Overexpression of Proteasome �5 Subunit Increases the Amount ofAssembled Proteasome and Confers Ameliorated Response toOxidative Stress and Higher Survival Rates*□S
Received for publication, November 17, 2004, and in revised form, January 10, 2005Published, JBC Papers in Press, January 20, 2005, DOI 10.1074/jbc.M413007200
Niki Chondrogianni‡§, Christos Tzavelas‡¶, Alexander J. Pemberton�**, Ioannis P. Nezis‡,A. Jennifer Rivett�, and Efstathios S. Gonos‡ ‡‡
From the ‡National Hellenic Research Foundation, Institute of Biological Research and Biotechnology, 48 VasileosConstantinou Avenue, Athens 116 35, Greece and �Department of Biochemistry, University of Bristol, School of MedicalSciences, Bristol BS8 1TD, United Kingdom
The proteasome is the major cellular proteolytic ma-chinery responsible for the degradation of both normaland damaged proteins. Proteasomes play a fundamentalrole in retaining cellular homeostasis. Alterations of pro-teasome function have been recorded in various biologi-cal phenomena including aging. We have recently shownthat the decrease in proteasome activity in senescent hu-man fibroblasts relates to the down-regulation of �-typesubunits. In this study we have followed our preliminaryobservation by developing and further characterizing anumber of different human cell lines overexpressing the�5 subunit. Stable overexpression of the �5 subunit inWI38/T and HL60 cells resulted in elevated levels of other�-type subunits and increased levels of all three protea-some activities. Immunoprecipitation experiments haveshown increased levels of assembled proteasomes in sta-ble clones. Analysis by gel filtration has revealed that therecorded higher level of proteasome assembly is directlylinked to the efficient integration of “free” (not inte-grated) �-type subunits identified to accumulate in vec-tor-transfected cells. In support we have also found lowproteasome maturation protein levels in �5 transfectants,thus revealing an increased rate/level of proteasome as-sembly in these cells as opposed to vector-transfectedcells. Functional studies have shown that �5-overexpress-ing cell lines confer enhanced survival following treat-ment with various oxidants. Moreover, we demonstratethat this increased rate of survival is due to higher deg-radation rates following oxidative stress. Finally, becauseoxidation is considered to be a major factor that contrib-utes to aging and senescence, we have overexpressed the�5 subunit in primary IMR90 human fibroblasts and ob-served a delay of senescence by 4–5 population doublings.In summary, these data demonstrate the phenotypic ef-fects following genetic up-regulation of the proteasomeand provide insights toward a better understanding ofproteasome regulation.
Protein degradation is a major intracellular function, whichis responsible not only for housekeeping but also for the regu-lation of important cellular functions, such as homeostasis andsurvival. The proteasome is a major cellular non-lysosomalthreonine protease. Through its catabolic functions, it is impli-cated in many cellular processes including removal of abnor-mal, misfolded, denatured, or otherwise damaged proteins aswell as normal proteins (reviewed in Refs. 1–3). The 20S pro-teasome, a 700-kDa multisubunit enzyme complex, is a stack offour heptameric rings with the two outer �-type subunits ringsembracing two central head-to-head oriented rings containing�-type subunits. The inside rings give rise to the central cavityof the particle, where the catalytic sites of the complex areconfined. Three of the seven �-type subunits are proteolyticallyactive in the mature 20S proteasome, namely, the �1, �2, and �5
subunits. The proteasome has a broad specificity hydrolyzingpeptide bonds on the carboxyl site of hydrophobic (CT-L),1
acidic (PGPH), and basic (T-L) amino acids (reviewed in Refs.4–6). The 20S proteasome is also central to the ATP/ubiquitin-dependent intracellular protein degradation pathway, where itrepresents the proteolytic core of the 26S complex (20S corecapped on each side by 19S regulatory complexes; reviewed inRefs. 7 and 8).
Alterations in proteasome function have been found in manybiological processes including aging (reviewed in Refs. 3, 9, and10). We (11, 12) and others (13–19) have reported loss of pro-teasome function upon aging of several human tissues as wellas in senescent primary cultures. Our work in human embry-onic fibroblast cultures undergoing replicative senescence hasshown that the reduced levels of proteasomal activities duringthe process are accompanied by lower proteasome content andprotein expression levels of some, but interestingly not all,proteasome subunits. Specifically, we have found that loss ofproteasome function is due to lower levels of �-type subunits(the “rate-limiting” subunits), whereas �-type subunits are inexcess as “free” subunits in senescent cells (12). Finally, thefundamental link between cellular senescence and proteasomefunction is further supported by our recent study (20), in whichwe demonstrate that when the proteasome is partially inhib-ited in young primary fibroblast cultures, a senescence-likephenotype is triggered.
* This work was supported in part by European Union FOOD/FP-6“Zincage” Grant FOOD-CT-2003-506850 (to E. S. G.). The costs of pub-lication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked “advertisement”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
□S The on-line version of this article (available at http://www.jbc.org)contains Supplemental Fig. 1.
§ Recipient of a Ph.D. fellowship from the Bodosaki Foundation.¶ Recipient of a postdoctoral fellowship from Greek State’s Scholar-
ship Foundation (IKY).** Funded by a Cancer Research-UK studentship.‡‡ To whom correspondence should be addressed. Tel.: 30-210-
Data in the literature are very limited regarding the activa-tion/up-regulation of the proteasome (reviewed in Refs. 7 and8). Gaczynska et al. (21) have succeeded in the enhancement ofhydrophobic (CT-L) and basic (T-L) proteasome activities fol-lowing transfection of lymphoblasts and HeLa cells with the �5i
subunit, along with an increase of basic activity (T-L) followingtransfection of the same cell lines with the �1i subunit. Thesame group has also shown the stimulation of acidic activity(PGPH) after overexpression of �1 subunit in HeLa cells (22).Our preliminary efforts to “activate” the proteasome revealedthat stable overexpression of �1 or �5 subunits results in in-creased proteasome activities and enhanced cellular survivalfollowing treatment with proteasome inhibitors or H2O2 (12).Furthermore, and in accordance with the earlier results of Rockand co-workers (22), we have observed a co-regulation of �1 and�5 subunits (12). Apart from these pilot proteasome subunittransfection studies, few additional data exist regarding pro-teasome activation. Davies and co-workers (23) have shownthat poly(ADP-ribose) polymerase binds to nuclear protea-somes during oxidative stress by H2O2, leading to increasedproteasomal activities. More recently, Ustrell et al. (24) haveidentified a novel nuclear protein, PA200, that activates thenuclear proteasome following �-irradiation, thus indirectly in-volving the proteasome in DNA repair. Finally, Friguet andco-workers have described the ability of a lipid algae extract(Phaeodactylum tricornutum) to stimulate 20S proteasomepeptidase activities following keratinocyte UVA and UVB irra-diation.2 In summary, although these preliminary observationsindicate that the proteasome can be activated through otherpathways than the standard 19S and 11S complexes, there isstill limited knowledge regarding the molecular mechanisms ofsuch activation.
Data have been emerging recently with regard to the regu-lation of proteasome assembly in mammals (reviewed in Ref.25). Proteasome biogenesis is a precisely ordered multistepevent involving the biosynthesis of all subunits, their assembly,and maturation processes. POMP (or human/mouse Ump1 orproteassemblin; Refs. 26–28) is a factor that has been charac-terized as a human homologue of the yeast proteasome matu-ration factor Ump1 (29). Ump1 has been identified as an ac-cessory protein, a short-lived chaperone, that is required andessential for correct maturation and normal proteasome assem-bly in yeast (29). POMP is respectively responsible for theproteasome maturation and biogenesis of mammalian 20S pro-teasome; it is a constituent of a mammalian proteasome assem-bly intermediate that is detected only in precursor, inactivefractions (26) and also becomes the first substrate of the ma-ture proteasome (27). It has been reported to interact with �1i,�1, �5, �6, and �7 subunits but not with �-type subunits (30).
In this study, we have investigated proteasome up-regula-tion by means of stably overexpressing the �5 subunit in dif-ferent cell lines. We show that this overexpression leads toincreased levels of assembled and functional proteasome. Thisregulation is linked with the efficient integration of “free” (notintegrated) �-type subunits identified to accumulate in vector-transfected cells. Additionally, we clearly demonstrate thatproteasome up-regulated cell lines confer enhanced survivalagainst various oxidants. Finally, we provide evidence thatoverexpression of the �5 subunit delays senescence in IMR90normal fibroblasts.
EXPERIMENTAL PROCEDURES
Reagents and Antibodies—LLVY-AMC, LLE-NA, LSTR-AMC, andMG132 as well as primary proteasomal antibodies against �4 (XAPC7, C6;
PW8120), �6 (C2; PW8100), �7 (C8; PW8110), �1 (Y, delta; PW8140), �2 (Z;PW8145), and �5 (X, MB1, �; PW8895) subunits were purchased fromAffiniti Research Products Ltd. Primary antibody against POMP was agenerous gift from Dr. E. Kruger (26). Primary antibody against �-actin(sc1616) and secondary antibodies were purchased from Santa Cruz Bio-technology Inc. Oxidized proteins were detected with anti-dinitrophenolantibody from the OxyBlotTM Protein Oxidation Detection Kit (Qbiogene).
Cell Lines and Culture Conditions—Human embryonic fibroblastsIMR90 and WI38/T (SV40 T Ag WI38 VA 13 cell line) were obtainedfrom the European Collection of Cell Cultures and maintained in Dul-becco’s modified Eagle’s medium (Invitrogen) supplemented with 10%(v/v) fetal bovine serum (Invitrogen), 100 units/ml penicillin, 100 �g/mlstreptomycin, 2 mM glutamine, and 1% nonessential amino acids (com-plete medium). HL60 cells were a kind gift of Dr. D. Rickwood andcultured at concentrations between 0.5 and 1 � 106 cells/ml in RPMI1640 medium (Invitrogen) supplemented with 10% (v/v) fetal bovineserum and BASH (7.36 mg/ml sodium pyruvate, 0.454 mg/ml hypoxan-thine, 4.18 mg/ml penicillin, 4.5 mg/ml streptomycin, and 4.53 �g/lvitamin B12). Cells were fed �16 h prior to the assay, and cell numberwas determined in duplicates using a Coulter Z2 counter (CoulterCorp.). Normal fibroblasts were subcultured when cells reached conflu-ence at a split ratio of 1:2 until they entered senescence.
Stable Transfections—An expression vector encoding for the full-length �5 subunit cDNA (plasmid pBJ1-neo.�5) was the generous gift ofDrs. K. Tanaka and K. Rock (22, 31). WI38/T cells were transfected witheither empty vector or �5 plasmid by using the electroporation method.In brief, 107 cells were mixed with 50 �g of plasmid DNA (plasmidpBJ1-neo.�5 or empty vector) and electroporated at 260 V, 960 micro-farads (Gene PulserTM; Bio-Rad). 2.1 � 105 IMR90 cells were trans-fected with either plasmid or empty vector by using the EffecteneTM
transfection reagent (Qiagen) according to the manufacturer’s instruc-tions. Transfected cells (IMR90 or WI38/T cells) were split 48 h aftertransfection and maintained in complete medium containing 400 �g/mlG418. Colonies of stable transfectants were isolated following 3–4weeks of selection and propagated into cell lines.
HL60 cells were transfected as originally described by Bildirici et al.(32). In brief, 4 � 105 cells were washed twice with immunoporationwashing medium (IPBiosciences) and then resuspended in immunopo-ration medium (Immunoporation Ltd.) containing antibody-coated Im-munofect beads (IPBiosciences) in a ratio of 20 beads/cell. 0.2 �g ofplasmid DNA (plasmid pBJ1-neo.�5 or empty vector) was added, andthen cells were incubated in an end-over-end mixer at 40 rpm for 6 h atroom temperature. Cells were separated from the beads using a mag-netic separator (IPBiosciences) (32, 33), centrifuged, and suspended in1 ml of complete medium. Three days after transfection, G418 wasadded to the culture medium at a concentration of 800 �g/ml. Stabletransfectants were clonally selected by serially diluting 103 cells in96-well dishes covered with a small layer of 1% high purity agarose(Invitrogen) containing 800 �g/ml G418.
Survival Assays—104 WI38/T or 2 � 106 HL60 cells stably trans-fected with empty vector or �5 plasmid were seeded in 6-well plates induplicates. Cells were then treated immediately (HL60) or allowed torecover for 24 h prior to treatment (WI38/T) with 10 or 20 �M tBHP, 100or 300 �M H2O2, 1% or 2% EtOH or being subjected to metal catalyzedoxidation (treatment with 0.1 mM FeCl3/0.5 mM ADP in the presence of25 mM L-ascorbic acid) for 2.5 h at 37 °C in fresh medium. Treatedcultures were washed thoroughly in PBS, maintained in complete me-dium for 7 days, and counted. Each experiment was performed at leastthree times. For immunoblot detection of carbonyl groups into proteins,105 cells were seeded in 6-well plates in normal medium and treatedwith 300 �M H2O2 for 30 min. Proteins were extracted immediatelyafter treatment and 24 h after treatment.
Immunofluorescence Antigen Staining and Confocal Laser ScanningMicroscope Analysis—For immunofluorescence labeling, cells grown oncoverslips were washed in ice-cold PBS and subsequently fixed with 4%freshly prepared paraformaldehyde in PBS followed by cell permeabi-lization with 0.2% Triton X-100 in PBS. Immunolabeling of protea-somes was carried out using antibodies against the �7 and �5 subunits.The antibodies were diluted in PBS containing 0.1% Tween 20 and 3%bovine serum albumin (blocking buffer). The secondary anti-mouseIgG/fluorescein isothiocyanate-conjugated antibody and anti-rabbitIgG/TRITC-conjugated antibody were diluted 1:250 in blocking buffer.Images of the mounted coverslips were taken by using a Nikon PCM2000 confocal laser scanning microscope. Routine procedures, appliedas controls to demonstrate the specificity of the antibody used, were asfollows: (a) the usage of normal serum instead of the reactive antibody,and (b) omission of the first antibody. All controls appeared free of anyimmunofluorescence background.
2 C. Nizaed, B. Friguet, M. Moreau, A. L. Bulteau, and A. Saunois,patent PCT Application WO 02/080876.
Proteasome Peptidase Assays and Protein Determination—CT-L,PGPH, and T-L activities of the proteasome in crude extracts wereassayed with hydrolysis of the fluorogenic peptides LLVY-AMC, LLE-NA, and LSTR-AMC, respectively, for 30 min at 37 °C, as describedpreviously (11). Proteasome activity was determined as the differencebetween the total activity of crude extracts or fractions and the remain-ing activity in the presence of 20 �M MG132. Assays of 26S proteasomeswere carried out in 25 mM Tris/HCl buffer, pH 7.5, containing 5 mM ATP(34). Fluorescence was measured using a PerkinElmer Life Sciences650-40 fluorescence spectrophotometer. Protein concentrations weredetermined using the Bradford method with bovine serum albumin asa standard.
Immunoblot Analysis—Cells were harvested (for IMR90 and WI38/Tcell lines) or collected (for HL60 cell lines) at the indicated time points,lysed in non-reducing Laemmli buffer, and fractionated by SDS-PAGE(12% separating gel) according to standard procedures (35). After elec-trophoresis, proteins were transferred to nitrocellulose membranes forblotting with appropriate antibodies. Secondary antibodies conjugatedwith horseradish peroxidase and enhanced chemiluminescence wereused to detect the bound primary antibodies. Immunoblot detection ofcarbonyl groups into proteins was performed with OxyBlotTM ProteinOxidation Detection Kit (Qbiogene) according to the manufacturer’sinstructions. Protein loading was tested by stripping each membraneand reprobing it with a �-actin antibody.
Preparation of Cell Extracts and Separation of Proteasome Complexesby Gel Filtration—WI38/T and HL60 cells (vector- and �5-transfectedcells) were lysed in 20 mM Tris/HCl buffer, pH 7.5, containing 5 mM ATPand 0.2% Nonidet P-40. Extracts were centrifuged at 13,000 rpm for 10min at 4 °C, and then an equal amount of protein (1.5 mg) from emptyvector or �5 subunit transfectants was fractionated by gel filtrationusing an Amersham Biosciences Superose 6 HR10/30 fast protein liquidchromatography column equilibrated in 20 mM Tris/HCl buffer, pH 7.5,containing 10% glycerol, 5 mM ATP, and 100 mM NaCl (36). Fractionswere collected, and samples were analyzed by SDS-PAGE and immu-noblot analysis. Proteasome fractions were identified by quantifying theCT-L activity in duplicate experiments.
Immunoprecipitation of Proteasomes—Immunoprecipitated protea-somes were prepared as follows: cell monolayers of 80% confluentWI38/T vector- and �5-transfected cultures were washed and scrapedinto ice-cold PBS containing 10 mM phenylmethylsulfonyl fluoride and10 �g/ml aprotinin. 107 HL60 vector- and �5-transfected cells werecollected accordingly. For radioactive labeled immunoprecipitation,prior to washing and scraping/collection, cells were starved in methi-onine-deficient medium for 1 h, followed by pulse labeling with 100�Ci/ml [35S]methionine for 4 h in methionine-deficient medium. Col-lected cells were diluted directly in 20 mM Tris/HCl buffer, pH 7.5,containing 5 mM ATP, 10% glycerol, 0.2% Nonidet P-40, 10 mM phen-ylmethylsulfonyl fluoride, and 10 �g/ml aprotinin (lysis buffer). Cellextracts (equilibrated in lysis buffer) were then cleared by addingnormal mouse serum and protein A-agarose beads for 3 h at 4 °C.Meanwhile, protein A-agarose beads equilibrated in lysis buffer werecoupled with 1–2 �g of antibody against the �6 subunit for 3 h at 4 °Cwith constant rocking. The target antigen was then immunoprecipi-tated by adding the pre-cleared extracts in the pre-coupled antibodyagainst �6 subunit-protein A. Binding reactions were performed withconstant rocking at 4 °C overnight. Immunoprecipitated protein com-plexes were collected; washed four times in 50 mM Tris/HCl buffer, pH7.5, containing 5 mM ATP, 75 mM NaCl, 10% glycerol, and 0.2% TritonX-100 (washing buffer); and eluted from the agarose beads by boiling for5 min in non-reducing Laemmli buffer. Controls were used to demon-strate the specificity of the observed immunoprecipitations. Followingimmunoprecipitation of proteasomes, the number of counts due to in-corporated radioactive [35S]methionine in the immunoprecipitated pro-teasomes was determined using a Wallac 1409 DSA liquid scintillationcounter. Nonradioactive samples were processed for one-dimensionalSDS-PAGE as described.
Analysis of Protein Turnover—Quantification of degradation of met-abolically radiolabeled proteins in cell cultures was performed as fol-lows: cells were starved in methionine-deficient medium for 1 h, incu-bated with [35S]methionine in methionine-deficient medium for 4 h at37 °C, washed twice in PBS, and treated with oxidants (metal catalyzedoxidation and tBHP) for 30 min as described. Following treatment, cellswere washed thoroughly in PBS and maintained in normal medium forthe indicated time points up to 72 h. The degradation of metabolicallylabeled proteins was quantified by adding an equal volume of 20%trichloroacetic acid on 1 ml of culture supernatant for 30 min on ice.Scintillation counting of the acid-soluble counts in the supernatant wasperformed using a Wallac 1409 DSA liquid scintillation counter after
centrifugation for 10 min at 13,000 rpm.Statistical Analysis and Densitometry—Statistical calculations and
graphs were performed with Microsoft Excel software. Data were ana-lyzed by single-factor analysis of variance, and p value was used todetermine the level of significance (p � 0.05). All values were reportedas mean (the average of three independent experiments) � S.E., unlessotherwise indicated. Densitometric analysis was performed with GelAnalyzer Version 1.0.
RESULTS
Overexpression of the �5 Catalytic Subunit in WI38/T andHL60 Cells Leads to Increased Proteasomal Activities and In-creased Protein Expression Levels of Proteasome Subunits—Weaimed to reveal changes of the overall proteasome status andfunction in WI38/T and HL60 �5-transfected cells. Several sta-ble WI38/T and HL60 clones were selected and propagated intocell lines. These cell lines exhibited similar growth rates andmorphological characteristics as compared with the parentalcell lines, even after several months of cultivation. Two rep-resentative cell lines, WI38/T/�5.8 and HL60/�5.3, were cho-sen for additional studies. As shown in Fig. 1, all three majorproteasomal activities (CT-L, PGPH, and T-L) were found tobe significantly increased in both WI38/T and HL60 clones(p � 0.05) as compared with their control counterparts. Theactivities were increased from 1.3- to 2.1-fold, with the CT-Land PGPH activities being the most affected. Transfectionwith vector alone (pBJ1-neo) had no effect on proteasomeactivities.
Next we examined whether there is a quantitative differencein proteasome content in �5 transfectants. A detailed immuno-blot analysis of several representative 20S proteasomal sub-units was performed. As shown in Fig. 2, clones overexpressingthe �5 subunit also overexpressed the �1 and �2 subunits inboth the WI38/T and HL60 cell lines. In order to further verifythese data and investigate the proteasome localization and
FIG. 1. Specific proteasomal activities in �5-overexpressingcell lines. Percentage of CT-L, PGPH, and T-L activities in vector(CON)- and �5-transfected (A) WI38/T/�5.8 and (B) HL60/�5.3 cells. Useof a proteasome inhibitor (MG132) in control reactions ensured thespecificity of the enzymatic reactions. Mean value of activities was setat 100% in vector-transfected cells. Each column shows the average ofthree independent experiments, and error bars denote S.E. All threeproteasome activities were found increased in �5-overexpressing cells.
distribution, we immunolocalized representative proteasomesubunits in WI38/T/�5.8 and control cells. Several �- and �-typesubunits were examined. Examples of �5 and �7 subunits areshown in Fig. 3. An enhanced immunolocalization signal wasobserved in �5 transfectants. In all cases, antigens were foundto be distributed mainly in the nucleoplasm. A dispersed punc-tate pattern was also evident in the cytoplasm. There was nomajor qualitative change in the intracellular distribution pat-tern of the examined subunits between transfectants and con-trols, although we observed a trend of nuclear and perinuclearaccumulation in �5 transfectants. These differences were alsovisible when both images were superimposed to demonstratethe co-localization of the two examined subunits, as shown inFig. 3C. Thus, we concluded that overexpression of the �5
subunit results in increased proteasome activities, a phenom-
enon accompanied by elevated levels of other proteasome sub-units. Importantly, these intriguing findings are consistent indifferent cell lines.
Overexpression of the �5 Catalytic Subunit in WI38/T andHL60 Cells Leads to Increased Levels of Assembled Protea-some—We investigated whether the observed up-regulation ofproteasome subunits affects the overall amount of assembledproteasome in WI38/T and HL60 clones. Proteasome content inboth clones and control cells was initially investigated by im-munoprecipitation followed by immunoblot analysis. As shownin Fig. 4 (right panels), the expression levels of the �2, �4, �6,and �7 subunits were increased in the elution of the precipi-tated proteasome obtained by cell extracts from clones as com-pared with the respective elution of control cells. In contrast, asshown in Figs. 1 and 4 (left panels), in total cell lysates therewas no difference in the amount of �-type subunits betweenclones and control cells. Given the fact that the antibodyagainst the �6 subunit that was used for the immunoprecipi-tation experiments recognizes and precipitates the assembledproteasome (37), the increased levels of the examined subunitsdetected in the elution as opposed to the total cell extractsindicate an overall higher amount of assembled proteasomes inWI38/T and HL60 �5 cell lines. Quantification of the relativeproportion of each immunoprecipitated subunit betweenWI38/T transfectants and control cell lines revealed an �1.6-fold increase in �5.8 transfectants of the �2 (1.65-fold), �4
(1.60-fold), and �7 (1.55-fold) subunits, with the exception ofthe �6 subunit, which exhibited a lower 1.3-fold increase.This finding regarding �6 may relate to the fact that theimmunoprecipitating antibody used also recognizes free �6
subunit in control cells (see the data presented in Fig. 6).
FIG. 3. Proteasome subcellular distribution in WI38/T/�5 cells.Confocal micrographs demonstrating in situ localization of proteasomes invector (CON; left panels)- and �5-transfected WI38/T/�5.8 (right panels)cells. A, labeling with antibody against the �5 proteasome subunit; B,labeling with antibody against the �7 proteasome subunit; and C, �5 and�7 labeling overlapping localization signals. Nuclei (Nu) and cytoplasm(Cyt) are indicated by arrows. �5-Overexpressing WI38/T cells reveal anenhanced immunolocalization signal of the examined subunits.
FIG. 4. Proteasome immunoprecipitation followed by immu-noblot analysis of proteasome subunits in �5-overexpressingcell lines. Immunoblot analysis of whole extracts used for the immu-noprecipitation (left panels) and immunoprecipitated proteasomes(right panels) of representative subunits of 20S complex (�2, 30 kDa; �4,27.9 kDa; �6, 29.5 kDa; and �7, 28.4 kDa) in vector (CON)- and �5-transfected (A) WI38/T/�5.8 and (B) HL60/�5.3 cells. Immunoprecipita-tions were initiated by using the same amount of total crude extracts.All subunits tested were found to be increased in �5-overexpressingcells in the immunoprecipitated proteasomes as compared with thewhole extracts.
FIG. 2. Protein levels of proteasome subunits in �5-overex-pressing cell lines. Immunoblot analysis of representative subunits of20S complex, �-type subunits (�1, 25.3 kDa; �2, 30 kDa; and �5, 22.9kDa), and �-type subunits (�6, 29.5 kDa; and �7, 28.4 kDa) in vector(CON)- and �5-transfected (A) WI38/T/�5.8 and (B) HL60/�5.3 cells.Equal protein loading was verified by stripping the membranes andreprobing them with �-actin (43 kDa) antibody (bottom panels). Proteinexpression levels of catalytic subunits (�-type subunits) were elevatedin �5-overexpressing cells.
Similar results for all four subunits were obtained in HL60cells (data not shown). Results were further confirmed byemploying a two-dimensional gel electrophoresis analysis fol-lowing immunoprecipitation of assembled proteasomes (datanot shown). Thus, a higher level of proteasome assembly inclones is suggested.
We further investigated this increase of mature proteasomein clones by measuring the stability of newly synthesized as-sembled proteasome in WI38/T and HL60 �5 cells and controlcells for a period of 40 h. Cells were radiolabeled for 4 h with[35S]methionine, and proteasome was immunoprecipitated atdifferent time points after withdrawing [35S]methionine. Asseen in Fig. 5, in both WI38/T/�5.8 and HL60/�5.3 cells (graybars) the radiolabeled subunits were used directly after thelabeling procedure. In contrast, in control cells (black bars),there was a lag phase of 16 h (for WI38/T cells) and 24 h (forHL60 cells) before observing a decrease similar to the onerecorded in clones. In particular, in control cells we observed amaximum increase of the amount of radiolabeled/immunopre-cipitated proteasome of �1.8-fold in WI38/T cells and 1.3-fold inHL60 cells 4 and 16 h after removal of [35S]methionine, respec-tively. This intriguing finding indicates that the radiolabeledproteasome subunits are not totally used at once as they areproduced, but they are stored and used later, when cells aredeficient for these subunits. In contrast, this is not the case inWI38/T/�5.8 and HL60/�5.3 cell lines because the radiolabeledproteasome subunits are used directly upon their production.
We further examined the proteasome complexes and theassembled proteasomes in WI38/T and HL60 �5 clones andcontrols by gel filtration. In accordance with above-stated data,an increase in proteasome complexes was observed in WI38/T/�5.8 and HL60/�5.3 cell lines as shown by the exhibited CT-Lactivity (Fig. 6A for WI38/T cells and Fig. 7A for HL60 cells)and the immunodetection of representative 20S subunits (datanot shown). However, some significant differences were foundin the immunoblots of late fractions. Specifically, we have
detected enhanced levels of the �4, �6, and �7 proteasomesubunits in fractions 32–46 from control cells as compared withthe corresponding �5 cell lines (Fig. 6B for WI38/T cells andFig. 7B for HL60 cells). �-Type subunits were not found toaccumulate in these fractions in either cell line (data notshown). This observation indicates that �-type subunits maynot integrate efficiently in assembled proteasomes in controlcells and, moreover, that proteasomes may not mature in thecontrol cells as efficiently as they do in the �5 clones. A similarobservation was recorded in human embryonic fibroblast cul-tures undergoing senescence (12), in which an accumulation of“free” (not integrated) subunits was detected in the late frac-tions of senescent cell extracts but not in young cell extracts(see also “Discussion”).
Because POMP has been identified as a factor responsible forthe maturation and biogenesis of 20S proteasome, we investi-gated its levels in the different fractions of clones and controlcells. In accordance with the findings of Kruger and co-workers(26) (see also “Discussion”), POMP was accumulated in theinactive proteasome precursor fractions 27–31 of WI38/T andHL60 control cells (Figs. 6C and 7C, respectively). No protea-some activity was recorded in these fractions (see Figs. 6A and7A). In agreement with the previously reported data, low levelsof POMP were found in �5 clones, thus revealing an increasedrate/level of proteasome assembly in these cells.
Overexpression of the �5 Catalytic Subunit in WI38/T andHL60 Cells Leads to Better Response to Oxidative Stress andAmeliorated Survival—Having established that overexpressionof the �5 catalytic subunit results in higher proteasome content,next we addressed whether this “proteasome up-regulation” re-sults in functional differences. For these studies, we treated bothWI38/T/�5.8 and HL60/�5.3 cell lines and controls with severaloxidants, and we determined cell survival. Specifically, cell lineswere exposed to 10 or 20 �M tBHP, 100 or 300 �M H2O2, 1% or 2%EtOH or were subjected to metal catalyzed oxidation for 2.5 h,and their survival capacity was recorded following a recovery
FIG. 5. Immunoprecipitation of la-beled proteasomes from whole ex-tracts of �5-overexpressing cell lines.Cells were labeled with [35S]methioninefor 4 h, and proteasomes were immuno-precipitated at the indicated time points(0, 2, 4, 16, 24, and 40 h after labeling) byusing a mouse monoclonal antibodyagainst the �6 subunit (conditions forimmunoprecipitation of 26S complex) asdescribed under “Experimental Proce-dures.” Immunoprecipitations were initi-ated by using the same amount of totalcrude extracts from vector (CON)- and �5-transfected (A) WI38/T/�5.8 and (B)HL60/�5.3 cells. Mean value of recordedcounts of vector- and �5-transfected cellsat 0 h post-labeling was set at 100%.Values represent the average of two inde-pendent experiments for each time point.Newly synthesized labeled subunits areused directly in the �5-overexpressing celllines.
period of 7 days (Fig. 8). Both �5 cell lines exhibited highersurvival rates as compared with the corresponding control cells.Specifically, WI38/T/�5.8 cells exhibited higher survival rates of1.2–1.8-fold for EtOH, 1.8–2.0-fold for H2O2, 1.8–1.9-fold fortBHP, and 2.1-fold for metal-catalyzed oxidation as comparedwith control cells. Similarly, HL60/�5.3 cells exhibited highersurvival rates of 1.6–1.7-fold for EtOH, 1.4–1.5-fold for H2O2,1.8–1.9-fold for tBHP, and 2.3-fold for metal-catalyzed oxidationas compared with control cells. Analysis of oxidized proteins inHL60 cells following treatment with 300 �M H2O2 revealed lowerlevels of oxidized proteins in �5 transfectants both right aftertreatment and following a recovery period of 24 h (Fig. 9). Similarresults have been obtained in WI38/T cells (12). Thus, we con-clude that the differences in the proteolytic activities of �5-over-expressing cell lines can be translated into functional differencesof the proteasome because transfectants exhibit an increasedcapacity to cope better with various oxidants by decreasing theiroxidative load.
Finally, we examined the degradation rates of cellular pro-
teins following treatment with oxidants. WI38/T and HL60 �5
clones and control cells were labeled with [35S]methionine,treated with 10 �M tBHP, or subjected to metal-catalyzed oxi-dation, and effects on the degradation rates were monitored forup to 72 h after treatment. As shown in Fig. 10, WI38/T/�5.8and HL60/�5.3 cells constantly exhibited a higher rate of deg-radation compared with the corresponding rates of control cellsafter both treatments. These data further support our assump-tion that �5 clones respond efficiently to oxidative stress due tomore efficient proteolytic machinery.
Overexpression of the �5 Catalytic Subunit Delays Senescencein Human Primary IMR90 Fibroblasts—Normal human fibro-blasts undergo a limited number of divisions in culture, and theyprogressively reach a state of irreversible growth arrest, a proc-ess termed replicative senescence. Replicative senescence is ac-companied by gradual accumulation of oxidized/damaged pro-teins (reviewed in Ref. 38) as well as by loss of proteasomefunction (reviewed in Refs. 39 and 40). Thus we investigated theeffects on replicative life span of normal human fibroblasts fol-
FIG. 6. Gel filtration analysis of proteasome in WI38/T/�5.8 cell line. Cell extracts from vector (CON)- and �5-transfected WI38/T/�5.8 cellswere chromatographed on a Superose 6 HR10/30 fast protein liquid chromatography column equilibrated in 20 mM Tris/HCl buffer, pH 7.5,containing 10% glycerol, 5 mM ATP, and 100 mM NaCl. A, assays of 26S proteasomes for CT-L activity. Values represent the average of twoindependent experiments. B, late gel filtration fractions (fractions 32–46) were immunoblotted with antibodies against �-type subunits (�4, 27.9kDa; �6, 29.5 kDa; and �7, 28.4 kDa) as indicated. C, immunoblot analysis of POMP in fractions 24–31. Because proteasome activities are detectedin fractions 17–24 (A), precursor proteasomes are expected after fraction 25. “Free” (not integrated) �-type subunits and precursor proteasomecomplexes were more abundant in extracts from vector-transfected WI38/T cells.
lowing stable overexpression of the �5 subunit. Stable trans-fection experiments were attempted in both IMR90 and WI38primary cells, but few IMR90 clones were efficiently propa-gated. This is not surprising because the difficulties regard-ing efficient introduction of a given construct into primaryhuman cells are well documented (reviewed in Ref. 41). Thuswe have characterized all the IMR90 transfectants, and wehave kept them in culture until they reached senescence. Inaccordance with the previously reported data, all �5 cloneswere found to carry enhanced proteolytic activities and ex-pression levels of proteasome subunits (data not shown).More importantly, clones performed, on average, 4.43 � 0.95more population doublings as compared with control cells.These data further support our previously reported findingsthat overexpression of the �5 subunit ameliorates cell sur-vival, and additionally, they indicate that proteasome up-regulation in primary cells delays senescence.
DISCUSSION
In this article, we have shown that overexpression of the �5
proteasome subunit in different cell lines leads to enhanced
proteasome activities, increased protein expression levels ofthe proteasome subunits, and efficiently assembled protea-some. We have found that this increased amount of assem-bled proteasome is translated to more functional proteasomethat confers enhanced survival following treatment with ox-idants, primarily through a higher rate of degradation. Fi-nally, we provide evidence that proteasome up-regulation isdue, in part, to the recruitment of “free” (not integrated)�-type subunits.
In a previous work (12), pilot analysis has indicated enhancedproteasome function (i.e. elevated levels of CT-L and PGPH ac-tivities) after stable transfections of either the �1 or �5 catalyticsubunit in WI38/T cells. In this article and after screening a largenumber of stable clones, we show that all three major proteolyticactivities (CT-L, PGPH, and T-L) are enhanced not only inWI38/T immortalized fibroblasts but also in two other unrelatedcell lines, namely, HL60 (a leukemic cell line) and IMR90 (normalfibroblasts), following overexpression of the �5 subunit. This find-ing further strengthens the suggestion that proteasome can beup-regulated through additional molecular pathways, apart from
FIG. 7. Gel filtration analysis of proteasome in the HL60/�5.3 cell line. Cell extracts from vector (CON)- and �5-transfected HL60/�5.3 cellswere processed as described in Fig. 6. A, assays of 26S proteasomes for CT-L activity. Values represent the average of two independentexperiments. B, late gel filtration fractions (fractions 32–46) were immunoblotted with antibodies against �-type subunits (�4, 27.9 kDa; �6, 29.5kDa; and �7, 28.4 kDa) as indicated. C, immunoblot analysis of POMP in fractions 24–31. Because proteasome activities are detected in fractions19–26 (A), precursor proteasomes are expected in later fractions. “Free” (not integrated) �-type subunits and precursor proteasome complexes weremore abundant in extracts from vector-transfected HL60 cells.
its association with internal protein activators, such as 19S or11S complexes (reviewed in Refs. 7 and 8), poly(ADP-ribose)polymerase (23), or PA200 (24).
To understand the mechanism(s) of proteasome up-regula-tion, we have taken a detailed molecular and biochemical ap-proach to WI38/T and HL60 �5-overexpressing clones. Ouranalysis indicates that there is a common regulation betweenthe �-type subunits because overexpression of the �5 subunithas resulted in the overexpression of other �-type subunits. In
support of our assumption, previous studies have shown thatoverexpression of either the �1 or �5 subunit resulted in over-expression of the other subunit (12, 22). Furthermore, ourpreliminary analysis and our previously reported data demon-strate that stable overexpression of the �1 subunit in WI38/Tcells also results in elevated levels of all three proteasomalactivities, increased amounts of other subunits, and, finally,enhanced survival of the stable clones following treatment witheither proteasome inhibitors or oxidants (12).3 Based on thesedata, we speculate that overexpression of just one �-type sub-unit (�5 or �1, according to our results) is sufficient to increaseproteasome assembly and function. A potential auto-regulatorymechanism regarding proteasome has also been suggested byother studies. Davies and co-workers (reviewed in Ref. 2) indi-rectly implied that few proteasomal subunits may be regulatedin the same way because daily treatment with an �6 antisenseoligonucleotide severely depressed the intracellular levels ofseveral, but not all, proteasome subunits in both cultured liverepithelial cells and K562 human hemopoietic cells. In addition,a common transcriptional regulation of the proteasomal sub-units has been reported in yeast (42–44), in which 26 of the 32proteasomal subunit genes have been found to be preceded in
3 N. Chondrogianni and E. S. Gonos, unpublished observations.
FIG. 8. Survival rates of �5-overexpressing cell lines following treatments with oxidants. Number of vector (CON)- and �5-transfected (A)WI38/T/�5.8 and (B) HL60/�5.3 cells following a single treatment with 1% or 2% EtOH, 100 or 300 �M H2O2, 10 or 20 �M tBHP or after being subjectedto metal-catalyzed oxidation after a recovery period of 1 week. Each column shows the average of three independent experiments, and error bars denoteS.E. �5-Overexpressing cell lines proliferate significantly (p � 0.05) better than the vector-transfected cells.
FIG. 9. Levels of oxidized proteins in HL60/�5.3 cells. Analysis ofoxidized proteins by oxyblot (for conditions, see “Experimental Proce-dures”) in vector (CON)- and �5-transfected HL60/�5.3 cells following asingle treatment with 300 �M H2O2, either immediately after the treat-ment (no recovery) or after a recovery period of 24 h (24 recovery). Equalprotein loading was verified by the use of a �-actin (43 kDa) antibody(bottom panel). Molecular mass (in kDa) is shown to the left of the blot.Levels of oxidized proteins were decreased in �5-overexpressing cells.
their promoters by proteasome-associated control element(PACE). RPN4 has been found to be the factor that binds onthis element and transcriptionally activates these genes (42).However, no human homologue of RPN4 has been identifiedthus far. Additional support regarding auto-regulation of pro-teasome subunits comes from the work of Wojcik andDeMartino (45) in S2 cells derived from Drosophila. Silencingof different proteasomal genes, by the use of small interferingRNA technology, resulted in reduction of the mRNA level ofthe respective targeted subunit but, moreover, changes in themRNA levels of several other (but not all) non-targeted sub-units. More recently, Meiners et al. (46) have also suggestedan auto-regulatory feedback mechanism that allows the com-pensation of reduced proteasome activity in mammalian cellsafter exposure to proteasome inhibitors. Their results clearly
show that although a concerted regulation of proteasomegenes takes place after proteasome inhibition, all subunitsare not up-regulated to the same extent. These results are inagreement with our suggestion regarding a different regula-tion between different proteasome subunits.
Because the �-type subunits are co-regulated and conse-quently overexpressed following transfection of the �5 subunit,where does the cell find additional �-type subunits to producemore assembled proteasome? The gel filtration results clearlyshow that there are “free” (not integrated in active protea-somes) �-type subunits in the control cells as opposed to the �5
clones (see also below). The reported immunoprecipitation ex-periments show that when �5 subunit is overexpressed, thewhole proteasome is up-regulated and more efficiently assem-bled. This is in accordance with our previously reported dataregarding the assembly of the proteasome in early- and late-passage human embryonic fibroblasts (12), in which the �-typesubunits appear to be “rate-limiting” subunits in late-passagecells, whereas �-type subunits were also found in excess, being“free” (not integrated). Radiolabeled immunoprecipitation ex-periments in this study demonstrate a lag phase of 16–24 hregarding the use and integration of labeled, newly synthesizedsubunits into assembled proteasomes in control cells but not inthe �5 clones. We suggest that in clones in which the �-typesubunits are in excess, radiolabeled subunits can be used atonce because they can find their �-type partners for the assem-bly process. Thus, we hypothesize that the intracellular levelsof �-type subunits may determine the amount of assembledproteasome. This hypothesis is further strengthened by the factthat we have never detected any free �-type subunits in all celllines studied.
The gel filtration analysis also shows that there are moreinactive proteasome precursor fractions in the control cells asopposed to the �5 clones. It is known that there are assemblyintermediates (13S and 16S proteasome precursor complexes)in the formation of mammalian 20S proteasomes (Refs. 47–50;reviewed in Ref. 51) that are inactive. During proteasome frac-tionation, late fractions of control cells (in which no activitywas detected) reveal higher amounts of these complexes ascompared with the corresponding late fractions of the �5 clones.Furthermore, our data concerning POMP expression analysisstrengthen our assumption that in control cells there are moreprecursor and thus inactive complexes. In accordance with ourfindings, Witt et al. (26) have reported that in cell lines thataccumulate precursor proteasome complexes, POMP is stabi-lized and accumulated in the inactive proteasome precursorfractions. Finally, we have found more abundant �-type sub-units (�4, �6, and �7) in late fractions derived from control cellsthan in ones from �5 clones. Although little is known about theearly steps in the assembly of subunits in eukaryotes, humansubunit �7, when overexpressed in Escherichia coli, has beenshown to spontaneously form double ring-like structures (52).The �6 and �1 subunits are unable to form ring-like structureswhen expressed by themselves; however, they are incorporatedin such assemblies when they are co-expressed with �7 (53).Similar observations have also been reported for the �5 subunit(54). Moreover, Seemuller and co-workers (55) have character-ized the proteasome as a self-compartmentalizing protease,where proteasomal �-type subunits spontaneously self-assem-ble into seven-membered rings that serve as templates intowhich �-type subunits are incorporated. Mayer and co-workers(56) have recently shown that several �-type subunits arefound in low-density fractions, giving the molecular basis forthe formation of 13S and 16S assembly intermediates. Theysuggest that in addition to the 13S precursor complex, there aresmaller complexes that are composed of at least the �4 and �7
FIG. 10. Rate of overall proteolysis following oxidative stressin �5-overexpressing cell lines. Rate of overall proteolysis followingmetal catalyzed oxidation (A and B) and tBHP treatment (C and D) invector (CON)- and �5-transfected WI38/T/�5.8 (A and C) and HL60/�5.3(B and D) cells for the indicated post-labeling time points. Degradation(i.e. recorded scintillation counts) for vector-transfected cells at 0.5 hwas set to 100%. Values represent the average of two independentexperiments for each time point. Overall proteolysis was increased in�5-overexpressing cells following treatment with oxidants.
subunits (subunits that we have also identified in the latefractions). Our results are in accordance with the existence ofsuch complexes. It is shown here that an excess of �-typesubunits in the control cells could possibly give rise to suchearly complexes (because we have detected �4, �6, and �7 sub-units in different late fractions), whereas these complexes arenot detected in the �5 clones, possibly because they are quicklyused to produce the higher amounts of assembled proteasomes.Finally, it is also feasible that these subunits act as free mono-mers. Jorgensen and Hendil (57) have reported that the �5
subunit seems to exhibit an RNase activity even as a freemonomer. In addition, the S5a/Rpn10 yeast subunit of the 19Scomplex has been identified to exist in two forms, as either aproteasome subunit or a free form (58).
A major point of this study is that the “proteasome up-regulated” cell lines have increased capacity to cope with var-ious stresses. Accumulation of abnormal proteins is determinedby their rates of formation, but their rates of hydrolysis andelimination are of equal importance. Because the proteasomalsystem is responsible for the degradation of non-functionalproteins as well as for the cellular response to oxidative stress(reviewed in Ref. 59), it is expected that cells possessing ele-vated proteasome activities, like our �5 clones, exhibit en-hanced survival to the different oxidants used in this study. Inaddition we demonstrate that �5 clones exhibit higher degra-dation rates following oxidative stress. Thus the increased ca-pacity to respond to oxidative stress of the clones relates to theelevated protein turnover.
In a previous work, we have studied proteasome activity infibroblasts derived from healthy donors of different ages(18–80 years) including centenarians, because they representthe best model of successful aging (reviewed in Ref. 60). Wehave found that healthy centenarians possess an active protea-some (11). Davies and co-workers (2) have hypothesized that theability of the proteasome to degrade oxidized proteins serves as asecondary cellular antioxidant defense system. In the currentstudy, we show that overexpression of the �5 subunit in primaryIMR90 fibroblasts delays senescence. Because oxidation is a ma-jor factor that contributes to aging and replicative senescence(reviewed in Ref. 38), the retardation of the senescent phenotypeobserved in IMR90 clones could be considered an amelioratedresponse against oxidative stress.
Proteasome function appears to be dependent on both geneticand environmental factors. This study highlights the observa-tion that the proteasome can be genetically “up-regulated,”thus resulting in enhanced cellular capacity against oxidants.Because the proteasome is down-regulated in several biologicalprocesses, including aging and diseases, anti-aging/therapeuticstrategies should be aimed at proteasome activation and iden-tification of the rules that govern proteasome assembly andregulation. Furthermore, the search for compounds that mayactivate the proteasome is expected to be of great interest.
Acknowledgments—We thank Dr. B. Friguet for critical reading ofthe manuscript and Dr. I. P. Trougakos for helpful discussions duringthe course of this work. We are grateful to Drs. E. Kruger, D. Rickwood,K. Rock, and K. Tanaka for cell lines, antibodies, and plasmids.Drs. H. Moutsopoulos and D. Liakos are acknowledged for the use ofmicroscopy facilities.
REFERENCES
1. Coux, O., Tanaka, K., and Goldberg, A. L. (1996) Annu. Rev. Biochem. 65,801–847
2. Grune, T., Reinheckel, T., and Davies, K. J. (1997) FASEB J. 11, 526–5343. Gaczynska, M., Osmulski, P. A., and Ward, W. F. (2001) Mech. Ageing Dev.
122, 235–2544. Rivett, A. J. (1993) Biochem. J. 291, 1–105. Tanaka, K. (1998) J. Biochem. (Tokyo) 123, 195–2046. Voges, D., Zwickl, P., and Baumeister, W. (1999) Annu. Rev. Biochem. 68,
1015–10687. Ciechanover, A. (1998) EMBO J. 17, 7151–7160
8. DeMartino, G. N., and Slaughter, C. A. (1999) J. Biol. Chem. 274,22123–22126
9. Merker, K., and Grune, T. (2000) Exp. Gerontol. 35, 779–78610. Chondrogianni, N., Fragoulis, E. G., and Gonos, E. S. (2002) Biogerontology 3,
121–12311. Chondrogianni, N., Petropoulos, I., Franceschi, C., Friguet, B., and Gonos,
E. S. (2000) Exp. Gerontol. 35, 721–72812. Chondrogianni, N., Stratford, F. L. L., Trougakos I. P., Friguet, B., Rivett,
A. J., and Gonos, E. S. (2003) J. Biol. Chem. 278, 28026–2803713. Bulteau, A. L., Petropoulos, I., and Friguet, B. (2000) Exp. Gerontol. 35,
767–77714. Bulteau, A. L., Szweda, L. I., and Friguet, B. (2002) Arch. Biochem. Biophys.
397, 298–30415. Merker, K., Sitte, N., and Grune, T. (2000) Arch. Biochem. Biophys. 375,
50–5416. Petropoulos, I., Conconi, M., Wang, X., Hoenel, B., Bregegere, F., and Friguet,
B. (2000) J. Gerontol. 55, 220–22717. Sitte, N., Merker, K., von Zglinicki, T., Grune, T., and Davies K. J. A. (2000)
FASEB J. 14, 2495–250218. Sitte, N., Merker, K., von Zglinicki, T., Grune, T., and Davies, K. J. A. (2000)
FASEB J. 14, 2503–251019. Sitte, N., Merker, K., von Zglinicki, T., and Grune, T. (2000) Free Radic. Biol.
Med. 28, 701–70820. Chondrogianni, N., and Gonos E. S. (2004) Biogerontology 5, 55–6121. Gaczynska, M., Rock, K., Spies, T., and Goldberg, A. L. (1994) Proc. Natl. Acad.
Sci. U. S. A. 91, 9213–921722. Gaczynska, M., Goldberg, A. L., Tanaka, K., Hendil, K. B., and Rock, K. (1996)
J. Biol. Chem. 271, 17275–1728023. Ullrich, O., Reinheckel, T., Sitte, N., Haas, R., Grune, T., and Davies, K. J.
(1999) Proc. Natl. Acad. Sci. U. S. A. 96, 6223–622824. Ustrell, V., Hoffman, L., Pratt, G., and Rechsteiner, M. (2002) EMBO J. 13,
3516–352525. Kruger, E., Kloetzel, P. M., and Enenkel, C. (2001) Biochimie (Paris) 83,
289–29326. Witt, E., Zantopf, D., Schmidt, M., Kraft, R., Kloetzel, P. M., and Kruger, E.
(2000) J. Mol. Biol. 301, 1–927. Burri, L., Hockendorff, J., Boehm, U., Klamp, T., Dohmen, R. J., and Levy, F.
(2000) Proc. Natl. Acad. Sci. U. S. A. 97, 10348–1035328. Griffin, T. A., Slack, J. P., McCluskey, T. S., Monaco, J., and Colbert, R. A.
(2000) Mol. Cell. Biol. Res. Commun. 3, 212–21729. Ramos, P. C., Hockendorff, J., Johnson, E. S., Varshavsky, A., and Dohmen,
R. J. (1998) Cell 92, 489–49930. Jayarapu, K., and Griffin, T. A. (2004) Biochem. Biophys. Res. Commun. 314,
523–52831. Akiyama, K., Yokota, K., Kagawa, S., Shimbara, N., Tamura, T., Akioka, H.,
Nothwang, H. G., Noda, C., Tanaka, K., and Ichihara, A. (1994) Science 265,1231–1234
32. Bildirici, L., Smith, P., Tzavelas, C., Horefti, E., and Rickwood, D. (2000)Nature 405, 298
33. Tzavelas, C., Bildirici, L., and Rickwood, D. (2004) BioTechniques 37, 276–278,280–281
34. Rivett, A. J., Savory, P. J., and Djaballah H. (1994) Methods Enzymol. 244,331–350
35. Harlow, E., and Lane, D. (1999) Using Antibodies: A Laboratory Manual, ColdSpring Harbor Laboratory Press, Cold Spring Harbor, NY
36. Rivett, A. J., Bose, S., Pemberton, A. J., Brooks, P. B., Onion, O., Shirley, D.,Stratford, F. L. L., and Forti, K. (2002) Exp. Gerontol. 37, 1217–1222
37. Hendil, K. B., Kristensen, P., and Uerkvitz, W. (1995) Biochem. J. 305,245–252
38. Trougakos, I. P., Chondrogianni, N., Pimenidou, A., Katsiki, M., Tzavelas, C.,and Gonos, E. S. (2003) in Modulating Aging and Longevity (Rattan, S. I. S.,ed) pp. 65–83, Kluwer Academic Publishers, Dordrecht, The Netherlands
39. Petropoulou, C., Chondrogianni, N., Simoes, D., Agiostratidou, G., Drosopou-los, N., Kotsota, V., and Gonos, E. S. (2000) Ann. N. Y. Acad. Sci. 908,133–142
40. Friguet, B., Bulteau, A. L., Chondrogianni, N., Conconi, M., and Petropoulos,I. (2000) Ann. N. Y. Acad. Sci. 908, 143–154
41. Doyle, A., and Griffiths J. B. (1998) Cell and Tissue Culture: LaboratoryProcedures in Biotechnology, Wiley and Sons Ltd., West Sussex, UnitedKingdom
42. Mannhaupt, G., Schnall, R., Karpov, V., Vetter, I., and Feldmann, H. (1999)FEBS Lett. 450, 27–34
43. Xie, Y., and Varshavsky, A. (2001) Proc. Natl. Acad. Sci. U. S. A. 98,3056–3061
44. Owsianik, G., Balzi, I. L., and Ghislain M. (2002) Mol. Microbiol. 43,1295–1308
45. Wojcik, C., and DeMartino, G. N. (2002) J. Biol. Chem. 277, 6188–619746. Meiners, S., Heyken, D., Weller, A., Ludwig, A., Stangl, K., Kloetzel, P. M., and
Kruger, E. (2003) J. Biol. Chem. 278, 21517–2152547. Frentzel, S., Pesold-Hurt, B., Seelig, A., and Kloetzel P. M. (1994) J. Mol. Biol.
236, 975–98148. Yang, Y., Fruh, K., Ahn, K., and Peterson, P. A. (1995) J. Biol. Chem. 270,
27687–2769449. Nandi, D., Woodward, E., Ginsburg, D. B., and Monaco, J. J. (1997) EMBO J.
16, 5363–537550. Schmidtke, G., Schmidt, M., and Kloetzel P. M. (1997) J. Mol. Biol. 268,
95–10651. Heinemeyer, W., Ramos, P. C., and Dohmen, R. J. (2004) Cell. Mol. Life Sci. 61,
1562–157852. Gerards, W. L., Enzlin, J., Haner, M., Hendriks, I. L., Aebi, U., Bloemendal, H.,
and Bohelens W. (1997) J. Biol. Chem. 272, 10080–1008653. Gerards, W. L., de Jong, W. W., Bloemendal, H., and Boelens, W. (1998) J. Mol.
54. Yao, Y., Toth, C. R., Huang, L., Wong, M. L., Dias, P., Burlingame, A. L.,Coffino, P., and Wang, C. C. (1999) Biochem. J. 344, 349–358
55. Baumeister, W., Walz, J., Zuhl, F., and Seemuller E. (1998) Cell 92,367–380
56. Apcher, G.-S., Maitland, J., Dawson, S., Sheppard, P., and Mayer, R. J. (2004)FEBS Lett. 569, 211–216
57. Jorgensen, L., and Hendil, K. B. (1999) Mol. Biol. Rep. 26, 119–12358. van Nocker, S., Sadis, S., Rubin, D. M., Glickman, M., Fu, H., Coux, O., Wefes,
I., Finley, D., and Vierstra, R. D. (1996) Mol. Cell. Biol. 16, 6020–602859. Grune, T., Jung, T., Merker, K., and Davies, K. J. A. (2004) Int. J. Biochem.
Cell Biol. 36, 2519–253060. Gonos, E. S. (2000) Exp. Gerontol. 35, 15–21
![[Vierstra, 2003 TIPS]. Ubiquitin/26S proteasome pathway Ub + ATP E1 E3 E2 Target Ub Target 26S proteasome UbiquitinationProteolysis + ATP Simplified.](https://static.documents.pub/doc/80x56/56649c7d5503460f94932c85/vierstra-2003-tips-ubiquitin26s-proteasome-pathway-ub-atp-e1-e3-e2-target.jpg)
