Sumoylation as a Signal for Polyubiquitylation and Proteasomal Degradation

The small ubiquitin-related modifier (SUMO) is a versatile cellular tool to modulate a protein's function. SUMO modification is a reversible process analogous to ubiquitylation. The consecutive actions of E1, E2 and E3 enzymes catalyze the attachment of SUMO to target proteins, while deconjugation is promoted by SUMO specific proteases. Contrary to the long-standing assumption that SUMO has no role in proteolytic targeting and rather acts as an antagonist of ubiquitin in some cases, it has recently been discovered that sumoylation itself can function as a secondary signal mediating ubiquitin-dependent degradation by the proteasome. The discovery of a novel family of RING finger ubiquitin ligases bearing SUMO interaction motifs implicated the ubiquitin system in the control of SUMO modified proteins. SUMO modification as a signal for degradation is conserved in eukaryotes and ubiquitin ligases that specifically recognize SUMO-modified proteins have been discovered in species ranging from yeasts to humans. This review summarizes what is known about these ligases and their role in controlling sumoylated proteins.

INTRODUCTION

Posttranslational protein modification is one of the main cellular mechanisms to regulate the fate of a protein. Ubiquitin and SUMO (small ubiquitin-like modifier) are the most prominent members of a conserved family of ubiquitin-like (UBL) posttranslational modifiers (for a recent review, see ref. 1). UBL proteins share an analogous structure termed the ubiquitin fold. In addition, they have a similar conjugating machinery comprised of specific activating and conjugating enzymes, termed E1 and E2, respectively, that catalyse the attachment of an UBL via its C-terminal glycine residue to a lysine residue of the target protein via an amide bond. Substrate specificity is achieved by the activity of additional UBL ligases (E3), which form complexes with specific E2 proteins (Fig. 1).

Figure 1.

Comparison of ubiquitin and SUMO conjugation systems. Ubiquitin (Ub) and SUMO conjugation involves the activities of related enzymes. A) After processing of its precursor forms by deubiquitylating enzymes (Dub), Ub is activated by Ub-activating enzyme (E1) und subsequently conjugated to substrates by complexes of Ub-conjugating enzymes (E2) and Ub ligases (E3). Attachment of Ub to Ub leads to the formation of substrate attached chains. Ub isopeptidases (Dub) can regenerate free Ub from substrates. B) The analogous enzymes as shown in A) are shown for the SUMO system. Here the activating enzyme is composed of two subunits (Uba2 and Aos1) and only a single conjugating enzyme (Ubc9) is used. Desumoylating enzymes are Ulp1 and Ulp2 in budding yeast and several SENPs in mammals.

Invertebrates and yeasts express only one variant of SUMO, while mammals express three conjugatable SUMO variants. In Saccharomyces cerevisiae, SUMO is encoded by a single gene (SMT3, for Suppressor of Mif Two), which was originally isolated as a high-copy suppressor of mutations affecting the centromeric protein Mif2.^2,3 While the budding yeast SMT3 gene is essential for viability, its fission yeast orthologue pmt3 is not.^4,5 Mutants lacking pmt3, however, display severe defects in genome maintenance and are barely viable.⁵ The three mammalian SUMO isoforms (SUMO-1,-2 and -3) differ in sequence and function. SUMO-1 is up to 44% identical to SUMO-2/-3, while SUMO-2 and SUMO-3 are 97% identical. The latter are assumed to carry out largely overlapping functions. While sumoylation in general appears to be essential in mammals, as can be concluded from the lethality of UBC9 ablation,⁶ conjugation of SUMO-2 and SUMO-3 appear to be sufficient to compensate for a loss of SUMO-1 because mice with an inactivated SUMO-1 gene were reported to be viable.⁷ Unlike SUMO-1, SUMO-2/-3 supply the main reservoir of free SUMO for conjugation in response to certain stress stimuli.^8-10

Figure 2.

Yeast and mammalian SUMO orthologues and paralogues. A) Shown is a schematic representation of the single essential budding yeast SUMO (Smt3) and the three conjugatable SUMO isoforms found in mammals (SUMO-1, SUMO-2 and SUMO-3). The latter are nearly identical differing only in 3 residues. The positions of Lys (K) residues are indicated. Sumoylation consensus sites (Ψ KxD/E, in which Ψ is a hydrophobic residue and x a variable residue) that are involved in the formation of SUMO chains are marked with asterisks. B) Depicted are the different types of SUMO modification referred to in the main text.

SUMO can be attached either to a single or to multiple lysine (Lys) residues within a target protein (mono- or multisumoylation), which are often part of a ΨKxD/E consensus motif that is directly recognized by the SUMO conjugation enzyme Ubc9 (Fig. 2).^11,12 Similar to ubiquitin, SUMO is also attached to Lys residues within SUMO itself, which leads to the formation of SUMO chains (polysumoylation).^13,14 SUMO chain formation mainly involves Lys residues within the above-mentioned consensus motifs, three of which (K11, K15 and K19) are present in an N-terminal domain of S. cerevisiae SUMO, while only one (K11) is found in SUMO-2 and SUMO-3 of mammals (Fig. 2).^11,13,15 SUMO-1 in contrast, lacks a Lys residue in such a consensus motif, which appears to be the reason why it does not form chains efficiently.¹³ SUMO-1, however, may be attached to Lys residues within SUMO-2/3 chains thereby preventing the elongation of such chains.¹⁶

The SUMO pathway has important functions in the regulation of a large variety of cellular processes such as progression through the cell cycle, transcriptional regulation, DNA repair, stress responses, cancer and aging.^17-27 The molecular mechanisms underlying the role of SUMO modification in these processes is still incomplete. Generally, with the exception of RanGAP1, only a small fraction of a given protein pool is sumoylated at a certain point in time.¹⁹ Supported by exemplary cases, it is assumed that SUMO modification alters protein function by affecting protein-protein interactions or subcellular localization. Until recently, SUMO and ubiquitin were thought to have opposing functions with sumoylation saving proteins from degradation by occupying the same lysine residue that is required for their ubiquitylation. The example leading to this concept was the modification of IκBα, an inhibitor of the transcriptional activator NF-κB.²⁸ Upon activation of the inflammatory response pathway, IκBα is ubiquitylated and degraded by the proteasome releasing active NF-κB.297 IκBα, however, can be sumoylated on the same lysine residue preventing its ubiquitylation and leading to a stabilization of the protein.^28,30 Based on this finding it has been suggested that SUMO modification of IκBα serves to inhibit the induction of NF-κB-dependent transcription.²⁸

Another example of SUMO-ubiquitin crosstalk was revealed by studies on postreplicative DNA repair and its key player PCNA (proliferating cell nuclear antigen). Also for PCNA, sumoylation and ubiquitylation target the same lysine residue, but in this case SUMO modification blocking ubiquitylation apparently does not only prevent ubiquitylation but in addition directs PCNA to a distinct interaction. PCNA is monoubiquitylated by Rad6 and Rad18 at the conserved lysine residue 164 in response to DNA damage, while the assembly of a Lys63-linked ubiquitin chain is catalyzed by a different subset of enzymes.^31,32 Monoubiquitylation triggers error-prone bypass replication, whereas polyubiquitylation results in an error-free bypass mode.³¹ SUMO conjugation to Lys164 seems to inhibit DNA repair when the RAD6 pathway is not functional.^33,34 Sumoylated yeast PCNA recruits the helicase Srs2, which prevents Rad51-dependent recombination by disrupting nucleoprotein filaments required for recombination and thereby prevents unwanted sister chromatid recombination during replication.³⁵ In this case, SUMO and ubiquitin can operate on the same lysine residue as a switch between different functional forms of a given protein.

The two examples described above gave only a glimpse to the range of functional interactions of these two Ubl modifier systems. Recent discoveries indicated that interconnections between SUMO and the ubiquitin/proteasome system (UPS) are far more widespread than anticipated. Contrary to previous assumptions, these studies have identified sumoylation as a targeting signal for ubiquitylation and ubiquitin-dependent degradation.^33,36-40 Our current understanding of these processes and their putative functions will be discussed in this review.

FUNCTIONAL INTERACTIONS OF SUMOYLATED PROTEINS AND THEIR REGULATION

Sumoylation of proteins alters their biochemical properties in a way that may either promote a specific function, for example by enabling or enhancing an interaction with another protein, or inhibit a function by preventing interactions. Examples, in which sumoylation was shown to promote specific protein interactions, are the binding of Ran-GAP-SUMO to RanBP2,^41-44 the binding of PCNA-SUMO to Srs2^34,35 and the interactions of PML-SUMO proteins with each other as well as with other proteins, which are thought to promote the formation of PML nuclear bodies.^45,46 The interaction of sumoylated proteins with their partners involves specific SUMO interaction motifs (SIMs) in the latter proteins (Figs. 3 and 4, see next section).⁴⁷ The identification of ubiquitin ligases that bind sumoylated proteins via such motifs led to the discovery of a novel targeting pathway, to which this review is devoted. In other cases, such as thymine-DNA glycosylase, sumoylation terminates a function or interaction of the modified protein.^48,49

Figure 3.

SUMO interaction motifs. A) Shown is the ribbon diagram of the SIMb from PIASx bound to human SUMO-1 (based on structure PDB: 2ASQ). B) Residue conservation of the three SIM types is shown in a sequence logo representation. Overall height of a position indicates its information content, height of individual residues indicates their frequency at that position. The residues are color coded (black: charged, green: polar, blue: hydrophobic). A color version of this image is available at www.landesbioscience.com/curie.

Figure 4.

Ubiquitin ligases recognizing SUMO (ULS). Shown are schematic representations of recently discovered SUMO binding ubiquitin ligases. In budding yeast two such enzymes, Uls1 and the heterodimeric Slx5-Slx8, were identified. In fission yeast, complexes formed by Slx8 and either Rfp1 or Rfp2 carry out similar function as Slx5-Slx8 in budding yeast. The human RNF4 protein can functionally replace the Slx8-based ligases in fission and budding yeasts. The positions and sequences of two different types of SIMs in these ligases and the RING motifs are depicted.

Considering the fact that of a given protein only a small fraction is commonly found in the sumoylated state and that this modification is transient, it is an important question how the modified state of the protein is regulated in the cell.¹⁹ The most direct way involves the activity of SUMO isopeptidases, which revert the substrate into its unmodified state (for recent reviews see refs. 50,51). While two SUMO isopeptidases, Ulp1 and Ulp2, were found in S. cerevisiae,^52-54 human cells express at least six such enzymes (SENP1,2,3,5,6 and 7).^50,51,55 Distinct specificities have been assigned to these enzymes. Ulp2 in yeast and SENP6 (also known as SUSP1) or SENP7 in mammals are efficient in the disassembly of SUMO chains (Fig. 1).^14,56 Aside from reversion of SUMO conjugation, another way to regulate the function of a sumoylated protein is to add further modifications. Conjugation of additional SUMO moieties to other Lys residues in a substrate (multisumoylation) or to an already conjugated SUMO (polysumoylation) may determine an altered fate of the so modified protein. As discussed in more detail below, recent evidence indicates that polysumoylation serves as a preferred recognition signal for specific ubiquitin ligases, which may ultimately target a SUMO substrate for ubiquitin-dependent degradation by the proteasome.

SUMO INTERACTION MOTIFS (SIM)

Although ubiquitin and SUMO share considerable similarity in both sequence and structure, the nature and binding mode of their interaction partners are fundamentally different. Ubiquitin can be bound by more than 10 different classes of autonomously folded recognition domains; the interaction typically involves a hydrophic surface region surrounding the critical Ile44 residue in ubiquitin.⁵⁷ By contrast, SUMO does not seem to employ folded interaction domains, but seems to be exclusively recognized by short conserved motifs called SIMs (for SUMO interaction motifs). The first motif to specifically recognize SUMO was described in 2000 by Minty et al.⁵⁸ A consensus motif, consisting of four hydrophobic residues followed by a serine-rich spacer and a group of acidic residues, was identified in human SAE2, PML and the PIAS family. Residues of this motif—particularly those within the hydrophobic part—were shown to be important for SUMO recognition. Since then, many more SUMO binding proteins in mammals and yeast have been identified and the definition of the SIM consensus has undergone several rounds of refinement.^36,44,59-61 At present, most experimentally proven SUMO-binding motifs can be classified by their residue conservation into three major SIM types (Fig. 3B).³⁶

• SIMa: Motifs belonging to the SIMa type are characterized by four consecutive hydrophobic residues, immediately followed by a mixed cluster of Ser/Asp/Glu residues. When analyzing the sequence conservation of SIMa motifs, a certain variability at the third hydrophobic position becomes apparent. This position is not only less conserved than the other hydrophobic positions, it can even accommodate nonhydrophobic residues.
• SIMr: The second SIM type resembles the SIMa, but has a reversed orientation. In the SIMr motif, the four hydrophobic positions are preceded by an acidic cluster. In keeping with the inverse theme, the second hydrophobic position of SIMr—which corresponds to the third hydrophobic position of SIMa—is the most variable one and can occasionally accommodate nonhydrophobic residues.
• SIMb: The third SIM type is shorter than SIMa, but generally better conserved and easier to recognize. Most type b SIMs strictly adhere to the consensus sequence V-I-D-L-T, with some variability in the first two hydrophobic positions. The third position has a strong preference for Asp, unlike the usually hydrophobic residue found at the corresponding position of SIMa. Several SIMb motifs, including those of the PIAS family, are followed by a serine/acidic region resembling that of SIMa. However, this stretch is not crucial for SIMb function, as there are several documented instances of non-acidic SIMb motifs, including those of the RNF4 family.

While most established SIMs follow one of the three consensus motifs, there is also evidence that the three SIM classes form a continuum. The SIMb, when followed by an acidic stretch, is not much different from a SIMa, with the third (variable) hydrophobic position filled by an Asp residue. In addition, there are putative SIMs consisting of four hydrophobic residues flanked by acidic residues on both sides; thus precluding a classification as SIMa or SIMr. It is at present not clear if the SIMs class has a major influence on the recognition properties of the SIM. Mammals and several other taxa have multiple SUMO versions, often with differing properties and expression patterns. It has been proposed that in human SIMs the acidic stretch is important mainly for SUMO-1 binding, while SUMO-2 binds equally well to SIMs with and without acidic regions.⁶¹ However, this proposal is mainly based on the highly atypical and not conserved SIM of the human TTRAP protein; its generalization clearly requires further investigation. Interestingly, there are several examples where orthologous proteins from closely related species use different SIM types at a corresponding position, which supports the idea that different SIMs might be functionally interchangeable.

The recognition mode of SUMO by SIMs has been elucidated by a series of structural studies.^59,61-63 Three high-resolution structures of SIMs in contact with SUMO are currently available: the SIMb of PIASx bound to SUMO-1 (PDB: 2ASQ),⁶³ the SIMb of MCAF1 bound to SUMO-3 (PDB: 2RPQ)⁶³ and a (partial) SIMr of RANBP2 bound to SUMO-1 (PDB:1Z5S).⁶³ As shown in Figure 3A, the hydrophobic residues of the PIASx-derived SIMb form a short β strand that interacts with SUMO at a region formed by its β sheet 2 and the α helix, which is remote from the Ile44 patch that is used by ubiquitin for its interactions. The SIMb of MCAF1 is closely related to that of PIASx; both consist of the archetypical V-I-D-L-T motif, followed by a serine/acidic stretch. The binding mode of the two SIMb examples is also similar, with the highly conserved Asp3 of the SIMb core contacting a lysine residue conserved in all human SUMO paralogs. The orientation of the conserved Thr5 at the C-terminal boundary of the SIMb differs between the two structures; the same is true for the flanking serine/acidic stretch, which does not seem to participate in any crucial SUMO contact. The structure of the reversed SIM of RANBP2 demonstrates that the SIMr indeed recognizes SUMO in a reversed orientation as compared to the SIMb structures. As the acidic residues of the SIMr are not included in the structure, their contribution to SUMO recognition by the SIMr motif cannot be judged.

Recently, another layer of complexity has been added to the regulation of SUMO/SIM interaction. As mentioned above and shown in Figure 3, several SIMs of all types contain serine residues at positions that are occupied by aspartate or glutamate in other SIM instances. Already in 2006, it had been noticed that SIM-associated serines can be phosphorylated,⁶¹ adding a negative charge and making them similar to an acidic amino acid. More recently, it was shown that casein kinase 2 (CK2) phosphorylates serine residues in the SIMs of PIAS1, PML and PMSCL1 and that this phosphorylation is required for efficient recognition of both human SUMO-1 and SUMO-2.⁶⁴ The enhanced SUMO binding was shown to depend on a conserved lysine residue (Lys39 in SUMO-1, Lys35 in SUMO-2). Most likely, the phospho-serine residues form a salt bridge with the SUMO lysine. In the NMR structure of the PIASx SIMb, which is closely related to that of PIAS1, these contacts are not visible—probably due to the fact that the PIASx SIMb has been studied in the nonphosphorylated form.⁵⁹ The full extent of the crosstalk between kinase signaling and SUMO/SIM recognition remains to be explored—serine residues at susceptible positions are rather common in different SIM subtypes.

PROTEASOMAL DEGRADATION OF SUMO CONJUGATES

The sumoylation state of a given protein is a dynamic equilibrium that is regulated mainly by the conjugating and deconjugating activities of the SUMO pathway. Desumoylation is also required to counteract the formation of SUMO chains. Genetic analysis revealed that Ulp2, one of the two desumoylating enzymes in S. cerevisiae, is mainly responsible for this activity in this organism.¹⁴ Mutants lacking Ulp2 are viable but have severe growth defects and are sensitive to various stresses.^53,54 Accumulation of polysumoylated proteins appears to be responsible for these defects.¹⁴ These high molecular weight SUMO conjugates (HMW-SC) are not detectable in yeast strains expressing a mutant SUMO variant (Smt3-R11,15,19) lacking lysine residues critical for SUMO chain formation (Fig. 2A). Importantly, this variant of SUMO is able to suppress the defects of the ulp2Δ mutant.¹⁴

HMW-SC occur not only in the ulp2Δ mutant but also in mutants deficient in the UPS such as ubc4Δ ubc5Δ, or upon inhibition of the proteasome.³⁶ The UPS apparently contributes to the removal of certain types of sumoylated proteins. Experiments with the Smt3-R11,15,19 mutant indicated that chain formation promotes targeting of sumoylated proteins in the UPS. Consistent with this notion, conjugates of the chain forming SUMO-2/3 isoforms accumulated in human cells to much higher levels upon inhibition of the proteasome than those of SUMO-1, which does not form chains efficiently.^13,36 The preference for a proteolytic targeting of polysumoylated proteins can be related to the properties of SUMO-specific ubiquitin ligases (see below).

UBIQUITIN LIGASES RECOGNIZING SUMOYLATED PROTEINS IN S. CEREVISIAE

In the UPS, recognition and targeting of substrates for proteasomal degradation is mediated by a class of enzymes known as ubiquitin protein ligases, E3s or recognins, which form complexes with ubiquitin-conjugating enzymes (E2s).^65-67 The largest group of ubiquitin ligases is characterized by a structural domain termed the RING finger, in which two zinc atoms are complexed by Cys and His residues.⁶⁸

In S. cerevisiae, a new subgroup of RING finger ubiquitin ligases was discovered recently. One of their characteristic properties is that they contain multiple SIMs for binding to SUMO (Fig. 4). Therefore they were called SUMO-targeted ubiquitin ligases (StUbL) or ubiquitin ligases for SUMO conjugates (ULS) (Fig. 3A).^36-38,69 We and others identified the yeast proteins Slx5 (alias Hex3) and Uls1 (alias Ris1) as SUMO interactors in yeast two-hybrid screens.^36,60,70,71 Both proteins contain multiple putative SIMs (Fig. 3B and 4).^36,60 Slx5 contains two SIMa and one SIMb sequences, whereas Uls1 contains two SIMa and two SIMb sequences. Deletion of the SLX5 or ULS1 gene in yeast leads to the accumulation of HMW-SC.^36,37,72 Deletion of both genes results in an even stronger effect on SUMO conjugates pattern.³⁶

PROPERTIES AND FUNCTIONS OF slx5-slx8

SLX5 and SLX8 were identified in a genetic screen that selected for genes required in the absence of the only RecQ type DNA helicase Sgs1 in budding yeast.⁷³ Mutations in WRN and BLM, two human homologues of Sgs1, cause premature aging (Werner's syndrome) or genome instability often leading to cancer (Bloom's syndrome). Sgs1 is involved in various processes including replication and the S-phase checkpoint, double-strand break repair, recombination and telomere maintenance (for a review see ref. 74). Mutations in SLX5 and SLX8 were also found to cause a synthetic growth defect with mutations in the SRS2 gene, which encodes another DNA helicase.^75,76 As mentioned earlier, Srs2 is recruited to replication forks by sumoylated PCNA and prevents unscheduled recombination (for review see refs. 26,77,78). SRS2 mutations were also found to be synthetically lethal in combination with mutations affecting the SUMO-deconjugating enzyme Ulp1.⁷⁹ Ulp1 mutations on the other hand were suppressed by overexpression of SLX5.³⁷ Interestingly, sgs1, srs2 and slx5 mutants all accumulate HMW-SC to a similar extent.³⁸ Together these observation are consistent with the idea that mutations affecting the two helicases Sgs1 and Srs2 accumulate sumoylated proteins and that Ulp1 and Slx5-Slx8 are required to limit the abundance of such conjugates either by desumoylation or by proteolytic targeting (Fig. 5).^37,38 If uncontrolled, such conjugates apparently lead to an arrest in cell division. Interestingly, the activity of the Ulp2 protein, which cleaves SUMO chains was shown to be required to recover from checkpoint arrests that were for example induced with hydroxyurea.⁵¹ These observations together pointed to the fact that SUMO modification and its subsequent processing by Ulps, or its recognition by Slx5 and Slx8 are linked to processes of DNA repair, recombination and replication fork stability. In agreement with this notion, slx5 and slx8 mutants, similar to ulp2, are hypersensitive to DNA damage, or the replication inhibitor hydroxyurea, they accumulate gross chromosomal rearrangements (GCRs) as well as shortened telomeres and display rDNA hyperrecombination.^80-86 Consistent with an involvement in these processes, Slx5 or Slx8 have been localized to replication centers and damage-induced Rad52 nuclear foci.^82,87 Another study reported the recruitment of DNA double strand breaks to the nuclear pore complex in a Slx5-Slx8-dependent manner.⁸⁸

Figure 5.

Role of SUMO-targeted ubiquitin ligases in genome stability. Processing of stalled replication forks and double strand breaks as well as homologous recombination repair (HRR) and DNA damage repair require the action of the ULS enzymes Slx5-Slx8 in budding yeast and Rfp1,2-Slx8 in fission yeast. These types of genomic instability coincide with the appearance of increased amounts of SUMO conjugates, which can lead to an arrest in the cell division cycle if they are not sufficiently controlled by Slx5-Slx8-mediated ubiquitylation and subsequent proteasomal degradation or by desumoylation.

Figure 6.

Ubiquitin-dependent proteolytic control of SUMO conjugates. A) Shown is a schematic model of the control of the sumoylated state of substrate proteins in budding yeast. Sumoylation on one side is controlled by the activity of SUMO ligases such as Siz1 and Siz2 and on the other side by the desumoylating enzymes Ulp1 and Ulp2, which have different activities towards SUMO chains. In addition, sumoylation in particular polysumoylation may trigger ubiquitylation by SUMO recognizing Ub ligases (ULS) and subsequent degradation by the proteasome. Ub may be attached either to Lys (K) residues of the substrate or in the SUMO chain. B) Model of ATO-induced SUMO-dependent proteolytic targeting of PML. ATO (arsenic trioxide) treatment of cells or APL patients results in an increased sumoylation of PML, which mediates its subsequent ubiquitylation by RNF4 and in turn its degradation by the proteosome. The molecular mechanism of ATO-induced PML sumoylation is still unclear (indicated by question marks). It may either involve stimulation of PML phosphorylation by ERK2 or CK2, or changes in the regulation of enzymes of the SUMO system.

Slx5 and Slx8 are both RING finger proteins that form a heterodimer.^72,89 Aside from interaction between Slx5 and SUMO,^36-38 Slx8 by itself was shown to bind to SUMO in vitro, indicating that the Slx5-Slx8 heterodimer has multiple SUMO binding sites distributed between the two polypeptides (Fig. 4).³⁸ Genetic and biochemical data support a view, in which Slx5-Slx8 has a role as a SUMO-dependent ubiquitin ligase that acts together with the redundant E2s Ubc4 or Ubc5 to target sumoylated proteins for degradation by the proteasome (Fig. 6A).^36-38 Consistent with this notion, in vitro experiments revealed that Slx5-Slx8 has ubiquitin ligase activity.^36-38,69 Slx8 and Slx5-Slx8 complexes, in contrast to Slx5 alone, were shown to display autoubiquitylation activity in vitro.^36,37 Based on these and additional data, it was concluded that the core ubiquitylation activity is inherent to Slx8 and that Slx5 enhances this activity by a function relying on its RING domain and by binding to conjugated SUMO on a target protein.³⁷ The latter conclusion was derived from in vitro experiments that used Rad52 or a Rad52-SUMO fusion as substrates. In these assays, the Rad52-SUMO fusion protein was preferentially ubiquitylated.³⁷

Binding experiments that used immobilized Slx5-Slx8 produced in E. coli revealed that this ligase preferentially binds to high molecular weight SUMO conjugates suggesting that poly- or multisumoylated proteins might be the preferred substrates of this ligase.³⁶ Supporting this notion, another study provided convincing in vitro evidence, using autosumoylated forms of the yeast SUMO ligase Siz2 as substrates, that poly-SUMO chains act as a preferred targeting signal for Slx5-Slx8-dependent ubiquitylation and suggested that ubiquitin modification occurs mainly on the terminal SUMO moiety.³⁸

A recent study identified the transcriptional regulator Mot1 as a first in vivo substrate of Slx5-Slx8 in yeast.⁶⁹ In an earlier study by the same group, mutations in the SLX5 and SLX8 genes were found to suppress a mutation in the MOT1 gene.⁷² Interestingly, the same allele was also suppressed by mutations in the genes encoding components of the SUMO conjugation system. These authors propose a role for Slx5-Slx8-mediated SUMO-targeted ubiquitylation in the quality control of misfolded nuclear proteins. The increased turnover rates detected for the Mot1 mutant depended on SUMO modification and Slx5-Slx8. In additon, the turnover of wild-type Mot1 was stimulated by the same components when cells were treated with the amino acid analogue canavanine.⁶⁹

What remains puzzling is the observation that combinations of mutations that interfere with SUMO conjugation (uba2 or ubc9), or with the formation of SUMO chains (smt3-R11,15,19), with slx5 or slx8 mutations cause either synthetic lethality or sickness.^36,72 Intuitively, one would have expected the opposite, that the inability to turnover sumoylated substrates might be suppressed by a reduction in their formation. Indeed, a mutation of the S. pombe SUMO ligase Pli1 was found to suppress phenotypic effects of mutations in the Rfp1,2-Slx8 ligase, which is related to Slx5-Slx8 as discussed in a later section. A similar interrelation was also observed for mutations in the ULP2 gene leading to a loss of a SUMO deconjugating enzyme, which were suppressed by mutations impairing the SUMO-activating enzyme or the SUMO precursor processing protease Ulp1.^53,54 The unsuspected synthetic effects of mutants affecting Slx5-Slx8 and SUMO conjugation could be explained by putative SUMO-independent functions of Slx5-Slx8, which might become essential when sumoylation is compromised. Another possibility is that low levels of sumoylation could be sufficient to direct certain essential substrates towards Slx5-Slx8, but at the same time might be insufficient to terminate the function of the same substrates more directly by generating multi- or polysumoylated forms of them. The latter explanation, however, is somewhat at odds with the observed binding preference of Slx5-Slx8 for HMW-SCs. Additional studies are required to resolve this issue.

PROPERTIES AND FUNCTIONS OF Uls1/Ris1

Compared to Slx5-Slx8, even less is known about the specific functions of the other SUMO-binding ubiquitin ligase Uls1/Ris1 in S. cerevisae. Uls1 is a member of the SWI/SNF family of DNA-dependent ATPases, for which a role in antagonizing silencing during mating-type switching was reported.⁹⁰ The presence of a RING domain as well as of multiple putative SIMs (Fig. 4) and the fact that Uls1 interacts with SUMO and Ubc4 suggested a role for Uls1 in SUMO-dependent ubiquitylation.³⁶ In support of this hypothesis, the deletion of both SLX5 and ULS1 yielded a synthetic effect both on yeast growth and on the accumulation of HMW-SC.³⁶ The recently identified interaction between Uls1 and sumoylated Ebp2, a yeast protein related to Epstein-Barr virus nuclear antigen 1-binding protein 2, is also consistent with a significance of the SUMO binding properties of Uls1.⁹¹ Apparent changes in the overall Ebp2 levels as a consequence of ULS1 deletion, however, were not observed in this study suggesting that Uls1 may not control the stability of Ebp2. Polysumoylated forms of Ebp2, which are likely to be the preferred substrates of Uls1, however, may be too low in abundance and too heterogeneous to detect stabilizing effects of a uls1 mutation. While genetic evidence strongly suggests that Uls1 acts as a SUMO-dependent ubiquitin ligase with functions partially overlapping with those of Slx5-Slx8, biochemical prove of this activity is still missing.

SUMO TARGETED UBIQUITIN LIGASE rfp1,2-slx8 IN SCHIZOSACCHAROMYCES, POMBE

Ubiquitin ligases for SUMO modified proteins were also discovered in fission yeast.^33,39 The functionally redundant proteins Rfp1 and Rfp2 are RING finger proteins with multiple SUMO interacting motifs. By themselves, these polypeptides lack ligase activity, but upon interaction with Slx8 they form an enzyme that ubiquitylates Rad60, a protein that resembles a fusion of two SUMO domains,^92,93 or GST-SUMO in vitro.^33,39 Similar to the slx5Δ uls1Δ mutant in S. cerevisae, the fission yeast double mutant rfp1Δ rfp2Δ accumulates high molecular weight SUMO conjugates. Cells lacking Rfp1,2-Slx8 display genomic instability and hypersensitivity to genotoxic stress. Much in contrast to what was discussed above for Slx5-Slx8 in budding yeast, these defects can be suppressed by the deletion of the major fisson yeast SUMO ligase Pli1.³⁹ These data indicate that Rfp1,2-Slx8 is required to remove sumoylated targets that are generated as a result of DNA damage or replication fork arrests (for a recent review, see ref. 86). Together these studies showed that Rfp1,2-Slx8 complexes are SUMO-targeted ubiquitin ligases that are functionally related to the Slx5-Slx8 complex of S. cerevisiae.

MAMMALIAN SUMO TARGETED UBIQUITIN LIGASE RNF4

No apparent human homolog of the S. cerevisiae ULS proteins Slx5-Slx8 or Uls1 could be found in the human genome, but the S. pombe proteins Rfp1 or Rfp2, which form redundant complexes with Slx8 in fission yeast, show weak homology to the human protein RNF4 (alias SNURF).³⁹ RNF4 is a nuclear RING finger protein with ubiquitin ligase activity.⁹⁴ RNF4 can functionally complement the rfp1Δ rfp2Δ or slx8Δ mutations in S. pombe and slx5Δ or slx8Δ in S. cerevisiae suggesting that RNF4 alone provides a similar activity as the Rfp1,2-Slx8 or Slx5-Slx8 heterodimeric enzymes in these yeasts.^33,36,39,95

RNF4 contains four putative SIMs, which are located closely spaced near its N terminus (3 of which are shown in Fig. 4).⁸⁶ In vitro studies showed that binding of RNF4 to polySUMO chains is mainly dependent on SIM2 and 3, which both fit into the 'type b' class, with some contribution from SIM4, which is of 'type a'. Complete loss of SUMO binding was observed when all four SIMs were mutated.⁹⁶ Even though a contribution of SIM1 remained uncertain, these data demonstrated that multiple SIMs are required for efficient binding of RNF4 to SUMO-2 chains. This study also revealed, using a mixture of SUMO-2 conjugates, that for chains comprising 3 or more SUMO-2 moieties increasing chain length correlated with efficiency of RNF4 binding, while mono-SUMO or di-SUMO binding was undetectable under these conditions.⁹⁶ SUMO-2 chains were efficiently polyubiquitylated by RNF4 in vitro, while mono-SUMO was a poor substrate. Together these data characterized RNF4 as a mammalian poly-SUMO-dependent ubiquitin ligase that is functionally related to yeast SUMO-targeted ubiquitin ligases.^33,36,39,96

REGULATION OF PML BY RNF4-MEDIATED SUMO-DEPENDENT PROTEOLYSIS

A hint to a physiological substrate of RNF4 was provided by the observation that it interacted with the promyelocytic leukemia protein (PML) in a SUMO-dependent manner.⁹⁷ PML and its sumoylation are essential for the formation of PML nuclear bodies (PML-NBs),^98-100 large protein complexes detected in mammalian cell nuclei.¹⁰¹ Many other proteins such as Daxx, Sp100 and RNF4 are recruited to these dynamic PML-NBs.^97,101 Complex formation between these proteins is provided by SUMO/SIM interactions, which contribute to the interactions of sumoylated and unmodified PML-NBs proteins and thereby to the formation of these subnuclear structures.^45,46 PML-NBs have been implicated in the regulation of many important cellular functions including, transcription, apoptosis, DNA repair, antiviral defence, tumour suppression, senescence, proteolysis and storage of nuclear proteins.¹⁰¹

PML is also known as TRIM19 because it belongs to the tripartite motif family of proteins, which are characterized by a RING domain followed by two B-boxes and a coiled-coil region (for review see ref. 102). Eleven PML isoforms could be isolated from human cells that all differ at their C termini due to alternative splicing. Most isoforms are nuclear proteins; only two of them lack the NLS and are therefore cytoplasmic.^101,102 PML has three sumoylation sites K65, K160 and K490 (or K442 depending on the splice variant) and a single SIM of the 'a type' downstream of K490/442.¹⁰² These properties are critical for the role of PML as a scaffold protein of PML nuclear bodies. PML was first identified in patients suffering from acute promyelocytic leukemia (APL). This leukemia is caused by a chromosomal translocation (t(15;17)) in myelocytic progenitor cells. This translocation results in a fusion of the N-terminal part of PML to the C terminus of retinoic receptor alpha (PML-RARα).^103,104 As a consequence, these progenitor cells proliferate without proper differentiation. This leukemia is treated with all-trans retinoic acid (ATRA) or arsenic trioxide (ATO), both of which induce terminal differentiation of these cells. High doses of ATO, in addition, induce apoptosis of APL cells (for a recent review see ref. 105).

A milestone finding towards an understanding of the therapeutic effect of ATO was that it induces sumoylation of PML-RARα and its subsequent degradation by the proteasome.¹⁰⁶ The discovery of SUMO-dependent ubiquitin ligases described above together with the observation that PML and RNF4 interact in a SUMO-dependent manner suggested that the sumoylation of PML may target it for ubiquitylation by RNF4 (Fig. 6B).^33,36,39,97 This conjecture has been proven recently identifying PML as a first bona fide substrate of SUMO-dependent ubiquitylation.^10,96,107 Knockdown of RNF4, mutation of K160 of PML to arginine, or simultaneous knockdown of SUMO-1,-2 and -3 isoforms all impaired ATO-induced degradation of PML indicating that attachment of SUMO to K160 is essential for this proteolytic targeting mechanism.^96,107 Importantly, SUMO-ubiquitin hybrid conjugates of PML could be detected both in vivo, upon ATO treatment, and in vitro.^10,96 Mass spectrometric analyses of such conjugates indicated that ubiquitin is either conjugated to lysine residues of SUMO, including K11, or to various lysine residues in the C-terminal part of PML. Importantly, K160 was not found to be ubiquitylated in these experiments.^96,107 Consistent with an earlier observation that SUMO-2/3 conjugates are accumulating dramatically upon inhibition of the proteasome,³⁶ RNF4-mediated ubiquitylation and subsequent degradation of PML was promoted most effectively by SUMO-2/3 modification.^10,96,107 The situation, however, may not be quite that simple because SUMO-1 modified forms of PML were also ubiquitylated by RNF4 in a SUMO-dependent manner in vitro and a knockdown of SUMO-2/3 was insufficient to prevent ATO-induced degradation of PML.^10,107 In addition, it was suggested that modification of K65, either by SUMO-1 or SUMO-2/3, influences sumoylation at K160 indicating that there may be a crosstalk between modification of individual attachment sites by the different SUMO isoforms.¹⁰⁷

POSSIBLE TARGETS OF ATO LEADING TO DEGRADATION OF PML

Stress induction appears to be a key trigger for increased SUMO-2/3 modification. SUMO-2/3 modification as well as subsequent ubiquitylation can be observed after many stresses including heat, osmotic or oxidative stress (H₂O₂), or treatment with ethanol or ATO. ^8,10 The molecular targets relevant for these effects are still largely unknown. Among the possibilities is that compounds such as H₂O₂ or ATO cause an inhibition of certain cystein-containing enzymes mediating conjugation or deconjugation of SUMO isoforms or ubiquitin.^55,108 Inhibition of SUMO chain depolymerising isopeptidase activity could for example explain the observed increased levels of high molecular weight SUMO-2/3 conjugates following ATO or H²O² treatment. Additional signals that specifically induce SUMO-dependent ubiquitylation, however, must be involved in the effect of ATO on PML, because up-regulation of SUMO-2/3 modification alone was not sufficient to induce PML degradation.¹⁰ In line with this conclusion is the observation that down-regulation of the SUMO-chain specific protease SENP6 does not lead to smaller PML-NBs but instead to larger ones.⁵⁶

Another possibility is that ATO-induced phosphorylation of the PML protein through a mitogen-activated protein (MAP) kinase pathway activates the RNF4-dependent pathway (Fig. 6B), as it was shown to cause increased PML sumoylation and PML-mediated apoptosis.¹⁰⁹ These phosphorylations occur near the N terminus and also between the third sumoylation site and the SIM in PML. Other serine residues of PML, which are part of its SIM, were found to be phosphorylated by CK2 after osmotic stress.¹¹⁰ This study suggested that phosphorylation of the SIM in PML is necessary for stress induced degradation. The same phosphorylation was recently shown to enhance its interaction of PML with SUMO.⁶⁴ Another study, however, showed that, while this SIM is important for ATO-induced degradation of PML, its phosphorylation is not.¹¹¹ A possible way to explain the need for the SIM in RNF4-mediated degradation could be that it promotes autosumoylation of PML by recruiting SUMO-2/3-loaded Ubc9. A similar role of SIMs in promoting autosumoylation was observed for other proteins including human Daxx (death domain associated) protein.^46,112-114 This mechanism is reminiscent of an analogous mechanism for noncovalent ubiquitin/substrate interactions leading to monoubiquitylation.¹¹⁵

In conclusion, while the identification of SUMO-dependent ubiquitin ligases has significantly increased our understanding of ATO-induced degradation of PML, more studies are required to fully understand its mechanism in detail. ATO targeting of PML is of considerable interest also for the treatment of cancer types other than APL. Recent results have indicated that ATO treatment prior to conventional chemotherapy contributes to the activation and subsequent eradication of quiescent cancer initiating cells in chronic myeloid leukaemia.¹¹⁶ Other important questions concern the physiological functions of the RNF4 pathway and its other substrates and how their targeting is regulated for example by cellular stress response pathways. RNF4 interacts with a variety of transcription factors that are known to be both sumoylated and ubiquitylated suggesting that it is involved in their regulation (for a recent review see ref. 86).

SUMO-DEPENDENT REGULATION OF HIF1α

SUMO-dependent proteolytic targeting was shown to contribute also to the regulation of hypoxia-inducible factor 1α (HIF1α) under hypoxic conditions.⁴⁰ Inactivation of the desumoylating enzyme SENP1 led to a stabilization of HIF1α under these conditions. According to the model proposed by these authors, SUMO-1 modification of HIF1α promotes its recognition by the von Hippel-Lindau (VHL) protein, a subunit of a ubiquitin ligase and subsequent degradation by the proteasome.⁴⁰ These results are still under discussion as other studies reported stabilizing effects of sumoylation on HIF1α under hypoxic conditions.^117-119

CONCLUSION

The studies summarized above indicated that the UPS, via specific ubiquitin ligases, participates in the regulation of SUMO target proteins. Studies in budding and fission yeast have implicated ubiquitylation of SUMO targets in the control of genomic stability, while studies in mammalian cells on the functions of RNF4 have linked this mechanism to the regulation of PML, nuclear bodies and of transcription factors. For an individual substrate multiple scenarios can be envisioned. The sole function of sumoylation of a certain substrate may be to target it for ubiquitin-dependent degradation by the proteasome. In other cases, SUMO-mediated ubiquitylation may also have nonproteolytic targeting functions. For the bulk of targets, sumoylation is likely to serve to transiently regulate protein interactions. The transient nature of this modification on one side is provided by de-sumoylating activities and on the other side by formation of SUMO chains and proteolytic targeting via SUMO-dependent ubiquitylation. The latter scenario resembles the suicide model for activation of certain transcription factors that require monoubiquitylation for their activation, which however is also the beginning of their end since subsequent formation of a ubiquitin chain leads to their degradation.^120,121 This mechanism is thought to ascertain a transient nature of gene activation by so regulated transcription factors. This scenario also resembles that of other protein modifications such as phosphorylation, which can be reversed by dephosphorylation, but which can also trigger ubiquitin-dependent degradation.¹²² Future studies on individual targets will reveal which of the described scenarios is relevant to the function of ubiquitin ligases recognizing SUMO modified proteins.

Acknowledgments

Work in the authors' laboratories is supported by grants from the Deutsche Forschungsgemeinschaft (SPP1365 and SFB635).

REFERENCES

1.

Hochstrasser M. Origin and function of ubiquitin-like proteins. Nature. 2009;458:422–429. [PMC free article: PMC2819001] [PubMed: 19325621]

2.

Meluh PB, Koshland D. Evidence that the MIF2 gene of Saccharomyces cerevisiae encodes a centromere protein with homology to the mammalian centromere protein CENP-C. Mol Biol Cell. 1995;6:793–807. [PMC free article: PMC301241] [PubMed: 7579695]

3.

Mannen H, Tseng HM, Cho CL, et al. Cloning and expression of human homolog HSMT3 to yeast SMT3 suppressor of MIF2 mutations in a centromere protein gene. Biochem Biophys Res Commun. 1996;222:178–180. [PubMed: 8630065]

4.

Johnson ES, Schwienhorst I, Dohmen RJ, et al. The ubiquitin-like protein Smt3p is activated for conjugation to other proteins by an Aos1p/Uba2p heterodimer. EMBO J. 1997;16:5509–5519. [PMC free article: PMC1170183] [PubMed: 9312010]

5.

Tanaka K, Nishide J, Okazaki K, et al. Characterization of a fission yeast SUMO-1 homologue, pmt3p, required for multiple nuclear events, including the control of telomere length and chromosome segregation. Mol Cell Biol. 1999;19:8660–8672. [PMC free article: PMC85006] [PubMed: 10567589]

6.

Nacerddine K, Lehembre F, Bhaumik M, et al. The SUMO pathway is essential for nuclear integrity and chromosome segregation in mice. Dev Cell. 2005;9:769–779. [PubMed: 16326389]

7.

Zhang FP, Mikkonen L, Toppari J, et al. Sumo-1 function is dispensable in normal mouse development. Mol Cell Biol. 2008;28:5381–5390. [PMC free article: PMC2519746] [PubMed: 18573887]

8.

Saitoh H, Hinchey J. Functional heterogeneity of small ubiquitin-related protein modifiers SUMO-1 versus SUMO-2/3. J Biol Chem. 2000;275:6252–6258. [PubMed: 10692421]

9.

Golebiowski F, Matic I, Tatham MH, et al. System-wide changes to SUMO modifications in response to heat shock. Sci Signal. 2009;2 [PubMed: 19471022]

10.

Weisshaar SR, Keusekotten K, Krause A, et al. Arsenic trioxide stimulates SUMO-2/3 modification leading to RNF4-dependent proteolytic targeting of PML. FEBS Lett. 2008;582:3174–3178. [PubMed: 18708055]

11.

Johnson ES, Blobel G. Cell cycle-regulated attachment of the ubiquitin-related protein SUMO to the yeast septins. J Cell Biol. 1999;147:981–994. [PMC free article: PMC2169351] [PubMed: 10579719]

12.

Bernier-Villamor V, Sampson DA, Matunis MJ, et al. Structural basis for E2-mediated SUMO conjugation revealed by a complex between ubiquitin-conjugating enzyme Ubc9 and RanGAP1. Cell. 2002;108:345–356. [PubMed: 11853669]

13.

Tatham MH, Jaffray E, Vaughan OA, et al. Polymeric chains of SUMO-2 and SUMO-3 are conjugated to protein substrates by SAE1/SAE2 and Ubc9. J Biol Chem. 2001;276:35368–35374. [PubMed: 11451954]

14.

Bylebyl GR, Belichenko I, Johnson ES. The SUMO isopeptidase Ulp2 prevents accumulation of SUMO chains in yeast. J Biol Chem. 2003;278:44112–44120. [PubMed: 12941945]

15.

Ulrich HD. The fast-growing business of SUMO chains. Mol Cell. 2008;32:301–305. [PubMed: 18995828]

16.

Matic I, van Hagen M, Schimmel J, et al. In vivo identification of human SUMO polymerization sites by high accuracy mass spectrometry and an in-vitro to in vivo strategy. Mol Cell Proteomics. 2007 [PMC free article: PMC3840926] [PubMed: 17938407]

17.

Muller S, Hoege C, Pyrowolakis G, et al. SUMO, ubiquitin's mysterious cousin. Nat Rev Mol Cell Biol. 2001;2:202–210. [PubMed: 11265250]

18.

Seeler JS, Dejean A. Nuclear and unclear functions of SUMO. Nat Rev Mol Cell Biol. 2003;4:690–699. [PubMed: 14506472]

19.

Johnson ES. Protein modification by SUMO. Annu Rev Biochem. 2004;73:355–382. [PubMed: 15189146]

20.

Hay RT. SUMO: a history of modification. Mol Cell. 2005;18:1–12. [PubMed: 15808504]

21.

Kim JH, Choi HJ, Kim B, et al. Roles of sumoylation of a reptin chromatin-remodelling complex in cancer metastasis. Nat Cell Biol. 2006;8:631–639. [PubMed: 16699503]

22.

Hoeller D, Hecker CM, Dikic I. Ubiquitin and ubiquitin-like proteins in cancer pathogenesis. Nat Rev Cancer. 2006;6:776–788. [PubMed: 16990855]

23.

Geiss-Friedlander R, Melchior F. Concepts in sumoylation: a decade on. Nat Rev Mol Cell Biol. 2007;8:947–956. [PubMed: 18000527]

24.

Bischof O, Dejean A. SUMO is growing senescent. Cell Cycle. 2007;6:677–681. [PubMed: 17374992]

25.

Seeler JS, Bischof O, Nacerddine K, et al. SUMO, the three Rs and cancer. Curr Top Microbiol Immunol. 2007;313:49–71. [PubMed: 17217038]

26.

Bergink S, Jentsch S. Principles of ubiquitin and SUMO modifications in DNA repair. Nature. 2009;458:461–467. [PubMed: 19325626]

27.

Dohmen RJ. SUMO protein modification. Biochim Biophys Acta. 2004;1695:113–131. [PubMed: 15571812]

28.

Desterro JM, Rodriguez MS, Hay RT. SUMO-1 modification of IkappaBalpha inhibits NF-kappaB activation. Mol Cell. 1998;2:233–239. [PubMed: 9734360]

29.

Skaug B, Jiang X, Chen ZJ. The role of ubiquitin in NF-kappaB regulatory pathways. Annu Rev Biochem. 2009;78:769–796. [PubMed: 19489733]

30.

Ulrich HD. Mutual interactions between the SUMO and ubiquitin systems: a plea of no contest. Trends Cell Biol. 2005;15:525–532. [PubMed: 16125934]

31.

Hoege C, Pfander B, Moldovan GL, et al. RAD6-dependent DNA repair is linked to modification of PCNA by ubiquitin and SUMO. Nature. 2002;419:135–141. [PubMed: 12226657]

32.

Ulrich HD, Jentsch S. Two RING finger proteins mediate cooperation between ubiquitin-conjugating enzymes in DNA repair. Embo J. 2000;19:3388–3397. [PMC free article: PMC313941] [PubMed: 10880451]

33.

Sun H, Leverson JD, Hunter T. Conserved function of RNF4 family proteins in eukaryotes: targeting a ubiquitin ligase to SUMOylated proteins. Embo J. 2007;26:4102–112. [PMC free article: PMC2230674] [PubMed: 17762864]

34.

Pfander B, Moldovan GL, Sacher M, et al. SUMO-modified PCNA recruits Srs2 to prevent recombination during S phase. Nature. 2005;436:428–433. [PubMed: 15931174]

35.

Papouli E, Chen S, Davies AA, et al. Crosstalk between SUMO and ubiquitin on PCNA is mediated by recruitment of the helicase Srs2p. Mol Cell. 2005;19:123–133. [PubMed: 15989970]

36.

Uzunova K, Gottsche K, Miteva M, et al. Ubiquitin-dependent proteolytic control of SUMO conjugates. J Biol Chem. 2007;282:34167–34175. [PubMed: 17728242]

37.

Xie Y, Kerscher O, Kroetz MB, et al. The yeast HEX3-SLX8 heterodimer is a ubiquitin ligase stimulated by substrate sumoylation. J Biol Chem. 2007 [PubMed: 17848550]

38.

Mullen JR, Brill SJ. Activation of the Slx5-Slx8 ubiquitin ligase by poly-small ubiquitin-like modifier conjugates. J Biol Chem. 2008;283:19912–19921. [PMC free article: PMC2459270] [PubMed: 18499666]

39.

Prudden J, Pebernard S, Raffa G, et al. SUMO-targeted ubiquitin ligases in genome stability. Embo J. 2007;26:4089–4101. [PMC free article: PMC2230673] [PubMed: 17762865]

40.

Cheng J, Kang X, Zhang S, et al. SUMO-specific protease 1 is essential for stabilization of HIF1alpha during hypoxia. Cell. 2007;131:584–595. [PMC free article: PMC2128732] [PubMed: 17981124]

41.

Mahajan R, Delphin C, Guan T, et al. A small ubiquitin-related polypeptide involved in targeting RanGAP1 to nuclear pore complex protein RanBP2. Cell. 1997;88:97–107. [PubMed: 9019411]

42.

Saitoh H, Pu R, Cavenagh M, et al. RanBP2 associates with Ubc9p and a modified form of RanGAP1. Proc Natl Acad Sci USA. 1997;94:3736–3741. [PMC free article: PMC20510] [PubMed: 9108047]

43.

Matunis MJ, Wu J, Blobel G. SUMO-1 modification and its role in targeting the Ran GTPase-activating protein, RanGAP1, to the nuclear pore complex. J Cell Biol. 1998;140:499–509. [PMC free article: PMC2140169] [PubMed: 9456312]

44.

Song J, Durrin LK, Wilkinson TA, et al. Identification of a SUMO-binding motif that recognizes SUMO-modified proteins. Proc Natl Acad Sci USA. 2004;101:14373–14378. [PMC free article: PMC521952] [PubMed: 15388847]

45.

Shen TH, Lin HK, Scaglioni PP, et al. The mechanisms of PML-nuclear body formation. Mol Cell. 2006;24:331–339. [PMC free article: PMC1978182] [PubMed: 17081985]

46.

Lin DY, Huang YS, Jeng JC, et al. Role of SUMO-interacting motif in Daxx SUMO modification, subnuclear localization and repression of sumoylated transcription factors. Mol Cell. 2006;24:341–354. [PubMed: 17081986]

47.

Kerscher O. SUMO junction-what's your function- New insights through SUMO-interacting motifs. EMBO Rep. 2007;8:550–555. [PMC free article: PMC2002525] [PubMed: 17545995]

48.

Hardeland U, Steinacher R, Jiricny J, et al. Modification of the human thymine-DNA glycosylase by ubiquitin-like proteins facilitates enzymatic turnover. EMBO J. 2002;21:1456–1464. [PMC free article: PMC125358] [PubMed: 11889051]

49.

Steinacher R, Schar P. Functionality of human thymine DNA glycosylase requires SUMO-regulated changes in protein conformation. Curr Biol. 2005;15:616–623. [PubMed: 15823533]

50.

Hay RT. SUMO-specific proteases: a twist in the tail. Trends Cell Biol. 2007;17:370–376. [PubMed: 17768054]

51.

Mukhopadhyay D, Dasso M. Modification in reverse: the SUMO proteases. Trends Biochem Sci. 2007;32:286–295. [PubMed: 17499995]

52.

Li SJ, Hochstrasser M. A new protease required for cell-cycle progression in yeast. Nature. 1999;398:246–251. [PubMed: 10094048]

53.

Li SJ, Hochstrasser M. The yeast ULP2 (SMT4) gene encodes a novel protease specific for the ubiquitin-like Smt3 protein. Mol Cell Biol. 2000;20:2367–2377. [PMC free article: PMC85410] [PubMed: 10713161]

54.

Schwienhorst I, Johnson ES, Dohmen RJ. SUMO conjugation and deconjugation. Mol Gen Genet. 2000;263:771–786. [PubMed: 10905345]

55.

Xu Z, Chan HY, Lam WL, et al. SUMO Proteases: Redox Regulation and Biological Consequences. Antioxid Redox Signal. 2009;11:1453–1484. [PubMed: 19186998]

56.

Mukhopadhyay D, Ayaydin F, Kolli N, et al. SUSP1 antagonizes formation of highly SUMO2/3-conjugated species. J Cell Biol. 2006;174:939–949. [PMC free article: PMC2064386] [PubMed: 17000875]

57.

Hofmann K. Ubiquitin-binding domains and their role in the DNA damage response. DNA Repair (Amst). 2009;8:544–556. [PubMed: 19213613]

58.

Minty A, Dumont X, Kaghad M, et al. Covalent modification of p73alpha by SUMO-1, Two-hybrid screening with p73 identifies novel SUMO-1-interacting proteins and a SUMO-1 interaction motif. J Biol Chem. 2000;275:36316–36323. [PubMed: 10961991]

59.

Song J, Zhang Z, Hu W, et al. Small ubiquitin-like modifier (SUMO) recognition of a SUMO binding motif: a reversal of the bound orientation. J Biol Chem. 2005;280:40122–40129. [PubMed: 16204249]

60.

Hannich JT, Lewis A, Kroetz MB, et al. Defining the SUMO-modified proteome by multiple approaches in Saccharomyces cerevisiae. J Biol Chem. 2005;280:4102–4110. [PubMed: 15590687]

61.

Hecker CM, Rabiller M, Haglund K, et al. Specification of SUMO1- and SUMO2-interacting motifs. J Biol Chem. 2006;281:16117–16127. [PubMed: 16524884]

62.

Sekiyama N, Ikegami T, Yamane T, et al. Structure of the small ubiquitin-like modifier (SUMO)-interacting motif of MBD1-containing chromatin-associated factor 1 bound to SUMO-3. J Biol Chem. 2008;283:35966–35975. [PubMed: 18842587]

63.

Reverter D, Lima CD. Insights into E3 ligase activity revealed by a SUMO-RanGAP1-Ubc9-Nup358 complex. Nature. 2005;435:687–692. [PMC free article: PMC1416492] [PubMed: 15931224]

64.

Stehmeier P, Muller S. Phospho-regulated SUMO interaction modules connect the SUMO system to CK2 signaling. Mol Cell. 2009;33:400–409. [PubMed: 19217413]

65.

Hershko A, Ciechanover A. The ubiquitin system. Annu Rev Biochem. 1998;67:425–479. [PubMed: 9759494]

66.

Varshavsky A. Regulated protein degradation. Trends Biochem Sci. 2005;30:283–286. [PubMed: 15950869]

67.

Weissman AM. Themes and variations on ubiquitylation. Nat Rev Mol Cell Biol. 2001;2:169–178. [PubMed: 11265246]

68.

Joazeiro CA, Weissman AM. RING finger proteins: mediators of ubiquitin ligase activity. Cell. 2000;102:549–552. [PubMed: 11007473]

69.

Wang Z, Prelich G. Quality control of a transcriptional regulator by SUMO-targeted degradation. Mol Cell Biol. 2009;29:1694–1706. [PMC free article: PMC2655623] [PubMed: 19139279]

70.

Uetz P, Giot L, Cagney G, et al. A comprehensive analysis of protein-protein interactions in Saccharomyces cerevisiae. Nature. 2000;403:623–627. [PubMed: 10688190]

71.

Ito T, Chiba T, Ozawa R, et al. A comprehensive two-hybrid analysis to explore the yeast protein interactome. Proc Natl Acad Sci USA. 2001;98:4569–4574. [PMC free article: PMC31875] [PubMed: 11283351]

72.

Wang Z, Jones GM, Prelich G. Genetic analysis connects SLX5 and SLX8 to the SUMO pathway in Saccharomyces cerevisiae. Genetics. 2006;172:1499–1509. [PMC free article: PMC1456262] [PubMed: 16387868]

73.

Mullen JR, Kaliraman V, Ibrahim SS, et al. Requirement for three novel protein complexes in the absence of the Sgs1 DNA helicase in Saccharomyces cerevisiae. Genetics. 2001;157:103–118. [PMC free article: PMC1461486] [PubMed: 11139495]

74.

Cobb JA, Bjergbaek L. RecQ helicases: lessons from model organisms. Nucleic Acids Res. 2006;34:4106–4114. [PMC free article: PMC1616958] [PubMed: 16936315]

75.

Pan X, Ye P, Yuan DS, et al. A DNA integrity network in the yeast Saccharomyces cerevisiae. Cell. 2006;124:1069–1081. [PubMed: 16487579]

76.

Collins SR, Miller KM, Maas NL, et al. Functional dissection of protein complexes involved in yeast chromosome biology using a genetic interaction map. Nature. 2007;446:806–810. [PubMed: 17314980]

77.

Ulrich HD. PCNASUMO and Srs2: a model SUMO substrate-effector pair. Biochem Soc Trans. 2007;35:1385–1388. [PubMed: 18031227]

78.

Watts FZ. Sumoylation of PCNA: Wrestling with recombination at stalled replication forks. DNA Repair (Amst). 2006;5:399–403. [PubMed: 16368276]

79.

Soustelle C, Vernis L, Freon K, et al. A new Saccharomyces cerevisiae strain with a mutant Smt3-deconjugating Ulp1 protein is affected in DNA replication and requires Srs2 and homologous recombination for its viability. Mol Cell Biol. 2004;24:5130–5143. [PMC free article: PMC419856] [PubMed: 15169880]

80.

Azam M, Lee JY, Abraham V, et al. Evidence that the S.cerevisiae Sgs1 protein facilitates recombinational repair of telomeres during senescence. Nucleic Acids Res. 2006;34:506–516. [PMC free article: PMC1342037] [PubMed: 16428246]

81.

Zhang C, Roberts TM, Yang J, et al. Suppression of genomic instability by SLX5 and SLX8 in Saccharomyces cerevisiae. DNA Repair (Amst). 2006;5:336–346. [PubMed: 16325482]

82.

Burgess RC, Rahman S, Lisby M, et al. The Slx5-Slx8 complex affects sumoylation of DNA repair proteins and negatively regulates recombination. Mol Cell Biol. 2007;27:6153–6162. [PMC free article: PMC1952148] [PubMed: 17591698]

83.

Takahashi Y, Dulev S, Liu X, et al. Cooperation of sumoylated chromosomal proteins in rDNA maintenance. PLoS Genet. 2008;4 [PMC free article: PMC2563031] [PubMed: 18846224]

84.

Putnam CD, Hayes TK, Kolodner RD. Specific pathways prevent duplication-mediated genome rearrangements. Nature. 2009;460:984–989. [PMC free article: PMC2785216] [PubMed: 19641493]

85.

Rouse J. Control of genome stability by SLX protein complexes. Biochem Soc Trans. 2009;37:495–510. [PubMed: 19442243]

86.

Heideker J, Perry JJ, Boddy MN. Genome stability roles of SUMO-targeted ubiquitin ligases. DNA Repair (Amst). 2009;8:517–524. [PMC free article: PMC2685196] [PubMed: 19233742]

87.

Cook CE, Hochstrasser M, Kerscher O. The SUMO-targeted ubiquitin ligase subunit Slx5 resides in nuclear foci and at sites of DNA breaks. Cell Cycle. 2009;8:1080–1089. [PMC free article: PMC2700622] [PubMed: 19270524]

88.

Nagai S, Dubrana K, Tsai-Pflugfelder M, et al. Functional targeting of DNA damage to a nuclear pore-associated SUMO-dependent ubiquitin ligase. Science. 2008;322:597–602. [PMC free article: PMC3518492] [PubMed: 18948542]

89.

Yang L, Mullen JR, Brill SJ. Purification of the yeast Slx5-Slx8 protein complex and characterization of its DNA-binding activity. Nucleic Acids Res. 2006;34:5541–5551. [PMC free article: PMC1635298] [PubMed: 17020915]

90.

Zhang Z, Buchman AR. Identification of a member of a DNA-dependent ATPase family that causes interference with silencing. Mol Cell Biol. 1997;17:5461–5472. [PMC free article: PMC232395] [PubMed: 9271422]

91.

Shirai C, Mizuta K. SUMO mediates interaction of Ebp2p, the yeast homolog of Epstein-Barr virus nuclear antigen 1-binding protein 2, with a RING finger protein Ris1p. Biosci Biotechnol Biochem. 2008;72:1881–1886. [PubMed: 18603780]

92.

Novatchkova M, Bachmair A, Eisenhaber B, et al. Proteins with two SUMO-like domains in chromatin-associated complexes: the RENi (Rad60-Esc2-NIP45) family. BMC Bioinformatics. 2005;6 [PMC free article: PMC549199] [PubMed: 15698469]

93.

Prudden J, Perry JJ, Arvai AS, et al. Molecular mimicry of SUMO promotes DNA repair. Nat Struct Mol Biol. 2009;16:509–516. [PMC free article: PMC2711901] [PubMed: 19363481]

94.

Häkli M, Lorick KL, Weissman AM, et al. Transcriptional coregulator SNURF (RNF4) possesses ubiquitin E3 ligase activity. FEBS Lett. 2004;560:56–62. [PubMed: 14987998]

95.

Kosoy A, Calonge TM, Outwin EA, et al. Fission yeast rnf4 homologs are required for DNA repair. J Biol Chem. 2007;282:20388–20394. [PubMed: 17502373]

96.

Tatham MH, Geoffroy MC, Shen L, et al. RNF4 is a poly-SUMO-specific E3 ubiquitin ligase required for arsenic-induced PML degradation. Nat Cell Biol. 2008;10:538–546. [PubMed: 18408734]

97.

Häkli M, Karvonen U, Janne OA, et al. SUMO-1 promotes association of SNURF (RNF4) with PML nuclear bodies. Exp Cell Res. 2005;304:224–233. [PubMed: 15707587]

98.

Muller S, Matunis MJ, Dejean A. Conjugation with the ubiquitin-related modifier SUMO-1 regulates the partitioning of PML within the nucleus. EMBO J. 1998;17:61–70. [PMC free article: PMC1170358] [PubMed: 9427741]

99.

Ishov AM, Sotnikov AG, Negorev D, et al. PML is critical for ND10 formation and recruits the PML-interacting protein daxx to this nuclear structure when modified by SUMO-1. J Cell Biol. 1999;147:221–234. [PMC free article: PMC2174231] [PubMed: 10525530]

100.

Zhong S, Muller S, Ronchetti S, et al. Role of SUMO-1-modified PML in nuclear body formation. Blood. 2000;95:2748–2752. [PubMed: 10779416]

101.

Bernardi R, Pandolfi PP. Structure, dynamics and functions of promyelocytic leukaemia nuclear bodies. Nat Rev Mol Cell Biol. 2007;8:1006–1016. [PubMed: 17928811]

102.

Nisole S, Stoye JP, Saib A. TRIM family proteins: retroviral restriction and antiviral defence. Nat Rev Microbiol. 2005;3:799–808. [PubMed: 16175175]

103.

de The H, Chomienne C, Lanotte M, et al. The t(15;17) translocation of acute promyelocytic leukaemia fuses the retinoic acid receptor alpha gene to a novel transcribed locus. Nature. 1990;347:558–561. [PubMed: 2170850]

104.

Kakizuka A, Miller WH Jr, Umesono K, et al. Chromosomal translocation t(15;17) in human acute promyelocytic leukemia fuses RAR alpha with a novel putative transcription factor, PML. Cell. 1991;66:663–674. [PubMed: 1652368]

105.

Wang ZY, Chen Z. Acute promyelocytic leukemia: from highly fatal to highly curable. Blood. 2008;111:2505–515. [PubMed: 18299451]

106.

Lallemand-Breitenbach V, Zhu J, Puvion F, et al. Role of promyelocytic leukemia (PML) sumolation in nuclear body formation, 11S proteasome recruitment and As2O3-induced PML or PML/retinoic acid receptor alpha degradation. J Exp Med. 2001;193:1361–1371. [PMC free article: PMC2193303] [PubMed: 11413191]

107.

Lallemand-Breitenbach V, Jeanne M, Benhenda S, et al. Arsenic degrades PML or PML-RARalpha through a SUMO-triggered RNF4/ubiquitin-mediated pathway. Nat Cell Biol. 2008;10:547–555. [PubMed: 18408733]

108.

Bossis G, Melchior F. Regulation of SUMOylation by Reversible Oxidation of SUMO Conjugating Enzymes. Mol Cell. 2006;21:349–357. [PubMed: 16455490]

109.

Hayakawa F, Privalsky ML. Phosphorylation of PML by mitogen-activated protein kinases plays a key role in arsenic trioxide-mediated apoptosis. Cancer Cell. 2004;5:389–401. [PubMed: 15093545]

110.

Scaglioni PP, Yung TM, Choi SC, et al. CK2 mediates phosphorylation and ubiquitin-mediated degradation of the PML tumor suppressor. Mol Cell Biochem. 2008;316:149–154. [PubMed: 18566754]

111.

Percherancier Y, Germain-Desprez D, Galisson F, et al. Role of SUMO in RNF4-mediated promyelocytic leukemia protein (PML) degradation: sumoylation of PML and phospho-switch control of its SUMO binding domain dissected in living cells. J Biol Chem. 2009;284:16595–16608. [PMC free article: PMC2713554] [PubMed: 19380586]

112.

Knipscheer P, Flotho A, Klug H, et al. Ubc9 sumoylation regulates SUMO target discrimination. Mol Cell. 2008;31:371–382. [PubMed: 18691969]

113.

Meulmeester E, Kunze M, Hsiao HH, et al. Mechanism and consequences for paralog-specific sumoylation of ubiquitin-specific protease 25. Mol Cell. 2008;30:610–619. [PubMed: 18538659]

114.

Cho G, Lim Y, Golden JA. SUMO interaction motifs in Sizn1 are required for promyelocytic leukemia protein nuclear body localization and for transcriptional activation. J Biol Chem. 2009;284:19592–19600. [PMC free article: PMC2740585] [PubMed: 19416967]

115.

Hoeller D, Crosetto N, Blagoev B, et al. Regulation of ubiquitin-binding proteins by monoubiquitination. Nat Cell Biol. 2006;8:163–169. [PubMed: 16429130]

116.

Ito K, Bernardi R, Morotti A, et al. PML targeting eradicates quiescent leukaemia-initiating cells. Nature. 2008;453:1072–1078. [PMC free article: PMC2712082] [PubMed: 18469801]

117.

Bae SH, Jeong JW, Park JA, et al. Sumoylation increases HIF-1alpha stability and its transcriptional activity. Biochem Biophys Res Commun. 2004;324:394–400. [PubMed: 15465032]

118.

Carbia-Nagashima A, Gerez J, Perez-Castro C, et al. RSUME, a Small RWD-Containing Protein, Enhances SUMO Conjugation and Stabilizes HIF-1alpha during Hypoxia. Cell. 2007;131:309–323. [PubMed: 17956732]

119.

Geoffroy MC, Hay RT. An additional role for SUMO in ubiquitin-mediated proteolysis. Nat Rev Mol Cell Biol. 2009;10:564–568. [PubMed: 19474794]

120.

Salghetti SE, Caudy AA, Chenoweth JG, et al. Regulation of transcriptional activation domain function by ubiquitin. Science. 2001;293:1651–1653. [PubMed: 11463878]

121.

Conaway RC, Brower CS, Conaway JW. Emerging roles of ubiquitin in transcription regulation. Science. 2002;296:1254–1258. [PubMed: 12016299]

122.

Hunter T. The age of crosstalk: phosphorylation, ubiquitination and beyond. Mol Cell. 2007;28:730–738. [PubMed: 18082598]