HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 22, 19961-19966, May 31, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/22/19961    most recent M201799200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hang, J. Articles by Dasso, M. Search for Related Content PubMed PubMed Citation Articles by Hang, J. Articles by Dasso, M. Social Bookmarking                      What's this?

Association of the Human SUMO-1 Protease SENP2 with the Nuclear Pore^*

Jun Hang and Mary Dasso

From the Laboratory of Gene Regulation and Development, NICHD, National Institutes of Health, Bethesda, Maryland 20892-5431

Received for publication, February 22, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

SUMO-1 is a small ubiquitin-like protein that can be covalently conjugated to other proteins. A family of proteases catalyzes deconjugation of SUMO-1-containing species. Members of this family also process newly synthesized SUMO-1 into its conjugatable form. To understand these enzymes better, we have examined the localization and behavior of the human SUMO-1 protease SENP2. Here we have shown that SENP2 associates with the nuclear face of nuclear pores and that this association requires protein sequences near the N terminus of SENP2. We have also shown that SENP2 binds to Nup153, a nucleoporin that is localized to the nucleoplasmic face of the pore. Nup153 binding requires the same domain of SENP2 that mediates its targeting in vivo. Removal of the Nup153-interacting region of SENP2 results in a significant change in the spectrum of SUMO-1 conjugates within the cell. Our results suggest that association with the pore plays an important negative role in the regulation of SENP2, perhaps by restricting its activity to a subset of the conjugated proteins within the nucleus.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

SUMO-1¹ is a ubiquitin-like protein that can be covalently conjugated to other proteins through an isopeptide linkage in a manner similar to ubiquitin (1). The SUMO-1 conjugation pathway utilizes proteins that both show sequence similarity to analogous enzymes in the ubiquitin pathway and utilize similar biochemical mechanisms (1). A large and growing number of SUMO-1 conjugation substrates have been reported in vertebrates (1). Notably, the profile of SUMO-1-conjugated proteins changes substantially in response to altered cellular conditions (see Ref. 2), suggesting that there are mechanisms to control the specificity of conjugation and/or deconjugation of SUMO-1 differentially between distinct substrates.

Unlike enzymes of the SUMO-1 conjugation pathway, enzymes involved in SUMO processing and deconjugation are not closely related by sequence to their ubiquitin counterparts. Rather, known SUMO proteases share sequence homology in their catalytic domains, which is more nearly conserved to viral proteases (3). Two SUMO proteases have been described in budding yeast, Ulp1p (3) and Ulp2p/Smt4p (4, 5). Ulp1p is concentrated near the nuclear periphery (5) and interacts with nuclear pore components in two-hybrid assays (6), while Ulp2p is localized throughout the nucleus (5). ULP1 is an essential gene, and temperature-sensitive (ts) Ulp1p mutants arrest at the G2/M transition of the cell cycle. Ulp2 is not essential, but it is required for normal meiotic development, for regulation of spindle checkpoint arrest, and for chromatin condensation of rDNA during mitosis (4, 5). Interestingly, these proteins do not appear to act in a simple complementary manner, since ulp1-ts/ulp2-null double mutants grow better than ulp2 single mutants under a variety of conditions (5).

In mammals, data base searches find at least seven members of the SUMO protease family (7), some of which have now been confirmed to act as SUMO proteases in vitro (8-11). Outside of their conserved catalytic domain, these proteases possess non-conserved N-terminal extensions of varying lengths and relatively short non-conserved C-terminal sequences. SENP2 was discovered both through its homology to other SUMO proteases (7) and its interactions with murine Axin, a regulator of the Wnt signaling pathway (12). When overexpressed in tissue culture cells or under in vitro conditions, the murine SENP2 homologue (Smt3IP2) cleaves conjugates of SUMO-1, SUMO-2, and SUMO-3 (11).

Here we have shown that full-length human SENP2 associates with nuclear pores in a manner similar to Ulp1 in yeast. This association occurs exclusively with the nuclear face of the pore and requires sequences near the N terminus of SENP2. We have also shown that SENP2 binds specifically to Nup153, a nucleoporin localized to the nucleoplasmic face of the nuclear pore and that this association requires the same domain of SENP2 that mediates its targeting in vivo. Remarkably, a mutant SENP2 protein that is unable to bind Nup153 is significantly more effective in promoting deconjugation of SUMO-1-conjugated species, indicating that localization of SENP2 to the nuclear pore plays an important role in spatially restricting the activity of this enzyme.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Antibodies-- Affinity-purified antibodies against Xenopus Nup153 were a kind gift from K. Ullman (Huntsman Cancer Institute, Salt Lake City). Affinity-purified antibodies against Xenopus Nup98 were a kind gift from M. Powers (Emory University, Atlanta). The mouse monoclonal antibody mAb414 against nucleoporins, mouse monoclonal antibody against RanGAP1, and anti-V5 antibody were purchased from BAbCO (Richmond, CA), Zymed Laboratories, Inc. (South San Francisco, CA), and Strategene (La Jolla, CA) respectively. Mouse monoclonal anti-FLAG antibody and rabbit polyclonal antibody against the green fluorescence protein (GFP) were purchased from CLONTECH, Inc. (Palo Alto, CA). Rhodamine- and Alexa 488-conjugated secondary antibodies were purchased from Molecular Probes, Inc. (Eugene, OR).

DNA Constructs and Protein Expression-- The SENP2 cDNA was amplified from the human universal QUICK-Clone^TM cDNA (CLONTECH). The sequence of the encoded protein was identical to human cDNAs that have been reported by other laboratories (GenBank^TM accession numbers NM021627 and AF151697). The encoded protein has previously been designated as SENP2, and we will use this nomenclature throughout this report. We fused the SENP2 coding sequence in-frame to the 3'-end of the GFP coding sequence by insertion between the EcoRI and SalI sites of pEGFP-C2 (CLONTECH). Similarly, we prepared a vector encoding a version of SENP2 with an N-terminal FLAG tag by insertion of the SENP2 coding region between EcoRI and SalI sites of pCMV-Tag2B (Strategene). Truncation mutants were generated by PCR using Pfu DNA polymerase (Strategene) and subcloned into the same vectors.

The SENP2 truncation mutants were also subcloned into pGEX4T-1 between EcoRI and SalI for expression of glutathione S-transferase (GST) fusion proteins in bacteria. Expression of the recombinant GST·SENP2 fusion proteins was induced with 0.05 mM isopropyl-1-thio--D-galactopyranoside at room temperature for 4 h and purified according to the manufacturer's instructions (Amersham Biosciences). A cDNA encoding the mature form of SUMO-1 (amino acids 1-97) was subcloned into pcDNA4/HisMax C (Invitrogen) between the BamHI and NotI restriction sites, allowing expression with six histidine and Xpress epitope tags at its N terminus. A human RanGAP1 cDNA was amplified from the pET-RanGAP1 plasmid kindly provided by Volker Gerke (University of Muenster, Muenster, Germany). The cDNA was cloned into the pcDNA3.1/V5-His-TOPO TA cloning vector (Invitrogen), allowing its expression with the V5 epitope and six histidine tags at its C terminus. A form of RanGAP1 that cannot be modified by SUMO-1 was generated by PCR using Pfu DNA polymerase to produce a point mutation that substituted lysine 524 with an arginine residue.

Cell Culture, Transfection, and Immunofluorescence-- COS7 and HeLa cells were cultured in Dulbecco's modified Eagle's medium (Invitrogen) containing 10% fetal bovine serum, 100 µg/ml penicillin, 100 µg/ml streptomycin, and 2 mM glutamine (Biofluids, Rockville, MD). In all cases, the cells were transfected using Effectene transfection reagent (Qiagene Inc., Valencia, CA). To measure the effect of SENP2 expression on the overall profile of conjugated substrates (Fig. 5A), COS7 cells were grown in 6-well plates and transfected with vectors expressing His-Xpress-tagged SUMO-1 and EGFP·SENP2, as indicated. To measure the effect of SENP2 expression specifically on RanGAP1 conjugation (Fig. 5B), COS7 cells were transfected with vectors expressing EGFP or EGFP·SENP2, as well as vectors expressing V5-tagged wild type or unconjugatable mutant human RanGAP1. In Fig. 5, A and B, the cells were washed twice with cold phosphate-buffered saline 24 h after transfection and suspended in 125 µl of boiling 2× SDS sample buffer, followed by brief sonication. The samples were analyzed by Western blotting as described below.

HeLa cells were grown for immunofluorescence on glass coverslips. In all figures except Fig. 1B, the cells were fixed for 5 min with 4% formaldehyde plus 2% sucrose in KB buffer (10 mM Tris-Cl, pH 7.7, 150 mM NaCl, 0.1% bovine serum albumin) and permeabilized with 0.1% Triton X-100 in KB buffer. In Fig. 1B, HeLa cells were fixed with 3% formaldehyde for 20 min at room temperature, and cells were alternatively permeabilized with Triton X-100 as described above or permeabilized with 0.004% digitonin at 4 °C for 15 min. In all cases, the coverslips were incubated for 1 h at room temperature in primary and secondary antibodies that had been diluted in KB buffer with 2% normal horse serum. Hoechst 33258 was added to the secondary antibody incubation to stain the DNA. After each incubation, the cells were rinsed five times for 2 min in KB buffer with 2% normal horse serum. The coverslips were mounted in Vectashield (Vector Laboratories, Burlingame, CA). Images were captured on a Zeiss Axioskop microscope with a Hamamatsu Orca II digital CCD camera (Carl Zeiss, Inc., Thornwood, NY) and Openlab software (Improvision, Inc., Lexington, MA).

Western Blotting-- Proteins were resolved in 4-20% SDS-PAGE gel and transferred to polyvinylidene difluoride membrane. The blots were blocked in 5% nonfat dry milk in phosphate-buffered saline (Biofluids) containing 0.15% Tween-20 (PBS-T) at room temperature for 1 h, incubated in primary and horseradish peroxidase-labeled secondary antibodies for 1 h each, thoroughly rinsed with PBS-T after each incubation, and detected using ECL Plus reagents (Amersham Biosciences).

GST Pull-down Assays-- Xenopus interphase egg extracts were prepared exactly as described elsewhere (13). 15 µg of purified recombinant GST·SENP2 fusion proteins were incubated with 15 µl of egg extract plus 500 µl of buffer B (20 mM Tris-Cl, pH 8.0, 50 mM NaCl, 0.5 mM dithiothreitol, 2.5 mM MgCl₂, 0.1% Triton X-100, 10% glycerol) for 5 h at 4 °C. 20 µl of glutathione-agarose beads that had previously been equilibrated with buffer B were added to the reaction and incubated overnight at 4 °C. The beads were washed four times with 1.5 ml of buffer B, and the bound proteins were eluted using 30 µl of SDS sample buffer. The products were subjected to Western blotting using the indicated antibodies.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

SENP2 Is Localized to the Nucleoplasmic Side of the Nuclear Pore-- To better understand the function of individual SUMO proteases, we cloned the human homologue of SENP2 by PCR and performed experiments to confirm that a recombinant fragment of the human SENP2 protein (amino acids 178-590, containing the putative catalytic region) had activity in assays for SUMO-1 processing and isopeptide cleavage of bonds between SUMO-1 and RanGAP1 (data not shown). Our results were very similar to those previously described for the in vitro analysis of mouse SENP2 (11). To pursue a better understanding of SENP2 in vivo, we subcloned the SENP2 open reading frame into plasmid vectors for the expression of fusion proteins encoding SENP2 tagged with either green fluorescent protein (EGFP·SENP2) or a FLAG epitope (FLAG-SENP2).

To determine the localization of SENP2, we examined the localization of a fusion between GFP and the full SENP2 coding region (EGFP·SENP2) in HeLa cells. When expressed at low levels (Fig. 1A), this protein co-localized with the nuclear envelope, whereas EGFP alone was diffusely distributed. At higher levels of expression, we found the fusion protein not only at the nuclear envelope but also within inclusions in the nucleus (data not shown). EGFP·SENP2 distribution overlapped with immunofluorescent staining using the monoclonal antibody 414 (mAb414), which recognizes a family of FXFG-containing nuclear pore proteins (Fig. 1A). Notably, the distribution of eGFP·SENP2 was more restricted than mAb414 staining and appeared to be found primarily on the nuclear side of the nuclear envelope (Fig. 1A, lower panel). The FLAG-SENP2 localized similarly (data not shown), suggesting that the pattern of localization was independent of the fusion moiety. In addition, polyclonal rabbit antibodies directed against SENP2 showed strong staining of the nuclear envelope in immunofluorescence experiments (data not shown), suggesting that the endogenous SENP2 protein is also localized to the nuclear envelope. Together, these observations demonstrate unambiguously that SENP2 is associated with the nuclear envelope and strongly indicate that SENP2 is resident on the nuclear face of the nuclear pore.

View larger version (66K): [in this window] [in a new window]   Fig. 1.   SENP2 localizes to the nuclear side of the pore during interphase. A, HeLa cells were transiently transfected with EGFP·SENP2, fixed, and stained for immunofluorescence using the monoclonal antibody mAb414, which recognizes components of the nuclear pore. The cells were also stained using Hoechst 33258 DNA dye. The cells were examined by confocal laser microscopy. EGFP·SENP2 is shown in green, mAb414 staining is shown in red, and DNA is shown in blue in the right column only. The bottom panels show an enlargement of a partial area of a transfected HeLa cell. B, HeLa cells were transiently transfected with EGFP·SENP2 and stained for immunofluorescence using rabbit polyclonal antibodies against the EGFP moiety (shown in red). The upper panels show cells that were permeabilized with digitonin before incubation with primary antibodies; the lower panels show cells that were permeabilized with Triton X-100.

To test more directly whether SENP2 is restricted to the nuclear side of the pore, we examined EGFP·SENP2 accessibility to anti-GFP antibody staining under different permeabilization conditions (Fig. 1B). When cells were permeabilized with digitonin, which disrupts the plasma membrane but leaves the nuclear envelope intact (14), we observed no staining with anti-GFP antibodies. However, it was clear that the fusion protein was still present because the GFP emission was observed on the nuclear envelope. By contrast, permeabilization with a detergent that disrupts the nuclear envelope (Triton X-100) allowed staining, indicating that the fusion protein could be recognized by anti-eGFP antibodies. These observations show that SENP2 is localized to the nucleoplasmic side of the nuclear envelope.

The SENP2 N Terminus Is Necessary and Sufficient for Nuclear Envelope Targeting-- We wished to determine which sequences within SENP2 are required for its correct targeting to the nuclear envelope. To do this, we made a series of deletions in the SENP2 fusion protein, encompassing both the N and C termini (Fig. 2A). Deletion of as few as 30 amino acids from the N terminus of SENP2 disrupted its association to the pore (Fig. 2B). By contrast, fusion proteins that were extensively deleted from the C terminus, including one that retained only 70 amino acids of the N terminus of SENP2 (EGFP·SENP2-(1-70)), were able to localize correctly to the nuclear envelope (Fig. 3). A similar deletion analysis using FLAG-SENP2 fusion proteins provided essentially identical results (data not shown), confirming that this finding was independent of the fusion epitope used. These observations show that the sequences within the N-terminal 70 amino acids of SENP2 are both necessary and sufficient for its correct localization at the nuclear pore.

View larger version (40K): [in this window] [in a new window]   Fig. 2.   Sequences at the N terminus of SENP2 are required for localization to the nuclear pore. A, schematic representation of eGFP deletion constructs. Association to the pore and binding to hNup153 data are summarized on the right. N/D indicates binding or association has not been determined for this construct. The hatched box indicates the conserved isopeptidase domain. B, eGFP fusion proteins with the indicated N-terminal deletions of SENP2 were transfected into HeLa cells. Localization of the fusion proteins was visualized by eGFP fluorescence (left column, green). DNA stained with Hoechst 33258 dye is shown in the center column (blue), and merged images are shown in the right column. Similar results were observed using indirect immunofluorescence with anti-FLAG antibodies against FLAG-tagged SENP2 proteins.

View larger version (34K): [in this window] [in a new window]   Fig. 3.   Sequences at the N terminus of SENP2 are sufficient for localization to the nuclear pore. EGFP fusion proteins with the indicated segments of SENP2 were transfected into HeLa cells. Localization of the fusion proteins was visualized by eGFP fluorescence (left column, green). DNA stained with Hoechst 33258 dye is shown in the center column (blue), and merged images are shown in the right column. Similar results were observed using indirect immunofluorescence with anti-FLAG antibodies against FLAG-tagged SENP2 proteins.

The N Terminus of SENP2 Interacts with Nup153-- To test whether SENP2 binds to nuclear pore proteins, we utilized interphase Xenopus egg extracts (13) as a source of unassembled, soluble nuclear pore components. We incubated the egg extracts with recombinant GST fusion proteins that encoded different regions of the SENP2 protein. After incubation, we purified the fusion proteins by affinity chromatography and examined whether the samples specifically retained extract proteins that could be recognized in Western blot assays by mAb414 (Fig. 4, upper panel).

View larger version (61K): [in this window] [in a new window]   Fig. 4.   The N terminus of SENP2 interacts with Nup153. The indicated GST·SENP2 fusion proteins were incubated with Xenopus egg extracts. Resultant complexes were purified by affinity to glutathione-Sepharose. The purified proteins were subjected to Western blot analysis using mAb414 (upper panel), anti-Nup153 (middle panel), and anti-Nup98 (lower panel) antibodies. In each panel, the lane order is as follows: starting material (lane 1); GST pull-down products using GST (lane 2); GST·SENP2-(101-590) (lane 3); GST·SENP2-(201-590) (lane 4); GST·SENP2-(1-170) (lane 5); GST·SENP2-(1-70) (lane 6).

Fusion proteins encoding the N terminus of SENP2 (e.g. GST·SENP2-(1-170) and GST·SENP2-(1-70)) specifically retained a mAb414-reactive band that migrated with a mobility corresponding to 180 kDa on gel electrophoresis. Previous characterization of mAb414-reactive proteins in Xenopus egg extracts (15) suggested that the 180-kDa band was likely to be the nucleoporin Nup153 (16). To confirm this identification, we subjected the same samples to Western blotting using antibodies against the Xenopus Nup153 (17). This analysis showed that Nup153 associated with GST·SENP2-(1-170) and GST·SENP2-(1-70) (Fig. 4, lanes 5 and 6) but did not associate with GST or with fusion proteins containing any other region of SENP2 (Fig. 4).

Several additional observations suggested that the retention of Nup153 was highly specific. First, no other mAb414-reactive bands were specifically retained in association with GST or any of the GST fusion proteins. Second, Western blots using antibodies directed against a GLFG nucleoporin associated with the nucleoplasmic side of the pore (18), Nup98, did not recognize any proteins associated with SENP2 (Fig. 4, lower panel). Taken together, our findings strongly suggest that SENP2 interacts with Nup153 in Xenopus egg extracts. This finding is consistent with the localization of SENP2 determined in Figs. 1 and 2, because Nup153 has been demonstrated to be a component of the nuclear basket in vertebrates (19). The specificity of these interactions was independently confirmed by directed two-hybrid analysis in which strong interactions were observed between Nup153 and either full-length SENP2 or the N-terminal domain of SENP2 (data not shown). By contrast, no specific interactions were observed between Nup153 and SENP2 sequences outside of the first 100 amino acids.

SENP2 Targeting Regulates Its Activity against Cellular Substrates-- To determine whether SENP2 localization has any role in regulation of its activity, we transfected COS7 cells with wild type EGFP·SENP2, a mutant lacking the N-terminal 70 amino acids of SENP2 (EGFP·SENP2-(71-590)), and a mutant in which a critical cysteine in the predicted active site of the enzyme was changed to serine (EGFP·SENP2-C/S). EGFP·SENP2-C/S correctly targeted to the nuclear pore in a manner that was indistinguishable from the wild type protein (data not shown). To monitor the pattern of SUMO-1 conjugation within the transfected cells, we simultaneously transfected with a lower concentration of a plasmid expressing tagged SUMO-1 protein (His-Xpress-SUMO-1) and assayed the conjugation of the tagged SUMO-1 by Western blot (Fig. 5).

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Removal of the Nup153 binding region alters SENP2 activity in vivo. A, removal of the N-terminal targeting domains results in hyperactivity of SENP2 against most SUMO-1-conjugated species. Lanes 1-4, EGFP, EGFP·SENP2, EGFP·SENP2-(71-590), and EGFP·SENP2-C/S were co-transfected into COS7 cells with His-Xpress-SUMO-1 (10:1). Lane 5, the COS7 cells were transfected with EGFP·SENP2 in the absence of His-Xpress-SUMO-1. The total cell lysates were subjected to Western blotting with antibodies against the Xpress tag (upper panel) or against the GFP moiety of the EGFP·SENP2 fusion proteins (lower panel). B, overexpression of SENP2 does not alter SUMO-1 conjugation of RanGAP1. Plasmids expressing EGFP·SENP2 were co-transfected into COS7 cells with constructs expressing wild type or mutant V5-tagged RanGAP1 (10:1), as indicated. The total cell lysates were subjected to Western blotting with antibodies against the V5 epitope (upper panel) or against the GFP moiety of the EGFP·SENP2 fusion proteins (lower panel). The unmodified form of RanGAP1 is identified in the control lane where a RanGAP1 mutant lacking the SUMO-1 modification site (RanGAP1-K524R) is expressed.

We observed that His-Xpress-SUMO-1 became conjugated to a variety of proteins when it was co-transfected with a vector expressing EGFP, most easily observed on Western blots as a high molecular weight smear. This spectrum was slightly attenuated when the EGFP expression vector was replaced with a vector expressing wild type SENP2. By contrast, when EGFP·SENP2-C/S was co-expressed with the tagged SUMO-1 it dramatically increased the level of SUMO-1 conjugation within transfected cells, indicating that it was not only enzymatically inactive but also functioned in a dominant fashion to disrupt deconjugation by endogenous SENP2. We do not currently know the mechanism whereby EGFP·SENP2-C/S inhibited the endogenous protein.

Most remarkably, expression of EGFP·SENP2-(71-590) caused the loss of almost all conjugated forms of His-Xpress-SUMO-1, indicating that SENP2 was much more effective in deconjugation of nuclear proteins when it was no longer tethered to the nuclear pore. These results suggest that association with the pore may play an important negative role in the regulation of SENP2, restricting its activity to a subset of the conjugated proteins within the nucleus and that allowing the protein to access other locations within the nucleus promotes inappropriate deconjugation of SUMO-1 species, which are not normally substrates for this enzyme.

Because SENP2 is resident at the nuclear pore, we also examined the capacity of overexpressed SENP2 to alter the conjugation status of RanGAP1 in vivo. RanGAP1 is localized in the cytosol, where SUMO-1 conjugation targets it to the nuclear pore through association with a large nucleoporin, RanBP2 (20, 21). In this experiment, we co-transfected a plasmid expressing V5-tagged RanGAP1 with the plasmid expressing EGFP·SENP2 (Fig. 5B). As a control, we performed similar experiments with a mutant version of RanGAP1 lacking the single lysine residue that becomes modified by SUMO-1 conjugation (RanGAP1-K524R, Ref. 22). EGFP·SENP2 expression did not substantially alter the pattern of RanGAP1 modification. Moreover, under conditions of moderate or even massive EGFP·SENP2 overexpression, RanGAP1 staining at the nuclear pore was retained (data not shown), further arguing that its SUMO-1 modification status is not regulated by SENP2 despite the fact that RanGAP1 can be deconjugated from SUMO-1 in vitro by SENP2 (Ref. 11, data not shown). These observations suggest that the primary substrate of SENP2 at the nuclear pore is unlikely to be RanGAP1.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

SUMO proteases share homology in their catalytic domains but otherwise have widely divergent primary sequences in their N- and C-terminal domains. We have found that human SENP2 localizes to the nucleoplasmic face of nuclear pores through sequences in the N-terminal domain. We have further found that this domain of SENP2 interacts with Nup153, a nucleoporin on the nucleoplasmic face of the nuclear pore. Together, our observations suggest that at least one function of the divergent regions in SUMO proteases is their proper localization within the nucleus. Notably, elimination of the N-terminal domain of SENP2 not only allows it to relocalize to other parts of the nucleus but also increases its capacity to deconjugate SUMO-1-conjugated species within the cell. These observations suggest that targeting SENP2 to the nuclear pore is a mechanism to sequester it from SUMO-1-conjugated proteins in the nuclear interior. Alternatively, the N-terminal domain may have a role in negatively regulating SENP2 through additional mechanisms.

Our findings suggest that interactions between Nup153 and SENP2 may be responsible for the localization of SENP2 at the nuclear pore (Fig. 4). Nup153 is associated with the basket structure on the nucleoplasmic side of the pore (19), and it has been implicated in multiple aspects of nuclear transport and pore structure (17, 23). The association between Nup153 and SENP2 is therefore entirely consistent with immunofluorescence data showing that SENP2 was found only on the nucleoplasmic side of the nuclear envelope (Fig. 1). Ulp1p, a protease for budding yeast SUMO-1 (Smt3), is found at the nuclear pore (5) and has been reported to bind Nup42p (6). Like Nup153, Nup42p is an FG-repeat containing nucleoporin (24). These observations may suggest a conserved role for SUMO-1 deconjugation in the regulation of pore activity. However, there are likely to be some differences between vertebrates and yeast in the details of this function because Nup42p is localized on the cytosolic face of the nuclear envelope in yeast (24).

Although SENP2 was localized to the nuclear pore, we have not yet found any role for SENP2 in nuclear trafficking. Overexpression of wild type or mutant versions of SENP2 did not significantly alter nuclear import or export of a GFP-labeled chimeric model substrate consisting of HIV-1 Rev and a hormone-inducible nuclear localization sequence (Rev-GR-GFP, Ref. 25) (data not shown). Although such negative results do not exclude the possibility that some nuclear transport pathways are controlled by SENP2, they suggest that modulation of SENP2 activity does not grossly alter the role of Nup153 in pore structure or nuclear import.

Interestingly, Pichler et al. (26) have demonstrated that RanBP2 is an E3 enzyme for SUMO-1 that can catalyze both its own hyperconjugation to SUMO-1 and the conjugation of Sp100 in vitro. Pichler et al. (26) have proposed a model wherein RanBP2 may couple nuclear import with the conjugation of a subset of SUMO-1 substrates (26). It would be attractive to speculate that SENP2 might have a role in transport-linked deconjugation of the same subset of SUMO-1-conjugated proteins. Notably, translocation through the nuclear pore is not essential for their efficient conjugation of Sp100 by RanBP2 (26). Furthermore, nonconjugatable forms of Sp100 show no defects in nuclear localization in vivo (27), suggesting that SUMO-1 modification cannot be essential for its nuclear import. These observations further indicate that pore-associated SUMO-1 conjugation and deconjugation activities are unlikely to be involved in nuclear transport per se. An alternative function might be the specific marking of newly imported proteins, perhaps to direct their localization after nuclear entry or to regulate their activity before their assembly into macromolecular complexes within the nucleus (28).

SENP2 is closely related to a rat SUMO-1 protease, Axam, which has been reported as an Axin-binding protein (11, 12). Axam antagonizes the binding of Dvl-1 to Axin and suppresses GSK-3-dependent phosphorylation in the Axin complex, thereby enhancing -catenin degradation (12). Enhancement of -catenin degradation does not require SUMO protease activity (11), suggesting that SENP2 may have multiple, independent functions. Earlier reports on Axam also indicated that Axam and SENP2 (SMT3IP2) are localized to the cytosol (11, 12). We believe that the difference between these observations and ours was that both reports visualized the localization of N-terminal-truncated forms of Axam or SENP2, missing 72 or 42 amino acids, respectively. We suspect that expression levels and cell type-specific differences may also contribute to the differences between our observations and those reports, since we did not observe localization of truncated forms of SENP2 in the cytosol. We cannot currently explain the relationship between the APC/-catenin pathway and SENP2, but it will be of interest to examine whether SENP2 or other SUMO-1 pathway enzymes regulate nuclear translocation or subnuclear localization of any of the proteins of the APC/-catenin pathway.

In summary, we have shown that full-length SENP2 localizes to the nuclear face of the nuclear pore. This localization is likely to be achieved through interaction with Nup153. Appropriate localization to the pore plays an important role for the correct regulation of SENP2 because its mislocalization leads to the inappropriate deconjugation of many SUMO-1-conjugated species. It will be of interest in the future to determine which SUMO-1-conjugated species are normally the targets of SENP2 activity.

	ACKNOWLEDGEMENTS

We thank Jomon Joseph and Tadashi Anan for help in generating constructs expressing tagged human RanGAP1 and SUMO-1. We thank Shyh-Han Tan for help with confocal microscopy. Finally, we thank Alexei Arnaoutouv, Yoshiaki Azuma, Byrn Booth Quimby, and Shyh-Han Tan for critical reading of this report.

	FOOTNOTES

^* This work was supported by Human Science Frontiers Program Research Grant RG0229/1999-M and by NICHD, National Institutes of Health intramural funds (Project No. Z01 HD 01902-05).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed. Tel.: 301-402-1005; Fax: 301-402-1323; E-mail: mdasso@helix.nih.gov.

Published, JBC Papers in Press, March 14, 2002, DOI 10.1074/jbc.M201799200

	ABBREVIATIONS

The abbreviations used are: SUMO, small ubiquitin-like modifier; GFP, green fluorescence protein; EGFP, enhanced GFP; GST, glutathione S-transferase; HIV-1, human immunodeficiency virus, type 1; Axam, Axin associating molecule; APC, adenomatous polyposis coli.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				C. D. Lima and D. Reverter Structure of the Human SENP7 Catalytic Domain and Poly-SUMO Deconjugation Activities for SENP6 and SENP7 J. Biol. Chem., November 14, 2008; 283(46): 32045 - 32055. [Abstract] [Full Text] [PDF]

				N. Fujita, M. Hayashi-Nishino, H. Fukumoto, H. Omori, A. Yamamoto, T. Noda, and T. Yoshimori An Atg4B Mutant Hampers the Lipidation of LC3 Paralogues and Causes Defects in Autophagosome Closure Mol. Biol. Cell, November 1, 2008; 19(11): 4651 - 4659. [Abstract] [Full Text] [PDF]

				A. Lewis, R. Felberbaum, and M. Hochstrasser A nuclear envelope protein linking nuclear pore basket assembly, SUMO protease regulation, and mRNA surveillance J. Cell Biol., August 27, 2007; 178(5): 813 - 827. [Abstract] [Full Text] [PDF]

				X. M. Xu, A. Rose, S. Muthuswamy, S. Y. Jeong, S. Venkatakrishnan, Q. Zhao, and I. Meier NUCLEAR PORE ANCHOR, the Arabidopsis Homolog of Tpr/Mlp1/Mlp2/Megator, Is Involved in mRNA Export and SUMO Homeostasis and Affects Diverse Aspects of Plant Development PLANT CELL, May 1, 2007; 19(5): 1537 - 1548. [Abstract] [Full Text] [PDF]

				T. Makhnevych, C. Ptak, C. P. Lusk, J. D. Aitchison, and R. W. Wozniak The role of karyopherins in the regulated sumoylation of septins J. Cell Biol., April 9, 2007; 177(1): 39 - 49. [Abstract] [Full Text] [PDF]

				R. Zunino, A. Schauss, P. Rippstein, M. Andrade-Navarro, and H. M. McBride The SUMO protease SENP5 is required to maintain mitochondrial morphology and function J. Cell Sci., April 1, 2007; 120(7): 1178 - 1188. [Abstract] [Full Text] [PDF]

				M. D. Benson, Q.-J. Li, K. Kieckhafer, D. Dudek, M. R. Whorton, R. K. Sunahara, J. A. Iniguez-Lluhi, and J. R. Martens SUMO modification regulates inactivation of the voltage-gated potassium channel Kv1.5 PNAS, February 6, 2007; 104(6): 1805 - 1810. [Abstract] [Full Text] [PDF]

				M. C. Jones, L. Fusi, J. H. Higham, H. Abdel-Hafiz, K. B. Horwitz, E. W.-F. Lam, and J. J. Brosens Regulation of the SUMO pathway sensitizes differentiating human endometrial stromal cells to progesterone PNAS, October 31, 2006; 103(44): 16272 - 16277. [Abstract] [Full Text] [PDF]

				S. J. Choi, S. S. Chung, E. J. Rho, H. W. Lee, M. H. Lee, H.-S. Choi, J. H. Seol, S. H. Baek, O. S. Bang, and C. H. Chung Negative Modulation of RXR{alpha} Transcriptional Activity by Small Ubiquitin-related Modifier (SUMO) Modification and Its Reversal by SUMO-specific Protease SUSP1 J. Biol. Chem., October 13, 2006; 281(41): 30669 - 30677. [Abstract] [Full Text] [PDF]

				H. Wu, L. Sun, Y. Zhang, Y. Chen, B. Shi, R. Li, Y. Wang, J. Liang, D. Fan, G. Wu, et al. Coordinated Regulation of AIB1 Transcriptional Activity by Sumoylation and Phosphorylation J. Biol. Chem., August 4, 2006; 281(31): 21848 - 21856. [Abstract] [Full Text] [PDF]

				A. Di Bacco, J. Ouyang, H.-Y. Lee, A. Catic, H. Ploegh, and G. Gill The SUMO-Specific Protease SENP5 Is Required for Cell Division Mol. Cell. Biol., June 15, 2006; 26(12): 4489 - 4498. [Abstract] [Full Text] [PDF]

				Y. Itahana, E. T. H. Yeh, and Y. Zhang Nucleocytoplasmic Shuttling Modulates Activity and Ubiquitination-Dependent Turnover of SUMO-Specific Protease 2 Mol. Cell. Biol., June 15, 2006; 26(12): 4675 - 4689. [Abstract] [Full Text] [PDF]

				L. Gong and E. T. H. Yeh Characterization of a Family of Nucleolar SUMO-specific Proteases with Preference for SUMO-2 or SUMO-3 J. Biol. Chem., June 9, 2006; 281(23): 15869 - 15877. [Abstract] [Full Text] [PDF]

				V. Yurchenko, Z. Xue, and M. J. Sadofsky SUMO Modification of Human XRCC4 Regulates Its Localization and Function in DNA Double-Strand Break Repair. Mol. Cell. Biol., March 1, 2006; 26(5): 1786 - 1794. [Abstract] [Full Text] [PDF]

				A. Yueh, J. Leung, S. Bhattacharyya, L. A. Perrone, K. de los Santos, S.-y. Pu, and S. P. Goff Interaction of Moloney Murine Leukemia Virus Capsid with Ubc9 and PIASy Mediates SUMO-1 Addition Required Early in Infection J. Virol., January 1, 2006; 80(1): 342 - 352. [Abstract] [Full Text] [PDF]

				T. D. Rockel, D. Stuhlmann, and A. von Mikecz Proteasomes degrade proteins in focal subdomains of the human cell nucleus J. Cell Sci., November 15, 2005; 118(22): 5231 - 5242. [Abstract] [Full Text] [PDF]

				T. G. Hofmann, E. Jaffray, N. Stollberg, R. T. Hay, and H. Will Regulation of Homeodomain-interacting Protein Kinase 2 (HIPK2) Effector Function through Dynamic Small Ubiquitin-related Modifier-1 (SUMO-1) Modification J. Biol. Chem., August 12, 2005; 280(32): 29224 - 29232. [Abstract] [Full Text] [PDF]

				I. M. Veltman, L. A. Vreede, J. Cheng, L. H.J. Looijenga, B. Janssen, E. F.P.M. Schoenmakers, E. T.H. Yeh, and A. G. van Kessel Fusion of the SUMO/Sentrin-specific protease 1 gene SENP1 and the embryonic polarity-related mesoderm development gene MESDC2 in a patient with an infantile teratoma and a constitutional t(12;15)(q13;q25) Hum. Mol. Genet., July 15, 2005; 14(14): 1955 - 1963. [Abstract] [Full Text] [PDF]

				S. Chupreta, S. Holmstrom, L. Subramanian, and J. A. Iniguez-Lluhi A Small Conserved Surface in SUMO Is the Critical Structural Determinant of Its Transcriptional Inhibitory Properties Mol. Cell. Biol., May 15, 2005; 25(10): 4272 - 4282. [Abstract] [Full Text] [PDF]

				J. Cheng, N. D. Perkins, and E. T. H. Yeh Differential Regulation of c-Jun-dependent Transcription by SUMO-specific Proteases J. Biol. Chem., April 15, 2005; 280(15): 14492 - 14498. [Abstract] [Full Text] [PDF]