- Email: [email protected]
SUMO-1 transiently localizes to Cajal bodies in mammalian neurons
Journal of Structural Biology 163 (2008) 137–146
Contents lists available at ScienceDirect
Journal of Structural Biology journal homepage: www.elsevier.com/locate/yjsbi
SUMO-1 transiently localizes to Cajal bodies in mammalian neurons J. Navascues 1,2, R. Bengoechea 2, O. Tapia, I. Casafont, M.T. Berciano, M. Lafarga * Departamento de Anatomia y Biologia Celular, Unidad de Biomedicina (C.S.I.C), Universidad de Cantabria, Santander, y Centro de Investigación Biomedica en Red sobre Enfermedades Neurodegenerativas (CIBERNED), Spain
a r t i c l e
i n f o
Article history: Received 6 February 2008 Received in revised form 21 April 2008 Accepted 25 April 2008 Available online 6 May 2008 Keywords: Cajal bodies SUMO-1 conjugation Coilin SMN Neurons Osmotic stress Methyltransferase inhibitors
a b s t r a c t Cajal bodies (CBs) are nuclear organelles involved in the maturation of small nuclear ribonucleoproteins required for the processing of pre-mRNAs. They concentrate coilin, splicing factors and the survival of motor neuron protein (SMN). By using immunocytochemistry and transfection experiments with GFP– SUMO-1, DsRed1-Ubc9, GFP–coilin and GFP–SMN constructs we demonstrate the presence of SUMO-1 and the SUMO conjugating enzyme (Ubc9) in a subset of CBs in undifferentiated neuron-like UR61 cells. Furthermore, SUMO-1 is transiently localized into neuronal CBs from adult nervous tissue in response to osmotic stress or inhibition of methyltransferase activity. SUMO-1-positive CBs contain coilin, SMN and small nuclear ribonucleoproteins, suggesting that they are functional CBs involved in pre-mRNA processing. Since coilin and SMN have several putative motifs of SUMO-1 modification, we suggest that the sumoylation of coilin and/or SMN might play a role in the molecular reorganization of CBs during the neuronal differentiation or stress–response. Ó 2008 Elsevier Inc. All rights reserved.
1. Introduction Cajal bodies (CBs) are nuclear structures discovered by Ramón y Cajal (1903) in mammalian neurons. They concentrate their signature protein, p80 coilin (Andrade et al., 1991; Raska et al., 1991), small nuclear ribonucleoproteins (snRNPs) and the survival of motor neuron (SMN) protein (Matera and Frey, 1998; Carvalho et al., 1999; Gall, 2000). In addition to these components, CBs share with the nucleolus the proteins fibrillarin and Nopp140 and some small nucleolar ribonucleoproteins (snoRNPs) (Raska et al., 1990; Cioce and Lamond, 2005), and also contain basal transcription factors and several kinases (for a review, see Gall, 2000, 2001). CBs are involved in the biogenesis of snRNP and snoRNP required for the processing of pre-mRNAs and pre-rRNAs, respectively (Gall, 2000; Carmo-Fonseca, 2002; Cioce and Lamond, 2005). CBs are ubiquitous nuclear organelles in most mammalian cells. Mammalian neurons contain prominent CBs and their number correlates with both nucleolar and cell body size and transcriptional activity (Lafarga et al., 1991; Pena et al., 2001; Berciano et al., 2007). An important aspect of the cell biology of CBs is to establish the molecular mechanisms that govern their formation. Self-association of coilin may provide a molecular scaffold for the nucleation
* Corresponding author. Fax: +34 942 201903. E-mail address: [email protected] (M. Lafarga). 1 Present address: Departamento de Nutrición Vegetal, Estación Experimental de Aula Dei, C.S.I.C., Zaragoza, Spain. 2 These authors have contributed equally to the present paper. 1047-8477/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jsb.2008.04.013
of other molecular components (Shpargel et al., 2003). Thus, it has been demonstrated that recruitment of SMN protein and spliceosomal snRNPs to CBs is dependent on the coilin C-terminal arginine/glycine (RG) domain (Hebert et al., 2001), and also that the affinity of the coilin for SMN is regulated by the methylation status of arginines within the coilin RG box (Hebert et al., 2002). Furthermore, the residual CBs observed in neurons and fibroblasts from coilin knockout mice contain fibrillarin, Nopp140 and U3 snoRNA, but fail to recruit snRNPs and SMN complex, indicating that coilin is necessary for the normal recruitment of SMN and snRNPs to CBs (Tücker et al., 2001). The small ubiquitin-like modifier 1 (SUMO-1) serves as a post-translational modification (sumoylation) of its protein substrates (Melchior, 2000). This modification requires the sequential participation of an E1 activating enzyme, the E2 conjugating enzyme Ubc9 and an E3 ligase (for review, see Hay, 2005; Geiss-Friedlander and Melchior, 2007). Conjugation with SUMO-1 is a reversible process, and SUMO proteases deconjugate SUMO-1 from target proteins (Gill, 2004; Hay, 2005). Within the cell nucleus, SUMO-1 is diffusely distributed throughout the nucleoplasm and concentrated in promyelocytic leukemia (PML) bodies (Ishov et al., 1999; Villagra et al., 2004), SUMO-1 nuclear bodies (Navascues et al., 2007) and the nuclear envelope (Gill, 2004; Villagra et al., 2004). The nuclear pattern of SUMO-1 expression is consistent with its role in regulation of gene expression, DNA repair, traffic nucleo-cytoplasmic and subnuclear relocalization of its target proteins (Melchior, 2000; Gill, 2004; Hay, 2005).
138
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
The UR61 cell subline, derived from PC12 cells, is a useful model system for studying cell-signaling pathways involved in neuritogenesis and neuronal differentiation (Guerrero et al., 1988; Navascues et al., 2004, 2007), as well as for analyzing the organization of the cell nucleus in neuron-like culture cells. For example, we have previously shown (Navascues et al., 2004) that the number of CBs and gems, coilin-negative nuclear bodies that concentrate SMN protein (Liu and Dreyfuss, 1996), as well as the composition of CBs, change during the dexamethasone (Dex)-induced differentiation of UR61 cells. Regarding the possible sumoylation of CB-associated proteins, bioinformatic analysis with SUMOplotTM software (http://www.abgent.com/doc/sumoplot) identified several possible sumoylation sites in rat coilin and SMN proteins. Moreover, previous biochemical studies have shown that the SUMO-1 E3 ligase PIASy (Sachdev et al., 2001) directly interacts with coilin (Sun et al., 2005). The aim of this study is to investigate the possible participation of SUMO-1 in the molecular reorganization of CBs associated with cellular differentiation and with the neuronal response to stress. As cellular models we use the neuron-like differentiation of UR61 cells (Navascues et al., 2004, 2007) and two neuronal models of cellular stress induced in living rats by hyperosmolar treatment, in the osmosensitive supraoptic nuclei (SON) neurons (Lafarga et al., 1998), or by treatment with the methyltransferase inhibitor adenosine dialdehyde (Adox) in trigeminal ganglia (TG) neurons. The results show the localization of SUMO-1 in a subset of CBs of undifferentiated UR61 cells and in stressed neurons, particularly during the reformation phase of CBs, but not in normal rat neurons. Overexpression of SUMO-1 or Ubc9 in undifferentiated UR61 cells confirmed the presence of both proteins in a subset of CBs. Collectively, these results suggest that the sumoylation of certain CB-associated proteins may be involved in the molecular reorganization of neuronal CBs associated with differentiation and response to stress.
out at 8:00 AM and free access to food and water. The experimental procedures were performed under the approved national guidelines for animal care and in accordance with guidelines established by the International Association for the Study of Pain (Zimmerman, 1983). To induce osmotic stress, animals were given a single intraperitoneal injection of 1.5 M NaCl (18 ml/kg), as previously described (Lafarga et al., 1998). After 0, 0.5, 2, 6, 12 and 24 h rats were sacrificed and tissue samples of SON were processed for immunofluorescence. To inhibit methyltransferase, animals were given a single intravenous injection of Adox (100 mM/ kg). The animals were sacrificed after 0, 0.5, 4 and 24 h post-injection. As neuronal tissue samples we selected sensory ganglia since these nervous centers are free of the blood brain barrier, which facilitates the access of the Adox to neurons. 2.3. Light microscopy immunocytochemistry and quantification
UR61 cells were cultured in RPMI 1640 medium supplemented with 10% normal calf serum, 100 U/ml gentamycin, as described previously (Greene and Tischler, 1976) and grown on coverslips. To induce a neuron-like differentiation, cultures were exposed to 0.2 lM Dex for 0, 12 and 24 h (Guerrero et al., 1988). Mouse neuroblastoma cells Neuro2A (N2A) were grown in Petri dishes with Dulbecco’s modified Eagle’s medium (DMEM) (Gibco–BRL), supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 U/ml streptomycin. Transfections were performed with the plasmids expressing GFP–SMN (Navascues et al., 2004; Sleeman et al., 2003), GFP–coilin (Shpargel et al., 2003), GFP–SUMO-1, which codifies for active SUMO-1 processed by SUMO proteases to expose a carboxyl terminal diglycine motif (Gostissa et al., 1999; Desterro et al., 2005), and pDsRed1-C1-Ubc9 (kindly provided by Dr. Ronald Hay, University of Dundee, Scotland). Undifferentiated UR61 cells were transfected for 18 h using FuGene 6 transfection reagent (Roche) according to the manufacturer’s instructions. For the inhibition of methylation, undifferentiated UR61 cells were incubated with the vehicle (DMSO) or with the methyltransferase inhibitor adenosine dialdehyde (Adox, Sigma) at a final concentration of 100 lM for 4 h.
For indirect immunofluorescence, UR61 cells were grown on 10 10-mm glass coverslips. The cells were washed twice in phosphate-buffered saline (PBS), fixed with 3.7% formaldehyde (freshly prepared from paraformaldehyde) in PBS for 10 min at room temperature. For the preparation of neuronal tissue samples, rats were perfused through the ascending aorta with 3.7% formaldehyde (freshly prepared from paraformaldehyde) in PBS, pH 7.4, for 15 min at room temperature. SON and TG tissue samples were dissected from 500 lm-thick vibratome slices. The tissue fragments were washed in PBS for 1 h and cut into small fragments and individually transferred to a drop of PBS on a siliconized slide. Then, a coverslip was applied on top of the slide and the tissue was squashed by percussion with a histological needle. The preparation was then frozen in dry ice, and the coverslip removed using a razor blade (Pena et al., 2001). The slides with adhered neurons were sequentially dehydrated in 96% and 70% alcohol at 4 °C for 10 min and rinsed in PBS. For the immunostaining, both UR61 cell samples and neuronal tissue preparations were sequentially incubated with 0.5% Triton X-100 in PBS for 30 min, 0.1 M glycine in PBS for 30 min, and 0.01% Tween 20 in PBS for 5 min. The samples were incubated for 1 h with the primary antibody diluted in PBS, washed in PBS containing 0.05% Tween 20, incubated for 45 min with the appropriate secondary antibodies conjugated with FITC or TexasRed (Jackson ImmunoResearch Laboratories, West Grove, PA) and mounted with VectaShield (Vector Laboratories, Burlingame, CA). Confocal microscopy was performed with a laser scanning microscope (LSM 510; Carl Zeiss, Germany) by using excitation wavelengths of 488 nm (for FITC) and 543 nm (for Texas Red). Each channel was recorded independently, and pseudocolor images were generated and superimposed. TIFF images were transferred to Adobe Photoshop 7.0 software (Adobe Systems Inc.) for presentation. For the quantitative analysis of the proportion of CBs that concentrate SUMO-1, UR61 cells and SON and TG neurons were double immunolabeled with the anti-coilin and anti-SUMO-1 antibodies. Coilin positive and SUMO-1 positive CBs versus coilin positive but SUMO-1-negative CBs were counted by direct examination of the different focal planes throughout cell nuclei, using a 63 oil objective. Three samples of each experimental group were used and at least 100 cells for each cell type were counted. Data were analyzed by the StatView 4.5 software and using the statistical test ANOVA and v2. Statistical significance was established at p < 0.05.
2.2. Animals and treatments
2.4. Antibodies
Male, 3-month-old rats of the Sprague–Dawley strain were kept on a 12 h day, 12 h night light timing regime with lights
The following primary antibodies were used. Rabbit polyclonal anti-coilin antibody serum 204.3 (Bohmann et al., 1995); rabbit
2. Materials and methods 2.1. Cell culture and transfection assays
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
polyclonal and mouse monoclonal anti-SUMO-1 antibodies (Santa Cruz Biotechnology and Zymed, 21C7, respectively), mouse monoclonal anti-U2B” antibody (4G3, Euro Diagnostica, The Netherlands) that recognizes the spliceosomal U2B” snRNP protein (Habets et al., 1989); mouse monoclonal anti-SMN antibody (BD, Transduction Laboratories, San Jose, CA) and mouse monoclonal anti-2,2,7-trimethylguanosine cap (TMG-cap) of spliceosomal snRNPs (Oncogene Research).
139
cine) for 1 h at 37 °C. After washing, the sections were incubated with the goat anti-mouse IgG antibody coupled to 15 nm gold particles (BioCell, UK: diluted 1:50 in PBS containing 1% BSA) Following immunogold labeling, the grids were stained with uranyl acetate and lead citrate and examined with a Phillips EM208 electron microscope operated at 60 kV. As controls, sections were treated as described but with the primary antibody omitted. 3. Results
2.5. Immunoelectron microscopy 3.1. Neuron-like UR61 cells contain a subset of CBs enriched in SUMO-1 For immunoelectron microscopy, UR61 were fixed with 3.7% paraformaldehyde in 0.1 M cacodylate buffer for 1 h at room temperature. The cells were scraped from the dishes, transferred to an Eppendorf tube, and centrifuged for 1 min in a microfuge to pellet the cultures. Pellets were washed with 0.1 M cacodylate buffer, dehydrated in increasing concentrations of methanol at 20 °C, embedded in lowicryl K4M at 20 °C and polymerized with ultraviolet irradiation. Ultrathin sections were sequentially incubated with 0.1 M glycine in PBS (15 min), 5% BSA in PBS (20 min) and the primary mouse monoclonal anti-SUMO-1 antibody (diluted 1:10 in 50 mM Tris–HCl, pH 7.6 containing 1% BSA and 0.1 M gly-
We have previously shown that the neuron-like differentiation of UR61 cells is accompanied by variations in the number and composition of CBs (Navascues et al., 2004). In order to extend these observations, we have investigated whether SUMO-1 may be involved in the molecular reorganization of CBs during the differentiation of UR61 cells. Double immunofluorescence experiments for SUMO-1 and coilin revealed the colocalization of these two proteins in a subset of CBs in both undifferentiated and Dex-induced differentiated UR61 cells (Fig. 1A–F). Given the molecular heterogeneity of CBs in UR61
Fig. 1. Immunofluorescence analysis of undifferentiated (A–C) and Dex-treated (24 h) UR61 cells (D–F) showing the presence of SUMO-1 in CBs. (A–F) Double immunolabeling for coilin and SUMO-1 illustrates the colocalization of both molecular components in CBs. (G–I) UR61 cell containing SUMO-1 in a CB that concentrates spliceosomal snRNPs identified with an antibody that recognizes the TMG-cap of snRNAs. (J–L) Colocalization of SUMO-1 and coilin in two CBs of a neuroblastoma N2A cell. Scale bar = 5 lm.
140
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
cells (Navascues et al., 2004), we next investigated whether SUMO1 positive CBs contain spliceosomal snRNPs identified with the anti-TMG-cap antibody. Co-immunostaining for SUMO-1 and TMG-cap clearly showed the accumulation of snRNP splicing factors in the CBs enriched in SUMO-1 (Fig. 1G–I). CBs containing SUMO-1 were also found in mouse neuroblastoma cells N2A (Fig. 1J–L). The presence of SUMO-1 in CBs from UR61 cells was confirmed with immunoelectron microscopy. Gold particles of SUMO-1 immunoreactivity decorated the typical coiled threads of CBs (Fig. 2A). Our next step was to determine whether the proportion of cells containing at least one CB with SUMO-1, and that of SUMO-1-positive CBs, change during the Dex-induced differentiation of UR61 cells. Both values significantly decreased upon Dex treatment for 24 h (Fig. 2B and C). Furthermore, although the total number of CBs (coilin-positive nuclear bodies) per cell significantly increased during neuron-like differentiation, the number of SUMO-1-positive CBs per cell decreased upon Dex treatment for 24 h (Fig. 2D). These results indicate that the localization of SUMO-1 to CBs is preferentially associated with the undifferentiated state of UR61 cells. To investigate the possible influence of cell proliferation on the localization of SUMO-1 to CBs, undifferentiated UR61 cells were growtharrested by serum starvation for 18 h. No significant changes in the proportion of SUMO-1-positive CBs, and of cells containing these bodies, were detected in serum-starved cells, compared with cells cultured in a medium with serum (data not shown). The accumulation of SUMO-1 in a subset of CBs was confirmed in undifferentiated UR61 transiently expressing GFP–SUMO-1, GFP–coilin or GFP–SMN. Transfection experiments revealed that GFP–SUMO-1 was readily detectable in a subset of CBs immunolabeled with the anti-coilin antibody (Fig. 3A–C). Transient expression of GFP–coilin resulted in a nucleoplasmic diffuse staining pattern in addition to nuclear foci being identified as CBs. Immunostaining experiments confirmed the colocalization of endogenous SUMO-1 and GFP–coilin in certain CBs, whereas other SUMO-1 positive nuclear bodies were free of the fusion protein (Fig. 3D–F). Similar results were observed in UR61 cells transfected with the GFP–SMN construct. The fusion protein appeared distributed diffusely throughout the cytoplasm and in nuclear and cytoplasmic foci, as previously reported (Navascues et al., 2004). The SUMO-1 immunoreactivity signal colocalized with some nuclear GFP–SMN foci (Fig. 3G–I). Having established the presence of SUMO-1 in a subset of CBs, we next studied whether Ubc9, an essential enzyme in the sumoylation cycle (Desterro et al., 1997; Hay, 2005), localized to CBs. Transient expression of DsRed1-Ubc9 in undifferentiated UR61 cells resulted in the formation of nuclear foci of the exogenous Ubc9, some of which containing endogenous coilin were identified as CBs (Fig. 3J–L). Since coilin methylation regulates CB formation (Hebert et al., 2002), we wanted to assay the behavior of SUMO-1 positive CBs under conditions of reduced methyltransferase activity. We therefore investigated the effect of the methyltransferase inhibitor Adox (100 lM) on the localization of SUMO-1 to CBs in undifferentiated UR61 cells. A short treatment with Adox for 4 h, which does not affect cell survival, resulted in a significant increase in the proportion of cells containing SUMO-1-positive CBs (56 ± 0.4% vs 39.1 ± 0.5% in untreated cells, means ± SD) and that of CBs with SUMO-1 (50.1 ± 0.7% vs 37.1 ± 0.3% in untreated cells, means ± SD). Thus, inhibition of methylation increases the association of SUMO-1 with CBs. 3.2. Hyperosmolar stress induced a transient localization of SUMO-1 to CBs in supraoptic neurons We next investigated whether CBs containing SUMO-1 also occur in mammalian neurons from adult nervous tissue. As a
Fig. 2. (A) Immunogold electron microscopy illustrates the presence of gold particles of SUMO-1 immunoreactivity on the coiled threads of a CB. Scale bar = 150 nm. (B and C) Quantitative analysis of the proportion of cells containing at least one CB with SUMO-1 (B), and that of CBs with SUMO-1 (C) during the Dex-induced differentiation of UR61 cells. Bars represent means ± standard deviation. (D) Distribution of coilin-positive and SUMO-1-negative, and coilin-positive and SUMO-1positive CBs during the differentiation of UR61 cells. Bars represent means ± standard deviation.
cellular model system we used rat SON neurons, neurosecretory cells which produce and release the antidiuretic hormone vasopressin involved in the regulation of water and electrolytic balance (Burbach et al., 2001). Previous studies in these neurons have shown that CBs undergo a transcription-dependent disassembly/reassembly cycle in response to the osmotic stress in-
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
141
Fig. 3. (A–C) A UR61 cell expressing GFP–SUMO-1 exhibits several nuclear foci of the fusion protein. One of these foci, identified as a CB, is immunoreactive for coilin. (D–F) Overexpression of GFP–coilin wild-type in an undifferentiated UR61 cell illustrates the colocalization of the fusion protein and the endogenous SUMO-1 in one CB. Another SUMO-1 positive nuclear body is avoid of the fusion protein. (G–I) Overexpression of GFP–SMN wild-type in an undifferentiated UR61 cell showing prominent nuclear and cytoplasmic foci of GFP–SMN, in addition to a diffuse cytoplasmic pool. Note that one of the two nuclear foci appears intensely immunolabeled for SUMO-1. (J–L) Overexpression of DsRed1-Ubc9 in undifferentiated UR61 cells shows the concentration of the fusion protein in a CB immunostained for coilin. Note the presence of a DsRed1-Ubc9-positive and coilin-negative nuclear body. Scale bar = 5 lm.
duced by the intraperitoneal injection of hypertonic saline (HS) solutions (Lafarga et al., 1998). This disassembly/assembly cycle of CBs provides a good model to analyze the possible participation of SUMO-1 in the molecular reorganization of CBs associated with neuronal stress. Double-labeling experiments to detect SUMO-1 and coilin in control SON neurons revealed the typical diffuse nucleoplasmic signal of SUMO-1 and the absence of colocalization of SUMO-1 and coilin in CBs (Fig. 4A–C). As early as 30 min after HS treatment, CBs tended to disappear and coilin relocalized in small perinucleolar caps (Raska et al., 1990; Santama et al., 1996; Lafarga et al., 1998) free of SUMO-1 signal (Fig. 4D–F). After 2 h and 6 h of HS treatment, co-localization of SUMO-1 and coilin was observed in a subset of CBs free in the nucleoplasm or, most frequently, attached to the nucleolus (Fig. 4G–L). By 24 h of induction of the osmotic stress, prominent CBs were observed but most of them lacked SUMO-1 immunoreactivity (Fig. 4M–O). The quantitative analysis of CBs with and without SUMO-1 was performed on squash preparations of SON neurons co-immunostained for coilin and SUMO-1 (Fig. 5A and B). Figs. 5C and D
show that, 2 h after HS treatment, the proportion of cells containing at least one CB with SUMO-1, and that of SUMO-1-positive CBs, increased considerably. The two values increased further at 6 h, progressively decreasing at 12 and 24 h after the HS treatment. Curiously, the mean number of CBs (with and without SUMO-1) per neuron remarkably decreased at 0.5 h after the induction of the osmotic stress. This was followed by a rebound effect showing a progressive increase of CBs, over the control value, at 6, 12 and 24 h after HS treatment (Fig. 5E). Through this disassembly/reassembly cycle, the highest number of SUMO-1 positive CBs per neuron was detected during the period of reformation of CBs, at 6 h after the induction of the osmotic stress (Fig. 5E). 3.3. The reduction of methyltransferase activity induces a transient localization of SUMO-1 to CBs in trigeminal ganglia neurons Having established that the inhibition of methylation with Adox treatment increases the number of SUMO-1 positive CBs in undifferentiated UR61 cells, we next studied whether this
142
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
Fig. 4. Immunofluorescence analysis of distribution of SUMO-1 and coilin in SON neurons from control (A–C) and osmotically stressed (OS) rats at different time points of hypertonic saline injection (D–O). (A–C) In control neurons, SUMO-1 showed a diffuse nucleoplasmic pattern, excluding the nucleolus, but did not concentrate in the CB labeled with the anti-coilin antibody. (D–F) After 30 min of hypertonic saline injection, coilin immunolabeling appeared as perinucleolar caps. (G–L) Following 2 and 6 h of osmotic stress induction, SUMO-1 was diffusely distributed throughout the nucleoplasm, and concentrated in both intranucleolar foci free of coilin (H), and in nucleolus-attached CBs. (M–O) Upon 24 h of hypertonic saline injection, numerous and prominent coilin-positive and SUMO-1-negative CBs were observed. Scale bar = 5 lm.
effect is extensive to neurons from adult nervous tissue. As a cellular model system we used primary sensory neurons from TG, peripheral nervous centers free of the blood brain barrier. As occurs with SON neurons, CBs of control TG neurons lack SUMO-1 (Fig. 6A–C). Upon treatment with Adox for 4 h, a subset of neuronal CBs showed colocalization of SUMO-1 and coilin (Fig. 6D–F). At this time point of Adox treatment, 36 ± 3.5% (means ± SD) of neurons contained at least one CB with SUMO1 and 24.1 ± 2.4% (means ± SD) of CBs were SUMO-1-positive.
However, SUMO-1 was rarely observed in CBs after longer treatment with the methylation inhibitor for 24 h (Fig. 6G–I). Reduction of the methyltransferase activity with Adox also induced changes in the mean number of CBs per neuron, with an initial and significant decrease after 0.5 h of Adox treatment, compared with control neurons, followed by a dramatic increase after 4 h, returning to control values after 24 h of Adox treatment (Fig. 7A–E). The presence of SUMO-1 in CBs correlates with a dramatic increase in the mean number of total CBs per neuron
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
143
Fig. 5. Quantitative analysis of the distribution of SUMO-1-positive CBs in SON neurons at different time points of osmotic stress induction. (A and B) Illustrative example of dissociated SON neurons immunolabeled for coilin (A) and SUMO-1 (B) used for the quantitative analysis. Arrows indicate the colocalization of SUMO-1 and coilin in CBs after 6 h of osmotic stress (OS). Scale bar = 10lm. (C–E) Histograms represent the proportion of neurons containing at least on CB with SUMO-1 (C), and that of SUMO-1-positive CBs (D), as well as the number of CBs per neuron (E, with and without SUMO-1) at different time points of osmotic stress induction. Bars represent means ± standard deviation.
(Fig. 7E). Curiously, although previous studies have shown that Adox treatment affects the interaction between coilin and SMN (Hebert et al., 2002; Boisvert et al., 2002), the formation of gems was not induced in TG neurons. We next investigated whether the CBs with SUMO-1 induced by Adox treatment in TG neurons represent canonical CBs which, in addition to coilin, contain spliceosomal snRNPs and SMN (Gall 2000; Cioce and Lamond, 2005). Double immunolabeling experiments revealed the colocalization of SUMO-1 with SMN or with spliceosomal snRNPs in a subset of CBs (Fig. 6J–O). 4. Discussion Modulation of protein activities by SUMO-dependent modification is known to play a regulatory role in many cellular processes (for review, see Hay, 2005; Geiss-Friedlander and Melchior, 2007). In this study we show that SUMO-1 accumulates in a subset of CBs in undifferentiated UR61 cells and in stressed neurons, and also that GFP-tagged SUMO-1 is concentrated in certain CBs. These findings suggest that certain CB proteins may be modified by SUMO-1. Interestingly, we have also demonstrated that the E2 enzyme Ubc9, the sole conjugating enzyme for sumoylation (Desterro et al., 1997), localizes to a subset
of CBs in UR61 cells expressing DsRed-1-Ubc9. This raises the possibility that certain protein targets may be modified by SUMO-1 in CBs. In this way, bioinformatic analysis with SUMOplotTM software identified four lysines in rat coilin (84, 280, 518 and 522) and two lysines in rat SMN (53 and 117) with a high probability of SUMO-1 modification. Moreover, recent biochemical studies have demonstrated that the SUMO E3 ligase PIASy (Sachdev et al., 2001) interacts directly with coilin in vitro and in vivo (Sun et al., 2005). However, at present, binding studies have been unable to demonstrate the conjugation of coilin with SUMO-1 (Sun et al., 2005). The presence of SUMO-1 in several of the CBs provides additional support in favor of the molecular heterogeneity of these structures (for review, see Cioce and Lamond, 2005). In the case of the UR61 cells, our results indicate that the differentiation of UR61 cells is accompanied by a reduction of CBs containing SUMO-1. Thus, the higher proportion of SUMO-1-positive CBs reported here correlates with an undifferentiated state of the UR61 cells. Furthermore, it is independent of cell proliferation. Our previous studies have shown a molecular reorganization of CBs in differentiating UR61 cells. Thus, SMN is globally upregulated and progressively recruited to CBs as UR61 cells shift to a differentiate state (Navascues et al., 2004). Although the functional sig-
144
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
Fig. 6. Double immunolabeling experiments on dissociated trigeminal ganglia neurons from control (A–C) and experimental rats treated with Adox for 4 h (D–F and J–O). Costaining for SUMO-1 and coilin revealed the absence the colocalization of these two molecules in the CBs of both control (A–C) and Adox-treated neurons for 24 h (G–I). Adox treatment for 4 h induced a proliferation of CBs some of them concentrated SUMO-1 (D–F). (J–O) Double immunofluorescence analysis for SUMO-1 and SMN (J–L) or SUMO-1 and spliceosomal snRNPs, identified with the anti-U2B” antibody (M–O), showed the concentration of SMN and pre-mRNA splicing factors in SUMO-1-positive CBs after 4 h of Adox treatment. Note the immunolabeling of the nuclear envelope for SMN (J) and SUMO-1 (N), as well as the intranucleolar localization of SUMO-1 (K and N). Scale bar = 5 lm.
nificance of the localization of SUMO-1 to CBs is unknown, we suggest that the reversible sumoylation of certain CB proteins, particularly coilin and/or SMN, may be involved in the molecular reorganization of CBs during the neuron-like differentiation of UR61 cells. We have previously demonstrated that SON neurons undergo a profound reorganization of CBs in response to the osmotic stress (Lafarga et al., 1998). This response is characterized by an initial phase (0.5 h after HS treatment) of CB disassembly, relocalization of coilin as perinucleolar caps and decrease in transcrip-
tional activity, followed by a period of CB reformation (Lafarga et al., 1998; present results). We now demonstrate that SUMO1 transiently localizes to CBs during the reformation process. Interestingly, SUMO-1 colocalizes with coilin in some CBs attached to the nucleolus, whereas the adjacent smaller perinucleolar caps of coilin lack SUMO-1. Our previous studies have shown that the perinucleolar caps are progressively replaced by typical CBs during the osmotic stress–response, suggesting that the caps represent CB precursors (Lafarga et al., 1998). Indeed, perinucleolar caps of coilin are observed in mouse embryos before the
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
145
Fig. 7. Treatment with AdOx induces changes in CB number in trigeminal ganglia neurons. (A–D) Immunofluorescence patterns of coilin in control and Adox-tretaed neurons. Typical CBs were observed in neurons of the control (A) and after 24 h of Adox treatment (D). At 0.5 h of Adox treatment coilin relocalized around of the nucleolus (B), whereas a proliferation of CBs was detected at 4 h of treatment (C). (E) Histogram of the distribution of CBs. Note the presence of CBs containing SUMO-1 at 4 h of Adox treatment. Bars represent means ± standard deviation.
appearance of the first CBs (Ferreira and Carmo-Fonseca, 1995). In this context, we suggest that the sumoylation of certain CB proteins, particularly coilin and/or SMN, may be involved in the disassembly/reformation cycle of neuronal CBs induced by osmotic stress. It is also interesting to note that the neuronal stress induced by inhibition of the protein methyltransferase activity in TG neurons is also associated with a transient appearance of SUMO-1-positive CBs. As occurs in SON neurons, the presence of SUMO-1 in a subset of CBs correlates with a stage of reformation of numerous CBs after an earlier stage of CB disassembly and perinucleolar relocalization of the coilin. Although Adox treatment could affect the methylation of several nuclear proteins (Boisvert et al., 2002), previous studies in HeLa cells have shown that this methyltransferase inhibitor reduces the levels of coilin methylation and the interaction of coilin and SMN (Hebert et al., 2002; Boisvert et al., 2002). Thus, coilin methylation seems to be important for the localization of coilin and SMN in CBs (Hebert et al., 2002; Boisvert et al. 2002). Curiously, in TG neurons from adult tissues, as occurs in HeLa and UR61 cells (Hebert et al., 2002; Boisvert et al., 2002; Navascues et al., 2004), Adox treatment does not induce the redistribution of SMN into gems. It does, however, cause the appearance of SUMO-1-positive CBs in TG neurons and also increases the number of these nuclear bodies in undifferentiated UR61 cells. Moreover, SUMO-1-positive CBs contain SMN and snRNPs suggesting that they are functional CBs involved in the biogenesis of spliceosomal snRNPs (Gall 2000; Cioce and Lamond, 2005). Coilin and SMN are putative substrates of SUMO-1 and their reversible sumoylation might play a role in the molecular reassembly of CBs during the neuronal stress–response. Further transfection experiments with constructs expressing mutated coilin or SMN in putative sumoylation lysines will be necessary to determine whether CBs are sumoylation sites for these CB proteins.
Acknowledgments The authors wish to thank R. García-Ceballos for technical assistance. We are also grateful to Dr. A.I. Lamond (University of Dundee, Scotland) for generously providing anti-coilin antibody (204.10) and the plasmid expressing GFP–SMN, Dr. G. Matera (Case Western University, Cleveland, USA) for the plasmid expressing GFP–coilin and Dr. Joana Desterro (IMM, Lisbon, Portugal) for the plasmid expressing GFP-SUMO-1. This work was supported by the following grants: ‘‘Dirección General de Investigacion” of Spain (BFU2005-01030), and Centro de Investigación Biomedica en Red sobre Enfermedades Neurodegenerativas (CIBERNED; CB06/05/ 0037) from Spain. References Andrade, L.E., Chan, E.K., Raska, I., Peebles, C.L., Roos, G., Tan, E.M., 1991. Human antibodies to a novel protein of the nuclear coiled body: immunological characterization and cDNA cloning of p-80-coilin. J. Exp. Med. 173, 1407– 1419. Berciano, M.T., Novell, M., Villagra, N.T., Casafont, I., Bengoechea, R., Val-Bernal, F., Lafarga, M., 2007. Cajal body number and nucleolar size correlate with cell body mass in human sensory ganglia neurons. J. Struct. Biol. 158, 410–420. Bohmann, K., Ferreira, J., Lamond, A.I., 1995. Mutational analysis of p80 coilin indicates a functional interaction between coiled bodies and nucleolus. J. Cell Biol. 68, 817–831. Boisvert, F.M., Cote, J., Boulanger, M.C., Cleroux, P., Bachand, F., Autexier, C., Richard, S., 2002. Symmetrical dimethylarginine methylation is required for the localization of SMN in Cajal bodies and premRNA splicing. J. Cell Biol. 159, 957–969. Burbach, J.P., Luckman, S.M., Murphy, D., Gainer, H., 2001. Gene regulation in the magnocellular hypothalamo-neurohypophysial system. Physiol. Rev. 81, 1197– 1267. Cajal, S.R., 1903. Un sencillo método de coloración selectiva del retículo protoplasmático y sus efectos en los diversos órganos nerviosos de vertebrados e invertebrados. Trab. Lab. Invest. Biol. 2, 129–221. Carmo-Fonseca, M., 2002. New clues to the function of the Cajal body. EMBO Rep. 3, 726–727.
146
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
Carvalho, T., Almeida, F., Calapez, A., Lafarga, M., Berciano, M.T., Carmo-Fonseca, M., 1999. The spinal muscular atrophy disease gene product, SMN: a link between snRNP biogenesis and the Cajal (coiled) body. J. Cell Biol. 147, 715– 728. Cioce, M., Lamond, A.I., 2005. Cajal bodies: a long history of discovery. Annu. Rev. Cell. Dev. Biol. 21, 105–131. Desterro, J.M., Thompson, J., Hay, R.T., 1997. Ubch9 conjugates SUMO but not ubiquitin. FEBS Lett. 417, 297–300. Desterro, J.M., Keegan, L.P., Jaffray, L., Hay, R.T., O’Connell, M.A., Carmo-Fonseca, M., 2005. SUMO-1 modification alters ADAR1 editing activity. Mol. Biol. Cell 16, 5115–5126. Ferreira, J., Carmo-Fonseca, M., 1995. The biogenesis of the coiled body during early mouse development. Development 121, 601–612. Gall, J.G., 2000. Cajal bodies: the first 100 years. Annu. Rev. Cell. Dev. Biol. 16, 273– 300. Gall, J.G., 2001. A role for Cajal bodies in assembly of the nuclear transcription machinery. FEBS Lett. 498, 164–167. Geiss-Friedlander, R., Melchior, F., 2007. Concepts in sumoylation: a decade on. Nat. Rev. Mol. Cell Biol. 8, 947–956. Gill, G., 2004. SUMO and ubiquitin in the nucleus: different functions, similar mechanisms? Genes Dev. 18, 2046–2059. Gostissa, M., Hengstermann, A., Fogal, V., Sandy, P., Scwarz, S.E., Scheffner, M., Del Sal, G., 1999. Activation of p53 by conjugation to the ubiquitin-like protein SUMO-1. EMBO J. 21, 1456–1464. Greene, L.A., Tischler, A.S., 1976. Establishment of a noradrenergic clonal line of rat adrenal pheochomocytoma cells which respond to nerve growth factor. Proc. Natl. Acad. Sci. USA 73, 2424–2428. Guerrero, I., Pellicer, A., Burstein, D.E., 1988. Dissociation of c-fos from ODC expresión and neuronal differentiation in a PC12 subline stably transfected with an inducible N-Ras oncogene. Biochem. Biophys. Res. 150, 1185–1192. Habets, W.J., Hoet, M.H., De Jong, B.A., Van der Kemp, A., Van Venrooij, W.J., 1989. Mapping of B cell epitopes on small nuclear ribonucleoproteins that react with human antibodies as well as with experimentally-induced mouse monoclonal antibodies. J. Immunol. 143, 2560–2566. Hay, R.T., 2005. SUMO: a history of modification. Mol. Cell 18, 1–12. Hebert, M.D., Szymczyk, P.W., Shpargel, K.B., Matera, A.G., 2001. Coilin forms the bridge between Cajal bodies and SMN, the spinal muscular atrophy protein. Genes Dev. 15, 2720–2729. Hebert, M.D., Shpargel, K.B., Ospina, J.K., Tucker, K.E., Matera, A.G., 2002. Coilin methylation regulates nuclear body formation. Dev. Cell 3, 329– 337. Ishov, A.M., Sotnikov, A.G., Negorev, D., Vladimirova, O.V., Neff, N., Kamitani, T., Yeh, E.T.H., Strauss III, J.F., Maul, G.G., 1999. PML is critical for ND10 42 formation and recruits the PML-interacting protein Daxx to this nuclear structure when modified by SUMO-1. J. Cell Biol. 147, 212–233. Lafarga, M., Andres, M.A., Berciano, M.T., Maquiera, E., 1991. Organization of nucleoli and nuclear bodies in osmotically stimulated supraoptic neurons of the rat. J. Comp. Neurol. 308, 329–339.
Lafarga, M., Berciano, M.T., Garcia-Segura, L.M., Andres, M.A., Carmo-Fonseca, M., 1998. Acute osmotic/stress stimuli induce a transient decrease of transcriptional activity in the neurosecretory of supraoptic nuclei. J. Neurocytol. 27, 205–217. Liu, Q., Dreyfuss, G., 1996. A novel nuclear structure containing the survival of motor neurons protein. EMBO J. 15, 3555–3565. Matera, A.G., Frey, M.R., 1998. Nuclear structure ’98. Coiled bodies and gems: janus or gemini? Am. J. Hum. Genet. 63, 317–321. Melchior, F., 2000. SUMO-nonclassical ubiquitin. Annu. Rev. Cell. Dev. Biol. 16, 591– 626. Navascues, J., Berciano, M.T., Tücker, K.E., Lafarga, M., Matera, G., 2004. Targeting SMN to Cajal bodies and nuclear gems during neuritogenesis. Chromosoma 112, 398–409. Navascues, J., Bengoechea, R., Tapia, O., Vaque, J.P., Lafarga, M., Berciano, M.T., 2007. Characterization of a new SUMO-1 nuclear body (SNB) enriched in pCREB, CBP and c-Jun in neuron-like UR61 cells. Chromosoma 116, 441–451. Pena, E., Berciano, M.T., Fernandez, R., Ojeda, J.L., Lafarga, M., 2001. Neuronal body size correlates with the number of nucleoli and Cajal bodies, and with the organization of the splicing machinery in rat trigeminal ganglion neurons. J. Comp. Neurol. 430, 250–263. Raska, I., Ochs, R.L., Andrade, L.E.C., Chan, E.K.L., Burlingame, R., Peebles, C., Gruol, D., Tan, E.M., 1990. Association between nucleolus and the coiled body. J. Struct. Biol. 104, 120–127. Raska, I., Andrade, L.E.C., Ochs, R.L., Chan, E.K.L., Chang, C.-M., Roos, G., Tan, E.M., 1991. Immunological and structural studies of the nuclear coiled body with autoimmune antibodies. Exp. Cell Res. 195, 27–37. Sachdev, S., Bruhn, L., Sieber, H., Pichler, A., Melchior, F., Grosschedl, R., 2001. PIASy, a nuclear matrix-associated SUMO E3 ligase, represses LEF1 activity by sequestration into nuclear bodies. Genes Dev. 15, 3088–3103. Santama, N., Dotti, C.G., Lamond, A.I., 1996. Neuronal differentiation in the rat hippocampus involves a stage-specific reorganization of subnuclear structure both in vivo and in vitro. Eur. J. Neurosci. 8, 892–905. Shpargel, K.B., Ospina, J.K., Tucker, K.E., Matera, A.G., Hebert, M.D., 2003. Control of Cajal body number is mediated by the coilin C-terminus. J. Cell Sci. 14, 9–16. Sleeman, J.E., Trinkle-Mulcahy, L., Prescott, A.R., Ogg, S.C., Lamond, A.I., 2003. Cajal body proteins SMN and coilin show differential dynamic behaviour in vivo. J. Cell Sci. 116, 2039–2050. Sun, J., Hongzhi, X., Subramony, S.H., Hebert, M.D., 2005. Interactions between Coilin and PIASy partially link Cajal bodies to PML bodies. J. Cell Sci. 118, 4995– 5003. Tücker, K.E., Berciano, M.T., Jacobs, E.Y., LePage, E.F., Shpargel, K.B., Rossire, J.J., Chan, E.K., Lafarga, M., Conlon, R.A., Matera, A.G., 2001. Residual Cajal bodies in coilin knockout mice fail to recruit Sm snRNPs and SMN, the spinal muscular atrophy gene product. J. Cell Biol. 154, 293–307. Villagra, N.T., Berciano, J., Altable, M., Navascues, J., Casafont, I., Lafarga, M., Berciano, M.T., 2004. PML bodies in reactive sensory ganglion neurons of the Guillain-Barré syndrome. Neurobiol. Dis. 16, 158–168. Zimmerman, M., 1983. Ethical guidelines for investigation of experimental pain in conscious animals. Pain 16, 109–110.
Contents lists available at ScienceDirect
Journal of Structural Biology journal homepage: www.elsevier.com/locate/yjsbi
SUMO-1 transiently localizes to Cajal bodies in mammalian neurons J. Navascues 1,2, R. Bengoechea 2, O. Tapia, I. Casafont, M.T. Berciano, M. Lafarga * Departamento de Anatomia y Biologia Celular, Unidad de Biomedicina (C.S.I.C), Universidad de Cantabria, Santander, y Centro de Investigación Biomedica en Red sobre Enfermedades Neurodegenerativas (CIBERNED), Spain
a r t i c l e
i n f o
Article history: Received 6 February 2008 Received in revised form 21 April 2008 Accepted 25 April 2008 Available online 6 May 2008 Keywords: Cajal bodies SUMO-1 conjugation Coilin SMN Neurons Osmotic stress Methyltransferase inhibitors
a b s t r a c t Cajal bodies (CBs) are nuclear organelles involved in the maturation of small nuclear ribonucleoproteins required for the processing of pre-mRNAs. They concentrate coilin, splicing factors and the survival of motor neuron protein (SMN). By using immunocytochemistry and transfection experiments with GFP– SUMO-1, DsRed1-Ubc9, GFP–coilin and GFP–SMN constructs we demonstrate the presence of SUMO-1 and the SUMO conjugating enzyme (Ubc9) in a subset of CBs in undifferentiated neuron-like UR61 cells. Furthermore, SUMO-1 is transiently localized into neuronal CBs from adult nervous tissue in response to osmotic stress or inhibition of methyltransferase activity. SUMO-1-positive CBs contain coilin, SMN and small nuclear ribonucleoproteins, suggesting that they are functional CBs involved in pre-mRNA processing. Since coilin and SMN have several putative motifs of SUMO-1 modification, we suggest that the sumoylation of coilin and/or SMN might play a role in the molecular reorganization of CBs during the neuronal differentiation or stress–response. Ó 2008 Elsevier Inc. All rights reserved.
1. Introduction Cajal bodies (CBs) are nuclear structures discovered by Ramón y Cajal (1903) in mammalian neurons. They concentrate their signature protein, p80 coilin (Andrade et al., 1991; Raska et al., 1991), small nuclear ribonucleoproteins (snRNPs) and the survival of motor neuron (SMN) protein (Matera and Frey, 1998; Carvalho et al., 1999; Gall, 2000). In addition to these components, CBs share with the nucleolus the proteins fibrillarin and Nopp140 and some small nucleolar ribonucleoproteins (snoRNPs) (Raska et al., 1990; Cioce and Lamond, 2005), and also contain basal transcription factors and several kinases (for a review, see Gall, 2000, 2001). CBs are involved in the biogenesis of snRNP and snoRNP required for the processing of pre-mRNAs and pre-rRNAs, respectively (Gall, 2000; Carmo-Fonseca, 2002; Cioce and Lamond, 2005). CBs are ubiquitous nuclear organelles in most mammalian cells. Mammalian neurons contain prominent CBs and their number correlates with both nucleolar and cell body size and transcriptional activity (Lafarga et al., 1991; Pena et al., 2001; Berciano et al., 2007). An important aspect of the cell biology of CBs is to establish the molecular mechanisms that govern their formation. Self-association of coilin may provide a molecular scaffold for the nucleation
* Corresponding author. Fax: +34 942 201903. E-mail address: [email protected] (M. Lafarga). 1 Present address: Departamento de Nutrición Vegetal, Estación Experimental de Aula Dei, C.S.I.C., Zaragoza, Spain. 2 These authors have contributed equally to the present paper. 1047-8477/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jsb.2008.04.013
of other molecular components (Shpargel et al., 2003). Thus, it has been demonstrated that recruitment of SMN protein and spliceosomal snRNPs to CBs is dependent on the coilin C-terminal arginine/glycine (RG) domain (Hebert et al., 2001), and also that the affinity of the coilin for SMN is regulated by the methylation status of arginines within the coilin RG box (Hebert et al., 2002). Furthermore, the residual CBs observed in neurons and fibroblasts from coilin knockout mice contain fibrillarin, Nopp140 and U3 snoRNA, but fail to recruit snRNPs and SMN complex, indicating that coilin is necessary for the normal recruitment of SMN and snRNPs to CBs (Tücker et al., 2001). The small ubiquitin-like modifier 1 (SUMO-1) serves as a post-translational modification (sumoylation) of its protein substrates (Melchior, 2000). This modification requires the sequential participation of an E1 activating enzyme, the E2 conjugating enzyme Ubc9 and an E3 ligase (for review, see Hay, 2005; Geiss-Friedlander and Melchior, 2007). Conjugation with SUMO-1 is a reversible process, and SUMO proteases deconjugate SUMO-1 from target proteins (Gill, 2004; Hay, 2005). Within the cell nucleus, SUMO-1 is diffusely distributed throughout the nucleoplasm and concentrated in promyelocytic leukemia (PML) bodies (Ishov et al., 1999; Villagra et al., 2004), SUMO-1 nuclear bodies (Navascues et al., 2007) and the nuclear envelope (Gill, 2004; Villagra et al., 2004). The nuclear pattern of SUMO-1 expression is consistent with its role in regulation of gene expression, DNA repair, traffic nucleo-cytoplasmic and subnuclear relocalization of its target proteins (Melchior, 2000; Gill, 2004; Hay, 2005).
138
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
The UR61 cell subline, derived from PC12 cells, is a useful model system for studying cell-signaling pathways involved in neuritogenesis and neuronal differentiation (Guerrero et al., 1988; Navascues et al., 2004, 2007), as well as for analyzing the organization of the cell nucleus in neuron-like culture cells. For example, we have previously shown (Navascues et al., 2004) that the number of CBs and gems, coilin-negative nuclear bodies that concentrate SMN protein (Liu and Dreyfuss, 1996), as well as the composition of CBs, change during the dexamethasone (Dex)-induced differentiation of UR61 cells. Regarding the possible sumoylation of CB-associated proteins, bioinformatic analysis with SUMOplotTM software (http://www.abgent.com/doc/sumoplot) identified several possible sumoylation sites in rat coilin and SMN proteins. Moreover, previous biochemical studies have shown that the SUMO-1 E3 ligase PIASy (Sachdev et al., 2001) directly interacts with coilin (Sun et al., 2005). The aim of this study is to investigate the possible participation of SUMO-1 in the molecular reorganization of CBs associated with cellular differentiation and with the neuronal response to stress. As cellular models we use the neuron-like differentiation of UR61 cells (Navascues et al., 2004, 2007) and two neuronal models of cellular stress induced in living rats by hyperosmolar treatment, in the osmosensitive supraoptic nuclei (SON) neurons (Lafarga et al., 1998), or by treatment with the methyltransferase inhibitor adenosine dialdehyde (Adox) in trigeminal ganglia (TG) neurons. The results show the localization of SUMO-1 in a subset of CBs of undifferentiated UR61 cells and in stressed neurons, particularly during the reformation phase of CBs, but not in normal rat neurons. Overexpression of SUMO-1 or Ubc9 in undifferentiated UR61 cells confirmed the presence of both proteins in a subset of CBs. Collectively, these results suggest that the sumoylation of certain CB-associated proteins may be involved in the molecular reorganization of neuronal CBs associated with differentiation and response to stress.
out at 8:00 AM and free access to food and water. The experimental procedures were performed under the approved national guidelines for animal care and in accordance with guidelines established by the International Association for the Study of Pain (Zimmerman, 1983). To induce osmotic stress, animals were given a single intraperitoneal injection of 1.5 M NaCl (18 ml/kg), as previously described (Lafarga et al., 1998). After 0, 0.5, 2, 6, 12 and 24 h rats were sacrificed and tissue samples of SON were processed for immunofluorescence. To inhibit methyltransferase, animals were given a single intravenous injection of Adox (100 mM/ kg). The animals were sacrificed after 0, 0.5, 4 and 24 h post-injection. As neuronal tissue samples we selected sensory ganglia since these nervous centers are free of the blood brain barrier, which facilitates the access of the Adox to neurons. 2.3. Light microscopy immunocytochemistry and quantification
UR61 cells were cultured in RPMI 1640 medium supplemented with 10% normal calf serum, 100 U/ml gentamycin, as described previously (Greene and Tischler, 1976) and grown on coverslips. To induce a neuron-like differentiation, cultures were exposed to 0.2 lM Dex for 0, 12 and 24 h (Guerrero et al., 1988). Mouse neuroblastoma cells Neuro2A (N2A) were grown in Petri dishes with Dulbecco’s modified Eagle’s medium (DMEM) (Gibco–BRL), supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 U/ml streptomycin. Transfections were performed with the plasmids expressing GFP–SMN (Navascues et al., 2004; Sleeman et al., 2003), GFP–coilin (Shpargel et al., 2003), GFP–SUMO-1, which codifies for active SUMO-1 processed by SUMO proteases to expose a carboxyl terminal diglycine motif (Gostissa et al., 1999; Desterro et al., 2005), and pDsRed1-C1-Ubc9 (kindly provided by Dr. Ronald Hay, University of Dundee, Scotland). Undifferentiated UR61 cells were transfected for 18 h using FuGene 6 transfection reagent (Roche) according to the manufacturer’s instructions. For the inhibition of methylation, undifferentiated UR61 cells were incubated with the vehicle (DMSO) or with the methyltransferase inhibitor adenosine dialdehyde (Adox, Sigma) at a final concentration of 100 lM for 4 h.
For indirect immunofluorescence, UR61 cells were grown on 10 10-mm glass coverslips. The cells were washed twice in phosphate-buffered saline (PBS), fixed with 3.7% formaldehyde (freshly prepared from paraformaldehyde) in PBS for 10 min at room temperature. For the preparation of neuronal tissue samples, rats were perfused through the ascending aorta with 3.7% formaldehyde (freshly prepared from paraformaldehyde) in PBS, pH 7.4, for 15 min at room temperature. SON and TG tissue samples were dissected from 500 lm-thick vibratome slices. The tissue fragments were washed in PBS for 1 h and cut into small fragments and individually transferred to a drop of PBS on a siliconized slide. Then, a coverslip was applied on top of the slide and the tissue was squashed by percussion with a histological needle. The preparation was then frozen in dry ice, and the coverslip removed using a razor blade (Pena et al., 2001). The slides with adhered neurons were sequentially dehydrated in 96% and 70% alcohol at 4 °C for 10 min and rinsed in PBS. For the immunostaining, both UR61 cell samples and neuronal tissue preparations were sequentially incubated with 0.5% Triton X-100 in PBS for 30 min, 0.1 M glycine in PBS for 30 min, and 0.01% Tween 20 in PBS for 5 min. The samples were incubated for 1 h with the primary antibody diluted in PBS, washed in PBS containing 0.05% Tween 20, incubated for 45 min with the appropriate secondary antibodies conjugated with FITC or TexasRed (Jackson ImmunoResearch Laboratories, West Grove, PA) and mounted with VectaShield (Vector Laboratories, Burlingame, CA). Confocal microscopy was performed with a laser scanning microscope (LSM 510; Carl Zeiss, Germany) by using excitation wavelengths of 488 nm (for FITC) and 543 nm (for Texas Red). Each channel was recorded independently, and pseudocolor images were generated and superimposed. TIFF images were transferred to Adobe Photoshop 7.0 software (Adobe Systems Inc.) for presentation. For the quantitative analysis of the proportion of CBs that concentrate SUMO-1, UR61 cells and SON and TG neurons were double immunolabeled with the anti-coilin and anti-SUMO-1 antibodies. Coilin positive and SUMO-1 positive CBs versus coilin positive but SUMO-1-negative CBs were counted by direct examination of the different focal planes throughout cell nuclei, using a 63 oil objective. Three samples of each experimental group were used and at least 100 cells for each cell type were counted. Data were analyzed by the StatView 4.5 software and using the statistical test ANOVA and v2. Statistical significance was established at p < 0.05.
2.2. Animals and treatments
2.4. Antibodies
Male, 3-month-old rats of the Sprague–Dawley strain were kept on a 12 h day, 12 h night light timing regime with lights
The following primary antibodies were used. Rabbit polyclonal anti-coilin antibody serum 204.3 (Bohmann et al., 1995); rabbit
2. Materials and methods 2.1. Cell culture and transfection assays
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
polyclonal and mouse monoclonal anti-SUMO-1 antibodies (Santa Cruz Biotechnology and Zymed, 21C7, respectively), mouse monoclonal anti-U2B” antibody (4G3, Euro Diagnostica, The Netherlands) that recognizes the spliceosomal U2B” snRNP protein (Habets et al., 1989); mouse monoclonal anti-SMN antibody (BD, Transduction Laboratories, San Jose, CA) and mouse monoclonal anti-2,2,7-trimethylguanosine cap (TMG-cap) of spliceosomal snRNPs (Oncogene Research).
139
cine) for 1 h at 37 °C. After washing, the sections were incubated with the goat anti-mouse IgG antibody coupled to 15 nm gold particles (BioCell, UK: diluted 1:50 in PBS containing 1% BSA) Following immunogold labeling, the grids were stained with uranyl acetate and lead citrate and examined with a Phillips EM208 electron microscope operated at 60 kV. As controls, sections were treated as described but with the primary antibody omitted. 3. Results
2.5. Immunoelectron microscopy 3.1. Neuron-like UR61 cells contain a subset of CBs enriched in SUMO-1 For immunoelectron microscopy, UR61 were fixed with 3.7% paraformaldehyde in 0.1 M cacodylate buffer for 1 h at room temperature. The cells were scraped from the dishes, transferred to an Eppendorf tube, and centrifuged for 1 min in a microfuge to pellet the cultures. Pellets were washed with 0.1 M cacodylate buffer, dehydrated in increasing concentrations of methanol at 20 °C, embedded in lowicryl K4M at 20 °C and polymerized with ultraviolet irradiation. Ultrathin sections were sequentially incubated with 0.1 M glycine in PBS (15 min), 5% BSA in PBS (20 min) and the primary mouse monoclonal anti-SUMO-1 antibody (diluted 1:10 in 50 mM Tris–HCl, pH 7.6 containing 1% BSA and 0.1 M gly-
We have previously shown that the neuron-like differentiation of UR61 cells is accompanied by variations in the number and composition of CBs (Navascues et al., 2004). In order to extend these observations, we have investigated whether SUMO-1 may be involved in the molecular reorganization of CBs during the differentiation of UR61 cells. Double immunofluorescence experiments for SUMO-1 and coilin revealed the colocalization of these two proteins in a subset of CBs in both undifferentiated and Dex-induced differentiated UR61 cells (Fig. 1A–F). Given the molecular heterogeneity of CBs in UR61
Fig. 1. Immunofluorescence analysis of undifferentiated (A–C) and Dex-treated (24 h) UR61 cells (D–F) showing the presence of SUMO-1 in CBs. (A–F) Double immunolabeling for coilin and SUMO-1 illustrates the colocalization of both molecular components in CBs. (G–I) UR61 cell containing SUMO-1 in a CB that concentrates spliceosomal snRNPs identified with an antibody that recognizes the TMG-cap of snRNAs. (J–L) Colocalization of SUMO-1 and coilin in two CBs of a neuroblastoma N2A cell. Scale bar = 5 lm.
140
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
cells (Navascues et al., 2004), we next investigated whether SUMO1 positive CBs contain spliceosomal snRNPs identified with the anti-TMG-cap antibody. Co-immunostaining for SUMO-1 and TMG-cap clearly showed the accumulation of snRNP splicing factors in the CBs enriched in SUMO-1 (Fig. 1G–I). CBs containing SUMO-1 were also found in mouse neuroblastoma cells N2A (Fig. 1J–L). The presence of SUMO-1 in CBs from UR61 cells was confirmed with immunoelectron microscopy. Gold particles of SUMO-1 immunoreactivity decorated the typical coiled threads of CBs (Fig. 2A). Our next step was to determine whether the proportion of cells containing at least one CB with SUMO-1, and that of SUMO-1-positive CBs, change during the Dex-induced differentiation of UR61 cells. Both values significantly decreased upon Dex treatment for 24 h (Fig. 2B and C). Furthermore, although the total number of CBs (coilin-positive nuclear bodies) per cell significantly increased during neuron-like differentiation, the number of SUMO-1-positive CBs per cell decreased upon Dex treatment for 24 h (Fig. 2D). These results indicate that the localization of SUMO-1 to CBs is preferentially associated with the undifferentiated state of UR61 cells. To investigate the possible influence of cell proliferation on the localization of SUMO-1 to CBs, undifferentiated UR61 cells were growtharrested by serum starvation for 18 h. No significant changes in the proportion of SUMO-1-positive CBs, and of cells containing these bodies, were detected in serum-starved cells, compared with cells cultured in a medium with serum (data not shown). The accumulation of SUMO-1 in a subset of CBs was confirmed in undifferentiated UR61 transiently expressing GFP–SUMO-1, GFP–coilin or GFP–SMN. Transfection experiments revealed that GFP–SUMO-1 was readily detectable in a subset of CBs immunolabeled with the anti-coilin antibody (Fig. 3A–C). Transient expression of GFP–coilin resulted in a nucleoplasmic diffuse staining pattern in addition to nuclear foci being identified as CBs. Immunostaining experiments confirmed the colocalization of endogenous SUMO-1 and GFP–coilin in certain CBs, whereas other SUMO-1 positive nuclear bodies were free of the fusion protein (Fig. 3D–F). Similar results were observed in UR61 cells transfected with the GFP–SMN construct. The fusion protein appeared distributed diffusely throughout the cytoplasm and in nuclear and cytoplasmic foci, as previously reported (Navascues et al., 2004). The SUMO-1 immunoreactivity signal colocalized with some nuclear GFP–SMN foci (Fig. 3G–I). Having established the presence of SUMO-1 in a subset of CBs, we next studied whether Ubc9, an essential enzyme in the sumoylation cycle (Desterro et al., 1997; Hay, 2005), localized to CBs. Transient expression of DsRed1-Ubc9 in undifferentiated UR61 cells resulted in the formation of nuclear foci of the exogenous Ubc9, some of which containing endogenous coilin were identified as CBs (Fig. 3J–L). Since coilin methylation regulates CB formation (Hebert et al., 2002), we wanted to assay the behavior of SUMO-1 positive CBs under conditions of reduced methyltransferase activity. We therefore investigated the effect of the methyltransferase inhibitor Adox (100 lM) on the localization of SUMO-1 to CBs in undifferentiated UR61 cells. A short treatment with Adox for 4 h, which does not affect cell survival, resulted in a significant increase in the proportion of cells containing SUMO-1-positive CBs (56 ± 0.4% vs 39.1 ± 0.5% in untreated cells, means ± SD) and that of CBs with SUMO-1 (50.1 ± 0.7% vs 37.1 ± 0.3% in untreated cells, means ± SD). Thus, inhibition of methylation increases the association of SUMO-1 with CBs. 3.2. Hyperosmolar stress induced a transient localization of SUMO-1 to CBs in supraoptic neurons We next investigated whether CBs containing SUMO-1 also occur in mammalian neurons from adult nervous tissue. As a
Fig. 2. (A) Immunogold electron microscopy illustrates the presence of gold particles of SUMO-1 immunoreactivity on the coiled threads of a CB. Scale bar = 150 nm. (B and C) Quantitative analysis of the proportion of cells containing at least one CB with SUMO-1 (B), and that of CBs with SUMO-1 (C) during the Dex-induced differentiation of UR61 cells. Bars represent means ± standard deviation. (D) Distribution of coilin-positive and SUMO-1-negative, and coilin-positive and SUMO-1positive CBs during the differentiation of UR61 cells. Bars represent means ± standard deviation.
cellular model system we used rat SON neurons, neurosecretory cells which produce and release the antidiuretic hormone vasopressin involved in the regulation of water and electrolytic balance (Burbach et al., 2001). Previous studies in these neurons have shown that CBs undergo a transcription-dependent disassembly/reassembly cycle in response to the osmotic stress in-
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
141
Fig. 3. (A–C) A UR61 cell expressing GFP–SUMO-1 exhibits several nuclear foci of the fusion protein. One of these foci, identified as a CB, is immunoreactive for coilin. (D–F) Overexpression of GFP–coilin wild-type in an undifferentiated UR61 cell illustrates the colocalization of the fusion protein and the endogenous SUMO-1 in one CB. Another SUMO-1 positive nuclear body is avoid of the fusion protein. (G–I) Overexpression of GFP–SMN wild-type in an undifferentiated UR61 cell showing prominent nuclear and cytoplasmic foci of GFP–SMN, in addition to a diffuse cytoplasmic pool. Note that one of the two nuclear foci appears intensely immunolabeled for SUMO-1. (J–L) Overexpression of DsRed1-Ubc9 in undifferentiated UR61 cells shows the concentration of the fusion protein in a CB immunostained for coilin. Note the presence of a DsRed1-Ubc9-positive and coilin-negative nuclear body. Scale bar = 5 lm.
duced by the intraperitoneal injection of hypertonic saline (HS) solutions (Lafarga et al., 1998). This disassembly/assembly cycle of CBs provides a good model to analyze the possible participation of SUMO-1 in the molecular reorganization of CBs associated with neuronal stress. Double-labeling experiments to detect SUMO-1 and coilin in control SON neurons revealed the typical diffuse nucleoplasmic signal of SUMO-1 and the absence of colocalization of SUMO-1 and coilin in CBs (Fig. 4A–C). As early as 30 min after HS treatment, CBs tended to disappear and coilin relocalized in small perinucleolar caps (Raska et al., 1990; Santama et al., 1996; Lafarga et al., 1998) free of SUMO-1 signal (Fig. 4D–F). After 2 h and 6 h of HS treatment, co-localization of SUMO-1 and coilin was observed in a subset of CBs free in the nucleoplasm or, most frequently, attached to the nucleolus (Fig. 4G–L). By 24 h of induction of the osmotic stress, prominent CBs were observed but most of them lacked SUMO-1 immunoreactivity (Fig. 4M–O). The quantitative analysis of CBs with and without SUMO-1 was performed on squash preparations of SON neurons co-immunostained for coilin and SUMO-1 (Fig. 5A and B). Figs. 5C and D
show that, 2 h after HS treatment, the proportion of cells containing at least one CB with SUMO-1, and that of SUMO-1-positive CBs, increased considerably. The two values increased further at 6 h, progressively decreasing at 12 and 24 h after the HS treatment. Curiously, the mean number of CBs (with and without SUMO-1) per neuron remarkably decreased at 0.5 h after the induction of the osmotic stress. This was followed by a rebound effect showing a progressive increase of CBs, over the control value, at 6, 12 and 24 h after HS treatment (Fig. 5E). Through this disassembly/reassembly cycle, the highest number of SUMO-1 positive CBs per neuron was detected during the period of reformation of CBs, at 6 h after the induction of the osmotic stress (Fig. 5E). 3.3. The reduction of methyltransferase activity induces a transient localization of SUMO-1 to CBs in trigeminal ganglia neurons Having established that the inhibition of methylation with Adox treatment increases the number of SUMO-1 positive CBs in undifferentiated UR61 cells, we next studied whether this
142
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
Fig. 4. Immunofluorescence analysis of distribution of SUMO-1 and coilin in SON neurons from control (A–C) and osmotically stressed (OS) rats at different time points of hypertonic saline injection (D–O). (A–C) In control neurons, SUMO-1 showed a diffuse nucleoplasmic pattern, excluding the nucleolus, but did not concentrate in the CB labeled with the anti-coilin antibody. (D–F) After 30 min of hypertonic saline injection, coilin immunolabeling appeared as perinucleolar caps. (G–L) Following 2 and 6 h of osmotic stress induction, SUMO-1 was diffusely distributed throughout the nucleoplasm, and concentrated in both intranucleolar foci free of coilin (H), and in nucleolus-attached CBs. (M–O) Upon 24 h of hypertonic saline injection, numerous and prominent coilin-positive and SUMO-1-negative CBs were observed. Scale bar = 5 lm.
effect is extensive to neurons from adult nervous tissue. As a cellular model system we used primary sensory neurons from TG, peripheral nervous centers free of the blood brain barrier. As occurs with SON neurons, CBs of control TG neurons lack SUMO-1 (Fig. 6A–C). Upon treatment with Adox for 4 h, a subset of neuronal CBs showed colocalization of SUMO-1 and coilin (Fig. 6D–F). At this time point of Adox treatment, 36 ± 3.5% (means ± SD) of neurons contained at least one CB with SUMO1 and 24.1 ± 2.4% (means ± SD) of CBs were SUMO-1-positive.
However, SUMO-1 was rarely observed in CBs after longer treatment with the methylation inhibitor for 24 h (Fig. 6G–I). Reduction of the methyltransferase activity with Adox also induced changes in the mean number of CBs per neuron, with an initial and significant decrease after 0.5 h of Adox treatment, compared with control neurons, followed by a dramatic increase after 4 h, returning to control values after 24 h of Adox treatment (Fig. 7A–E). The presence of SUMO-1 in CBs correlates with a dramatic increase in the mean number of total CBs per neuron
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
143
Fig. 5. Quantitative analysis of the distribution of SUMO-1-positive CBs in SON neurons at different time points of osmotic stress induction. (A and B) Illustrative example of dissociated SON neurons immunolabeled for coilin (A) and SUMO-1 (B) used for the quantitative analysis. Arrows indicate the colocalization of SUMO-1 and coilin in CBs after 6 h of osmotic stress (OS). Scale bar = 10lm. (C–E) Histograms represent the proportion of neurons containing at least on CB with SUMO-1 (C), and that of SUMO-1-positive CBs (D), as well as the number of CBs per neuron (E, with and without SUMO-1) at different time points of osmotic stress induction. Bars represent means ± standard deviation.
(Fig. 7E). Curiously, although previous studies have shown that Adox treatment affects the interaction between coilin and SMN (Hebert et al., 2002; Boisvert et al., 2002), the formation of gems was not induced in TG neurons. We next investigated whether the CBs with SUMO-1 induced by Adox treatment in TG neurons represent canonical CBs which, in addition to coilin, contain spliceosomal snRNPs and SMN (Gall 2000; Cioce and Lamond, 2005). Double immunolabeling experiments revealed the colocalization of SUMO-1 with SMN or with spliceosomal snRNPs in a subset of CBs (Fig. 6J–O). 4. Discussion Modulation of protein activities by SUMO-dependent modification is known to play a regulatory role in many cellular processes (for review, see Hay, 2005; Geiss-Friedlander and Melchior, 2007). In this study we show that SUMO-1 accumulates in a subset of CBs in undifferentiated UR61 cells and in stressed neurons, and also that GFP-tagged SUMO-1 is concentrated in certain CBs. These findings suggest that certain CB proteins may be modified by SUMO-1. Interestingly, we have also demonstrated that the E2 enzyme Ubc9, the sole conjugating enzyme for sumoylation (Desterro et al., 1997), localizes to a subset
of CBs in UR61 cells expressing DsRed-1-Ubc9. This raises the possibility that certain protein targets may be modified by SUMO-1 in CBs. In this way, bioinformatic analysis with SUMOplotTM software identified four lysines in rat coilin (84, 280, 518 and 522) and two lysines in rat SMN (53 and 117) with a high probability of SUMO-1 modification. Moreover, recent biochemical studies have demonstrated that the SUMO E3 ligase PIASy (Sachdev et al., 2001) interacts directly with coilin in vitro and in vivo (Sun et al., 2005). However, at present, binding studies have been unable to demonstrate the conjugation of coilin with SUMO-1 (Sun et al., 2005). The presence of SUMO-1 in several of the CBs provides additional support in favor of the molecular heterogeneity of these structures (for review, see Cioce and Lamond, 2005). In the case of the UR61 cells, our results indicate that the differentiation of UR61 cells is accompanied by a reduction of CBs containing SUMO-1. Thus, the higher proportion of SUMO-1-positive CBs reported here correlates with an undifferentiated state of the UR61 cells. Furthermore, it is independent of cell proliferation. Our previous studies have shown a molecular reorganization of CBs in differentiating UR61 cells. Thus, SMN is globally upregulated and progressively recruited to CBs as UR61 cells shift to a differentiate state (Navascues et al., 2004). Although the functional sig-
144
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
Fig. 6. Double immunolabeling experiments on dissociated trigeminal ganglia neurons from control (A–C) and experimental rats treated with Adox for 4 h (D–F and J–O). Costaining for SUMO-1 and coilin revealed the absence the colocalization of these two molecules in the CBs of both control (A–C) and Adox-treated neurons for 24 h (G–I). Adox treatment for 4 h induced a proliferation of CBs some of them concentrated SUMO-1 (D–F). (J–O) Double immunofluorescence analysis for SUMO-1 and SMN (J–L) or SUMO-1 and spliceosomal snRNPs, identified with the anti-U2B” antibody (M–O), showed the concentration of SMN and pre-mRNA splicing factors in SUMO-1-positive CBs after 4 h of Adox treatment. Note the immunolabeling of the nuclear envelope for SMN (J) and SUMO-1 (N), as well as the intranucleolar localization of SUMO-1 (K and N). Scale bar = 5 lm.
nificance of the localization of SUMO-1 to CBs is unknown, we suggest that the reversible sumoylation of certain CB proteins, particularly coilin and/or SMN, may be involved in the molecular reorganization of CBs during the neuron-like differentiation of UR61 cells. We have previously demonstrated that SON neurons undergo a profound reorganization of CBs in response to the osmotic stress (Lafarga et al., 1998). This response is characterized by an initial phase (0.5 h after HS treatment) of CB disassembly, relocalization of coilin as perinucleolar caps and decrease in transcrip-
tional activity, followed by a period of CB reformation (Lafarga et al., 1998; present results). We now demonstrate that SUMO1 transiently localizes to CBs during the reformation process. Interestingly, SUMO-1 colocalizes with coilin in some CBs attached to the nucleolus, whereas the adjacent smaller perinucleolar caps of coilin lack SUMO-1. Our previous studies have shown that the perinucleolar caps are progressively replaced by typical CBs during the osmotic stress–response, suggesting that the caps represent CB precursors (Lafarga et al., 1998). Indeed, perinucleolar caps of coilin are observed in mouse embryos before the
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
145
Fig. 7. Treatment with AdOx induces changes in CB number in trigeminal ganglia neurons. (A–D) Immunofluorescence patterns of coilin in control and Adox-tretaed neurons. Typical CBs were observed in neurons of the control (A) and after 24 h of Adox treatment (D). At 0.5 h of Adox treatment coilin relocalized around of the nucleolus (B), whereas a proliferation of CBs was detected at 4 h of treatment (C). (E) Histogram of the distribution of CBs. Note the presence of CBs containing SUMO-1 at 4 h of Adox treatment. Bars represent means ± standard deviation.
appearance of the first CBs (Ferreira and Carmo-Fonseca, 1995). In this context, we suggest that the sumoylation of certain CB proteins, particularly coilin and/or SMN, may be involved in the disassembly/reformation cycle of neuronal CBs induced by osmotic stress. It is also interesting to note that the neuronal stress induced by inhibition of the protein methyltransferase activity in TG neurons is also associated with a transient appearance of SUMO-1-positive CBs. As occurs in SON neurons, the presence of SUMO-1 in a subset of CBs correlates with a stage of reformation of numerous CBs after an earlier stage of CB disassembly and perinucleolar relocalization of the coilin. Although Adox treatment could affect the methylation of several nuclear proteins (Boisvert et al., 2002), previous studies in HeLa cells have shown that this methyltransferase inhibitor reduces the levels of coilin methylation and the interaction of coilin and SMN (Hebert et al., 2002; Boisvert et al., 2002). Thus, coilin methylation seems to be important for the localization of coilin and SMN in CBs (Hebert et al., 2002; Boisvert et al. 2002). Curiously, in TG neurons from adult tissues, as occurs in HeLa and UR61 cells (Hebert et al., 2002; Boisvert et al., 2002; Navascues et al., 2004), Adox treatment does not induce the redistribution of SMN into gems. It does, however, cause the appearance of SUMO-1-positive CBs in TG neurons and also increases the number of these nuclear bodies in undifferentiated UR61 cells. Moreover, SUMO-1-positive CBs contain SMN and snRNPs suggesting that they are functional CBs involved in the biogenesis of spliceosomal snRNPs (Gall 2000; Cioce and Lamond, 2005). Coilin and SMN are putative substrates of SUMO-1 and their reversible sumoylation might play a role in the molecular reassembly of CBs during the neuronal stress–response. Further transfection experiments with constructs expressing mutated coilin or SMN in putative sumoylation lysines will be necessary to determine whether CBs are sumoylation sites for these CB proteins.
Acknowledgments The authors wish to thank R. García-Ceballos for technical assistance. We are also grateful to Dr. A.I. Lamond (University of Dundee, Scotland) for generously providing anti-coilin antibody (204.10) and the plasmid expressing GFP–SMN, Dr. G. Matera (Case Western University, Cleveland, USA) for the plasmid expressing GFP–coilin and Dr. Joana Desterro (IMM, Lisbon, Portugal) for the plasmid expressing GFP-SUMO-1. This work was supported by the following grants: ‘‘Dirección General de Investigacion” of Spain (BFU2005-01030), and Centro de Investigación Biomedica en Red sobre Enfermedades Neurodegenerativas (CIBERNED; CB06/05/ 0037) from Spain. References Andrade, L.E., Chan, E.K., Raska, I., Peebles, C.L., Roos, G., Tan, E.M., 1991. Human antibodies to a novel protein of the nuclear coiled body: immunological characterization and cDNA cloning of p-80-coilin. J. Exp. Med. 173, 1407– 1419. Berciano, M.T., Novell, M., Villagra, N.T., Casafont, I., Bengoechea, R., Val-Bernal, F., Lafarga, M., 2007. Cajal body number and nucleolar size correlate with cell body mass in human sensory ganglia neurons. J. Struct. Biol. 158, 410–420. Bohmann, K., Ferreira, J., Lamond, A.I., 1995. Mutational analysis of p80 coilin indicates a functional interaction between coiled bodies and nucleolus. J. Cell Biol. 68, 817–831. Boisvert, F.M., Cote, J., Boulanger, M.C., Cleroux, P., Bachand, F., Autexier, C., Richard, S., 2002. Symmetrical dimethylarginine methylation is required for the localization of SMN in Cajal bodies and premRNA splicing. J. Cell Biol. 159, 957–969. Burbach, J.P., Luckman, S.M., Murphy, D., Gainer, H., 2001. Gene regulation in the magnocellular hypothalamo-neurohypophysial system. Physiol. Rev. 81, 1197– 1267. Cajal, S.R., 1903. Un sencillo método de coloración selectiva del retículo protoplasmático y sus efectos en los diversos órganos nerviosos de vertebrados e invertebrados. Trab. Lab. Invest. Biol. 2, 129–221. Carmo-Fonseca, M., 2002. New clues to the function of the Cajal body. EMBO Rep. 3, 726–727.
146
J. Navascues et al. / Journal of Structural Biology 163 (2008) 137–146
Carvalho, T., Almeida, F., Calapez, A., Lafarga, M., Berciano, M.T., Carmo-Fonseca, M., 1999. The spinal muscular atrophy disease gene product, SMN: a link between snRNP biogenesis and the Cajal (coiled) body. J. Cell Biol. 147, 715– 728. Cioce, M., Lamond, A.I., 2005. Cajal bodies: a long history of discovery. Annu. Rev. Cell. Dev. Biol. 21, 105–131. Desterro, J.M., Thompson, J., Hay, R.T., 1997. Ubch9 conjugates SUMO but not ubiquitin. FEBS Lett. 417, 297–300. Desterro, J.M., Keegan, L.P., Jaffray, L., Hay, R.T., O’Connell, M.A., Carmo-Fonseca, M., 2005. SUMO-1 modification alters ADAR1 editing activity. Mol. Biol. Cell 16, 5115–5126. Ferreira, J., Carmo-Fonseca, M., 1995. The biogenesis of the coiled body during early mouse development. Development 121, 601–612. Gall, J.G., 2000. Cajal bodies: the first 100 years. Annu. Rev. Cell. Dev. Biol. 16, 273– 300. Gall, J.G., 2001. A role for Cajal bodies in assembly of the nuclear transcription machinery. FEBS Lett. 498, 164–167. Geiss-Friedlander, R., Melchior, F., 2007. Concepts in sumoylation: a decade on. Nat. Rev. Mol. Cell Biol. 8, 947–956. Gill, G., 2004. SUMO and ubiquitin in the nucleus: different functions, similar mechanisms? Genes Dev. 18, 2046–2059. Gostissa, M., Hengstermann, A., Fogal, V., Sandy, P., Scwarz, S.E., Scheffner, M., Del Sal, G., 1999. Activation of p53 by conjugation to the ubiquitin-like protein SUMO-1. EMBO J. 21, 1456–1464. Greene, L.A., Tischler, A.S., 1976. Establishment of a noradrenergic clonal line of rat adrenal pheochomocytoma cells which respond to nerve growth factor. Proc. Natl. Acad. Sci. USA 73, 2424–2428. Guerrero, I., Pellicer, A., Burstein, D.E., 1988. Dissociation of c-fos from ODC expresión and neuronal differentiation in a PC12 subline stably transfected with an inducible N-Ras oncogene. Biochem. Biophys. Res. 150, 1185–1192. Habets, W.J., Hoet, M.H., De Jong, B.A., Van der Kemp, A., Van Venrooij, W.J., 1989. Mapping of B cell epitopes on small nuclear ribonucleoproteins that react with human antibodies as well as with experimentally-induced mouse monoclonal antibodies. J. Immunol. 143, 2560–2566. Hay, R.T., 2005. SUMO: a history of modification. Mol. Cell 18, 1–12. Hebert, M.D., Szymczyk, P.W., Shpargel, K.B., Matera, A.G., 2001. Coilin forms the bridge between Cajal bodies and SMN, the spinal muscular atrophy protein. Genes Dev. 15, 2720–2729. Hebert, M.D., Shpargel, K.B., Ospina, J.K., Tucker, K.E., Matera, A.G., 2002. Coilin methylation regulates nuclear body formation. Dev. Cell 3, 329– 337. Ishov, A.M., Sotnikov, A.G., Negorev, D., Vladimirova, O.V., Neff, N., Kamitani, T., Yeh, E.T.H., Strauss III, J.F., Maul, G.G., 1999. PML is critical for ND10 42 formation and recruits the PML-interacting protein Daxx to this nuclear structure when modified by SUMO-1. J. Cell Biol. 147, 212–233. Lafarga, M., Andres, M.A., Berciano, M.T., Maquiera, E., 1991. Organization of nucleoli and nuclear bodies in osmotically stimulated supraoptic neurons of the rat. J. Comp. Neurol. 308, 329–339.
Lafarga, M., Berciano, M.T., Garcia-Segura, L.M., Andres, M.A., Carmo-Fonseca, M., 1998. Acute osmotic/stress stimuli induce a transient decrease of transcriptional activity in the neurosecretory of supraoptic nuclei. J. Neurocytol. 27, 205–217. Liu, Q., Dreyfuss, G., 1996. A novel nuclear structure containing the survival of motor neurons protein. EMBO J. 15, 3555–3565. Matera, A.G., Frey, M.R., 1998. Nuclear structure ’98. Coiled bodies and gems: janus or gemini? Am. J. Hum. Genet. 63, 317–321. Melchior, F., 2000. SUMO-nonclassical ubiquitin. Annu. Rev. Cell. Dev. Biol. 16, 591– 626. Navascues, J., Berciano, M.T., Tücker, K.E., Lafarga, M., Matera, G., 2004. Targeting SMN to Cajal bodies and nuclear gems during neuritogenesis. Chromosoma 112, 398–409. Navascues, J., Bengoechea, R., Tapia, O., Vaque, J.P., Lafarga, M., Berciano, M.T., 2007. Characterization of a new SUMO-1 nuclear body (SNB) enriched in pCREB, CBP and c-Jun in neuron-like UR61 cells. Chromosoma 116, 441–451. Pena, E., Berciano, M.T., Fernandez, R., Ojeda, J.L., Lafarga, M., 2001. Neuronal body size correlates with the number of nucleoli and Cajal bodies, and with the organization of the splicing machinery in rat trigeminal ganglion neurons. J. Comp. Neurol. 430, 250–263. Raska, I., Ochs, R.L., Andrade, L.E.C., Chan, E.K.L., Burlingame, R., Peebles, C., Gruol, D., Tan, E.M., 1990. Association between nucleolus and the coiled body. J. Struct. Biol. 104, 120–127. Raska, I., Andrade, L.E.C., Ochs, R.L., Chan, E.K.L., Chang, C.-M., Roos, G., Tan, E.M., 1991. Immunological and structural studies of the nuclear coiled body with autoimmune antibodies. Exp. Cell Res. 195, 27–37. Sachdev, S., Bruhn, L., Sieber, H., Pichler, A., Melchior, F., Grosschedl, R., 2001. PIASy, a nuclear matrix-associated SUMO E3 ligase, represses LEF1 activity by sequestration into nuclear bodies. Genes Dev. 15, 3088–3103. Santama, N., Dotti, C.G., Lamond, A.I., 1996. Neuronal differentiation in the rat hippocampus involves a stage-specific reorganization of subnuclear structure both in vivo and in vitro. Eur. J. Neurosci. 8, 892–905. Shpargel, K.B., Ospina, J.K., Tucker, K.E., Matera, A.G., Hebert, M.D., 2003. Control of Cajal body number is mediated by the coilin C-terminus. J. Cell Sci. 14, 9–16. Sleeman, J.E., Trinkle-Mulcahy, L., Prescott, A.R., Ogg, S.C., Lamond, A.I., 2003. Cajal body proteins SMN and coilin show differential dynamic behaviour in vivo. J. Cell Sci. 116, 2039–2050. Sun, J., Hongzhi, X., Subramony, S.H., Hebert, M.D., 2005. Interactions between Coilin and PIASy partially link Cajal bodies to PML bodies. J. Cell Sci. 118, 4995– 5003. Tücker, K.E., Berciano, M.T., Jacobs, E.Y., LePage, E.F., Shpargel, K.B., Rossire, J.J., Chan, E.K., Lafarga, M., Conlon, R.A., Matera, A.G., 2001. Residual Cajal bodies in coilin knockout mice fail to recruit Sm snRNPs and SMN, the spinal muscular atrophy gene product. J. Cell Biol. 154, 293–307. Villagra, N.T., Berciano, J., Altable, M., Navascues, J., Casafont, I., Lafarga, M., Berciano, M.T., 2004. PML bodies in reactive sensory ganglion neurons of the Guillain-Barré syndrome. Neurobiol. Dis. 16, 158–168. Zimmerman, M., 1983. Ethical guidelines for investigation of experimental pain in conscious animals. Pain 16, 109–110.