Ubiquitin-dependent Proteolytic Control ofSUMO Conjugates*□S

Received for publication, April 17, 2007, and in revised form, August 21, 2007 Published, JBC Papers in Press, August 29, 2007, DOI 10.1074/jbc.M706505200

Kristina Uzunova‡1, Kerstin Gottsche‡, Maria Miteva‡2, Stefan R. Weisshaar‡§, Christoph Glanemann‡3,Marion Schnellhardt‡, Michaela Niessen‡, Hartmut Scheel¶, Kay Hofmann¶, Erica S. Johnson�,Gerrit J. K. Praefcke‡§, and R. Jurgen Dohmen‡4

From the ‡Institute for Genetics, University of Cologne, Zulpicher Strasse 47, D-50674 Cologne, Germany, the §Center for MolecularMedicine Cologne, University of Cologne, Zulpicher Strasse 47, D-50674 Cologne, Germany, the ¶Bioinformatics Department,Miltenyi Biotec GmbH, MACSmolecular Business Unit, Nattermannallee 1, D-50829 Cologne, Germany, and the �Department ofBiochemistry and Molecular Biology, Thomas Jefferson University, Philadelphia, Pennsylvania 19107

Posttranslational protein modification with small ubiquitin-relatedmodifier (SUMO) is an important regulatorymechanismimplicated in many cellular processes, including several of bio-medical relevance. We report that inhibition of the proteasomeleads to accumulation of proteins that are simultaneously con-jugated to both SUMO and ubiquitin in yeast and in humancells. A similar accumulation of such conjugates was detected inSaccharomyces cerevisiae ubc4 ubc5 cells as well as in mutantslacking twoRING finger proteins, Ris1 andHex3/Slx5-Slx8, thatbind to SUMO as well as to the ubiquitin-conjugating enzymeUbc4. In vitro, Hex3-Slx8 complexes promote Ubc4-dependentubiquitylation. Together these data identify a previously unrec-ognized pathway thatmediates the proteolytic down-regulationof sumoylated proteins. Formation of substrate-linked SUMOchains promotes targeting of SUMO-modified substrates forubiquitin-mediated proteolysis. Genetic and biochemical evi-dence indicates that SUMO conjugation can ultimately lead toinactivation of sumoylated substrates by polysumoylationand/or ubiquitin-dependent degradation. Simultaneous inhibi-tion of both mechanisms leads to severe phenotypic defects.

Small ubiquitin-related modifier (SUMO),5 which is struc-turally related to ubiquitin, is conjugated posttranslationally to

a large number of substrates (1–5). The enzymes mediatingSUMO conjugation are similar to those that catalyze the trans-fer of ubiquitin (3, 4). The functions of these twomodifications,however, are distinct. Posttranslational modification of pro-teins with certain types of ubiquitin chains serves as a second-ary degradation signal that targets such proteins for degrada-tion by the 26 S proteasome (6). SUMO modification, incontrast, is not thought to result in proteolytic targeting (1–3,7). Among the many functions of SUMOmodification are reg-ulation of transcription, nuclear transport, formation of sub-nuclear structures, cell cycle progression, andDNArepair (1–5,7–9). Several substrates can be modified either by ubiquitin orSUMO (10). Modification of PCNA on a specific Lys residueby ubiquitylation mediates DNA repair (10, 11). Sumoyla-tion of the same Lys, in contrast, mediates interaction withthe Srs2 helicase, which results in inhibition of recombina-tion during DNA replication (12, 13). In this example,sumoylation and ubiquitylation appear to direct proliferat-ing cell nuclear antigen into distinct functions by promotingalternative interactions. Sumoylation of I�B� on Lys21 hasbeen proposed to prevent its ubiquitylation and subsequentdegradation (14). It has also been shown that SUMO-1 mod-ification of a pathogenic fragment of Huntingtin enhancesstability of this fragment, thereby increasing neurodegenera-tion, whereas ubiquitylation reduces fragment stability (15).Several recent studies suggested that, similar to ubiquitin,

SUMO can form substrate-linked chains. In Saccharomycescerevisiae, SUMO chain formation does not appear to serve anessential function (16). The reduced ability to remove SUMOchains in the ulp2� mutant, however, appears to be the mainreason for its severe phenotypic defects (16). In mammals,SUMO-2 and SUMO-3 form chains, whereas SUMO-1,which lacks the appropriate Lys residues close to the N terminus,apparently does not (17). The physiological role of SUMO chainsin dividing cells remains largely unclear. They appear to beinvolved in synaptonemal complex formation inmeiosis (18).The levels of SUMO-conjugated versions of substrate pro-

teins are currently thought to be regulated by a dynamic equi-

* This work was supported in part by National Institutes of Health GrantGM62268 (to E. S. J.), European Union Grant MERG-CT-2004-006344 (toG. J. K. P.), and Deutsche Forschungsgemeinschaft Grant SFB635 (toR. J. D.). The costs of publication of this article were defrayed in part bythe payment of page charges. This article must therefore be herebymarked “advertisement” in accordance with 18 U.S.C. Section 1734solely to indicate this fact.

□S The on-line version of this article (available at http://www.jbc.org) containssupplemental Figs. S1–S5 and supplemental Tables S1 and S2.

1 Supported by a Deutsche Forschungsgemeinschaft Graduate Program“Genetics of Cellular Systems” predoctoral fellowship. Present address:Research Institute of Molecular Pathology, Dr. Bohrgasse 7, A-1030 Vienna,Austria.

2 Supported by a Center for Molecular Medicine Cologne predoctoral fellowship.3 Supported by a North Rhine-Westphalia Graduate School “Functional

Genetics and Genomics” predoctoral fellowship. Present address: Cen-ter for Neurologic Diseases, Brigham and Women’s Hospital, HarvardMedical School, Boston, MA 02115.

4 To whom correspondence should be addressed: Institute for Genetics, Uni-versity of Cologne, Zulpicher Str. 47, D-50674 Cologne, Germany. Tel.:49-221-470-4862; E-mail: j.dohmen@uni-koeln.de.

5 The abbreviations used are: SUMO, small ubiquitin-related modifier; GST,glutathione S-transferase; PBS, phosphate-buffered saline; SIP, SUMO-in-

teracting protein; HA, hemagglutinin; SIM, SUMO interaction motif; GAD,Gal4 activation domain; HMW-SC, high molecular weight SUMO conju-gate; E1, ubiquitin-activating enzyme; E2, ubiquitin carrier protein; E3,ubiquitin-protein isopeptide ligase; UPS, ubiquitin/proteasome system;A�, amyloid � peptide.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 282, NO. 47, pp. 34167–34175, November 23, 2007© 2007 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

NOVEMBER 23, 2007 • VOLUME 282 • NUMBER 47 JOURNAL OF BIOLOGICAL CHEMISTRY 34167

by guest on May 29, 2020

http://ww

w.jbc.org/

Dow

nloaded from

librium of conjugation and deconjugation. We report here thatsumoylated proteins can, in addition, be proteolytically down-regulated by the ubiquitin/proteasome system. In S. cerevisiae,a pair of RING type ubiquitin ligases, Ris1 and Hex3/Slx5-Slx8,cooperate with the conjugating enzymes Ubc4 and Ubc5 in theproteolytic control of sumoylated substrates.

EXPERIMENTAL PROCEDURES

Yeast Strains and Methods—Yeast strains and plasmids arelisted in the supplemental tables. Yeast two-hybrid interac-tion cloning was carried out as described using Smt3 as a baitand yeast proteins or protein fragments expressed from agenomic DNA or cDNA libraries as prey (19, 20). Positiveclones were selected on dropout medium lacking adenine.Library plasmids were isolated from positive clones and ana-lyzed by sequencing.Western Blot Analysis of Yeast Crude Extracts—Yeast crude

extracts were prepared using an alkaline lysis/trichloroaceticacid precipitation protocol as described (21). Separating andstacking gels were analyzed by Western blotting with poly-clonal rabbit anti-SUMO serum or monoclonal anti-Cdc11antibodies as described (16, 22).Analysis of Ubiquitin-SUMO Hybrid Conjugates—Extracts

from strains transformed with pKU103 (expressing His6-ubiq-uitin) or a corresponding empty plasmid (p416GAL1; see sup-plemental Table S2) were prepared as described above and pro-cessed for pulldown experiments using nickel-nitrilotriaceticacid-agarose (Qiagen) as described (21). The data shown in Fig.3A have been reproduced once in a similar experiment.GST Pulldown Assay—For expression of GST or GST fusion

proteins, Escherichia coli strain BL21 (DE3) (Stratagene) wastransformed with the corresponding plasmids. Protein expres-sion was induced with 1 mM isopropyl �-D-thiogalactopyrano-side for 3 h at 37 °C. The cells corresponding to 10A600 nm unitswere collected by centrifugation, resuspended in 600 �l ofphosphate-buffered saline (21) with 0.1% Triton X-100 (PBST)supplemented with a protease inhibitor mixture (RocheApplied Science), and subjected to mechanical rupture usingglass beads. The cell debris was removed by centrifugation, andthe supernatants were applied onto 30�l of glutathione-Sepha-rose resin (GEHealthcare). After incubation for 2 h at 4 °Cwithrotation, the beads were collected by centrifugation (4 °C, 1000rpm for 1 min) and washed three times with PBST. Yeast cellextracts, prepared by mechanical rupture using mortar andpestle in PBST, were incubated with the beads for 2 h at 4 °Cwith rotation. The beads were then washed four times withextraction buffer, resuspended in loading buffer containing 0.1M dithiothreitol, and incubated at 100 °C for 5 min before SDS-PAGE. GST pulldown assays shown in Figs. 1B and 3B havebeen reproduced once with similar results.In Vitro Ubiquitylation and SUMO Binding Assays—Hex3-

Slx8-dependent ubiquitylation reactions were carried out in avolumeof 60�l and contained 2�g of ubiquitin (Sigma), 1�g ofubiquitin-activating enzyme (Boston Biochem), 0.04 �g puri-fied His6-Ubc4 expressed in E. coli, 1.6 �g of purified GST-Hex3-Slx8-FLAG co-expressed in E. coli, 2 mM ATP, 2 mM di-thiothreitol, 30 mM Tris-HCl, pH 7.4, 5 mM MgCl2, 100 mM

NaCl, and an ATP-regenerating system (creatine phosphate/phospho-creatine kinase). The reactions were incubated for 2 hat 37 °C. Results similar to those shown in Fig. 3C have beenreproducedmore than three times. To assay for binding of yeastSUMO conjugates, cell pellets corresponding to 20 A600 wereresuspended in 500 �l of lysis buffer (50 mM Tris-HCl, pH 7.5,10% glycerol, 10 mM MgCl2, 1 mM EDTA, 100 mM NaCl) sup-plemented with a protease inhibitor mixture (Roche AppliedScience), 1 mM phenylmethylsulfonyl fluoride, and 10 mM

N-ethylmaleimide. After incubation at 4 °C for 30min, the cellswere lysed with glass beads. The cell debris was removed bycentrifugation, and the supernatants were applied to glutathi-one-Sepharose-bound GST-Hex3/Slx8-FLAG. After incuba-tion for 4 h at 4 °C with rotation, the beads were washed threetimeswith lysis buffer. Boundmaterial was eluted by boiling thebeads in SDS/loading buffer and subjected to Western blotanalysis. Results similar to those shown in Fig. 4 have beenreproduced three times.Analysis of SUMO Conjugates in HeLa Cells—HeLa B cells

(ECACC 85060701) were grown in Eagle’s minimum essentialmedium containing 2 mM glutamine, 1% nonessential aminoacids, 10% fetal bovine serum, and penicillin. Treatment with20�MMG132 (inMe2SO; 2�l/1ml of culture)was done for 8 h.The cells were washed once with PBS before extract prepara-tion using the alkaline lysis/trichloroacetic acid precipitationprocedure described above. SUMO-1 was detected with amouse monoclonal antibody (Zymed Laboratories Inc.),SUMO-2 and SUMO-3 were detected with a rabbit polyclonalantiserum (Abcam), and �-tubulin was detected with a mousemonoclonal antibody (Sigma). For immunoprecipitation ex-periments with FLAG-tagged SUMO, the cells were transientlytransfected using GeneJuice (Novagen) 24 h prior to the addi-tion of MG132. Constructs used for the transfection werepCMV-2b-SUMO1 or -SUMO3, both expressing matureFLAG-tagged versions of the respective SUMO isoforms, or theempty pCMV-2b vector. For the FLAG pulldown assays, �5 �106 cells were suspended and washed in PBS, resuspended in240 �l of lysis buffer (1.85 M NaOH, 7.4% �-mercaptoethanol),and incubated on ice for 10 min. After the addition of 240 �l of50% trichloroacetic acid and incubation on ice for 15 min, theextracts were collected by centrifugation, and the pellet waswashed once with ice-cold acetone. After resuspension in240 �l of TSG buffer (0.5 M Tris base, 6.5% SDS, 12% glycerol,100 mM dithiothreitol), the sample was used for anti-FLAGimmunoprecipitations. The immunoprecipitations wereperformed with 240 �l of extract diluted into 10 ml of radio-immune precipitation assay buffer (50 mM Tris-HCl, pH 7.5,150 mMNaCl, 5 mM EDTA, 1% Triton X-100, protease inhib-itors (Roche Applied Science), 20mMN-ethylmaleimide, and50 �l of anti-FLAG resin (Sigma)) with overnight rotation at4 °C. The resins were washed five times with radioimmuneprecipitation assay buffer containing 0.1% SDS, and proteinswere eluted in 2� SDS loading buffer. Ubiquitylated forms ofFLAG-SUMO conjugates were detected byWestern blottingwith anti-ubiquitin antibody (P4D1; Santa Cruz). Resultsconfirming those shown in Fig. 5 have been reproduced twotimes.

Ubiquitin-dependent Proteolysis of SUMO Conjugates

34168 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 282 • NUMBER 47 • NOVEMBER 23, 2007

by guest on May 29, 2020

http://ww

w.jbc.org/

Dow

nloaded from

RESULTS

Properties of SUMO-interacting Proteins (SIPs) in S.cerevisiae—Using the yeast two-hybrid system with S. cerevi-siae SUMO (Smt3) as bait, we identified five strongly interact-ing SIPs, some of which were represented in multiple overlap-ping clones (Fig. 1A). One clone encoded a peptide of Siz1, a

known SUMO ligase (23). To testwhether interaction of SUMO withthe other four identified SIPs, Nis1/Jip1 (24), Fir1 (25), Hex3/Slx5 (26,27), and Ris1/Dis1 (28), could alsobe detected with their full-lengthversions and whether the interac-tionwas noncovalent, the respectiveSIP-encoding genes were chro-mosomally tagged to express HAepitope-marked versions. In pull-down assays, these four SIPs spe-cifically bound to GST-SUMOexpressed in E. coli, but not to GST,confirming that they where able tobind SUMO (Fig. 1B). Because thebinding assays were performedunder conditions (on ice and with-out addition of ATP) that precludedin vitro SUMO conjugation, theinteractions between these SIPs andSUMO were not a consequence ofcovalent modification of SIPs bySUMO but were apparently nonco-valent in nature. Profile-basedsequence comparisons (22) of thepeptides encoded by these two-hy-brid clones allowed us to identify ashort (�10 residue) peptide motifthat is encoded by all clones. Inthese putative SUMO interactionmotifs (termed a “type a” SIM), apatch of 3–4 hydrophobic residues,is followed by a stretch of 3–4 acidicresidues (Fig. 1C). All two-hybridclones derived from Ris1 and Hex3contain two type a SIMs (Fig. 1, Aand C). Ris1, Nis1, Fir1, and Hex3were also identified as two-hybridinteractors of SUMO in other stud-ies (29–31). One of these studies(31) predicted some of the sameputative SUMO interacting motifsthat were identified by our analysis.Because of different criteria includ-ing a lysine residue upstream of thehydrophobic patch (shown in greenin Fig. 1C), these authors, however,predicted only the SIMs in Nis1 andFir1 as well as SIM1 of Ris1. Thefunctional significance of thesemotifs and the role of these SIPs in

the SUMO conjugation system remained unknown. Severalstudies on mammalian proteins that interacted with SUMO-1,however, suggested that SIMs related to the ones describedhere represent functionally conserved SUMO-binding motifs(32–35). A related but distinct SUMO-bindingmotif has previ-ously been detected in RanBP2 (33). In this case the motif

Hex3/ Uls2

Siz1

Nis1

Ris1/ Uls1

Fir1

407

876

1619

619

904

RING

RING

S/P-RING

4888

543127

627247672250

868102867287

4071240752

876585

407235407352

A

C

Nis1 387 KSNPIIIPDSQDDSIL Fir1 755 KMVEVILLDEDEDVGL Ris1(1) 367 KNSSIIILSDEDESGA Ris1(2) 539 STQQILVDEAENQLNK Hex3(1) 21 NETVILIDSDKEEDAS Hex3(2) 473 KEETIIVTDDDLAKTL Siz1 477 PSEPIIINLDSDDDEP

SUMO interaction motif (SIMa) RING domain

Hex3 PVCCLCGA...ASCGHAFCGRCFA...KLCPADSCK Ris1 MTCFWCME...TGCGHLICDTCIE...IPCKDCQ

1330 1333 1348 1353 1356 1382 13851350

494 497 556 558 561 564 599 604

E

B

extract GST GST- SUMO

Nis1-HA

Fir1-HA

Ris1-HA

Hex3-HA

Slx8 175 EEQQVLQISDDDFQEE Uba2(1) 505 GNGSIILFSDEEGDTM Uba2(2) 579 GKDGIVILDDDEGEIT Wss1 243 SSLEVVILDDDDEVLP RanBP2 2625 DSPSDDDVLIVYELTP

pulldowns

Slx8 YRCPICFE...TLCGHVFCCPCLF...GHCALCR

206 209 221 223 226 229 246 249

Hex3(3) 110 DPSAHYVDLDQEPGSETLETPRTIQVDNTNGYLNDN Ris1(3) 1 MAAVPTIDLTLADSDNEDIFHSFSSSTSVDKIDIRK Ris1(4) 464 HECNSSLDTLKKNHQKLLKDLNSRESELRNALSCCK RNF4 41 TAGDEIVDLTCESLEPVVVDLTHNDSVVIVDERRRP Rfp1 9 IDESSVIDLTRSPSPPVETSISSTNIIDLDAIPDDS Rfp2 14 RLSPEVIDLTEDIEDD-GADVSEVTLLDLTRIPEFQ PIAS1 454 NKKVEVIDLTIDSSSDEEEEEPSAKRTCPSLSPTSP PIASx 464 KKKVDVIDLTIESSSDEEEDPPAKRKCIFMSETQSS

DSUMO interaction motif (SIMb)

type a

type b

*

**

FIGURE 1. SIPs. A, schematic representation of five proteins that were identified as two-hybrid interactors withyeast SUMO/Smt3. The position of putative SIMs (type a in blue; type b in green) and of RING domains (purple)are indicated. SIMs that were experimentally shown to be important for SUMO binding are indicated byasterisks. The sequence stretches encoded in the various two-hybrid clones are given below each protein.B, SIPs bind SUMO in vitro. GST and GST-SUMO expressed in E. coli were bound to glutathione-Sepharose andthen incubated with crude extracts from yeast cells expressing genomically HA-tagged versions of SIPs iden-tified in the two-hybrid screen. Bound material was assayed by SDS-PAGE and anti-HA Western blotting. Theresults were reproduced once. C, alignment of putative type a SIMs (boxed) present in the SIPs shown in A(upper part), as well as in other yeast proteins and mammalian RanBP2 (lower part). Common features of theseSIMs appear to be 3– 4 hydrophobic residues (shown in blue) followed by several acidic ones (shown in red).D, alignment of putative type b SUMO interaction motifs (boxed) present in Hex3 and Ris1, in Rfp1 and Rfp2from S. pombe, as well in human RNF4, PIAS1, and PIASx. E, alignment of RING domains in Hex3, Ris1, and Slx8.The relevant Cys and His residues are shown in purple and a larger font. Hex3 and Slx8 are known to form acomplex (see main text).

Ubiquitin-dependent Proteolysis of SUMO Conjugates

NOVEMBER 23, 2007 • VOLUME 282 • NUMBER 47 JOURNAL OF BIOLOGICAL CHEMISTRY 34169

by guest on May 29, 2020

http://ww

w.jbc.org/

Dow

nloaded from

appears to be in an inversed orientation, i.e. the acidic residuesprecede a hydrophobic stretch. Using profile-based data basesearches, we found relatedmotifs in Slx8, a protein that forms afunctional complex with Hex3 (see above), the Uba2 protein, asubunit of SUMO activating enzyme (36), and Wss1, a proteinthat was previously linked to the SUMO system, although themolecular details of this connection are unknown (37) (Fig. 1C).For Nis1 we obtained in vivo evidence for an interactionbetween the inferred SIMs and SUMO conjugates (supplemen-tal Figs. S1 and S2). Residues both of the hydrophobic stretch aswell as of the acidic patch are important determinants withinthis motif. The SUMO binding properties of Hex3-Slx8 andRis1 appear to be more complex. Similar mutations as for Nis1,when introduced into the two putative type a SIMs of Ris1, hadno detectable effect (data not shown). We used the two-hybridassay to investigate the SUMO binding properties of Hex3 andSlx8 in more detail. We could not detect any two-hybrid inter-actionwith SUMO for Slx8 (data not shown). The SUMO inter-action of the Hex3 segment found in the two-hybrid screen wasonly slightly affected by mutation of the type a SIM2 butstrongly reduced by mutation of SIM1. The double mutantbehaved similarly to the SIM1 singlemutant (supplemental Fig.S3). These results indicated that Ris1 and Hex3 contain multi-ple SUMO-binding sites or possibly interactwith other SUMO-binding proteins. A study fromBoddy and colleagues (38) iden-tified Schizosaccharomyces pombe proteins that appear to befunctionally related to the S. cerevisiae Hex3-Slx8 complex. S.pombe Rfp1 and Rfp2 are redundant proteins that interact withS. pombe Slx8. No apparent sequence homologue of Hex3/Slx5is encoded in the S. pombe genome. DNA damage hypersensi-tivity of double mutants lacking Rfp1 and Rfp2 is comple-mented by the human Protein RNF4, which is related insequence to Rfp1 and Rfp2 (39). None of these proteins displaysextensive sequence similarity to Hex3. A small sequence motif(consensus: (I/V)DL(T/D)), which occurs twice each in Rfp1,Rfp2, and RNF4, however, is also present once in Hex3 andtwice in Ris1. This motif resembles the core sequence of theSIM in PIAS proteins (Fig. 1D) (33). We therefore askedwhether this motif, which we tentatively termed “type b” SIM,contributed to the SUMO binding property of Hex3. To testthis, wemutated the VDLDmotif to AAAD and tested how thismutation affected SUMO binding in the two-hybrid assay.Although the “type b” SIM mutation in Hex3 alone had only asmall effect on SUMO binding, it led to a loss of detectableinteractionwhen combinedwithmutation of type a SIM1 (sup-plemental Fig. S3). Together these data indicated that bothtypes of SIMs in Hex3 contribute to SUMO binding. The com-plex arrangement of SIMs in Hex3 is likely to underlie the pref-erence for binding tomultiply sumoylated proteins (see below).In addition to SIMs,Hex3, Slx8, and Ris1, share a RINGdomain(Fig. 1,A andD) similar to those found inmany ubiquitin ligases(40).Looking for evidence of functional roles of the identified SIPs

in the SUMO system, we askedwhether overexpression of theirGal4 activation domain (GAD) fusions that were obtained inthe two-hybrid screen would have an effect on the pattern ofSUMOconjugates in the cell. Expression ofGAD-Ris1 orGAD-Nis1, and to a lesser extent of GAD-Siz1, GAD-Fir1 or GAD-

Hex3, led to accumulation of high molecular weight SUMOconjugates (HMW-SC) (supplemental Fig. S1). It was reportedrecently that S. cerevisiae SUMO formsHMW-SC in vivo, someof which apparently bear substrate-attached SUMO chains(16). Such polysumoylated proteins accumulated in mutantsthat lacked the SUMO deconjugating enzyme Ulp2. Similar tothe conjugates observed upon overexpression of SIPs, theseconjugates in the ulp2� mutant were most prominentlydetected upon anti-SUMOWestern blotting on the top of sep-arating SDS-polyacrylamide gels and even more strikingly soupon blotting of the corresponding stacking gels (16). Identicaleffects as with the two-hybrid constructs were also obtainedwhen we overexpressed native full-length Nis1 (supplementalFigs. S2 and S4). Together these results suggested that theseSIPs bind and, upon overexpression, stabilize SUMO chains.Accumulation of HMW-SC in these experiments depended onthe presence of lysines residues in SUMO (Lys11, Lys15, andLys19) that have been implicated in formation of SUMO chains(supplemental Fig. S4) (16).Although the effects of overexpressed SIM-containing frag-

ments of SIPs on SUMO conjugate stability described aboveprovided independent in vivo evidence for their SUMObindingproperty, the physiological function of these proteins remainedunclear. It should be noted in this context that the GAD-Ris1and GAD-Hex3 constructs used in these experiments lackedthe RING domains (supplemental Fig. S1). We therefore askedwhether deletion of SIP-encoding genes had an effect onSUMO conjugate patterns. Significantly increased amounts ofHMW-SC were detected in extracts of the ris1� mutant and inthe hex3� strain, and even more strikingly so in the doublemutant (Fig. 2A). Deletion of NIS1 or FIR1, in contrast, had nodetectable effects on SUMO patterns. These results togetherwith the presence of RING domains in both proteins suggestedthat Ris1 andHex3might be ubiquitin ligases (E3s) that controlthe levels of SUMO conjugates. Hex3 was described to be in acomplex with Slx8, another protein containing a RING fingerdomain (26, 41, 42). Hex3 and slx8 null mutants have similarphenotypes and effects on SUMO conjugate accumulation(data not shown), suggesting that the two proteins function as aheterodimer similar to other RING finger ubiquitin ligases suchas BRCA1/BARD1 (27, 43).Degradation of SUMO Conjugates by the UPS—The results

obtained with the ris1� and hex3�mutants prompted us to askwhether the UPS is involved in a control of SUMO conjugates.Consistent with a role of Ris1- and Hex3-dependent ubiquitinconjugation in controlling the level of SUMO conjugates, weobserved a similar accumulation of HMW-SC in a mutant(uba1-ts) carrying a temperature-sensitive ubiquitin-activating(E1) enzyme (22) (Fig. 2B). We went on to test various ubcmutants (ubc1�, ubc2�, ubc4�, ubc5�, ubc4� ubc5�, ubc6�ubc7�, ubc8�, ubc10�, and ubc13�) lacking ubiquitin-conju-gating (E2) enzymes. The ubc4� ubc5� strain lacking theredundant Ubc4 and Ubc5 enzymes (44) displayed a strikingaccumulation of HMW-SC, suggesting that these enzymesmediate a proteolytic control of such conjugates (Fig. 2B). Thepatterns detected in the other ubc mutants, in contrast, weresimilar to those in the wild type (data not shown). Consistentwith a role for the proteasome in the control of SUMO conju-

Ubiquitin-dependent Proteolysis of SUMO Conjugates

34170 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 282 • NUMBER 47 • NOVEMBER 23, 2007

by guest on May 29, 2020

http://ww

w.jbc.org/

Dow

nloaded from

gates, we also observed accumula-tion of HMW-SC in the protea-some-deficient rpt6/cim3 mutant(Fig. 2B) (45). Independent evidencefor the importance of the protea-some in controlling SUMO conju-gates was obtained with the pdr5�mutant, which is particularly sensi-tive to proteasome inhibitors (46).Treatment of this mutant withMG132 resulted in a striking accu-mulation of HMW-SC (Fig. 2B).These results suggested that Hex3and Ris1 together with Ubc4/5mediate a proteolytic control ofSUMOconjugates in targeting themfor degradation by the proteasome.A prediction of the above model

was that SUMO conjugates shouldbe ubiquitylated prior to their deg-radation. To test this prediction, weexpressed His6-tagged ubiquitin(His6-Ub) in wild-type cells as wellas in various mutant strains.Extracts from these cells were pro-duced under denaturing conditionsfollowed by affinity purification ofHis6-Ub conjugates. These conju-gates were then analyzed by West-ern blotting with anti-SUMO anti-serum. SUMO conjugates weredetected when His6-Ub pulldownassays were performedwith extractsfrom wild-type cells, and particu-larly so with extracts from thepdr5� mutant treated with protea-some inhibitor (Fig. 3A). In contrast,ubiquitylated forms of SUMO con-jugates were absent from extractsderived from the ubc4 ubc5 or thehex3� ris1� mutants. Anotherinteresting finding of this analysiswas that particularly strong signalsof ubiquitylated HMW-SC weredetected in pulldown assays fromulp2� extracts. Because Ulp2 hasbeen shown to counteract SUMOchain formation (16), we askedwhether inhibition of SUMO chainformation would affect the forma-tion of ubiquitin-SUMO hybridconjugates. To test this, we used astrain (ulp2� smt3-R11,15,19) inwhich formation of SUMO chainswas impaired because of the muta-tion of lysine residues (to arginines)in three SUMO attachment siteswithin the N-terminal domain of

25

32.5

175

62

83

47.5

stacking gel

fir1 nis1

ris1

hex3

ris1

hex3

wt wt ubc4

ubc5

cim3-1

pdr5

+ DMSO

pdr5

+ MG13

2

pdr5

uba1

-ts26

wt cim3-1

uba1

-ts26

30 C 37 C

Cdc11

A B

anti- SUMO blots

HMW- SC

kDa

anti-

SU

MO

blo

ts

Cdc11

C

wt ulp2

ris1

hex3

smt3-

R11,15

,19

ubc4

ubc5

smt3-

R11,15

,19

ulp2

smt3-

R11,15

,19

smt3-

R11,15

,19

ris1

hex3

ubc4

ubc5

D

1 2 3 4 5 6 7 8

30 C

1

2 3

4

5

6 7

8

37 C

1

2 3

4

5

6 7

8

25

32.5

175

62

83

47.5

stacking gel

kDa

FIGURE 2. SUMO conjugates are stabilized in UPS mutants. A, Hex3 and Ris1 have a function in control-ling SUMO conjugate levels. Shown is an anti-SUMO Western blot analysis of crude extracts from S.cerevisiae strains with the indicated genotypes. B, the UPS controls the levels of SUMO conjugates in yeastcells. A mutation affecting the essential ubiquitin-activating enzyme (uba1-ts26), deletion of the genesencoding the ubiquitin-conjugating enzymes Ubc4 and Ubc5, as well as an inhibition of proteasomefunction either by a mutation (cim3-1) of its Rpt6 subunit or with the proteasome inhibitor MG132 in thepdr5� strain led to an accumulation of SUMO conjugates including high molecular weight forms (HMW-SC). The cells were grown either at 30 °C, and, where indicated, treated for 4 h with 20 �M MG132 (dis-solved in Me2SO (DMSO)), or shifted to 37 °C for 4 h. C, the smt3-R11,15,19 mutation reduces the accumu-lation of high molecular weight SUMO conjugates in ulp2�, hex3� ris1�, or ubc4� ubc5�. Shown is ananti-SUMO Western blot analysis or crude extracts from S. cerevisiae strains with the indicated genotypes.SUMO conjugates patterns of all the strains shown in A–C were analyzed at least three times with similarresults. D, inhibition of SUMO chain formation and SUMO conjugate degradation cause synthetic growthinhibition. Growth properties of the same strains as in C were assayed on YPD plates incubated at 30 or37 °C for 2 days. wt, wild type.

Ubiquitin-dependent Proteolysis of SUMO Conjugates

NOVEMBER 23, 2007 • VOLUME 282 • NUMBER 47 JOURNAL OF BIOLOGICAL CHEMISTRY 34171

by guest on May 29, 2020

http://ww

w.jbc.org/

Dow

nloaded from

SUMO. In contrast to the congenic ulp2� strain, this mutantdid not yield appreciable amounts of ubiquitin-SUMO hybridconjugates (Fig. 3A). Together these in vivo data indicated thatHex3 and Ris1 recognize sumoylated proteins and togetherwith Ubc4,5 mediate their ubiquitylation and that SUMOchains that are stabilized in ulp2� promote formation ofHMWubiquitin-SUMO hybrid conjugates.RING finger type ubiqutin ligases recognize and bind their

substrates, either directly or via substrate binding ancillary fac-tors, and bring them into close contact with their cognate E2s,which are bound to the RING domains (47, 48). We thereforeasked whether Hex3 and Ris1 present in yeast extracts couldbindUbc4. Pulldown experimentswithGST-Ubc4 expressed inE. coli confirmed that both Hex3 and Ris1 bind to Ubc4 (Fig.3B). Experiments described above suggested that Hex3 andRis1 bind sumoylated substrates via complex SUMO-bindingsites and to the Ubc4,5 E2 enzymes via their RING domains. Toinactivate the RING domains of Hex3 and Ris1, wemutated therespective second cysteine residues to serine residues (Fig. 1D).Strains expressing versions of Ris1 or Hex3 carrying thesemutations in their RING domain instead of their wild-type

counterparts accumulatedHMW-SC to similar extents as ris1�or hex3�, respectively, indicating that the RING domains areessential for the observed function of these proteins in the pro-teolytic control of SUMO conjugates (data not shown).Ubiquitin Ligase Activity of Hex3-Slx8 in Vitro—To obtain

biochemical evidence, beyond their interaction with the Ubc4enzyme, for a function of Ris1 and Hex3-Slx8 as ubiquitinligases, we expressed HA-Ris1 and expressed GST-Hex3together with Slx8-FLAG in E. coli. While HA-Ris1 wasexpressed only poorly and remained insoluble, we were able torecover active GST-Hex3-Slx8-FLAG heterodimer from E. coliextracts. As shown in Fig. 3C, Hex3-Slx8 promoted the forma-tion of ubiquitin chains in vitro. Similar to autoubiquitylationreactions that have been observed for other RING type ligases(49), these ubiquitin chains were attached to a large extend toHex3 itself (data not shown). We went on to ask whether theGST-Hex3-Slx8 complex would be able to bind to SUMO con-jugates present in wild type, ulp2�, or hex3� ris1�. In theseexperiments, we found a selective enrichment of HMW-SC,because they are particularly accumulating in ulp2� or hex3�ris1�, among those retained on the GST-Hex3-Slx8 matrix(Fig. 4). These findings were consistent with the idea thatHMW-SC are preferred substrates of the Hex3-Slx8 ubiquitinligase.Multi-level Control of SUMO Conjugates—Mutants lacking

Hex3 and/or Ris1, as well as othermutants deficient in theUPS,accumulated HMW-SC, which at first glance appeared to besimilar to those observed in ulp2� mutants (16) (Fig. 2). Forulp2� it was concluded that formation of such conjugates oflow electrophoretic mobility required SUMO chain formation(polysumoylation) because they were undetectable in strainsthat expressed a mutant SUMO lacking lysine residues (Lys11,Lys15, and Lys19) critical for linking SUMO to SUMO (16). Wetherefore askedwhether it is a function of proteolytic control toprevent a toxic accumulation of polysumoylated proteins. Toinvestigate this question, we replaced the genomic SMT3 geneencoding SUMOby amutant gene expressing SUMO, in whichLys11, Lys15, and Lys19 were replaced by arginine residues(referred to as SUMO-R11,15,19) in wild type, ulp2�, hex3�ris1�, and ubc4� ubc5� (Fig. 2, C and D). Consistent with pre-vious data, SUMO-R11,15,19 expression had little effect on thegrowth properties of wild-type cells but resulted in a strikingsuppression of the ulp2� growth defects (16) (Fig. 2D). Surpris-ingly, however, we found that SUMO-R11,15,19 expressionresulted in a severe synthetic growth inhibitionwhen combinedwith hex3� ris1� and also when combined with ubc4 ubc5 (Fig.2D). Anti-SUMO Western blot analysis revealed that elimina-tion of the branch sites in SUMO resulted in a drastic reductionof HMW-SC not only in ulp2� but also in the hex3� ris1� andubc4� ubc5�mutants (Fig. 2C). Together our data suggest thatformation of SUMO chains and degradation of SUMO conju-gates may have overlapping functions.Proteolytic Control of SUMO Conjugates in Human Cells—

The observed degradation of SUMO conjugates in S. cerevisiaeprompted us to ask whether such a mechanism is also detecta-ble in mammals, whose SUMO conjugation system is morecomplex. In mammals, SUMO-2 and SUMO-3, which are 95%identical (therefore called SUMO-2/3 from hereon), appear to

25

32.5

175

62

83

47.5

kDa

*

175

83

anti-HA blots

no ta

g

Hex3-

HA

Ris1-H

A

GST-Ubc

4

GSTGST-U

bc4

GST

Hex3

Ris1

extracts pulldown

kDa

A

- + - + - + - + - - + +His6-Ub

stacking gel

wt hex3

ris1

+ DM

SO

ubc4

ubc

5

ulp2 +

DMSO

+ M

G132

+ M

G132

anti- SUMO blots

pdr5

B ATP

Hex3-Slx8Ubc4

+++

++-

-++

+-+

anti-Ubiquitin blot

C

- +

ulp2sm

t3-R

11,1

5,19

FIGURE 3. Hex3 and Ris1 cooperate with Ubc4,5 to mediate ubiquityla-tion of SUMO conjugates in S. cerevisiae. A, extracts from cells of the indi-cated genotypes that were transformed with a plasmid expressing His6-Ub orwith an empty vector were subjected to nickel-nitrilotriacetic acid pulldownassays. The proteasome inhibitor hypersensitive strain pdr5� was treated for4 h with Me2SO (DMSO) or Me2SO � MG132 (final concentration, 20 �M)before extract preparation. The bound material was eluted with 500 mM imid-azole and analyzed by anti-SUMO Western blotting. B, extracts fromuntagged yeast cells and from strains expressing genomically HA taggedHex3 or Ris1 were subjected to pulldown assays with GST-Ubc4 expressed inE. coli. Extracts (amounts correspond to 5% of the input in the pulldownassay) and bound material were analyzed by anti-HA Western blotting. Anasterisk marks the position of a protein that cross-reacted with the antibody.C, Hex3-Slx8-dependent in vitro ubiquitylation. The indicated componentswere incubated in the presence of ubiquitin-activating enzyme for 2 h at 37 °Cand analyzed by anti-ubiquitin Western blotting. wt, wild type.

Ubiquitin-dependent Proteolysis of SUMO Conjugates

34172 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 282 • NUMBER 47 • NOVEMBER 23, 2007

by guest on May 29, 2020

http://ww

w.jbc.org/

Dow

nloaded from

form SUMO chains, whereas SUMO-1 does not (see Introduc-tion). We analyzed SUMO-1 as well as SUMO-2/3 conjugatesin HeLa cells after treatment with proteasome inhibitor (Fig. 5,A and B). Inhibition of the proteasome withMG132 resulted ina dramatic accumulation of high molecular weight SUMO-2/3conjugates (Fig. 5B), similar to what we had observed in a com-parable experimentwith yeast cells. Detectionwithmonoclonalantibody specific for SUMO-1, in contrast, showed that pat-terns of SUMO-1 conjugates were less affected by proteasomeinhibition (Fig. 5A). Using HeLa cells that were transfected toexpress FLAG-tagged SUMO-1 or SUMO-3, we detectedSUMO-ubiquitin hybrid conjugates after treatment with pro-teasome inhibitor and FLAG pulldown (Fig. 5C). High molecu-larweight ubiquitylated formswere observed both for SUMO-1and SUMO-3 conjugates. Because of the stronger accumulationof SUMO-3 conjugates upon treatmentwith proteasome inhib-itor, however, much higher amounts of ubiquitylated SUMO-3conjugates were detected. Together these data indicated that inparticular conjugates formed by the SUMO-2/3 isoform aresubject to a proteolytic control by the UPS in human cells, sim-ilar towhatwe observed for SUMOconjugates in yeast. The factthat this mechanism appears not to apply to SUMO-1 conju-gates to the same extent, however, suggested that a preferentialproteolytic control of SUMO-2/3 conjugates provides a func-

tional distinction from SUMO-1 in humans. These findingsraised the possibility that SUMO-2/3 chain formationmay con-tribute to proteolytic control of conjugates of these SUMOparalogues.

DISCUSSION

Contrary to what was believed previously, we have discov-ered that SUMOmodification can serve as a targeting signal inthe ubiquitin/proteasome system (Fig. 6). We show that twoRING finger proteins with SUMO interaction motifs, Ris1 andHex3-Slx8, are required to ubiquitylate SUMO conjugates in S.cerevisiae cells. In vitro experiments (Fig. 3C) confirmed thatthe Hex3-Slx8 heterodimer produced in E. coli possesses ubiq-uitin ligase activity. A parallel study by Hochstrasser and co-

GST GST-Hex3+ Slx8-FLAG GST GST-Hex3+

Slx8-FLAG

bound material unbound material

anti-SUMO blot

extract

input

175

83

62

47.5

32.5

25

kDa

stacking gel

wt

ulp2

hex3

ris

1

4 5 6 7 8 9- 10 11 12 13 14 15-1 2 3w

t

ulp2

hex3

ris

1

wt

ulp2

hex3

ris

1

wt

ulp2

hex3

ris

1

wt

ulp2

hex3

ris

1

FIGURE 4. Preferential binding of high molecular mass SUMO conjugatesto Hex3-Slx8. Glass bead extracts from cells with the indicated genotypeswere analyzed in a GST pulldown assay either with GST or with GST-Hex3-Slx8-FLAG produced in E. coli. In lanes 1–3, samples (amounts correspond tothose used for GST pulldown assays) of the extracts were loaded. In lanes 4 –9,samples of the unbound material, and in lanes 10 –15, samples of the materialbound to the GST matrix were loaded. The samples were analyzed by SDS-PAGE and anti-SUMO Western blotting. The separating and the stacking partsof the gel were simultaneously blotted in this experiment. The positions ofthe prestained molecular mass markers are indicated. wt, wild type.

16.5

25

32.5

175

62

83

47.5

stacking gel

16.5

25

32.5

175

62

83

47.5

stacking gel

+ DM

SO

+ M

G132

+ DM

SO

+ M

G132

-

anti- SUMO-1 blots

anti- SUMO-2/3blots

A B

anti-Tubulin blots

-

kDakDa

C-3-1_FLAG-SUMO -3-1_

anti-Ubiquitin blot

MG132 / anti-FLAG IP

stacking gel

anti-FLAG blot

-HC

-LC

*

25

32.5

175

62

83

47.5

kDa

FIGURE 5. Ubiquitylation and proteasome-mediated degradation ofSUMO conjugates in human cells. HeLa cells were incubated for 8 h eitherwithout any addition, with the addition of Me2SO (DMSO), or with the addi-tion of Me2SO � MG132 (final concentration, 20 �M). A, whole cell extractswere analyzed by SDS-PAGE and anti-SUMO-1 Western blotting. To compareloading, the blot was reprobed with anti-tubulin antibody. B, same as in A, butwith anti-SUMO-2/3 antibody. C, detection of hybrid SUMO-ubiquitin conju-gates in HeLa cells. Extracts from cells transiently transfected to express FLAG-SUMO-1 or FLAG-SUMO-3 and treated for 8 h with MG132 were subjected toanti-FLAG pulldown followed by anti-ubiquitin and anti-FLAG Western blot-ting. wt, wild type; IP, immunoprecipitation.

Ubiquitin-dependent Proteolysis of SUMO Conjugates

NOVEMBER 23, 2007 • VOLUME 282 • NUMBER 47 JOURNAL OF BIOLOGICAL CHEMISTRY 34173

by guest on May 29, 2020

http://ww

w.jbc.org/

Dow

nloaded from

workers (50) provided independent in vitro evidence for a stim-ulation of Hex3-Slx8-mediated substrate ubiquitylation bysumoylation of the substrate. Together these data confirm theconclusion derived from our in vivo analysis that this novelubiquitin ligase mediates targeting of sumoylated proteins tothe proteasome. A similar activity remains to be confirmed invitro for Ris1, which was insoluble upon expression in E. coli.Similar properties of Ris1 and Hex3-Slx8, which include thepresence of RING domains, binding of SUMO conjugates andUbc4, accumulation of SUMO conjugates in the null mutants,and the synthetic effects of the ris1 hex3 doublemutant, suggestthat Ris1, similar to Hex3-Slx8, can function as a SUMO-de-pendent ubiquitin ligase. We therefore propose the terms Uls1and Uls2 (ubiquitin ligases for SUMO conjugates 1 and 2) asalternative names for these proteins to indicate this function(Fig. 6). In S. pombe, a complex of Slx8 with either Rfp1 or Rfp2,two proteins with weak sequence similarity to human RNF4,apparently perform functions similar to Hex3-Slx8 in S. cerevi-siae (38). S. pombe rfp1 rfp2 mutants were complemented byhuman RNF4, indicating that this small RING domain protein,which was previously shown to bind SUMO and to auto-ubiq-uitylate in vitro (51, 52), may have a function similar to the S.pombe Rfp proteins in human cells (39). Consistent with thisnotion, we found that the human RNF4, but not PML, anotherRING finger protein that binds to SUMO, efficiently comple-ments both hex3� and slx8� mutations (supplemental Fig. S5).These data indicate that ubiquitylation of SUMO conjugates isconserved from yeast to humans. Consistent with this notion,we observed an accumulation of ubiquitylated SUMO-2/3 con-jugates in human cells upon inhibition of the proteasome, sug-gesting that a proteolytic control of SUMO conjugates is oper-ating in humans as well. SUMO-1 conjugates, in contrast, weremuch less affected. A possible explanation for this differencebetween conjugates of these isoforms is that SUMO-1, becauseof a lack of appropriate attachment sites in the N-terminaldomain, does not efficiently form chains, whereas SUMO-2/3does (17). In this context it is noteworthy that higher levels ofSUMO-2/3 conjugates are detected after subjecting cells tostresses such as heat (53). This indicates that under these con-ditions either higher conjugation rates are induced, or turnoverof SUMO-2/3 substrates is slowed down because of a stress-

related overload of the UPS. If the latter effect is the cause forthe stress induction of SUMO-2/3 conjugates, these dataresemble the ones obtained after inhibition of the proteasomeusing MG132 (Fig. 5B). If a given substrate can be modifiedeither by SUMO-1 or by SUMO-2/3, then modification withthe former may lead to stabilization whereas modification withthe latter may lead to a destabilization. Noteworthy in this con-text is the observation that the amount of amyloid � peptide(A�), which is generated from amyloid precursor protein andplays a prominent role inAlzheimer’s disease, drops upon over-expression of SUMO-2. Overexpression of SUMO-2(K11R), inwhich SUMO chain formation is blocked, in contrast, results inincreased amounts of A� (54). Even though no sumoylatedforms of amyloid precursor protein were detected in this study,its SUMO-1 modification was detected in a later proteomicstudy (5, 55). Together with our results, these data are consist-ent with a role for SUMO-2/3 chains in regulating A� forma-tion via controlling stability of amyloid precursor protein.Interestingly, inhibition of the proteasome was also reported toenhance A� formation (56).Similar to SUMO-2/3 in mammals, yeast SUMO was

reported to form chains, but the functional significance remainedunclear (16). Our data indicate that formation of SUMO chains(polysumoylation) promotes ubiquitin-mediated targeting. Arequirement for multiple SUMOmoieties in efficient targeting isconsistent with the observation that Hex3-Slx8 and Ris1 containmultiple putative SUMO interactionmotifs (Fig. 1).Protein modification by SUMO in many cases is a transient

and cell cycle-controlled process that is critically controlled byconjugation and deconjugation (21, 57–60). SUMO modifica-tion has been shown tomediate specific interactions, for exam-ple between SUMO-PCNA and Srs2 (12, 13). Our current datasuggest that several mechanisms, including desumoylation,polysumoylation, and degradation, may be employed to termi-nate such transient interactions.It is alsopossible that for certain substrates theprimary function

of SUMO attachment is to target them for ubiquitin-dependentproteolysis. In this caseSUMOattachment couldprovidea secondlevel of control in the regulationofproteindestruction.Themech-anismdiscovered in this studymayprovideaplausible explanationfor the observed induction of turnover of certainmammalian pro-teins following their sumoylation (61–63).

Acknowledgments—We thank Michael N. Boddy and Mark Hoch-strasser for communicating data prior to publication.We are gratefulto Ingrid Schwienhorst for generating the GBD-SUMO construct usedas bait in the two-hybrid screen and to Christiane Horst for technicalassistance.We thank StevenElledge, Phil James,Gregor Jansen, StefanJentsch, Mark Hochstrasser, Kiran Madura, and Carl Mann for gen-erously providing yeast strains, plasmids, and two-hybrid libraries.

Addendum—While this paper was in press, another study by Sun etal. provided evidence for a SUMO-dependent ubiquitin ligase activ-ity of Rfp1,2-Slx8 and of RNF4 (64).

REFERENCES1. Melchior, F., Schergaut, M., and Pichler, A. (2003) Trends Biochem. Sci.

28, 612–6182. Seeler, J. S., and Dejean, A. (2003) Nat. Rev. Mol. Cell. Biol. 4, 690–699

substratedegradation by the 26S proteasome

Ulp2

Ubc4,5Uls1,2

Ubc4,5Uls1,2

Ulp2

ubiquitin

SUMO

FIGURE 6. Proteolytic control of SUMO conjugates in S. cerevisiae. Shownis a model of the role of SUMO chain formation and SUMO conjugate degra-dation. Substrates marked either by multi-sumoylation (upper branch) or bypolysumoylation (SUMO chain formation, lower branch; link between upperand lower branch) are recognized and ubiquitylated by Uls1,2 (Uls1/Ris1 andUls2/Hex3-Slx8) together with the conjugating enzymes Ubc4 and Ubc5.SUMO chains apparently enhance targeting, which is indicated by the thickarrow. SUMO chain formation is counteracted by the activity of the Ulp2 des-umoylating enzyme. SUMO chain formation and SUMO conjugate degrada-tion are two alternative mechanisms to inactivate mono- or multi-sumoy-lated substrates.

Ubiquitin-dependent Proteolysis of SUMO Conjugates

34174 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 282 • NUMBER 47 • NOVEMBER 23, 2007

by guest on May 29, 2020

http://ww

w.jbc.org/

Dow

nloaded from

3. Johnson, E. S. (2004) Annu. Rev. Biochem. 73, 355–3824. Dohmen, R. J. (2004) Biochim. Biophys. Acta 1695, 113–1315. Kerscher, O., Felberbaum, R., andHochstrasser,M. (2006)Annu. Rev. Cell

Dev. Biol. 22, 159–1806. Pickart, C. M., and Fushman, D. (2004) Curr. Opin. Chem. Biol. 8,

610–6167. Hay, R. T. (2005)Mol. Cell 18, 1–128. Kirkin, V., and Dikic, I. (2007) Curr. Opin. Cell Biol. 19, 199–2059. Zhao, X., and Blobel, G. (2005) Proc. Natl. Acad. Sci. U. S. A. 102,

4777–478210. Ulrich, H. D. (2005) Trends Cell Biol. 15, 525–53211. Hoege, C., Pfander, B., Moldovan, G. L., Pyrowolakis, G., and Jentsch, S.

(2002) Nature 419, 135–14112. Pfander, B., Moldovan, G. L., Sacher, M., Hoege, C., and Jentsch, S. (2005)

Nature 436, 428–43313. Papouli, E., Chen, S., Davies, A. A., Huttner, D., Krejci, L., Sung, P., and

Ulrich, H. D. (2005)Mol. Cell 19, 123–13314. Desterro, J. M., Rodriguez, M. S., and Hay, R. T. (1998) Mol. Cell 2,

233–23915. Steffan, J. S., Agrawal, N., Pallos, J., Rockabrand, E., Trotman, L. C., Slepko,

N., Illes, K., Lukacsovich, T., Zhu, Y. Z., Cattaneo, E., Pandolfi, P. P.,Thompson, L. M., and Marsh, J. L. (2004) Science 304, 100–104

16. Bylebyl, G. R., Belichenko, I., and Johnson, E. S. (2003) J. Biol. Chem. 278,44112–44120

17. Tatham, M. H., Jaffray, E., Vaughan, O. A., Desterro, J. M., Botting, C. H.,Naismith, J. H., and Hay, R. T. (2001) J. Biol. Chem. 276, 35368–35374

18. Cheng, C. H., Lo, Y. H., Liang, S. S., Ti, S. C., Lin, F. M., Yeh, C. H., Huang,H. Y., and Wang, T. F. (2006) Genes Dev. 20, 2067–2081

19. James, P., Halladay, J., and Craig, E. A. (1996) Genetics 144, 1425–143620. Bai, C., Elledge, S. J. (1997)Methods Enzymol. 283, 141–15621. Johnson, E. S., and Blobel, G. (1999) J. Cell Biol. 147, 981–99422. Palanimurugan, R., Scheel, H., Hofmann, K., and Dohmen, R. J. (2004)

EMBO J. 23, 4857–486723. Johnson, E. S., and Gupta, A. A. (2001) Cell 106, 735–74424. Iwase, M., and Toh-e, A. (2001). Genes Genet. Syst. 76, 335–34325. Russnak, R., Pereira, S., and Platt, T. (1996) Gene Expr. 6, 241–25826. Mullen, J. R., Kaliraman, V., Ibrahim, S. S., and Brill, S. J. (2001) Genetics

157, 103–11827. Wang, Z., Jones, G. M., and Prelich, G. (2006) Genetics 172, 1499–150928. Zhang, Z., and Buchman, A. R. (1997)Mol. Cell. Biol. 17, 5461–547229. Uetz, P., Giot, L., Cagney, G., Mansfield, T. A., Judson, R. S., Knight, J. R.,

Lockshon, D., Narayan, V., Srinivasan, M., Pochart, P., Qureshi-Emili, A.,Li, Y., Godwin, B., Conover, D., Kalbfleisch, T., Vijayadamodar, G., Yang,M., Johnston, M., Fields, S., and Rothberg, J. M. (2000) Nature 403,623–627

30. Ito, T., Chiba, T., Ozawa, R., Yoshida, M., Hattori, M., and Sakaki, Y.(2001) Proc. Natl. Acad. Sci. U. S. A. 98, 4569–4574

31. Hannich, J. T., Lewis, A., Kroetz, M. B., Li, S. J., Heide, H., Emili, A., andHochstrasser, M. (2005) J. Biol. Chem. 280, 4102–4110

32. Minty, A., Dumont, X., Kaghad, M., and Caput, D. (2000) J. Biol. Chem.275, 36316–36323

33. Song, J., Durrin, L. K., Wilkinson, T. A., Krontiris, T. G., and Chen, Y.(2004) Proc. Natl. Acad. Sci. U. S. A. 101, 14373–14378

34. Song, J., Zhang, Z., Hu, W., and Chen, Y. (2005) J. Biol. Chem. 280,40122–40129

35. Hecker, C. M., Rabiller, M., Haglund, K., Bayer, P., and Dikic, I. (2006)J. Biol. Chem. 281, 16117–16127

36. Johnson, E. S., Schwienhorst, I., Dohmen, R. J., and Blobel, G. (1997)EMBO J. 16, 5509–5519

37. Biggins, S., Bhalla, N., Chang, A., Smith, D. L., and Murray, A. W. (2001)Genetics 159, 453–470

38. Prudden, J., Pebernard, S., Raffa, G., Slavin, D. A., Perry, J. J. P., Tainer, J. A.,McGowan, C. H., and Boddy, M. N. (2007). EMBO J. 26, 4089–4101

39. Kosoy, A., Calonge, T. M., Outwin, E. A., and O’Connell, M. J. (2007).J. Biol. Chem. 282, 20388–20394

40. Joazeiro, C. A., and Weissman, A. M. (2000) Cell 102, 549–55241. Yang, L., Mullen, J. R., and Brill, S. J. (2006) Nucleic Acids Res. 34,

5541–555142. Zhang, C., Roberts, T. M., Yang, J., Desai, R., and Brown, G. W. (2006)

DNA Repair 5, 336–34643. Brzovic, P. S., Rajagopal, P., Hoyt, D. W., King, M. C., and Klevit, R. E.

(2001) Nat. Struct. Biol. 8, 833–83744. Seufert, W., and Jentsch, S. (1990) EMBO J. 9, 543–55045. Ghislain, M., Udvardy, A., and Mann, C. (1993) Nature 366, 358–36246. Fleming, J. A., Lightcap, E. S., Sadis, S., Thoroddsen, V., Bulawa, C. E., and

Blackman, R. K. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 1461–146647. Zheng, N., Wang, P., Jeffrey, P. D., and Pavletich, N. P. (2000) Cell 102,

533–53948. Petroski,M. D., andDeshaies, R. J. (2005)Nat. Rev.Mol. Cell. Biol. 6, 9–2049. Lorick, K. L., Jensen, J. P., Fang, S., Ong, A.M., Hatakeyama, S., andWeiss-

man, A. M. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 11364–1136950. Xie, Y., Kerscher, O., Kroetz, M. B., McConchie, H. F., Sung, P., and

Hochstrasser, M. (September 11, 2007). J. Biol. Chem. 10.1074/jbc.M706025200

51. Hakli, M., Karvonen, U., Janne, O. A., and Palvimo, J. J. (2005) Exp. CellRes. 304, 224–233

52. Hakli, M., Lorick, K. L., Weissman, A. M., Janne, O. A., and Palvimo, J. J.(2004) FEBS Lett. 560, 56–62

53. Saitoh, H., and Hinchey, J. (2000) J. Biol. Chem. 275, 6252–625854. Li, Y., Wang, H., Wang, S., Quon, D., Liu, Y. W., and Cordell, B. (2003)

Proc. Natl. Acad. Sci. U. S. A. 100, 259–26455. Gocke, C. B., Yu, H., and Kang, J. (2005) J. Biol. Chem. 280, 5004–501256. Flood, F., Murphy, S., Cowburn, R. F., Lannfelt, L., Walker, B., and John-

ston, J. A. (2005) Biochem. J. 385, 545–55057. Li, S. J., and Hochstrasser, M. (1999) Nature 398, 246–25158. Schwienhorst, I., Johnson, E. S., andDohmen, R. J. (2000)Mol. Gen. Genet.

263, 771–78659. Bachant, J., Alcasabas, A., Blat, Y., Kleckner, N., and Elledge, S. J. (2002)

Mol. Cell 9, 1169–118260. Stead, K., Aguilar, C., Hartman, T., Drexel, M., Meluh, P., and Guacci, V.

(2003) J. Cell Biol. 163, 729–74161. Lallemand-Breitenbach, V., Zhu, J., Puvion, F., Koken, M., Honore, N.,

Doubeikovsky, A., Duprez, E., Pandolfi, P. P., Puvion, E., Freemont, P., andde The, H. (2001) J. Exp. Med. 193, 1361–1371

62. Isik, S., Sano, K., Tsutsui, K., Seki,M., Enomoto, T., Saitoh,H., andTsutsui,K. (2003) FEBS Lett. 546, 374–378

63. Ghioni, P., D’Alessandra, Y., Mansueto, G., Jaffray, E., Hay, R. T., LaMan-tia, G., and Guerrini, L. (2005) Cell Cycle 4, 183–190

64. Sun, H., Leverson, J., and Hunter, T., (2007) EMBO J. 26, 4102–4112

Ubiquitin-dependent Proteolysis of SUMO Conjugates

NOVEMBER 23, 2007 • VOLUME 282 • NUMBER 47 JOURNAL OF BIOLOGICAL CHEMISTRY 34175

by guest on May 29, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Erica S. Johnson, Gerrit J. K. Praefcke and R. Jürgen DohmenGlanemann, Marion Schnellhardt, Michaela Niessen, Hartmut Scheel, Kay Hofmann, Kristina Uzunova, Kerstin Göttsche, Maria Miteva, Stefan R. Weisshaar, Christoph

Ubiquitin-dependent Proteolytic Control of SUMO Conjugates

doi: 10.1074/jbc.M706505200 originally published online August 29, 20072007, 282:34167-34175.J. Biol. Chem. 

  10.1074/jbc.M706505200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

  http://www.jbc.org/content/suppl/2007/08/30/M706505200.DC1

  http://www.jbc.org/content/282/47/34167.full.html#ref-list-1

This article cites 64 references, 27 of which can be accessed free at

by guest on May 29, 2020

http://ww

w.jbc.org/

Dow

nloaded from