sites activated in uiuo [19]. Moreover, many key enzymes of DNA replication are linked to the matrix and could be part of the replitase complex [20, 211. These enzymes, like DNA polymerase (Y, DNA topoisomerase II, and RNA primase, are intimately linked to the matrix, being largely resistant to high salt extractions [ 22,231, though a partial release of topoisomerase II has been described to occur after matrix treatment with reducing agents [24]. Until now, in the nucleolus, the presence of topoisomerase I has been reported by immunocytochemistry [25] and its involvement in the functional organization of nucleolar genes has been suggested in yeast [26] and in mammalian cells [27]. The observed localization of the 150/180-kDa isoform of topoisomerase II in the nucleolus and, after the nuclear matrix purification, in the nucleolar remnant, that is the nucleoskeletal element which organizes the fixed sites for the transcription of the nucleolar genes, strongly suggests that the 150/180-kDa isoform could play a key role for the spatial organization and for the regulation of transcription of the nucleolus-associated chromatin. The research has been partially supported by grant PF Ingegneria Genetica from C.N.R. Italy and 40% grants from Minister0 della Universitl e della Ricerca. The skillful preparation of the photographic material was done by Mr. Aurelio Valmori.
REFERENCES 1.
Earnshaw, 1716-1725.
W. C., and Heck,
M. M. S. (1985)
J. Cell Bill.
100,
2. van der Velden, 3.
H. M., van Willigen, G., Wetzels, R. H., and Wanka, F. (1984) FEES I&t. 171,13-16. Earnshaw, W. C., Halligan, B., Cooke, C. A., Heck, M. M. S., and Liu, L. F. (1985) J. Cell Biol. 100, 1706-1715.
4. Nelson,
Liu,
5.
C., and Heck,
6. I.
W. G., 18’7-191. Earnshaw, W. 279-288. Heller, R. A., Brutlag, D. L. Cockerill,
L. F., and Coffey, M.
S. (1988)
Nature Cancer
322, Cell. 6,
Shelton, E. R., Dietrich, V., Elgin, S. C. R., and (1986) J. Bill. Chem. 261,8063-8069.
8. Gasser,
P. N., and Garrard, S. M., and Laemmli,
9.
F. H., Hofmann,
Drake,
M.
D. S. (1986)
W. T. (1986) Cell 44,273-282. U. K. (1986) Cell 46,521-530.
G. A., Bartus,
H. F., Mattern,
M.
R.,
466
10. 11. 12.
13.
ZINI ET AL. Crooke, S. T., and Mirabelli, C. K. (1989) Biochemistry 28, 8154-8160. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275. Wang, J. C. (1985) Annu. Reu. Biochem. 54,665-697. Chung, T. D. Y., Drake, F. H., Tan, K. B., Per, S. R., Crooke, S. T., and Mirabelli, C. K. (1989) PFOC. Natl. Acad. Sci. USA 86, 9431-9435. Kozasa, T., Itoh, H., Tsukamoto, T., and Kaziro, Y. (1986) Proc. Natl.
Acad.
Sci. USA
86,2081-2085.
14. Willems, M., Wagner, E., Laing, R., and Penmann, S. (1968) J. Mol. Biol. 32, 211-220. 15. Razin, S. V. (1987) BioEssays 6, 19-23. 16. Pardoll, D. M., Vogelstein, B., and Coffey, D. S. (1980) Cell 22, 527-536. 11. Smith, H. C., Puvion, E., Bucholtz, L. A., and Berezney, R. (1984) J. Cell Biol. 99, 1794-1802. 18. Zini, N., Mazzotti, G., Santi, P., Rizzoli, R., Galanzi, A., Rana, R., and Maraldi, N. M. (1989) Histochemistry 91, 199-204. 19. Berezney, R., and Buchholtz, L. (1981) Exp. Cell Res. 132,1-13. Received July 2, 1991 Revised version received February 14, 1992
20. Tubo, R. A., and Berezney, R. (1987) J. Biol. Chem. 262,66376642. 21. McConnell, M., Whalen, A. M., Smith, D. E., and Fisher, P. A. (1987) J. Cell Biol. 105, 108771098. 22. Littlewood, T. D., Hancock, D. C., and Evan, G. I. (1987) J. Cell Sci. 88,65-72.
23. Fisher, P. A., Liu, L., McConnell, M., Greanleaf, A., Lee, J., and Smith, D. E. (1989) J. Biol. Chem. 264, 3464-3469. 24. Kaufmann, S. H., and Shaper, J. H. (1991) Exp. Cell Res. 192, 511-523. 25. Raska, I., Reimer, E., Jarnik, M., Kostrouch, Z., and Raska, K., Jr. (1989) Biol. Cell. 65, 79-82. 26. Hirano, T., Konoha, G., Toda, T., and Yanagida, M. (1989) J. CellBiol. 108, 243-253. 27. Belenguer, P., Baldin, V., Mathieu, C., Prats, H., Bensaid, M., Bouche, G., and Amalric, F. (1989) Nucleic Acids Res. 17,66256636. 28. Negri, C., Chiesa, R., Cerino, A., Bestagno, M., Sala, C., Zini, N., Maraldi, N. M., and Astraldi Ricotti, G. C. B. (1992) Exp. Cell Res., 200,452-459.
CELL
RESEARCH
200,460-466
(1992)
The 180-kDa lsoform of Topoisomerase II Is Localized in the Nucleolus and Belongs to the Structural Elements of the Nucleolar Remnant NICOLETTA *Z&it&o
ZINI,*
ALBERTO M. MARTELLI,t PATRIZIA SABATELLI,* SPARTACO SANTI,$ CLAUDIA GIULIA C. B. ASTALDI RICOTTI,~ AND NADIR M. MARALDI**’
di Citomorfologia Normale e Patologica, Consiglio Nazionale delle Richerche, c/o I.O.R., Normale, Universitd di Bologna, Bologna, Italy; *Lab. Biologia Cellulare e Microscopia and $Istituto di Genetica Biochimica ed Evoluzionistica de1 Consiglio Nazionale
Monoclonal antibodies raised against two isoforms (170 and 150/180 kDa) of DNA topoisomerase II showed distinct fluorescence patterns in HeLa cells in different moments of the cell cycle (C. Negri et al., 1992, Exp. Cell Res. 200, 452-469). The ultrastructural distribution of the 160/180-kDa isoform, which in immunotluorescence showed a localization into the nucleolar region, has been analyzed by electron microscopy with a gold-conjugated secondary antibody in HeLa and KS62 cells. The results indicate that this isoform of the enzyme is exclusively localized in the nucleolus, mainly in the dense fibrillar component, while the nucleoplasm of interphase cells and the chromosomes of mitotic cells are completely negative. The antibody also reacts with the nucleolus of isolated nuclei and with the nucleolar remnant of purified nuclear matrices. A quantitative evaluation of the label distribution indicates that the percentage of label in the nucleolar remnant of isolated matrix is almost identical to that of the nucleolus in whole cells. The interaction with the insoluble proteins of the isolated nuclear matrix is also demonstrated by quantitative immunoblotting in which the M6Ab specifically stains a unique band corresponding to the 150/180-kDa isoform of topoisomerase II. The localization of the 150/180-kDa isoform of topoisomerase II in the nucleolar remnant strongly suggests that it represents a structural element for the spatial organization and for the regulation of transcription of the ribosomal genes. 0 1992 Academic PresB,
Inc.
INTRODUCTION Two types of DNA topoisomerase are present in eukaryotic cells; while topoisomerase I is a quite ubiquitous and constant soluble enzyme, topoisomerase II is a structural, largely insoluble protein that quantitatively
1 To whom
reprint
requests
should
be addressed.
0014-4827/92 $5.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
460
NEGRI,~
Bologna, Italy; tlsituto di Anatomia Elettronica, I.O.R., Bologna, Italy; delle Richerche, Pavia, Ztaly
Umana
varies from cell type to cell type and during the cell cycle [l]. Topoisomerase II is one of the major components of the nuclear matrix, conceivably involved in the maintenance of the chromatin loop architecture during interphase and in chromosome condensation [Z, 31. The localization of topoisomerase II at the attachment sites of chromatin loops to the matrix is suggested by numerous evidences: topoisomerase II is intimately associated with newly replicated DNA molecules [4]; by immunofluorescence topoisomerase II occupies a limited subdomain within the chromosome [l] and by ultrastructural immunocytochemistry the localization is defined at the base of the chromatin loops in isolated chromosomes [5]. Therefore, while the role of topoisomerase II in DNA replication is widely demonstrated, its involvement in transcription is uncertain. In fact, topoisomerase II is not concentrated at puffs on polytene chromosomes [6]; however, several findings indicate that enhancer sequences are linked to the nuclear matrix through topoisomerase II binding [7, 81. Two isoforms of topoisomerase II have been recently identified and characterized, a 170-kDa isoform almost undetectable in nonproliferating cells and a 180-kDa isoform prevailingly present in interphase cells; the hypothesis has been advanced that the 180-kDa isoform could be involved in ribosomal RNA transcription [9]. We produced several monoclonal antibodies specific to 170- and 180-kDa topoisomerase II isoforms. The anti-180-kDa MoAb has been referred to as anti-1501 180-kDa, since the 180-kDa topoisomerase II degrades to 150-kDa after its isolation [28]. In this work the 150/180-kDa isoform of topoisomerase II has been identified at the ultrastructural level in the nucleolus of interphase cells, while it is undetectable in mitotic chromosomes; moreover, it remains detectable as an insoluble structural element in the nucleolar remnant of isolated nuclear matrix. This suggests that this isoform of topoisomerase II might play a struc-
NUCLEOLAR
tural-enzymatic tion.
function
MATERIALS
in ribosomal
AND
LOCALIZATION
RNA transcrip-
METHODS
Cell cultures. The K562 human myeloid cell line was maintained in RPM1 1640 (GIBCO, Grand Island, NY), buffered with 25 n&f Hepes, supplemented with 10% fetal calf serum (FCS; GIBCO), 4 mM glutamine, 100 U/ml penicillin, and 0.1 mg/ml streptomycin. The cells were subcultured in fresh medium three times per week. HeLa cells, grown as monolayers, were cultured in F12 medium (GIBCO) supplemented with 10% FCS, 4 n&f glutamine, 100 U/ml penicillin, and 0.1 mg/ml streptomycin, trypsinized when confluent (normally every 3-4 days), and splitted 1:lO in fresh medium. Mitotic-arrested cells were obtained by adding colcemid (GIBCO, 0.2 pg/ml final concentration) to exponentially growing cultures and incubating for 24 h at 37°C in 5% CO, atmosphere. Exponentially growing K562 cells were washed ZsoZution of nucki. twice with Dulbecco’s phosphate-buffered saline (PBS), pH 7.4, and then resuspended to a concentration of lo7 cell/ml in 10 mM Tris-Cl, pH 7.4,2 mM MgCl*, for 5 min at 0°C. Triton X-100 was then added to 0.3% (w/v). Cells were sheared through a 22-gauge needle. Nuclei were separated by centrifugation at SOOg for 8 min. Isolated nuclei were washed twice with 10 mM Tris-Cl, pH 7.4, 2 mM MgCl, and finally resuspended to 2 mg DNA/ml in 0.25 M sucrose in 10 mM Tris-Cl, pH 7.4, 5 n-J4 MgCl,. All buffers for the isolation of nuclei and of nuclear matrices contained the following protease inhibitors: 1 mM phenylmethylsulfonyl fluoride, 0.5 mikf N-ethylmaleimide, 1 pgl ml leupeptin, 0.2 @g/ml aprotinin (Sigma, St. Louis, MO). Nuclei isolated from K562 cells Preparation of nuclear matrices. were incubated in the resuspending buffer for 45 min at 37°C and then digested with 30 U DNase I (EC 3.1.21.1) (Sigma) per milligram DNA for 45 min at 0°C. At the end of the digestion an equal volume of 4 M NaCl in 10 m&f Tris-Cl, pH 7.4,0.2 mM MgClz was added, followed by 8 vol of 2 M NaCl, 10 m&f Tris-Cl, pH 7.4, 0.2 n-&f MgCl, (high salt buffer), and centrifuged at 1OOOg for 10 min. The pellet was washed with 10 mM Tris-Cl, pH 7.4,0.2 n&f MgCl, (low salt buffer), and centrifuged at 1OOOg for 10 min. Antibodies. MoAbs to topoisomerase II from HeLa cells have been prepared and tested as described in the accompanying paper [28]. The screened hybridoma supernatants have been used undiluted or diluted up 1:5 for electron microscope immunocytochemistry. Electron microscopy immunocytochemistry. The samples were fixed in 1% glutaraldehyde in 0.1 M phosphate buffer, pH 7.2, dehydrated up to 70% ethanol, and embedded in London Resin White (LR White; Polyscience, Warrington, PA). The polymerization was carried out at 0°C. Thin sections were treated with normal goat serum (Sigma) for 30 min at room temperature, incubated for 16 h at 4’C with MoAbs anti-150/180-kDa topoisomerase II (diluted up to 1:5), and then incubated with a goat anti-mouse IgG conjugated with lo-nm colloidal gold particles (Bio Cell, Cardiff, UK) diluted 1:20 in 0.02 M Tris-Cl, pH 8.2, containing 0.1% bovine serum albumin (BSA) (Sigma), for 1 h at room temperature. The sections were stained with aqueous uranyl acetate and lead citrate. Controls consisted of sections not incubated with primary antibody. Observations were carried out with Zeiss EM 109 or Philips 400 electron microscope. Protein was Zmmunoblots of nuclear and nuclear matrix proteins. quantitated according to Lowry et al. [lo]. For immunoblotting analysis, nuclear proteins (350 pg), nuclear matrix proteins recovered from an equivalent number of nuclei (122 pg, since the recovery of nuclear proteins in the matrix fraction was 35%), and proteins extracted by high and low salt buffers [228 pg, obtained by trichloroacetic acid (TCA) precipitation of the extracts] were separated on a 8% polyacrylamide-0.1% sodium dodecyl sulfate
OF
180-kDa
TOPOISOMERASE
II
461
gel and then electrophoretically transferred to 2.5-cm wide X 8.5-cm long pieces of nitrocellulose paper in 0.192 M glycine, 0.025 M Tris, 20% methanol, pH 8.3, at 4°C for 24 h at 50 V (constant). Equal strips were then cut from the nitrocellulose pieces and saturated overnight at 4°C in PBS, pH 7.4, containing 3% BSA. The strips were then reacted for 6 h at room temperature in PBS-O.l% BSA containing a 1:lO dilution of the supernatant of anti-150/180-kDa topoisomerase II MoAb. After four washes (5 min each) in PBS, 0.1% Tween 20, the strips were incubated for 2 h at room temperature in PBS, 0.1% BSA, 0.1% Tween 20 containing a 1:500 dilution of an alkaline phosphatase-conjugated goat anti-mouse IgG (Sigma). After five washes in PBS-Tween 20 as above, antibody binding was detected in 100 n-&f Tris-Cl, pH 9.5, 4 n&f MgCl,, 0.1 mg/ml p-nitrotetrazolium blue, 0.05 mg/ml5-bromo-4-chloro-3-indolyl phosphate (Sigma). Quantitative evaluations were performed by densitometric scanning of the strips using a 1312 ISCO gel scanner.
RESULTS
Electron Microscope Immunocytochemistry
The immunocytochemical method that has been selected is based on a postembedding technique in which the antibody reacts against the antigen exposed on the surface of the thin sections. In order to maintain the antigenic specificity, the fixation has been done with a low concentration of glutaraldehyde and the dehydration with organic solvent has been limited up to 70% ethanol. The embedding has been performed in the hydrophilic LR White which also contributes to the preservation of the reactivity of the antigen sites. With these treatments, the cell morphology is quite preserved at the cytoplasmic level and almost perfect at the nuclear level. After the incubation with the MoAbs against the l70and the 150/180-kDa topoisomerase II, and with a secondary antibody conjugated with colloidal gold, the background is negligible and control sections, incubated only with the secondary antibody, are completely unlabeled. The observations were carried out in HeLa and in K562 cell lines either in exponentially growing cultures arrested in mitosis by colcemide or in quiescent cultures in which interphase cells prevail. In agreement with immunofluorescence findings [28], the group of MoAbs against the 170-kDa isoform label the nucleoplasm while the nucleolus is negative (data not reported). By using the group of MoAbs against the 150/180kDa isoform, in interphase HeLa (Fig. 1A) and K562 cells (Fig. 1B) the gold particles appear localized in the nucleolus, mainly over the dense fibrillar component (DFC) or at its border, while the heterochromatin associated with the nucleolus, the interchromatin granules, and the peripheral clumps of heterochromatin are almost unlabeled. In K562 mitotic cells arrested in metaphase, neither the chromosomes nor the cytoplasm are labeled (Fig. 1C). In order to determine whether this topoisomerase II
FIG. 1. MoAbs to the 150/180-kDa topoisomerase II isoform. Goat anti-mouse IgG conjugated with 10 nm colloidal gold. (A) HeLa cells. The labeling is present in the nucleolus (N) and mainly in the DFC. (B, C) K562 cells. In interphase cells, the nucleolus (N) is specifically labeled (B), while, in mitotic cells, the chromosomes (CH) are almost unlabeled (C). Bars, 0.5 @cm.
NUCLEOLAR
EPIG.
2.
LOCALIZATION
OF
180-kDa
TOPOISOMERASE
II
Nuclei and nuclear matrices isolated from K562 cells. MoAbs to the X0/180-kDa topoisomerase II isoform. Goat anti-mouse with 10 nm colloidal gold. (A) In isolated nuclei, the labeling is present mainly in the nucleolar area (N). In the nucleoplasm, the chromatin (CH) and the interchromatin granules (IG) are almost unlabeled. (B) In the purified nuclear matrix the label is present nut :leolar remnant (NR), while the inner matrix (IM) and the nuclear lamina (NL) are almost negative. Bars, 0.5 pm. con dugated
463
IgG 110th in the
464
ZINI ET AL. DEF
150kDao
whole nucleus. This figure is maintained also in purified nuclear matrices in which the label in the nucleolar remnant represents about 97% of that present in the whole matrix. This indicates that the HO-kDa topoisomerase II isoform is resistant to the treatments with high salt and could be considered as a structural element of the nuclear matrix and mainly, as indicated by the quantitative evaluation of the label distribution, of the nucleolar remnant. Quantitation of Immunoblots of Nuclear and Nuclear Matrix Proteins
FIG. 3. Immunoblotting analysis of the reactivity of MoAbs to the 150/180-kDa topoisomerase II isoform in nuclear and nuclear matrix fractions. Nuclear proteins, nuclear matrix proteins from an equivalent number of nuclei, and proteins extracted by high and low salt buffers were electrophoretically transferred to nitrocellulose paper and then probed with the MoAb to 150/180-kDa topoisomerase II (see Materials and Methods). (Lanes A, B) Total nuclear proteins. (Lanes C, D) Nuclear matrix proteins. (Lanes E, F) Nuclear proteins soluble in high and low salt buffers and precipitated by TCA. In B, D, and F the strips were reacted with the secondary antibody only. The arrow corresponds to exactly 150-kDa.
isoform could be considered a nuclear structural component, as the 170-kDa isoform detected in the chromosome scaffold [11, its immunolocalization has been done in nuclear and subnuclear fractions. After the isolation procedure for obtaining purified nuclei from K562 cells, the nuclear fraction appears homogeneous and free from cytoplasmic contaminations. The nuclear components are well preserved and the nuclear envelope is completely removed by the detergent treatment. The gold particles are present only on the nucleolus. Also in this case the DFC is the site of more intense labeling (Fig. 2A). Isolated nuclei were then treated with nucleases and with high and low salt extractions in order to obtain the residual insoluble nuclear matrix fraction. The morphology of purified nuclear matrices is well preserved, displaying its main constituents, that is, the nuclear lamina, the inner nuclear matrix network, and the nucleolar remnant; the immunogold reaction is almost exclusively present on the nucleolar remnant (Fig. 2B). Quantitative
Evaluation of the Label Distribution
The specificity of the immunolabeling for the nucleolar components, both in intact cells as well as in isolated nuclei and nuclear matrices, has been confirmed by a quantitative evaluation of the label distribution on the nucleolar area versus the nucleoplasmic area. In intact cells and in isolated nuclei, the label in the nucleolar area represents about 98% of the label present in the
In order to determine whether the extracting procedures for obtaining the nuclear matrix affect the recovery of the HO-kDa protein, we performed Western blot assays on nuclear and nuclear matrix proteins, in which the MoAbs to 150/180-kDa protein appear to be specific for a band of 150-kDa molecular weight (Fig. 3). A quantitative evaluation by densitometric scanning indicates that about 95% of the immunoreaction detectable in the total nuclear proteins (Fig. 3, lane A) is present in the nuclear matrix proteins (Fig. 3, lane Cl. No reaction is detectable in the proteins extracted by high and low salt buffers obtained by TCA precipitation of the extracts (Fig. 3, lane El. DISCUSSION The enzymatic functions of topoisomerase II have been mainly investigated in vitro and the mechanism of cleavage of both the DNA strands in the presence of ATP for relaxing superhelical twist in replicating DNA molecules has been widely clarified [ll]. More undefined are the roles of this enzyme in vivo, because other enzymes share the DNA relaxation properties of topoisomerase II. Moreover, the variations of topoisomerase II level and activity during the cell cycle have not been completely clarified. In fact, while the decrease of topoisomerase II content in the resting cell population could reflect its instability accompanying the programmed chromosome decondensation [5], the possible role of topoisomerase II during interphase in relationship with transcription is not well defined [6-81. On the other hand, the process of transcription seemsto require, at the chromatin loop level, deep structural transitions, involving the relaxation of the superhelical twist. Topoisomerase II plays a dual catalytic/structural of the nuclear role; in fact it is the main constituent matrix and of the chromosome scaffold, and the existence of matrix-associated regions consensus sequences interdispersed in the genome suggests that each chromatin
loop
is bound
to the nucleoskeleton
through
to-
poisomerase II itself [8]. Moreover, the actual localization of topoisomerase II at the central scaffold of the
NUCLEOLAR
LOCALIZATION
chromosomes has been documented by immunocytochemical methods in light and electron microscopy [5], demonstrating that this enzyme is involved also in the structural modifications of the matrix that occur during the cell cycle. Recently, the identification of two isoforms of topoisomerase II, characterized at the genetic and immunocytochemical levels [9, 121, provides new insight for the interpretation of the functional activity of the enzyme. In fact, the 170- and 180-kDa forms of topoisomerase II, both ATP-dependent and capable of relaxing DNA superhelical twist, require different KC1 concentrations for their catalytic activity and exhibit different heat denaturation and a selective sensitivity to inhibitors [9]. More interestingly, the variations of activity and concentration of the two isoforms are cell-cycle specific; in fact, the 170-kDa level decreases within resting cells, while the 180-kDa level diminishes in proliferating cells and seems not to be required for chromosome condensation [9]. It has been therefore hypothesized that the two isoforms play different functions in the cell. In fact, the 170-kDa isoform is inhibited by A-T rich oligomers, while the 180-kDa isoform is not; this suggests that the 170-kDa form is particularly localized in the A-T rich sequences present in the replication origins at the base of the chromatin loops [9]. On the contrary, G-C rich sequences have been found in promoter regions of genes of houskeeping proteins [ 131 and in nucleolar DNA [ 141. Therefore it has been hypothesized that 180-kDa topoisomerase II could be involved in transcription, especially of ribosomal RNA genes [9]. The reported results indicate that the MoAbs against the 150/180-kDa isoform do not react at all with chromosomes and with the nucleoplasm of interphase cells, but only with the nucleolus. These findings strongly support the hypothesis that the 180-kDa isoform of topoisomerase II, detectable only in interphase cells, is not a component of the chromosome scaffold involved in DNA replication, but an integral element of the nucleolus. In fact, the specific nucleolar immunolabeling with the anti-150/180-kDa MoAbs is retained also in isolated nuclei and, more significantly, in purified nuclear matrices. The results obtained with quantitative immunoblotting analysis indicate that almost all the 150/180-kDa topoisomerase II isoform is recovered in the nuclear matrix fraction. In fact, when proteins from equivalent amounts of nuclei and nuclear matrices are reacted with the MoAbs against the 150/180-kDa topoisomerase II, the immunoblotted bands obtained exhibit almost the same density. On the contrary, no immunoreaction is detectable in the soluble proteins precipitated by TCA from the proteins extracted by high and low salt buffers. It is worth noting that about 35% of total nuclear protein is retained in the final matrix fraction.
OF
180-kDa
TOPOISOMERASE
465
II
The quantitative immunoelectronmicroscopy analysis, moreover, indicates that almost all the 150/180kDa topoisomerase II isoform present in the nuclear matrix is detectable in the nucleolar remnant and not in the other matrix components. The nuclear matrix, aside from the structural role in arranging the chromatin loops, is considered to be involved in DNA replication and transcription [15]. The more striking evidences are represented by the enrichment of the matrix residual DNA in newly replicated DNA [16-181 and by the capability of isolated nuclear matrix to carry out in vitro DNA synthesis on the replication sites activated in uiuo [19]. Moreover, many key enzymes of DNA replication are linked to the matrix and could be part of the replitase complex [20, 211. These enzymes, like DNA polymerase (Y, DNA topoisomerase II, and RNA primase, are intimately linked to the matrix, being largely resistant to high salt extractions [ 22,231, though a partial release of topoisomerase II has been described to occur after matrix treatment with reducing agents [24]. Until now, in the nucleolus, the presence of topoisomerase I has been reported by immunocytochemistry [25] and its involvement in the functional organization of nucleolar genes has been suggested in yeast [26] and in mammalian cells [27]. The observed localization of the 150/180-kDa isoform of topoisomerase II in the nucleolus and, after the nuclear matrix purification, in the nucleolar remnant, that is the nucleoskeletal element which organizes the fixed sites for the transcription of the nucleolar genes, strongly suggests that the 150/180-kDa isoform could play a key role for the spatial organization and for the regulation of transcription of the nucleolus-associated chromatin. The research has been partially supported by grant PF Ingegneria Genetica from C.N.R. Italy and 40% grants from Minister0 della Universitl e della Ricerca. The skillful preparation of the photographic material was done by Mr. Aurelio Valmori.
REFERENCES 1.
Earnshaw, 1716-1725.
W. C., and Heck,
M. M. S. (1985)
J. Cell Bill.
100,
2. van der Velden, 3.
H. M., van Willigen, G., Wetzels, R. H., and Wanka, F. (1984) FEES I&t. 171,13-16. Earnshaw, W. C., Halligan, B., Cooke, C. A., Heck, M. M. S., and Liu, L. F. (1985) J. Cell Biol. 100, 1706-1715.
4. Nelson,
Liu,
5.
C., and Heck,
6. I.
W. G., 18’7-191. Earnshaw, W. 279-288. Heller, R. A., Brutlag, D. L. Cockerill,
L. F., and Coffey, M.
S. (1988)
Nature Cancer
322, Cell. 6,
Shelton, E. R., Dietrich, V., Elgin, S. C. R., and (1986) J. Bill. Chem. 261,8063-8069.
8. Gasser,
P. N., and Garrard, S. M., and Laemmli,
9.
F. H., Hofmann,
Drake,
M.
D. S. (1986)
W. T. (1986) Cell 44,273-282. U. K. (1986) Cell 46,521-530.
G. A., Bartus,
H. F., Mattern,
M.
R.,
466
10. 11. 12.
13.
ZINI ET AL. Crooke, S. T., and Mirabelli, C. K. (1989) Biochemistry 28, 8154-8160. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275. Wang, J. C. (1985) Annu. Reu. Biochem. 54,665-697. Chung, T. D. Y., Drake, F. H., Tan, K. B., Per, S. R., Crooke, S. T., and Mirabelli, C. K. (1989) PFOC. Natl. Acad. Sci. USA 86, 9431-9435. Kozasa, T., Itoh, H., Tsukamoto, T., and Kaziro, Y. (1986) Proc. Natl.
Acad.
Sci. USA
86,2081-2085.
14. Willems, M., Wagner, E., Laing, R., and Penmann, S. (1968) J. Mol. Biol. 32, 211-220. 15. Razin, S. V. (1987) BioEssays 6, 19-23. 16. Pardoll, D. M., Vogelstein, B., and Coffey, D. S. (1980) Cell 22, 527-536. 11. Smith, H. C., Puvion, E., Bucholtz, L. A., and Berezney, R. (1984) J. Cell Biol. 99, 1794-1802. 18. Zini, N., Mazzotti, G., Santi, P., Rizzoli, R., Galanzi, A., Rana, R., and Maraldi, N. M. (1989) Histochemistry 91, 199-204. 19. Berezney, R., and Buchholtz, L. (1981) Exp. Cell Res. 132,1-13. Received July 2, 1991 Revised version received February 14, 1992
20. Tubo, R. A., and Berezney, R. (1987) J. Biol. Chem. 262,66376642. 21. McConnell, M., Whalen, A. M., Smith, D. E., and Fisher, P. A. (1987) J. Cell Biol. 105, 108771098. 22. Littlewood, T. D., Hancock, D. C., and Evan, G. I. (1987) J. Cell Sci. 88,65-72.
23. Fisher, P. A., Liu, L., McConnell, M., Greanleaf, A., Lee, J., and Smith, D. E. (1989) J. Biol. Chem. 264, 3464-3469. 24. Kaufmann, S. H., and Shaper, J. H. (1991) Exp. Cell Res. 192, 511-523. 25. Raska, I., Reimer, E., Jarnik, M., Kostrouch, Z., and Raska, K., Jr. (1989) Biol. Cell. 65, 79-82. 26. Hirano, T., Konoha, G., Toda, T., and Yanagida, M. (1989) J. CellBiol. 108, 243-253. 27. Belenguer, P., Baldin, V., Mathieu, C., Prats, H., Bensaid, M., Bouche, G., and Amalric, F. (1989) Nucleic Acids Res. 17,66256636. 28. Negri, C., Chiesa, R., Cerino, A., Bestagno, M., Sala, C., Zini, N., Maraldi, N. M., and Astraldi Ricotti, G. C. B. (1992) Exp. Cell Res., 200,452-459.
