Defining the SUMO-modified Proteome by Multiple Approaches in Saccharomyces cerevisiae*

SUMO, or Smt3 in Saccharomyces cerevisiae, is a ubiquitin-like protein that is post-translationally attached to multiple proteins in vivo. Many of these substrate modifications are cell cycle-regulated, and SUMO conjugation is essential for viability in most eukaryotes. However, only a limited number of SUMO-modified proteins have been definitively identified to date, and this has hampered study of the mechanisms by which SUMO ligation regulates specific cellular pathways. Here we use a combination of yeast two-hybrid screening, a high copy suppressor selection with a SUMO isopeptidase mutant, and tandem mass spectrometry to define a large set of proteins (>150) that can be modified by SUMO in budding yeast. These three approaches yielded overlapping sets of proteins with the most extensive set by far being those identified by mass spectrometry. The two-hybrid data also yielded a potential SUMO-binding motif. Functional categories of SUMO-modified proteins include SUMO conjugation system enzymes, chromatin- and gene silencing-related factors, DNA repair and genome stability proteins, stress-related proteins, transcription factors, proteins involved in translation and RNA metabolism, and a variety of metabolic enzymes. The results point to a surprisingly broad array of cellular processes regulated by SUMO conjugation and provide a starting point for detailed studies of how SUMO ligation contributes to these different regulatory mechanisms. SUMO, or Smt3 in Saccharomyces cerevisiae, is a ubiquitin-like protein that is post-translationally attached to multiple proteins in vivo. Many of these substrate modifications are cell cycle-regulated, and SUMO conjugation is essential for viability in most eukaryotes. However, only a limited number of SUMO-modified proteins have been definitively identified to date, and this has hampered study of the mechanisms by which SUMO ligation regulates specific cellular pathways. Here we use a combination of yeast two-hybrid screening, a high copy suppressor selection with a SUMO isopeptidase mutant, and tandem mass spectrometry to define a large set of proteins (>150) that can be modified by SUMO in budding yeast. These three approaches yielded overlapping sets of proteins with the most extensive set by far being those identified by mass spectrometry. The two-hybrid data also yielded a potential SUMO-binding motif. Functional categories of SUMO-modified proteins include SUMO conjugation system enzymes, chromatin- and gene silencing-related factors, DNA repair and genome stability proteins, stress-related proteins, transcription factors, proteins involved in translation and RNA metabolism, and a variety of metabolic enzymes. The results point to a surprisingly broad array of cellular processes regulated by SUMO conjugation and provide a starting point for detailed studies of how SUMO ligation contributes to these different regulatory mechanisms. Many types of post-translational protein modifications alter protein function, and in some cases the modifying group is itself a protein. The prototypical example of this is ubiquitin, a small, highly conserved polypeptide that is reversibly linked to many different proteins (probably thousands) in the cell (1Peng J. Schwartz D. Elias J.E. Thoreen C.C. Cheng D. Marsischky G. Roelofs J. Finley D. Gygi S.P. Nat. Biotechnol. 2003; 21: 921-926Crossref PubMed Scopus (1278) Google Scholar). Polypeptides distinct from but related to ubiquitin, called ubiquitin-like proteins or Ubls, can also be ligated to proteins (2Hochstrasser M. Nat. Cell Biol. 2000; 2: E153-E157Crossref PubMed Scopus (362) Google Scholar, 3Schwartz D.C. Hochstrasser M. Trends Biochem. Sci. 2003; 28: 321-328Abstract Full Text Full Text PDF PubMed Scopus (316) Google Scholar). Ligation to each Ubl has unique mechanistic and functional consequences. SUMO (Smt3 in the yeast Saccharomyces cerevisiae) is a divergent Ubl that has crucial roles in many organisms (4Seeler J.S. Dejean A. Nat. Rev. Mol. Cell. Biol. 2003; 4: 690-699Crossref PubMed Scopus (569) Google Scholar, 5Johnson E.S. Annu. Rev. Biochem. 2004; 73: 355-382Crossref PubMed Scopus (1358) Google Scholar). Vertebrates have four SUMO variants, SUMO1–SUMO4, whereas yeasts have only one. Only human SUMO1, and not the other SUMO variants, can substitute for the essential yeast protein (6Johnson P.R. Hochstrasser M. Trends Cell Biol. 1997; 7: 408-413Abstract Full Text PDF PubMed Scopus (68) Google Scholar). Activation and conjugation reactions involving SUMO have much in common with those of ubiquitin (reviewed in Refs. 2Hochstrasser M. Nat. Cell Biol. 2000; 2: E153-E157Crossref PubMed Scopus (362) Google Scholar, 5Johnson E.S. Annu. Rev. Biochem. 2004; 73: 355-382Crossref PubMed Scopus (1358) Google Scholar, and 7Melchior F. Annu. Rev. Cell Dev. Biol. 2000; 16: 591-626Crossref PubMed Scopus (646) Google Scholar). Attachment of substrates to Smt3/SUMO depends on a heterodimeric SUMO-activating enzyme (E1), 1The abbreviations used are: E1, SUMO-activating enzyme; E2, SUMO-conjugating enzyme; E3, SUMO-protein ligase; ORF, open reading frame; MS, mass spectrometry; GBD, Gal4 DNA-binding domain; X-gal, 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside.1The abbreviations used are: E1, SUMO-activating enzyme; E2, SUMO-conjugating enzyme; E3, SUMO-protein ligase; ORF, open reading frame; MS, mass spectrometry; GBD, Gal4 DNA-binding domain; X-gal, 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside. called Uba2-Aos1 in yeast, and a SUMO-conjugating enzyme (E2), Ubc9. Both enzymes form transient thiolester bonds with the C terminus of SUMO. Substrate recognition factors (E3s) that stimulate the transfer of SUMO from E2 to substrate have also been identified. SUMO, like ubiquitin, is always synthesized in precursor form, requiring enzymatic removal of a C-terminal peptide. Specialized proteases, called Ubl-specific proteases or Ulps, are responsible for these SUMO processing reactions and for reversing the post-translational attachment of SUMO to proteins. Unlike ubiquitin ligation, many sumoylation sites in proteins match a short consensus sequence, hKXE, where h represents a hydrophobic residue, X is any residue, and lysine (K) is the site of SUMO attachment (5Johnson E.S. Annu. Rev. Biochem. 2004; 73: 355-382Crossref PubMed Scopus (1358) Google Scholar). This consensus can be rationalized by structural data showing the direct binding of the Ubc9 E2 to the conserved elements of this sequence (8Bernier-Villamor V. Sampson D.A. Matunis M.J. Lima C.D. Cell. 2002; 108: 345-356Abstract Full Text Full Text PDF PubMed Scopus (456) Google Scholar, 9Hochstrasser M. Mol. Cell. 2002; 9: 453-454Abstract Full Text Full Text PDF PubMed Scopus (14) Google Scholar). This E2-substrate interaction is unlikely to be sufficient for specific substrate discrimination in the cell, so E3 factors are expected to be important for achieving the requisite in vivo specificity. The SUMO system has been implicated in multiple physiological pathways (3Schwartz D.C. Hochstrasser M. Trends Biochem. Sci. 2003; 28: 321-328Abstract Full Text Full Text PDF PubMed Scopus (316) Google Scholar, 4Seeler J.S. Dejean A. Nat. Rev. Mol. Cell. Biol. 2003; 4: 690-699Crossref PubMed Scopus (569) Google Scholar, 5Johnson E.S. Annu. Rev. Biochem. 2004; 73: 355-382Crossref PubMed Scopus (1358) Google Scholar). SMT3, as well as the genes encoding the SUMO E1, E2, and Ulp1 protease, are all essential for viability in budding yeast, and cells with a conditional allele of either UBC9 or ULP1 are defective in cell cycle progression. The first identified target of SUMO conjugation was vertebrate RanGAP1, a protein required for nucleocytoplasmic trafficking. Modification by SUMO is required for RanGAP1 localization to the nuclear pore complex. Among the many (>50) mammalian targets for SUMO now known are RanBP2, another nuclear pore complex component; PML, a nuclear protein that is altered in certain leukemias; Sp100, an autoantigen in primary biliary cirrhosis; the p53 tumor suppressor; the Sp3 transcription factor; and the NF-κB inhibitor IκB. Even this partial list of in vivo targets makes it clear that the SUMO system must exert a major influence on mammalian growth and regulation. Misregulation of the system is likely to contribute to tumorigenesis, abnormal inflammatory responses, and autoimmune defects. In yeast, a much limited number of substrates has been The first was a of the which are proteins essential for sumoylation has E.S. G. J. Cell Biol. PubMed Scopus Google Scholar, E.S. Cell. Full Text Full Text PDF PubMed Scopus Google Scholar). is also in and this not required for the essential of the contribute to the of the J. A. Mol. Cell. 2002; 9: Full Text Full Text PDF PubMed Scopus Google Scholar). SUMO is also ligated to the DNA protein and sumoylation the DNA repair of G. 2002; PubMed Scopus Google Scholar, 2003; PubMed Scopus Google Scholar). all of these substrate proteins are essential for in is sumoylation required for the protein or proteins that must be for cell in yeast to be identified. proteins involved in and have also been to be in yeast, but the sites of sumoylation not M. V. J. Cell Biol. 2003; PubMed Scopus Google Scholar, D. F. A. Cell. 2004; Full Text Full Text PDF PubMed Scopus Google Scholar). of the of specific SUMO-protein modifications in be in other eukaryotes. However, a much list of proteins in this must first be this we have and data a surprisingly large and set of SUMO system the different of substrate have yielded overlapping but distinct of proteins, the of multiple approaches to the SUMO and and as F. in Scholar). used for of yeast F. in Scholar). The used in this study and and and as in and Scholar). of used are in The for all other the two-hybrid was Hochstrasser M. Cell. Full Text PDF PubMed Scopus Google Scholar). the for of was with and on to for cells that the by The of yeast with genes to the was from in a for the two-hybrid was by a that the open reading J. PubMed Google Scholar). of this that either or C-terminal of by and of the to or by and to was by of sequence encoding first and ligation was by of the a that a site and encoding a by the recognition sequence and a with for by a and a recognition The was with and and ligated to the DNA encoding was by with and and the yeast that been with and and encoding SUMO also by and for by DNA with SUMO was to the sequence encoding the Gal4 DNA-binding This was used as for a two-hybrid of yeast of partial ligated to a sequence encoding the Gal4 J. PubMed Google Scholar). DNA from the was yeast the and on for The three genes and each the of a different that on the and to on to and on to from yeast that and on The the two-hybrid to the two-hybrid and the the and J. PubMed Google Scholar). the for with the SUMO and in a that also ULP1 on a for of of of growth was by of the Hochstrasser M. Mol. Cell. Biol. 2000; PubMed Scopus Google with a yeast on and as a in to as on the The to the DNA was from and to for the of high from identified. in these by DNA and data the four DNA and was used to the genes responsible for the to either a for h or a for The was with in for h and three with for to in with with for The was with and for h with in with binding was SUMO with a yeast SUMO Hochstrasser M. PubMed Scopus Google Scholar). of from yeast cells in of to an of of cells with and in cells to a to in and in of with three a The was for and the was with a from the on The was to of for a and three with of proteins with in and and the was of with in which of binding in for the to a and with of and of The in of of was in the on a for The was and with to all of the protein. of is the from the of and E2 the of The with with for h and four with to by and by and protein with and with The with and with for h with The with to the The by tandem mass spectrometry the protein of and D. Nat. Biotechnol. PubMed Scopus Google as by D. J. A. Mol. Cell 2003; 2: Full Text Full Text PDF PubMed Scopus Google Scholar). a was to a a and to a with by The was the a and to a high and with an tandem mass The was by a cycle as D. J. A. Mol. Cell 2003; 2: Full Text Full Text PDF PubMed Scopus Google Scholar). to The a data of yeast protein from the Saccharomyces a of the on a J. PubMed Scopus Google and for high protein the D. J. A. Mol. Cell 2003; 2: Full Text Full Text PDF PubMed Scopus Google Scholar). The data sets the and J. 2002; PubMed Scopus Google and gene functional categories J. A. J. M. J. M. M. J.E. A. D. D. V. M. J. M. V. 2004; PubMed Google the J. 2002; PubMed Scopus Google Scholar). have a of approaches to proteins in yeast. These different expected to different but related of SUMO-protein ligation and the roles of SUMO in the for protein that can be to other proteins an of in an two-hybrid for SUMO to the C terminus of the Gal4 DNA-binding a that can SUMO not Unlike most a in the two-hybrid can be by ligation of the to a substrate protein to the Gal4 attached to an yeast protein that can with the and with an be it is to from the substrates and proteins that to SUMO or SUMO-protein The proteins be SUMO or enzymes of the SUMO yeast and identified of different proteins that two-hybrid interaction with three each the of a different J. PubMed Google any of the two-hybrid likely to on SUMO ligation, we a which the of of SUMO and be attached to other proteins E.S. G. J. 1997; 16: PubMed Scopus Google Scholar). of the to SUMO ligation the interaction was with in as for binding and in also the of these to Ulp1 isopeptidase most SUMO-protein are in cells this enzyme Hochstrasser M. PubMed Scopus Google Scholar). the also to Ulp1 some SUMO be substrates for Ulp1 in which for the and In all but of the identified proteins that also with SUMO in the of high SUMO for substrates or proteins that Among the identified proteins are proteins involved in RNA metabolism, DNA protein and gene of the proteins, and been linked to the SUMO in studies E.S. Cell. Full Text Full Text PDF PubMed Scopus Google Scholar, A. PubMed Google Scholar). is an involved in protein substrates are not The of is but it is a high copy suppressor of a and have been to to which are the most proteins in yeast. The to the cell and another two-hybrid also to this other proteins, and a a sequence for E3 these proteins be SUMO enzymes as Both and a form of SUMO for of a two-hybrid in the ubiquitin be with and but the of these proteins that with in the two-hybrid the which that sumoylation was in to it is in in vivo and to the use of the two-hybrid to substrates for SUMO we a to in vivo protein M. M.J. V. A. Sci. A. PubMed Scopus Google an was used to from the was by we a major and a which a form of a of SUMO was the the be by the SUMO. a was by and the was in These data that be by either SUMO or to a of the mass SUMO-binding the two-hybrid four proteins and to a form of SUMO and by Ulp1 These factors for proteins that to SUMO or SUMO used the SUMO-binding in these proteins to for potential conserved elements PubMed Scopus Google Scholar). consensus sequence for a conserved was The consensus is by a of or by a of or used A. A. 2002; Google to of the consensus sequence in the yeast Only proteins the four of was in the of a conserved with multiple with which we identified in two-hybrid as a protein that to SUMO but not is to that the interaction and is by sumoylation of the protein The consensus sequence we is related but not to a SUMO-binding from a combination of and with the SUMO-binding protein J. Sci. A. 2004; PubMed Scopus Google Scholar). The consensus sequence in this is related to the hydrophobic in The that to a of SUMO that a The in consensus sequence contribute to the interaction with this SUMO interaction on the was by A. M. D. J. Biol. 2000; Full Text Full Text PDF PubMed Scopus Google Scholar). an of hydrophobic and a C-terminal of which is to the consensus we by A. M. D. J. Biol. 2000; Full Text Full Text PDF PubMed Scopus Google the of as well as the for SUMO The not to to SUMO, on data J. Sci. A. 2004; PubMed Scopus Google but it be important for the on the SUMO extensive binding and studies be to the and sufficient of the by these different of results that two-hybrid can sumoylation substrates and potential factors as well as of the SUMO ligation of is another potential to these be linked to specific physiological pathways by the SUMO with of either Ulp1 or be to the to SUMO from specific the substrate of the protein to the SUMO by a binding of the high copy of the growth of the a yeast four different that growth of cells identified the gene responsible in each as in the was to as of of these genes also the of the to the this proteins that in two-hybrid and in is in vivo. However, the high copy only the first of which the but by with the from the two-hybrid to be important for SUMO of the SUMO a SUMO-protein a form that is or it Ubc9 and the The most and the essential of in yeast as well where it with in the of J. A. Mol. Cell. 2002; 9: Full Text Full Text PDF PubMed Scopus Google Scholar). we identified a as a high copy suppressor of not for not so this is an unlikely of to the of the has been linked to the of so it is that high of in a that the for sumoylation of or In high copy of an SUMO isopeptidase yielded SUMO substrates and a SUMO of these proteins have been linked to by other this as a to proteins of to the of proteins has also been cell by a of proteins and tandem mass of from the the of SUMO we used a in which the Ulp1 isopeptidase gene been in of many ULP1 is but a high copy a form of SUMO is growth is the growth is Hochstrasser M. PubMed Scopus Google Scholar). The of ULP1 and with viability with a encoding either SUMO or has a by a and the recognition sequence for The of the terminus of SUMO, the the of SUMO conjugation with and also but the SUMO viability of an of sumoylation of substrates by a The and modified by a sequence encoding the tandem A. J.S. 2003; PubMed Scopus Google Scholar). with or either or SUMO from a in and for h with in from these cells by the was with the and cells in the ULP1 allele was with and cells showing of proteins identified by two-hybrid with SUMO high copy of and with a in a of this is expected to any SUMO isopeptidase in the and to protein which the of proteins that are in with SUMO-modified with the first on a that the with from the first was an proteins by with protease, which the and SUMO in of the proteins by protein and The the by these that SUMO highly by the of SUMO SUMO we used a highly and on high of protein to tandem mass spectrometry the of from target proteins and of the SUMO the was to in to to and The a data of yeast gene and high by of the of proteins identified by proteins identified by a each These proteins a broad of functional categories and that the proteins identified SUMO-protein we first any of these proteins as The was with a yeast to the used that the SUMO the of proteins from this not any yeast proteins, that all or all of the proteins identified in the SUMO-modified or to a protein. number of the proteins we by identified in other studies three and that of and the of SUMO also a of the proteins by for SUMO by an the in used that the copy of a gene of with a sequence encoding the which a protein A. J.S. 2003; PubMed Scopus Google Scholar). These have ULP1 and genes and not the target proteins. cell was with an a or a with an that the protein protein the primary these for a to an form be in cells an SUMO This can be in for a protein that regulates the and for an and for the of the which is a for RNA of SUMO conjugation to substrates and to conjugation of These data that multiple factors in and which the proteins we are likely to be regulated by an essential DNA-binding protein involved in the form was also but this required of a Hochstrasser M. PubMed Scopus Google Scholar). In a the of proteins identified by yeast two-hybrid with a SUMO high copy suppressor of and of proteins. These three approaches different and yielded different of proteins, but proteins identified by of the the was the most of the approaches likely SUMO substrates identified only by the two-hybrid and was identified in the two-hybrid and the high copy suppressor but not by These results that of of proteins is of specific substrates of the yeast SUMO but also the of not that all SUMO substrates have been identified. proteins to be modified by SUMO in and not in any of are many certain proteins have been the some genes either have been or in a number of in the DNA was in In these some of the cell cycle are expected to be and certain proteins only be in of these M. V. J. Cell Biol. 2003; PubMed Scopus Google Scholar). proteins are only modified by SUMO specific with the of SUMO and was the large of SUMO of from SUMO in each and this limited to substrate to this are being of on in three other mass spectrometry studies have been in which of yeast SUMO-protein was J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google Scholar, J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google Scholar, E.S. J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google Scholar). of the proteins identified in study and in these other is in In all four yeast in we of the proteins by J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google these much was with the other studies and Even J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google cells specific which are expected to a distinct array of proteins, the with modified proteins was high and with cells to and of studies on yeast protein of proteins in J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google E.S. J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google in a is likely that the much proteins identified in study and that of J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google with the other studies the extensive used in the In and that of J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google used the of J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google and E.S. J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google Scholar). in the of by a protein and by was for protein that the proteins highly by E.S. J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google the protein and an that most of the proteins in not In all but the J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google used cell and it is unlikely that proteins that are not but to proteins This is with the in Only three are modified by SUMO E.S. G. J. Cell Biol. PubMed Scopus Google Scholar). However, all with another to form the that form the three studies identified only the three that are whereas J. Biol. 2004; Full Text Full Text PDF PubMed Scopus Google four in on the we that the number of proteins as proteins in study is This not that the physiological of SUMO-modified protein we is on this is the that only a of a protein is to SUMO in vivo. This be a but it a of sumoylation that can be by the of the identified proteins are highly and only a the be by high of a not that the sumoylation is The which are in a cell not this for essential in E.S. G. J. Cell Biol. PubMed Scopus Google Scholar, E.S. Cell. Full Text Full Text PDF PubMed Scopus Google Scholar). and studies with each protein are required to the mechanistic of a SUMO Functional variety of SUMO system substrates have been identified in distinct functional Here we on some is that a number of metabolic enzymes identified. The of these are enzymes. all enzymes that to are in protein with the most The of this is but of the by SUMO is of protein a in is also by the of from the large and from the one. sumoylation protein the or regulates the of a specific of also be modified a point in the cell The protein is J. G. F. M. Finley D. Cell. 2000; Full Text Full Text PDF PubMed Scopus Google and this is also in SUMO-protein SUMO and many SUMO system enzymes are in the with a large number of SUMO-modified proteins in are known to be Both transcription factors to all three of RNA and transcription factors the targets identified as a variety of DNA repair proteins. In we identified many factors that in structural gene and genome multiple proteins in each of these categories that SUMO regulates many and by multiple mechanisms. proteins also of these in the but and are proteins. This that SUMO is not limited to proteins like ubiquitin, be used as a is that proteins and that we identified by as proteins that are modified by SUMO by the two-hybrid study to with SUMO In the pathways regulated by ubiquitin, a number of proteins have a are modified by ubiquitin and This can of proteins that on a interaction is by a ubiquitin to a site on the protein or on another protein. a also be for the in This essential a and is regulated by proteins, as the and proteins J. Cell Biol. 2003; PubMed Scopus Google Scholar, A. 2004; PubMed Scopus Google Scholar). identified and as proteins in also have a SUMO-binding that and proteins be regulated by SUMO in to ubiquitin of protein three other proteins identified as SUMO and have a in This is also by that of and are all modified by SUMO in vivo. the mechanistic of sumoylation to these protein be an important for Schwartz for of the yeast for the use of the and and Schwartz for on the with