![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 18, Issue 3, 1107-1117, March 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*Department of Molecular Radiobiology, Division of Genome Biology, Research Institute for Radiation Biology and Medicine, and Department of Oral Maxillofacial Pathobiology, Division of Frontier Medical Science, Graduate School of Biomedical Sciences, Hiroshima University, Hiroshima 734-8553, Japan
Submitted November 17, 2006;
Revised December 15, 2006;
Accepted December 29, 2006
Monitoring Editor: A. Gregory Matera
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Nucleoli are generally composed of three morphologically distinct regions; the fibrillar centers containing RNA polymerase I and its associated transcription factors, a dense fibrillar component that surrounds the fibrillar centers, and a granular component that is the site of preribosome assembly and preribosomal particle synthesis (Stoykova et al., 1985). At the onset of mitosis, rDNA transcription is repressed by cyclin-dependent kinase (CDK) 1/Cdc2 cyclin-dependent kinase–Cyclin B-directed phosphorylation of components of the rDNA transcription machinery, and nucleolar disassembly is induced (Heix et al., 1998; Sirri et al., 2000). The RNA synthesis machinery for rDNA in the fibrillar centers remains associated with mitotic chromosomes in an inactive state at nucleolar-organizing regions (Stoykova et al., 1985). Other RNA processing components, such as fibrillarin, nucleolar protein Nop52, NPM1/nucleophosmin/B23/Numatrin/Nucleolar protein NO38, and nucleolin/C23/100-kDa nucleolar protein, are located at the periphery of chromosomes, away from nucleolar-organizing regions during mitosis (Jimenez-Garcia et al., 1994; Pinol-Roma, 1999; Savino et al., 1999; Dousset et al., 2000). CDK1-Cyclin B also phosphorylates NPM1 and nucleolin (Peter et al., 1990), and the disassembly is apparently driven by mitotic phosphorylation of nucleolar components. Inhibition of CDK1-Cyclin B during mitosis is sufficient to cause the resumption of rDNA transcription, but it is not sufficient to restore proper processing of RNA biogenesis or total rebuilding of the nucleolar processing machinery (Sirri et al., 2002). Thus, repression of the nucleolar assembly is spatiotemporally and multifactorially regulated.
Aurora-B is a conserved protein kinase essential for the segregation of eukaryotic chromosomes (Andrews et al., 2003; Carmena and Earnshaw, 2003; Yanagida, 2005), and it forms the mitotic passenger protein complex with inner centromere protein (INCENP), Survivin, and Borealin/Dasra (Carmena and Earnshaw, 2003; Gassmann et al., 2004; Sampath et al., 2004; Giet et al., 2005). In early mitosis, Aurora-B associates with centromeric heterochromatin and then moves to the midzone spindle in anaphase, to the cleavage furrow in telophase, and finally to both ends of the midbody in cytokinesis (Schumacher et al., 1998; Terada et al., 1998; Adams et al., 2001; Murata-Hori et al., 2002). Mitotic activation of Aurora-B is triggered by autophosphorylation, which is stimulated by association with and phosphorylation of its substrate INCENP. Survivin also stimulates Aurora-B activity by a direct interaction, and it is a substrate for Aurora-B (Wheatley et al., 2001). Borealin is also an Aurora-B substrate but does not stimulate Aurora-B activity. There are at least two different types of Aurora-B complex present during mitosis, suggesting that each complex targets different sites or contains different components (Gassmann et al., 2004; Giet et al., 2005). In addition, the mitotic passenger movements of Aurora-B are accompanied by continuous exchange with the surrounding cytoplasmic pool (Murata-Hori et al., 2002). The requirements for centromere or cleavage furrow targeting of Aurora-B and chromosomal and cytoskeletal Aurora-B substrates for each mitotic process are now being elucidated; however, the mitotic functions of Aurora-B may not be limited to cytokinesis or other processes of chromosome segregation.
Here, we report a novel substrate of Aurora-B, which contains an NOL1/NOP2/sun domain and is the second member of the NSUN family (NSUN2) of RNA methyltransferase. We have shown that NSUN2 has RNA methyltransferase activity that is repressed by Aurora-B–provoked phosphorylation and that the phosphorylation causes NSUN2 to dissociate from its nucleolar binding protein, NPM1. Phosphorylation of NSUN2 by Aurora-B seems to represent a novel mechanism for the regulation of nucleolar architecture and nucleic acids metabolism during mitosis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Aurora Inhibitors
Hesperadin and ZM447439 were obtained from Boehringer Ingelheim (Ingelheim, Germany) and AstraZeneca Pharmaceuticals LP (Wilmington, DE), respectively. Synchronized HeLa cells were treated with 100 nM Hesperadin or 2 µM ZM447439 for 2 h (from hours 8–10).
Mammalian Expression Plasmids and Transfection
5'-Rapid amplification of cDNA ends (RACE) polymerase chain reaction (PCR) products were generated using primers corresponding to peptides lep56 and lep77-1 (Supplemental Figure S1). The full-length cDNA was isolated from the Okayama-Berg full-length cDNA library (produced from HeLa cells) by using the colony hybridization technique with the 5'-RACE PCR products. The obtained full-length cDNA was then sequenced. The coding region of this cDNA was subcloned into pcDNA3.1(–) for mammalian expression of tag-free NSUN2, generating plasmid pcDNA3.1(–)-NSUN2. The coding region was also subcloned into pcDNA3.1/His for mammalian expression of Xpress-NSUN2, generating plasmid pcDNA3.1-Xpress-NSUN2. The following mutant NSUN2 expression plasmids were also constructed: pcDNA3.1(–)-NSUN2-SA and pcDNA3.1-Xpress-NSUN2-SA, producing dephosphorylation-mimic NSUN2-S139A, in which Ser139 of NSUN2 was replaced with Ala by site-directed mutation at T410G; and pcDNA3.1(–)-NSUN2-SE and pcDNA3.1-Xpress-NSUN2-SE, producing phosphorylation-mimic NSUN2-S139E in which Ser139 of NSUN2 was replaced with Glu by site-directed mutation at T410G and C411A. Other mammalian expression vectors were also generated: pcDNA3-FLAG-Aurora-A-WT and -K/R (Tatsuka et al., 2005); pcDNA3-FLAG-Aurora-B-WT and -K/R (Kanda et al., 2005); pcDNA3-FLAG-Aurora-C-WT and -K/R (constructed from human Aurora-C cDNA isolated from the HeLa cDNA library by reverse transcription-PCR); and pEGFP-
Cyclin B1 (constructed from the Cyclin B1 expression plasmid supplied by M. Brandeis, Department of Genetics, Silberman Institute of Life Sciences, The Hebrew University of Jerusalem, Jerusalem, Israel). Transfection was performed using Lipofectamine 2000 (Invitrogen, Carlsbad, CA).
Short Hairpin RNA (shRNA) Constructs
Oligonucleotides corresponding to Aurora or NSUN2 genes were synthesized as shown in Supplemental Figure S2. The oligonucleotides were then ligated into pSUPERIOR.PURO (OligoEngine). The mixture of each Aurora-shRNA was used to repress Aurora expression in HeLa cells. HeLa cell lines expressing each NSUN2-shRNA plasmid construct were established by puromycin selection (Supplemental Figure S3). We used clone 5 from NSUN2-shRNA 3 construct-transfected HeLa cells as HeLa-NSUN2-KD cells.
Antibodies
We raised a rabbit polyclonal antibody (
H3-P) against a phosphorylated synthetic Histone H3 peptide (ARKS*TGGKAPRKQL, where S* indicates the phosphorylated serine). Collected serum was affinity purified using the peptide. Rabbit polyclonal antibodies, NSUN2 and NSUN2-full, were raised against the C-terminal peptide (GCDPAGVHPPR) of NSUN2 and bacterially expressed full-length His-NSUN2, respectively. The bacterially expressed protein was isolated from the lysate of pRSET(C)-NSUN2-WT–expressing Escherichia coli cells by using a Talon nickel column (BD Biosciences, San Jose, CA) (Supplemental Figure S4). Other antibodies used in the experiments included the following: polyclonal rabbit anti-Cdc25A (Jinno et al., 1994) and anti-Human Aurora-C antibodies (38-9400; Zymed Laboratories, South San Francisco, CA); and monoclonal antibodies against FLAG (M5; Sigma-Aldrich, St. Louis, MO), Xpress (Invitrogen), -tubulin (CLT-9002; Cedarlane Laboratories, Hornby, Ontario, Canada), enhanced green fluorescent protein (EGFP; JL-8; BD Biosciences), Aurora-A (clone 4; BD Biosciences), Aurora-B (Kanda et al., 2005), fibrillarin (38F3; Abcam, Cambridge, United Kingdom), NPM1 (B0556; Sigma-Aldrich), and nucleolin (sc-8031; Santa Cruz Biotechnology).
Immunoblot and Immunoprecipitation
Cells were lysed in TBSN buffer (20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1 mM EDTA, 5 mM EGTA, 0.5 mM Na3VO4, 20 mM p-nitrophenyl phosphate, 1 mM phenylmethylsulfonyl fluoride, protease inhibitor cocktail [Sigma-Aldrich], and 0.5% NP-40) on ice for 10 min. Cell lysates were then frozen in liquid nitrogen and thawed in water at 20°C. Thawed lysates were vortexed and centrifuged (15,000 rpm for 5 min), and the cleared extracted lysates were used for immunoblotting. For immunoprecipitation, the lysate (500 µg/tube) was incubated with rotation for 4 h at 4°C with antibody (10 µg)-coupled protein G-Sepharose beads. The beads were washed four times with TBSN buffer. The final volume was adjusted to 25 µl per tube, and the samples were processed for SDS-PAGE. Proteins were transferred to a polyvinylidene difluoride membrane for immunoblotting or were stained with Coomassie brilliant blue. For Edman degradation analysis, relevant areas of the gel were excised.
Indirect Immunofluorescence
Cells were fixed with methanol, blocked with 1% bovine serum albumin (BSA) in phosphate-buffered saline (PBS)(–) and were then incubated for 1 h with rabbit polyclonal antibody against NSUN2 (1:200), and a monoclonal antibody (mAb) against fibrillarin (1:500), NPM1 (1:200), or nucleolin (1:100). Cells were washed three times with PBS(–) containing 1% BSA and were then incubated for 1 h with secondary antibodies and 0.1 µg/ml 4',6-diamidino-2-phenylindole dihydrochloride. Alexa 568-conjugated anti-rabbit or anti-mouse antibodies (1:1000; Invitrogen) and fluorescein isothiocyanate-conjugated anti-rabbit or anti-mouse antibodies (1:150; Chemicon International, Temecula, CA) were used as secondary antibodies. Stained cells were mounted with an anti-fade fluorescent mounting medium (Dako UK, Ely, Cambridgeshire, United Kingdom) and observed under a LSM510META laser scanning microscope (Carl Zeiss, Jena, Germany).
In Vitro Phosphorylation
Purified NSUN2 proteins were used as substrates for in vitro phosphorylation by GST-fused Aurora-B. The reaction was performed for 30 min at 25°C in 100 µl of a reaction mixture containing 25 mM Tris, pH 7.5, 2 mM MgCl2, 100 µM ATP, 100 µM [
-32P]ATP, 0.1 µM calyculin A, 80 µg/ml glutathione S-transferase (GST)-Aurora-B, and 300 µg/ml NSUN2. Histone H3 was used as a positive control substrate. The reaction mixtures were processed for SDS-PAGE and were analyzed by autoradiography with x-ray film (Fuji, Tokyo, Japan).
Methyltransferase Assay
His-NSUN2 was phosphorylated by anti-Aurora-B antibody-immunoprecipitated Aurora-B (from nocodazole-treated HeLa cells) or anti-GST antibody-immunoprecipitated GST-Aurora-B. Both Aurora-B samples contain protein G-Sepharose beads. The Aurora-B kinase activity was monitored by the phosphorylation of H3 histone. Aurora-B-K/R was also used as control experiments. For MOCK kinase reaction, nonimmune IgG-immunoprecipitated sample and anti-GST antibody-immunoprecipitated GST were used. The kinase reacted NSUN2 proteins were used for the methyltransferase assay after removal of the beads containing Aurora-B. NSUN2-S139A and NSUN2-S139E proteins generated by the TNT in vitro transcription/translation (rabbit reticulocyte) system (Promega, Madison, WI) were also used for the methyltransferase assay. [35S]methionine (GE Healthcare, Little Chalfont, Buckinghamshire, United Kingdom) incorporation followed by SDS-PAGE and fluorography confirmed that a protein of the expected size was translated, and the NSUN2 protein amount was estimated by immunoblot.
In methyltransferase assay for DNA, lambda DNA (0.03 pmol) was reacted with the kinase-reacted His-NSUN2 (0.5 µM) in NE buffer 2 reaction buffer (New England Biolabs, Beverly, MA) containing 160 µM S-adenosylmethionine (SAM). Sss1 (New England Biolabs) and Dnmt1 (New England Biolabs) were used as positive and negative control, respectively. After reaction at 37°C for 1 h, the reacted lambda DNA was digested with BstU I (New England Biolabs) and analyzed by agarose gel electrophoresis.
In methyltransferase assay for hemimethylated DNA, hemimethylated mimic substrate poly(dI:dC) (Roche Diagnostics, Mannheim, Germany) (0.2 pmol) or tRNA (purified from E. coli; Sigma-Aldrich) (4 nmol) was reacted with the kinase reacted His-NSUN2 (0.5 µM) or TNT-produced NSUN2 (0.5 µM) in Dnmt1 reaction buffer (New England Biolabs) containing 6.5 µM S-adenosyl-L-[methyl-3H]methionine (specific activity 555 GBq/mmol; GE Healthcare) and RNasein (40 U/µl; Promega). After reaction at 37°C for 30 min (for His-NSUN2) or 5 min (for TNT-produced NSUN2), unincorporated labeled compound was removed by using MicroSpin G25 columns (GE Healthcare), and incorporated radioactivity was measured by liquid scintillation counting.
Ag-Nucleolar Organizer Region (NOR) Protein Silver Staining Method
To visualize the active nucleolus, selective silver staining for the Ag-NOR proteins (Robert-Fortel et al., 1993) was performed. Preparations were washed for 5 min in 50% ethanol and in deionized water. The slides were then covered with a freshly produced staining solution prepared by mixing equal volumes of the following solutions: 1) 0.5 g/ml AgNO3 (Merck, Darmstadt, Germany) in deionized water and 2) and 1 mg/ml gelatin dissolved in 2% (by volume) formic acid. The slides were incubated for 15–20 min at room temperature. After staining, slides were vigorously washed in distilled water, dehydrated in ethanol, and mounted in alcohol-Eukitt.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
H3-P) recognized mitotic phosphorylated Histone H3 (Figure 1A, H3-P) as well as an 
100-kDa protein in mitotic cells (Figure 1A, arrowhead). Both this 100-kDa phosphorylated protein and phosphorylated Histone H3 were repressed by transfection with a kinase-negative form of Aurora-B (Aurora-B-K/R) but not by transfection with the kinase-negative form of Aurora-A (Aurora-A-K/R) (Figure 1A).
|
H3-P (Figure 1D, asterisk). The protein was extracted from the gel band, and partial amino acid sequences were obtained by Edman degradation. We determined the sequences of four peptide fragments (lep56, lep60, lep77-1, and lep77-2; Supplemental Figure S1, red letters). One fragment, lep60, was identical to a sequence in putative protein BC001041
[GenBank]
, but we did not find proteins homologous with the other three peptides. We then performed PCR with primers corresponding to peptides lep56 and lep77-1 (Supplemental Figure S1, underlined blue letters), followed by 5'-RACE. Thus, the PCR product was used to clone the full-length cDNA from a HeLa cDNA library (Supplemental Figure S1). This gene product contains an NOL1/NOP2/sun (NSUN) domain and is currently classified as the second member of the NSUN family (NSUN2).
NSUN2 Is Concentrated in Nucleoli
We raised a rabbit polyclonal antibody (NSUN2) against the C-terminal peptide (GCDPAGVHPPR) of NSUN2. The affinity-purified antibody recognized a protein at the expected size (100 kDa) on immunoblots of lysates from both HeLa and normal human diploid fibroblasts (NHDFs) (Figure 2A). The levels of NSUN2 were markedly different between HeLa and NHDF cells (Figure 2A, arrowhead). Indirect immunofluorescence showed that NSUN2 is located in the nucleoplasm and is concentrated in the nucleoli in both cell lines (Figure 2, B and C). Simultaneous staining for nucleolar protein fibrillarin, NPM1, or nucleolin confirmed this nucleolar localization (Figure 2, B and C). When observed under higher magnification, NSUN2 and these nucleolar proteins were colocalized in the dense fibrillar component and the granular component, and none were significantly associated with the fibrillar centers (Figure 2D). The subnucleolar localization of NSUN2 was absent after RNase treatment (data not shown). During mitosis, NSUN2 is translocated into perichromosomal and cytoplasmic regions, accompanied by nucleolar disassembly, until the next G1 phase when nucleoli reassemble (Figure 2E). All immunoblot and immunofluorescence experiments shown in Figure 2 were confirmed using NSUN2-full (data not shown).
|
|
H3-P or NSUN2, and then performed immunoblotting with the opposite antibody. Immunoprecipitated NSUN2 was recognized by H3-P (Figure 3C, middle). In addition, NSUN2 was one of the proteins immunoprecipitated by H3-P (Figure 3C, right).
Antibody H3-P was raised against a synthetic phosphorylated peptide corresponding to the Ser10 phosphorylation site in Histone H3 (ARKS*TGGKAPRKQL, where S* represents phosphorylated serine). We detected the same RKS* sequence around the 139 site (Supplemental Figure S1). This suggests that NSUN2 is phosphorylated at Ser139. To confirm this observation, we used site-directed mutagenesis to replace Ser139 of NSUN2 with Ala. We then produced full-length wild-type and Ser139Ala mutants of NSUN2 (NSUN2-WT and NSUN2-S139A, respectively) in E. coli. The purified proteins (Supplemental Figure S4) were used in an in vitro kinase assay with either N-terminal GST-fused recombinant Aurora-B-WT or Aurora-B-K/R. We found that NSUN2 was phosphorylated by Aurora-B-WT but not by Aurora-B-K/R (Figure 3D). In addition, NSUN2-S139A was not phosphorylated by Aurora-B-WT (Figure 3D). We also confirmed that the in vitro phosphorylated band for His-tagged NSUN2 was recognized by H3-P (Figure 3E). Thus, it seems that Ser139 on NSUN2 is phosphorylated by Aurora-B.
We next expressed NSUN2 proteins containing N-terminal Xpress tags in HeLa cells and synchronized the cells in mitosis by treatment with nocodazole. Immunoprecipitation with anti-Xpress mAb (Xpress) followed by immunoblotting with H3-P detected NSUN2-WT but not NSUN2-S139A (Supplemental Figure S5). Together with our other results, this suggests that H3-P recognizes NSUN2 phosphorylated during mitosis at Ser139 as well as Histone H3 phosphorylated at Ser10. This phosphorylation was repressed by Aurora-B-K/R (Figure 1A). In addition, a 2-h treatment with Hesperadin or ZM447439 to inhibit Aurora-B repressed mitotic phosphorylation of NSUN2 in the synchronized HeLa cells (Figure 3F).
Because Aurora-B plays a central role in spindle checkpoint regulation, the repression of NSUN2 phosphorylation by Aurora-B-K/R or the inhibitors may be due to an inability to accumulate mitotic cells caused by the absence of spindle checkpoint regulation. To determine whether this is the case, we used a destruction-defective Cyclin B (Cyclin B1), which induces mitotic arrest, even in cells expressing Aurora-B-K/R or an Aurora-B shRNA, or treatment with an Aurora-B inhibitor (data not shown). Cotransfection of HeLa cells with Cyclin B1 and Aurora-B-K/R repressed mitotic phosphorylation of NSUN2 (Figure 3G, lane 8). Similar repression was observed in cells cotransfected with the expression vectors for Cyclin B1 and Aurora-C-K/R (Figure 3G, lane 10). In contrast, in the Aurora shRNA experiments, only the Aurora-B shRNA was able to inhibit NSUN2 phosphorylation in the presence of Cyclin B1 expression (Figure 3H, lanes 6, 8, 9, and 11). Based on these results, we concluded that NSUN2 is phosphorylated by Aurora-B.
Conservation of Aurora-B Phosphorylation Site of NSUN2 in Vertebrates
The human NSUN2 gene belongs to the methyltransferase superfamily and is located on chromosome 5p15.31. Saccharomyces cerevisiae Ncl1/Trm4 is related to human NSUN2 (35% similarity). We also detected a remote similarity between NSUN2 and predicted proteins of Drosophila melanogaster and Caenorhabditis elegans (Figure 4A, Dm and Ce, respectively). Among vertebrates, NSUN2 is highly conserved (Figure 4B). The Aurora-B phosphorylation site at Ser139, which lies within the NOL1/NOP2/sun domain (residues 102-427 in human NSUN2), is also conserved among vertebrates (Figure 4B, red letters). In contrast, nonvertebrates other than Schizosaccharomyces pombe lack the putative Aurora-B phosphorylation site in this domain (Figure 4C).
|
NSUN2 (Figure 5A, IP: NSUN2; IB: NPM1). In addition, in the immunoprecipitation experiments, association between NSUN2 and NPM1 was not observed in nocodazole-treated mitotic cells, whereas the association of NSUN2 with nucleolin was unchanged (Figure 5A, IP: NSUN2, IB: NPM1 and IP: NSUN2, IB: nucleolin). We confirmed these data both by immunoblot using the supernatant of immunoprecipitated samples and by immunoprecipitation using the opposite combination of antibodies (IP: NPM1 or nucleolin; IB: NSUN2) (data not shown). Thus, NSUN2 phosphorylated by Aurora-B during mitosis does not seem to be associated with NPM1.
|
We next examined the effects of the NSUN2 mutants. HeLa cells were transfected with plasmids expressing Xpress-tagged NSUN2-WT, NSUN2-S139A, or NSUN2-S139E. Although Xpress-tagged NSUN2-WT was phosphorylated in mitotic cells (Supplemental Figure S5C), its association with NPM1 was not detected, even in interphase cells (data not shown).
We speculated that the long N-terminal His-Xpress polypeptide disrupts the NSUN2 structure necessary for association with NPM1. We also examined tag-free NSUN2 expressed in NSUN2 knockdown (KD) cells. HeLa cells were transfected with the vector for expressing NSUN2 shRNA, and stable clones with low NSUN2 expression (HeLa-NSUN2-KD cells) were established (Supplemental Figure S3). HeLa-NSUN2-KD cells were transfected with expression plasmids for NSUN2-WT, NSUN2-S139A, or NSUN2-S139E (without tag). When NSUN2-WT was expressed in HeLa-NSUN2-KD cells, the association of NSUN2 with NPM1 and nucleolin was detected in interphase cells (Figure 5C, lane 3). The association of NSUN2 with NPM1 was reduced in mitotic cells, but the association with nucleolin was not (Figure 5C, lane 4). In contrast, the phosphorylation site mutant NSUN2-S139A was associated with NPM1, even during mitosis (Figure 5C, lanes 5 and 6). Finally, the phosphorylation-mimic mutant NSUN2-S139E was associated with nucleolin, but not NPM1 (Figure 5C, lanes 8 and 7, respectively). These data indicate that Aurora-B-mediated phosphorylation of NSUN2 at Ser139 during mitosis directs the dissociation of NPM1 from the NSUN2-nucleolin complex.
Aurora-B-provoked Phosphorylation Affects Methyltransferase Activities of NSUN2
Next, we tested whether phosphorylation status of NSUN2 provoked by Aurora-B affects the methyltransferase activity. NSUN2 contains an NOL1/NOP2/sun domain and relates to yeast protein Ncl1, which is a methyl transferase that generates m5C in RNA. As expected, NSUN2 has no methyltransferase activity in DNA that is not methylated on either strand (Supplemental Figure S6). In contrast, NSUN2 has a methyltransferase activity in hemimethylated DNA. When poly(dI:dC) was used as substrate, NSUN2 stimulated the incorporation of tritirated methyl group from methyl donor SAM into the substrate (Figure 6A, **). In addition, NSUN2 has methyltransferase activity in RNA. When tRNA was used as substrate, the activity was also found (Figure 6B, **). The phosphorylation of NSUN2 provoked by either nocodazole-treated HeLa cells-derived immunoprecipitated Aurora-B or GST-Aurora-B suppressed NSUN2 methyltransferase activities (Figure 6, A and B). Such a suppression was not found when Aurora-B-K/R was used (Figure 6, A and B). The effect of phosphorylation in NSUN2 was also confirmed by using NSUN2-S139A and NSUN2-S139E proteins generated by the TNT in vitro transcription/translation (rabbit reticulocyte) system (Figure 6C). Therefore, as schematically illustrated in Figure 7, we propose a model that the NSUN2 is negatively regulated by Aurora-B-provoked phosphorylation during mitosis.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
NSUN2 was recently identified in a large-scale screening of nuclear phosphoproteins in HeLa cells by mass spectroscopic analysis of tryptic peptides (Beausoleil et al., 2004). This study identified three phosphorylation sites on NSUN2 (Ser456, Ser593, and Ser743). It is possible that Ser139 was not identified as one of the phosphorylation sites due to technical limitations of the large-scale screening by using trypsin. Furthermore, because the cells were not synchronized, mitosis-specific phosphorylation would have been diluted. Whether the three identified sites are constitutively or periodically phosphorylated during the cell cycle was not determined. Together, the current results and those of Beausoleil et al. (2004) indicate that phosphorylation of NSUN2 on Ser139 is mediated by Aurora-B and that this is one of many sites phosphorylated in vivo.
Like Aurora-B, Aurora-C is a mitotic passenger protein that is active during mitosis (Li et al., 2004; Sasai et al., 2004; Dutertre et al., 2005). We also found that kinase-negative Aurora-C-K/R represses mitotic phosphorylation of NSUN2 due to overlapping repression of Aurora-B; however, we found that the level of Aurora-C was very low in HeLa cells and that repression of expression by an shRNA does not affect the phosphorylation of NSUN2 at Ser139 in mitotic cells. Therefore, Aurora-C may not make a substantial contribution to the phosphorylation of NSUN2, and Aurora-B may be the protein kinase critical for its phosphorylation at Ser139 during mitosis.
NSUN2 Is a Nucleolar Protein That Has RNA Methyltransferase Activity
NSUN2 is a novel mammalian RNA methyltransferase that was independently discovered in a study of myc-induced proliferation mediators in epithelial cells (Frye and Watt, 2006). The original working name of the protein in our laboratory was substrate of AIM-1/Aurora-B kinase (SAKI; accession no. AB255451), whereas that reported by Frye and Watt (2006) was myc-induced SUN-domain-containing protein (MISU; accession no. DQ490066). The protein is now designated "NSUN2." NSUN2 contains an NOL1/NOP2/sun domain. This domain is found in archaeal, bacterial, and eukaryotic proteins. In the archaea and bacteria, the majority of proteins possessing this domain have an S-adenosyl methionine binding-domain and are related to E. coli Sun/Fmu protein, which catalyzes 5-methylation of C967 within 16S rRNA (Foster et al., 2003). In S. cerevisiae, the nucleolar protein Nop2 is known to be a probable RNA m5C methyltransferase (Katz et al., 2003) and is essential for processing and maturation of rRNA (de Beus et al., 1994; Hong et al., 1997). Yeast Nop2 is thought to be closely related to human NOL1/proliferation-associated nucleolar antigen p120 (Fonagy et al., 1989; de Beus et al., 1994). Yeast contain Ncl1, another protein related to Nop2 (Wu et al., 1998), but unlike Nop2, Ncl1 is not essential for cell survival. Yeast Ncl1, also called as Trm4, has tRNA m5C methyltransferase activity, and it probably catalyzes the methylation of other RNA molecules (Motorin and Grosjean, 1999). Like the relationship between yeast Nop2 and human NOL1, there may be substantial identity between yeast Ncl1 and human NSUN2.
Similar to the case of yeast Ncl1, repression of NSUN2 expression in HeLa cells by shRNA did not affect cell growth properties such as doubling time and the ability of anchorage-independent growth (data not shown). Thus, like Ncl1, NSUN2 is probably nonessential for growth. Interestingly, disruption of Ncl1 in yeast increases the sensitivity to rRNA-targeting antibiotics, such as paromomycin (Wu et al., 1998). In addition, under disruption of the tRNA methyltransferase Trm8, disruption of Ncl1 greatly reduces tRNA stability (Alexandrov et al., 2006). The evolutionary conservation of NSUN2 suggests that it plays important biological roles other than the nonessential methylation of RNAs. This is also supported by the fact that NSUN2 expression is markedly elevated in cancer cells (Figure 2A; our unpublished data).
Here, we showed that NSUN2 methylated hemimethylated DNA, poly(dI:dC), as well as tRNA. This raises the possibility that NSUN2 is implicated in alterations of not only RNA methylation patterns but also genomic methylation patterns. Unlike prokaryotic restriction DNA methyltransferases that have innate sequence specificity, eukaryotic DNA cytosine methyltransferases exhibit a preference for hemimethylated substrates with indirect sequence recognition. Mammalian DNA methyltransferase-2 (Dnmt2) methylates tRNA, suggesting eukaryotic DNA methyltransferases were derived from ancestral RNA methyltransferases rather than prokaryotic restriction DNA methyltransferases (Goll et al., 2006). Thus, eukaryotic RNA or DNA methyltransferases possibly have broad substrate specificity in nucleic acids, even though their sequence and their organization of catalytic motifs are characteristic of RNA or DNA.
Aurora-B Regulates NSUN2-NPM1 Association and NSUN2 Methyltransferase Activities
During interphase, EGFP-Aurora-B is placed in nucleoplasm but is not located in nucleolus (our unpublished data). During prophase, EGFP-Aurora-B rapidly spreads throughout the chromatin (Murata-Hori et al., 2002). Consequently, Histone H3 is phosphorylated on Ser10 (Crosio et al., 2002). Simultaneously, NSUN2 is rapidly phosphorylated by Aurora-B. NSUN2 phosphorylation probably occurs in parallel with nucleolar disassembly during mitosis. CDK1-Cyclin B is a major regulator of this disassembly but is not the sole driving factor (Sirri et al., 2002). When ribosome disassembles, nucleolar proteins may be phosphorylated by multiple kinases at multiple sites. NPM1 contains four or five CDK1-Cyclin B phosphorylation sites (Thr4, Thr199, Thr219, Thr234, and Thr237) (Peter et al., 1990; Okuwaki et al., 2002), and the interaction of NPM1 with nucleolin is inhibited during mitosis (Liu and Yung, 1999). Here, we have found that NSUN2 is associated with nucleolar proteins NPM1 and nucleolin and that the interaction of NSUN2 with NPM1 is inhibited during mitosis. This dissociation is driven by Aurora- B-provoked phosphorylation of NSUN2. Thus, Aurora-B is also a mitotic kinase that participates to nucleolar disassembly-reassembly processes.
The C-terminal region of NPM1.1, which has a 35-amino acid C-terminal extension, was shown to bind RNA during interphase and to lose this ability upon phosphorylation during mitosis (Okuwaki et al., 2002). Our results show that NSUN2 binds only to NPM1.1, indicating that the interaction must be mediated by the C-terminal region of NPM1.1. We found that NPM1.1 can associate with NSUN2 but that these proteins dissociate during mitosis when NSUN2 is phosphorylated on Ser139 by Aurora-B. This dissociation may also occur in concert with CDK1 phosphorylation of NPM1.1. However, this dissociation is primarily regulated by Aurora-B-provoked phosphorylation of NSUN2, because the Aurora-B kinase inhibitor Hesperadin is able to inhibit this dissociation during mitosis.
Inhibition of CDK1 leads to resumption of rDNA transcription, but it is not sufficient to build a functioning rRNA processing complex in mitotic cells (Sirri et al., 2002). Inhibition of Aurora-B is also not sufficient to activate constitutively rRNA processing in mitosis. Silver staining to visualize the nucleolus confirms that Hesperadin is not able to completely inhibit nucleolar disassembly. The Ag-NOR is not sufficiently spread in Hesperadin-treated mitotic chromosomes, but the number and the size of NOR are not drastically altered. It is possible that this is due to impaired chromosome organization induced by Aurora-B inhibition. Thus, in Aurora-B–inhibited cells, Ag-NOR nucleolar disassembly–reassembly processes are likely to be intact.
The methyltransferase activities of NSUN2 are suppressed by Aurora-B–provoked phosphorylation. During normal mitosis, NSUN2 should be inactive due to its phosphorylation by Aurora-B. It remains uncertain what are precise target nucleic acids of NSUN2 in the cells, although NSUN2 may have broad in vivo substrate specificities including DNA and RNAs such as rRNA, tRNA or microRNAs. The elucidation of the in vivo targets of NSUN2 helps us to understand the physiological effects of the phosphorylation of NSUN2 provoked by Aurora-B.
In summary, Aurora-B regulates not only chromosome segregation but also nucleolar protein function during mitosis. Our findings presented here possibly create a new aspect to investigate how nucleic acid modifications are regulated in mitotic cells.
|
ACKNOWLEDGMENTS |
|---|
|
Footnotes |
|---|
The online version of this article contains supplemental material at MBC Online (http://www.molbiolcell.org). 
Address correspondence to: Masaaki Tatsuka (haruo{at}hiroshima-u.ac.jp)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Alexandrov, A., Chernyakov, I., Gu, W., Hiley, S. L., Hughes, T. R., Grayhack, E. J., Phizicky, E. M. (2006). Rapid tRNA decay can result from lack of nonessential modifications. Mol. Cell 21, 87–96.[CrossRef][Medline]
Andrews, P. D., Knatko, E., Moore, W. J., Swedlow, J. R. (2003). Mitotic mechanics: the auroras come into view. Curr. Opin. Cell Biol 15, 672–683.[CrossRef][Medline]
Beausoleil, S. A., Jedrychowski, M., Schwartz, D., Elias, J.E., Villen, J., Li, J., Cohn, M. A., Cantley, L. C., Gygi, S. P. (2004). Large-scale characterization of HeLa cell nuclear phosphoproteins. Proc. Natl. Acad. Sci. USA 101, 12130–12135.
Carmena, M. and Earnshaw, W. C. (2003). The cellular geography of aurora kinases. Nat. Rev. Mol. Cell Biol 4, 842–854.[CrossRef][Medline]
Crosio, C., Fimia, G. M., Loury, R., Kimura, M., Okano, Y., Zhou, H., Sen, S., Allis, C. D., Sassone-Corsi, P. (2002). Mitotic phosphorylation of histone H3, spatio-temporal regulation by mammalian Aurora kinases. Mol. Cell. Biol 22, 874–885.
de Beus, E., Brockenbrough, J. S., Hong, B., Aris, J. P. (1994). Yeast NOP2 encodes an essential nucleolar protein with homology to a human proliferation marker. J. Cell Biol 127, 1799–1813.
Dimario, P. J. (2004). Cell and molecular biology of nucleolar assembly and disassembly. Int. Rev. Cytol 239, 99–178.[Medline]
Dousset, T., Wang, C., Verheggen, C., Chen, D., Hernandez-Verdun, D., Huang, S. (2000). Initiation of nucleolar assembly is independent of RNA polymerase I transcription. Mol. Biol. Cell 11, 2705–2717.
Dutertre, S., Hamard-Peron, E., Cremet, J. Y., Thomas, Y., Prigent, C. (2005). The absence of p53 aggravates polyploidy and centrosome number abnormality induced by Aurora-C overexpression. Cell Cycle 4, 1783–1787.[Medline]
Fonagy, A., Henning, D., Jhiang, S., Haidar, M., Busch, R. K., Larson, R., Valdez, B., Busch, H. (1989). Cloning of the cDNA and sequence of the human proliferating-cell nucleolar protein P120. Cancer Commun 1, 243–251.[Medline]
Foster, P. G., Nunes, C. R., Greene, P., Moustakas, D., Stroud, R. M. (2003). The first structure of an RNA m5C methyltransferase, Fmu, provides insight into catalytic mechanism and specific binding of RNA substrate. Structure 11, 1609–1620.[Medline]
Frye, M. and Watt, F. M. (2006). The RNA methyltransferase Misu (NSun2) mediates Myc-induced proliferation and is upregulated in tumors. Curr. Biol 16, 971–981.[CrossRef][Medline]
Gassmann, R., Carvalho, A., Henzing, A. J., Ruchaud, S., Hudson, D. F., Honda, R., Nigg, E. A., Gerloff, D. L., Earnshaw, W. C. (2004). Borealin: a novel chromosomal passenger required for stability of the bipolar mitotic spindle. J. Cell Biol 166, 179–191.
Giet, R., Petretti, C., Prigent, C. (2005). Aurora kinases, aneuploidy and cancer, a coincidence or a real link? Trends Cell Biol 15, 241–250.[CrossRef][Medline]
Goll, M. G., Kirpekar, F., Maggert, K. A., Yoder, J. A., Hsieh, C. L., Zhang, X., Golic, K. G., Jacobsen, S. E., Bestor, T. H. (2006). Methylation of tRNAAsp by the DNA methyltransferase homolog Dnmt2. Science 311, 395–398.
Heix, J., Vente, A., Voit, R., Budde, A., Michaelidis, T. M., Grummt, I. (1998). Mitotic silencing of human rRNA synthesis: inactivation of the promoter selectivity factor SL1 by cdc2/cyclin B-mediated phosphorylation. EMBO J 17, 7373–7381.[CrossRef][Medline]
Hong, B., Brockenbrough, J. S., Wu, P., Aris, J. P. (1997). Nop2p is required for pre-rRNA processing and 60S ribosome subunit synthesis in yeast. Mol. Cell Biol 17, 378–388.[Abstract]
Jimenez-Garcia, L. F., Segura-Valdez, M. L., Ochs, R. L., Rothblum, L. I., Hannan, R., Spector, D. L. (1994). Nucleologenesis: U3 snRNA-containing prenucleolar bodies move to sites of active pre-rRNA transcription after mitosis. Mol. Biol. Cell 5, 955–966.[Abstract]
Jinno, S., Suto, K., Nagata, A., Igarashi, M., Kanaoka, Y., Nojima, H., Okayama, H. (1994). Cdc25A is a novel phosphatase functioning early in the cell cycle. EMBO J 13, 1549–1556.[Medline]
Kanda, A., Kawai, H., Suto, S., Kitajima, S., Sato, S., Takata, T., Tatsuka, M. (2005). Aurora-B/AIM-1 kinase activity is involved in Ras-mediated cell transformation. Oncogene 24, 7266–7272.[CrossRef][Medline]
Katz, J. E., Dlakic, M., Clarke, S. (2003). Automated identification of putative methyltransferases from genomic open reading frames. Mol. Cell Proteomics 2, 525–540.
Li, X., Sakashita, G., Matsuzaki, H., Sugimoto, K., Kimura, K., Hanaoka, F., Taniguchi, H., Furukawa, K., Urano, T. (2004). Direct association with inner centromere protein (INCENP) activates the novel chromosomal passenger protein, Aurora-C. J. Biol. Chem 279, 47201–47211.
Liu, H. T. and Yung, B. Y. (1999). In vivo interaction of nucleophosmin/B23 and protein C23 during cell cycle progression in HeLa cells. Cancer Lett 144, 45–54.[CrossRef][Medline]
Motorin, Y. and Grosjean, H. (1999). Multisite-specific tRNA:m5C-methyltransferase (Trm4) in yeast Saccharomyces cerevisiae: identification of the gene and substrate specificity of the enzyme. RNA 5, 1105–1118.[Abstract]
Murata-Hori, M., Tatsuka, M., Wang, Y. L. (2002). Probing the dynamics and functions of aurora B kinase in living cells during mitosis and cytokinesis. Mol. Biol. Cell 13, 1099–1108.
Okuwaki, M., Tsujimoto, M., Nagata, K. (2002). The RNA binding activity of a ribosome biogenesis factor, nucleophosmin/B23, is modulated by phosphorylation with a cell cycle-dependent kinase and by association with its subtype. Mol. Biol. Cell 13, 2016–2030.
Peter, M., Nakagawa, J., Doree, M., Labbe, J. C., Nigg, E. A. (1990). Identification of major nucleolar proteins as candidate mitotic substrates of cdc2 kinase. Cell 60, 791–801.[CrossRef][Medline]
Pinol-Roma, S. (1999). Association of nonribosomal nucleolar proteins in ribonucleoprotein complexes during interphase and mitosis. Mol. Biol. Cell 10, 77–90.
Robert-Fortel, I., Junera, H. R., Geraud, G., Hernandez-Verdun, D. (1993). Three-dimensional organization of the ribosomal genes and Ag-NOR proteins during interphase and mitosis in PtK1 cells studied by confocal microscopy. Chromosoma 102, 146–157.[CrossRef][Medline]
Sampath, S. C., Ohi, R., Leismann, O., Salic, A., Pozniakovski, A., Funabiki, H. (2004). The chromosomal passenger complex is required for chromatin- induced microtubule stabilization and spindle assembly. Cell 118, 187–202.[CrossRef][Medline]
Sasai, K., et al. (2004). Aurora-C kinase is a novel chromosomal passenger protein that can complement Aurora-B kinase function in mitotic cells. Cell Motil. Cytoskeleton 59, 249–263.[CrossRef][Medline]
Savino, T. M., Bastos, R., Jansen, E., Hernandez-Verdun, D. (1999). The nucleolar antigen Nop52, the human homologue of the yeast ribosomal RNA processing RRP1, is recruited at late stages of nucleologenesis. J. Cell Sci 112, 1889–1900.[Abstract]
Schumacher, J. M., Golden, A., Donovan, P. J. (1998). AIR-2, An Aurora/Ipl1-related protein kinase associated with chromosomes and midbody microtubules is required for polar body extrusion and cytokinesis in Caenorhabditis elegans embryos. J. Cell Biol 143, 1635–1646.
Sirri, V., Hernandez-Verdun, D., Roussel, P. (2002). Cyclin-dependent kinases govern formation and maintenance of the nucleolus. J. Cell Biol 156, 969–981.
Sirri, V., Roussel, P., Hernandez-Verdun, D. (2000). In vivo release of mitotic silencing of ribosomal gene transcription does not give rise to precursor ribosomal RNA processing. J. Cell Biol 148, 259–270.
Stoykova, A. S., Dabeva, M. D., Dimova, R. N., Hadjiolov, A. A. (1985). Ribosome biogenesis and nucleolar ultrastructure in neuronal and oligodendroglial rat brain cells. J. Neurochem 45, 1667–1676.[CrossRef][Medline]
Tatsuka, M., Sato, S., Kitajima, S., Suto, S., Kawai, H., Miyauchi, M., Ogawa, I., Maeda, M., Ota, T., Takata, T. (2005). Overexpression of Aurora-A potentiates HRAS-mediated oncogenic transformation and is implicated in oral carcinogenesis. Oncogene 24, 1122–1127.[CrossRef][Medline]
Terada, Y., Tatsuka, M., Suzuki, F., Yasuda, Y., Fujita, S., Otsu, M. (1998). AIM-1, a mammalian midbody-associated protein required for cytokinesis. EMBO J 17, 667–676.[CrossRef][Medline]
Wheatley, S. P., Carvalho, A., Vagnarelli, P., Earnshaw, W. C. (2001). INCENP is required for proper targeting of Survivin to the centromeres and the anaphase spindle during mitosis. Curr. Biol 11, 886–890.[CrossRef][Medline]
Wu, P., Brockenbrough, J. S., Paddy, M. R., Aris, J. P. (1998). NCL1, a novel gene for a non-essential nuclear protein in Saccharomyces cerevisiae. Gene 220, 109–117.[CrossRef][Medline]
Yanagida, M. (2005). Basic mechanism of eukaryotic chromosome segregation. Philos. Trans. R. Soc. Lond. B. Biol. Sci 360, 609–621.[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
