Proteome analysis of a rat liver nuclear insoluble protein fraction and localization of a novel protein, ISP36, to compartments in the interchromatin space Masashi Segawa1, Koko Niino1, Reiko Mineki2, Naoko Kaga2, Kimie Murayama2, Kenji Sugimoto3, Yuichi Watanabe4, Kazuhiro Furukawa1,5 and Tsuneyoshi Horigome1,5,6 Natural Science Course, Graduate School of Science and Technology, Niigata University, Japan Division of Proteomics and Biomolecular Science, Biomedical Research Center, Juntendo University Graduate School of Medicine, Tokyo, Japan Division of Applied Biochemistry, Graduate School of Agriculture and Biological Sciences, Osaka Prefecture University, Japan Department of Biology, Niigata University, Japan Department of Chemistry, Niigata University, Japan Center for Transdisciplinary Research, Niigata University, Japan Keywords ISP36, interchromatin space protein, nuclear matrix protein, nuclear protein proteome, insoluble protein proteome Correspondence T Horigome, Department of Chemistry, Faculty of Science, Niigata University, Igarashi-2, Niigata 950-2181, Japan Fax: +81 25 262 6160 Tel: +81 25 262 6160 E-mail: thori@chem.sc.niigata-u.ac.jp (Received 27 March 2005, revised July 2005, accepted July 2005) doi:10.1111/j.1742-4658.2005.04847.x A rat liver nuclear insoluble protein fraction was analyzed to investigate candidate proteins participating in nuclear architecture formation Proteins were subjected to two-dimensional separation by reversed-phase HPLC in 60% formic acid and SDS ⁄ PAGE The method produced good resolution of insoluble proteins One hundred and thirty-eight proteins were separated, and 28 of these were identified The identified proteins included one novel protein, seven known nuclear proteins and 12 known nuclear matrix proteins The novel 36 kDa protein was further investigated for its subnuclear localization The human ortholog of the protein was expressed in Escherichia coli and antibodies were raised against the recombinant protein Exclusive localization of the protein to the nuclear insoluble protein fraction was confirmed by cell fractionation followed by immunoblotting Immunostaining of mouse C3H cells suggested that the 36 kDa protein was a constituent of an insoluble macromolecular complex spread throughout the interchromatin space of the nucleus The protein was designated ‘interchromatin space protein of 36 kDa’, ISP36 In the interior of the cell nucleus, individual reactions contributing to nuclear function occur at ‘specific domains’, rather than in the whole nuclear interior To generate these specific domains, higher order organization of the nuclear interior is necessary Each chromosomal DNA is anchored to a specific ‘chromosome territory’ and does not mix with other chromosomal DNA in the nucleus [1] It is known that proteins important for particular nuclear functions are not spread throughout the whole nuclear interior, but are rather localized to specific compartments as speckles, meshwork structures and so on [2] Within the compartments, proteins participating in a specific function are often present as a complex [2] It is also known that many protein complexes are recruited to, stored in and function in interchromosomal compartments [2] Precise control of these subnuclear compartment structures is crucial for nuclear functions The highly dynamic and flexible structures of the nuclear interior compartments are based on the nuclear envelope which is composed of nuclear membranes, nuclear pore complexes (NPC) and the nuclear lamina Disruption of the integrity of the nuclear lamina by mutation of lamin A ⁄ C causes Emery-Dreifuss Abbreviations DAPI, 4¢,6-diamidino-2-phenylindole; DMEM, Dulbecco’s modified Eagle’s medium; GFP, green fluorescent protein; ISP36, interchromatin space protein of 36 kDa; NPC, nuclear pore complex; PMSF, phenylmethylsulfonyl fluoride; TFA, trifluoroacetic acid FEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS 4327 Interchromatin space protein ISP36 muscular dystrophy and familial partial lipodystrophy [3,4] Mutations of an inner nuclear membrane protein, emerin, also cause Emery-Dreifuss muscular dystrophy [4] These observations suggest that structural components of the nuclear framework are important for the structural integrity and functions of the nucleus [3,4] By contrast, inner nuclear structures have been supposed to be maintained by nuclear matrix filaments [5], but such filaments have not yet been clearly demonstrated in vivo and remain under debate [6,7] Some nuclear proteins present in the interior, such as actin [8], lamins [9,10], NuMA [11,12] and the NPC-associated proteins Nup153 [13] and Tpr [14], are known to form filaments or dot-like structures in the interior of the nucleus However, whether these proteins are responsible for the inner nuclear structure is not yet clear Recently, many integral nuclear membrane proteins have been found by proteome analyses, and some of these are thought to be linked to a variety of dystrophies [15] These findings strongly suggest that the structure and functions of the nuclear envelope are more complicated than previously expected from limited protein members Similar to the nuclear envelope, it is likely that many unknown proteins participate in maintaining the inner nuclear structures such as chromosomal territories, interchromosomal compartments and many kinds of speckles Therefore, searches for and analyses of novel nuclear structural proteins are necessary to gain further insight into the inner nuclear structure, nuclear compartmentalization and higher order nuclear structure and functions In this study, we analyzed a rat liver nuclear insoluble protein fraction to search for candidate novel nuclear structural proteins High resolution over a wide molecular mass range of insoluble proteins was achieved by two-dimensional separation using polymer-based reversed-phase HPLC in 60% formic acid and SDS ⁄ PAGE The separated proteins were identified by partial amino acid sequencing or MS, and consisted of novel proteins along with many known structural proteins One of these novel proteins was shown to be located to compartments in the interchromatin space in the nucleus The protein is suggested to be a constituent of an insoluble macromolecular complex spread throughout the interchromatin space Results We previously developed a reversed-phase HPLC system using polymer-based columns for membrane protein separation [16] Membrane fractions often contain some proteins that are strongly bound to the columns 4328 M Segawa et al and can not be eluted by any solvents other than a strong alkaline solution Therefore, we optimized the separation conditions for membrane proteins by using a polymer-based column in the presence of 60% formic acid [16] The column can be washed with an alkaline solution We applied this method to the separation of a rat liver nuclear insoluble protein fraction as follows Rat liver nuclei were separated by sucrose density gradient centrifugation and treated with RNase A ⁄ DNase I to remove chromatin The obtained fraction contained the nuclear envelope and some of the nuclear matrix proteins This fraction was further extracted with 2% (v ⁄ v) Triton X-100 ⁄ 0.3 m KCl to remove lipids and proteins bound to the nuclear insoluble proteins by ionic and hydrophobic interactions The obtained ‘nuclear insoluble fraction’ was then dissolved in 100% formic acid, applied to a Poros 10R1 column and eluted with a linear gradient of n-butanol in 60% (v ⁄ v) formic acid The elution profile is shown in Fig 1A The collected fractions were further separated by SDS ⁄ PAGE (Fig 1B) One hundred and thirty-eight components, including a protein as large as 217 kDa, were separated Selected protein bands derived from 21 mg of the nuclear insoluble fraction were excised from the gel and digested with trypsin Next, the obtained peptides were separated by octylsilica reversed-phase HPLC and sequenced using a protein sequencer One example is shown in Fig 2A Nineteen samples were analyzed by Edman degradation and 15 samples could be identified from the obtained partial amino acid sequences and the molecular masses estimated by SDS ⁄ PAGE (Table 1) Peptides derived from samples < 10 lg were determined by peptide mass finger printing (Table 1) Eighteen samples were analyzed using the method and 14 samples could be identified In the case of band in Fig 1B, the partial amino acid sequences were further confirmed by MS ⁄ MS analysis (Fig 2B) As shown in Table 1, many filament proteins were identified in the nuclear insoluble protein fraction The major components found in Fig 1B, i.e bands 7, 8, 13, 17, 19, 23 and 25, were lamin A, lamin C, lamin B1, vimentin, keratin type 2, keratin type and actin, respectively, which are all filament proteins Uricase (band 15 in Fig 1B) probably represented a contaminating protein derived from peroxisome cores, because peroxisome cores are composed of uricase protein crystals, often contaminate heavy subcellular fractions such as rough microsomes and are not solubilized by mild detergents [17] We found a novel protein of 36 kDa, corresponding to bands in Fig 1B A cDNA sequence encoding the 36 kDa protein was found in the Universal Protein Resource database (UniProt accession no Q96QD9) FEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS M Segawa et al Interchromatin space protein ISP36 Fig Elution profile of a nuclear insoluble fraction from a Poros-HPLC column (A) A rat liver nuclear insoluble fraction (7 mg) was applied to a Poros 10R1 column (7.5 · 75 mm, 10 lm porous polystyrene beads; PerSeptive Biosystems, Cambridge, MA, USA) buffered with solvent A (60% formic acid, 0.1% TFA, v ⁄ v) and eluted with a 140-min linear gradient from to 40% of Solvent B (33% n-butanol, 60% formic acid, 0.1% TFA, v ⁄ v) at 1.5 mLỈmin)1 Next, the concentration of Solvent B was brought to 100% in The eluate was fractionated as indicated (B) ⁄ 100 volumes of the fractions were analyzed by 8.5% SDS ⁄ PAGE, and the gel was stained with silver The positions of the molecular mass standards are shown on the left The numbered protein bands were identified by amino acid sequencing or MS after tryptic digestion However, the protein had not yet been characterized, although it was referred to as ‘putative 40-2-3 protein’ Therefore, we further determined the localization of this protein in the nucleus as described below The other proteins are discussed in detail in the Discussion Predicted amino acid sequences for the murine, human and bovine 36 kDa protein (no in Table 1) were found in the UniProt database The amino acid sequence of the human ortholog is shown in Fig 3A We designated the protein as ‘interchromatin space protein of 36 kDa’ (ISP36), because the protein was shown to be localized to compartments in the interchromatin space by the results presented below ISP36 is a basic protein with a calculated isoelectric point of 12.5, and contains nuclear localization signal motifs, a helix-turn-helix DNA-binding motif and an intermediate filament protein-like structure as shown in Fig 3A The amino acid sequences of the human, mouse and bovine ISP36 were very similar, as shown in Fig 3B Therefore, we used human ISP36 to raise antibodies against the protein Human ISP36 was expressed as a His6-tagged protein in Escherichia coli and purified by Ni-agarose affinity chromatography in the presence of m urea because the expressed protein formed inclusion bodies (Fig 4A) The purified protein was used to raise antibodies in rabbits The generated antiserum reacted with the antigen, i.e His6-tagged human ISP36 (Fig 4, B3) The antiserum was then applied to the rat liver nuclear insoluble fraction from which ISP36 was originally purified As shown in Fig 4B, a 36 kDa protein was clearly stained with the anti-ISP36 serum FEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS These results indicate that the antibodies raised against human ISP36 cross-react with rat ISP36, and that ISP36 is really present in the rat liver nuclear insoluble fraction Next, we determined the subcellular localization of ISP36 using the antibodies and the rat liver subcellular fractions As shown in Fig 5A, ISP36 was localized to the nuclear fraction (Fig 5A, ‘Anti ISP36’ lane 4) Moreover, the protein was completely localized to the insoluble fraction among the subnuclear fractions (Fig 5B, ‘Anti ISP36’ lane 8) As shown in Fig 5A, affinity-purified anti-ISP36 specifically reacted with ISP36 Therefore, we applied the antibodies to immunostaining of cultured cells When mouse C3H cells were stained with affinity-purified anti-ISP36, the cell nuclei were specifically stained with speckles (Fig 6) When the ISP36 staining was compared with the DNA staining, the ISP36 speckles were dense in regions where the DNA was weakly stained (Fig 6, and 3) When immunoglobulin fractions prepared from the preimmune rabbit were used instead of the affinity-purified anti-ISP36, no staining was observed (Fig 6, 1) These results suggested that ISP36 was present in compartments in the interchromatin space When green fluorescent protein (GFP)– ISP36 was transiently expressed in mouse C3H cells, the fusion protein was completely localized to the nuclei (Fig 7) and appeared as speckles similar to the immunostaining of ISP36 in Fig Within the nucleus, the protein again seemed to be localized to compartments in the interchromatin space (Fig 7) To further confirm the localization of ISP36, an inter4329 Interchromatin space protein ISP36 M Segawa et al stained by 4¢,6-diamidino-2-phenylindole (DAPI; blue) and ISP36 stained with anti-ISP36 (red) were completely separate within the nucleus When two daughter cells in early G1 phase were observed from the upper side, chromatin and ISP36 were present as speckles without overlapping (Fig 9, panel 1) These staining patterns showed that ISP36 may be constrained as a constituent of an insoluble macromolecular complex spread throughout the interchromatin space In prophase, ISP36 staining was partly lost in the interchromatin regions but accumulated in a limited area (Fig 9, panel 5) In prometaphase and metaphase, ISP36 staining was excluded from the interchromatin region and found in the periphery of condensed chromosomes (Fig 9, panels and 7) Some of the ISP36 did not diffuse into the cytoplasm and remained near the chromosomes These results suggest that ISP36 may still be a constituent of some macro-structure In anaphase, most ISP36 disappeared from the near-chromosomal region (Fig 9, panel 8) Taking the results of the immunostaining and GFP– ISP36 expression together, we conclude that ISP36 is a protein present in compartments in the interchromatin space and may be a constituent of an intranuclear structure that is spread throughout the sinusoidal interchromatin space Discussion Fig Identification of proteins by partial amino acid sequencing (A) and MS (B) (A) Twenty micrograms of protein in Fig 1B was digested in-gel with trypsin, and the generated peptides were applied to a reversed-phase HPLC equipped with an octylsilica column (4.8 · 250 mm; Capcel Pack, Shiseido, Tokyo) The peptides were eluted with a linear gradient of 5–75% (v ⁄ v) acetonitrile containing 0.1% (v ⁄ v) TFA at 0.4 mLỈmin)1 The amino acid sequences of the separated peptides numbered 1–3 were determined using a 470A Protein Sequencer (Applied Biosystems) (B) Three micrograms of protein in Fig 1B was digested in-gel with trypsin, and the tryptic peptides were analyzed using an API QSTAR Pulsar (Applied Biosystems) equipped with a micro-HPLC The total ion chromatogram and amino acid sequences estimated by MS ⁄ MS analysis are shown chromatin granule cluster protein, SC35, was costained with ISP36 (Fig 8) The antibody staining patterns were processed using a deconvolution method to improve the images Most of the SC35 speckles (green) were colocalized with the ISP36 staining (red) in the interphase cell nuclei, such that the color became yellow (Fig 8) These results confirmed the presence of ISP36 in compartments in the interchromatin space To precisely examine the localization of ISP36 in the cell cycle, we observed cell nuclei in the different cellcycle stages (Fig 9) As shown in Fig 9, chromatin 4330 We used a perfusion-type polystyrene resin column to separate the nuclear insoluble protein fraction in the presence of 60% (v ⁄ v) formic acid because the system exhibits high resolution for membrane protein separation [16] In addition, column maintenance is easier than for a silica-based column, because the column can be cleaned with 0.5 m NaOH and still maintain high resolution The column showed high resolution in the separation of nuclear insoluble proteins (Fig 1) Protein recovery was expected to be high, because recoveries of > 70% have been shown previously, even for membrane proteins > 140 kDa [16] Partial amino acid sequencing with a protein sequencer and a peptide mass finger printing method using MS were applied to identify the proteins separated by this method (Table 1) If greater amounts of starting material are used, more proteins of lower abundance may be identified because the reverse-phase HPLC system can easily be scaled up This method should therefore be applicable to other proteome analyses of insoluble protein fractions The proteins identified in this study could be classified into five groups as shown in Table Group A contains proteins that are known to be nuclear matrix FEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS M Segawa et al Interchromatin space protein ISP36 Table Proteins identified by partial amino acid sequencing or MS The proteins identified in the rat liver nuclear insoluble fraction are summarized Each ‘No.’ corresponds to that in Fig 1B ‘UniProt Accession no.’ shows the accession number in the Universal Protein Resource Database ‘Molecular mass’ values were obtained from the amino acid sequences ‘S’ and ‘M’ under ‘Identification’ indicate identification by partial amino acid sequencing with a protein sequencer and peptide mass finger printing, respectively ‘No of peptides’ indicates the number of peptides of which partial amino acid sequences were determined Proteins 2, 20 and 28 were further confirmed by amino acid sequencing via nanoelectrospray ionization and quadrupole time-of-flight mass spectrometry Grouping of identified proteins A, Proteins that are known to be nuclear matrix proteins; B, proteins that are known to be nuclear proteins; C, proteins that have not previously been reported to be present in the nucleus; D, proteins derived from contaminating organelles, namely peroxisomes; and E, novel proteins found in this study No Protein 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Heterogeneous nuclear ribonucleoprotein G 36 kDa protein (ISP36) Ribosomal protein L6 Zn finger protein 75 Peptidyl-prolyl-cis-trans isomerase B precursor Fragment of lamin A or C Lamin A Lamin C or lamin A fragment Fragment of lamin A or C Fragment of lamin A or C Fragment of lamin A or C ATP binding cassette, subfamily C, member Lamin B1 Nuclear pore complex protein 98 Uricase Heterogeneous nuclear ribonucleoprotein M Vimentin Keratin, type cytoskeletal Keratin, type cytoskeletal Interleukin enhancer binding factor Keratin, type cytoskeletal 18 Heat shock cognate 71 kDa protein Keratin, type cytoskeletal 18 Actin-binding protein ACF7, neural isoform Actin (a, b or c) 26 27 28 29 Lamina-associated polypeptide 1b ATP binding RNA helicase C06E 1.10 Nuclear pore complex protein 160 Importin b-1 subunit a UniProt Accession no Molecular mass (kDa) Identification No of peptides Sequence coverage (%) Group P84586, P38159 Q96QD9 P21533 P84587, P51815 P24368 P48679, P11516 P48679 P11516, P48679 P48679, P11516 P48679, P11516 P48679, P11516 Q5T2B1 P70615 P49793 P09118 Q62826 P31000 Q10758 Q10758 Q9JIL3 Q5BJY9 P84588 P63018 P05784 7513586a P68134, P60710 P63259 Q5PQX1 P34305 Q9Z0W3 P70168 42 36 34 75 23 36, 48 74 65 74,65 47, 36 65 175 66 98 35 77 54 54 54 98 47 S M M S M M S S S M M M S M S S S M S M M – – – 1 – – – – – 12 – – 68 59 – 69 43, 32 – – – 51, 42 64 29 – 31 – – – 81 – 48 58 B E B A A A A A A A A C A B D A A A A B A 71 47 217 42 S S M S 1 – – – 25 – B A C A 67 130 160 98 S S M M – – – 13 30 A A B B Accession no in the NCBI protein database proteins Filamentous proteins listed in Table 1, i.e lamin B1, lamin A, lamin C, vimentin, keratin types and and actin, were all included in this group Lamins are known to form a meshwork structure underneath the inner nuclear membrane and function to maintain the nuclear architecture [3] The majority of vimentin may be bound to the nuclear lamina from the outside the nucleus [18] The presence of actin in the nucleus has been demonstrated by immunogold staining with partial digestion of the surface lamina of the nuclear matrix to allow penetration of the gold particles into the nuclear matrix [19] and the export FEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS mechanism from the nucleus has been well characterized [20] Nuclear actin may be involved in the regulation of nuclear processes, such as chromatin remodeling [21] LAP1 is an intrinsic inner nuclear membrane protein that participates in anchoring the nuclear lamina to the inner nuclear membrane [22] ZAN 75 protein was recently reported to associate with the nuclear matrix [23] and function as a transcriptional activator [24] Group B contains proteins that are known to be present in the nucleus, but have not yet been reported to be present in the nuclear matrix fraction It was recently reported that a protein 4331 Interchromatin space protein ISP36 Fig Amino acid sequence of human ISP36 (A) The amino acid sequence of human ISP36 (GenBank Accession no AJ344096) Motifs found in the sequence by programs, ‘PROFILE’ and ‘MOTIF FINDER’, were underlined (B) Amino acid sequence identities between human ISP36 and its orthologs The percentages of identical and similar amino acids are indicated M Segawa et al complex containing Nup160 functions to anchor the NPC to the chromatin during NPC reconstitution in late anaphase to early telophase in mitosis, because depletion of the protein complex from a nuclear reconstitution system caused loss of NPC in the reconstituted nuclear envelope [25] NPC protein 98 binds to the protein complex [25] Therefore, these proteins seem to have requisite functions for NPC reconstitution on chromatin Another group B protein, heterogeneous nuclear ribonucleoprotein G, functions in alternative splicing [26] Interleukin enhancer-binding factor binds protein-arginine methyltransferase and contains double-stranded RNA-binding motifs [27] It is not yet known whether these insoluble nuclear proteins participate in nuclear architecture formation Further characterization of these proteins will prove interesting Group C contains proteins that have not previously been reported to be present in the nucleus ATP-binding cassette, subfamily C, member (mrp2) is a member of the multidrug resistance protein family [28] Most members of this family of proteins are located only in plasma membrane, but one member is also located in intracellular vesicles [28] Actin-binding protein ACF7, neural isoform 2, is a member of the dystonin subfamily and the beta-spectrin superfamily Fig Preparation of an anti-ISP36 serum (A) Expression of human His6-ISP36 in E coli and its purification using Ni-agarose beads A1 and A2 are lysates of E coli transformed with the pET28c plasmid encoding His6-ISP36 before and after ISP36 induction, respectively A3 contains lg of His6-ISP36 purified by Ni-agarose beads A1–A3 are stained with Coomassie Brilliant Blue A4–A6 are the same as A1–A3 except that the samples were blotted onto a nitrocellulose sheet, incubated with anti-His sera and then alkaline phosphatase-conjugated secondary antibodies, and developed with 5-bromo-4-chloro-3-indolyl phosphate (B) Purified His6-ISP36 (B1–B3) and a rat liver nuclear insoluble fraction (B4–B6) were electrophoresed and stained with Coomassie Brilliant Blue (B1 and B4) or transferred to a nitrocellulose sheet (B2, B3, B5 and B6) followed by immunostaining For the immunostaining, nitrocellulose strips were incubated with 200-fold diluted anti-ISP36 serum (B3 and B6) or preimmune serum (B2 and B5) and then alkaline phosphatase-conjugated anti-(rabbit IgG), and developed as in (A) The arrowheads indicate ISP36 The bars at the left of panels indicate the positions of marker proteins of 97, 66, 43 and 29 kDa from top to bottom 4332 FEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS M Segawa et al Interchromatin space protein ISP36 A B Fig Subcellular localization of ISP36 (A) Rat liver cytosol (A1), microsome (A2), mitochondria (A3) and nuclear (A4) fractions were electrophoresed, transferred to nitrocellulose sheets and then incubated with preimmune IgG fraction or affinity-purified anti-ISP36 sera After incubation with peroxidase-conjugated secondary antibodies, the sheets were developed with H2O2 and diaminobenzidine (B) A rat liver nuclear fraction was treated with DNase–RNase and the solubilized fraction (chromatin fraction, B5) was separated by centrifugation The pellet was extracted with 0.5 M NaCl (salt extract fraction, B6) and 2% (v ⁄ v) Triton X-100, 0.3 M KCl (Tx ⁄ KCl extract fraction, B7), followed by separation of the insoluble fraction (insoluble fraction, B8) These factions were immunoblotted as in (A) The details of the fractionation are described in the Experimental Procedures Molecular mass markers are same as in Fig [29] This protein functions in cytoplasmic microtubule dynamics to facilitate actin–microtubule interactions [29] The majority of this protein is located in cytoplasmic filaments [29], and whether a proportion of this protein is localized to the nucleus to interact with nuclear actin needs to be examined Group D contains a protein that appears to be a contaminant from peroxisomes as mentioned in the Results section [17].Group E in Table contains the 36 kDa protein The36 kDa protein was characterized in this study and designated ISP36 ISP36 was localized to compartments in the interchromatin space (Figs 6–9) The interchromatin space starts at the nuclear pores [30], and expands between chromatin territories and into their interior [1] The surfaces of compact chromatin domains provide a functionally relevant barrier that can be penetrated by single proteins or small protein aggregates, but not by larger macromolecular complexes above a certain threshold size [1] Therefore, ISP36 is suggested to be a constituFEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS Fig Immunostaining of mouse C3H cells with anti-ISP36 serum Mouse C3H cells cultured on glass coverslips were treated with preimmune IgG fraction (1) or affinity-purified anti-(human ISP36) sera (2 and 3) and then stained with Cy3-conjugated anti-(rabbit IgG) After counterstaining of DNA with DAPI, the cells were observed by fluorescence microscopy The round structure in the DAPI staining corresponds to a cell nucleus Bar, 10 lm Fig Localization of GFP–ISP36 protein transiently expressed in mouse C3H cells A plasmid harboring GFP-tagged ISP36 cDNA was transfected into mouse C3H cells and the GFP–ISP36 protein was transiently expressed The cells were immobilized, and the DNA was stained with DAPI and observed by fluorescence microscopy The round structure in the DAPI staining corresponds to a cell nucleus Bar, 10 lm ent of an insoluble macromolecular complex spread throughout the interchromatin space, because: (a) the protein was fractionated into the nuclear insoluble fraction in the native state (Fig 5); (b) the protein was 4333 Interchromatin space protein ISP36 Fig Localization of ISP36 and SC35 in mouse C3H cell nuclei Mouse C3H cells in interphase were stained by affinity-purified anti-ISP36 as in Fig (red) An SR splicing factor, SC35, which is known to be present in interchromatin granule clusters within the interchromatin space, was stained with an monoclonal anti-SC35 serum (aSC35; BD Biosciences) followed by FITC-labeled anti(mouse IgG) (green), and then observed by fluorescence microscopy The digital image data of a Z series were captured using a CCD camera, processed by deconvolution software (AUTODEBLUR; AUTOQUANT) and presented in 3D (VOLOCITY, IMPROVISION) The round structure corresponds to a cell nucleus Bar, 10 lm localized to compartments in the interchromatin space by anti-ISP36 staining (Figs and 9); (c) the immunostaining did not expand into the chromosomal compartment or the cytoplasm (Figs and 9); and (d) some ISP36 did not diffuse into the cytoplasm and remained near chromatin in metaphase (Fig 9, panel 7) The interchromatin space contains other macromolecular complexes that are required for replication, translation, splicing and repair [1,31] Therefore, ISP36 may also participate in these nuclear functions We examined the in vitro DNA binding ability of ISP36 expressed in E coli because the protein contains a helix-turn-helix DNA-binding motif (Fig 3) ISP36 bound to a nonspecific double-stranded DNA fragment in 0.14 m KCl, although the binding strength was moderate, i.e bound DNA was dissociated by 0.2 m KCl (data not shown) To elucidate the function of this novel protein in compartments in the interchromatin space, analyses of its dynamics using GFP– ISP36 and searches for binding proteins both in vivo and in vitro are necessary as the next step 4334 M Segawa et al Fig Intranuclear localization of ISP36 throughout the cell cycle Mouse C3H cells were stained with affinity-purified anti-(human ISP36) sera (red) and DAPI (blue) as in Fig 6, and then observed by fluorescence microscopy The digital image data were processed as in Fig 1, 2: early G1; 3, 4: G2 to prophase; 5: prophase; 6: prometaphase; 7: metaphase; and 8: anaphase and represent side views of stained nuclei Bars, 10 lm Experimental procedures Buffers Buffers were as follows Buffer A: 10 mm Tris ⁄ HCl, pH 7.4, 0.2 mm MgSO4; m urea buffer: 10 mm Tris, m urea, 0.1 m NaH2PO4, pH 8.0; NaCl ⁄ Pi: 1.5 mm NaH2PO4, 8.1 mm Na2HPO4, 145 mm NaCl Animals Animal care, housing and killing were in accordance with the guidelines of the Ministry of Education, Science, Sports and Culture of Japan for the use of laboratory animals Cell fractionation Subcellular fractions of rat liver were prepared as described previously [32] FEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS M Segawa et al Preparation of rat liver subnuclear fractions Rat liver nuclei were isolated from fasting rats as described previously [33] Next, the isolated nuclei were suspended in 50 mm Tris ⁄ HCl buffer, pH 7.4, containing 0.25 m sucrose, mm MgSO4, and proteinase inhibitors i.e mm phenylmethylsulfonyl fluoride (PMSF), mm benzamidine, 10 lgỈmL)1 of leupeptin and lgỈmL)1 each of antipain, chymostatin and pepstatin A (150–300 U nucleiỈmL)1) The amount of material derived from one A260 unit and mL of isolated nuclei ( · 106 nuclei) was defined as unit (1 U) The suspension was treated with DNase I and RNase A (final concentration of 125 lgỈmL)1 for each enzyme) at °C for h and then centrifuged at 800 g for 15 The supernatant was designated the ‘chromatin fraction’ The pellet was suspended in buffer A (10 mm Tris ⁄ HCl, pH 7.4, 0.2 mm MgSO4) containing protease inhibitors, and the concentration was adjusted to 167 mL)1 The suspension was added to vol of buffer A containing 625 mm NaCl, protease inhibitors and a ⁄ 25 vol of 2-mercaptoethanol After incubation for 30 min, the suspension was centrifuged at 10 000 g for 15 The supernatant was designated the ‘salt extract fraction’ The pellet was washed with buffer A containing 500 mm NaCl and then suspended in buffer A containing protease inhibitors, followed by adjustment of the volume to 500 mL)1 The suspension was added to an equal volume of 40 mm Mes ⁄ KCl, pH 6.0, 4% (v ⁄ v) Triton X-100, 0.6 m KCl, 20% (w ⁄ v) sucrose, mm EDTA containing protease inhibitors and incubated for 30 The suspension was centrifuged at 20 000 g for 30 The supernatant and pellet were designated the ‘Tx ⁄ KCl extract fraction’ and ‘insoluble fraction’, respectively Purification and identification of proteins in the nuclear insoluble fraction Proteins in the rat liver nuclear insoluble fraction were separated by reversed-phase HPLC in 60% (v ⁄ v) formic acid, followed by SDS ⁄ PAGE as described previously [16] For the view of protein elution pattern, gels were stained by silver (Fig 1B) However, for further analysis of interesting proteins, gels were stained with Coomassie Brilliant Blue and protein bands were excised from the gel The excised proteins were identified by mass spectrometry or partial amino acid sequencing using a protein sequencer Samples obtained in quantities > 10 lg were identified by the sequencer Tryptic digestion of the proteins excised from the gel was performed as described previously [34] The mixtures of tryptic peptides were completely dried by vacuum centrifugation, and then dissolved in 2% (v ⁄ v) trifluoroacetic acid (TFA) Next, peptide mapping was performed using a model TSQ 700 triple-stage electrospray ionization quadrupole mass spectrometer (Thermo Electron, San Jose, CA, USA) with a Gilson HPLC system (Gilson, Villiers-le-Bel, FEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS Interchromatin space protein ISP36 France) The HPLC conditions were as follows: Magic C18 column (0.2 · 50 mm; Michrom BioResource, Auburn, CA) eluted with 0.1% (v ⁄ v) formic acid (solvent C) and 0.1% (v ⁄ v) formic acid in 90% (v ⁄ v) CH3CN (solvent D), using a program of 5% (v ⁄ v) solvent D for min, a gradient of 1.05% per for 90 and 100% solvent D for min, at a flow rate of lLỈmin)1 using an Accurate splitter (1 : 100 v ⁄ v; LC Packing, San Francisco, CA, USA) The MS conditions were as follows: ion spray voltage of 2.2–2.7 kV, electron multiplier voltage of 1000 eV, and mass range of 220–2500 (4 s) The proteins were identified using the prowl search engine (http://www.prowl rockefeller.edu/) and the NCBI database The API QSTAR pulsar hybrid mass spectrometer system, consisting of nanoelectrospray ionization and quadrupole time-of-flight, with a microliquid chromatograph (Magic 2002; Michrom BioResource) was used for MS ⁄ MS analysis to confirm the amino acid sequences of the peptides The MS conditions for MS ⁄ MS analysis were as follows: ion spray voltage of 3.0–3.8 kV, electron multiplier voltage of 2200 V, nitrogen 10 unit curtain gas, nitrogen 10 unit collision gas and collision energy of 20–55 eV In cases of partial amino acid sequencing, the peptides generated from the protein were purified by octylsilica reversed-phase HPLC and the partial amino acid sequences were determined using a 470A Protein Sequencer (Applied Biosystems) Proteins were identified from their molecular masses and the obtained partial amino acid sequences using the program fasta (http://www.fasta.genome.ad.jp/) DNA constructs ISP36 cDNA clones for protein expression were prepared using the full-length human ISP36 cDNA sequence A human ISP36 cDNA clone in pSPORT1 (clone ID: DKFZp761B1514; RZPD Deutsches Ressourcenzentrum fur Genomforschung GmbH, Berlin, Germany) was used as a template for PCR with the following synthetic oligonucleotide primers: N, 5¢-CATGAACCGGTTTGGTAC-3¢ as the 5¢ primer, and C, 5¢-CTATCCCACGGTGACAA AGC-3¢ as the 3¢ primer The sequence between these primers was amplified by PCR using Ex Taq DNA polymerase (Takara, Tokyo, Japan) The PCR product was digested with HindIII and XhoI, and the resulting DNA fragment was inserted into the HindIII ⁄ XhoI sites of pET28c for His6-tagged protein expression (Novagen, Madison, WI, USA), or the HindIII ⁄ SalI sites of pEGFP-C2 for GFPtagged protein expression (Clontech, Mountain View, CA, USA), resulting in pEGFP-ISP36 Expression and purification of His6-ISP36 The pET28c expression plasmid encoding His6-ISP36 was transformed into E coli DE3 (BL21) cells, and the cells were grown in Luria–Bertani medium Expression of the 4335 Interchromatin space protein ISP36 fusion protein was induced by addition of 0.1 mm isopropyl thio-b-d-galactoside, followed by incubation for h at 30 °C The bacterial cells were collected and disrupted by sonication, resulting in the recovery of ISP36 as inclusion bodies in the precipitate fraction Next, the inclusion bodies were dissolved in 10 mm Tris, m urea, 0.1 m NaH2PO4, pH 8.0 (8 m urea buffer), and the resulting supernatant was loaded onto a column (1.4 · cm) packed with mL of Ni-NTA beads (Qiagen) The column was washed extensively with m urea buffer containing 10 mm imidazole and the bound His6-ISP36 was then eluted with m urea buffer containing m imidazole and stored at )80 °C until use Preparation of rabbit anti-ISP36 sera An anti-ISP36 serum was raised in female rabbits by immunization with gel pieces containing His6-ISP36 and Freund’s adjuvant [32] The IgG fraction was purified from the antiserum by ammonium sulfate precipitation Purification of anti-ISP36 sera M Segawa et al Cell culture and immunofluorescence Mouse C3H cells were grown in Dulbecco’s modified Eagle’s medium (Nissui Pharmaceutical, Tokyo, Japan) containing 10% (v ⁄ v) fetal bovine serum at 37 °C in a 5% (v ⁄ v) CO2 atmosphere [34] Aliquots of the cells (1–3 · 105) were grown on glass coverslips in 35 mm dishes for 48 h, fixed with 4% paraformaldehyde for 20 and treated with 0.1% (v ⁄ v) Triton X-100 for The fixed cells were incubated with affinity-purified anti-ISP36 sera or preimmune IgG (final concentration, lgỈmL)1), and then with Cy3-conjugated goat anti-rabbit sera (1 : 1500; Jackson Immunoresearch Laboratories, West Grove, PA, USA) After counterstaining with DAPI (1 lgỈmL)1), the cells were observed under a fluorescence microscope (Eclipse E600; Nikon) equipped with a PlanApo ·60 objective (NA 1.40; Nikon) Digital images were obtained using luminavision software (Mac version 1.57; Mitani Corp., Tokyo, Japan) controlling an ORCA-ER cooled CCD camera (Hamamatsu Photonics, Shizuoka, Japan) and a Z-axis motor (Ludl Electronic Products Ltd, Hawthorne, NY, USA) for collecting Z-series optical sections (0.1 lm intervals) His6-ISP36 purified as above (about mg) was dialyzed against 0.9% (w ⁄ v) NaCl, and then precipitated by addition of volumes of cold acetone The precipitate was dissolved in 0.3 m triethanolamine ⁄ HCl, pH 8.0, m urea, and reacted with mL Tresyl Sepharose 4B (Toso, Tokyo, Japan) at room temperature for 2–3 days on a rotator After incubation with 0.1 m monoethanolamine, pH 8.0, m urea at room temperature for h, the affinity resin bearing His6ISP36 was thoroughly washed with 20 mm Tris ⁄ HCl, pH 8.0, 0.1% (w ⁄ v) SDS, 0.5 m NaCl Immunoglobulin fractions obtained from the anti-ISP36 serum were incubated with the affinity resin at °C overnight After washing with 20 mm Tris ⁄ HCl, pH 7.0, 150 mm NaCl, 0.1% (v ⁄ v) Triton X-100, the purified antibodies were eluted with 0.1 m glycine ⁄ HCl (pH 2.5) and immediately neutralized with a Tris buffer The pEGFP-ISP36 plasmid (16 lg) was mixed with 0.6 · 107 cells in 0.5 mL of 30 mm NaCl, 120 mm KCl, mm Na2HPO4, 1.5 mm MgCl2 and electroporated into mouse C3H cells in a cuvette with electrodes at 0.4 cm intervals at a setting of 220 V and a 960 lF capacitor using a Gene Pulser and Capacitance Extender (Bio-Rad, Hercules, CA, USA) After the pulse, the electroporated cells were incubated with an equal volume of serum-free MEM (Nissui Pharmaceutical) for 10 at room temperature, and then aliquots of the cells were grown in an excess amount of Dulbecco’s modified Eagle’s medium on glass coverslips in 35 mm dishes for subsequent microscopic observation Immunoblotting Acknowledgements Samples separated by SDS ⁄ PAGE were electrotransferred onto a nitrocellulose filter The filter was incubated with 1.5 mm NaH2PO4, 8.1 mm Na2HPO4, 145 mm NaCl (NaCl ⁄ Pi) containing 5% nonfat dry milk for h, and then with the same buffer containing the anti-ISP36 serum (1 : 200, v ⁄ v) for h, followed by three washes in NaCl ⁄ Pi containing 0.05% (v ⁄ v) Tween-20 for each The filter was then incubated with horseradish peroxidase-conjugated goat anti-(rabbit IgG) (1 : 1000, v ⁄ v) in NaCl ⁄ Pi containing 5% (w ⁄ v) nonfat dry milk for h, washed five times with NaCl ⁄ Pi containing 0.05% (v ⁄ v) Tween-20 for each, and detected using H2O2 and 3,3¢-diaminobenzidine This work was supported by a grant-in-aid from the Ministry of Education, Science, Sports and Culture of Japan and grants from the Biodesign Research Project of RIKEN and for Project Research of Niigata University 4336 Transient expression of GFP-ISP36 References Cremer T & Cremer C (2001) Chromosome territories, nuclear architecture and gene regulation in mammalian cells Nat Rev Genet 2, 292–301 FEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS M Segawa et al Lamond AI & Spector DL (2003) Nuclear speckles: a model for nuclear organelles Nat Rev Mol Cell Biol 4, 605–612 Hutchison CJ (2002) Lamins: building blocks or regulators of gene expression? Nat Rev Mol Cell Biol 3, 848– 858 Ostlund C & Worman HJ (2003) Nuclear envelope proteins and neuromuscular diseases Muscle Nerve 27, 393–406 Nickerson JA (2001) Experimental observations of a nuclear matrix J Cell Sci 114, 463–474 Pederson T (2000) Half a century of ‘the nuclear matrix’ Mol Biol Cell 11, 799–805 Martelli AM, Falcieri E, Zweyer M, Bortul R, Tabellini G, Cappellini A, Cocco L & Manzoli L (2002) The controversial nuclear matrix: a balanced point of view Histol Histopathol 17, 1193–1205 Amankwah KS & De Boni U (1994) Ultrastructural localization of filamentous actin within neuronal interphase nuclei in situ Exp Cell Res 210, 315–325 Hozak P, Sasseville AMJ, Raymond Y & Cook PR (1995) Lamin proteins form an internal nucleoskeleton as well as a peripheral lamina in human cells J Cell Sci 108, 635–644 10 Jagatheesan G, Thanumalayan S, Muralikrishna B, Rangaraj N, Karande AA & Parnaik VK (1999) Colocalization of intranuclear lamin foci with RNA splicing factors J Cell Sci 112, 4651–4661 11 Gueth-Hallonet C, Wang J, Harborth J, Weber K & Osborn M (1998) Induction of a regular nuclear lattice by overexpression of NuMA Exp Cell Res 243, 434– 452 12 Harborth J, Wang J, Gueth-Hallonet C, Weber K & Osborn M (1999) Self assembly of NuMA: multiarm oligomers as structural units of a nuclear lattice EMBO J 18, 1689–1700 13 Codes VC, Reidenbach S, Kohler A, Stuurman N, van Driel R & Franke WW (1993) Intranuclear filaments containing a nuclear pore complex protein J Cell Biol 123, 1333–1344 14 Codes VC, Reidenbach S, Rackwitz HR & Franke WW (1997) Identification of protein p270 ⁄ Tpr as a constitutive component of the nuclear pore complex-attached intranuclear filaments J Cell Biol 136, 515–529 15 Schirmer EC, Florens L, Guan T, Yates JR III & Gerace L (2003) Nuclear membrane proteins with potential disease links found by subtractive proteomics Science 301, 1380–1382 16 Kikuchi N, Yamakawa Y, Ichimura T, Omata S & Horigome T (1997) Reversed-phase high-performance liquid chromatography of membrane proteins on perfusion-type polystyrene resin columns in 60% formic acid Chromatography 18, 175–184 17 Kreibich G, Ulrich BL & Sabatini DD (1978) Proteins of rough microsomal membranes related to ribosome FEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS Interchromatin space protein ISP36 18 19 20 21 22 23 24 25 26 27 28 29 30 binding I Identification of ribophorins I and II, membrane proteins characteristic of rough microsomes J Cell Biol 77, 464–487 Fey EG, Wan KM & Penman S (1984) Epithelial cytoskeletal framework and nuclear matrix-intermediate filament scaffold: three-dimensional organization and protein composition J Cell Biol 98, 1973–1984 Nakayasu H & Ueda K (1985) Ultrastructural localization of actin in nuclear matrices from mouse leukemia L5178Y cells Cell Struct Funct 10, 305–309 Wada A, Fukuda M, Mishima M & Nishida E (1998) Nuclear export of actin: a novel mechanism regulating the subcellular localization of a major cytoskeletal protein EMBO J 17, 1635–1641 Rando OJ, Zhao K & Crabtree GR (2000) Searching for a function for nuclear actin Trends Cell Biol 10, 92–97 Martin L, Crimaudo C & Gerace L (1995) cDNA cloning and characterization of lamina-associated polypeptide 1C (LAP1C), an integral protein of the inner nuclear membrane J Biol Chem 270, 8822–8828 Lee JY, Kambe M, Hayashi M & Takenaga K (1998) Cloning and characterization of a novel zinc finger protein that associates with nuclear matrix DNA Cell Biol 17, 849–858 Lee JY, Nakane Y, Koshikawa N, Nakayama K, Hayashi M & Takenaga K (2000) Characterization of a zinc finger protein ZAN75: nuclear localization signal, transcriptional activator activity, and expression during neuronal differentiation of P19 cells DNA Cell Biol 19, 227–224 Walther TC, Alves A, Pickersgill H, Loiodice I, Hetzer M, Galy V, Hulsmann BB, Kocher T, Wilm M, Allen T et al (2003) The conserved Nup107–160 complex is critical for nuclear pore complex assembly Cell 113, 195– 206 Nasim MT, Chernova TK, Chowdhury HM, Yue BG & Eperon IC (2003) HnRNP G and Tra2beta: opposite effects on splicing matched by antagonism in RNA binding Hum Mol Genet 12, 1337–1348 Tang J, Kao PN & Herschman HR (2000) Protein-arginine methyltransferase I, the predominant protein-arginine methyltransferase in cells, interacts with and is regulated by interleukin enhancer-binding factor J Biol Chem 275, 19866–19876 Borst P, Evers R, Kool M & Wijnholds J (1999) The multidrug resistance protein family Biochim Biophys Acta 1461, 347–357 Karakesisoglou I, Yang Y & Fuchs E (2000) An epidermal plakin that integrates actin and microtubule networks at cellular junctions J Cell Biol 149, 195– 208 Visser AE, Jaunin F, Fakan S & Aten JA (2000) High resolution analysis of interphase chromosome domains J Cell Sci 113, 2585–2593 4337 Interchromatin space protein ISP36 31 Lamond AI & Spector DL (2003) Nuclear speckles: a model for nuclear organelles Nat Rev Mol Cell Biol 4, 605–612 32 Kawahire S, Tachibana T, Umemoto M, Yoneda Y, Imai N, Saito M, Ichimura T, Omata S & Horigome T (1996) Subcellular distribution and phosphorylation of the nuclear localization signal binding protein, NBP60 Exp Cell Res 222, 385–394 33 Kita K, Omata S & Horigome T (1993) Purification and characterization of a nuclear pore glycoprotein complex containing p62 J Biochem (Tokyo) 113, 377– 382 4338 M Segawa et al 34 Castellanos-Serra L, Proenza W, Huerta V, Moriz R & Simpson R (1999) Proteome analysis of polyacrylamide gel-separated proteins visualized by reversible negative staining using imidazole-zinc salts Electrophoresis 20, 732–737 35 Sugimoto K, Tasaka H & Dotsu M (2001) Molecular behavior in living mitotic cells of human centromere heterochromatin protein HPLalpha ectopically expressed as a fusion to red fluorescent protein Cell Struct Funct 26, 705–718 FEBS Journal 272 (2005) 4327–4338 ª 2005 FEBS ... in Fig Within the nucleus, the protein again seemed to be localized to compartments in the interchromatin space (Fig 7) To further confirm the localization of ISP36, an inter4329 Interchromatin. .. Peptidyl-prolyl-cis-trans isomerase B precursor Fragment of lamin A or C Lamin A Lamin C or lamin A fragment Fragment of lamin A or C Fragment of lamin A or C Fragment of lamin A or C ATP binding cassette, subfamily... interchromosomal compartments and many kinds of speckles Therefore, searches for and analyses of novel nuclear structural proteins are necessary to gain further insight into the inner nuclear structure, nuclear