1706
Chem. Res. Toxicol. 2004, 17, 1706-1715
Global Shifts in Protein Sumoylation in Response to Electrophile and Oxidative Stress Linda L. Manza,† Simona G. Codreanu,‡ Sheryl L. Stamer,‡ Darrin L. Smith,† K. Sam Wells,§ Richard L. Roberts,| and Daniel C. Liebler*,‡,⊥,# Department of Pharmacology and Toxicology, College of Pharmacy, The University of Arizona, Tucson, Arizona 85721, and Departments of Biochemistry, Molecular Physiology and Biophysics, Pathology, and Pharmacology, and Mass Spectrometry Research Center, Vanderbilt University School of Medicine, Nashville, Tennessee 37232-0146 Received August 20, 2004
Human small ubiquitin-like modifier (sumo) proteins include sumo-1 and the less studied, nearly identical sumo-2 and sumo-3 proteins. Whereas the structurally related ubiquitin molecule targets proteins for degradation, sumo provides a distinct, yet poorly understood regulatory signal. Protein sumoylation is sensitive to diverse cellular stresses, yet the targets of sumoylation in stress are unknown. We studied protein sumoylation in HEK293 cells exposed to hydrogen peroxide, alkylating agents, and the lipid oxidation-derived electrophile 4-hydroxynonenal, which is an ubiquitous product of lipid oxidation associated with oxidative stress. Confocal immunofluorescence microscopy indicated that in unstressed cells sumo-1 targeted nuclear proteins, whereas sumo-2/3 targeted proteins in both nuclei and cytoplasm. Western blot analyses revealed changes in sumo-1 and sumo-2/3 targeting patterns with stress. We used immunoaffinity chromatography to harvest sumo-associated proteins from HA-sumo-1and HA-sumo-3-expressing HEK293 cells both before and after treatment with 4-hydroxynonenal. Multidimensional liquid chromatography-tandem mass spectrometry analyses identified 54 HA-sumo-1-associated proteins and 38 HA-sumo-3-associated proteins. Major protein targets included RNA binding and processing proteins, transcription factors, metabolic enzymes, and cytoskeletal regulators. Treatment with 4-hydroxynonenal caused a near-complete redistribution of sumo-1 and sumo-3 to different protein targets, which included chaperones, antioxidant, and DNA damage signaling proteins. A 10-15% overlap of sumo-1 and sumo-3 targets before and after stress suggests that sumo proteins target distinct protein groups. The results suggest that reactive electrophiles not only directly modify proteins but also lead to indirect changes in endogenous protein modifications that regulate protein functions.
Introduction Sumo (small ubiquitin-like modifier) proteins display similarities to ubiquitin in both the structure and the biochemistry of their activation and attachment to target proteins (1-3). Sumo proteins are expressed in all eukaryotes. The human genome encodes at least four sumo proteins, which include the three best-characterized family members [sumo-1 (SMT-3C), sumo-2 (SMT-3A), and sumo-3 (SMT-3B)] and a recently discovered fourth sumo protein (4). Sumo-1 displays approximately 44% sequence identity to sumo-2 and sumo-3, which have largely identical (95%) sequences (5). Sumo-1 modification has been most extensively studied, and over 50 sumo-1 targets have been identified to date (1, 6). These * To whom correspondence should be addressed. † Department of Pharmacology and Toxicology, The University of Arizona. ‡ Department of Biochemistry, Vanderbilt University School of Medicine. § Department of Molecular Physiology and Biophysics, Vanderbilt University School of Medicine. | Department of Pathology, Vanderbilt University School of Medicine. ⊥ Department of Pharmacology, Vanderbilt University School of Medicine. # Mass Spectrometry Research Center, Vanderbilt University School of Medicine.
include transcription factors and their cofactors and regulators, nuclear body proteins, nuclear pore complex proteins, DNA repair proteins, and viral proteins (1). In contrast, only a handful of sumo-2/3 targets are known (7-10). Sumoylation of target proteins is rapidly reversed by isopeptidases, which hydrolyze the sumo C-terminal amide bond to acceptor Lys residues (1, 11). Sumo modification occurs preferentially at ΨKxE sequences, where Ψ is a hydrophobic amino acid residue (I, L, V, or F) (12). Unlike sumo-1, sumo-2, and sumo-3, both contain a sumoylation consensus motif and can form polysumo chains on protein substrates (13). Whether “mixed” polysumo chains containing different sumo proteins or whether multiple monosumoylations can occur on a single protein is unknown, as is the functional significance of polysumo chains (13). The functions of sumo differ from those of ubiquitin. Whereas polyubiquitylation via K-48-linked chains tags proteins for degradation by the 26S proteasome system (14), sumoylation does not. Sumo modification instead governs subnuclear localization and assembly of multiprotein complexes, particularly in nuclear bodies, nuclear pores, and other nuclear substructures (reviewed in refs 1-3). Interplay of sumo and ubiquitin provides biochemical switches for control of cell signaling and DNA repair
10.1021/tx049767l CCC: $27.50 © 2004 American Chemical Society Published on Web 11/23/2004
Sumoylation and Protein Damage
pathways. Sumoylation and ubiquitylation of IκBR occur on the same lysine residue, and the former stabilizes IκBR and prevents NFκB activation (15). Similarly, competing poly(K63)ubiquitylation vs monoubiquitylation or sumoylation of the same lysine residue in the DNA repair protein PCNA directs error-free replicative bypass vs error-prone translesion bypass (16, 17). An intriguing aspect of sumo proteins is the responsiveness of protein sumoylation to cellular stresses that involve protein modification or damage. Hinchey and Saitoh observed rapid and reversible accumulation of high molecular mass sumo-2/3-modified protein conjugates and comcommitant depletion of the free sumo-2/3 pools upon exposure to heat shock, hydrogen peroxide, and ethanol (18). Similar studies in Arabidopsis thaliana revealed the existence of multiple sumo proteins, including Arabidopsis sumo-1/2, which underwent rapid, reversible incorporation into high molecular mass protein conjugates in response to heat and cold shock, toxic metals, and oxidative stress (19). Stress-associated protein modifications frequently occur through the formation of oxidants and electrophiles derived from lipid peroxidation during oxidative stress (20). In addition, many xenobiotics induce oxidative stress or undergo metabolism to protein-modifying electrophiles (21, 22). Recent reports indicate that protein modification by reactive electrophiles affects protein folding and stabilizes misfolded protein forms. Conway et al. (23) demonstrated that oxidizable catecholamine-like molecules stabilize neurotoxic R-synuclein protofibrils. Stabilization of protofibrils by dopamine involved covalent modification of the R-synuclein protein. Cholesterol oxidation products (24) and lipid oxidation-derived aldehydes (24, 25) covalently modify amyloid β peptide and accelerate its amyloidogenesis in vitro. Ubiquitylation of proteins in response to protein-damaging stress results from the coordinated action of chaperone proteins and E3 ubiquitin ligases that recognize non-native protein structures characteristic of misfolded proteins (reviewed in ref 26). We hypothesize that similar mechanisms account for the observed changes in protein sumoylation in response to protein-damaging stresses. To test this hypothesis, we studied chemical stressinduced changes in protein sumoylation in HEK293 cells. Electrophiles and hydrogen peroxide all produced perturbations in the distribution of both sumo-1 and sumo2/3 protein conjugates. Multidimensional liquid chromatography-tandem mass spectrometry (LC-MS/MS)-based proteomic analysis of sumo protein conjugates identified sumoylated and sumo-associated proteins, including proteins involved in RNA processing and transport, genome maintenance, intermediary metabolism, and cytoskeletal regulation. Stress induced a near complete shift of both sumo-1 and sumo-2/3 to a different set of protein targets, including proteins that function as chaperones and components of cellular stress responses. The results suggest that protein damage by reactive electrophiles leads to additional changes in proteome regulation by altering the distribution of endogenous protein modifiers.
Experimental Procedures Reagents. Antibodies directed against sumo-1 (FL-101) were obtained from Santa Cruz (Santa Cruz, CA). Antibodies against sumo-3 were purchased from Zymed (South San Francisco, CA).
Chem. Res. Toxicol., Vol. 17, No. 12, 2004 1707 Hydrogen peroxide was from Sigma (St. Louis, MO). IodoacetylLC biotin (IAB), 1-biotinamido-4-(4′-[maleimidoethyl-cyclohexane]carboxamido)butane (biotin-BMCC), tris(carboxyethyl)phosphine, and dithiothreitol were from Pierce (Rockford, IL), and 4-hydroxy-2-nonenal (HNE) was from Cayman Chemicals (Ann Arbor, MI). Mammalian expression constructs for HAsumo-1 and HA-sumo-3 (pcDNA3-HA-sumo-1 and pcDNA3-HAsumo-3) (13) were generously provided by Prof. R. T. Hay (University of St. Andrews, United Kingdom). Cell Culture, Transfection, and Treatments. HEK293 cells were obtained frozen at low passage from Master Cell Bank cultures from GIBCO Life Technologies (Grand Island, NY). Cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum and antibiotic-antimycotic at 5 mL/L. Cells were trypsinized (0.05% trypsin) and seeded at a final density of 1.5 × 105 cells/mL into 75 cm2 flasks or 100 mm plates containing 15 mL of culture medium. Cells were incubated at 37 °C and 95% air/5% CO2 and reached confluency after approximately 4 days. Experiments were conducted on cells between passages 15 and 45 postrecovery from cryopreservation. Confluent cells in 100 mm dishes were transfected with either pcDNA3-HA-sumo-1, pcDNA3-HA-sumo-3, or pcDNA3-HA vectors using Lipofectamine 2000 reagent at a ratio of 1:3 DNA/ lipofectamine and then were serum-deprived for 6 h during the transfection process. Stable transfectants were selected in the presence of 800 mg/mL. Geneticin and single cell colonies were isolated to generate stable transfectants expressing HA-sumo proteins or HA-vector only control cells (HA-null cells). Confluent cells in 100 mm plates were washed with phosphatebuffered saline and treated with HNE, hydrogen peroxide, IAB, or BMCC at doses of 10, 100, and 250 µM or equal volumes of vehicle delivered in 4 mL of nonsupplemented DMEM without phenol red. The vehicle was ethanol at 0.1% total volume. To measure cytotoxicity, an aliquot of culture medium was taken for the assay of lactate dehydrogenase (LDH) leakage using an in vitro LDH-based toxicology assay kit (Sigma). Cell death was calculated as a percent of total LDH measured in untreated cells. Immunoblot Analyses. Cells were lysed in 10 mM Tris-HCl, pH 7.4, containing 1% Triton X-100 and protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 10 µg/mL leupeptin, 10 mM N-ethylmaleimide, 10 µg/mL aprotinin, 10 µg/mL soybean trypsin inhibitor, and 25 mM iodoacetamide). Cell lysate proteins were diluted 1:1 (v/v) with Laemmli buffer and separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis in 12% Tris HCl Ready-Gels (Bio-Rad, Hercules, CA). Resolved proteins then were electrophoretically transferred to polyvinylidene fluoride membranes, which were blocked with 5% milk in TBST buffer (20 mM Tris HCl, pH 7.5, 200 mM NaCl, and 0.1% Tween 20) and then probed with anti-sumo-1 or anti-sumo-3 at a 1:1000 dilution for 2 h at 25 °C. After treatment with secondary antibody [mouse anti-IgG for sumo-1 and anti-rabbit IgG for sumo-3 (Santa Cruz)] at 1:2000 dilution in 1% milk-TBST buffer, immunostained proteins were detected by enhanced chemiluminescence with Western blotting luminol reagent (Santa Cruz). Confocal Microscopy. HEK293 cells were cultured in polyL-lysine-coated chambers for 24-48 h and then treated with either vehicle (0.1% ethanol) or 250 µM HNE in DMEM for 30 min at 37 °C. Cells were then fixed with 0.6% formaldehyde in Hanks’ balanced saline solution (HBSS) for 5 min, washed with methanol/0.1% Triton X-100 for 2 min, and then washed with HBSS. The cells then were blocked with 1% milk in HBSS for 5 min and then incubated with primary antibody (anti-sumo-1, anti-sumo-3, or anti-HA) at 20 µg/mL for 3 h. The cells were then washed with HBSS and incubated with secondary antibody (Cy5 mouse anti-IgG for sumo-1, Cy3 mouse anti-IgG for sumo3, or HA) in 1% milk in HBSS for 1 h and then washed with HBSS. The chamber slides were mounted in Cytoseal (Stevens Scientific, Riverdale, NJ) and sealed under no. 1.5 cover slips. Images were acquired with an LSM510 confocal microscope
1708
Chem. Res. Toxicol., Vol. 17, No. 12, 2004
Manza et al.
using either Plan-Neofluar 40×/1.3 or Plan-Apochromat 63×/ 1.4 objectives (Carl Zeiss, Go¨ttingen, Germany). Cy3 fluorescence was excited at 543 nm, and emission was detected through an LP560 barrier filter; Cy5 fluorescence was excited at 633 nm, and emission was detected through an LP650 barrier filter. Affinity Capture and Digestion of HA-Sumo-Associated Proteins. HEK293 cell lysates were fractionated with a CelLytic NuCLEAR extraction kit (Sigma) to produce nuclear and cytosolic protein fractions for HA-sumo-1 cells, HA-sumo-3 cells, and HA-null cells. Aliquots of these fractions were applied to affinity columns containing rat monoclonal anti-HA affinity matrix (3F10) (Roche Diagnostics, Indianapolis, IN) at a concentration of 5 mg lysate protein/0.3 mL settled resin volume and incubated overnight at 4 °C. After the columns were washed with 20 bed volumes of wash buffer (20 mM Tris HCl, pH 7.4, 100 mM NaCl, 0.1 mM EDTA, and 0.05% Tween 20), the HAtagged proteins were eluted by washing with 2 mL of 100 mM glycine, pH 2. Ultrafree-MC low binding regenerated cellulose centrifugal spin filter devices with a 5000 MWCO were obtained from Millipore (Billerica, MA). Prior to sample addition, spin filters were rinsed with sequential 300 µL washes of methanol and distilled water by centrifugation through the filters at 4500g at 20 °C. Protein samples in 100 mM glycine, pH 2, were added to the upper chamber of the spin filter, and samples then were centrifuged at 4500g to pass the solution through the filter. The proteins on the filter were washed with 300 µL each of distilled water and 0.1 M ammonium bicarbonate followed each time by centrifugation at 4500g. The filtrates were discarded. The protein was resuspended in 100 µL of 0.1 M ammonium bicarbonate containing 4 mM tris(carboxyethyl)phosphine and 10 mM dithiothreitol and incubated at 50 °C for 15 min. Iodoacetamide was added to a final concentration of 20 mM for 15 min to convert thiols to carboxamidomethyl derivatives. Modified porcine sequencing grade trypsin (Promega, Madison, WI) then was added in a 1:50 protein:trypsin ratio, and the samples were incubated at 37 °C for 18-24 h. Tryptic peptides then were collected by centrifugation through the filter at 4500g, and the filtrate was acidified with 0.5 µL of concentrated formic acid for LC-MS/MS analysis. Multidimensional LC-MS/MS Analyses and Protein Identification. Strong cation exchange (SCX) fractionation of peptide mixtures was done by a modification described previously (27). Peptides were fractionated on a polysulfoethyl-A, 5 µm particle size, 100 mm × 2.1 mm column (Nest Group, Southboro, MA). Samples were loaded in 10 mM ammonium formate, pH 3.0/acetonitrile (75:25, v/v) for 5 min at 0.2 mL/ min followed by a 30 min linear gradient to 200 mM ammonium formate, pH 8.0/acetonitrile (75:25, v/v) with a 10 min wash at the final mobile phase composition. Fifteen 3 min fractions were collected. Fractions 1-3 and 14 and 15 were combined, and then, all fractions were evaporated in vacuo and resuspended in water/acetonitrile/formic acid (98:2:0.1, v/v/v) for LC-MS/MS analysis. LC-MS/MS analyses were done on a ThermoFinnigan LCQ Deca XP ion trap instrument equipped with a ThermoFinnigan microelectrospray source (Thermo Electron, San Jose, CA), an Agilent 1100 series HPLC pump (Agilent, Palo Alto, CA), and a Famos autosampler (Dionex, Sunnyvale, CA). LC-MS/MS analyses were done by reverse phase chromatography on a 11 cm fused silica capillary column (100 µm ID) packed with Monitor C-18 (5 µm) (Column Engineering, Ontario, CA) and eluted first with water/acetonitrile/formic acid (98:2:0.1, v/v/v) for 5 min. A linear gradient then increased acetonitrile to 60% by 45 min, to 80% by 47 min, and then held at this solvent composition for another 18 min. MS/MS spectra were acquired in data-dependent scanning mode with one full scan followed by one MS/MS scan on the most intense precursor with dynamic exclusion of previously selected precursors for a period of 3 min. MS/MS spectra were matched to database sequences with TurboSequest (Thermo Electron).
Figure 1. Effects of alkylating agents and hydrogen peroxide on protein sumoylation in HEK293 cells. Cell lysates were analyzed following treatment for 30 min with compounds: 1, MW standards; 2, DMEM control; 3, DMEM + 0.1% DMSO (biotin-BMCC and IAB vehicle control); 4, 10 µM biotin-BMCC; 5, 100 µM IAB; 6, 100 µM HNE; 7, 100 µM hydrogen peroxide; and 8, DMEM + 0.1% ethanol (HNE vehicle control). The DMEM control lane (2) was omitted in the sumo-3 gel on the right. Sumoylated proteins were detected with antibodies to sumo-1 and sumo-3 as indicated. The anti-sumo-3 antibody detects both sumo-2 and sumo-3. Sequest outputs were analyzed with a custom-designed software and database system called CHIPS (Complete Heirarchical Integration of Protein Searches), which enables filtering of Sequest output files based on Sequest output parameters for sequence-spectrum matches and other criteria. Sequencespectrum assignments were accepted based on the following filtering criteria: All peptide sequence assignments were required to result from fully tryptic cleavages of the corresponding proteins. Assignments for singly charged ions were accepted when an XCorr score was >2.0, and all putative matches were confirmed by visual inspection of the spectra. Assignments for doubly charged ions were accepted when an XCorr score was >3.0, and for XCorr scores between 2.0 and 3.0, all putative matches were confirmed by visual inspection. Assignments for triply charged ions were accepted when an XCorr score was >3.0, and all putative matches were confirmed by visual inspection. Peptide identifications for all 15 SCX fractions in each sample were recombined with CHIPS, and the lists of corresponding proteins were compared. Protein identifications (including single peptide assignments) generated from HA-null samples (nuclei or cytosol, each either vehicle- or HNE-treated) were subtracted from lists of identified proteins from the corresponding HA-sumo-1 and HA-sumo-3 samples. Peptide and protein assignments corresponding to nuclei and cytosolic fractions from each sample were then combined to generate final lists of HA-sumo-1- and HA-sumo-3-associated proteins, which were further edited to include only those proteins identified by at least two different peptide sequences (Table 2). Complete lists of accepted sequence-spectrum matches for each sample (HAsumo-1, HA-sumo-3, and HA-null; nuclei and cytosol; each vehicle- and HNE-treated, 12 groups total) are presented in the Supporting Information (Tables 1-12).
Results Perturbation of Protein Sumoylation by Electrophile and Oxidative Stress. Treatment of HEK293 cells with electrophiles and prooxidants induced changes in the molecular mass distribution of sumo protein conjugates as detected by western blot analyses of cell lysates (Figure 1). In unstressed cells, major sumo-1modified proteins are RanGAP-1 (6, 28), which migrates at approximately 90 kDa, and another unidentified protein at approximately 40 kDa (Figure 1). Sumo-2/3 targeted different proteins in unstressed cells (the anti-
Sumoylation and Protein Damage
Chem. Res. Toxicol., Vol. 17, No. 12, 2004 1709
Figure 2. Dose-dependent effect of HNE on protein sumoylation in HEK293 cells. Cell lysates were analyzed following treatment for 30 min with 100 µM HNE. Sumoylated proteins were detected with antibodies to sumo-1 and sumo-3 as indicated.
sumo-3 antibody detects both sumo-2 and sumo-3). Consistent with a previous report (18), unconjugated sumo-1 was not detected in these analyses, whereas unconjugated sumo-2/3 was detected. Treatment with the model thiol reactive electrophiles biotin-BMCC, IAB, and the prototypical lipid peroxidation product HNE all increased high molecular mass conjugates of both sumo-1 and sumo-2/3 (Figure 1). The oxidant hydrogen peroxide produced a qualitatively different effect on sumoylation; higher molecular mass sumo-1 modifications were increased, whereas sumo-2/3 modifications were decreased. Levels of unconjugated sumo-2/3 declined significantly with HNE, apparently reflecting the incorporation of sumo-2/3 into high molecular mass protein conjugates. The most dramatic effect on both sumo-1 and sumo-2/3 was caused by HNE, which produced a dose-dependent increase in the formation of higher molecular mass sumo-1 and sumo-2/3 conjugates (Figure 2). Generation and Characterization of HEK293 Cells Expressing HA-Sumo-1 and HA-Sumo-3. To capture sumo-protein conjugates for identification, we generated HEK293 cells that stably express either HA-sumo-1 or HA-sumo-3. (We did not generate and study HA-sumo2-transfected cells because we assumed for purposes of these studies that the nearly identical sumo-2 and sumo-3 proteins would have essentially identical distributions and protein targets. This assumption requires verification in future work.) To provide a control for proteomic analyses (see below), we also generated HAnull transfectants with a pCDNA3-HA vector lacking a sumo coding sequence. Confocal microscopy indicated that both sumo-1 and sumo-2/3 were primarily localized to the nucleus in nontransfected HEK293 cells (Figure 3). As reported previously (18), both sumo-1 and sumo2/3 exhibited intense staining in nuclear bodies. In contrast to sumo-1, sumo-2/3 also exhibits detectable cytoplasmic staining, both in nontransfected cells and in HA-sumo-3-transfected cells. In HA-sumo-1- and HAsumo-3-transfected 293 cells, the HA-sumo proteins displayed essentially identical localization to their native counterparts (Figure 3). Western blot analyses of sumo proteins with anti sumo-1 and anti-sumo-2/3 antibodies indicated that overall levels of sumo proteins were comparable in transfected vs nontransfected cells (data not shown). Treatment of HA-sumo-1- and HA-sumo-3transfected cells with HNE produced shifts in HA-sumo protein conjugate patterns on Western blots similar to
Figure 3. Confocal immunofluorescence microscopy of untransfected HEK293 cells probed with anti-sumo-1 (A) and antisumo-3 (B), HA-sumo-1-transfected cells probed with anti-HA (C), and HA-sumo-3-transfected cells probed with anti-HA (D). Table 1. Protein Identifications in HA-Null, HA-Sumo-1, and HA-Sumo-3 Data Setsa
a
data set
protein IDsa
HA-null HA-null, HNE HA-sumo-1 HA-sumo-1, HNE HA-sumo-3 HA-sumo-3, HNE
345 377 507 254 461 356
Includes single peptide identifications.
those observed in untransfected cells (data not shown). These results collectively indicate that HA-sumo-1 and HA-sumo-3 proteins expressed in stably transfected 293 cells appear to represent the distribution and properties of the native sumo proteins. Identification of Sumo Protein Targets. HAsumo-1 and HA-sumo-3 protein conjugates were isolated from 293 cells by capture with an immobilized monoclonal antibody directed against the HA sequence tag (YPYDVPDYA). To improve the efficiency of HA-tagged protein capture, nuclear and cytosolic protein fractions were prepared from cell lysates prior to anti-HA capture. After washing to remove unbound proteins, bound proteins were eluted with 100 mM glycine, pH 2. Proteins eluted from the anti-HA beads include (i) HA-sumoprotein conjugates, (ii) proteins that are not sumoylated yet are tightly bound to sumoylated proteins, and (iii) nonspecifically bound proteins. To distinguish the latter group, we performed parallel analyses on nuclear and cytosolic fractions from “HA-null” cells transfected with pCDNA3-HA without a sumo coding sequence. Numbers of sequence-spectrum matches for HA-null samples were comparable to those in HA-sumo-1 and HA-sumo-3 samples (Table 1). Thus, a significant fraction of the proteins identified in HA-sumo-1 and HA-sumo-3 data sets also were present in the corresponding HA-null data sets. All proteins identified in these HA-null analyses thus were subtracted from protein identification lists for HA-sumo-1- and HA-sumo-3-associated proteins. Finally, lists of proteins identified in the nuclear and cytosolic fractions were combined to generate lists of HA-sumo1-associated proteins (Table 2) and HA-sumo-3-associated
1710
Chem. Res. Toxicol., Vol. 17, No. 12, 2004
Manza et al.
Table 2. HA-Sumo-1-Associated Proteins from Vehicle- and HNE-Treated HEK293 Cells
protein
accession no.
no. of peptides
cons. sequence
phosphoribosylaminoimidazole carboxylase GTP-binding nuclear protein RAN (TC4) karyopherin (importin) β 1 leucine-rich protein cystatin B 26S proteasome non-ATPase regulatory subunit 4 60S ribosomal protein L12 60S ribosomal protein L23a DEAD box polypeptide 17 isoform p82
HA-sumo-1, vehicle-treated gi|2507169 2 none gi|38455427 2 none gi|12408675 2 none gi|7387740 3 none gi|2780868 2 none gi|119717 2 none gi|24119203 2 99LKDE gi|2118336 2 247FKME gi|19913406 5 32AKKE, 1270VKVE gi|4506753 2 none gi|4503841 4 none gi|4507841 2 none gi|21361181 2 35LKKE gi|7512448 2 129IKSE gi|114549 4 none gi|189238 3 none gi|13786849 2 12LKEE, 57LKGE gi|26006838 2 none gi|21361368 2 none gi|4507521 4 351FKKE, 522LKKE gi|24415404 2 18AKNE, 3462VKSE, 4101LKVE, 4720IKGE, 5207LKPE gi|5453539 3 260LKSE gi|5453555 2 none gi|19923142 2 none gi|284289 3 484LKAE gi|4503117 2 none gi|2134662 2 131LKKE gi|4506597 2 none gi|17105394 2 none gi|38201710 2 none
DEAD/H box-5 (RNA helicase, 68kD)
gi|4758138
5
none
DEAD-box protein 3 (helicase-like protein 2)
gi|3023628
4
903VKQE, 1105VKQE
dyskerin
gi|4503337
2
fus-like protein
gi|7440064
2
38IKPE, 412 WKQE, 423 AKKE 336LKGE
hnRNPA3
gi|1710627
2
none
hnRNPF
gi|4826760
2
none
hnRNP homologue JKTBP
gi|7446333
4
182GKCE
hnRNPR
gi|5031755
5
365GKLE
HLA-B associated transcript-1; DEAD-box protein
gi|4758112
3
52LKPE
polypyrimidine tract binding protein, isoform a
gi|4506243
4
none
hnRNPA1
gi|22055112
2
none
splicing factor 3b, subunit 1
gi|6912654
2
806IKTE
splicing factor U2AF large chain
gi|107723
3
none
U5 snRNP-associated 102 kDa protein
gi|24212088
2
471AKLE, 682VKLE
U5 snRNP specific protein, 116 kDa
gi|4759280
2
none
progesterone binding protein receptor of activated protein kinase C 1 (RACK1) prothymosin R EBNA-2 coactivator (100 kDa) prohibitin thymopoietin, isoforms β/γ transcription factor BTF3 transcription factor NF-AT E1B protein, large T-antigen
gi|5729875 gi|5174447 gi|625274 gi|7657431 gi|4505773 gi|1174690 gi|115143 gi|1082855 gi|119030
2 2 2 2 2 2 2 3 6
135LKDE 270LKQE none none none 16LKSE, 400VKSE none none 103VKRE
peroxiredoxin 2 chaperonin containing TCP1, subunit 4 (δ) prefoldin 2 histone H2B.l fimbrin ezrin (p81) (cytovillin) tropomyosin R DNA topoisomerase II β DNA topoisomerase II R TATA binding protein interacting protein thyroid autoantigen ubiquinol-cytochrome c reductase core protein I ATPase, Na+/K+ transporting, R 1 γ-adaptin homologue DKFZp564D066.1 ATPB_HUMAN ATP synthase β chain glucose phosphate isomerase muscle LDH M chain phosphoglycerate mutase pyrroline-5-carboxylate synthetase transketolase MDN1, midasin homologue
function antioxidant defense chaperone chaperone chromatin structure cytoskeletal regulation cytoskeletal regulation cytoskeletal regulation DNA repair and maintenance DNA repair and maintenance DNA repair and maintenance DNA repair and maintenance electron transport ion transporter membrane/protein trafficking metabolism metabolism metabolism metabolism metabolism metabolism nuclear/unknown nucleic acid synthesis nucleocytoplasmic transport nucleocytoplasmic transport nucleocytoplasmic transport protease inhibitor protein degradation ribosomal protein ribosomal protein RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport signal transduction signal transduction transcriptional regulation transcriptional regulation transcriptional regulation transcriptional regulation transcriptional regulation transcriptional regulation viral protein
Sumoylation and Protein Damage
Chem. Res. Toxicol., Vol. 17, No. 12, 2004 1711
Table 2 (Continued) protein
accession no.
no. of peptides
cons. sequence
Cu/Zn superoxide dismutase calnexin heat shock 70 kDa protein binding protein (hop) oxygen-regulated protein precursor T-complex protein 1, β (TCP-1-β) H3 histone family, member T cofilin 1 (nonmuscle) plastin 3 electron transfer flavoprotein, R polypeptide Rab11B Rab1B adenylate kinase 2 isoform b crystallin, ζ ornithine aminotransferase, OAT voltage-dependent anion channel 1
HA-sumo-1, HNE gi|2982080 2 gi|2134858 2 gi|19923193 2 gi|5453832 4 gi|5453603 2 gi|4504299 2 gi|5031635 2 gi|7549809 3 gi|4503607 2 gi|4758986 2 gi|13569962 2 gi|7524346 2 gi|13236495 2 gi|1168056 2 gi|4507879 2
none 102GKWE 49AKSE 622LKEE none none 131IKHE none 125AKLE none none none none none none
ubiquitin-conjugating enzyme E2L 3 hnRNPR
gi|4507789 gi|5031755
2 2
none 365GKLE
hnRNPA2
gi|255313
2
none
hnRNPA1
gi|22055112
2
none
prohibitin eukaryotic translation initiation factor 5A
gi|4505773 gi|4503545
4 3
none none
HA-sumo-3, vehicle-treated gi|2507169 none gi|625305 4 46LKEE, 354FKKE, 1144LKTE smooth muscle myosin heavy chain 11, isoform SM1 gi|13124879 2 50IKEE, 361FKKE, 1043LKKE, 1151LKTE, 1509LKAE spectrin R chain, nonerythroid gi|88622 2 none transgelin 2 gi|4507357 2 none chromodomain helicase DNA binding protein 4 gi|4557453 2 710VKYE, 1227FKDE, 1303IKQE, 1564IKIE, 1571LKEE, 1765FKGE cytochrome c gi|11128019 2 none aldolase C, fructose-bisphosphatase gi|4885063 2 none ATP synthase β chain, mitochondrial gi|114549 2 none ATP synthase, R subunit, isoform 1, cardiac muscle gi|4757810 2 none fatty acid synthase (EC 2.3.1.85) gi|7433799 2 none glucose phosphate isomerase gi|189238 3 none long-chain fatty acid coenzyme A ligase 3 gi|4758330 2 none muscle LDH M chain gi|13786849 2 12LKEE, 57LKGE phosphoglycerate mutase gi|26006838 3 none transketolase gi|4507521 3 351FKKE, 522LKKE P32, acidic mitochondrial matrix protein gi|4930073 2 17IKEE GTP-binding nuclear protein RAN (TC4) gi|5453555 3 none 40S ribosomal protein S18 gi|6755368 2 none 60S ribosomal protein L18 gi|4506607 2 none 60S ribosomal protein L23a gi|17105394 3 none hnRNPF gi|4826760 3 none
peroxiredoxin 2 myosin heavy chain nonmuscle form A
hnRNP homologue JKTBP
gi|7446333
2
none
hnRNPR
gi|5031755
2
12LKEE
HLA-B associated transcript-1; DEAD-box protein
gi|4758112
2
52LKPE
hnRNP-E2
gi|6707736
2
none
PTB-associated splicing factor
gi|4826998
4
337IKLE
splicing factor, arginine/serine-rich 2
gi|6755478
U1 small nuclear ribonucleoprotein
gi|13635663
2
none
U5 snRNP specific protein, 116 kDa
gi|4759280
2
none
14-3-3 protein τ adenylyl cyclase-associated protein protein phosphatase 1G
gi|5803227 gi|5453595 gi|4505999
2 2 2
none 80LKLE none
none
function antioxidant defense chaperone chaperone chaperone chaperone chromatin structure cytoskeletal regulation cytoskeletal regulation electron transport membrane/protein trafficking membrane/protein trafficking metabolism metabolism metabolism mitochondrial ion channel/ apoptosis regulation protein degradation RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport transcriptional regulation translational regulation antioxidant defense cytoskeletal regulation cytoskeletal regulation cytoskeletal regulation cytoskeletal regulation DNA repair and maintenance electron transport metabolism metabolism metabolism metabolism metabolism metabolism metabolism metabolism metabolism mitochondrial matrix nucleocytoplasmic transport ribosomal protein ribosomal protein ribosomal protein RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport signal transduction signal transduction signal transduction
1712
Chem. Res. Toxicol., Vol. 17, No. 12, 2004
Manza et al.
Table 2 (Continued) protein
accession no.
no. of peptides
cons. sequence
receptor of activated protein kinase C 1 (RACK1) prothymosin R transcription factor NF-AT 45K chain, human elongation factor 1-β (EF-1-β) E1B protein, large T-antigen
HA-sumo-3, vehicle-treated gi|5174447 3 270LKQE gi|625274 2 none gi|1082855 4 none gi|4503477 2 none gi|119030 4 103VKRE
Cu/Zn superoxide dismutase programmed cell death 5 voltage-dependent anion channel 1 calnexin oxygen-regulated protein myotrophin DNA-activated protein kinase, catalytic subunit
HA-sumo-3, HNE-treated gi|2982080 3 gi|4759224 2 gi|4507879 2 gi|2134858 2 gi|5453832 2 gi|21956645 2 gi|1362789 5
reticulocalbin 1
gi|4506455
2
none 62VKPE none 524VKEE 622LKEE none 1212LKEE, 2756IKSE, 3688FKVE none
dihydrolipoamide dehydrogenase pyrroline-5-carboxylate synthetase transketolase 40S ribosomal protein S12 40S ribosomal protein S6 60S ribosomal protein L15 ATP-dependent RNA helicase A
gi|4557525 gi|21361368 gi|4507521 gi|133742 gi|6677809 gi|13385036 gi|3915658
2 3 2 2 2 2 3
409LKEE none 351FKKE, 522LKKE none none none 119LKAE, 364LKNE;
hnRNPL
gi|4557645
2
232LKIE
nucleolar protein NOP5/NOP58
gi|7706254
2
466VKVE
DEK oncogene (DNA binding) prohibitin prothymosin R zinc finger protein 9
gi|4503249 gi|4505773 gi|625274 gi|4827071
2 2 2 2
none none none none
proteins (Table 2). All protein identifications reported in Tables 1 and 2 represent identifications of at least two different peptides. We refer to the identified proteins as “associated” with HA-sumo-1 and HA-sumo-3 because these analyses do not strictly establish that the identified proteins are covalently modified by sumo proteins. Analysis of untreated HA-sumo-1 transfectants revealed 54 sumo-1-associated proteins (Table 2). Of these, the largest group (29%) was associated with RNA binding, processing, and transport and included five hnRNP proteins, four DEAD-box RNA helicases, and seven proteins involved in RNA binding and mRNA splicing or presplicing. In addition, transcriptional regulators (11%) and proteins involved in DNA repair and maintenance (7%) and in nucleocytoplasmic transport (6%) comprised protein target groups previously associated with sumo-1 (3, 29, 30). Six intermediary metabolic enzymes (11%) also were found as sumo-1-associated proteins. Other protein functional classes represented included cytoskeletal regulation, chaperones, signal transduction, antioxidant defense, and membrane trafficking. Of the 54 proteins identified, only 23 (43%) display a sumo consensus motif. Analysis of HA-sumo-3 transfectants identified 38 sumo-3-associated proteins (Table 2). As with the sumo1-associated proteins, the largest target group (23%) contained proteins with RNA binding, processing, and transport functions and included four hnRNP proteins, one DEAD-box RNA helicase, and four proteins involved in RNA binding, mRNA splicing. The other largest sumo-3 target group included enzymes of intermediary metabolism (23%). Other protein functional classes identified in the HA-sumo-3 analyses included cytoskeletal
function signal transduction transcriptional regulation transcriptional regulation translational regulation viral protein antioxidant defense apoptosis regulation apoptosis regulation chaperone chaperone cytoskeletal regulation DNA repair and maintenance membrane/protein trafficking metabolism metabolism metabolism ribosomal protein ribosomal protein ribosomal protein RNA binding, processing, and transport RNA binding, processing, and transport RNA binding, processing, and transport transcriptional regulation transcriptional regulation transcriptional regulation transcriptional regulation
regulators, ribosomal proteins, signal transduction proteins, and transcriptional regulators. Of the 38 proteins identified, 12 (32%) display a sumo consensus motif. There was significant concordance between proteins identified as sumo-1-associated and sumo-3-associated (Figure 4). A total of 14 proteins appeared as both sumo1- and sumo-3-associated. Moreover, sumo-3 peptides were identified in analyses of proteins captured from HAsumo-1-transfected cells. Treatment of cells with 250 µM HNE for 30 min dramatically changed the protein targeting of both sumo-1 and sumo-3. In HNE-treated, HA-sumo-1-transfected cells, 21 sumo-1-associated proteins were identified (Table 2). Only three of these were identified in untreated cells (Figure 4). HA-sumo-1-associated proteins in HNEtreated cells were distributed over most of the same functional categories as in untreated cells, although fewer RNA-associated proteins were targeted. Two of the proteins targeted in both untreated and HNE-treated cells were hnRNPs. Most notably, HNE treatment in-
Figure 4. Summary of HA-sumo-1- and HA-sumo-3-associated proteins before and after HNE treatment. See text for discussion.
Sumoylation and Protein Damage
creased apparent sumo-1 targeting of chaperone proteins, which are induced in stresses. In HA-sumo-3-transfected cells treated with HNE, 21 proteins were associated with sumo-3. As in HA-sumo1-transfected cells, HNE treatment also induced an almost complete redistribution of sumo equivalents; only one protein (transketolase) was identified as an HAsumo-3-associated protein both before and after HNE treatment (Table 2). Distribution among functional categories of HA-sumo-3-associated proteins in HNE treated cells was similar to that observed for HA-sumo-1-associated proteins after HNE. However, comparison of the lists of HA-sumo-1- and HA-sumo-3-associated proteins after HNE treatment indicates only four proteins common to both groups (Figure 4).
Discussion We undertook these studies in order to better describe the functional role of sumo proteins in cells. Although several interesting effects of protein sumoylation have been observed (1-3), sumoylation does not mimic the best-defined function of ubiquitinsthe tagging of proteins for degradation. However, previous work suggested that the distribution of sumo proteins was responsive to stress (18, 19). Our interest in the effects of covalent protein modification by reactive oxidants and electrophiles led us to examine the effect of these stresses on sumo proteins. While we were completing our work, three other groups reported MS-based proteomics approaches to identify targets for both sumo-1 (6, 30) and for sumo-2/3 (31). Together, these studies have greatly expanded the list of known sumo targets. However, we also have characterized protein sumoylation in the context of oxidative and alkylation stress. Our results indicate that this posttranslational modification is dramatically affected by stress and that electrophile adduction leads to additional changes in proteome regulation by sumo proteins. Our identification of sumo-associated proteins was based on the capture of HA-tagged sumo proteins and their conjugates and binding partners on bead-linked anti-HA monoclonal antibodies. This approach inevitably results in both specific and nonspecific capture of proteins. To correct for nonspecific protein background, we analyzed proteins captured in an identical workup from HA-null cells. These analyses yielded almost as many protein identifications as did those from HA-sumo-1 and HA-sumo-3 cells (Table 1). The occurrence of a high background in proteome analyses of this type has been observed in other studies (32, 33) and underscores the need for appropriate controls. This high background also means that considerable MS instrument time was “wasted” on MS/MS analyses of peptides from the contaminating proteins and thus may have reduced detection of specifically captured proteins. This is most likely why we did not observe sumo-1-derived peptides in our analyses. Tryptic digestion of sumo-1 would yield only nine peptides of 6-25 amino acids in length, and detection of these may have been consistently precluded by coelution of other more abundant peptides. We did observe sumo-3 peptides in the HA-null control samples, which is consistent with its apparent high abundance (18). Our protein identification criteria required at least two peptide identifications for each protein identification. This led us to reject other potential sumo-associated
Chem. Res. Toxicol., Vol. 17, No. 12, 2004 1713
protein identifications, including the prototypical sumo-1 target RanGAP, for which only one peptide was detected (Supporting Information, Table 5). Our exclusion of proteins identified in HA-null controls also may have led to the rejection of abundant proteins as possible sumo targets (see below). Despite our use of the HA-null controls, our LC-MS/MS analyses inevitably resulted in detection of peptides in the HA-sumo-1 and HA-sumo-3 cells that were nonspecifically captured yet were not detected in the HA-null controls. This reflects the somewhat stochastic nature of ion selection for MS/MS in datadependent analyses for all but the most abundant peptides (34, 35). Our results suggest that proteins involved in RNA binding, processing, and transport constitute a major class of sumo substrates. We identified six hnRNP proteins, four DEAD-box RNA helicases, and several other RNA binding proteins as sumo-associated proteins. Vassileva et al. reported that that hnRNP C and M can undergo sumoylation by a reconstituted sumoylation system in vitro (29). hnRNP proteins are relatively abundant species, and peptides corresponding to hnRNP A1, B1, C, G, H2, L, and U proteins were detected in HAnull cells, thus excluding them as possible sumo-associated proteins in our analytical scheme. This exclusion here should not necessarily rule out these proteins as potential sumo targets, as may be confirmed by additional studies. Our results, together with previous work indicating that sumo-1 modifies proteins of the nuclear pore complex, suggests that sumoylation plays an important role for sumo in the regulation of RNA processing and nucleocytoplasmic transport. Another novel observation from our studies is the identification of enzymes of intermediary metabolism as sumo targets. The sumo-associated enzymes that we found were LDH and phosphoglycerate mutase, which function in glycolysis, and glucose phosphate isomerase and transketolase, which operate in the pentose phosphate pathway. Of these apparent sumo targets, only two (LDH and transketolase) have sumo consensus motifs. The appearance of the mitochondrial enzyme ATP synthase β chain in our analyses probably reflects mitochondrial contamination of the cytosolic and nuclear fractions, as several abundant mitochondrial proteins were detected in most of the sample sets. Another interesting sumo-associated protein subgroup consists of proteins associated with cytoskeletal structures. Of the seven proteins in this group, three have sumo consensus motifs. Particularly notable in this respect is the identification of two different myosin heavy chains as sumo-3-associated proteins. Given that the numbers of peptides identified are roughly proportional to the abundance of a protein in the sample (36), the identification of six different myosin heavy chain peptides suggests that myosin is among the most abundant sumo3-associated proteins. Both of the myosin heavy chain proteins identified in our analyses contain multiple sumo consensus motifs. The apparent modification of myosin may also provide an explanation for the detection of extranuclear sumo-3 immunoreactivity in HEK293 cells (Figure 3). Oxidative and alkylation stress produced a dramatic shift in the patterns of sumo-associated proteins in HEK293 cells (Figures 1-3). Our results indicate that both the sumo-1 and the sumo-2/3 pools are highly responsive to stress. Western blot analyses indicated that
1714
Chem. Res. Toxicol., Vol. 17, No. 12, 2004
the shifts in protein sumoylation patterns with the different agents studies were similar. Most striking is the near complete redistribution of sumo proteins to different targets in response to stress. Only three of 21 apparent sumo-1 targets and one of 21 apparent sumo-3 targets were associated with sumo proteins in unstressed cells. Two other characteristics of this stress-induced shift are notable. First, HNE caused a net decrease in both sumo-1- (45%) and sumo-3-associated proteins (45%). Second, about 15% of the sumo-associated proteins identified in unstressed cells were associated with both sumo-1 and sumo-3, whereas 10% were both sumo-1- and sumo-3-associated after HNE. This latter observation raises the intriguing possibility that sumo-1 and sumo2/3 modifications occur on the same proteins. Sumo modification of proteins may involve either (i) single modifications with either sumo-1 or sumo-2/3, (ii) polysumoylation involving sumo-2/3 chains (possibly including a sumo-1 “cap” (13), or (iii) polysumoylation involving multiple monomeric modifications by sumo-1 and/or sumo-2/3. The available information does not allow us to distinguish between these types of modifications for the proteins identified as sumo-associated in our studies. However, it appears likely that sumo-1 and sumo-2/3 modifications can occur on the same substrate, given that we detected sumo-3 peptides in HA-sumo-1associated proteins. Western blot analyses indicate an increase in high molecular mass sumoylated proteins with oxidant and electrophile treatment (Figures 1 and 2). Our observation that a small number of sumoassociated proteins are detected after HNE may reflect a smaller number of targets that become multiply sumoylated. Further analyses of individual sumo target proteins before and after stress will be needed to distinguish proteins that are singly- and multiply sumoylated from those that are modified by polysumo chains. The dramatic shift in sumo-associated proteins with stress suggests that sumo proteins may help to regulate cellular stress responses. To explain stress responsive sumoylation changes, two broad hypotheses may be formulated that follow from known effects of sumo modification. First, sumo modification can compete with ubiquitylation and thus prevent protein degradation (1517). Thus, sumoylation may stabilize chaperones and other cellular protective enzymes detected as sumoassociated protein-stressed cells. The chaperones oxygenregulated protein and calnexin were associated with both sumo-1 and sumo-3 exclusively in stressed cells. Other sumo-1- or sumo-3-associated proteins detected in stress included the HSP 70 binding protein hop, the chaperonin component TCP-1 β, and calnexin (Table 2). Another protein that is both sumo-1- and sumo-3-associated in stress is Cu/Zn superoxide dismutase, which scavenges superoxide and prevents the formation of more reactive oxidants (37). Another interesting observation is the detection of DNA-activated protein kinase catalytic subunit as a sumo-3-associated protein after HNE treatment. This kinase is a member of the PI3K family and contributes to DNA repair and DNA damage responses (38, 39). The catalytic subunit contains three consensus sumoylation motifs and was represented in our data set by five different peptides, suggesting its relatively high abundance as a sumo-3-associated conjugate. Chaperones and antioxidant proteins are relatively abundant, and a number of these were detected in analyses of HA-null cells, thus excluding them from consideration as sumo
Manza et al.
targets. Thus, our approach may have significantly underestimated the sumoylation of chaperones and other stress response proteins. A second effect of sumoylation is to target protein substrates to specific subcellular organelles or substructures, such as nuclear bodies or the nuclear pore complex. For example, sumoylation of RanGAP governs its localization both to the nuclear pore complex and to the kinetochore and mitotic spindle complexes (28, 40, 41). The cytosolic protein prohibitin, which was found to be both sumo-1- and sumo-3-associated in stressed cells, is an inhibitor of E2F transcription factors involved in regulation of cell cycle progression (42, 43). Because prohibitin appears to have cytosolic localization, yet exerts nuclear effects, it seems reasonable to postulate that sumoylation facilitates nuclear localization of prohibitin and its effects on E2F-mediated transcription. The effects of sumo modfication in the context of environmental stresses are probably much more diverse than our present data indicate. However, our findings underscore the sensitivity of protein sumoylation to stress and the diversity of targets affected. Future work to characterize the underlying mechanisms will advance our understanding of the relationship between protein adduction, sumoylation, proteome management, and cellular responses to stress.
Acknowledgment. We thank Dr. Dean Ballard for helpful comments on our manuscript. This work was supported by NIH Grants ES11811, ES000267, and ES00694. S.C. was supported by NIH Training Grant ES007028. Supporting Information Available: Complete lists of accepted sequence spectrum matches for each sample. This material is available free of charge via the Internet at http:// pubs.acs.org.
References (1) Melchior, F., Schergaut, M., and Pichler, A. (2003) SUMO: Ligases, isopeptidases and nuclear pores. Trends Biochem. Sci. 28 (11), 612-618. (2) Muller, S., Ledl, A., and Schmidt, D. (2004) SUMO: A regulator of gene expression and genome integrity. Oncogene 23 (11), 19982008. (3) Seeler, J. S., and Dejean, A. (2003) Nuclear and unclear functions of SUMO. Nat. Rev. Mol. Cell Biol. 4 (9), 690-699. (4) Bohren, K. M., Nadkarni, V., Song, J. H., Gabbay, K. H., and Owerbach, D. (2004) A M55V polymorphism in a novel SUMO gene (SUMO-4) differentially activates heat shock transcription factors and is associated with susceptibility to type I diabetes mellitus. J. Biol. Chem. 279 (26), 27233-27238. (5) Su, H. L., and Li, S. S. (2002) Molecular features of human ubiquitin-like SUMO genes and their encoded proteins. Gene 296 (1-2), 65-73. (6) Zhao, Y., Kwon, S. W., Anselmo, A., Kaur, K., and White, M. A. (2004) Broad-spectrum Identification of cellular SUMO substrate proteins. J. Biol. Chem. (7) Rosendorff, A., Illanes, D., David, G., Lin, J., Kieff, E., and Johannsen, E. (2004) EBNA3C coactivation with EBNA2 requires a SUMO homology domain. J. Virol. 78 (1), 367-377. (8) Holmstrom, S., Van Antwerp, M. E., and Iniguez-Lluhi, J. A. (2003) Direct and distinguishable inhibitory roles for SUMO isoforms in the control of transcriptional synergy. Proc. Natl. Acad. Sci. U.S.A. 100 (26), 15758-15763. (9) Eaton, E. M., and Sealy, L. (2003) Modification of CCAAT/ enhancer-binding protein-beta by the small ubiquitin-like modifier (SUMO) family members, SUMO-2 and SUMO-3. J. Biol. Chem. 278 (35), 33416-33421. (10) Isik, S., Sano, K., Tsutsui, K., Seki, M., Enomoto, T., Saitoh, H., and Tsutsui, K. (2003) The SUMO pathway is required for selective degradation of DNA topoisomerase IIbeta induced by a catalytic inhibitor ICRF-193(1). FEBS Lett. 546 (2-3), 374-378.
Sumoylation and Protein Damage (11) Hemelaar, J., Borodovsky, A., Kessler, B. M., Reverter, D., Cook, J., Kolli, N., Gan-Erdene, T., Wilkinson, K. D., Gill, G., Lima, C. D., Ploegh, H. L., and Ovaa, H. (2004) Specific and covalent targeting of conjugating and deconjugating enzymes of ubiquitinlike proteins. Mol. Cell. Biol. 24 (1), 84-95. (12) Rodriguez, M. S., Dargemont, C., and Hay, R. T. (2001) SUMO-1 conjugation in vivo requires both a consensus modification motif and nuclear targeting. J. Biol. Chem. 276 (16), 12654-12659. (13) Tatham, M. H., Jaffray, E., Vaughan, O. A., Desterro, J. M., Botting, C., Naismith, J. H., and Hay, R. T. (2001) Polymeric chains of SUMO-2 and SUMO-3 are conjugated to protein substrates by SAE1/SAE2 and Ubc9. J. Biol. Chem. 276 (38), 35368-35374. (14) Glickman, M. H., and Ciechanover, A. (2002) The ubiquitinproteasome proteolytic pathway: Destruction for the sake of construction. Physiol. Rev. 82 (2), 373-428. (15) Desterro, J. M., Rodriguez, M. S., and Hay, R. T. (1998) SUMO-1 modification of IkappaBalpha inhibits NF-kappaB activation. Mol. Cell 2 (2), 233-239. (16) Hoege, C., Pfander, B., Moldovan, G. L., Pyrowolakis, G., and Jentsch, S. (2002) RAD6-dependent DNA repair is linked to modification of PCNA by ubiquitin and SUMO. Nature 419 (6903), 135-141. (17) Stelter, P., and Ulrich, H. D. (2003) Control of spontaneous and damage-induced mutagenesis by SUMO and ubiquitin conjugation. Nature 425 (6954), 188-191. (18) Saitoh, H., and Hinchey, J. (2000) Functional heterogeneity of small ubiquitin-related protein modifiers SUMO-1 versus SUMO2/3. J. Biol. Chem. 275 (9), 6252-6258. (19) Kurepa, J., Walker, J. M., Smalle, J., Gosink, M. M., Davis, S. J., Durham, T. L., Sung, D. Y., and Vierstra, R. D. (2003) The small ubiquitin-like modifier (SUMO) protein modification system in Arabidopsis. Accumulation of SUMO1 and -2 conjugates is increased by stress. J. Biol. Chem. 278 (9), 6862-6872. (20) Marnett, L. J., Riggins, J. N., and West, J. D. (2003) Endogenous generation of reactive oxidants and electrophiles and their reactions with DNA and protein. J. Clin. Invest. 111 (5), 583593. (21) Evans, D. C., Watt, A. P., Nicoll-Griffith, D. A., and Baillie, T. A. (2004) Drug-protein adducts: An industry perspective on minimizing the potential for drug bioactivation in drug discovery and development. Chem. Res. Toxicol. 17 (1), 3-16. (22) Hinson, J. A., Pumford, N. R., and Nelson, S. D. (1994) The role of metabolic activation in drug toxicity. Drug Metab. Rev. 26 (12), 395-412. (23) Conway, K. A., Rochet, J. C., Bieganski, R. M., and Lansbury, P. T., Jr. (2001) Kinetic stabilization of the alpha-synuclein protofibril by a dopamine-alpha-synuclein adduct. Science 294 (5545), 13461349. (24) Zhang, Q., Powers, E. T., Nieva, J., Huff, M. E., Dendle, M. A., Bieschke, J., Glabe, C. G., Eschenmoser, A., Wentworth, P., Jr., Lerner, R. A., and Kelly, J. W. (2004) Metabolite-initiated protein misfolding may trigger Alzheimer’s disease. Proc. Natl. Acad. Sci. U.S.A. 101 (14), 4752-4757. (25) Boutaud, O., Ou, J. J., Chaurand, P., Caprioli, R. M., Montine, T. J., and Oates, J. A. (2002) Prostaglandin H2 (PGH2) accelerates formation of amyloid beta1-42 oligomers. J. Neurochem. 82 (4), 1003-1006. (26) Cyr, D. M., Hohfeld, J., and Patterson, C. (2002) Protein quality control: U-box-containing E3 ubiquitin ligases join the fold. Trends Biochem. Sci. 27 (7), 368-375. (27) Adkins, J. N., Varnum, S. M., Auberry, K. J., Moore, R. J., Angell, N. H., Smith, R. D., Springer, D. L., and Pounds, J. G. (2002) Toward a human blood serum proteome: Analysis by multidimensional separation coupled with mass spectrometry. Mol. Cell. Proteomics 1 (12), 947-955.
Chem. Res. Toxicol., Vol. 17, No. 12, 2004 1715 (28) Matunis, M. J., Coutavas, E., and Blobel, G. (1996) A novel ubiquitin-like modification modulates the partitioning of the RanGTPase-activating protein RanGAP1 between the cytosol and the nuclear pore complex. J. Cell Biol. 135 (6, Part 1), 1457-1470. (29) Vassileva, M. T., and Matunis, M. J. (2004) SUMO modification of heterogeneous nuclear ribonucleoproteins. Mol. Cell. Biol. 24 (9), 3623-3632. (30) Li, T., Evdokimov, E., Shen, R. F., Chao, C. C., Tekle, E., Wang, T., Stadtman, E. R., Yang, D. C., and Chock, P. B. (2004) Sumoylation of heterogeneous nuclear ribonucleoproteins, zinc finger proteins, and nuclear pore complex proteins: A proteomic analysis. Proc. Natl. Acad. Sci. U.S.A. (31) Vertegaal, A. C., Ogg, S. C., Jaffray, E., Rodriguez, M. S., Hay, R. T., Andersen, J. S., Mann, M., and Lamond, A. I. (2004) A proteomic study of SUMO-2 target proteins. J. Biol. Chem. (32) Ranish, J. A., Yi, E. C., Leslie, D. M., Purvine, S. O., Goodlett, D. R., Eng, J., and Aebersold, R. (2003) The study of macromolecular complexes by quantitative proteomics. Nat. Genet. 33 (3), 349355. (33) Lindstrom, D. L., Squazzo, S. L., Muster, N., Burckin, T. A., Wachter, K. C., Emigh, C. A., McCleery, J. A., Yates, J. R., III, and Hartzog, G. A. (2003) Dual roles for Spt5 in pre-mRNA processing and transcription elongation revealed by identification of Spt5-associated proteins. Mol. Cell. Biol. 23 (4), 1368-1378. (34) Spahr, C. S., Davis, M. T., McGinley, M. D., Robinson, J. H., Bures, E. J., Beierle, J., Mort, J., Courchesne, P. L., Chen, K., Wahl, R. C., Yu, W., Luethy, R., and Patterson, S. D. (2001) Towards defining the urinary proteome using liquid chromatography-tandem mass spectrometry. I. Profiling an unfractionated tryptic digest. Proteomics 1 (1), 93-107. (35) Yi, E. C., Marelli, M., Lee, H., Purvine, S. O., Aebersold, R., Aitchison, J. D., and Goodlett, D. R. (2002) Approaching complete peroxisome characterization by gas-phase fractionation. Electrophoresis 23 (18), 3205-3216. (36) Gao, J., Opiteck, G. J., Friedrichs, M. S., Dongre, A. R., and Hefta, S. A. (2003) Changes in the protein expression of yeast as a function of carbon source. J. Proteome Res. 2 (6), 643-649. (37) Halliwell, B. (1996) Antioxidants in human health and disease. Annu. Rev. Nutr. 16, 33-50. (38) Hartley, K. O., Gell, D., Smith, G. C., Zhang, H., Divecha, N., Connelly, M. A., Admon, A., Lees-Miller, S. P., Anderson, C. W., and Jackson, S. P. (1995) DNA-dependent protein kinase catalytic subunit: A relative of phosphatidylinositol 3-kinase and the ataxia telangiectasia gene product. Cell 82 (5), 849-856. (39) Panta, G. R., Kaur, S., Cavin, L. G., Cortes, M. L., Mercurio, F., Lothstein, L., Sweatman, T. W., Israel, M., and Arsura, M. (2004) ATM and the catalytic subunit of DNA-dependent protein kinase activate NF-kappaB through a common MEK/extracellular signalregulated kinase/p90(rsk) signaling pathway in response to distinct forms of DNA damage. Mol. Cell. Biol. 24 (5), 1823-1835. (40) Matunis, M. J., Wu, J., and Blobel, G. (1998) SUMO-1 modification and its role in targeting the Ran GTPase-activating protein, RanGAP1, to the nuclear pore complex. J. Cell Biol. 140 (3), 499509. (41) Joseph, J., Tan, S. H., Karpova, T. S., McNally, J. G., and Dasso, M. (2002) SUMO-1 targets RanGAP1 to kinetochores and mitotic spindles. J. Cell Biol. 156 (4), 595-602. (42) Wang, S., Nath, N., Adlam, M., and Chellappan, S. (1999) Prohibitin, a potential tumor suppressor, interacts with RB and regulates E2F function. Oncogene 18 (23), 3501-3510. (43) Fusaro, G., Wang, S., and Chellappan, S. (2002) Differential regulation of Rb family proteins and prohibitin during camptothecininduced apoptosis. Oncogene 21 (29), 4539-4548.
TX049767L
