Monoubiquitination in Proteasomal Degradation Ze’Ev A

COMMENTARY COMMENTARY

Monoubiquitination in proteasomal degradation Ze’ev A. Ronaia,1

Conjugation of ubiquitin to target proteins is a finely thedurationandmagnitudeofaprotein’s activity by pro- tuned process involving a reaction, culminating in the moting its clearance (4, 5). conjugation of a single ubiquitin (1). The first step is The diverse functions associated with protein conjugation of a single ubiquitin molecule to the sub- ubiquitination require exquisite regulation to ensure strate’s protein amino group (monoubiquitination) or that the cellular machinery can distinguish between to multiple amino groups (multimonoubiquitination), ubiquitination that marks a protein for translocation which can remain as it is or be further extended by from one that targets the protein for elimination. This additional ubiquitin molecules to form elongated process is accomplished, in part, through the length, chains (polyubiquitination). The process is reversible type, and organization of ubiquitin molecules on the because deubiquitinating enzymes can partially trim substrate. On a simplistic level, monoubiquitination the polyubiquitin chains or completely strip all ubiquitin has largely been linked to chromatin regulation, protein molecules from the protein (2, 3). Protein ubiquitination sorting, and trafficking, whereas polyubiquitination has is of great physiological significance and plays a role in been associated with protein signaling and clearance every cellular process. Examples include determining through proteasomal or autophagic degradation (5–7). protein subcellular localization, regulating signaling An additional layer of regulatory complexity results from (e.g., transcription factors, protein kinases), and controlling the different topologies of polyubiquitin chains conjugated through various lysine residues (e.g., K6, K11, K29, K48, and K63). Different polyubiquitin chain topologies have been documented in vitro and in vivo, and they Required ubiquitin code for recognition by the 26S proteasome have been modeled structurally (8–11). In particular, high- Identified by cellular proteomic screen resolution MS/MS has revealed the highly heterogenous nature of polyubiquitin chains and their topologies. poly-Ub proteins mono /multimono-Ub proteins Cellular homeostasis critically depends on maintain- yeast (≈75%) > human (≈50%) yeast (≈25%) < human (≈50%) ing a balance between the synthesis of proteins and the clearance of proteins that have fulfilled their cellular duty (e.g., transient signaling) or have aberrant structure (e.g., genetic or cellular stress-induced misfolding). Protein degradation, which is primarily carried out by proteasomes, requires structural organization of the proteasome and Physical characteristics: Physical characteristics: of ubiquitin-binding subunits within the 19S proteasome protein size protein size subcomplex. Efficient entry and processing of proteins with no preference mostly smaller proteins monoubiquitin, multimonoubiquitin, or heterogenous protein structure protein structure ubiquitin linkages into the proteasome also depend H: mostlyly proteins H: proteins with less on deubiquitinating enzymes and other regulatory factors – with regions of structural disorder (6, 12 17). structural disorder Y: no preference The study by Braten et al. in PNAS (18) demon- Y: no preference strates that mono- and multimonoubiquitinated proteins are commonly processed by proteasomes, more Ub-site structure Ub-site structure than we have appreciated to date. To identify and char- H: more disordered H: less disordered Y: less disordered Y: more disordered acterize monoubiquitinated proteasomal substrates, the authors silenced expression of endogenous wild-type ubiquitin and expressed a nonpolymerizable Fig. 1. Recognition of ubiquitinated proteins by the proteasome. The outline lysineless ubiquitin. The ectopically expressed ubiquitin summarizes the findings by Braten et al. (18). H, human; Y, yeast. can modify each lysine in the target substrate only once

aTumor Initiation and Maintenance Program, Sanford Burnham Prebys Medical Discovery Institute, La Jolla, CA 92037 Author contributions: Z.A.R. wrote the paper. The author declares no conflict of interest. See companion article on page E4639. 1Email: [email protected].

8894–8896 | PNAS | August 9, 2016 | vol. 113 | no. 32 www.pnas.org/cgi/doi/10.1073/pnas.1610186113 Downloaded by guest on September 28, 2021 and cannot be further conjugated, creating only monoubiquiti- this. Proteins of higher organisms often contain large regions that be- nated and multimonoubiquitinated proteins. Braten et al. (18) ex- come intrinsically disordered upon posttranslational modification or amined both human and yeast cells using this approach, which association with other proteins; such regions can also oscillate be- allowed a comparison of the significance of monoubiquitination tween ordered and disordered forms. Bioinformatics modeling has in protein clearance in both species. The conditions used to analyze shown that disordered proteins are enriched in eukaryotes compared the yeast and human cells differed technically. In yeast cells, the with prokaryotes and are associated with signaling diversity [achieved ubiquitin genes were encoded by a dual promoter system, where by posttranslational modification (20–24), of which ubiquitination is galactose-dependent expression is used to achieve tight repression one example]. Of note, the presence of disordered regions in a pro- of synthetic wild-type ubiquitin and copper to induce lysineless ubiq- tein is a required signal for degradation, as evidenced by the prox- uitin expression. In human U2OS osteosarcoma cells, endoge- imity of long-disordered regions to ubiquitin ligase recognition – nous ubiquitin was silenced with tetracycline-inducible shRNA and motifs and ubiquitin-acceptor lysine residues (20 23). Such re- then wild-type or lysineless ubiquitin was expressed by adenovi- arrangement is thought to enable unfolding of the substrate ral infection. A CRISPR-based approach for expression of the after engagement by the proteasome. Whether proteins with lysineless ubiquitin gene may more closely resemble the ap- larger disordered regions are generally more dependent on proach used in the yeast system. However, the subsequent analyt- p97, proteasome ubiquitin binding sites, and ATPases, is ical steps were similar in yeast and human cells. In both cases, among the explanations that may enable proteasome recognition by different conformation/degree of disordered proteins. differentially ubiquitinated proteins were identified by MS-based stable isotope-labeling by amino acids in cell culture and label-free quantification. The classification of proteins as monoubiquitin- and Protein structural disorder (i.e., disordered polyubiquitin-dependent proteasome substrates was based on regions that lack a defined tertiary structure) has the ratio of signal intensity between wild-type and lysineless ubiq- previously been proposed to be a determinant uitin. Braten et al. (18) note that a subset of proteins classified as monoubiquitinated is expected to include proteins harboring mul- of protein ubiquitination and proteasomal tiple monoubiquitinated moieties conjugated to multiple different recognition; the Braten et al. study provides lysines. Multiple ubiquitin molecules spread across a single pro- new data to support this. tein were shown to provide a degradation signal that can at times mimic polyubiquitination (12). Because polyubiquitinated The analysis by Braten et al. (18) suggests that, although structural substratesexploremultipleconfigurationsontheproteasomes disorder is seen in both monoubiquitinated and polyubiquitinated through stochastic binding (17), one wonders whether this can proteins in yeast, it is largely confined to polyubiquitinated proteins also be the case for multimonoubiquitinated proteins. Along in humans. Consistent with the notion that polyubiquitination provides these lines, SIM domains and SUMOylation may be part of the a stronger signal, structurally disordered proteins may benefit from the recognition signal for monoubiquitinated substrates. extended ubiquitination chains. The observed enrichment of polyubi- ∼ Surprisingly, only 25% of proteins degraded by the proteasome quitinated proteins in humans may be a result of factors that define were mono- (or multimono) ubiquitinated in yeast (82), compared with disordered regions, such as posttranslational modifications or the prox- ∼ 50% in human (220), whereas 303 and 416 proteins are depen- imity to ubiquitinated sites. Other considerations that may underlie the dent on polyubiquitination for proteasomal recognition and pro- differences seen in yeast and human cells include the replicative cessing in yeast and human, respectively (Fig. 1). This unexpected and physiological state of the cells (e.g., higher replication of the observation led Braten et al. (18) to suggest that monoubiquitination U2OS cells, differing levels of internal stress, and disparate de- is much more common in human cells than in yeast. One explana- grees of signaling pathway activation, to name a few). Experimen- tion for this difference might be the greater structural diversity of tally, it may be possible to establish a series of different structurally monoubiquitinated proteins in human cells (see below). The authors disordered domains within the same protein, enabling assessment identify distinct populations of proteasomally processed ubiquitin- of the relationship between structural disorder, the ability to un- modified proteins, leading them to conclude that proteasomal dergo monoubiquitination and polyubiquitination, and recogni- processing of monoubiquitinated proteins is also more common tion by the proteasome. in human cells than in yeast. Several considerations may shed further light on the physio- Further classification of the monoubiquitinated and polyubi- logical significance of these observations. For example, is a quitinated proteins using bioinformatics tools points to an associ- monoubiquitinated protein subject to polyubiquitination under ation between protein size, biological process, and sequence in the physiological conditions, where the prevalence of lysineless ubiquitin ubiquitinated domain. In both humans and yeast, shorter proteins does not preclude chain formation? Are there physiological condi- (<150 amino acids) were more likely to undergo monoubiquitina- tions that may favor the monoubiquitination state over multimono- tion, confirming the earlier observations by the Ciechanover group or polyubiquitnation? Similarly, it will be of interest to determine (15, 19). Braten et al. (18) suggest that monoubiquitination may whether a monoubiquitinated protein can out-compete the multi- provide a weaker targeting signal to the proteasomes, which is mono- or polyubiquitinated protein for proteasome recognition? sufficient for degradation of smaller proteins. Because longer pro- The tools used in the Braten et al. (18) study allow us to further teins were associated with either monoubiquitination or polyubiqui- study the ubiquitination requirements for proteasomal degradation. tination, properties such as structural disorder are among factors Spatiotemporal analyses of genetically comparable systems could that may govern their recognition by proteasomes. help clarify how changes in ubiquitination/degradation of regulatory Protein structural disorder (i.e., disordered regions that lack a pathways affect their recognition by the proteasomes. Similarly, fur- defined tertiary structure) has previously been proposed to be a ther studies will be required to determine the contribution of cofac- determinant of protein ubiquitination and proteasomal recognition tors, such as p97, ATPases, and deubiquitinating enzymes, to the (20–22); the Braten et al. (18) study provides new data to support recognition of ubiquitin-modified disordered regions by the

Ronai PNAS | August 9, 2016 | vol. 113 | no. 32 | 8895 Downloaded by guest on September 28, 2021 proteasomes. In all, the findings of Braten et al. (18) are expected to disordered proteins may allow us to better define novel approaches catalyze new lines of studies aimed at better understanding the to alter their stability or activity for therapeutic purposes. mechanisms underlying protein recognition by proteasomes, and the role of monoubiquitination vs. polyubiquitination in the context Acknowledgments of disordered regions in proteasome recognition. The possibility to The author’s research is supported by National Cancer Institute Grant CA CA197465, predict the type of ubiquitination that takes place in structurally and the Hervey Family Nonendowment Fund at The San Diego Foundation.

1 Hershko A, Ciechanover A (1998) The ubiquitin system. Annu Rev Biochem 67:425–479. 2 Eletr ZM, Wilkinson KD (2014) Regulation of proteolysis by human deubiquitinating enzymes. Biochim Biophys Acta 1843(1):114–128. 3 Chau V, et al. (1989) A multiubiquitin chain is confined to specific lysine in a targeted short-lived protein. Science 243(4898):1576–1583. 4 Varshavsky A (2012) The ubiquitin system, an immense realm. Annu Rev Biochem 81:167–176. 5 Husnjak K, Dikic I (2012) Ubiquitin-binding proteins: Decoders of ubiquitin-mediated cellular functions. Annu Rev Biochem 81:291–322. 6 Finley D (2009) Recognition and processing of ubiquitin-protein conjugates by the proteasome. Annu Rev Biochem 78:477–513. 7 Khaminets A, Behl C, Dikic I (2016) Ubiquitin-dependent and independent signals in selective autophagy. Trends Cell Biol 26(1):6–16. 8 Ye Y, Rape M (2009) Building ubiquitin chains: E2 enzymes at work. Nat Rev Mol Cell Biol 10(11):755–764. 9 Akutsu M, Dikic I, Bremm A (2016) Ubiquitin chain diversity at a glance. J Cell Sci 129(5):875–880. 10 Chen ZJ (2012) Ubiquitination in signaling to and activation of IKK. Immunol Rev 246(1):95–106. 11 Harper JW, Schulman BA (2006) Structural complexity in ubiquitin recognition. Cell 124(6):1133–1136. 12 Dimova NVHN, et al. (2012) APC/C-mediated multiple monoubiquitylation provides an alternative degradation signal for cyclin B1. Nat Cell Biol 14(2):168–176. 13 Finley D, Chen X, Walters KJ (2016) Gates, channels, and switches: Elements of the proteasome machine. Trends Biochem Sci 41(1):77–93. 14 Rosenzweig R, Osmulski PA, Gaczynska M, Glickman MH (2008) The central unit within the 19S regulatory particle of the proteasome. Nat Struct Mol Biol 15(6): 573–580. 15 Shabek N, et al. (2012) The size of the proteasomal substrate determines whether its degradation will be mediated by mono- or polyubiquitylation. Mol Cell 48(1): 87–97. 16 Thrower JS, Hoffman L, Rechsteiner M, Pickart CM (2000) Recognition of the polyubiquitin proteolytic signal. EMBO J 19(1):94–102. 17 Lu YLB, King RW, Finley D, Kirschner MW (2015) Substrate degradation by the proteasome: A single-molecule kinetic analysis. Science 348(6231):1250834. 18 Braten O, et al. (2016) Numerous proteins with unique characteristics are degraded by the 26S proteasome following monoubiquitination. Proc Natl Acad Sci USA 113:E4639–E4647. 19 Ciechanover A, Stanhill A (2014) The complexity of recognition of ubiquitinated substrates by the 26S proteasome. Biochim Biophys Acta 1843(1):86–96. 20 Zhang Y, Stec B, Godzik A (2007) Between order and disorder in protein structures: Analysis of “dual personality” fragments in proteins. Structure 15(9): 1141–1147. 21 Dunker AK, Silman I, Uversky VN, Sussman JL (2008) Function and structure of inherently disordered proteins. Curr Opin Struct Biol 18(6):756–764. 22 Guharoy M, Bhowmick P, Tompa P (2016) Design principles involving protein discorder facilitate specific substrate selection and degradation by the ubiquitin- proteasome system. J Biol Chem 291(13):6723–6731. 23 Aufderheide A, Unverdorben P, Baumeister W, Förster F (2015) Structural disorder and its role in proteasomal degradation. FEBS Lett 589(19 Pt A):2552–2560. 24 Yu H, Kago G, Yellman CM, Matouschek A (May 27, 2016) Ubiquitin-like domains can target to the proteasome but proteolysis requires a disordered region. EMBO J, 10.15252/embj.201593147.

8896 | www.pnas.org/cgi/doi/10.1073/pnas.1610186113 Ronai Downloaded by guest on September 28, 2021