Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Inhibition of the 26 S Proteasome by Polyubiquitin Chains Synthesized to Have Defined Lengths

Inhibition of the 26 S Proteasome by Polyubiquitin Chains Synthesized to Have Defined Lengths THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 272, No. 38, Issue of September 19, pp. 23712–23721, 1997 © 1997 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Inhibition of the 26 S Proteasome by Polyubiquitin Chains Synthesized to Have Defined Lengths* (Received for publication, June 2, 1997) Julia Piotrowski‡, Richard Beal§, Laura Hoffman¶, Keith D. Wilkinsoni, Robert E. Cohen**, and Cecile M. Pickart‡ ‡‡ From the ‡Department of Biochemistry, School of Hygiene and Public Health, Johns Hopkins University, Baltimore, Maryland 21205, the §Department of Biochemistry, School of Medicine, State University of New York, Buffalo, New York 14214, the ¶Department of Biochemistry, University of Utah Medical School, Salt Lake City, Utah 84132, the iDepartment of Biochemistry, School of Medicine, Emory University, Atlanta, Georgia 30322, and the **Department of Biochemistry, College of Medicine, University of Iowa, Iowa City, Iowa 52242 Ubiquitin is a covalent signal that targets cellular pro- tion of high volume and high selectivity: most short lived pro- teins to the 26 S proteasome. Multiple ubiquitins can be teins are degraded in this pathway, but the half-lives of indi- ligated together through the formation of isopeptide vidual substrate proteins can be regulated acutely and 48 76 bonds between Lys and Gly of successive ubiquitins. independently. Such a polyubiquitin chain constitutes a highly effective Ubiquitin acts as a degradation signal by virtue of covalent proteolytic targeting signal, but its mode of interaction ligation to target proteins. Ubiquitination occurs through the with the proteasome is not well understood. Experi- formation of an isopeptide bond between the COOH terminus ments to address this issue have been limited by diffi- 76 of ubiquitin (Gly ) and an internal Lys residue of the target culties in preparing useful quantities of polyubiquitin protein. This modification confers recognition by the multisub- chains of uniform length. We report a simple method for unit 26 S proteasome; the target protein is degraded, but ubiq- large scale synthesis of Lys -linked polyubiquitin uitin is regenerated for use in subsequent proteolytic cycles chains of defined length. In the first round of synthesis, (1–3). Specificity in ubiquitin-mediated proteolysis appears to two ubiquitin derivatives (K48C-ubiquitin and Asp - arise primarily in the ubiquitin attachment step, which in- ubiquitin) were used as substrates for the well charac- volves the sequential formation of ubiquitin thiol ester adducts terized ubiquitin-conjugating enzyme E2-25K. Diubiq- of ubiquitin-activating (E1), -conjugating (E2), and ligase (E3) uitin blocked at the nascent proximal and distal chain enzymes (1, 4, 5). Recent studies suggest that for a given termini was obtained in quantitative yield. Appropri- substrate, a specific E3 carries out the substrate ubiquitination ately deblocked chains were then combined to synthe- step, whereas a specific E2 charges the E3 with ubiquitin (5). size higher order chains (tetramer and octamer in the The ligation of multiple ubiquitins increases the rate of sub- present study). Deblocking was achieved either enzy- strate degradation (6 – 8), although the exact nature of the matically (proximal terminus) or by chemical alkylation (distal terminus). Chains synthesized by this method dependence is unclear. Multiple ubiquitination can occur were used to obtain the first quantitative information through the ligation of ubiquitin monomers to several sub- concerning the influence of polyubiquitin chain length strate Lys residues (9) but more typically involves the assembly on binding to the 26 S proteasome; this was done on the substrate of a polymeric, isopeptide-linked ubiquitin through comparison of different length (unanchored) chain (10). Multiple Lys residues of ubiquitin, including Lys , 11 29 48 63 polyubiquitin chains as inhibitors of ubiquitin-conju- Lys , Lys , Lys , and Lys , can serve as sites of polyubiq- gate degradation. K was found to decrease ;90-fold, 0.5 uitin chain initiation and/or elongation (10 –15). However, from 430 to 4.8 mM, as the chain was lengthened from two 48 Lys -linked chains represent the most commonly utilized deg- to eight ubiquitins. The implications of these results for radation signal in the ubiquitin pathway. This conclusion is the molecular basis of chain recognition are discussed. supported by several lines of evidence, including the results of in vitro biochemical analyses (10, 16) and the lethality of the K48R mutation in Saccharomyces cerevisiae (12). The targeting The conserved protein ubiquitin functions in diverse biolog- 48 ability of Lys -linked polyubiquitin chains apparently arises ical processes, including oncogenesis, cell cycle progression, from their high affinity for the 26 S proteasome, which may be antigen presentation, and programmed cell death (for review, due in part to the exposure of a regular array of hydrophobic see Refs. 1 and 2). These and most other functions of ubiquitin patches on the chain surface (17). reflect its role as an essential cofactor in an energy-dependent The 26 S proteasome is assembled from catalytic (20 S) and proteolytic pathway whose hallmarks are an unusual combina- regulatory (19 S) subcomplexes (3, 18 –20). The crystal struc- tures of 20 S proteasomes from archaebacteria and yeast show that access to the proteolytic active sites is highly restricted * This research was supported in part by National Institutes of (21, 22). Thus the target protein must be unfolded before deg- Health Grants DK46984 (to C. M. P), GM30308 (to K. D. W.), GM37666 (to R. E. C.), and GM37009 and the Lucille P. Markey Charitable Trust radation. Targeting of the ubiquitinated substrate to the pro- (to M. Rechsteiner). The costs of publication of this article were de- teasome is an activity of the 19 S complex, as suggested by the frayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The abbreviations used are: E1, ubiquitin-activating enzyme; E2, ‡‡ Recipient of a National Institutes of Health research career devel- ubiquitin-conjugating enzyme; E3, ubiquitin-protein ligase; Ubal, ubiq- opment award. To whom correspondence should be addressed: Dept. of uitin aldehyde; IPTG, isopropyl-b-D-thiogalactopyranoside; H , hexa- Biochemistry, Johns Hopkins University, 615 North Wolfe St., Balti- histidine; DTT, dithiothreitol; Bis-Tris, 2-[bis(2-hydroxyethyl)amino]-2- more, MD 21205. Tel.: 410-614-4554; Fax: 410-955-2926; E-mail: (hydroxymethyl)-propane-1,3-diol; Ub , polyubiquitin chain composed [email protected]. of n ubiquitins; ATPgS, adenosine 59-O-(thiotriphosphate). 23712 This paper is available on line at http://www.jbc.org This is an Open Access article under the CC BY license. Inhibition of 26 S Proteasome by Polyubiquitin Chains 23713 into pDS78 from which this same fragment had been excised. Plasmid ubiquitin independence of protein degradation catalyzed by the pDS78, provided by M. Rechsteiner (University of Utah) encodes an 20 S proteasome (3, 18 –20). The 19 S complex contains a polyu- H -tagged version of human ubiquitin under the control of an IPTG- biquitin chain-binding protein known as S5a, MBP1, or MCB1 inducible promoter. The BglII site in pDS78 occurs just after the H tag; (23, 24), and multiple subunits harboring ATP binding sites (3, the NH -terminal sequence is MHHHHHHGEFQ, where Q corresponds 25). However, most of the subunits of the 19 S complex are to Q2 in wild type human ubiquitin. functionally uncharacterized. One or more of these subunits Ubiquitin Expression and Purification—Expression of K48C-, K48R-, and Asp -ubiquitins (encoded by pET/pRSET-based plasmids) was in- must be an additional polyubiquitin-binding protein because a duced by the addition of 0.4 mM IPTG to cultures of appropriately yeast mcb1D strain is viable and competent in ubiquitin-medi- transformed E. coli BL21(DE3)pLysS cells growing at 37 °C. IPTG was ated proteolysis (26). added once an A of ;0.6 had been reached, and growth was continued Biochemically useful quantities of substrates, i.e. polyubiq- for 4 h more. Cells were harvested, frozen, and thawed in buffer con- uitinated target proteins, are a prerequisite for dissecting the taining 50 mM Tris-HCl (24% base) and 0.4 mg/ml lysozyme, supple- mechanistic coordination of chain recognition, substrate un- mented with 1 mM EDTA and 10 mM DTT in the case of K48C-ubiquitin (2 ml of buffer/g of cells). Efficient lysis occurred over several minutes. folding, and peptide bond hydrolysis by the 26 S proteasome. DNase I (20 mg/ml) and MgCl (10 mM) were then added to digest DNA. Ideally these substrates should be homogeneous by several Expression of H -tagged ubiquitins in strain M15 (harboring the lac different criteria: overall purity, polyubiquitin chain length, repressor-expressing plasmid pDMI,1) was carried out similarly, except chain linkage, and site of chain ligation to the target protein. So that cells were lysed using a French press. In all cases, crude soluble far, such well defined substrates have been unattainable be- lysates were produced by centrifugation at 9,000 3 g for 20 min. cause of the low purity and abundance of E3 enzymes, as well Non-ubiquitin proteins were precipitated at 0 °C with perchloric acid (3.5% w/v). Ubiquitin was resolved from the dialyzed supernatant by as the uncontrolled character of chain elongation as catalyzed cation exchange chromatography on S-Sepharose (Pharmacia Biotech by available E2 and E3 enzymes. To overcome the chain elon- Inc.) as described (36) (K48C- and Asp -ubiquitins) or by gradient gation problem, we have implemented a novel method that elution from S-Sepharose at pH 6.1 (H -tagged ubiquitins). For K48C utilizes the well characterized ubiquitin-conjugating enzyme mutants, all steps were carried out in the presence of 1 mM DTT and 0.1 E2-25K to generate Lys -linked polyubiquitin chains of de- mM EDTA. The ubiquitin derivatives utilized in this work were recov- fined length. This method avoids uncontrolled polymerization ered at 25–100 mg of purified protein/liter of cell culture. Ubiquitin concentrations were determined by A , assuming an absorbance of by utilizing, in each round of synthesis, two chain reactants, 0.16 for a 1 mg/ml solution (36). one reversibly capped at its proximal, and the other at its Analytical Alkylation Kinetics—Kinetics were monitored at 37 °C in distal, terminus. This method generates “unanchored” chains incubations containing 0.2 M Tris-HCl (50% base), 1 mM EDTA, 1 mM that bind to the 26 S proteasome and inhibit the degradation of cysteine, or 0.5 mM K48C-ubiquitin, and 0 – 64 mM alkylating reagent; polyubiquitinated lysozyme. This inhibition assay was used to or Tris (pH 8.0), was replaced by Bis-Tris propane (pH 8.5). At timed evaluate, for the first time, the relative binding of different intervals, aliquots were assayed for remaining thiol with dithionitro- benzoate (37). Values of k were obtained from linear plots of ln A /A length chains to the 26 S proteasome. obs t 0 against time and were corrected by subtracting the value of k meas- obs EXPERIMENTAL PROCEDURES ured in the absence of alkylating agent. Preparative Alkylation (Distal Terminus Deblocking)—For the exper- Ethyleneimine was purchased from ChemService, stored at 5 °C, and iments shown in Table I and Fig. 3, conditions were as follows. Reac- used within several days of opening the sealed ampule. Except where tions with ethyleneimine contained 50 mM ethyleneimine and the pu- noted, other reagents were purchased from Sigma. Protein iodination rified ubiquitin chain at 2–20 mg/ml (#1 ml). Other conditions were the with chloramine T was carried out as described (4). Tryptic digestion of same as for kinetic measurements (above). After 1 h, the reaction ubiquitin derivatives (5 mg/ml total ubiquitin and 5% w/w trypsin) and mixture was dialyzed against 5 mM ammonium acetate (pH 5.5), 0.1 mM reversed phase high performance liquid chromatography separation of EDTA, and 1 mM DTT to remove residual ethyleneimine. Alkylation of peptides were carried out as described (27, 28). K48C-ubiquitin (1 mg/ml) with bromoethylamine (50 mM, Aldrich) was Enzymes and Proteins—Fraction II was prepared from rabbit reticu- carried out similarly. For alkylation of K48C-ubiquitin (1 mg/ml) with locyte-rich whole blood (4). E1 was purified from bovine erythrocytes or TM N-(iodoethyl)trifluoroacetamide (Aminoethyl-8 , Pierce), 3.4-h incu- rabbit liver (29). Recombinant bovine E2-25K harboring two catalyti- bations were done at 50 °C in 0.2 M Bis-Tris propane (as above). The cally silent mutations was expressed in Escherichia coli using the alkylating agent was dissolved in methanol and added in two equal vector pET3d-C170S,F174L-25K and purified as described (30). Yeast portions at time zero and1hto give a final concentration of 3.1 mM. ubiquitin hydrolase-1 was prepared by a slight modification of the After addition of DTT to 0.25 mM, the reaction was dialyzed as above. method described previously (31). Ubal was prepared by either of two Proximal Terminus Deblocking—The purified ubiquitin chain (2–50 methods described previously (32, 33). mg/ml, #1 ml) was incubated with 15 mg/ml yeast ubiquitin hydrolase-1 Ubiquitin Plasmids—All coding sequences were derived from the for1hat37°Cin50mM Tris-HCl (24% base), 0.1 mM EDTA, and ;0.5 synthetic human ubiquitin gene described by Ecker et al. (34). Plasmids mM DTT. The reaction was passed through a 0.5-ml Q-Sepharose col- encoding K48C- and K48R-ubiquitin were prepared by mutagenesis of umn (Pharmacia; preequilibrated in the same buffer) to absorb the pPLhUb (35) as described (17). The coding regions of the resulting enzyme; the flow-through was collected into a Centricon-10 microcon- pPLhUb plasmids were amplified using flanking primers Ub5F (59- centrator (Amicon). The column was washed with 3 volumes of buffer. tgcatttatttgcatacattca-39) and Ub3F (59-cgaattcgagctcggatcctcaa-39). The combined flow-through and wash fractions were concentrated. The resulting products were digested with NdeI and BamHI and sub- Protein Degradation Assays—The production of acid-soluble radioac- cloned into NdeI/BamHI-restricted pET3a to produce pET3a-UbK48C 125 125 6 5 tivity from I-lysozyme or I-lactalbumin (;10 cpm/mg; ;10 cpm/ and pET3a-UbK48R. 25-ml incubation) was monitored in incubations with rabbit reticulocyte The plasmid pRSUbD, encoding Asp -ubiquitin, was prepared as fraction II (;1.5 mg/ml protein) as described (17). Rates obtained in the follows. A vector library encoding a variety of single amino acids COOH- presence of wild type ubiquitin or ubiquitin derivative (12 mM) were terminal to ubiquitin was constructed by polymerase chain reaction. To corrected using blanks obtained by omitting ubiquitin. In some cases create this amino acid library at position 77, the coding region of aliquots were withdrawn during the steady state of degradation for pRSUb was amplified with a degenerate 39-primer which contained all analysis of the levels of ubiquitin-substrate conjugates after electro- possible codons followed by a stop codon and a HindIII site. The primer phoresis and autoradiography (17). For the experiment involving dis- sequences were 59-atccatatgcagatcttcg-39 and 59-caagcttctaNNNaccac- assembly of engineered Ub (see the last entry in Table I), 4.5 mM Ub cacgaagtc-39. The polymerase chain reaction products from this reaction 4 4 (potentially yielding 18 mM monoubiquitin) was preincubated for 30 min were subcloned en masse into pCRII (Invitrogen). Plasmids containing under the conditions of the degradation assay, except that ATP, regen- inserts were sequenced, and an insert encoding Asp at the COOH erating system, and labeled substrate were omitted. These components terminus was subcloned into pRSET using the NdeI and HindIII sites. were added to initiate the assay. For the control, the preincubation A plasmid for IPTG-induced expression of H ,K48C-ubiquitin was contained 18 mM wild type monoubiquitin. prepared by ligating the KpnI-BglII fragment (34) from pPLhUb-K48C Engineered Polyubiquitin Chains: Synthesis and Purification—Incu- bations of 0.2–2.0 ml contained 50 mM Tris-HCl (50% base), 5 mM C. Larsen and K. Wilkinson, unpublished data. MgCl ,10mM phosphocreatine, 0.6 unit/ml each of creatine phosphoki- 2 23714 Inhibition of 26 S Proteasome by Polyubiquitin Chains FIG.1. Method for in vitro synthesis of defined length polyubiquitin chains. Ub is made by incubating E2- 25K with equal concentrations of ubiq- uitin capped at the future proximal chain terminus (Asp -ubiquitin, open circle), and ubiquitin capped at the future distal chain terminus (K48C-ubiquitin, filled circle). The proximal or distal terminus of the resulting dimer is deblocked by treat- ment with yeast ubiquitin hydrolase-1 (YUH1) or ethyleneimine, respectively. In the second cycle of synthesis, Ub is made by incubating the two single-deblocked dimers with E2-25K. See “Results and Discussion.” nase and inorganic pyrophosphatase, 4 mM ATP, and 0.5–1 mM DTT mM MgCl , 10% v/v glycerol, 10 mM DTT, 1.5 or 3 mM Ubal, 2 mM ATP, (37 °C); E1 and C170S,F174L-E2-25K were 0.2 mM and 20 mM, respec- 10 mM phosphocreatine, 0.3 unit/ml creatine phosphokinase, 0.3 tively. The two chain reactants were added at equal concentrations. unit/ml pyrophosphatase, 0.37 mg/ml 26 S proteasome (0.18 mM based Times required for completion of the conjugation reaction, usually 1– 4 on a molecular mass of 2.1 MDa), and ;5,000 – 8,000 cpm of I- h, were determined by examining reaction aliquots by SDS-polyacryl- lysozyme conjugates (;30 nM lysozyme; above). The lysozyme conju- amide gel electrophoresis. At the end of the incubation, the reaction was gates also contributed H -ubiquitin at a final concentration of ;10 mM. passed through a Q-Sepharose column (0.5–1 ml) at pH 7.6 to absorb Only a small fraction of this H -ubiquitin was conjugated to I-ly- the enzymes. In most cases the unabsorbed fraction was adjusted to pH sozyme; of the remainder, some was unconjugated, and some was con- 4.5 (for non-tagged chains) or 6.1 (for H -tagged chains) and applied to jugated to fraction II proteins (the conjugated:unconjugated ratio is an S-Sepharose column preequilibrated at the appropriate pH (Phar- unknown). Ubal was included to inhibit disassembly of polyubiquitin macia; 15 mg of ubiquitin/ml of resin), and the major chain product was chains by one or more isopeptidases in the 26 S proteasome prepara- purified by elution with a linear gradient of NaCl (0 – 0.5 M). Fractions tion; control experiments showed that 1.5–3 mM Ubal completely inhib- spanning the peak region were examined by SDS-polyacrylamide gel ited polyubiquitin chain disassembly but had no effect on the rate of electrophoresis and Coomassie staining before pooling. Concentration conjugate degradation. Incubations were quenched after 10 min by the and buffer exchange were carried out using Ultrafree-4 or -15 devices addition of trichloroacetic acid, and acid-soluble radioactivity was de- (Millipore). Ub and Ub were resolved by gel filtration on a 1 3 50-cm termined by counting an aliquot of the acid supernatant for 10 min. 4 8 column of Sephadex G-75 (Sigma) buffered with 50 mM ammonium Appropriate controls showed that degradation was linear with time for acetate (pH 5.5), 0.1 mM EDTA, and 1 mM DTT. ;15 min and that degradation was abolished if MgATP was omitted. Wild Type Polyubiquitin Chains—For Ub , the incubation (0.25 ml) Data were corrected by subtracting blanks derived from incubations in was carried out as described for engineered chains (above), except that which conjugates were replaced with unconjugated I-lysozyme. Con- the reactants were K48R-ubiquitin (;3 mg/ml) and Ub assembled from trol experiments showed that subjecting unconjugated I-lysozyme to wild type ubiquitin (;12 mg/ml). The Ub was largely des-Gly-Gly at its the same buffer sequence used in the Ni-NTA chromatography did not proximal terminus. After2hof incubation (37 °C), E1 and E2-25K were increase its susceptibility to degradation. Typically, 15–20% of the removed by passage through an anion exchange column (above), and conjugates were degraded in the positive control, and replicate assays Ub was purified by gradient elution (above) from a fast protein liquid agreed within 10 –15%. chromatography Mono S column (Pharmacia). Cation exchange was Because initial experiments showed that the addition of ubiquitin also used to resolve Ub and Ub from mixed chains assembled from chains to the assay significantly increased the recovery of acid-soluble 2 3 wild type ubiquitin (38). radioactivity by counteracting nonspecific absorption, all assays in- Preparation of H -Ubiquitin- I-Lysozyme Conjugates—Incubations cluded monoubiquitin as carrier. Most assays were supplemented with of 2 ml contained (pH 8.0, 37 °C) 50 mM Tris-HCl (50% base), 5 mM 230 mM monoubiquitin; in some cases, the concentration was 420 mM. MgCl ,5mM ATPgS (Boehringer Mannheim), 8 mM Ubal, 10 mM DTT, Data from the two sets of experiments could be fit by the same K 2 0.5 0.3 unit/ml inorganic pyrophosphatase, 3 mg/ml fraction II protein, 0.3 value for Ub , suggesting that monoubiquitin did not bind competi- 7 125 mg/ml H -ubiquitin, and ;4 3 10 cpm of I-lysozyme (40 –50 mg). tively with unanchored polyubiquitin chains. This was confirmed by After 2 h, 20 mM N-ethylmaleimide (to quench DTT and inactivate showing that ;50% inhibition by 29 mM Ub (see “Results”) was ob- ubiquitin pathway enzymes) was added together with 4 M urea and 10 served whether or not the assay was supplemented with 230 mM mM imidazole. The quenched incubation was applied to a 1-ml column monoubiquitin. of Ni-NTA agarose (Novagen) preequilibrated with buffer containing 50 Mass Spectrometry—Matrix-assisted laser desorption time-of-flight mM Tris-HCl (50% base), 0.15 M NaCl, 9 M urea, and 25 mM imidazole. mass spectrometry employed a PerSeptive Biosystems Voyager RP The loaded column was washed with equilibration buffer. H -ubiquitin spectrometer operated in linear mode at an accelerating voltage of 30 conjugates were eluted with 4 ml of equilibration buffer supplemented kV. Samples were prepared with either a-cyano-4-hydroxycinnamic with 1 M imidazole. The eluate was dialyzed overnight against 2 liters acid or 3,5-dimethoxy-4-hydroxycinnamic acid as the matrix compound, of buffer containing 20 mM Tris-HCl (pH 8.0), 1 mM EDTA, and 1 mM and the instrument was calibrated by use of the single-charged (m/z 5 DTT. The dialyzed sample was concentrated to ;0.5 ml by ultrafiltra- 14,314.2) and double-charged (m/z 5 7,157.6) cations from an internal tion (Centricon-10). Analysis by SDS-polyacrylamide gel electrophore- standard of chicken egg white lysozyme. sis and autoradiography showed that 15–20% of the I-lysozyme was conjugated to ubiquitin during the initial incubation. About 15% of the RESULTS AND DISCUSSION conjugates were recovered in the final eluate, and .90% of the I- Principle of the Method—Fig. 1 shows the method for syn- lysozyme in the final eluate was conjugated to ubiquitin. thesizing Lys -linked polyubiquitin chains of defined length. Assay of Ubiquitin Conjugate Degradation by the 26 S Proteasome— The approach utilizes ubiquitin-conjugating enzyme E2-25K to The 26 S proteasome was prepared as described (18). Standard incuba- 48 76 tions contained (25–50 ml; pH 7.3, 37 °C) 50 mM Tris-HCl (24% base), 5 generate a Lys -Gly isopeptide bond between two ubiquitin Inhibition of 26 S Proteasome by Polyubiquitin Chains 23715 derivatives. One of these derivatives is capped at its COOH terminus (future proximal chain terminus) by the presence of an extra residue, Asp . The other ubiquitin derivative is capped at the ubiquitin-accepting site (future distal chain ter- minus) by the presence of a Lys-to-Cys mutation at residue 48. In the first round of synthesis, these reactants are the two monomeric derivatives Asp -ubiquitin and K48C-ubiquitin. In principle, E2-25K should quantitatively convert these two re- 76 48 actants to Ub ; once Gly of K48C-ubiquitin is linked to Lys of Asp -ubiquitin, further polymerization is prevented by the 48 76 absence of additional Lys and Gly residues. Additional rounds of controlled conjugation are possible because the prox- imal or distal terminus of the double-capped dimer is easily deblocked; Asp can be removed catalytically by the ubiquitin- processing protease yeast ubiquitin hydrolase-1 (31), whereas chemical alkylation of Cys with ethyleneimine (below) cre- ates a lysine mimic (39). The two single-deblocked dimers should be quantitatively converted to Ub in the next round of synthesis. By using the appropriate combination of single- capped chain reactants, this method can in principle give rise FIG.2. Ethyleneimine alkylation kinetics. Incubations (pH 8.1, to a chain of any length. 37 °C) contained 64 mM ethyleneimine and 1 mM cysteine (●) or 0.5 m M K48C-ubiquitin (E). Remaining thiol was determined by Ellman’s test. Efficiency and Specificity of Deblocking—E2-25K catalyzes the synthesis of polyubiquitin chains harboring exclusively 48 76 Lys -Gly isopeptide bonds (28, 38) and is available in recom- these results, it appeared that ethyleneimine could best alkyl- binant form (40). E1 and E2 enzymes can transfer polyubiq- ate the distal Cys in a chain reactant at a rate competitive uitin chains (28, 41). Thus, there is no block to efficient and with undesired alternative reactions, such as Cys oxidation. specific conjugation provided the necessary chain reactants are As shown in Fig. 2 (open circles), ethyleneimine reacted with available. This is trivial in the first cycle of synthesis, which Cys in K48C-monoubiquitin about two times more slowly directly utilizes recombinant monomeric ubiquitins. In subse- than with free cysteine, presumably reflecting steric hindrance quent cycles, however, the efficiency of conjugation is highly from surrounding side chains. dependent on the efficiency with which the double-capped prod- The efficiency and specificity of alkylation were examined uct of the previous cycle is deblocked. In addition, the function- further by tryptic peptide mapping of K48C-ubiquitin after ality of long chains generated by this method is likely to depend reaction with each of the reagents discussed above (for condi- strongly on specificity in chemical alkylation; a low level of tions of the alkylation reactions, see “Experimental Proce- reaction with side chains other than Cys could be deleterious dures”). Trace a in Fig. 3 shows the map of wild type ubiquitin; after the multiple rounds of alkylation needed to produce long peptides 4 –11 are clearly resolved (27). The K48C mutation chains. abolishes a tryptic cleavage site (top, Fig. 3); this is reflected as We expected that yeast ubiquitin hydrolase-1-catalyzed loss of peptides 4 and 8, concomitant with the appearance of a proximal deblocking would proceed with high efficiency and new combined peptide, labeled 12 in Fig. 3, which contains specificity, and this proved to be the case. Removal of the 48 Cys (compare traces a and b). (The low yield of peptide 6 in proximal Asp residue was very rapid, proceeding to comple- trace b is not significant; peptides 5 and 6 are overlapping, and tion within 60 min in reactions involving mono-, di-, and tet- their relative yields were variable.) Because alkylation restores rameric ubiquitin derivatives, at concentrations as high as 50 a cleavage site at residue 48, the extent of alkylation could be 48 76 mg/ml. There was no evidence for cleavage of Lys -Gly quantitated by the reappearance of peptides 4 and 8 and the isopeptide bonds. The single-deblocked Ub derivative result- disappearance of peptide 12. The peptide map of ethylenei- ing from yeast ubiquitin hydrolase-1 treatment (see below) was mine-treated K48C-ubiquitin (trace c) shows essentially full found to have a molecular mass of 17,090.5 6 8.2 Da, which recovery of peptides 4 and 8 and a corresponding loss of peptide compared favorably with the calculated mass of 17,087.7 Da. 12, indicating .90% modification of Cys . As expected from It seemed likely that efficiency and specificity would be more their slower alkylation rates (above), bromoethylamine and problematic in the alkylation of Cys than in the removal of N-(iodoethyl)trifluoroacetamide gave only partial alkylation, Asp . Thus we began with a systematic comparison of three reflected in low yields of peptides 4 and 8 and a corresponding reagents that could add an aminoethyl moiety to Cys : ethyl- persistence of peptide 12 (Fig. 3, traces d and e). eneimine, bromoethylamine, and N-(iodoethyl)trifluoroacet- Inspection of trace c further suggests that ethyleneimine did TM amide (Aminoethyl-8 ). N-(Iodoethyl)trifluoroacetamide was not react significantly with other potential nucleophilic side 1 68 used to alkylate K48C-ubiquitin in an earlier study (16), but chains, such as Met and His ; modification of these side the efficiency and specificity of alkylation were not determined. chains would probably be manifested as shifts in peaks 9 and In initial studies we determined the rates at which moderate 11, respectively (27). Mass spectrometric analysis of alkylated concentrations of the three reagents alkylated free cysteine. diubiquitin (bearing a proximal Asp residue, see below) con- The pseudo-first-order reaction observed with 64 mM ethylenei- firmed this inference; the observed mass of 17,232.7 6 12.4 Da mine at pH 8.1 and 37 °C is shown in Fig. 2 (k 5 0.12 min , compared favorably with the calculated mass of 17,245.8 Da. obs filled circles). The reaction was second-order in [ethylenei- Taken together, these results indicate that ethyleneimine acts 21 21 48 mine], corresponding to k 5 2.4 M min (average of dupli- with high efficiency and specificity at the side chain of Cys ,in cate determinations). Reaction with 50 mM bromoethylamine at both mono- and dimeric ubiquitin derivatives. pH 8.1 and 37 °C followed k /[bromoethylamine] 5 0.6 M Functionality of Alkylated Ubiquitin—To address the func- obs 21 21 21 min , whereas a value of k 5 0.3 M min was determined tionality of alkylated K48C-ubiquitin, we tested its ability to for N-(iodoethyl)trifluoroacetamide at pH 8.5 and 37 °C. From support degradation in a modified reticulocyte lysate, fraction 23716 Inhibition of 26 S Proteasome by Polyubiquitin Chains FIG.3. Tryptic peptide mapping of K48C-ubiquitin and alkylated deriva- tives. Top, sequence of ubiquitin with tryptic peptides indicated (27). Lines and boldface numbers below the sequence in- dicate peptides visible in the peptide maps; lines above sequence indicate unre- solved peptides and peptide 12 in K48C- ubiquitin (see “Results and Discussion”). Bottom, tryptic maps: a, wild type ubiq- uitin; b, K48C-ubiquitin; c– e, K48C-ubiq- uitin after alkylation with ethyleneimine (c), bromoethylamine (d), or N-(iodoethyl) trifluoroacetamide (e). The origin of the small peak (●) eluting around the posi- tion of peptide 8 in the K48C-ubiquitin map is not known. Each trace derived from digestion of 0.5 mg of ubiquitin. Ab- sorbance was monitored at 215 nm (0.2 unit full scale). II (4), in which degradation is dependent upon the addition of TABLE I functional ubiquitin. Ubiquitin derivatives that cannot form Support of degradation by mutant ubiquitins and alkylated chains are expected to show impaired activity in this assay (10, derivatives 16). For the two substrates assayed, there was indeed a de- Assays of the production of acid-soluble radioactivity were carried out in fraction II; ubiquitin and derivatives were added at 12 mM (20 3 K creased rate of degradation when K48C-ubiquitin (or K48R- 0.5 for wild type). For methods used in generating alkylated derivatives, ubiquitin) was substituted for wild type ubiquitin (Table I), and see “Experimental Procedures.” Typical incubation times were 20 min this decreased rate correlated with a strongly reduced level of 125 125 for I-lactalbumin and2hfor I-lysozyme (pH 7.3, 37 °C). Data were very high molecular weight conjugates (Fig. 4, lanes 3 and 4 corrected by subtraction of the acid-soluble counts from an incubation versus 2). These conjugates presumably bear long polyubiquitin lacking ubiquitin. The rate thus obtained for a given derivative is expressed relative to the corrected rate for wild type ubiquitin meas- chains. ured in the same experiment. Relative rates are means 6 S.D. of Alkylation of K48C-ubiquitin with ethyleneimine partially triplicate determinations, except for the final entry, which is the aver- 125 125 ( I-lactalbumin) or fully ( I-lysozyme) restored activity in age of duplicate determinations. For details of the last experiment, degradation (Table I). As expected based on this result, alky- which involved engineered Ub deblocked at its distal terminus by alkylation with ethyleneimine, see “Experimental Procedures.” In some lation with ethyleneimine also restored the formation of very experiments aliquots were withdrawn to monitor the level of ubiquitin- high molecular weight conjugates of I-lactalbumin (as mon- substrate conjugates by SDS-polyacrylamide gel electrophoresis and itored by electrophoresis and autoradiography; Fig. 4, lane 5 autoradiography (see “Experimental Procedures” and Fig. 4). versus 2). We do not know why full activity in I-lactalbumin Steady-state degradation rate degradation was not restored after alkylation. However, results Ubiquitin 125 125 I-Lactalbumin I-Lysozyme with the other alkylating reagents (below) indicate that this is % wild type ubiquitin an intrinsic effect of the presence of the S-aminoethylcysteine moiety rather than a specific effect of ethyleneimine. Prelimi- K48C-ubiquitin 37.8 6 2.0 59.4 6 3.7 K48R-ubiquitin 9.3 6 1.7 35.7 6 8.2 nary concentration dependence studies, carried out with ethyl- K48C (ethyleneimine) 65.6 6 3.8 95.5 6 5.2 eneimine-treated K48C-ubiquitin, indicated that this was a K48C (bromoethylamine) 70.5 6 4.3 97.4 6 5.3 V effect (data not shown). These results show that S-amin- K48C (Aminoethyl-8) 62.9 6 2.7 91.6 6 5.1 max oethylcysteine at position 48 functions similarly to Lys in ubiq- Disassembled Ub 68.8 Not done uitination and conjugate degradation, confirming the conclu- Inhibition of 26 S Proteasome by Polyubiquitin Chains 23717 FIG.5. Synthesis of engineered polyubiquitin chains (Coo- massie-stained gel). Each lane contains 2– 6 mg of total ubiquitin; reactions were carried out at 37 °C. Lanes 1 and 2,Ub synthesis. The incubation contained 20 mg of each reactant in 2 ml (pH 7.3). Lane 1, FIG.4. Activities of ubiquitin derivatives in I-lactalbumin zero time; lane 2,4h. Lanes 3 and 4,Ub synthesis. The incubation conjugation (autoradiograph). Incubations (see Table I) were car- contained 7 mg of each reactant in 0.7 ml (pH 8.0). Lane 3, zero time; ried out in fraction II without ubiquitin (lane 1) or in the presence of lane 4,1h. Lanes 5 and 6,Ub synthesis controls. Double-capped dimer K48C-ubiquitin (12 mM) that had been alkylated with the indicated (10 mg/ml) was incubated for 1 h with 10 mg/ml dimer deblocked at the reagent (EI, ethyleneimine; BEA, bromoethylamine; AE8, N-(iodoeth- proximal terminus (lane 5) or at the distal terminus (lane 6) (pH 8.0). yl)trifluoroacetamide). Aliquots were withdrawn during the steady Lanes 7 and 8,Ub synthesis. The incubation contained 2 mg of each state of degradation and analyzed by electrophoresis and autoradiog- reactant in 0.6 ml (pH 8.0). Lane 7, zero time; lane 8, 1.5 h. raphy. LA, I-lactalbumin; cont., labeled contaminant. With K48C and K48R-ubiquitin, two ubiquitinated forms of lactalbumin are visible below the contaminant band. With K48C-ubiquitin, three ubiquitinated ger inhibition seen with I-lactalbumin probably reflects a forms are also seen above the contaminant band. difference in the rate-limiting step; conjugate formation is the slow step in I-lysozyme degradation (44), whereas conjugate degradation may be partly rate-limiting in I-lactalbumin sion of an earlier study by Gregori et al. (16). turnover. Lactalbumin degradation is thus expected to be Alkylation with N-(iodoethyl)trifluoroacetamide substan- more sensitive to features such as chain length, which influ- tially restored the function of K48C-ubiquitin (Table I), as seen ence conjugate recognition by the 26 S proteasome. In addition, previously (16). A similar result was obtained after alkylation we cannot exclude the possibility that non-polyubiquitinated with bromoethylamine (Table I). The functionality of the latter lysozyme conjugates are recognized better than comparable two derivatives is not inconsistent with their demonstrated forms of lactalbumin. substoichiometric alkylation (Fig. 3), in view of the high con- Second, K48R- and K48C-ubiquitin were not equivalent: the centration of ubiquitin in the assay (12 mM), the low K of 0.5 K48C mutation was significantly less inhibitory with both sub- ubiquitin in degradation (0.6 mM (42)), and the ease with which strates. In the case of I-lactalbumin, this difference was ubiquitin conjugates undergo disassembly and reassembly (12). probably caused by the greater ability of K48C-ubiquitin to As expected based on these considerations, these two alkylated support the formation of conjugates bearing multiple single derivatives largely restored the formation of high molecular ubiquitins (up to n ; 5; Fig. 4, lanes 3 versus 4). This effect may weight conjugates (Fig. 4, lanes 6 and 7 versus 2). be explained by decreased susceptibility to de-ubiquitination In a final experiment, we used alkylated Ub (resulting from because conjugates bearing K48C-ubiquitin, but not K48R- two cycles of synthesis, Fig. 1) as the source of ubiquitin to ubiquitin or alkylated K48C-ubiquitin, are resistant to disas- support degradation. Alkylated Ub was almost completely sembly by an isopeptidase associated with the 26 S proteasome disassembled by endogenous isopeptidases during a 30-min (45). The inhibitory effect of a given chain-terminating muta- preincubation in fraction II (without ATP), as indicated by the tion can also be influenced by substrate structure; with an results of Western blotting with anti-ubiquitin antibodies (not engineered b-galactosidase substrate, both K48R- and K48C- shown). After the addition of ATP, the monomers produced by ubiquitin inhibited degradation by more than 90% (10, 16). disassembly supported degradation at the same rate as alkyl- Here the similar behavior of the two mutant ubiquitins, and ated K48C-monoubiquitin (Table I). This result provides a the very dramatic inhibition, may be explained by the presence qualitative indication that isopeptidases recognize the modi- of only a single ubiquitination site in the target protein (10); fied isopeptide bond containing the S-aminoethylcysteine moi- regardless of the nature of the mutation at residue 48, only ety and shows that two rounds of alkylation did not result in monoubiquitinated b-galactosidase could be formed. gratuitous inhibitory modifications. Based on the functionality Large Scale Synthesis of Defined Length Polyubiquitin of products generated using ethyleneimine (Table I and Fig. 4) Chains—Synthesis of double-capped Ub from Asp -ubiquitin and on the high efficiency and specificity of this reagent (Figs. and K48C-ubiquitin is shown in lanes 1 and 2 of Fig. 5. This 2 and 3), we selected ethyleneimine as the alkylating agent in 2-ml incubation, carried out at pH 7.3, contained 40 mg of total large scale synthesis of polyubiquitin chains (Fig. 1 and below). ubiquitin; dimer synthesis was complete in 4 h. The reaction Two additional features of the functional data (Table I), was faster at pH 8, and this condition was used in subsequent although not directly relevant to our original objectives, de- synthetic reactions. Up to 60 mg of each monomer has been serve comment. First, the quantitative effect of blocking polyu- combined in such first-cycle reactions. biquitin chain formation was substrate-dependent; mutation of 48 125 Lys inhibited more strongly with I-lactalbumin than with I-lysozyme (Table I). Similarly, Hershko and co-workers (9, R. Beal, D. Toscano-Cantaffa, P. Young, and C. Pickart, manuscript 43) reported that reductive methylation of ubiquitin, which in preparation. blocks chain formation, suppresses I-lactalbumin degrada- Y. Lam, G. De Martino, C. Pickart, and R. Cohen, manuscript in tion more strongly than I-lysozyme degradation. The stron- preparation. 23718 Inhibition of 26 S Proteasome by Polyubiquitin Chains Synthesis of double-capped Ub is shown in lanes 3 and 4 of Fig. 5. This 0.7-ml incubation, carried out at pH 8, contained 7 mg of each single-deblocked Ub derivative; synthesis was es- sentially complete in 1 h. Up to 40 mg of each dimeric deriva- tive has been combined in second-cycle reactions. An additional round of deblocking and synthesis gave rise to double-blocked Ub (Fig. 5, lanes 7 and 8). In any given incubation, formation of the chain product was absolutely dependent upon the presence of both appropriately deblocked reactants, as shown for Ub synthesis in lanes 5 and 6 of Fig. 5. This last result confirms the very high linkage specificity of E2-25K (29, 38). A trace of apparent Ub in lane 4 (Fig. 5) may be because of the loss of Asp from the proximally capped dimer catalyzed by an E. coli carboxypeptidase that is present in trace amounts in some E2-25K preparations. Syn- thesis was very efficient; when the concentrations of the input reactants were equal, there was nearly complete conversion to the expected product (Fig. 5, lanes 3, 4, and 7, 8). Ub and Ub products were resolved from remaining reac- 2 4 tants by gradient cation exchange chromatography; Ub was purified by gel filtration (see “Experimental Procedures”). Typ- ically, the major chain product was obtained in about 60% yield FIG.6. Inhibition of the 26 S proteasome by unanchored Lys - from cation exchange columns, although yields were sometimes linked polyubiquitin chains. Degradation by purified 26 S protea- as high as 80%. Recoveries from the deblocking reactions were somes was assayed using purified H -ubiquitin- I-lysozyme conju- gates (see “Experimental Procedures”). The initial rate of conjugate higher: nearly 100% for the yeast ubiquitin hydrolase-1 reac- degradation in the presence of chains is expressed relative to the rate in tion and ;90% for the ethyleneimine reaction. a control reaction that lacked added polyubiquitin chains. Assays were Binding of Engineered Chains Inhibits Conjugate Degrada- supplemented with the indicated concentrations of Ub (L), Ub (M), 2 3 Ub (E), Ub (ƒ). Ub was double-blocked; in most cases Ub was also tion by the 26 S Proteasome—Because the rates of ubiquitina- 4 8 8 4 double-blocked (see “Results and Discussion”). The solid circle shows tion and de-ubiquitination, as well as the rate of conjugate data for wild type Ub (average of duplicate determinations). Lines for degradation, may contribute to the overall rate of degradation Ub (open circles) and Ub (inverted triangles) were obtained by fitting 4 8 in fraction II, it is difficult to draw quantitative conclusions the data using the program Enzfitter (Ub , 27.6 6 4.3 mM;Ub , 4.8 6 1.0 4 8 about conjugate recognition based solely on steady-state meas- mM). Points for Ub at 75 mM and Ub at 58 mM were omitted when fitting 4 8 the data. Points with error bars show the mean 6 S.D. of three or more urements in this system (above). We showed previously that a determinations. Points for Ub and Ub are averages of duplicate de- 48 2 3 mixture of unanchored Lys -linked chains inhibited overall terminations. Most other points are single determinations. For a sum- degradation in fraction II, indicating that chains compete with mary of K values, see Table II. 0.5 polyubiquitinated substrates for binding to chain recognition components of the 26 S proteasome (17). A higher resolution for binding. This conclusion is consistent with the the results version of this assay provided a convenient monitor of the shown in Table I. binding of engineered chains to the proteasome. These experi- Length Dependence of Inhibition—The rate of conjugate deg- ments utilized purified mammalian 26 S proteasomes (18) and radation is known to be facilitated by the presence of multiple purified I-lysozyme conjugates made with H -ubiquitin (see ubiquitins (6 – 8, 10, 16; Table I). In the case of conjugates “Experimental Procedures”). bearing chains, degradation is thought to increase with chain Ub inhibited the purified proteasome with a hyperbolic con- length. This may be caused both by enhanced binding of long centration dependence that corresponded to K 5 27.6 6 4.3 0.5 chains and their resistance to disassembly by a proteasome- mM (open circles, Fig. 6). Western analysis with anti-ubiquitin associated isopeptidase (45). With regard to binding, no quan- antibodies indicated that Ub was stable during the assay (data titative studies have addressed the form or basis of the pre- not shown). Inhibition must thus be caused by competition with sumptive length dependence. We show in Fig. 6 (inverted substrate conjugates. The presence of the proximal (Asp ) and triangles) that Ub inhibited the proteasome with a hyperbolic distal (Cys ) blocking residues did not influence binding be- concentration dependence that corresponded to K 5 4.8 6 1.0 0.5 cause similar inhibition was seen with double-blocked Ub 4 mM. This value is 5.8-fold smaller than the K for Ub . Given 0.5 4 versus each of the two single-blocked species (at 29 mM; these the errors in the respective measurements, the ratio could be as data are included in the open circles in Fig. 6). Failure of the high as 8 but no lower than 4. chain termini to influence binding is consistent with the expec- Inhibition by shorter chains was examined in a preliminary tation that substrate conjugates, which bear a macromolecule way. These chains were assembled from wild type ubiquitin; at the proximal chain terminus, will bind primarily through the absence of the S-aminoethylcysteine moiety should not interaction with the polyubiquitin chain. A series of hydropho- influence binding (above). As shown in Fig. 6, 78 mM Ub caused bic patches which has been implicated in the binding of chains ;50% inhibition (square), whereas 117 mM Ub caused only to the proteasome lies on the surface of the chain (17); this is ;21% inhibition (diamond). The observed inhibition is consist- also consistent with the observed absence of end effects. ent with K values of ;80 mM for Ub and ;430 mM for Ub , 0.5 3 2 Fig. 6 (filled circle) also shows the effect of wild type Ub at assuming a hyperbolic concentration dependence as is seen 29 mM. This chain had normal isopeptide linkages, with K48R- with longer chains. ubiquitin at the distal position (to facilitate synthesis; see The affinity of polyubiquitin chains for the proteasome thus “Experimental Procedures”). If anything, the wild type tet- increases ;90-fold as n increases from 2 to 8 (Table II). This is ramer inhibited more weakly than the engineered tetramers, substantially steeper than the chain length dependence ob- confirming that the presence of one or two S-aminoethylcys- served in the proteasome-catalyzed degradation of a series of teine moieties in the engineered tetramer was fully permissive a-globin conjugates (45). In the latter study, there was a rate Inhibition of 26 S Proteasome by Polyubiquitin Chains 23719 TABLE II Length dependence of inhibition of 26 S proteasome by unanchored polyubiquitin chains Data are taken from Fig. 6. For Ub and Ub , values of K were 2 3 0.5 calculated based on the inhibition observed at a single concentration of the respective chain. 0.5 Observed value Relative value mM 2 430 89.6 3 80 16.7 4 27.6 6 4.3 5.8 SCHEME I. Model of Ub which is based on the crystal structure 8 4.8 6 1.0 1 of Ub (48). The scheme depicts, as small circles, a series of surface 8 44 hydrophobic patches composed of the side chains of residues Leu , Ile , and Val . The solid circles are patches facing the viewer; the stippled increase of ;2-fold as n increased from 1 to 6 (degradation in circles are patches on the opposite face of the chain. the presence of Ubal (45)). This weaker dependence on n is probably caused by the structures of the a-globin conjugates, An alternative model postulates that two adjacent hydropho- which are linked to monoubiquitin(s) at low n values, and by a bic patches on the same face of the chain represent a minimal mixture of monoubiquitins and short chains at higher n values recognition element. The two ubiquitins bearing these patches, (46). It is likely that conjugates linked to monoubiquitin inter- e.g. ubiquitins 2 and 4 in Scheme I, are not directly covalently act with the proteasome differently than conjugates linked to linked. Therefore Ub lacks a recognition element and should polyubiquitin chains. cause negligible inhibition, as observed. Ub , if its conforma- Our results bear on the mechanisms by which polyubiquitin tion is Ub -like, contains one element, Ub contains two, and 4 4 chains may target substrates to the proteasome. The data Ub contains six. Ub inhibits about two times more weakly 8 3 unambiguously exclude a model in which the chain simply than predicted by this model; this could reflect the inability of amplifies the local concentration of monoubiquitin because this Ub to replicate faithfully the conformation seen in the Ub 3 4 model predicts that K will decrease in direct proportion to crystal (48). Like the previous model, this model predicts that 0.5 chain length; instead, K decreases by ;90-fold as n increases the affinity of Ub should be three times that of Ub , which is 0.5 8 4 by 4-fold (Table II). A model in which monoubiquitin is the close to the lower limit of 4 which can be accommodated by the recognition element is also inconsistent with the failure of current results (above). monoubiquitin to compete with polyubiquitin chains (see “Ex- These considerations serve to indicate that the enhanced perimental Procedures”). The results are generally consistent interaction of longer chains with the proteasome can be ex- with a model in which assembly of ubiquitin into a chain plained by a model in which the structure of the chain serves to creates and amplifies a specific structural element that is rec- create and amplify a minimum binding element that includes ognized by one or more binding components in the 19 S regu- two or more hydrophobic patches. Studies on the interaction of latory complex. A satisfying feature of this model is that it longer chains (n . 8) and mutant chains (below) with the could allow monoubiquitin to serve distinct functions from proteasome will be helpful in defining the precise nature of this polyubiquitin chains (e.g. 47). element or in developing alternative models. As noted above, When immobilized on a nitrocellulose membrane, subunit the observed ratio of K values for Ub and Ub is somewhat 0.5 4 8 S5a of the 19 S complex exhibits negligible binding of chains larger than the ratio predicted by two of the models discussed shorter than Ub , whereas for chains of n $ 4 binding increases above. However, “fraying” of the ends of the chain, which prob- steeply with chain length (23). This behavior led to the sugges- ably occurs in solution, would disproportionately weaken the tion that Ub harbors a minimum recognition element for S5a binding of short chains, leading to an increase in this ratio. (17, 23). We considered whether our data could be consistent In previous work we synthesized several Ub molecules bear- with a similar model for the proteasome. Scheme I shows a ing L8A,I44A-ubiquitin at one or more defined positions within model of Ub which is based on the crystal structure of Ub the chain. Because chains assembled solely from this mutant 8 4 (48). The scheme depicts a series of surface hydrophobic ubiquitin cannot target substrates for degradation (17), such 8 44 patches composed of the side chains of residues Lys , Ile , and defined oligomers may provide a way to test the models out- Val . Mutation of pairs of these residues to Ala abolishes lined above. We measured binding of these molecules to S5a binding of polyubiquitin chains to the proteasome and to iso- immobilized on a nitrocellulose membrane; the results (17) lated S5a, suggesting that these patches are part of a chain- agreed best with a four-ubiquitin recognition element. How- based recognition element (17). ever, the poor sensitivity of this binding assay, and the finding For the simple case in which Ub represents the actual that the yeast S5a homolog is nonessential (26), indicate that recognition element, and longer chains are assumed to have the these issues should be re-addressed with the 26 S proteasome. conformation seen in the Ub crystal, there are three elements Application of the method described here will facilitate the in Ub (ubiquitins 1– 4, 3– 6, and 5– 8 in Scheme I). This model assembly of mixed wild type/mutant polyubiquitin chains. predicts that Ub will bind three times more tightly than Ub ; Ubiquitin is present in cells at a concentration of ;20 mM,of 8 4 this factor is close to the lower limit of 4 which can be accom- which about half is in the form of conjugates (49). Although the modated by our binding data (above). The model is also con- length of the chain on a typical conjugate undergoing degrada- sistent with the observed negligible inhibition by Ub . Signifi- tion is not known, substrate-linked chains of n . 8 are difficult cant inhibition by Ub , although not predicted, might arise if to detect (e.g. 10, 16). Thus, it is reasonable to expect that Ub 3 8 the conformation of Ub is Ub -like, and a partial recognition should constitute an efficient targeting signal. The question 3 4 element is partially functional. A related model, in which ubiq- arises as to whether the binding reported here for Ub (K ; 8 0.5 uitins 2–5 and 4 –7 (Scheme I) are considered to be identical to 4 mM, Fig. 6) is sufficient to allow rapid degradation of poly- the elements specified above, predicts that Ub will bind five ubiquitinated proteins in vivo. This question cannot be an- times more tightly than Ub ; this prediction is in good agree- swered with certainty, but it is probable that affinities only ment with the data. modestly higher than those observed would be adequate be- 23720 Inhibition of 26 S Proteasome by Polyubiquitin Chains cause the high level of the proteasome in cells (;1% of cell E2-25K to conjugate engineered chains to the a-ubiquitin moi- protein (50)) means that recognition site(s) should be present at ety in a recombinant fusion protein such as ubiquitin-dihydro- a concentration of 1–2 mM. folate reductase or ubiquitin-b-galactosidase. Studies are in The K values measured in the current study may under- progress to explore the feasibility of this approach, which might 0.5 estimate the true affinities of polyubiquitinated conjugates for be facilitated by the use of tagged chains. H -K48C-ubiquitin two reasons. First, observed K values will exceed true K behaves identically to K48C-ubiquitin in synthesis of engi- 0.5 i values by a factor related to [S]/K . Although the concentration neered chains (data not shown). of ubiquitin- I-lysozyme conjugates in our assays was only Finally, the approach taken here should be applicable to ;30 nM, K is unknown. In addition, conjugated fraction II chains linked through Lys residues other than Lys .Ifan proteins were present at an undetermined concentration (“Ex- enzyme can be identified to catalyze formation of the desired perimental Procedures”). A finding that Ub inhibits the deg- isopeptide bond, application of the methodology described here radation of a Ub -RNase A conjugate with K ; 10 mM sug- could facilitate the structural and functional characterization 4 0.5 gests that the current K values may underestimate true of such novel chains. 0.5 affinity. However, such competition effects would not change Acknowledgments—We are grateful to Eileen Kasperek for generat- the relative affinities of different length polyubiquitin chains ing the K48C-ubiquitin plasmid, to Amy Lam for the purified yeast (Table II). ubiquitin hydrolase-1 enzyme, and to Christopher N. Larsen for pro- Second, the presence of a substrate moiety may lead to viding pRSUbD. We thank Leni Moldovan for assistance in preparing the H ,K48C-ubiquitin plasmid and for help in preliminary studies on tighter binding of conjugates than of unanchored polyubiquitin ethyleneimine reactivity. chains. This would be beneficial in minimizing inhibition by the end products of degradation. Depending upon the nature of the REFERENCES proteasomal site(s), such an effect might arise through elimi- 1. Ciechanover, A. (1994) Cell 79, 13–21 nation of the negative charge at the proximal terminus of the 2. Hochstrasser, M. (1996) Annu. Rev. Genet. 30, 405– 439 3. Rubin, D. M., and Finley, D. (1995) Curr. Biol. 5, 854 – 858 polyubiquitin chain or by hydrogen bonding to the isopeptide 4. Hershko, A., Heller, H., Elias, S., and Ciechanover, A. (1983) J. Biol. Chem. amide. These types of interactions, although dependent upon 258, 8206 – 8214 5. Scheffner, M., Nuber, U., and Huibregtse, J. M. (1995) Nature 373, 81– 83 the presence of a substrate, would be independent of its iden- 6. Hershko, A., Leshinsky, E., Ganoth, D., and Heller, H. (1984) Proc. Natl. Acad. tity and thus consistent with a universal targeting function for Sci. U. S. A. 81, 1619 –1623 7. Hough, R. Pratt, G., and Rechsteiner, M. (1986) J. Biol. Chem. 261, 2400 –2408 polyubiquitin chains. 8. Shaeffer, J. R., and Kania, M. A. (1995) Biochemistry 34, 4015– 4021 Our data provide little information regarding the nature of 9. Hershko, A., and Heller, H. (1985) Biochem. Biophys. Res. Commun. 128, proteasomal polyubiquitin binding sites. Inclusion of Ubal in 1079 –1086 10. Chau, V., Tobias, J. W., Bachmair, A., Marriott, D., Ecker, D. J., Gonda, D. K., the assays eliminated the possibility of a contribution from a and Varshavsky, A. (1989) Science 243, 1576 –1583 proteasome-associated isopeptidase (45). A value of K ; 4 mM 0.5 11. Arnason, T., and Ellison, M. J. (1994) Mol. Cell. Biol. 14, 7876 –7883 for Ub is substantially weaker than K ; 20 nM reported for 12. Finley, D., Sadis, S., Monia, B. P., Boucher, P., Ecker, D. J., Crooke, S. T., and 8 0.5 Chau, V. (1994) Mol. Cell. Biol. 14, 5501–5509 binding of an E2-Ub conjugate to immobilized S5a (15). Be- 13. Spence, J., Sadis, S., Haas, A. L., and Finley, D. (1995) Mol. Cell. Biol. 15, cause the present assay would not have detected binding that is 1265–1273 14. Johnson, E. S., Ma, P. C. M., Ota, I. M., and Varshavsky, A. (1995) J. Biol. not relevant to conjugate degradation, our results could be Chem. 270, 17442–17456 consistent with the suggestion that S5a does not play a major 15. Baboshina, O. V., and Haas, A. L. (1996) J. Biol. Chem. 271, 2823–2831 16. Gregori, L., Poosch, M. S., Cousins, G., and Chau, V. (1990) J. Biol. Chem. 265, role in proteolytic targeting (26). On the other hand, incorpo- 8354 – 8357 ration into the 19 S regulatory complex could alter the binding 17. Beal, R., Deveraux, Q., Xia, G., Rechsteiner, M., and Pickart, C. (1996) Proc. properties of S5a. Further work will be necessary to resolve Natl. Acad. Sci. U. S. A. 93, 861– 866 18. Hoffman, L., Pratt, G., and Rechsteiner, M. (1992) J. Biol. Chem. 267, this issue. 22362–22368 Concluding Remarks—The ability to synthesize polyubiq- 19. Chu-Ping, M., Slaughter, C. A., and DeMartino, G. N. (1992) Biochim. Biophys. uitin chains of defined length and structure solves one of two Acta 1119, 303–311 20. Dubiel, W., Ferrer, K., and Rechsteiner, M. (1995) Mol. Biol. Rep. 21, 27–34 significant problems associated with generating biochemically 21. Lowe, J., Stock, D., Jap, B., Zwickl, P., Baumeister, W., and Huber, R. (1995) useful quantities of substrates for the 26 S proteasome. It re- Science 268, 533–539 22. Groll, M., Ditzel, L., Lowe, J., Stock, D., Bochtler, M., Bartunik, H. D., and mains necessary to link these chains to a suitable target pro- Huber, R. (1997) Nature 386, 463– 471 tein. In principle, it should be possible to use a recombinant E2 23. Deveraux, Q., Ustrell, V., Pickart, C., and Rechsteiner, M. (1994) J. Biol. Chem. 269, 7059 –7061 protein to conjugate engineered chains to an in vitro substrate 24. Van Nocker, S., Deveraux, Q., Rechsteiner, M., and Vierstra, R. D. (1996) Proc. (41). However, because the target specificity of unassisted E2 Natl. Acad. Sci. U. S. A. 93, 856 – 860 conjugation is highly restricted (e.g. 51), it is unlikely that this 25. DeMartino, G. N., Moomaw, C. R., Zagnitko, O. P., Proske, R. J., Chu-Ping, M., Afendis, S. J., Swaffield, J. C., and Slaughter, C. A. (1994) J. Biol. Chem. approach will be broadly useful. For example, we used recom- 269, 20878 –20884 binant E2-20K (52) to conjugate engineered Ub to I-histone 4 26. Van Nocker, S., Sadis, S., Rubin, D. M., Glickman, M., Fu, H., Coux, O., Wefes, 2B, but the resulting conjugate was degraded at a negligible I., Finley, D., and Vierstra, R. D. (1996) Mol. Cell. Biol. 16, 6020 – 6028 27. Cox, M. J., Shapira, R., and Wilkinson, K. D. (1985) Anal. Biochem. 154, rate by the purified 26 S proteasome (data not shown). This is 345–352 not unexpected based on the low rate of histone degradation 28. Chen, Z., and Pickart, C. M. (1990) J. Biol. Chem. 265, 21835–21842 29. Pickart, C. M., and Vella, A. T. (1988) J. Biol. Chem. 263, 12028 –12035 observed in fraction II (53). Overall, it is likely that E3 en- 30. Haldeman, M. T., Xia, G., Kasperek, E. M., and Pickart, C. M. (1997) Biochem- zymes, together with their specific E2 partners, will represent istry, 36, in press 31. Miller, H. I., Henzel, W. J., Ridgway, J. B., Kuang, W.-J., Chisholm, V., and a more effective route to the ligation of engineered chains to Liu, C.-C. (1989) Bio/Technology 7, 698 –704 substrates. 32. Mayer, A. N., and Wilkinson, K. D. (1989) Biochemistry 28, 166 –172 One possible exception to an E3 requirement for generating 33. Dunten, R. L., and Cohen, R. E. (1989) J. Biol. Chem. 264, 16739 –16747 34. Ecker, D. J., Butt, T. R., Marsh, J., Sternberg, E. J., Margolis, N., Monia, B. P., proteasome substrates is provided by E2-25K itself. Several Jonnalagadda, S., Khan, M. I., Weber, P. L., Mueller, L., and Crooke, S. T. noncleavable ubiquitin fusion proteins are known to undergo (1987) J. Biol. Chem. 262, 14213–14221 polyubiquitination at Lys of the a-linked ubiquitin moiety, 35. Burch, T. J., and Haas, A. L. (1994) Biochemistry 33, 7300 –7308 36. Haas, A. L., and Wilkinson, K. D. (1985) Prep. Biochem. 15, 49–60 and conjugates of this structure are degraded by the 26 S 37. Ellman, G. L. (1959) Arch. Biochem. Biophys. 82, 70 –77 proteasome in yeast cells (54, 55). It should be possible to use 38. Pickart, C. M., Haldeman, M. T., Kasperek, E. M., and Chen, Z. (1992) J. Biol. Chem. 267, 14418 –14223 39. Cole, R. D. (1967) Methods Enzymol. 11, 315–317 40. Chen, Z., Niles, E. G., and Pickart, C. M. (1991) J. Biol. Chem. 266, W. Xu and R. Cohen, unpublished experiments. 15698 –15704 Inhibition of 26 S Proteasome by Polyubiquitin Chains 23721 41. van Nocker, S., and Vierstra, R. D. (1993) J. Biol. Chem. 268, 24766 –24773 49. Haas, A. L., and Bright, P. M. (1985) J. Biol. Chem. 260, 12464 –12473 42. Cox, M. J., Haas, A. L., and Wilkinson, K. D. (1986) Arch. Biochem. Biophys. 50. Tanaka, K., Ii, K., Ichihara, A., Waxman, L., and Goldberg, A. L. (1986) J. Biol. 250, 400 – 409 Chem. 261, 15197–15203 43. Hershko, A., Ganoth, D., Pehrson, J., Palazzo, R. E., and Cohen, L. H. (1991) 51. Pickart, C. M., and Rose, I. A. (1985) J. Biol. Chem. 260, 1573–1581 J. Biol. Chem. 266, 16376 –16379 52. Wefes, I., Kaiser, P., Schneider, R., Pickart, C. M., and Finley, D. (1995) Gene 44. Dunten, R. L., Cohen, R. E., Gregori, L., and Chau, V. (1991) J. Biol. Chem. (Amst.) 163, 321–322 266, 3260 –3267 53. Haas, A., Reback, P. M., Pratt, G., and Rechsteiner, M. (1990) J. Biol. Chem. 45. Lam, Y. A., Xu, W., DeMartino, G. N., and Cohen, R. E. (1997) Nature 385, 265, 21664 –21669 737–740 54. Johnson, E. S., Bartel, B., Seufert, W., and Varshavsky, A. (1992) EMBO J. 11, 46. Shaeffer, J. R. (1994) J. Biol. Chem. 269, 29530 –29536 487–505 47. Finley, D., Bartel, B., and Varshavsky, A. (1989) Nature 338, 394 – 401 55. Johnson, E. S., Ma, P. C. M., Ota, I. M., and Varshavsky, A. (1995) J. Biol. 48. Cook, W. J., Jeffrey, L. C., Kasperek, E. M., and Pickart, C. M. (1994) J. Mol. Chem. 270, 17442–17456 Biol. 236, 601– 609 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

Inhibition of the 26 S Proteasome by Polyubiquitin Chains Synthesized to Have Defined Lengths

Download PDF
10 pages

Loading next page...
 
/lp/unpaywall/inhibition-of-the-26-s-proteasome-by-polyubiquitin-chains-synthesized-krWE8LRTEB

References (62)

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.272.38.23712
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 272, No. 38, Issue of September 19, pp. 23712–23721, 1997 © 1997 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Inhibition of the 26 S Proteasome by Polyubiquitin Chains Synthesized to Have Defined Lengths* (Received for publication, June 2, 1997) Julia Piotrowski‡, Richard Beal§, Laura Hoffman¶, Keith D. Wilkinsoni, Robert E. Cohen**, and Cecile M. Pickart‡ ‡‡ From the ‡Department of Biochemistry, School of Hygiene and Public Health, Johns Hopkins University, Baltimore, Maryland 21205, the §Department of Biochemistry, School of Medicine, State University of New York, Buffalo, New York 14214, the ¶Department of Biochemistry, University of Utah Medical School, Salt Lake City, Utah 84132, the iDepartment of Biochemistry, School of Medicine, Emory University, Atlanta, Georgia 30322, and the **Department of Biochemistry, College of Medicine, University of Iowa, Iowa City, Iowa 52242 Ubiquitin is a covalent signal that targets cellular pro- tion of high volume and high selectivity: most short lived pro- teins to the 26 S proteasome. Multiple ubiquitins can be teins are degraded in this pathway, but the half-lives of indi- ligated together through the formation of isopeptide vidual substrate proteins can be regulated acutely and 48 76 bonds between Lys and Gly of successive ubiquitins. independently. Such a polyubiquitin chain constitutes a highly effective Ubiquitin acts as a degradation signal by virtue of covalent proteolytic targeting signal, but its mode of interaction ligation to target proteins. Ubiquitination occurs through the with the proteasome is not well understood. Experi- formation of an isopeptide bond between the COOH terminus ments to address this issue have been limited by diffi- 76 of ubiquitin (Gly ) and an internal Lys residue of the target culties in preparing useful quantities of polyubiquitin protein. This modification confers recognition by the multisub- chains of uniform length. We report a simple method for unit 26 S proteasome; the target protein is degraded, but ubiq- large scale synthesis of Lys -linked polyubiquitin uitin is regenerated for use in subsequent proteolytic cycles chains of defined length. In the first round of synthesis, (1–3). Specificity in ubiquitin-mediated proteolysis appears to two ubiquitin derivatives (K48C-ubiquitin and Asp - arise primarily in the ubiquitin attachment step, which in- ubiquitin) were used as substrates for the well charac- volves the sequential formation of ubiquitin thiol ester adducts terized ubiquitin-conjugating enzyme E2-25K. Diubiq- of ubiquitin-activating (E1), -conjugating (E2), and ligase (E3) uitin blocked at the nascent proximal and distal chain enzymes (1, 4, 5). Recent studies suggest that for a given termini was obtained in quantitative yield. Appropri- substrate, a specific E3 carries out the substrate ubiquitination ately deblocked chains were then combined to synthe- step, whereas a specific E2 charges the E3 with ubiquitin (5). size higher order chains (tetramer and octamer in the The ligation of multiple ubiquitins increases the rate of sub- present study). Deblocking was achieved either enzy- strate degradation (6 – 8), although the exact nature of the matically (proximal terminus) or by chemical alkylation (distal terminus). Chains synthesized by this method dependence is unclear. Multiple ubiquitination can occur were used to obtain the first quantitative information through the ligation of ubiquitin monomers to several sub- concerning the influence of polyubiquitin chain length strate Lys residues (9) but more typically involves the assembly on binding to the 26 S proteasome; this was done on the substrate of a polymeric, isopeptide-linked ubiquitin through comparison of different length (unanchored) chain (10). Multiple Lys residues of ubiquitin, including Lys , 11 29 48 63 polyubiquitin chains as inhibitors of ubiquitin-conju- Lys , Lys , Lys , and Lys , can serve as sites of polyubiq- gate degradation. K was found to decrease ;90-fold, 0.5 uitin chain initiation and/or elongation (10 –15). However, from 430 to 4.8 mM, as the chain was lengthened from two 48 Lys -linked chains represent the most commonly utilized deg- to eight ubiquitins. The implications of these results for radation signal in the ubiquitin pathway. This conclusion is the molecular basis of chain recognition are discussed. supported by several lines of evidence, including the results of in vitro biochemical analyses (10, 16) and the lethality of the K48R mutation in Saccharomyces cerevisiae (12). The targeting The conserved protein ubiquitin functions in diverse biolog- 48 ability of Lys -linked polyubiquitin chains apparently arises ical processes, including oncogenesis, cell cycle progression, from their high affinity for the 26 S proteasome, which may be antigen presentation, and programmed cell death (for review, due in part to the exposure of a regular array of hydrophobic see Refs. 1 and 2). These and most other functions of ubiquitin patches on the chain surface (17). reflect its role as an essential cofactor in an energy-dependent The 26 S proteasome is assembled from catalytic (20 S) and proteolytic pathway whose hallmarks are an unusual combina- regulatory (19 S) subcomplexes (3, 18 –20). The crystal struc- tures of 20 S proteasomes from archaebacteria and yeast show that access to the proteolytic active sites is highly restricted * This research was supported in part by National Institutes of (21, 22). Thus the target protein must be unfolded before deg- Health Grants DK46984 (to C. M. P), GM30308 (to K. D. W.), GM37666 (to R. E. C.), and GM37009 and the Lucille P. Markey Charitable Trust radation. Targeting of the ubiquitinated substrate to the pro- (to M. Rechsteiner). The costs of publication of this article were de- teasome is an activity of the 19 S complex, as suggested by the frayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The abbreviations used are: E1, ubiquitin-activating enzyme; E2, ‡‡ Recipient of a National Institutes of Health research career devel- ubiquitin-conjugating enzyme; E3, ubiquitin-protein ligase; Ubal, ubiq- opment award. To whom correspondence should be addressed: Dept. of uitin aldehyde; IPTG, isopropyl-b-D-thiogalactopyranoside; H , hexa- Biochemistry, Johns Hopkins University, 615 North Wolfe St., Balti- histidine; DTT, dithiothreitol; Bis-Tris, 2-[bis(2-hydroxyethyl)amino]-2- more, MD 21205. Tel.: 410-614-4554; Fax: 410-955-2926; E-mail: (hydroxymethyl)-propane-1,3-diol; Ub , polyubiquitin chain composed [email protected]. of n ubiquitins; ATPgS, adenosine 59-O-(thiotriphosphate). 23712 This paper is available on line at http://www.jbc.org This is an Open Access article under the CC BY license. Inhibition of 26 S Proteasome by Polyubiquitin Chains 23713 into pDS78 from which this same fragment had been excised. Plasmid ubiquitin independence of protein degradation catalyzed by the pDS78, provided by M. Rechsteiner (University of Utah) encodes an 20 S proteasome (3, 18 –20). The 19 S complex contains a polyu- H -tagged version of human ubiquitin under the control of an IPTG- biquitin chain-binding protein known as S5a, MBP1, or MCB1 inducible promoter. The BglII site in pDS78 occurs just after the H tag; (23, 24), and multiple subunits harboring ATP binding sites (3, the NH -terminal sequence is MHHHHHHGEFQ, where Q corresponds 25). However, most of the subunits of the 19 S complex are to Q2 in wild type human ubiquitin. functionally uncharacterized. One or more of these subunits Ubiquitin Expression and Purification—Expression of K48C-, K48R-, and Asp -ubiquitins (encoded by pET/pRSET-based plasmids) was in- must be an additional polyubiquitin-binding protein because a duced by the addition of 0.4 mM IPTG to cultures of appropriately yeast mcb1D strain is viable and competent in ubiquitin-medi- transformed E. coli BL21(DE3)pLysS cells growing at 37 °C. IPTG was ated proteolysis (26). added once an A of ;0.6 had been reached, and growth was continued Biochemically useful quantities of substrates, i.e. polyubiq- for 4 h more. Cells were harvested, frozen, and thawed in buffer con- uitinated target proteins, are a prerequisite for dissecting the taining 50 mM Tris-HCl (24% base) and 0.4 mg/ml lysozyme, supple- mechanistic coordination of chain recognition, substrate un- mented with 1 mM EDTA and 10 mM DTT in the case of K48C-ubiquitin (2 ml of buffer/g of cells). Efficient lysis occurred over several minutes. folding, and peptide bond hydrolysis by the 26 S proteasome. DNase I (20 mg/ml) and MgCl (10 mM) were then added to digest DNA. Ideally these substrates should be homogeneous by several Expression of H -tagged ubiquitins in strain M15 (harboring the lac different criteria: overall purity, polyubiquitin chain length, repressor-expressing plasmid pDMI,1) was carried out similarly, except chain linkage, and site of chain ligation to the target protein. So that cells were lysed using a French press. In all cases, crude soluble far, such well defined substrates have been unattainable be- lysates were produced by centrifugation at 9,000 3 g for 20 min. cause of the low purity and abundance of E3 enzymes, as well Non-ubiquitin proteins were precipitated at 0 °C with perchloric acid (3.5% w/v). Ubiquitin was resolved from the dialyzed supernatant by as the uncontrolled character of chain elongation as catalyzed cation exchange chromatography on S-Sepharose (Pharmacia Biotech by available E2 and E3 enzymes. To overcome the chain elon- Inc.) as described (36) (K48C- and Asp -ubiquitins) or by gradient gation problem, we have implemented a novel method that elution from S-Sepharose at pH 6.1 (H -tagged ubiquitins). For K48C utilizes the well characterized ubiquitin-conjugating enzyme mutants, all steps were carried out in the presence of 1 mM DTT and 0.1 E2-25K to generate Lys -linked polyubiquitin chains of de- mM EDTA. The ubiquitin derivatives utilized in this work were recov- fined length. This method avoids uncontrolled polymerization ered at 25–100 mg of purified protein/liter of cell culture. Ubiquitin concentrations were determined by A , assuming an absorbance of by utilizing, in each round of synthesis, two chain reactants, 0.16 for a 1 mg/ml solution (36). one reversibly capped at its proximal, and the other at its Analytical Alkylation Kinetics—Kinetics were monitored at 37 °C in distal, terminus. This method generates “unanchored” chains incubations containing 0.2 M Tris-HCl (50% base), 1 mM EDTA, 1 mM that bind to the 26 S proteasome and inhibit the degradation of cysteine, or 0.5 mM K48C-ubiquitin, and 0 – 64 mM alkylating reagent; polyubiquitinated lysozyme. This inhibition assay was used to or Tris (pH 8.0), was replaced by Bis-Tris propane (pH 8.5). At timed evaluate, for the first time, the relative binding of different intervals, aliquots were assayed for remaining thiol with dithionitro- benzoate (37). Values of k were obtained from linear plots of ln A /A length chains to the 26 S proteasome. obs t 0 against time and were corrected by subtracting the value of k meas- obs EXPERIMENTAL PROCEDURES ured in the absence of alkylating agent. Preparative Alkylation (Distal Terminus Deblocking)—For the exper- Ethyleneimine was purchased from ChemService, stored at 5 °C, and iments shown in Table I and Fig. 3, conditions were as follows. Reac- used within several days of opening the sealed ampule. Except where tions with ethyleneimine contained 50 mM ethyleneimine and the pu- noted, other reagents were purchased from Sigma. Protein iodination rified ubiquitin chain at 2–20 mg/ml (#1 ml). Other conditions were the with chloramine T was carried out as described (4). Tryptic digestion of same as for kinetic measurements (above). After 1 h, the reaction ubiquitin derivatives (5 mg/ml total ubiquitin and 5% w/w trypsin) and mixture was dialyzed against 5 mM ammonium acetate (pH 5.5), 0.1 mM reversed phase high performance liquid chromatography separation of EDTA, and 1 mM DTT to remove residual ethyleneimine. Alkylation of peptides were carried out as described (27, 28). K48C-ubiquitin (1 mg/ml) with bromoethylamine (50 mM, Aldrich) was Enzymes and Proteins—Fraction II was prepared from rabbit reticu- carried out similarly. For alkylation of K48C-ubiquitin (1 mg/ml) with locyte-rich whole blood (4). E1 was purified from bovine erythrocytes or TM N-(iodoethyl)trifluoroacetamide (Aminoethyl-8 , Pierce), 3.4-h incu- rabbit liver (29). Recombinant bovine E2-25K harboring two catalyti- bations were done at 50 °C in 0.2 M Bis-Tris propane (as above). The cally silent mutations was expressed in Escherichia coli using the alkylating agent was dissolved in methanol and added in two equal vector pET3d-C170S,F174L-25K and purified as described (30). Yeast portions at time zero and1hto give a final concentration of 3.1 mM. ubiquitin hydrolase-1 was prepared by a slight modification of the After addition of DTT to 0.25 mM, the reaction was dialyzed as above. method described previously (31). Ubal was prepared by either of two Proximal Terminus Deblocking—The purified ubiquitin chain (2–50 methods described previously (32, 33). mg/ml, #1 ml) was incubated with 15 mg/ml yeast ubiquitin hydrolase-1 Ubiquitin Plasmids—All coding sequences were derived from the for1hat37°Cin50mM Tris-HCl (24% base), 0.1 mM EDTA, and ;0.5 synthetic human ubiquitin gene described by Ecker et al. (34). Plasmids mM DTT. The reaction was passed through a 0.5-ml Q-Sepharose col- encoding K48C- and K48R-ubiquitin were prepared by mutagenesis of umn (Pharmacia; preequilibrated in the same buffer) to absorb the pPLhUb (35) as described (17). The coding regions of the resulting enzyme; the flow-through was collected into a Centricon-10 microcon- pPLhUb plasmids were amplified using flanking primers Ub5F (59- centrator (Amicon). The column was washed with 3 volumes of buffer. tgcatttatttgcatacattca-39) and Ub3F (59-cgaattcgagctcggatcctcaa-39). The combined flow-through and wash fractions were concentrated. The resulting products were digested with NdeI and BamHI and sub- Protein Degradation Assays—The production of acid-soluble radioac- cloned into NdeI/BamHI-restricted pET3a to produce pET3a-UbK48C 125 125 6 5 tivity from I-lysozyme or I-lactalbumin (;10 cpm/mg; ;10 cpm/ and pET3a-UbK48R. 25-ml incubation) was monitored in incubations with rabbit reticulocyte The plasmid pRSUbD, encoding Asp -ubiquitin, was prepared as fraction II (;1.5 mg/ml protein) as described (17). Rates obtained in the follows. A vector library encoding a variety of single amino acids COOH- presence of wild type ubiquitin or ubiquitin derivative (12 mM) were terminal to ubiquitin was constructed by polymerase chain reaction. To corrected using blanks obtained by omitting ubiquitin. In some cases create this amino acid library at position 77, the coding region of aliquots were withdrawn during the steady state of degradation for pRSUb was amplified with a degenerate 39-primer which contained all analysis of the levels of ubiquitin-substrate conjugates after electro- possible codons followed by a stop codon and a HindIII site. The primer phoresis and autoradiography (17). For the experiment involving dis- sequences were 59-atccatatgcagatcttcg-39 and 59-caagcttctaNNNaccac- assembly of engineered Ub (see the last entry in Table I), 4.5 mM Ub cacgaagtc-39. The polymerase chain reaction products from this reaction 4 4 (potentially yielding 18 mM monoubiquitin) was preincubated for 30 min were subcloned en masse into pCRII (Invitrogen). Plasmids containing under the conditions of the degradation assay, except that ATP, regen- inserts were sequenced, and an insert encoding Asp at the COOH erating system, and labeled substrate were omitted. These components terminus was subcloned into pRSET using the NdeI and HindIII sites. were added to initiate the assay. For the control, the preincubation A plasmid for IPTG-induced expression of H ,K48C-ubiquitin was contained 18 mM wild type monoubiquitin. prepared by ligating the KpnI-BglII fragment (34) from pPLhUb-K48C Engineered Polyubiquitin Chains: Synthesis and Purification—Incu- bations of 0.2–2.0 ml contained 50 mM Tris-HCl (50% base), 5 mM C. Larsen and K. Wilkinson, unpublished data. MgCl ,10mM phosphocreatine, 0.6 unit/ml each of creatine phosphoki- 2 23714 Inhibition of 26 S Proteasome by Polyubiquitin Chains FIG.1. Method for in vitro synthesis of defined length polyubiquitin chains. Ub is made by incubating E2- 25K with equal concentrations of ubiq- uitin capped at the future proximal chain terminus (Asp -ubiquitin, open circle), and ubiquitin capped at the future distal chain terminus (K48C-ubiquitin, filled circle). The proximal or distal terminus of the resulting dimer is deblocked by treat- ment with yeast ubiquitin hydrolase-1 (YUH1) or ethyleneimine, respectively. In the second cycle of synthesis, Ub is made by incubating the two single-deblocked dimers with E2-25K. See “Results and Discussion.” nase and inorganic pyrophosphatase, 4 mM ATP, and 0.5–1 mM DTT mM MgCl , 10% v/v glycerol, 10 mM DTT, 1.5 or 3 mM Ubal, 2 mM ATP, (37 °C); E1 and C170S,F174L-E2-25K were 0.2 mM and 20 mM, respec- 10 mM phosphocreatine, 0.3 unit/ml creatine phosphokinase, 0.3 tively. The two chain reactants were added at equal concentrations. unit/ml pyrophosphatase, 0.37 mg/ml 26 S proteasome (0.18 mM based Times required for completion of the conjugation reaction, usually 1– 4 on a molecular mass of 2.1 MDa), and ;5,000 – 8,000 cpm of I- h, were determined by examining reaction aliquots by SDS-polyacryl- lysozyme conjugates (;30 nM lysozyme; above). The lysozyme conju- amide gel electrophoresis. At the end of the incubation, the reaction was gates also contributed H -ubiquitin at a final concentration of ;10 mM. passed through a Q-Sepharose column (0.5–1 ml) at pH 7.6 to absorb Only a small fraction of this H -ubiquitin was conjugated to I-ly- the enzymes. In most cases the unabsorbed fraction was adjusted to pH sozyme; of the remainder, some was unconjugated, and some was con- 4.5 (for non-tagged chains) or 6.1 (for H -tagged chains) and applied to jugated to fraction II proteins (the conjugated:unconjugated ratio is an S-Sepharose column preequilibrated at the appropriate pH (Phar- unknown). Ubal was included to inhibit disassembly of polyubiquitin macia; 15 mg of ubiquitin/ml of resin), and the major chain product was chains by one or more isopeptidases in the 26 S proteasome prepara- purified by elution with a linear gradient of NaCl (0 – 0.5 M). Fractions tion; control experiments showed that 1.5–3 mM Ubal completely inhib- spanning the peak region were examined by SDS-polyacrylamide gel ited polyubiquitin chain disassembly but had no effect on the rate of electrophoresis and Coomassie staining before pooling. Concentration conjugate degradation. Incubations were quenched after 10 min by the and buffer exchange were carried out using Ultrafree-4 or -15 devices addition of trichloroacetic acid, and acid-soluble radioactivity was de- (Millipore). Ub and Ub were resolved by gel filtration on a 1 3 50-cm termined by counting an aliquot of the acid supernatant for 10 min. 4 8 column of Sephadex G-75 (Sigma) buffered with 50 mM ammonium Appropriate controls showed that degradation was linear with time for acetate (pH 5.5), 0.1 mM EDTA, and 1 mM DTT. ;15 min and that degradation was abolished if MgATP was omitted. Wild Type Polyubiquitin Chains—For Ub , the incubation (0.25 ml) Data were corrected by subtracting blanks derived from incubations in was carried out as described for engineered chains (above), except that which conjugates were replaced with unconjugated I-lysozyme. Con- the reactants were K48R-ubiquitin (;3 mg/ml) and Ub assembled from trol experiments showed that subjecting unconjugated I-lysozyme to wild type ubiquitin (;12 mg/ml). The Ub was largely des-Gly-Gly at its the same buffer sequence used in the Ni-NTA chromatography did not proximal terminus. After2hof incubation (37 °C), E1 and E2-25K were increase its susceptibility to degradation. Typically, 15–20% of the removed by passage through an anion exchange column (above), and conjugates were degraded in the positive control, and replicate assays Ub was purified by gradient elution (above) from a fast protein liquid agreed within 10 –15%. chromatography Mono S column (Pharmacia). Cation exchange was Because initial experiments showed that the addition of ubiquitin also used to resolve Ub and Ub from mixed chains assembled from chains to the assay significantly increased the recovery of acid-soluble 2 3 wild type ubiquitin (38). radioactivity by counteracting nonspecific absorption, all assays in- Preparation of H -Ubiquitin- I-Lysozyme Conjugates—Incubations cluded monoubiquitin as carrier. Most assays were supplemented with of 2 ml contained (pH 8.0, 37 °C) 50 mM Tris-HCl (50% base), 5 mM 230 mM monoubiquitin; in some cases, the concentration was 420 mM. MgCl ,5mM ATPgS (Boehringer Mannheim), 8 mM Ubal, 10 mM DTT, Data from the two sets of experiments could be fit by the same K 2 0.5 0.3 unit/ml inorganic pyrophosphatase, 3 mg/ml fraction II protein, 0.3 value for Ub , suggesting that monoubiquitin did not bind competi- 7 125 mg/ml H -ubiquitin, and ;4 3 10 cpm of I-lysozyme (40 –50 mg). tively with unanchored polyubiquitin chains. This was confirmed by After 2 h, 20 mM N-ethylmaleimide (to quench DTT and inactivate showing that ;50% inhibition by 29 mM Ub (see “Results”) was ob- ubiquitin pathway enzymes) was added together with 4 M urea and 10 served whether or not the assay was supplemented with 230 mM mM imidazole. The quenched incubation was applied to a 1-ml column monoubiquitin. of Ni-NTA agarose (Novagen) preequilibrated with buffer containing 50 Mass Spectrometry—Matrix-assisted laser desorption time-of-flight mM Tris-HCl (50% base), 0.15 M NaCl, 9 M urea, and 25 mM imidazole. mass spectrometry employed a PerSeptive Biosystems Voyager RP The loaded column was washed with equilibration buffer. H -ubiquitin spectrometer operated in linear mode at an accelerating voltage of 30 conjugates were eluted with 4 ml of equilibration buffer supplemented kV. Samples were prepared with either a-cyano-4-hydroxycinnamic with 1 M imidazole. The eluate was dialyzed overnight against 2 liters acid or 3,5-dimethoxy-4-hydroxycinnamic acid as the matrix compound, of buffer containing 20 mM Tris-HCl (pH 8.0), 1 mM EDTA, and 1 mM and the instrument was calibrated by use of the single-charged (m/z 5 DTT. The dialyzed sample was concentrated to ;0.5 ml by ultrafiltra- 14,314.2) and double-charged (m/z 5 7,157.6) cations from an internal tion (Centricon-10). Analysis by SDS-polyacrylamide gel electrophore- standard of chicken egg white lysozyme. sis and autoradiography showed that 15–20% of the I-lysozyme was conjugated to ubiquitin during the initial incubation. About 15% of the RESULTS AND DISCUSSION conjugates were recovered in the final eluate, and .90% of the I- Principle of the Method—Fig. 1 shows the method for syn- lysozyme in the final eluate was conjugated to ubiquitin. thesizing Lys -linked polyubiquitin chains of defined length. Assay of Ubiquitin Conjugate Degradation by the 26 S Proteasome— The approach utilizes ubiquitin-conjugating enzyme E2-25K to The 26 S proteasome was prepared as described (18). Standard incuba- 48 76 tions contained (25–50 ml; pH 7.3, 37 °C) 50 mM Tris-HCl (24% base), 5 generate a Lys -Gly isopeptide bond between two ubiquitin Inhibition of 26 S Proteasome by Polyubiquitin Chains 23715 derivatives. One of these derivatives is capped at its COOH terminus (future proximal chain terminus) by the presence of an extra residue, Asp . The other ubiquitin derivative is capped at the ubiquitin-accepting site (future distal chain ter- minus) by the presence of a Lys-to-Cys mutation at residue 48. In the first round of synthesis, these reactants are the two monomeric derivatives Asp -ubiquitin and K48C-ubiquitin. In principle, E2-25K should quantitatively convert these two re- 76 48 actants to Ub ; once Gly of K48C-ubiquitin is linked to Lys of Asp -ubiquitin, further polymerization is prevented by the 48 76 absence of additional Lys and Gly residues. Additional rounds of controlled conjugation are possible because the prox- imal or distal terminus of the double-capped dimer is easily deblocked; Asp can be removed catalytically by the ubiquitin- processing protease yeast ubiquitin hydrolase-1 (31), whereas chemical alkylation of Cys with ethyleneimine (below) cre- ates a lysine mimic (39). The two single-deblocked dimers should be quantitatively converted to Ub in the next round of synthesis. By using the appropriate combination of single- capped chain reactants, this method can in principle give rise FIG.2. Ethyleneimine alkylation kinetics. Incubations (pH 8.1, to a chain of any length. 37 °C) contained 64 mM ethyleneimine and 1 mM cysteine (●) or 0.5 m M K48C-ubiquitin (E). Remaining thiol was determined by Ellman’s test. Efficiency and Specificity of Deblocking—E2-25K catalyzes the synthesis of polyubiquitin chains harboring exclusively 48 76 Lys -Gly isopeptide bonds (28, 38) and is available in recom- these results, it appeared that ethyleneimine could best alkyl- binant form (40). E1 and E2 enzymes can transfer polyubiq- ate the distal Cys in a chain reactant at a rate competitive uitin chains (28, 41). Thus, there is no block to efficient and with undesired alternative reactions, such as Cys oxidation. specific conjugation provided the necessary chain reactants are As shown in Fig. 2 (open circles), ethyleneimine reacted with available. This is trivial in the first cycle of synthesis, which Cys in K48C-monoubiquitin about two times more slowly directly utilizes recombinant monomeric ubiquitins. In subse- than with free cysteine, presumably reflecting steric hindrance quent cycles, however, the efficiency of conjugation is highly from surrounding side chains. dependent on the efficiency with which the double-capped prod- The efficiency and specificity of alkylation were examined uct of the previous cycle is deblocked. In addition, the function- further by tryptic peptide mapping of K48C-ubiquitin after ality of long chains generated by this method is likely to depend reaction with each of the reagents discussed above (for condi- strongly on specificity in chemical alkylation; a low level of tions of the alkylation reactions, see “Experimental Proce- reaction with side chains other than Cys could be deleterious dures”). Trace a in Fig. 3 shows the map of wild type ubiquitin; after the multiple rounds of alkylation needed to produce long peptides 4 –11 are clearly resolved (27). The K48C mutation chains. abolishes a tryptic cleavage site (top, Fig. 3); this is reflected as We expected that yeast ubiquitin hydrolase-1-catalyzed loss of peptides 4 and 8, concomitant with the appearance of a proximal deblocking would proceed with high efficiency and new combined peptide, labeled 12 in Fig. 3, which contains specificity, and this proved to be the case. Removal of the 48 Cys (compare traces a and b). (The low yield of peptide 6 in proximal Asp residue was very rapid, proceeding to comple- trace b is not significant; peptides 5 and 6 are overlapping, and tion within 60 min in reactions involving mono-, di-, and tet- their relative yields were variable.) Because alkylation restores rameric ubiquitin derivatives, at concentrations as high as 50 a cleavage site at residue 48, the extent of alkylation could be 48 76 mg/ml. There was no evidence for cleavage of Lys -Gly quantitated by the reappearance of peptides 4 and 8 and the isopeptide bonds. The single-deblocked Ub derivative result- disappearance of peptide 12. The peptide map of ethylenei- ing from yeast ubiquitin hydrolase-1 treatment (see below) was mine-treated K48C-ubiquitin (trace c) shows essentially full found to have a molecular mass of 17,090.5 6 8.2 Da, which recovery of peptides 4 and 8 and a corresponding loss of peptide compared favorably with the calculated mass of 17,087.7 Da. 12, indicating .90% modification of Cys . As expected from It seemed likely that efficiency and specificity would be more their slower alkylation rates (above), bromoethylamine and problematic in the alkylation of Cys than in the removal of N-(iodoethyl)trifluoroacetamide gave only partial alkylation, Asp . Thus we began with a systematic comparison of three reflected in low yields of peptides 4 and 8 and a corresponding reagents that could add an aminoethyl moiety to Cys : ethyl- persistence of peptide 12 (Fig. 3, traces d and e). eneimine, bromoethylamine, and N-(iodoethyl)trifluoroacet- Inspection of trace c further suggests that ethyleneimine did TM amide (Aminoethyl-8 ). N-(Iodoethyl)trifluoroacetamide was not react significantly with other potential nucleophilic side 1 68 used to alkylate K48C-ubiquitin in an earlier study (16), but chains, such as Met and His ; modification of these side the efficiency and specificity of alkylation were not determined. chains would probably be manifested as shifts in peaks 9 and In initial studies we determined the rates at which moderate 11, respectively (27). Mass spectrometric analysis of alkylated concentrations of the three reagents alkylated free cysteine. diubiquitin (bearing a proximal Asp residue, see below) con- The pseudo-first-order reaction observed with 64 mM ethylenei- firmed this inference; the observed mass of 17,232.7 6 12.4 Da mine at pH 8.1 and 37 °C is shown in Fig. 2 (k 5 0.12 min , compared favorably with the calculated mass of 17,245.8 Da. obs filled circles). The reaction was second-order in [ethylenei- Taken together, these results indicate that ethyleneimine acts 21 21 48 mine], corresponding to k 5 2.4 M min (average of dupli- with high efficiency and specificity at the side chain of Cys ,in cate determinations). Reaction with 50 mM bromoethylamine at both mono- and dimeric ubiquitin derivatives. pH 8.1 and 37 °C followed k /[bromoethylamine] 5 0.6 M Functionality of Alkylated Ubiquitin—To address the func- obs 21 21 21 min , whereas a value of k 5 0.3 M min was determined tionality of alkylated K48C-ubiquitin, we tested its ability to for N-(iodoethyl)trifluoroacetamide at pH 8.5 and 37 °C. From support degradation in a modified reticulocyte lysate, fraction 23716 Inhibition of 26 S Proteasome by Polyubiquitin Chains FIG.3. Tryptic peptide mapping of K48C-ubiquitin and alkylated deriva- tives. Top, sequence of ubiquitin with tryptic peptides indicated (27). Lines and boldface numbers below the sequence in- dicate peptides visible in the peptide maps; lines above sequence indicate unre- solved peptides and peptide 12 in K48C- ubiquitin (see “Results and Discussion”). Bottom, tryptic maps: a, wild type ubiq- uitin; b, K48C-ubiquitin; c– e, K48C-ubiq- uitin after alkylation with ethyleneimine (c), bromoethylamine (d), or N-(iodoethyl) trifluoroacetamide (e). The origin of the small peak (●) eluting around the posi- tion of peptide 8 in the K48C-ubiquitin map is not known. Each trace derived from digestion of 0.5 mg of ubiquitin. Ab- sorbance was monitored at 215 nm (0.2 unit full scale). II (4), in which degradation is dependent upon the addition of TABLE I functional ubiquitin. Ubiquitin derivatives that cannot form Support of degradation by mutant ubiquitins and alkylated chains are expected to show impaired activity in this assay (10, derivatives 16). For the two substrates assayed, there was indeed a de- Assays of the production of acid-soluble radioactivity were carried out in fraction II; ubiquitin and derivatives were added at 12 mM (20 3 K creased rate of degradation when K48C-ubiquitin (or K48R- 0.5 for wild type). For methods used in generating alkylated derivatives, ubiquitin) was substituted for wild type ubiquitin (Table I), and see “Experimental Procedures.” Typical incubation times were 20 min this decreased rate correlated with a strongly reduced level of 125 125 for I-lactalbumin and2hfor I-lysozyme (pH 7.3, 37 °C). Data were very high molecular weight conjugates (Fig. 4, lanes 3 and 4 corrected by subtraction of the acid-soluble counts from an incubation versus 2). These conjugates presumably bear long polyubiquitin lacking ubiquitin. The rate thus obtained for a given derivative is expressed relative to the corrected rate for wild type ubiquitin meas- chains. ured in the same experiment. Relative rates are means 6 S.D. of Alkylation of K48C-ubiquitin with ethyleneimine partially triplicate determinations, except for the final entry, which is the aver- 125 125 ( I-lactalbumin) or fully ( I-lysozyme) restored activity in age of duplicate determinations. For details of the last experiment, degradation (Table I). As expected based on this result, alky- which involved engineered Ub deblocked at its distal terminus by alkylation with ethyleneimine, see “Experimental Procedures.” In some lation with ethyleneimine also restored the formation of very experiments aliquots were withdrawn to monitor the level of ubiquitin- high molecular weight conjugates of I-lactalbumin (as mon- substrate conjugates by SDS-polyacrylamide gel electrophoresis and itored by electrophoresis and autoradiography; Fig. 4, lane 5 autoradiography (see “Experimental Procedures” and Fig. 4). versus 2). We do not know why full activity in I-lactalbumin Steady-state degradation rate degradation was not restored after alkylation. However, results Ubiquitin 125 125 I-Lactalbumin I-Lysozyme with the other alkylating reagents (below) indicate that this is % wild type ubiquitin an intrinsic effect of the presence of the S-aminoethylcysteine moiety rather than a specific effect of ethyleneimine. Prelimi- K48C-ubiquitin 37.8 6 2.0 59.4 6 3.7 K48R-ubiquitin 9.3 6 1.7 35.7 6 8.2 nary concentration dependence studies, carried out with ethyl- K48C (ethyleneimine) 65.6 6 3.8 95.5 6 5.2 eneimine-treated K48C-ubiquitin, indicated that this was a K48C (bromoethylamine) 70.5 6 4.3 97.4 6 5.3 V effect (data not shown). These results show that S-amin- K48C (Aminoethyl-8) 62.9 6 2.7 91.6 6 5.1 max oethylcysteine at position 48 functions similarly to Lys in ubiq- Disassembled Ub 68.8 Not done uitination and conjugate degradation, confirming the conclu- Inhibition of 26 S Proteasome by Polyubiquitin Chains 23717 FIG.5. Synthesis of engineered polyubiquitin chains (Coo- massie-stained gel). Each lane contains 2– 6 mg of total ubiquitin; reactions were carried out at 37 °C. Lanes 1 and 2,Ub synthesis. The incubation contained 20 mg of each reactant in 2 ml (pH 7.3). Lane 1, FIG.4. Activities of ubiquitin derivatives in I-lactalbumin zero time; lane 2,4h. Lanes 3 and 4,Ub synthesis. The incubation conjugation (autoradiograph). Incubations (see Table I) were car- contained 7 mg of each reactant in 0.7 ml (pH 8.0). Lane 3, zero time; ried out in fraction II without ubiquitin (lane 1) or in the presence of lane 4,1h. Lanes 5 and 6,Ub synthesis controls. Double-capped dimer K48C-ubiquitin (12 mM) that had been alkylated with the indicated (10 mg/ml) was incubated for 1 h with 10 mg/ml dimer deblocked at the reagent (EI, ethyleneimine; BEA, bromoethylamine; AE8, N-(iodoeth- proximal terminus (lane 5) or at the distal terminus (lane 6) (pH 8.0). yl)trifluoroacetamide). Aliquots were withdrawn during the steady Lanes 7 and 8,Ub synthesis. The incubation contained 2 mg of each state of degradation and analyzed by electrophoresis and autoradiog- reactant in 0.6 ml (pH 8.0). Lane 7, zero time; lane 8, 1.5 h. raphy. LA, I-lactalbumin; cont., labeled contaminant. With K48C and K48R-ubiquitin, two ubiquitinated forms of lactalbumin are visible below the contaminant band. With K48C-ubiquitin, three ubiquitinated ger inhibition seen with I-lactalbumin probably reflects a forms are also seen above the contaminant band. difference in the rate-limiting step; conjugate formation is the slow step in I-lysozyme degradation (44), whereas conjugate degradation may be partly rate-limiting in I-lactalbumin sion of an earlier study by Gregori et al. (16). turnover. Lactalbumin degradation is thus expected to be Alkylation with N-(iodoethyl)trifluoroacetamide substan- more sensitive to features such as chain length, which influ- tially restored the function of K48C-ubiquitin (Table I), as seen ence conjugate recognition by the 26 S proteasome. In addition, previously (16). A similar result was obtained after alkylation we cannot exclude the possibility that non-polyubiquitinated with bromoethylamine (Table I). The functionality of the latter lysozyme conjugates are recognized better than comparable two derivatives is not inconsistent with their demonstrated forms of lactalbumin. substoichiometric alkylation (Fig. 3), in view of the high con- Second, K48R- and K48C-ubiquitin were not equivalent: the centration of ubiquitin in the assay (12 mM), the low K of 0.5 K48C mutation was significantly less inhibitory with both sub- ubiquitin in degradation (0.6 mM (42)), and the ease with which strates. In the case of I-lactalbumin, this difference was ubiquitin conjugates undergo disassembly and reassembly (12). probably caused by the greater ability of K48C-ubiquitin to As expected based on these considerations, these two alkylated support the formation of conjugates bearing multiple single derivatives largely restored the formation of high molecular ubiquitins (up to n ; 5; Fig. 4, lanes 3 versus 4). This effect may weight conjugates (Fig. 4, lanes 6 and 7 versus 2). be explained by decreased susceptibility to de-ubiquitination In a final experiment, we used alkylated Ub (resulting from because conjugates bearing K48C-ubiquitin, but not K48R- two cycles of synthesis, Fig. 1) as the source of ubiquitin to ubiquitin or alkylated K48C-ubiquitin, are resistant to disas- support degradation. Alkylated Ub was almost completely sembly by an isopeptidase associated with the 26 S proteasome disassembled by endogenous isopeptidases during a 30-min (45). The inhibitory effect of a given chain-terminating muta- preincubation in fraction II (without ATP), as indicated by the tion can also be influenced by substrate structure; with an results of Western blotting with anti-ubiquitin antibodies (not engineered b-galactosidase substrate, both K48R- and K48C- shown). After the addition of ATP, the monomers produced by ubiquitin inhibited degradation by more than 90% (10, 16). disassembly supported degradation at the same rate as alkyl- Here the similar behavior of the two mutant ubiquitins, and ated K48C-monoubiquitin (Table I). This result provides a the very dramatic inhibition, may be explained by the presence qualitative indication that isopeptidases recognize the modi- of only a single ubiquitination site in the target protein (10); fied isopeptide bond containing the S-aminoethylcysteine moi- regardless of the nature of the mutation at residue 48, only ety and shows that two rounds of alkylation did not result in monoubiquitinated b-galactosidase could be formed. gratuitous inhibitory modifications. Based on the functionality Large Scale Synthesis of Defined Length Polyubiquitin of products generated using ethyleneimine (Table I and Fig. 4) Chains—Synthesis of double-capped Ub from Asp -ubiquitin and on the high efficiency and specificity of this reagent (Figs. and K48C-ubiquitin is shown in lanes 1 and 2 of Fig. 5. This 2 and 3), we selected ethyleneimine as the alkylating agent in 2-ml incubation, carried out at pH 7.3, contained 40 mg of total large scale synthesis of polyubiquitin chains (Fig. 1 and below). ubiquitin; dimer synthesis was complete in 4 h. The reaction Two additional features of the functional data (Table I), was faster at pH 8, and this condition was used in subsequent although not directly relevant to our original objectives, de- synthetic reactions. Up to 60 mg of each monomer has been serve comment. First, the quantitative effect of blocking polyu- combined in such first-cycle reactions. biquitin chain formation was substrate-dependent; mutation of 48 125 Lys inhibited more strongly with I-lactalbumin than with I-lysozyme (Table I). Similarly, Hershko and co-workers (9, R. Beal, D. Toscano-Cantaffa, P. Young, and C. Pickart, manuscript 43) reported that reductive methylation of ubiquitin, which in preparation. blocks chain formation, suppresses I-lactalbumin degrada- Y. Lam, G. De Martino, C. Pickart, and R. Cohen, manuscript in tion more strongly than I-lysozyme degradation. The stron- preparation. 23718 Inhibition of 26 S Proteasome by Polyubiquitin Chains Synthesis of double-capped Ub is shown in lanes 3 and 4 of Fig. 5. This 0.7-ml incubation, carried out at pH 8, contained 7 mg of each single-deblocked Ub derivative; synthesis was es- sentially complete in 1 h. Up to 40 mg of each dimeric deriva- tive has been combined in second-cycle reactions. An additional round of deblocking and synthesis gave rise to double-blocked Ub (Fig. 5, lanes 7 and 8). In any given incubation, formation of the chain product was absolutely dependent upon the presence of both appropriately deblocked reactants, as shown for Ub synthesis in lanes 5 and 6 of Fig. 5. This last result confirms the very high linkage specificity of E2-25K (29, 38). A trace of apparent Ub in lane 4 (Fig. 5) may be because of the loss of Asp from the proximally capped dimer catalyzed by an E. coli carboxypeptidase that is present in trace amounts in some E2-25K preparations. Syn- thesis was very efficient; when the concentrations of the input reactants were equal, there was nearly complete conversion to the expected product (Fig. 5, lanes 3, 4, and 7, 8). Ub and Ub products were resolved from remaining reac- 2 4 tants by gradient cation exchange chromatography; Ub was purified by gel filtration (see “Experimental Procedures”). Typ- ically, the major chain product was obtained in about 60% yield FIG.6. Inhibition of the 26 S proteasome by unanchored Lys - from cation exchange columns, although yields were sometimes linked polyubiquitin chains. Degradation by purified 26 S protea- as high as 80%. Recoveries from the deblocking reactions were somes was assayed using purified H -ubiquitin- I-lysozyme conju- gates (see “Experimental Procedures”). The initial rate of conjugate higher: nearly 100% for the yeast ubiquitin hydrolase-1 reac- degradation in the presence of chains is expressed relative to the rate in tion and ;90% for the ethyleneimine reaction. a control reaction that lacked added polyubiquitin chains. Assays were Binding of Engineered Chains Inhibits Conjugate Degrada- supplemented with the indicated concentrations of Ub (L), Ub (M), 2 3 Ub (E), Ub (ƒ). Ub was double-blocked; in most cases Ub was also tion by the 26 S Proteasome—Because the rates of ubiquitina- 4 8 8 4 double-blocked (see “Results and Discussion”). The solid circle shows tion and de-ubiquitination, as well as the rate of conjugate data for wild type Ub (average of duplicate determinations). Lines for degradation, may contribute to the overall rate of degradation Ub (open circles) and Ub (inverted triangles) were obtained by fitting 4 8 in fraction II, it is difficult to draw quantitative conclusions the data using the program Enzfitter (Ub , 27.6 6 4.3 mM;Ub , 4.8 6 1.0 4 8 about conjugate recognition based solely on steady-state meas- mM). Points for Ub at 75 mM and Ub at 58 mM were omitted when fitting 4 8 the data. Points with error bars show the mean 6 S.D. of three or more urements in this system (above). We showed previously that a determinations. Points for Ub and Ub are averages of duplicate de- 48 2 3 mixture of unanchored Lys -linked chains inhibited overall terminations. Most other points are single determinations. For a sum- degradation in fraction II, indicating that chains compete with mary of K values, see Table II. 0.5 polyubiquitinated substrates for binding to chain recognition components of the 26 S proteasome (17). A higher resolution for binding. This conclusion is consistent with the the results version of this assay provided a convenient monitor of the shown in Table I. binding of engineered chains to the proteasome. These experi- Length Dependence of Inhibition—The rate of conjugate deg- ments utilized purified mammalian 26 S proteasomes (18) and radation is known to be facilitated by the presence of multiple purified I-lysozyme conjugates made with H -ubiquitin (see ubiquitins (6 – 8, 10, 16; Table I). In the case of conjugates “Experimental Procedures”). bearing chains, degradation is thought to increase with chain Ub inhibited the purified proteasome with a hyperbolic con- length. This may be caused both by enhanced binding of long centration dependence that corresponded to K 5 27.6 6 4.3 0.5 chains and their resistance to disassembly by a proteasome- mM (open circles, Fig. 6). Western analysis with anti-ubiquitin associated isopeptidase (45). With regard to binding, no quan- antibodies indicated that Ub was stable during the assay (data titative studies have addressed the form or basis of the pre- not shown). Inhibition must thus be caused by competition with sumptive length dependence. We show in Fig. 6 (inverted substrate conjugates. The presence of the proximal (Asp ) and triangles) that Ub inhibited the proteasome with a hyperbolic distal (Cys ) blocking residues did not influence binding be- concentration dependence that corresponded to K 5 4.8 6 1.0 0.5 cause similar inhibition was seen with double-blocked Ub 4 mM. This value is 5.8-fold smaller than the K for Ub . Given 0.5 4 versus each of the two single-blocked species (at 29 mM; these the errors in the respective measurements, the ratio could be as data are included in the open circles in Fig. 6). Failure of the high as 8 but no lower than 4. chain termini to influence binding is consistent with the expec- Inhibition by shorter chains was examined in a preliminary tation that substrate conjugates, which bear a macromolecule way. These chains were assembled from wild type ubiquitin; at the proximal chain terminus, will bind primarily through the absence of the S-aminoethylcysteine moiety should not interaction with the polyubiquitin chain. A series of hydropho- influence binding (above). As shown in Fig. 6, 78 mM Ub caused bic patches which has been implicated in the binding of chains ;50% inhibition (square), whereas 117 mM Ub caused only to the proteasome lies on the surface of the chain (17); this is ;21% inhibition (diamond). The observed inhibition is consist- also consistent with the observed absence of end effects. ent with K values of ;80 mM for Ub and ;430 mM for Ub , 0.5 3 2 Fig. 6 (filled circle) also shows the effect of wild type Ub at assuming a hyperbolic concentration dependence as is seen 29 mM. This chain had normal isopeptide linkages, with K48R- with longer chains. ubiquitin at the distal position (to facilitate synthesis; see The affinity of polyubiquitin chains for the proteasome thus “Experimental Procedures”). If anything, the wild type tet- increases ;90-fold as n increases from 2 to 8 (Table II). This is ramer inhibited more weakly than the engineered tetramers, substantially steeper than the chain length dependence ob- confirming that the presence of one or two S-aminoethylcys- served in the proteasome-catalyzed degradation of a series of teine moieties in the engineered tetramer was fully permissive a-globin conjugates (45). In the latter study, there was a rate Inhibition of 26 S Proteasome by Polyubiquitin Chains 23719 TABLE II Length dependence of inhibition of 26 S proteasome by unanchored polyubiquitin chains Data are taken from Fig. 6. For Ub and Ub , values of K were 2 3 0.5 calculated based on the inhibition observed at a single concentration of the respective chain. 0.5 Observed value Relative value mM 2 430 89.6 3 80 16.7 4 27.6 6 4.3 5.8 SCHEME I. Model of Ub which is based on the crystal structure 8 4.8 6 1.0 1 of Ub (48). The scheme depicts, as small circles, a series of surface 8 44 hydrophobic patches composed of the side chains of residues Leu , Ile , and Val . The solid circles are patches facing the viewer; the stippled increase of ;2-fold as n increased from 1 to 6 (degradation in circles are patches on the opposite face of the chain. the presence of Ubal (45)). This weaker dependence on n is probably caused by the structures of the a-globin conjugates, An alternative model postulates that two adjacent hydropho- which are linked to monoubiquitin(s) at low n values, and by a bic patches on the same face of the chain represent a minimal mixture of monoubiquitins and short chains at higher n values recognition element. The two ubiquitins bearing these patches, (46). It is likely that conjugates linked to monoubiquitin inter- e.g. ubiquitins 2 and 4 in Scheme I, are not directly covalently act with the proteasome differently than conjugates linked to linked. Therefore Ub lacks a recognition element and should polyubiquitin chains. cause negligible inhibition, as observed. Ub , if its conforma- Our results bear on the mechanisms by which polyubiquitin tion is Ub -like, contains one element, Ub contains two, and 4 4 chains may target substrates to the proteasome. The data Ub contains six. Ub inhibits about two times more weakly 8 3 unambiguously exclude a model in which the chain simply than predicted by this model; this could reflect the inability of amplifies the local concentration of monoubiquitin because this Ub to replicate faithfully the conformation seen in the Ub 3 4 model predicts that K will decrease in direct proportion to crystal (48). Like the previous model, this model predicts that 0.5 chain length; instead, K decreases by ;90-fold as n increases the affinity of Ub should be three times that of Ub , which is 0.5 8 4 by 4-fold (Table II). A model in which monoubiquitin is the close to the lower limit of 4 which can be accommodated by the recognition element is also inconsistent with the failure of current results (above). monoubiquitin to compete with polyubiquitin chains (see “Ex- These considerations serve to indicate that the enhanced perimental Procedures”). The results are generally consistent interaction of longer chains with the proteasome can be ex- with a model in which assembly of ubiquitin into a chain plained by a model in which the structure of the chain serves to creates and amplifies a specific structural element that is rec- create and amplify a minimum binding element that includes ognized by one or more binding components in the 19 S regu- two or more hydrophobic patches. Studies on the interaction of latory complex. A satisfying feature of this model is that it longer chains (n . 8) and mutant chains (below) with the could allow monoubiquitin to serve distinct functions from proteasome will be helpful in defining the precise nature of this polyubiquitin chains (e.g. 47). element or in developing alternative models. As noted above, When immobilized on a nitrocellulose membrane, subunit the observed ratio of K values for Ub and Ub is somewhat 0.5 4 8 S5a of the 19 S complex exhibits negligible binding of chains larger than the ratio predicted by two of the models discussed shorter than Ub , whereas for chains of n $ 4 binding increases above. However, “fraying” of the ends of the chain, which prob- steeply with chain length (23). This behavior led to the sugges- ably occurs in solution, would disproportionately weaken the tion that Ub harbors a minimum recognition element for S5a binding of short chains, leading to an increase in this ratio. (17, 23). We considered whether our data could be consistent In previous work we synthesized several Ub molecules bear- with a similar model for the proteasome. Scheme I shows a ing L8A,I44A-ubiquitin at one or more defined positions within model of Ub which is based on the crystal structure of Ub the chain. Because chains assembled solely from this mutant 8 4 (48). The scheme depicts a series of surface hydrophobic ubiquitin cannot target substrates for degradation (17), such 8 44 patches composed of the side chains of residues Lys , Ile , and defined oligomers may provide a way to test the models out- Val . Mutation of pairs of these residues to Ala abolishes lined above. We measured binding of these molecules to S5a binding of polyubiquitin chains to the proteasome and to iso- immobilized on a nitrocellulose membrane; the results (17) lated S5a, suggesting that these patches are part of a chain- agreed best with a four-ubiquitin recognition element. How- based recognition element (17). ever, the poor sensitivity of this binding assay, and the finding For the simple case in which Ub represents the actual that the yeast S5a homolog is nonessential (26), indicate that recognition element, and longer chains are assumed to have the these issues should be re-addressed with the 26 S proteasome. conformation seen in the Ub crystal, there are three elements Application of the method described here will facilitate the in Ub (ubiquitins 1– 4, 3– 6, and 5– 8 in Scheme I). This model assembly of mixed wild type/mutant polyubiquitin chains. predicts that Ub will bind three times more tightly than Ub ; Ubiquitin is present in cells at a concentration of ;20 mM,of 8 4 this factor is close to the lower limit of 4 which can be accom- which about half is in the form of conjugates (49). Although the modated by our binding data (above). The model is also con- length of the chain on a typical conjugate undergoing degrada- sistent with the observed negligible inhibition by Ub . Signifi- tion is not known, substrate-linked chains of n . 8 are difficult cant inhibition by Ub , although not predicted, might arise if to detect (e.g. 10, 16). Thus, it is reasonable to expect that Ub 3 8 the conformation of Ub is Ub -like, and a partial recognition should constitute an efficient targeting signal. The question 3 4 element is partially functional. A related model, in which ubiq- arises as to whether the binding reported here for Ub (K ; 8 0.5 uitins 2–5 and 4 –7 (Scheme I) are considered to be identical to 4 mM, Fig. 6) is sufficient to allow rapid degradation of poly- the elements specified above, predicts that Ub will bind five ubiquitinated proteins in vivo. This question cannot be an- times more tightly than Ub ; this prediction is in good agree- swered with certainty, but it is probable that affinities only ment with the data. modestly higher than those observed would be adequate be- 23720 Inhibition of 26 S Proteasome by Polyubiquitin Chains cause the high level of the proteasome in cells (;1% of cell E2-25K to conjugate engineered chains to the a-ubiquitin moi- protein (50)) means that recognition site(s) should be present at ety in a recombinant fusion protein such as ubiquitin-dihydro- a concentration of 1–2 mM. folate reductase or ubiquitin-b-galactosidase. Studies are in The K values measured in the current study may under- progress to explore the feasibility of this approach, which might 0.5 estimate the true affinities of polyubiquitinated conjugates for be facilitated by the use of tagged chains. H -K48C-ubiquitin two reasons. First, observed K values will exceed true K behaves identically to K48C-ubiquitin in synthesis of engi- 0.5 i values by a factor related to [S]/K . Although the concentration neered chains (data not shown). of ubiquitin- I-lysozyme conjugates in our assays was only Finally, the approach taken here should be applicable to ;30 nM, K is unknown. In addition, conjugated fraction II chains linked through Lys residues other than Lys .Ifan proteins were present at an undetermined concentration (“Ex- enzyme can be identified to catalyze formation of the desired perimental Procedures”). A finding that Ub inhibits the deg- isopeptide bond, application of the methodology described here radation of a Ub -RNase A conjugate with K ; 10 mM sug- could facilitate the structural and functional characterization 4 0.5 gests that the current K values may underestimate true of such novel chains. 0.5 affinity. However, such competition effects would not change Acknowledgments—We are grateful to Eileen Kasperek for generat- the relative affinities of different length polyubiquitin chains ing the K48C-ubiquitin plasmid, to Amy Lam for the purified yeast (Table II). ubiquitin hydrolase-1 enzyme, and to Christopher N. Larsen for pro- Second, the presence of a substrate moiety may lead to viding pRSUbD. We thank Leni Moldovan for assistance in preparing the H ,K48C-ubiquitin plasmid and for help in preliminary studies on tighter binding of conjugates than of unanchored polyubiquitin ethyleneimine reactivity. chains. This would be beneficial in minimizing inhibition by the end products of degradation. Depending upon the nature of the REFERENCES proteasomal site(s), such an effect might arise through elimi- 1. Ciechanover, A. (1994) Cell 79, 13–21 nation of the negative charge at the proximal terminus of the 2. Hochstrasser, M. (1996) Annu. Rev. Genet. 30, 405– 439 3. Rubin, D. M., and Finley, D. (1995) Curr. Biol. 5, 854 – 858 polyubiquitin chain or by hydrogen bonding to the isopeptide 4. Hershko, A., Heller, H., Elias, S., and Ciechanover, A. (1983) J. Biol. Chem. amide. These types of interactions, although dependent upon 258, 8206 – 8214 5. Scheffner, M., Nuber, U., and Huibregtse, J. M. (1995) Nature 373, 81– 83 the presence of a substrate, would be independent of its iden- 6. Hershko, A., Leshinsky, E., Ganoth, D., and Heller, H. (1984) Proc. Natl. Acad. tity and thus consistent with a universal targeting function for Sci. U. S. A. 81, 1619 –1623 7. Hough, R. Pratt, G., and Rechsteiner, M. (1986) J. Biol. Chem. 261, 2400 –2408 polyubiquitin chains. 8. Shaeffer, J. R., and Kania, M. A. (1995) Biochemistry 34, 4015– 4021 Our data provide little information regarding the nature of 9. Hershko, A., and Heller, H. (1985) Biochem. Biophys. Res. Commun. 128, proteasomal polyubiquitin binding sites. Inclusion of Ubal in 1079 –1086 10. Chau, V., Tobias, J. W., Bachmair, A., Marriott, D., Ecker, D. J., Gonda, D. K., the assays eliminated the possibility of a contribution from a and Varshavsky, A. (1989) Science 243, 1576 –1583 proteasome-associated isopeptidase (45). A value of K ; 4 mM 0.5 11. Arnason, T., and Ellison, M. J. (1994) Mol. Cell. Biol. 14, 7876 –7883 for Ub is substantially weaker than K ; 20 nM reported for 12. Finley, D., Sadis, S., Monia, B. P., Boucher, P., Ecker, D. J., Crooke, S. T., and 8 0.5 Chau, V. (1994) Mol. Cell. Biol. 14, 5501–5509 binding of an E2-Ub conjugate to immobilized S5a (15). Be- 13. Spence, J., Sadis, S., Haas, A. L., and Finley, D. (1995) Mol. Cell. Biol. 15, cause the present assay would not have detected binding that is 1265–1273 14. Johnson, E. S., Ma, P. C. M., Ota, I. M., and Varshavsky, A. (1995) J. Biol. not relevant to conjugate degradation, our results could be Chem. 270, 17442–17456 consistent with the suggestion that S5a does not play a major 15. Baboshina, O. V., and Haas, A. L. (1996) J. Biol. Chem. 271, 2823–2831 16. Gregori, L., Poosch, M. S., Cousins, G., and Chau, V. (1990) J. Biol. Chem. 265, role in proteolytic targeting (26). On the other hand, incorpo- 8354 – 8357 ration into the 19 S regulatory complex could alter the binding 17. Beal, R., Deveraux, Q., Xia, G., Rechsteiner, M., and Pickart, C. (1996) Proc. properties of S5a. Further work will be necessary to resolve Natl. Acad. Sci. U. S. A. 93, 861– 866 18. Hoffman, L., Pratt, G., and Rechsteiner, M. (1992) J. Biol. Chem. 267, this issue. 22362–22368 Concluding Remarks—The ability to synthesize polyubiq- 19. Chu-Ping, M., Slaughter, C. A., and DeMartino, G. N. (1992) Biochim. Biophys. uitin chains of defined length and structure solves one of two Acta 1119, 303–311 20. Dubiel, W., Ferrer, K., and Rechsteiner, M. (1995) Mol. Biol. Rep. 21, 27–34 significant problems associated with generating biochemically 21. Lowe, J., Stock, D., Jap, B., Zwickl, P., Baumeister, W., and Huber, R. (1995) useful quantities of substrates for the 26 S proteasome. It re- Science 268, 533–539 22. Groll, M., Ditzel, L., Lowe, J., Stock, D., Bochtler, M., Bartunik, H. D., and mains necessary to link these chains to a suitable target pro- Huber, R. (1997) Nature 386, 463– 471 tein. In principle, it should be possible to use a recombinant E2 23. Deveraux, Q., Ustrell, V., Pickart, C., and Rechsteiner, M. (1994) J. Biol. Chem. 269, 7059 –7061 protein to conjugate engineered chains to an in vitro substrate 24. Van Nocker, S., Deveraux, Q., Rechsteiner, M., and Vierstra, R. D. (1996) Proc. (41). However, because the target specificity of unassisted E2 Natl. Acad. Sci. U. S. A. 93, 856 – 860 conjugation is highly restricted (e.g. 51), it is unlikely that this 25. DeMartino, G. N., Moomaw, C. R., Zagnitko, O. P., Proske, R. J., Chu-Ping, M., Afendis, S. J., Swaffield, J. C., and Slaughter, C. A. (1994) J. Biol. Chem. approach will be broadly useful. For example, we used recom- 269, 20878 –20884 binant E2-20K (52) to conjugate engineered Ub to I-histone 4 26. Van Nocker, S., Sadis, S., Rubin, D. M., Glickman, M., Fu, H., Coux, O., Wefes, 2B, but the resulting conjugate was degraded at a negligible I., Finley, D., and Vierstra, R. D. (1996) Mol. Cell. Biol. 16, 6020 – 6028 27. Cox, M. J., Shapira, R., and Wilkinson, K. D. (1985) Anal. Biochem. 154, rate by the purified 26 S proteasome (data not shown). This is 345–352 not unexpected based on the low rate of histone degradation 28. Chen, Z., and Pickart, C. M. (1990) J. Biol. Chem. 265, 21835–21842 29. Pickart, C. M., and Vella, A. T. (1988) J. Biol. Chem. 263, 12028 –12035 observed in fraction II (53). Overall, it is likely that E3 en- 30. Haldeman, M. T., Xia, G., Kasperek, E. M., and Pickart, C. M. (1997) Biochem- zymes, together with their specific E2 partners, will represent istry, 36, in press 31. Miller, H. I., Henzel, W. J., Ridgway, J. B., Kuang, W.-J., Chisholm, V., and a more effective route to the ligation of engineered chains to Liu, C.-C. (1989) Bio/Technology 7, 698 –704 substrates. 32. Mayer, A. N., and Wilkinson, K. D. (1989) Biochemistry 28, 166 –172 One possible exception to an E3 requirement for generating 33. Dunten, R. L., and Cohen, R. E. (1989) J. Biol. Chem. 264, 16739 –16747 34. Ecker, D. J., Butt, T. R., Marsh, J., Sternberg, E. J., Margolis, N., Monia, B. P., proteasome substrates is provided by E2-25K itself. Several Jonnalagadda, S., Khan, M. I., Weber, P. L., Mueller, L., and Crooke, S. T. noncleavable ubiquitin fusion proteins are known to undergo (1987) J. Biol. Chem. 262, 14213–14221 polyubiquitination at Lys of the a-linked ubiquitin moiety, 35. Burch, T. J., and Haas, A. L. (1994) Biochemistry 33, 7300 –7308 36. Haas, A. L., and Wilkinson, K. D. (1985) Prep. Biochem. 15, 49–60 and conjugates of this structure are degraded by the 26 S 37. Ellman, G. L. (1959) Arch. Biochem. Biophys. 82, 70 –77 proteasome in yeast cells (54, 55). It should be possible to use 38. Pickart, C. M., Haldeman, M. T., Kasperek, E. M., and Chen, Z. (1992) J. Biol. Chem. 267, 14418 –14223 39. Cole, R. D. (1967) Methods Enzymol. 11, 315–317 40. Chen, Z., Niles, E. G., and Pickart, C. M. (1991) J. Biol. Chem. 266, W. Xu and R. Cohen, unpublished experiments. 15698 –15704 Inhibition of 26 S Proteasome by Polyubiquitin Chains 23721 41. van Nocker, S., and Vierstra, R. D. (1993) J. Biol. Chem. 268, 24766 –24773 49. Haas, A. L., and Bright, P. M. (1985) J. Biol. Chem. 260, 12464 –12473 42. Cox, M. J., Haas, A. L., and Wilkinson, K. D. (1986) Arch. Biochem. Biophys. 50. Tanaka, K., Ii, K., Ichihara, A., Waxman, L., and Goldberg, A. L. (1986) J. Biol. 250, 400 – 409 Chem. 261, 15197–15203 43. Hershko, A., Ganoth, D., Pehrson, J., Palazzo, R. E., and Cohen, L. H. (1991) 51. Pickart, C. M., and Rose, I. A. (1985) J. Biol. Chem. 260, 1573–1581 J. Biol. Chem. 266, 16376 –16379 52. Wefes, I., Kaiser, P., Schneider, R., Pickart, C. M., and Finley, D. (1995) Gene 44. Dunten, R. L., Cohen, R. E., Gregori, L., and Chau, V. (1991) J. Biol. Chem. (Amst.) 163, 321–322 266, 3260 –3267 53. Haas, A., Reback, P. M., Pratt, G., and Rechsteiner, M. (1990) J. Biol. Chem. 45. Lam, Y. A., Xu, W., DeMartino, G. N., and Cohen, R. E. (1997) Nature 385, 265, 21664 –21669 737–740 54. Johnson, E. S., Bartel, B., Seufert, W., and Varshavsky, A. (1992) EMBO J. 11, 46. Shaeffer, J. R. (1994) J. Biol. Chem. 269, 29530 –29536 487–505 47. Finley, D., Bartel, B., and Varshavsky, A. (1989) Nature 338, 394 – 401 55. Johnson, E. S., Ma, P. C. M., Ota, I. M., and Varshavsky, A. (1995) J. Biol. 48. Cook, W. J., Jeffrey, L. C., Kasperek, E. M., and Pickart, C. M. (1994) J. Mol. Chem. 270, 17442–17456 Biol. 236, 601– 609

Journal

Journal of Biological Chemistry – Unpaywall

Published: Sep 1, 1997

There are no references for this article.