The deubiquitinase USP36 promotes snoRNP group SUMOylation and is essential for ribosome biogenesis
2021; Springer Nature; Volume: 22; Issue: 6 Linguagem: Inglês
10.15252/embr.202050684
ISSN1469-3178
AutoresHyunju Ryu, Xiao‐Xin Sun, Yingxiao Chen, Yanping Li, Xiaoyan Wang, Roselyn S Dai, Hongming Zhu, John Klimek, Larry L. David, Lev M. Fedorov, Yoshiaki Azuma, Rosalie C. Sears, Mu‐Shui Dai,
Tópico(s)Peptidase Inhibition and Analysis
ResumoArticle14 April 2021free access Source Data The deubiquitinase USP36 promotes snoRNP group SUMOylation and is essential for ribosome biogenesis Hyunju Ryu Hyunju Ryu orcid.org/0000-0002-3188-7092 Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USAThese authors contributed equally to this work Search for more papers by this author Xiao-Xin Sun Xiao-Xin Sun Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USAThese authors contributed equally to this work Search for more papers by this author Yingxiao Chen Yingxiao Chen Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USAThese authors contributed equally to this work Search for more papers by this author Yanping Li Yanping Li Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Xiaoyan Wang Xiaoyan Wang Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Roselyn S Dai Roselyn S Dai Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Hong-Ming Zhu Hong-Ming Zhu Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author John Klimek John Klimek Department of Chemical Physiology & Biochemistry, School of Medicine, Oregon Health & Science University, Portland, OR, USA OHSU Proteomics Shared Resource, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Larry David Larry David Department of Chemical Physiology & Biochemistry, School of Medicine, Oregon Health & Science University, Portland, OR, USA OHSU Proteomics Shared Resource, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Lev M Fedorov Lev M Fedorov OHSU Transgenic Mouse Models Shared Resource, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Yoshiaki Azuma Yoshiaki Azuma orcid.org/0000-0002-9634-6676 Department of Molecular Biosciences, University of Kansas, Lawrence, KS, USA Search for more papers by this author Rosalie C Sears Rosalie C Sears Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Mu-Shui Dai Corresponding Author Mu-Shui Dai [email protected] orcid.org/0000-0001-9031-6962 Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Hyunju Ryu Hyunju Ryu orcid.org/0000-0002-3188-7092 Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USAThese authors contributed equally to this work Search for more papers by this author Xiao-Xin Sun Xiao-Xin Sun Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USAThese authors contributed equally to this work Search for more papers by this author Yingxiao Chen Yingxiao Chen Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USAThese authors contributed equally to this work Search for more papers by this author Yanping Li Yanping Li Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Xiaoyan Wang Xiaoyan Wang Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Roselyn S Dai Roselyn S Dai Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Hong-Ming Zhu Hong-Ming Zhu Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author John Klimek John Klimek Department of Chemical Physiology & Biochemistry, School of Medicine, Oregon Health & Science University, Portland, OR, USA OHSU Proteomics Shared Resource, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Larry David Larry David Department of Chemical Physiology & Biochemistry, School of Medicine, Oregon Health & Science University, Portland, OR, USA OHSU Proteomics Shared Resource, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Lev M Fedorov Lev M Fedorov OHSU Transgenic Mouse Models Shared Resource, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Yoshiaki Azuma Yoshiaki Azuma orcid.org/0000-0002-9634-6676 Department of Molecular Biosciences, University of Kansas, Lawrence, KS, USA Search for more papers by this author Rosalie C Sears Rosalie C Sears Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Mu-Shui Dai Corresponding Author Mu-Shui Dai [email protected] orcid.org/0000-0001-9031-6962 Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA Search for more papers by this author Author Information Hyunju Ryu1, Xiao-Xin Sun1, Yingxiao Chen1, Yanping Li1, Xiaoyan Wang1, Roselyn S Dai1, Hong-Ming Zhu1, John Klimek2,3, Larry David2,3, Lev M Fedorov4, Yoshiaki Azuma5, Rosalie C Sears1 and Mu-Shui Dai *,1 1Department of Molecular & Medical Genetics, Oregon Health & Science University, Portland, OR, USA 2Department of Chemical Physiology & Biochemistry, School of Medicine, Oregon Health & Science University, Portland, OR, USA 3OHSU Proteomics Shared Resource, Oregon Health & Science University, Portland, OR, USA 4OHSU Transgenic Mouse Models Shared Resource, Oregon Health & Science University, Portland, OR, USA 5Department of Molecular Biosciences, University of Kansas, Lawrence, KS, USA *Corresponding author: Tel: +1 503 494 9917; Fax: +1 503 494 6886; E-mail: [email protected] EMBO Reports (2021)22:e50684https://doi.org/10.15252/embr.202050684 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions Figures & Info Abstract SUMOylation plays a crucial role in regulating diverse cellular processes including ribosome biogenesis. Proteomic analyses and experimental evidence showed that a number of nucleolar proteins involved in ribosome biogenesis are modified by SUMO. However, how these proteins are SUMOylated in cells is less understood. Here, we report that USP36, a nucleolar deubiquitinating enzyme (DUB), promotes nucleolar SUMOylation. Overexpression of USP36 enhances nucleolar SUMOylation, whereas its knockdown or genetic deletion reduces the levels of SUMOylation. USP36 interacts with SUMO2 and Ubc9 and directly mediates SUMOylation in cells and in vitro. We show that USP36 promotes the SUMOylation of the small nucleolar ribonucleoprotein (snoRNP) components Nop58 and Nhp2 in cells and in vitro and their binding to snoRNAs. It also promotes the SUMOylation of snoRNP components Nop56 and DKC1. Functionally, we show that knockdown of USP36 markedly impairs rRNA processing and translation. Thus, USP36 promotes snoRNP group SUMOylation and is critical for ribosome biogenesis and protein translation. SYNOPSIS The deubiquitinase USP36 also promotes protein SUMOylation and is essential for ribosome biogenesis and protein translation. USP36 promotes nucleolar protein SUMOylation. USP36 is the first identified DUB/SUMO promoting dual function protein. USP36 promotes snoRNA protein group SUMOylation. USP36 is critical for rRNA processing, translation and cell growth. Introduction SUMOylation, a posttranslational modification of proteins by small ubiquitin-like modifiers (SUMOs), plays a crucial role in the regulation of diverse cellular processes, including transcription, chromatin dynamics, genome maintenance, DNA repair, RNA splicing and processing, cell cycle control, and metabolism and is essential for normal cell growth and animal development (Bergink & Jentsch, 2009; Finkbeiner et al, 2011; Jentsch & Psakhye, 2013; Chymkowitch et al, 2015; Nuro-Gyina & Parvin, 2015; Sarangi & Zhao, 2015; Zhao, 2018). Mammals express three main SUMO isoforms: SUMO2 and SUMO3 are 97% identical (referred to as SUMO2/3) and both are 45% identical to SUMO1. SUMOylation is ATP-dependent and occurs through sequential reactions involving a heterodimeric SUMO-activating enzyme SAE1/SAE2 (E1), a single SUMO-conjugating enzyme Ubc9 (E2) and one of a few SUMO ligases (E3) (Muller et al, 2001; Geiss-Friedlander & Melchior, 2007; Jentsch & Psakhye, 2013). Ubc9 transfers SUMO to substrate acceptor lysine (Lys, K) residues, which is facilitated by SUMO E3s, via an isopeptide linkage (Muller et al, 2001; Geiss-Friedlander & Melchior, 2007; Jentsch & Psakhye, 2013). The SUMO acceptor Lys is often present within a conserved ΨKxE motif, where Ψ is a large hydrophobic amino acid and x is any amino acid (Rodriguez et al, 2001; Sampson et al, 2001; Geiss-Friedlander & Melchior, 2007; Gareau & Lima, 2010). While there are over 600 ubiquitin E3s in human (Deshaies & Joazeiro, 2009; Metzger et al, 2012), only a small number of bona fide SUMO E3 have been identified, including the SP-RING family members PIAS1, PIAS3, PIASxα, PIASxβ, PIASy (Kahyo et al, 2001; Sachdev et al, 2001; Nakagawa & Yokosawa, 2002; Nishida & Yasuda, 2002; Schmidt & Muller, 2002), and Nse2 (Andrews et al, 2005; Potts & Yu, 2005; Berkholz et al, 2014), the nuclear pore Ran Binding Protein 2 (RanBP2) (Pichler et al, 2002; Reverter & Lima, 2005), and the ZNF451 family proteins (ZNF451-1, ZNF451-2, ZNF451-3) (Cappadocia et al, 2015; Eisenhardt et al, 2015; Varejao et al, 2020). In addition, a few other proteins have been shown to possess SUMO E3 activity such as Pc2, TRIM28, and SLX4 (Kagey et al, 2003; Ivanov et al, 2007; Guervilly et al, 2015). SUMO modification is also highly dynamic and reversible. Removal of SUMO from substrates (deSUMOylation) is catalyzed by a group of deSUMOylating enzymes or SUMO proteases, including SENP1-3 and SENP5-7, DESI-1, DESI-2, and USPL1 (Hickey et al, 2012; Kunz et al, 2018). SUMOylation also plays an important role in ribosome biogenesis, a multi-step cellular process for making the ribosome in eukaryotes, requiring synthesis of ribosomal RNA (rRNA) and ribosomal proteins, rRNA processing, the assembly of the mature ribosome subunits in the nucleolus and transport into the cytoplasm, as well as the participation of many accessory factors (Rodnina & Wintermeyer, 2009; Kressler et al, 2017; Tomecki et al, 2017). Impairment of ribosome biogenesis is associated with a group of diseases called ribosomopathies whereas aberrant over-activation of ribosome biogenesis is tightly linked to human cancers (Mills & Green, 2017; Pelletier et al, 2018). Therefore, it is crucial to understand how ribosome biogenesis is properly regulated during normal cell homeostasis. It was first found that many yeast ribosome biogenesis factors and pre-ribosomal particles are modified by SUMO, which is critical for efficient ribosome biogenesis (Panse et al, 2006). In human, a number of nucleolar proteins involved in ribosome biogenesis, including nucleophosmin (NPM) (Tago et al, 2005; Liu et al, 2007), nucleolin (Zhang et al, 2015), and Las1L (Finkbeiner et al, 2011; Castle et al, 2012), are modified by SUMOylation. Proteomic studies found that ribosome biogenesis-related proteins are one of the major classes of SUMOylated proteins in cells (Vertegaal et al, 2004; Matafora et al, 2009; Amente et al, 2012; Hendriks et al, 2014). Likewise, deSUMOylation is also important for ribosome biogenesis. Removal of SUMO from NPM by SENP3 is critical for 28S rRNA maturation and the subsequent nucleolar export of the 60S pre-ribosomal subunit (Haindl et al, 2008). DeSUMOylation of Las1L by SENP3 is essential for ribosome particles partitioning from the nucleolus to cytoplasm (Finkbeiner et al, 2011; Castle et al, 2012). SUMOylation of PELP1, a component of the PELP1-TEX10-WDR18 complex critical for ribosome maturation, promotes the recruitment of MDN1, an AAA ATPase that removes assembly factors from the pre-60S particles (Chen et al, 2018), to pre-60S particles, while deSUMOylation is needed to release both MDN1 and PELP1 from pre-ribosomes (Raman et al, 2016). Depletion of SENP3 inhibits rRNA processing reminiscent of the NPM knockdown (Yun et al, 2008). Further, Nop58 and Nhp2, components of the small nucleolar ribonucleoprotein (snoRNP) complexes responsible for rRNA 2′-O-ribose methylation and pseudouridylation and critical for rRNA processing (Mannoor et al, 2012; Watkins & Bohnsack, 2012; Lui & Lowe, 2013; Dupuis-Sandoval et al, 2015), are also modified by SUMO (Matic et al, 2010; Westman et al, 2010). SUMOylation of Nop58 is required for its high-affinity binding to snoRNAs and the nucleolar localization of snoRNAs (Westman et al, 2010). Thus, balanced levels of SUMOylation and deSUMOylation of ribosome biogenesis factors are critical for ribosome biogenesis and maturation in the nucleolus. However, how these nucleolar proteins are SUMOylated is less understood. In this study, we identified that the ubiquitin-specific protease USP36 mediates nucleolar SUMOylation. We show that USP36 directly interacts with SUMO2 and Ubc9, directly promotes the SUMOylation of Nop58 and Nhp2 in cells and in vitro, and enhances their binding to snoRNAs. Functionally, USP36 is required for multiple steps of rRNA processing and translation. Thus, USP36, a nucleolar deubiquitinating enzyme (DUB) (Endo et al, 2009; Sun et al, 2015), also acts to promote nucleolar SUMOylation and is critical for ribosome biogenesis. Results USP36 promotes SUMOylation in cells, mainly in the nucleolus In our initial investigation of the SUMO regulation of USP36, we serendipitously observed that USP36 markedly increased the levels of protein SUMOylation in cells. We confirmed that overexpression of USP36 significantly promoted SUMO2/3-conjugation in H1299 (Fig 1A), HeLa, and U2OS (Fig 1B) as well as other tested cell lines such as 293 (Fig EV1A) and breast cancer T47D (Fig EV1B) cells. This effect is specific for USP36 as SUMOylation was not increased by overexpression of either JOSD3, another nucleolar DUB (Fig EV1C), or the nuclear DUB USP7 (Fig EV1D). We focused on SUMO2 in this study but USP36 also promotes SUMO1 conjugation (Fig EV1E and F). Nickel (Ni2+-NTA beads) purification methods demonstrated that overexpression of USP36 drastically enhanced SUMOylation by exogenous His-SUMO2 in cells (Fig EV1G). To test whether endogenous USP36 also regulates the levels of protein SUMOylation in cells, we performed knockdown experiments. As shown in Fig 1C, knockdown of USP36 by two individual shRNAs drastically reduced the levels of SUMOylated species in H1299 cells. Similarly, knockdown of USP36 also reduced the levels of SUMOylated species in HeLa (Fig 1D) and U2OS (Fig 1E) cells. Thus, endogenous USP36 also promotes SUMOylation and this effect is not cell-type specific. Figure 1. USP36 promotes SUMOylation in cells A, B. Overexpression of USP36 promotes SUMOylation in cells. Whole cell lysates (WCL) from H1299 (A), U2OS and HeLa (B) cells transfected with control empty vector or Flag-USP36 were assayed by IB. C–E. Knockdown of USP36 reduces SUMOylation in cells. H1299 (C), HeLa (D), and U2OS (E) cells infected with scrambled (scr) or USP36 shRNA encoding lentiviruses were assayed by IB. F. USP36 promotes nucleolar SUMOylation. H1299 cells transfected with control or V5-USP36 were fractionated to the cytoplasm (Cyto), nucleoplasm (Np), and nucleolus (No) fractions and assayed by IB. Data information: SUMO2/3 conjugates are indicated in all the top panels. Source data are available online for this figure. Source Data for Figure 1 [embr202050684-sup-0003-SDataFig1.pdf] Download figure Download PowerPoint Click here to expand this figure. Figure EV1. USP36 promotes SUMOylation in cells A., B. USP36 promotes SUMOylation in cells. 293 (A) and T47D (B) cells transfected with control or Flag-USP36 plasmid were assayed by IB for SUMOylated proteins. C, D. USP36 promotion of SUMOylation is specific. H1299 cells transfected with increasing amounts of Flag-USP36, Flag-JOSD3 (C), or Flag-USP7 (D) were assayed by IB. E, F. USP36 promotes SUMOylation by SUMO1. H1299 cells (E) or HeLa and U2OS cells (F) transfected with Flag-USP36 were assayed by IB using anti-SUMO1 antibody. G. USP36 promotes SUMOylation by exogenous SUMO2. H1299 cells transfected with His-SUMO2 and/or Flag-USP36 were subjected to Ni2+-NTA agarose beads pulldown (PD) followed by IB using anti-SUMO2/3 antibody (top panel). WCL were also directly assayed by IB using indicated antibodies (bottom panel). H. Nucleolar localization of USP36. HeLa cells transfected with Flag-USP36 were stained with anti-Flag (red) and anti-B23 (green) antibodies followed by DAPI (blue). I. USP36 promotes nucleolar SUMOylation. HeLa cells transfected with control or V5-USP36 were fractionated to the cytoplasm (Cyto), nucleoplasm (Np), and nucleolus (No) fractions, followed by IB to detect the indicated proteins. Source data are available online for this figure. Download figure Download PowerPoint USP36 is mainly localized in the nucleolus (Endo et al, 2009; Sun et al, 2015) (Fig EV1H). To determine whether USP36 promotes SUMOylation in the nucleolus, we performed cell fractionation assays. As shown in Fig 1F, USP36 markedly increased the protein SUMOylation in the nucleolus, but not the nucleoplasm, in H1299 cells. Similar results were also observed in HeLa cells (Fig EV1I). Thus, USP36 mainly promotes SUMOylation of the nucleolar proteins. The USP36 N-terminus is necessary for promoting SUMOylation To understand how USP36 regulates SUMOylation, we examined whether it controls the levels of the SUMO pathway enzymes, given that USP36 is a DUB (Endo et al, 2009; Sun et al, 2015). However, USP36 neither increased the levels of SUMO E1 (SAE1 and SAE2) or SUMO E2 (Ubc9; Fig EV2A), nor reduced the levels of the nuclear SUMO proteases SENP1, SENP2, SENP3, SENP5, SENP6, SENP7, and USPL1 (Fig EV2B and C), suggesting that USP36 does not promote SUMOylation by controlling the levels of the SUMO pathway enzymes. These results prompted us to test whether USP36 directly promotes SUMOylation. We then examined which region of USP36 involves in promoting SUMOylation. We transfected cells with different fragments of USP36 and found that the N-terminal USP domain containing region (amino acids 1–420), but not the middle and C-terminal regions, is essential for promoting SUMOylation by SUMO2/3 (Fig 2A), suggesting that this N-terminal region might harbor SUMO E3 activity (Fig 2B). Click here to expand this figure. Figure EV2. USP36 neither induces the levels of SUMO E1 or E2 nor increases the levels of SUMO proteases A. USP36 does not increase the levels of SUMO E1 and E2. H1299 cells transfected with control or increased amounts of Flag-USP36 were assayed by IB using antibodies against indicated proteins. B, C. USP36 does not reduce the levels of SUMO proteases. H1299 cells transfected with Flag-USP36 were assayed by IB using antibodies against indicated SENP proteins (B) or with Flag-USPL1 (C) in the presence or absence of V5-USP36 were assayed by IB. D. H1299 cells transfected with V5-Ub together with Flag-USP36 or the indicated mutants were assayed by IB to detect total ubiquitination. E. WT USP36 and the H382A mutant, but not the C131A mutant, promote SUMOylation in the nucleolus. H1299 cells transfected with control or the indicated Flag-USP36 plasmids were subjected to nucleolar isolation, followed by IB. SUMO2/3 conjugates are indicated in the top panel. F. The N-terminus of USP36 interacts with Ubc9 in vitro. Purified His-Ubc9 was incubated with GST, GST-USP361–420, GST-USP36421–800, or GST-USP36801–1121. Bound protein was detected by IB. GST and GST-fusion proteins were shown in the bottom panel by coomassie staining. G. The N-terminal USP36 binds to SUMO in cells. H1299 cells transfected with either control or Flag-USP361–420 plasmid were assayed by IP using anti-Flag, followed by IB with anti-SUMO2/3 antibodies. H. Requirement of the SUMO–Ubc9 backside interaction for USP36's SUMO E3 activity. Recombinant T7-PARP1 protein (0.1 μM) was incubated with SUMO E1 (50 nM, Boston Biochem), Ubc9 (50 nM, WT or the F22A mutant), SUMO2 (4 μM, WT or the D63R mutant) in the presence of USP361–800 (50 nM) and/or ATP (2.5 mM) at 30°C for 5 h and then assayed by IB. I. USP36 interacts with Ubc9 in vitro. Purified His-Ubc9 (Wt or the F22A mutant) was incubated with GST or GST-USP361–800. Bound Ubc9 was detected by IB using anti-Ubc9 antibody. GST and GST-fusion proteins were detected by IB with anti-GST. Source data are available online for this figure. Download figure Download PowerPoint Figure 2. The N-terminus of USP36 is necessary for promoting SUMOylation H1299 cells were transfected with Flag-USP36 or the indicated deletion mutants and assayed by IB. FL, full length. Schematic diagram of USP36 indicating that the N-terminus of USP36, which contains the USP catalytic domain, is required for its SUMO E3 activity. Catalytic residues C131 and H382 are shown. H1299 cells transfected with WT USP36, the C131A, or H382A mutant were assayed by IB to detect SUMOylated proteins. H1299 cells transfected with WT USP36 or the indicated mutants and V5-SUMO2 were assayed by IB using anti-V5 antibody to detect total protein SUMOylation. H1299 cells transfected with WT USP36 or the indicated mutants and His-SUMO2 were subjected to Ni2+-NTA agarose beads pulldown (PD) followed by IB using anti-SUMO2/3 antibody. Source data are available online for this figure. Source Data for Figure 2 [embr202050684-sup-0004-SDataFig2.pdf] Download figure Download PowerPoint As this N-terminal region contains both DUB activity and the activity to promote SUMOylation, we next asked whether the SUMO promoting activity is related to the DUB activity. To this end, we examined the effect of catalytically inactive USP36 mutants (Figs 2B and EV2D) on SUMOylation. Interestingly, mutating the DUB catalytic Cys 131 to Ala (C131A) also abolished the activity of USP36 to promote SUMOylation by either endogenous SUMO (Fig 2C) or exogenously expressed SUMO2 (Fig 2D), whereas mutating the DUB catalytic proton acceptor His 382 to Ala (H382A) promotes SUMOylation as efficiently as wild-type (WT) USP36 (Fig 2C and D). Ni2+-NTA pull-down assays further confirmed that the H382A, but not the C131A, mutant enhanced SUMOylation by His-SUMO2 in cells (Fig 2E). Further, examination of SUMOylation using purified nucleoli also confirmed that the H382A, but not the C131A, mutant promotes SUMOylation of the nucleolar proteins (Fig EV2E). Thus, the SUMO promoting activity of USP36 does not depend on its DUB activity. These results also suggest that C131 may be structurally important for the proper conformation of USP36 to mediate SUMOylation. USP36 binds to SUMO and Ubc9 and mediates SUMOylation in vitro Next, we examined whether USP36 indeed possesses SUMO ligase activity. SUMO ligases bind to both Ubc9 and substrates to facilitate the transfer of SUMO to substrates and also bind to SUMO (Bernier-Villamor et al, 2002; Kagey et al, 2003; Yunus & Lima, 2009; Guervilly et al, 2015; Streich & Lima, 2016). Therefore, we first examined whether USP36 binds to Ubc9 and SUMO. Indeed, USP36 interacts with Ubc9 in cells as determined by co-immunoprecipitation (IP) assay (Fig 3A). Glutathione S-transferase (GST) pull-down assays by incubating recombinant Ubc9 with GST-fusion USP36 proteins showed that USP36 directly interacts with ubc9 in vitro (Fig 3B). The N-terminus (aa 1–420), but not the middle and C-terminal regions, of USP36 is also sufficient for binding to Ubc9 (Fig 3B and EV2F). To test whether USP36 interacts with SUMO, we performed co-IP assays. As shown in Fig 3C, Flag-USP36 binds to SUMO2/3-conjugated proteins in cells. Also, GST-USP36, but not GST alone, directly interacts with SUMO2, preferentially the polymeric SUMO2 chains, in vitro (Fig 3D). Again, the N-terminus (aa 1–420), but not the middle and C-terminal regions, binds to the SUMO2 chains in vitro (Fig 3D) and in cells (Fig EV2G). These results reveal that USP36 interacts with both Ubc9 and SUMO2 via its N-terminus. Figure 3. USP36 interacts with Ubc9 and SUMO and mediates SUMOylation in vitro USP36 interacts with Ubc9 in cells. H1299 cell transfected with the indicated plasmids was subjected to co-IP with anti-Flag antibody followed by IB. USP36 interacts with Ubc9 in vitro. Purified His-Ubc9 was incubated with GST, GST-USP361–800 or GST-USP361–420. Bound protein was detected by IB. GST and GST-fusion proteins were shown in the bottom panel by coomassie staining. USP36 interacts with SUMO in cells. H1299 cells transfected with either control or Flag-USP361–800 were assayed by co-IP using anti-Flag, followed by IB with anti-SUMO2/3 antibodies. USP36 interacts with SUMO2 chains in vitro. Recombinant SUMO2 chains were incubated with GST or indicated GST-USP36 fragments. Bound proteins were assayed by IB using anti-SUMO2/3 antibodies. Coomassie staining of the recombinant proteins was shown in bottom panel. USP36 SUMOylates PARP1 in vitro. Recombinant T7-PARP1 protein (0.1 μM) was incubated with SUMO E1 (30 nM), Ubc9 (50 nM), SUMO2 (4 μM) in the presence of USP361–800 (50 nM) and/or ATP (2.5 mM) at 30°C for 5 h and then assayed by IB. Source data are available online for this figure. Source Data for Figure 3 [embr202050684-sup-0005-SDataFig3.pdf] Download figure Download PowerPoint We next tested whether USP36 directly promotes SUMOylation in vitro using well-studied SUMO target PARP1 involved in cell cycle progression (Ryu et al, 2010a). As shown in Fig 3E, in vitro SUMOylation reaction using recombinant E1 (SAE1/SAE2), E2 (Ubc9), SUMO2, and ATP resulted in marginal SUMO conjugation of PARP1 (lane 2), consistent with the notion that SUMO E1 and E2 can mediate SUMOylation in vitro, although less efficiently (Bernier-Villamor et al, 2002; Geiss-Friedlander & Melchior, 2007; Ryu et al, 2010a; Ryu et al, 2010b). Notably, USP36 drastically increased PARP1 SUMOylation in vitro (compared lane 3 to lane 2, Fig 3E). These results suggest that USP36 may possess a novel SUMO E3 activity. As many SUMO E3s such as PIAS1 (Eisenhardt et al, 2019) and ZNF451 (Cappadocia et al, 2015; Eisenhardt et al, 2015; Koidl et al, 2016; Eisenhardt et al, 2019) family members require a SUMO interaction with the backside of Ubc9 for their SUMO E3 activity, we tested whether this interaction could also play a role in USP36's SUMO E3. As shown in Fig EV2H, either mutating Phe 22 of Ubc9 to Ala (Ubc9F22A), which weakens the SUMO–Ubc9 backside interaction (Capili & Lima, 2007), or mutating Asp 63 of SUMO2 to Arg (SUMO2D63R), which disrupts the SUMO–Ubc9 backside interaction (Knipscheer et al, 2007), markedly inhibited USP36 activity to promote PARP1 poly-SUMOylation. These results suggest that although USP36 directly interacts with Ubc9 or the F22A mutant (Fig EV2I), its full activity to promote poly-SUMOylation requires the SUMO–Ubc9 backside interaction, further suggesting that USP36 acts as a SUMO E3. USP36 promotes SUMOylation of Nop58 and Nhp2 in cells To identify proteins whose SUMOylation is increased by USP36, we performed proteomic analysis. SUMOylated proteins were purified from cells transfected with His-SUMO2 together with control or USP36 plasmid using Ni2+-NTA beads purification followed by mass spectrometry analysis. Among the SUMOylated proteins including USP36 itself (Dataset EV1), we found that the number of assigned MS2 spectra of Nop58 and a number of nucleolar proteins involved in ribosome biogenesis increased in the USP36 transfected sample, (Fig EV3A). Nop58 and Nhp2, components of the box C/D and box H/ACA snoRNP complexes, respectively, were previously shown to be SUMO substrates (Westman et al, 2010), yet a SUMO E3 mediating such SUMOylation has not been reported. Therefore, we tested whether USP36 promotes the SUMOylation of Nop58 and Nhp2. Indeed, in vivo SUMOylation assays using Ni2+-NTA pulldown showed that USP36 markedly promoted the SUMOylation of exogenous Nop58 (Fig EV3B) and Nhp2 (Fig EV3D) as well as endogenous Nop58 (Fig 4A) and Nhp2 (Fig 4B). Cell fractionation assays showed that USP36 drastically promoted the SUMOylation of both Nop58 and Nhp2 in the nucleolus (Fig 4C), consistent with its role in the nucleolar SUMOylation. We also confirmed that the two previously reported Lys residues, K467 and K487, in Nop58 and the single Lys 5 (K5) in Nhp2 were the SUMOylation sites. Mutation of the K467 and K487 to Arg in Nop58 (Nop582KR) (Fig EV3C) and K5 to R in Nhp2 (Nhp2K5R) (Fig EV3E) abrogated their SUMOylation in cells. Thus, both Nop58 and Nhp2 are the endogenous SUMO substrates of USP36. Of note, USP36 does not increase the SUMOylation of nucleolar protein NOLC1 (Fig EV3F), a previously identified SUMO substrate (Westman et al, 2010), suggesting that not all nucleolar proteins are SUMO targets of USP36. Consistently, knockdown of USP36 significantly reduced the levels of S
Referência(s)