Proc. NatI. Acad. Sci. USA Vol. 88, pp. 10312-10316, November 1991 Cell Biology
Nuclear matrins: Identification of the major nuclear matrix proteins HIROSHI NAKAYASU* AND RONALD BEREZNEYt Department of Biological Sciences, State University of New York at Buffalo, Buffalo, NY 14260
Communicated by Keith R. Porter, August 23, 1991 (received for review June 27, 1991)
A preparative two-dimensional polyacrylamABSTRACT ide gel system was used to separate and purify the major Coomassie blue-stained proteins from the isolated rat liver nuclear matrix. Approximately 12 major proteins were consistently found. Of these, 5 proteins represented identified proteins, including nuclear lamins A, B, and C, the nucleolar protein B-23, and residual components of core heterogeneous nuclear ribonucleoproteins. The remaining eight major proteins termed the nuclear matrins consisted of matrin 3 (125 kDa, slightly acidic), matrin 4 (105 kDa, basic), matrins D-G (60-75 kDa, basic), and matrins 12 and 13 (42-48 kDa, acidic). Peptide mapping and two-dimensional immunoblot studies indicate that matrins D-G compose two pairs of related proteins (matrins D/E and F/G) and that none of the matrins resemble the nuclear lamins or any of the other major proteins detected on our two-dimensional gels. Subfractionation immunoblot experiments demonstrated the nearly exclusive localization of matrins F/G and other matrins to the nuclear matrix fraction of the cell. These results were further supported by indirect immunofluorescence microscopy that showed a strictly interior nuclear localization of the matrins in intact cells in contrast to the peripherally located nuclear lamins. We conclude that the nuclear matrins are a major class of proteins of the nuclear matrix interior and are distinct from the nuclear lamins.
There is a growing awareness that many of the important answers to questions concerning the expression and regulation of the eukaryotic genome will require an understanding of the higher-order arrangement and function of the genetic components as they interact within the complex threedimensional architecture of the cell nucleus (1-13). The nuclear matrix, a residual nuclear structure that has been isolated from a wide variety of eukaryotic cells throughout the phylogenetic scale (1-3, 6, 7, 13-16), offers a potentially valuable in vitro approach for studying nuclear processes in relation to nuclear structure. Indeed, numerous studies have implicated the matrix as a site of organization for virtually all known nuclear processes (1-11, 13, 16), such as, DNA loop attachment, DNA replication, transcription, RNA splicing and transport, hormone receptor function, carcinogen binding, oncogene proteins, viral proteins, and protein phosphorylation. Despite this progress, our knowledge of the proteins that compose this proteinaceous nucleoskeletal structure is still in its infancy. In this study we have used high-resolution preparative PAGE to identify and purify many of the major Coomassie blue-stained nuclear matrix proteins. A class of nuclear matrix proteins termed the nuclear matrins is identified and characterized by peptide maps, polyclonal antibodies generated against the individual purified matrins, and indirect immunofluorescence microscopy.
MATERIALS AND METHODS Isolation of Nuclei and Nuclear Matrix. Nuclei and nuclear matrix were isolated from rat liver by a modification of the high-salt extraction procedures of Berezney and Coffey (6, 14, 15) as described (17). Phenylmethylsulfonyl fluoride (1 mM) and sodium tetrathionate (0.1 mM) were added to all the isolation solutions through the high salt extraction (2 M NaCl) step (17). Two-Dimensional (2D) Electrophoresis, Purification of Individual Nuclear Matrix Proteins, and Preparation of Chicken Polyclonal Antibodies. The 2D nonequilibrium pH gradient and SDS/PAGE system of O'Farrell et al. (18) was modified to optimize separation of the major nuclear matrix proteins on a large scale. Total rat liver nuclear matrix protein (2-3 mg) was separated on 13-cm tube gels for the firstdimensional pH gradient (pH 3-10 ampholytes, 400 V, 14-16 h, 22°C). The second-dimensional gels measured 15 cm x 30 cm x 1.5 mm and consisted of a 2-cm 3.5% polyacrylamide stacking gel on top of a 28-cm 5-12% gradient gel. Electrophoresis was performed at 15 V for 4 h followed by 250 V for 20-24 h at 4°C. The gels were stained with Coomassie blue and the major stained spots were excised and stored at -20°C. The individual proteins were further purified after extraction from the gel pieces (in 60 mM TrisHCl, pH 6.8/0.1% SDS/2% 2-mercaptoethanol) on one-dimensional (1D) SDS/PAGE gels (10% polyacrylamide) according to Laemmli (19) and stored as gel slices at -20°C. For polyclonal antibody production, multiple gel slices for each protein were washed four times with 10 mM sodium phosphate, pH 7.2/0.15 M NaCl and homogenized in the same buffer with 2 vol of Freund's complete or incomplete adjuvant. Intramuscular and subcutaneous injections into mature laying hens were given with complete adjuvant at day 1 (100-200 gg of protein) and every 2 weeks with incomplete adjuvant (50-100 ,ug of protein). Sera were prepared at day 28 and subsequently every 2 weeks. 2D Western Blot Analysis. Total nuclear matrix protein (100 ,ug) was separated on the 2D gel system described above on a miniscale (5.5-cm tube gels and 5 cm x 7.5 cm x 0.45 mm slab gels) and electrophoretically transferred to a nitrocellulose sheet (20). After blocking with 0.2% Tween 20/10 mM Tris-HCI, pH 7.4/0.15 M NaCl for 2-4 h, the blots were stained with india ink (21) or incubated for 12 h with the various polyclonal antibodies (1: 100 dilution of sera) followed by 2 h with alkaline phosphatase-conjugated goat antichicken IgG secondary antibodies (1:1000 dilution) and developed as described (22). Peptide Analysis of Purified Nuclear Matrix Proteins. Proteins were extracted from the gel slices as described above for the 2D gel pieces, precipitated with 3 vol of absolute ethanol (-20°C, 16 h), and dissolved in 0.1% SDS/60 mM Tris-HCI, pH 6.8. For 1D peptide analysis, 3 1,u of each purified protein Abbreviations: hnRNP, heterogeneous nuclear ribonucleoprotein; 1D and 2D, one and two dimensional, respectively. *Present address: Department of Pharmacology, Kyoto Prefectural University of Medicine, Kawaramachi-Hirokoji, Kamikyo, Kyoto 602, Japan. tTo whom reprint requests should be addressed.
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
10312
Cell
solution (3 jig of protein) was incubated with 2 ,1l of either L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated trypsin (80 ng; Sigma) or protease V8 (300 ng; Sigma) for 1 h at 370C and subjected to SDS/PAGE using either a 15% (protease V8 digested) or 12% (trypsin digested) minigel (19). The gels were stained with Coomassie blue. For 2D analysis, 14 pug of each nuclear matrix protein was incubated with 160 ng of trypsin (370C, 1 h). The reactions were stopped with 0.1 mM phenylmethylsulfonyl fluoride, and the peptides were precipitated with ethanol and subjected to 2D SDS/PAGE using the 10%o polyacrylamide minigel system described above. Affinity Purification of Chicken Polyclonal Antibodies and Indirect Immunofluorescence Microscopy. For immunofluorescence microscopy, the sera containing the polyclonal antibodies were first affinity purified on nitrocellulose. Nuclear matrix proteins from 2D gels were transferred to nitrocellulose (20). Individual spots corresponding to the major proteins were identified by brief staining with india ink (21), excised, and incubated with the corresponding chicken serum (16 h, room temperature). The filter spots were washed four times with 0.2% Tween 20/10 mM TrisHCl, pH 7.4/0.15 M NaCI and distilled water. Affinity-purified antibodies were eluted from the nitrocellulose with 0.1 M glycine hydrochloride (pH 2.2) on ice for 5 min and immediately neutralized with 3 M Trizma base. Immunofluorescence microscopy was performed on rat kangaroo kidney (PtK1) cells grown on coverslips as described (23).
RESULTS 2D Polyacrylamide Gel Electrophoresis and Identification of Nuclear Matrix Proteins. Nuclear matrix proteins from rat liver were separated on a preparative 2D SDS/PAGE system (18). Whereas >50 spots were detected on the 2D gels (Fig. 1), only =12 major Coomassie blue-stained proteins were present. Most of these major nuclear matrix proteins migrated between 60 and 200 kDa with an approximately equal distribution between the acidic and basic sides of the gel. We have numbered these major proteins from 1 to 15 including a more minor protein at position 2, which served as a control Basic
Acidic kDa
2 -
97-
1
II 68-
_*!(i 1t*B) 12
5(F)
6(D)
6(E)
57-
4
for certain experiments, and protein 1, which often appears as a more minor protein by Coomassie blue staining. Since our preparations include the nuclear lamins A, B, and C as major components, we have also labeled a series of four proteins, proteins D-G, that migrated at similar apparent molecular masses (60-75 kDa) but are much more basic in charge. The 2D pattern of the major Coomassie blue-stained polypeptides was extremely reproducible in rat liver and was also characteristic of nuclear matrix isolated from a variety of other mammalian cells including HeLa S-3 (24), Chinese hamster lung cells (Don line), mouse 3T3 fibroblasts, and PtK1 rat kangaroo kidney cells (data not shown). Relationship of Nuclear Matrix Proteins to Known Nuclear Proteins. To determine whether any of the major proteins other than the nuclear lamins were known nuclear proteins, we obtained a variety ofantibodies to nuclear proteins. These included antibodies to core heterogeneous nuclear ribonucleoproteins (hnRNPs) A, B, and C (25-27), other presumptive hnRNPs (26, 27), proteins of the small nuclear RNP particles (28, 29), the nucleolar specific proteins B-23 or numatrin (30, 31), and C-23 or nucleolin (32, 33). Immunoblot analysis showed that spot 14 is B-23 and spot 15 contains hnRNP core proteins (data not shown). All the other antibodies to RNP and nucleolar proteins reacted only with more minor polypeptides on the 2D gels (data not shown). Proteins 3, 4, 12, 13, and D-G are, therefore, uncharacterized major proteins of the nuclear matrix. We have termed these proteins the nuclear matrins to distinguish them from nuclear lamins A, B, and C. Purification and Peptide Mapping of Individual Nuclear Matrix Proteins. We next purified large quantities of individual major matrix proteins from 64 preparative 2D gels (Fig. 2c). The 1D V8 peptide maps of most of the proteins were very different (Fig. 2a). As reported (34), the peptide maps of lamins A and C were very similar to each other but quite distinct from that of lamin B. Since matrins D-G were very resistant to V8 protease, we used trypsin digestion. The digestion patterns of matrins D and E resembled each other a kDa
F G
D
E A
C
*
_
B 12 13 14 15 --
.....
..-
_
q S
-
A BO
*
G_
.
1t(C) A B C D
13
'4
15 (hn RNP)
FIG. 1. Preparative 2D PAGE of nuclear matrix proteins. Total nuclear matrix proteins from rat liver were separated on a preparative (2-3 mg of protein) PAGE system. The major Coomassie blue-stained proteins are numerically labeled including one minor spot (protein 2) and another spot (protein 1) that often stain less intensely than shown. Nuclear lamins A, B, and C are indicated with letters along with a group of matrin proteins that migrate in the same molecular weight range (matrins D-G). A broken line indicates the approximate position of pH 7 on the pH gradient. Molecular mass markers from top to bottom were thyroglobulin, phosphorylase b, bovine serum albumin, pyruvate kinase, ovalbumin, and lactate
dehydrogenase.
4
97684535241814-
45-
(B-23)
3
-7(G)
_W
10313
Proc. Natl. Acad. Sci. USA 88 (1991)
Biology: Nakayasu and Berezney
E F G
A B C D
E F
G
68- __ 45352418-
14c b FIG. 2. 1D peptide maps of nuclear matrix proteins. Nuclear matrix proteins purified from 2D gels (identified by lane labels) were digested with V8 protease (a) or trypsin (b) and subjected to 1D PAGE. (c) Purified proteins before protease digestion. Molecular mass markers were, from top to bottom, bovine serum albumin, ovalbumin, glyceraldehyde 3-phosphate, trypsinogen, lactoglobulin, and lysozyme.
10314
Proc. Natl. Acad Sci. USA 88
Cell Biology: Nakayasu and Berezney
but were very different from those of lamins A and C (Fig. 2b). Matrin G showed a digestion pattern distinct from the lamins and matrins D and E. The 2D peptide maps of lamins A and C were very similar with a few differences indicated in the areas of the gels enclosed by broken lines (Fig. 3 a and b). Lamin B had a very different peptide pattern (results not shown). The 2D peptide maps of the matrin pairs D/E (Fig. 3 c and d) and F/G (Fig. 3 e and f), respectively, were also similar but very different from either the lamins or each other. Characterization of Nuclear Matrins with Polyclonal Antibodies. Polyclonal antibodies were raised in chickens to the purified major matrix proteins. Positive sera were obtained for matrins 3, 4, 13, and A-G. A high degree of specificity was indicated on 2D Western blots. Anti-matrin 3 and 4 antibodies reacted only with matrins 3 and 4, respectively (Fig. 4 i and j). Both anti-lamin A and C antibodies reacted with both lamins A and C but not with lamin B (Fig. 4 c and d). This is consistent with the near identity of lamins A and C in primary structure (35-37). A similar cross reactivity was found with antibodies to matrins D and E (Fig. 4 e and]) and matrins F and G (Fig. 4 g and h), respectively. Antibodies to these proteins, however, did not react with lamins A, B, or C or other matrix proteins. Our immunological results suggest a relationship between the protein pairs lamins A/C, matrins D/E, and matrins F/G and are consistent with the 2D peptide map studies of Fig. 3, which suggested significant similarity in the primary sequence of these pairs of proteins. The chicken sera raised against matrin 13 showed a relatively weak signal on Western blots. It did, however, cross react with matrin 12 (data not shown) suggesting that proteins 12 and 13 may represent another example of a related matrin
a
b
4
B *
a
(1991)
A D F G C E
d
C *- (A,C)
AP- (AC)
f
e *r
(CE)
(0, E)
h
9 -I (F, G)
i
(3)
'
i
(FG)
'(4~)
pair. Cellular Distribution of the Nuclear Matrins. The subcellular distribution of the nuclear matrins in rat liver tissue was then examined on 1D immunoblots (Fig. 5). Matrins F/G were detected in isolated nuclei (lane 7), enriched in the nuclear matrix fraction (lane 11), and depleted in the high salt extract (lane 9), which contains the bulk chromatin and most of the nuclear protein. Matrin F/G was not detected in any of the other subcellular fractions. The absence of a visible
pH Gradient A
-
-
iC
e
1
BA
b
a
~laminA d
Lamin
C
FIG. 4. 2D Western blots of nuclear matrix proteins. (a) Coomassie blue-stained proteins. (b) Preimmune serum control. (c) Anti-lamin A. (d) Anti-lamin C. (e) Anti-matrin D. (f) Anti-matrin E. (g) Anti-matrin F. (h) Anti-matrin G. (i) Anti-matrin 3. (j) Antimatrin 4.
signal in the total liver homogenate fraction was anticipated since nuclear proteins are 200 proteins in the nuclear matrix. Stuurman et al. (46) have also found enormous complexity in the 2D profiles with the sensitive silver staining procedure. Despite this complexity, these studies have provided valuable information. For example, the total nuclear matrix proteins can be separated into two major classes: those found in a variety of cell lines (common matrix proteins) and those that are cell-type and differentiation-state dependent (13, 42, 46, 47). In this study we have developed a 2D PAGE system that optimally separates many of the major nuclear matrix proteins. We detected in rat liver nuclear matrix =12 major Coomassie blue-stained protein spots along with >50 more minor spots. A virtually identical pattern of major Coomassie blue-stained proteins was detected in 2D PAGE profiles of nuclear matrix obtained from a variety of mammalian cell
FIG. 6. Immunofluorescence staining of fixed cells by nuclear matrin antibodies. PtK1 cells were grown on coverslips and fixed with freshly depolymerized 3% (wt/vol) parafortnaldehyde, and immunofluorescence microscopy was performed with the antimatrin polyclonal antibodies. (a) Anti-lamin A/C. (c) Anti-matrin 3. (e) Anti-matrin 4. (g) Anti-matrin D/E. (i) Anti-matrin F/G. (k) Preimmune serum control. (b, d, f, h, j, and 1) Corresponding phase-contrast micrographs. (Bars = 4 jum.)
lines (ref. 24; H.N., L. Buchholtz, and R.B., unpublished data). Antibodies to known nuclear proteins revealed that five of the major Coomassie blue-stained proteins correspond to lamins A, B, and C, B-23 or numatrin, and hnRNP core proteins. The remaining eight proteins (termed nuclear matrins to distinguish them from the nuclear lamins) consisted of matrins 3, 4, D-G, 12, and 13. Matrin 3, a high molecular weight slightly acidic protein of 125 kDa, likely does not correspond to the nuclear matrix protein "mitotin," which migrates to a similar position on 2D gels but is found only in proliferating cells (48). Matrin 4 (105 kDa, basic) likely corresponds to p107, the nuclear-matrix-associated protein described by Smith et al. (49). Matrins 12 and 13 show no obvious relation to identified nuclear-matrix-associated proteins. A possible relationship to the 36- and 40-kDa nuclear matrix proteins reported by Lehner et al. (50) remains to be examined. Matrins D-G likely correspond to an ill-defined
10316
Cell Biology: Nakayasu and Berezney
cluster of 60- to 75-kDa basic nuclear matrix proteins observed in several studies (41, 43, 45, 46). Halikowski and Liew (51) have also identified a presumptive chromatin protein(s) termed B2 that migrated in the same general position as this basic cluster. B2, which is also a highly phosphorylated protein, was shown to be also associated with the nuclear matrix (52). Peptide mapping and antibody cross-reaction studies indicated that the nuclear matrins are distinct from the nuclear lamins. Immunofluorescence microscopy, furthermore, revealed a strictly interior nuclear location for the matrins as opposed to the peripherally located lamins. These results are consistent with Kaufmann and Shaper (45) who reported that components in the cluster of 60- to 75-kDa nuclear matrix proteins had 1D peptide maps different from the lamins and did not recognize anti-lamin antibodies on immunoblots. Our results further indicate that the four proteins identified in this cluster form two pairs of related proteins (matrins D/E and F/G). Matrins 12 and 13 may form a third pair and matrins 3 and 4 showed no relationship to any of the other matrix proteins. We propose that the nuclear matrins compose a broad family of structural proteins in the nucleus with potential subfamilies indicated by the various protein pair homologues. The actual relationship of each putative protein pair, however, will require more detailed molecular studies. In this regard, this laboratory has demonstrated that matrins D/E and F/G specifically bind DNA on 2D Southwestern blots, whereas matins 3, 12, and 13 do not (53). Moreover, the cDNA coding regions for matrins F/G and 3 have been sequenced and are-as predicted from this studycompletely unrelated (23, 54). Consistent with their DNA binding properties, the matrin F/G primary structure contains putative zinc-finger DNA binding motifs (54), but no known DNA binding motif was identified in matrin 3 (P. Belgrader and R.B., unpublished data). We thank the following investigators for their generous gifts of antibodies: Terrence Martin for antibodies to hnRNP core proteins (iD2), small nuclear RNP proteins (SmY12), and the 70-kDa U1 small nuclear RNP protein (2.73); Gideon Dreyfuss for antibodies to hnRNP C1 and C2 core proteins (2B12) and a 120-kDa hnRNP protein (3G6); Mark Olson for antibodies to B-23 (numatrin) and C-23 (nucleolin). We are extremely grateful to Linda A. Buchholtz and Steven Rosenbloom for their expert technical assistance. Jim Stamos provided the illustrations. This work was supported by National Institutes of Health Grant GM-23922 (to R.B.). 1. Berezney, R. & Coffey, D. S. (1976) Adv. Enzyme Regul. 14, 63-100. 2. Berezney, R. (1979) in The Cell Nucleus, ed. Busch, H. (Academic, New York), Vol. 7, pp. 413-456. 3. Shaper, J. H., Pardoll, D. M., Kaufmann, S. H., Barrack, E. R., Vogelstein, B. & Coffey, D. S. (1979) Adv. Enzyme Regul. 17, 213-248. 4. Berezney, R. (1985) UCLA Symp. Mol. Cell. Biol. 26, 99-117. 5. Berezney, R. (1991) J. Cell. Biochem. 47, 109-124. 6. Berezney, R. (1984) in Chromosomal Nonhistone Proteins, ed. Hnilica, L. S. (CRC, Boca Raton, FL), Vol. 4, pp. 119-180. 7. Bouteille, M., Bouvier, D. & Seve, A. P. (1983) Int. Rev. Cytol. 83, 135-182. 8. Jackson, D. A. (1986) Trends Biochem. Sci. 11, 249-252. 9. Gasser, S. M. & Laemmli, U. K. (1987) Trends Genetics 3, 16-22. 10. Bodnar, J. (1988) J. Theor. Biol. 132, 479-507. 11. Nelson, W. G., Pienta, K. J., Barrack, E. R. & Coffey, D. S. (1986) Annu. Rev. Biophys. Biophys. Chem. 15, 457-475. 12. Nigg, E. (1988) Int. Rev. Cytol. 110, 27-92. 13. Nickerson, J. A., He, D., Fey, E. G. & Penman, S. (1990) in The Eukaryotic Nucleus: Molecular Biochemistry and Macromolecular Assemblies, eds. Strauss, P. R. & Wilson, S. H. (Telford, Caldwell, NJ), Vol. 2, pp. 763-782. 14. Berezney, R. & Coffey, D. S. (1974) Biochem. Biophys. Res. Commun. 60, 1410-1417.
Proc. Natl. Acad Sci. USA 88 (1991) 15. Berezney, R. & Coffey, D. S. (1977) J. Cell Biol. 73, 616-637. 16. Agutter, P. S. & Richardson, J. C. W. (1980) J. Cell Sci. 44, 395-435. 17. Basler, J. B., Hastie, N. D., Pietris, D., Matsui, S., Sandberg, A. & Berezney, R. (1981) Biochemistry 20, 6921-6929. 18. O'Farrell, P. Z., Goodman, H. M. & O'Farrell, P. H. (1977) Cell 12, 1133-1142. 19. Laemmli, U. K. (1970) Nature (London) 227, 680-685. 20. Burnett, W. N. (1981) Anal. Biochem. 112, 195-203. 21. Hancock, K. & Tsang, V. C. (1983) Anal. Biochem. 133, 157-162. 22. Blake, M. S., Johnston, K. H., Russell-Johns, G. J. & Gotschlich, E. C. (1984) Anal. Biochem. 136, 175-179. 23. Belgrader, P., Dey, R. & Berezney, R. (1991) J. Biol. Chem. 266, 9893-9899. 24. Belgrader, P., Siegel, A. J. & Berezney, R. (1991) J. Cell Sci.
98, 281-291. 25. Kesser, G. P., Escarla-Wilke, J. & Martin, T. (1984) J. Biol. Chem. 259, 1827-1833. 26. Dreyfuss, G. (1986) Adv. Cell Biol. 2, 459-498. 27. Dreyfuss, G., Choi, Y. D. & Adam, S. A. (1986) Mol. Cell. Biol. 4, 1104-1114. 28. Pettersson, I., Hinterburger, M., Mimori, T., Gottlieb, E. & Steitz, J. (1984) J. Biol. Chem. 259, 5907-5914. 29. Billings, P. B., Allen, R. W., Jensen, F. C. & Hoch, S. O. (1982) J. Immunol. 128, 1176-1180. 30. Feurstein, N., Spiegel, S. & Mond, J. J. (1988) J. Cell Biol. 107, 1629-1642. 31. Feurstein, N., Chan, P. K. & Mond, J. J. (1988) J. Biol. Chem. 263, 10608-10612. 32. Olson, M. 0. J., Wallace, M. O., Herrera, A. M., MarshallCarlson, L. & Hunt, R. C. (1986) Biochemistry 25, 484-491. 33. Lapeyre, B., Bourbon, H. & Amalric, F. (1987) Proc. Natl. Acad. Sci. USA 84, 1472-1476. 34. Kaufmann, S. H., Gibson, W. & Shaper, J. H. (1983) J. Biol. Chem. 258, 2710-2719. 35. Laiberte, J.-F., Dagenais, A., Filion, M., Bibor-Hardy, V., Simard, R. & Royal, A. (1984) J. Cell Biol. 98, 980-985. 36. McKeon, F. D., Kirschner, M. N. & Caput, D. (1986) Nature (London) 319, 463-468. 37. Fisher, D. Z., Chandhary, N. & Blobel, G. (1986) Proc. Natl. Acad. Sci. USA 83, 6450-6454. 38. Fey, E. G., Krochmalnic, G. & Penman, S. (1986) J. Cell Biol. 102, 1654-1665. 39. He, D., Nickerson, J. A. & Penman, S. (1990) J. Cell Biol. 110, 569-580. 40. Peters, K. E., Okada, T. A. & Comings, D. E. (1982) Eur. J. Biochem. 129, 221-232. 41. Verheijen, R., Kuijpers, H., Vooijs, P., van Venrooij, W. & Ramaekers, F. (1986) J. Cell Sci. 86, 173-190. 42. Fey, E. G. & Penman, S. (1988) Proc. Natl. Acad. Sci. USA 85, 121-125. 43. Peters, K. E. & Comings, D. E. (1980) J. Cell Biol. 86, 135-155. 44. Milavetz, B. I. & Edwards, D. R. (1986) J. Cell. Physiol. 127, 388-395. 45. Kaufmann, S. H. & Shaper, J. H. (1984) Exp. Cell Res. 155, 477-495. 46. Stuurman, N., Meijne, A. M. L., van der Pol, A. J., de Jong, L., van Driel, R. & van Renswoude, J. (1990) J. Biol. Chem. 265, 5460-5465. 47. Dworetzky, S. I., Fey, E. G., Penman, S., Lian, J. B., Stein, J. L. & Stein, G. (1990) Proc. Natl. Acad. Sci. USA 87, 4605-4609.
48. Philipova, R. N., Zhelev, N. Z., Todorov, I. T. & Hadjiolov, A. A. (1987) Biol. Cell 60, 1-8. 49. Smith, H. C., Spector, D. L., Woodcock, L. F., Ochs, R. L. & Bhojee, J. (1985) J. Cell Biol. 101, 560-567. 50. Lehner, C. F., Eppenberger, H. M., Fakan, S. & Nigg, E. A. (1986) Exp. Cell Res. 162, 205-219. 51. Halikowski, M. J. & Liew, C. C. (1985) Biochem. J. 225, 357-363. 52. Halikowski, M. J. & Liew, C. C. (1987) Biochem. J. 225, 693-697. 53. Hakes, D. J. & Berezney, R. (1991) J. Biol. Chem. 266, 11131-11140. 54. Hakes, D. J. & Berezney, R. (1991) Proc. Natl. Acad. Sci. USA 88, 6186-6190.
Nuclear matrins: Identification of the major nuclear matrix proteins HIROSHI NAKAYASU* AND RONALD BEREZNEYt Department of Biological Sciences, State University of New York at Buffalo, Buffalo, NY 14260
Communicated by Keith R. Porter, August 23, 1991 (received for review June 27, 1991)
A preparative two-dimensional polyacrylamABSTRACT ide gel system was used to separate and purify the major Coomassie blue-stained proteins from the isolated rat liver nuclear matrix. Approximately 12 major proteins were consistently found. Of these, 5 proteins represented identified proteins, including nuclear lamins A, B, and C, the nucleolar protein B-23, and residual components of core heterogeneous nuclear ribonucleoproteins. The remaining eight major proteins termed the nuclear matrins consisted of matrin 3 (125 kDa, slightly acidic), matrin 4 (105 kDa, basic), matrins D-G (60-75 kDa, basic), and matrins 12 and 13 (42-48 kDa, acidic). Peptide mapping and two-dimensional immunoblot studies indicate that matrins D-G compose two pairs of related proteins (matrins D/E and F/G) and that none of the matrins resemble the nuclear lamins or any of the other major proteins detected on our two-dimensional gels. Subfractionation immunoblot experiments demonstrated the nearly exclusive localization of matrins F/G and other matrins to the nuclear matrix fraction of the cell. These results were further supported by indirect immunofluorescence microscopy that showed a strictly interior nuclear localization of the matrins in intact cells in contrast to the peripherally located nuclear lamins. We conclude that the nuclear matrins are a major class of proteins of the nuclear matrix interior and are distinct from the nuclear lamins.
There is a growing awareness that many of the important answers to questions concerning the expression and regulation of the eukaryotic genome will require an understanding of the higher-order arrangement and function of the genetic components as they interact within the complex threedimensional architecture of the cell nucleus (1-13). The nuclear matrix, a residual nuclear structure that has been isolated from a wide variety of eukaryotic cells throughout the phylogenetic scale (1-3, 6, 7, 13-16), offers a potentially valuable in vitro approach for studying nuclear processes in relation to nuclear structure. Indeed, numerous studies have implicated the matrix as a site of organization for virtually all known nuclear processes (1-11, 13, 16), such as, DNA loop attachment, DNA replication, transcription, RNA splicing and transport, hormone receptor function, carcinogen binding, oncogene proteins, viral proteins, and protein phosphorylation. Despite this progress, our knowledge of the proteins that compose this proteinaceous nucleoskeletal structure is still in its infancy. In this study we have used high-resolution preparative PAGE to identify and purify many of the major Coomassie blue-stained nuclear matrix proteins. A class of nuclear matrix proteins termed the nuclear matrins is identified and characterized by peptide maps, polyclonal antibodies generated against the individual purified matrins, and indirect immunofluorescence microscopy.
MATERIALS AND METHODS Isolation of Nuclei and Nuclear Matrix. Nuclei and nuclear matrix were isolated from rat liver by a modification of the high-salt extraction procedures of Berezney and Coffey (6, 14, 15) as described (17). Phenylmethylsulfonyl fluoride (1 mM) and sodium tetrathionate (0.1 mM) were added to all the isolation solutions through the high salt extraction (2 M NaCl) step (17). Two-Dimensional (2D) Electrophoresis, Purification of Individual Nuclear Matrix Proteins, and Preparation of Chicken Polyclonal Antibodies. The 2D nonequilibrium pH gradient and SDS/PAGE system of O'Farrell et al. (18) was modified to optimize separation of the major nuclear matrix proteins on a large scale. Total rat liver nuclear matrix protein (2-3 mg) was separated on 13-cm tube gels for the firstdimensional pH gradient (pH 3-10 ampholytes, 400 V, 14-16 h, 22°C). The second-dimensional gels measured 15 cm x 30 cm x 1.5 mm and consisted of a 2-cm 3.5% polyacrylamide stacking gel on top of a 28-cm 5-12% gradient gel. Electrophoresis was performed at 15 V for 4 h followed by 250 V for 20-24 h at 4°C. The gels were stained with Coomassie blue and the major stained spots were excised and stored at -20°C. The individual proteins were further purified after extraction from the gel pieces (in 60 mM TrisHCl, pH 6.8/0.1% SDS/2% 2-mercaptoethanol) on one-dimensional (1D) SDS/PAGE gels (10% polyacrylamide) according to Laemmli (19) and stored as gel slices at -20°C. For polyclonal antibody production, multiple gel slices for each protein were washed four times with 10 mM sodium phosphate, pH 7.2/0.15 M NaCl and homogenized in the same buffer with 2 vol of Freund's complete or incomplete adjuvant. Intramuscular and subcutaneous injections into mature laying hens were given with complete adjuvant at day 1 (100-200 gg of protein) and every 2 weeks with incomplete adjuvant (50-100 ,ug of protein). Sera were prepared at day 28 and subsequently every 2 weeks. 2D Western Blot Analysis. Total nuclear matrix protein (100 ,ug) was separated on the 2D gel system described above on a miniscale (5.5-cm tube gels and 5 cm x 7.5 cm x 0.45 mm slab gels) and electrophoretically transferred to a nitrocellulose sheet (20). After blocking with 0.2% Tween 20/10 mM Tris-HCI, pH 7.4/0.15 M NaCl for 2-4 h, the blots were stained with india ink (21) or incubated for 12 h with the various polyclonal antibodies (1: 100 dilution of sera) followed by 2 h with alkaline phosphatase-conjugated goat antichicken IgG secondary antibodies (1:1000 dilution) and developed as described (22). Peptide Analysis of Purified Nuclear Matrix Proteins. Proteins were extracted from the gel slices as described above for the 2D gel pieces, precipitated with 3 vol of absolute ethanol (-20°C, 16 h), and dissolved in 0.1% SDS/60 mM Tris-HCI, pH 6.8. For 1D peptide analysis, 3 1,u of each purified protein Abbreviations: hnRNP, heterogeneous nuclear ribonucleoprotein; 1D and 2D, one and two dimensional, respectively. *Present address: Department of Pharmacology, Kyoto Prefectural University of Medicine, Kawaramachi-Hirokoji, Kamikyo, Kyoto 602, Japan. tTo whom reprint requests should be addressed.
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
10312
Cell
solution (3 jig of protein) was incubated with 2 ,1l of either L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated trypsin (80 ng; Sigma) or protease V8 (300 ng; Sigma) for 1 h at 370C and subjected to SDS/PAGE using either a 15% (protease V8 digested) or 12% (trypsin digested) minigel (19). The gels were stained with Coomassie blue. For 2D analysis, 14 pug of each nuclear matrix protein was incubated with 160 ng of trypsin (370C, 1 h). The reactions were stopped with 0.1 mM phenylmethylsulfonyl fluoride, and the peptides were precipitated with ethanol and subjected to 2D SDS/PAGE using the 10%o polyacrylamide minigel system described above. Affinity Purification of Chicken Polyclonal Antibodies and Indirect Immunofluorescence Microscopy. For immunofluorescence microscopy, the sera containing the polyclonal antibodies were first affinity purified on nitrocellulose. Nuclear matrix proteins from 2D gels were transferred to nitrocellulose (20). Individual spots corresponding to the major proteins were identified by brief staining with india ink (21), excised, and incubated with the corresponding chicken serum (16 h, room temperature). The filter spots were washed four times with 0.2% Tween 20/10 mM TrisHCl, pH 7.4/0.15 M NaCI and distilled water. Affinity-purified antibodies were eluted from the nitrocellulose with 0.1 M glycine hydrochloride (pH 2.2) on ice for 5 min and immediately neutralized with 3 M Trizma base. Immunofluorescence microscopy was performed on rat kangaroo kidney (PtK1) cells grown on coverslips as described (23).
RESULTS 2D Polyacrylamide Gel Electrophoresis and Identification of Nuclear Matrix Proteins. Nuclear matrix proteins from rat liver were separated on a preparative 2D SDS/PAGE system (18). Whereas >50 spots were detected on the 2D gels (Fig. 1), only =12 major Coomassie blue-stained proteins were present. Most of these major nuclear matrix proteins migrated between 60 and 200 kDa with an approximately equal distribution between the acidic and basic sides of the gel. We have numbered these major proteins from 1 to 15 including a more minor protein at position 2, which served as a control Basic
Acidic kDa
2 -
97-
1
II 68-
_*!(i 1t*B) 12
5(F)
6(D)
6(E)
57-
4
for certain experiments, and protein 1, which often appears as a more minor protein by Coomassie blue staining. Since our preparations include the nuclear lamins A, B, and C as major components, we have also labeled a series of four proteins, proteins D-G, that migrated at similar apparent molecular masses (60-75 kDa) but are much more basic in charge. The 2D pattern of the major Coomassie blue-stained polypeptides was extremely reproducible in rat liver and was also characteristic of nuclear matrix isolated from a variety of other mammalian cells including HeLa S-3 (24), Chinese hamster lung cells (Don line), mouse 3T3 fibroblasts, and PtK1 rat kangaroo kidney cells (data not shown). Relationship of Nuclear Matrix Proteins to Known Nuclear Proteins. To determine whether any of the major proteins other than the nuclear lamins were known nuclear proteins, we obtained a variety ofantibodies to nuclear proteins. These included antibodies to core heterogeneous nuclear ribonucleoproteins (hnRNPs) A, B, and C (25-27), other presumptive hnRNPs (26, 27), proteins of the small nuclear RNP particles (28, 29), the nucleolar specific proteins B-23 or numatrin (30, 31), and C-23 or nucleolin (32, 33). Immunoblot analysis showed that spot 14 is B-23 and spot 15 contains hnRNP core proteins (data not shown). All the other antibodies to RNP and nucleolar proteins reacted only with more minor polypeptides on the 2D gels (data not shown). Proteins 3, 4, 12, 13, and D-G are, therefore, uncharacterized major proteins of the nuclear matrix. We have termed these proteins the nuclear matrins to distinguish them from nuclear lamins A, B, and C. Purification and Peptide Mapping of Individual Nuclear Matrix Proteins. We next purified large quantities of individual major matrix proteins from 64 preparative 2D gels (Fig. 2c). The 1D V8 peptide maps of most of the proteins were very different (Fig. 2a). As reported (34), the peptide maps of lamins A and C were very similar to each other but quite distinct from that of lamin B. Since matrins D-G were very resistant to V8 protease, we used trypsin digestion. The digestion patterns of matrins D and E resembled each other a kDa
F G
D
E A
C
*
_
B 12 13 14 15 --
.....
..-
_
q S
-
A BO
*
G_
.
1t(C) A B C D
13
'4
15 (hn RNP)
FIG. 1. Preparative 2D PAGE of nuclear matrix proteins. Total nuclear matrix proteins from rat liver were separated on a preparative (2-3 mg of protein) PAGE system. The major Coomassie blue-stained proteins are numerically labeled including one minor spot (protein 2) and another spot (protein 1) that often stain less intensely than shown. Nuclear lamins A, B, and C are indicated with letters along with a group of matrin proteins that migrate in the same molecular weight range (matrins D-G). A broken line indicates the approximate position of pH 7 on the pH gradient. Molecular mass markers from top to bottom were thyroglobulin, phosphorylase b, bovine serum albumin, pyruvate kinase, ovalbumin, and lactate
dehydrogenase.
4
97684535241814-
45-
(B-23)
3
-7(G)
_W
10313
Proc. Natl. Acad. Sci. USA 88 (1991)
Biology: Nakayasu and Berezney
E F G
A B C D
E F
G
68- __ 45352418-
14c b FIG. 2. 1D peptide maps of nuclear matrix proteins. Nuclear matrix proteins purified from 2D gels (identified by lane labels) were digested with V8 protease (a) or trypsin (b) and subjected to 1D PAGE. (c) Purified proteins before protease digestion. Molecular mass markers were, from top to bottom, bovine serum albumin, ovalbumin, glyceraldehyde 3-phosphate, trypsinogen, lactoglobulin, and lysozyme.
10314
Proc. Natl. Acad Sci. USA 88
Cell Biology: Nakayasu and Berezney
but were very different from those of lamins A and C (Fig. 2b). Matrin G showed a digestion pattern distinct from the lamins and matrins D and E. The 2D peptide maps of lamins A and C were very similar with a few differences indicated in the areas of the gels enclosed by broken lines (Fig. 3 a and b). Lamin B had a very different peptide pattern (results not shown). The 2D peptide maps of the matrin pairs D/E (Fig. 3 c and d) and F/G (Fig. 3 e and f), respectively, were also similar but very different from either the lamins or each other. Characterization of Nuclear Matrins with Polyclonal Antibodies. Polyclonal antibodies were raised in chickens to the purified major matrix proteins. Positive sera were obtained for matrins 3, 4, 13, and A-G. A high degree of specificity was indicated on 2D Western blots. Anti-matrin 3 and 4 antibodies reacted only with matrins 3 and 4, respectively (Fig. 4 i and j). Both anti-lamin A and C antibodies reacted with both lamins A and C but not with lamin B (Fig. 4 c and d). This is consistent with the near identity of lamins A and C in primary structure (35-37). A similar cross reactivity was found with antibodies to matrins D and E (Fig. 4 e and]) and matrins F and G (Fig. 4 g and h), respectively. Antibodies to these proteins, however, did not react with lamins A, B, or C or other matrix proteins. Our immunological results suggest a relationship between the protein pairs lamins A/C, matrins D/E, and matrins F/G and are consistent with the 2D peptide map studies of Fig. 3, which suggested significant similarity in the primary sequence of these pairs of proteins. The chicken sera raised against matrin 13 showed a relatively weak signal on Western blots. It did, however, cross react with matrin 12 (data not shown) suggesting that proteins 12 and 13 may represent another example of a related matrin
a
b
4
B *
a
(1991)
A D F G C E
d
C *- (A,C)
AP- (AC)
f
e *r
(CE)
(0, E)
h
9 -I (F, G)
i
(3)
'
i
(FG)
'(4~)
pair. Cellular Distribution of the Nuclear Matrins. The subcellular distribution of the nuclear matrins in rat liver tissue was then examined on 1D immunoblots (Fig. 5). Matrins F/G were detected in isolated nuclei (lane 7), enriched in the nuclear matrix fraction (lane 11), and depleted in the high salt extract (lane 9), which contains the bulk chromatin and most of the nuclear protein. Matrin F/G was not detected in any of the other subcellular fractions. The absence of a visible
pH Gradient A
-
-
iC
e
1
BA
b
a
~laminA d
Lamin
C
FIG. 4. 2D Western blots of nuclear matrix proteins. (a) Coomassie blue-stained proteins. (b) Preimmune serum control. (c) Anti-lamin A. (d) Anti-lamin C. (e) Anti-matrin D. (f) Anti-matrin E. (g) Anti-matrin F. (h) Anti-matrin G. (i) Anti-matrin 3. (j) Antimatrin 4.
signal in the total liver homogenate fraction was anticipated since nuclear proteins are 200 proteins in the nuclear matrix. Stuurman et al. (46) have also found enormous complexity in the 2D profiles with the sensitive silver staining procedure. Despite this complexity, these studies have provided valuable information. For example, the total nuclear matrix proteins can be separated into two major classes: those found in a variety of cell lines (common matrix proteins) and those that are cell-type and differentiation-state dependent (13, 42, 46, 47). In this study we have developed a 2D PAGE system that optimally separates many of the major nuclear matrix proteins. We detected in rat liver nuclear matrix =12 major Coomassie blue-stained protein spots along with >50 more minor spots. A virtually identical pattern of major Coomassie blue-stained proteins was detected in 2D PAGE profiles of nuclear matrix obtained from a variety of mammalian cell
FIG. 6. Immunofluorescence staining of fixed cells by nuclear matrin antibodies. PtK1 cells were grown on coverslips and fixed with freshly depolymerized 3% (wt/vol) parafortnaldehyde, and immunofluorescence microscopy was performed with the antimatrin polyclonal antibodies. (a) Anti-lamin A/C. (c) Anti-matrin 3. (e) Anti-matrin 4. (g) Anti-matrin D/E. (i) Anti-matrin F/G. (k) Preimmune serum control. (b, d, f, h, j, and 1) Corresponding phase-contrast micrographs. (Bars = 4 jum.)
lines (ref. 24; H.N., L. Buchholtz, and R.B., unpublished data). Antibodies to known nuclear proteins revealed that five of the major Coomassie blue-stained proteins correspond to lamins A, B, and C, B-23 or numatrin, and hnRNP core proteins. The remaining eight proteins (termed nuclear matrins to distinguish them from the nuclear lamins) consisted of matrins 3, 4, D-G, 12, and 13. Matrin 3, a high molecular weight slightly acidic protein of 125 kDa, likely does not correspond to the nuclear matrix protein "mitotin," which migrates to a similar position on 2D gels but is found only in proliferating cells (48). Matrin 4 (105 kDa, basic) likely corresponds to p107, the nuclear-matrix-associated protein described by Smith et al. (49). Matrins 12 and 13 show no obvious relation to identified nuclear-matrix-associated proteins. A possible relationship to the 36- and 40-kDa nuclear matrix proteins reported by Lehner et al. (50) remains to be examined. Matrins D-G likely correspond to an ill-defined
10316
Cell Biology: Nakayasu and Berezney
cluster of 60- to 75-kDa basic nuclear matrix proteins observed in several studies (41, 43, 45, 46). Halikowski and Liew (51) have also identified a presumptive chromatin protein(s) termed B2 that migrated in the same general position as this basic cluster. B2, which is also a highly phosphorylated protein, was shown to be also associated with the nuclear matrix (52). Peptide mapping and antibody cross-reaction studies indicated that the nuclear matrins are distinct from the nuclear lamins. Immunofluorescence microscopy, furthermore, revealed a strictly interior nuclear location for the matrins as opposed to the peripherally located lamins. These results are consistent with Kaufmann and Shaper (45) who reported that components in the cluster of 60- to 75-kDa nuclear matrix proteins had 1D peptide maps different from the lamins and did not recognize anti-lamin antibodies on immunoblots. Our results further indicate that the four proteins identified in this cluster form two pairs of related proteins (matrins D/E and F/G). Matrins 12 and 13 may form a third pair and matrins 3 and 4 showed no relationship to any of the other matrix proteins. We propose that the nuclear matrins compose a broad family of structural proteins in the nucleus with potential subfamilies indicated by the various protein pair homologues. The actual relationship of each putative protein pair, however, will require more detailed molecular studies. In this regard, this laboratory has demonstrated that matrins D/E and F/G specifically bind DNA on 2D Southwestern blots, whereas matins 3, 12, and 13 do not (53). Moreover, the cDNA coding regions for matrins F/G and 3 have been sequenced and are-as predicted from this studycompletely unrelated (23, 54). Consistent with their DNA binding properties, the matrin F/G primary structure contains putative zinc-finger DNA binding motifs (54), but no known DNA binding motif was identified in matrin 3 (P. Belgrader and R.B., unpublished data). We thank the following investigators for their generous gifts of antibodies: Terrence Martin for antibodies to hnRNP core proteins (iD2), small nuclear RNP proteins (SmY12), and the 70-kDa U1 small nuclear RNP protein (2.73); Gideon Dreyfuss for antibodies to hnRNP C1 and C2 core proteins (2B12) and a 120-kDa hnRNP protein (3G6); Mark Olson for antibodies to B-23 (numatrin) and C-23 (nucleolin). We are extremely grateful to Linda A. Buchholtz and Steven Rosenbloom for their expert technical assistance. Jim Stamos provided the illustrations. This work was supported by National Institutes of Health Grant GM-23922 (to R.B.). 1. Berezney, R. & Coffey, D. S. (1976) Adv. Enzyme Regul. 14, 63-100. 2. Berezney, R. (1979) in The Cell Nucleus, ed. Busch, H. (Academic, New York), Vol. 7, pp. 413-456. 3. Shaper, J. H., Pardoll, D. M., Kaufmann, S. H., Barrack, E. R., Vogelstein, B. & Coffey, D. S. (1979) Adv. Enzyme Regul. 17, 213-248. 4. Berezney, R. (1985) UCLA Symp. Mol. Cell. Biol. 26, 99-117. 5. Berezney, R. (1991) J. Cell. Biochem. 47, 109-124. 6. Berezney, R. (1984) in Chromosomal Nonhistone Proteins, ed. Hnilica, L. S. (CRC, Boca Raton, FL), Vol. 4, pp. 119-180. 7. Bouteille, M., Bouvier, D. & Seve, A. P. (1983) Int. Rev. Cytol. 83, 135-182. 8. Jackson, D. A. (1986) Trends Biochem. Sci. 11, 249-252. 9. Gasser, S. M. & Laemmli, U. K. (1987) Trends Genetics 3, 16-22. 10. Bodnar, J. (1988) J. Theor. Biol. 132, 479-507. 11. Nelson, W. G., Pienta, K. J., Barrack, E. R. & Coffey, D. S. (1986) Annu. Rev. Biophys. Biophys. Chem. 15, 457-475. 12. Nigg, E. (1988) Int. Rev. Cytol. 110, 27-92. 13. Nickerson, J. A., He, D., Fey, E. G. & Penman, S. (1990) in The Eukaryotic Nucleus: Molecular Biochemistry and Macromolecular Assemblies, eds. Strauss, P. R. & Wilson, S. H. (Telford, Caldwell, NJ), Vol. 2, pp. 763-782. 14. Berezney, R. & Coffey, D. S. (1974) Biochem. Biophys. Res. Commun. 60, 1410-1417.
Proc. Natl. Acad Sci. USA 88 (1991) 15. Berezney, R. & Coffey, D. S. (1977) J. Cell Biol. 73, 616-637. 16. Agutter, P. S. & Richardson, J. C. W. (1980) J. Cell Sci. 44, 395-435. 17. Basler, J. B., Hastie, N. D., Pietris, D., Matsui, S., Sandberg, A. & Berezney, R. (1981) Biochemistry 20, 6921-6929. 18. O'Farrell, P. Z., Goodman, H. M. & O'Farrell, P. H. (1977) Cell 12, 1133-1142. 19. Laemmli, U. K. (1970) Nature (London) 227, 680-685. 20. Burnett, W. N. (1981) Anal. Biochem. 112, 195-203. 21. Hancock, K. & Tsang, V. C. (1983) Anal. Biochem. 133, 157-162. 22. Blake, M. S., Johnston, K. H., Russell-Johns, G. J. & Gotschlich, E. C. (1984) Anal. Biochem. 136, 175-179. 23. Belgrader, P., Dey, R. & Berezney, R. (1991) J. Biol. Chem. 266, 9893-9899. 24. Belgrader, P., Siegel, A. J. & Berezney, R. (1991) J. Cell Sci.
98, 281-291. 25. Kesser, G. P., Escarla-Wilke, J. & Martin, T. (1984) J. Biol. Chem. 259, 1827-1833. 26. Dreyfuss, G. (1986) Adv. Cell Biol. 2, 459-498. 27. Dreyfuss, G., Choi, Y. D. & Adam, S. A. (1986) Mol. Cell. Biol. 4, 1104-1114. 28. Pettersson, I., Hinterburger, M., Mimori, T., Gottlieb, E. & Steitz, J. (1984) J. Biol. Chem. 259, 5907-5914. 29. Billings, P. B., Allen, R. W., Jensen, F. C. & Hoch, S. O. (1982) J. Immunol. 128, 1176-1180. 30. Feurstein, N., Spiegel, S. & Mond, J. J. (1988) J. Cell Biol. 107, 1629-1642. 31. Feurstein, N., Chan, P. K. & Mond, J. J. (1988) J. Biol. Chem. 263, 10608-10612. 32. Olson, M. 0. J., Wallace, M. O., Herrera, A. M., MarshallCarlson, L. & Hunt, R. C. (1986) Biochemistry 25, 484-491. 33. Lapeyre, B., Bourbon, H. & Amalric, F. (1987) Proc. Natl. Acad. Sci. USA 84, 1472-1476. 34. Kaufmann, S. H., Gibson, W. & Shaper, J. H. (1983) J. Biol. Chem. 258, 2710-2719. 35. Laiberte, J.-F., Dagenais, A., Filion, M., Bibor-Hardy, V., Simard, R. & Royal, A. (1984) J. Cell Biol. 98, 980-985. 36. McKeon, F. D., Kirschner, M. N. & Caput, D. (1986) Nature (London) 319, 463-468. 37. Fisher, D. Z., Chandhary, N. & Blobel, G. (1986) Proc. Natl. Acad. Sci. USA 83, 6450-6454. 38. Fey, E. G., Krochmalnic, G. & Penman, S. (1986) J. Cell Biol. 102, 1654-1665. 39. He, D., Nickerson, J. A. & Penman, S. (1990) J. Cell Biol. 110, 569-580. 40. Peters, K. E., Okada, T. A. & Comings, D. E. (1982) Eur. J. Biochem. 129, 221-232. 41. Verheijen, R., Kuijpers, H., Vooijs, P., van Venrooij, W. & Ramaekers, F. (1986) J. Cell Sci. 86, 173-190. 42. Fey, E. G. & Penman, S. (1988) Proc. Natl. Acad. Sci. USA 85, 121-125. 43. Peters, K. E. & Comings, D. E. (1980) J. Cell Biol. 86, 135-155. 44. Milavetz, B. I. & Edwards, D. R. (1986) J. Cell. Physiol. 127, 388-395. 45. Kaufmann, S. H. & Shaper, J. H. (1984) Exp. Cell Res. 155, 477-495. 46. Stuurman, N., Meijne, A. M. L., van der Pol, A. J., de Jong, L., van Driel, R. & van Renswoude, J. (1990) J. Biol. Chem. 265, 5460-5465. 47. Dworetzky, S. I., Fey, E. G., Penman, S., Lian, J. B., Stein, J. L. & Stein, G. (1990) Proc. Natl. Acad. Sci. USA 87, 4605-4609.
48. Philipova, R. N., Zhelev, N. Z., Todorov, I. T. & Hadjiolov, A. A. (1987) Biol. Cell 60, 1-8. 49. Smith, H. C., Spector, D. L., Woodcock, L. F., Ochs, R. L. & Bhojee, J. (1985) J. Cell Biol. 101, 560-567. 50. Lehner, C. F., Eppenberger, H. M., Fakan, S. & Nigg, E. A. (1986) Exp. Cell Res. 162, 205-219. 51. Halikowski, M. J. & Liew, C. C. (1985) Biochem. J. 225, 357-363. 52. Halikowski, M. J. & Liew, C. C. (1987) Biochem. J. 225, 693-697. 53. Hakes, D. J. & Berezney, R. (1991) J. Biol. Chem. 266, 11131-11140. 54. Hakes, D. J. & Berezney, R. (1991) Proc. Natl. Acad. Sci. USA 88, 6186-6190.
