WO2005021782A1 - Methods of identifying modulators of proteasome activity - Google Patents

Abstract

The methods, reagents, and systems of the invention are useful for assaying the activity of ubiquitin-mediated digestion by a proteasome, especially the 26S proteasome, in the presence or absence of certain proteins (e.g., Rad23, Dsk2, Rpn10, and homologues thereof) and their activity modulators (e.g., inhibitor or enhancer). The invention also provides methods to screen for compounds (small molecule, polypeptide, antibody, polynucleotide, organic or inorganic molecules, etc.) that inhibit or enhance the function of these certain proteins in the ubiquitin proteasome system.

Description

METHODS OF IDENTIFYING MODULATORS OF PROTEASOME ACTIVITY

Reference to Related Applications This application claims the benefit of U.S Provisional Application 60/498973, filed on August 29, 2003; and U.S Provisional Application 60/502541, filed on September 12, 2003. The entire contents of both U.S. provisional applications are incorporated herein by reference.

Background of the Invention Proteolysis by the ubiquitin-proteasome system (UPS) is required for the maintenance of cellular homeostasis (Hersh o and Ciechanover, Annu. Rev.

Biochem. 67: 425-479, 1998; and Pickart and Cohen, Nat. Rev. Mol. Cell Biol. 5: 177-187, 2004). Proteins destined to be degraded by the proteasome are marked for elimination by the covalent attachment of ubiquitin (Ub). The C terminus of Ub is linked by an isopeptide bond to the amino group of a lysine residue in the substrate. A multiubiquitin (multiUb) chain is formed by attachment of successive Ubs, primarily to the Lys48 residue of the distal-most Ub tethered to the substrate. Once the multiUb chain contains at least four Ubs (i.e., a tetraubiquitin chain), it can bind the proteasome and serve as a signal for degradation (Chau et al., Science 243: 1576-1583, 1989; and Thrower et al., EMBO J. 19: 94-102, 2000). Following specific binding, the ubiquitinated substrate is unfolded, deubiquitinated, and translocated by the 19S regulatory "cap" of the 26S proteasome into the 20S protease core, where it is proteolyzed to peptide remnants (Hershko and Ciechanover, Annu. Rev. Biochem. 67: 425-479, 1998; Verma et al, Science 298: 611-615, 2002; and Yao and Cohen, Nature 419: 403-407, 2002). Recognition of multiUb chains by the proteasome is central to Ub-selective degradation. The receptor(s) that mediates this process has thus been sought intensively. Over the past decade, three different classes of proteins have been advanced as candidate receptors that link Ub conjugates to the proteasome for degradation. RpnlO was the first protein that was shown to bind selectively to polyubiquitin (polyUb) chains. Because RpnlO is a bona fide stoichiometric subunit of the 26S proteasome, it was proposed that RpnlO is the multiUb chain receptor (Deveraux et al, J. Biol. Chem. 269: 7059-7061, 1994). However, even though proteasomal proteolysis is essential, RpnlO is dispensable for life in budding yeast (Fu et /., J. Biol. Chem. 273: 1970-1981, 1998; and van Nocker et al, Mol. Cell. Biol. 16: 6020-6028, 1996). Indeed, only one UPS substrate, Ub-proline-β- galactosidase (Ub-Pro-β-gal, or the related

has been shown to be stabilized in rpnlOΔ cells, and, paradoxically, Ub-Pro-β-gal turnover does not require the Ub binding domain of RpnlO (Fu et al, J. Biol. Chem. 273: 1970-1981, 1998). Additionally, RpnlO assembled into 26S proteasomes does not crosslink to a chemically reactive tetraubiquitin chain (Lam et al, Nature 416: 763- 767, 2002), and recombinant RpnlO inhibits proteolysis in frog extracts (Deveraux et al, J. Biol. Chem. 270: 29660-29663, 1995). Taken together, these observations raised doubts as to whether RpnlO functioned in the context of the 26S proteasome to recruit ubiquitinated substrates for degradation (Pickart and Cohen, Nat. Rev. Mol. Cell Biol. 5: 177-187, 2004). Attention was thus diverted to a second group of proteins exemplified by Rad23 and Dsk2. These proteins each contain a Ub-like domain (UbL) that binds the proteasome (Elsasser et al, Nat. Cell Biol. 4: 725-730, 2002; Saeki et al, Biochem. Biophys. Res. Commun. 296: 813-819 b, 2002; and Schauber et al, Nature 391: 715- 718, 1998) and UBA domains that bind multiUb chains (Rao and Sastry, J. Biol. Chem. 211: 11691-11695, 2002; and Wilkinson et al, Nat. Cell Biol. 3: 939-943, 2001). However, the role of Rad23 and Dsk2 in guiding multiUb chain-bearing substrates to the proteasome is equally controversial. Budding and fission yeast rad23Δ and dsk2Δ mutants accumulate reporter substrates and high molecular weight Ub conjugates, supporting a positive role for these proteins in the UPS (Chen and Madura, Mol. Cell. Biol. 22: 4902-4913, 2002; Funakoshi et al, Proc. Natl. Acad. Sci. USA 99: 745-750, 2002; Rao and Sastry, J. Biol. Chem. 277: 11691- 11695, 2002; Saeki et al, Biochem. Biophys. Res. Commun. 293: 986-992a, 2002; and Wilkinson et al, Nat. Cell Biol. 3: 939-943, 2001). However, rad23ΔrpnlOΔ double mutants are proficient in bulk turnover of short-lived proteins (Lambertson et al, Genetics 153: 69-79, 1999). Additionally, overexpression ofDsk2 or Rad23 in mammalian and yeast cells typically inhibits substrate turnover by the 26S proteasome (Kleijnen et al, Mol. Cell 6: 409-419, 2000; and Ortolan et al, Nat. Cell Biol. 2: 601-608, 2000) but can apparently stimulate turnover in some contexts (Funakoshi et al, Proc. Natl. Acad. Sci. USA 99: 745-750, 2002). Indeed, a key limitation to the argument that Rad23 and Dsk2 serve as substrate receptors is that such a role has never been directly demonstrated. In the only direct test so far of the hypothesis that Rad23 acts as a receptor that links substrates to the proteasome, it was shown that recombinant Rad23 actually inhibits substrate turnover by purified 26S proteasome in vitro (Raasi and Pickart, J. Biol. Chem. 278: 8951-8959, 2003). Similar results have been reported for RpnlO (Deveraux et al, J. Biol. Chem. 270: 29660-29663, 1995). In light of the lack of conclusive, direct evidence that Rad23 serves as a receptor to guide ubiquitinated substrates to the proteasome, other functions have been sought for this protein. Bioinformatics has revealed that the UBA domain is conserved in a number of enzymes of the UPS, including E2s, E3s, and Ub proteases (Ubps) (Hofmann and Bucher, Trends Biochem. Sci. 21: 172-173, 1996). Some members of the latter class, such as Ubpl4, bind polyUb chains and cleave them (Amerik et al, EMBOJ. 16: 4826-4838, 1997). Although binding of Rad23 to Ub conjugates did not cause cleavage of the Ub chain, it did inhibit Ub chain assembly (Ortolan et al, Nat. Cell Biol. 2: 601-608, 2000) as well as disassembly (Hartmann-Petersen et al. FEBSLett. 535: 77-81, 2003; and Raasi and Pickart, J. Biol. Chem. 278: 8951-8959, 2003), suggesting that Rad23 may promote degradation by serving as a shield that retards deubiquitination of substrates that are en route to the proteasome (Pickart and Cohen, Nat. Rev. Mol Cell Biol. 5: 177-187, 2004). To complicate matters further, a third candidate receptor (S6'/Rpt5) has recently been identified based on UV crosslinking of a tetra-Ub chain to purified 26S proteasomes (Lam et al, Nature 416: 763-767, 2002). Rpt5 is a member of the AAA ATPase family of enzymes, with an as yet undefined multiUb chain binding domain. A putative receptor function for Rpt5 is appealing based on precedent from other systems. The related AAA ATPases of bacterial compartmentalized proteases contribute to enzyme specificity by directly binding to short peptide degrons within substrates (Flynn et al, Mol. Cell 11: 671-683, 2003), and the mammalian AAA ATPase p97/Cdc48 promotes turnover of IkB by binding directly to multiubiquitin chains (Dai and Li, Nat. Cell Biol. 3: 740-744, 2001). However, a functional role for S6'/Rpt5 in recruiting ubiquitinated substrates to the proteasome has not been validated yet by either biochemical or genetic studies. Over the past few years it has become evident that the 26S proteasome may be an appealing target for therapeutic intervention in cancer, inflammation, ischemia, and other diseases. For example, Millennium Pharmaceuticals has shown in human phase II clinical trials that an inhibitor of the 20S core particle peptidases, known as VELCADE, is effective in treating patients suffering from relapsed multiple myeloma. VELCADE was approved by the U.S. Food and Drug Administration for treatment of this disease in May 2003. Since drugs that inhibit the proteasome peptidases have anti-cancer activity, drugs that inhibit other essential aspects of 26S proteasome function might likewise be candidate therapeutics. For example, molecules that block the targeting of ubiquitinated substrates to the 26S proteasome might be effective means to inhibit the ability of the 26S proteasome to degrade substrate proteins. Such molecules might be effective therapeutics to treat diseases such as cancer, inflammation, and ischemia. However, effective and focused drug development efforts in this area are severely hampered because too little is known about the mechanism of how a substrate with an attached tetra-ubiquitin chain is targeted to the proteasome for destruction. The studies summarized above highlight several key issues that need to be addressed before such drug development effort can be significantly advanced. For example, the nature of the primary gateway through which proteins targeted by the numerous cellular ubiquitin ligases are recognized by the proteasome and sent to meet their final fate is unclear. It is also unclear whether there is a single gateway (e.g., Rpt5) or multiple gateways (e.g., Rad23, RpnlO, and other Ub binding proteins), and, if the latter, whether the gateways function in parallel or in series. It is further unclear whether all ubiquitinated substrates are processed in the same manner, or whether there is an additional layer of substrate specificity downstream of the ubiquitin ligases. Finally, there is a lack of modulators (inhibitors or enhancers) capable of regulating this process. Summary of the Invention The invention provides a method for monitoring the promotion of degradation of ubiquitin-conjugated substrates by the 26S proteasome by proteins involved in ubiquitin chain recognition. Such proteins include Rad23, Dsk2, and RpnlO, as well as homologues thereof and other proteins determined to be involved in the recognition of ubiquitinated proteins. According to the methods of the invention, a proteasome preparation is provided that lacks one or more components involved in ubiquitin chain recognition, including, for example, Rad23, Dsk2, or RpnlO. According to the methods of the invention, a recombinant protein corresponding to the absent component can be added to the proteasome preparation to restore the function of protein degradation. The methods described herein can be employed to identify molecules (e.g., peptides, small molecules, or antibodies) that modulate the degradation-promoting activities of the Rad23, Dsk2, and RpnlO proteins and their homologues. Thus one aspect of the invention provides a system for assaying the activity of a ubiquitin-proteasome pathway, said system comprising: (1) a defective 26S proteasome preparation substantially unable to degrade a ubiquitin-conjugated target protein, (2) one or more functional multiubiquitin chain binding protein(s) (MCBP(s)) in an amount that substantially restores the ubiquitin-mediated proteasome degradation of said target protein to a wild-type level, with the proviso that said system is not a wild-type 26S proteasome preparation. In certain embodiments, the system is an in vitro system. In other embodiments, the system is an in vivo system. In certain embodiments, the MCBP(s) comprise one or more of: an RpnlO polypeptide, an Rad23 polypeptide, a Dsk2 polypeptide, a Cdc48/Ufdl/Npl4 complex, or an Rpt5 polypeptide. In certain embodiments, the MCBP(s) comprise one or more of: a Parkin polypeptide, a Ufd4 polypeptide, or an Hul5 polypeptide. In certain embodiments, the presence of said amount of one or more functional MCBP(s) in said defective 26S proteasome preparation restores at least about 30%, 50%, 60%, 70%, 80%, 90%, or nearly 100% of the wild-type level of ubiquitin-mediated proteasome degradation of said target protein. In certain embodiments, at least one of the MCBP(s) is from a species different from the species of the defective 26S proteasome preparation. In certain embodiments, the defective 26S proteasome preparation is from a non-human eukaryote, such as a yeast (S. cerevisiae or S. pombe, etc.). hi certain embodiments, the MCBP(s) comprise one or more human Rad23 homologs selected from hHR23 A and hHR23B, one or more human Dsk2 homologs selected from hPLIC-1 and hPLIC-2, or human RpnlO homolog S5A. In certain embodiments, the defective 26S proteasome preparation is from human. In certain embodiments, the defective 26S proteasome preparation has a diminished level of one or more MCBP(s). By "diminished level", it is meant to include diminished level of accumulation, diminished level of activity or a combination of both. In certain further embodiments, the defective 26S proteasome preparation contains one or more genetically-modified components, such as, for example, an epitope-tagged component. In certain embodiments, the diminished level of MCBP(s) is effectuated by selective chemical extraction of said MCBP(s), gene knock-out of said MCBP(s), immunodepletion or affinity depletion of said MCBP(s), orby R Ai of said MCBP(s). In certain such embodiments, the RNAi utilizes small inhibitor RNA (siRNA) or short hairpin RNA (shRNA). In certain embodiments, the diminished level is no more than about 30%), 20%, 10%>, 5%, or 1% of wild-type level. In certain embodiments, the amount of one or more functional MCBP(s) in said system is an optimum amount determined by titration. In certain embodiments, the system is an in vitro system, and said titration is performed by providing a range of different concentrations of MCBP(s). In certain embodiments, the system is an in vivo system, and said titration is performed by operatively linking the coding sequence of said MCBP(s) with a range of promoters with different transcription / translation strength. In certain embodiments, the amount of one or more functional MCBP(s) in said system is no more than 2-fold, preferably no more than 1.5-fold molar excess over the 26S proteasomes. In certain embodiments, the target protein is one or more of: Sicl, Farl,

Clb2, Gic2, CPY*, Cln2, or a homolog thereof. hi certain embodiments, the system lacks one or more MCBP(s). In certain such embodiments, the lacking MCBP(s) are different from said MCBP(s) in the system. In certain embodiments, the defective 26S proteasome preparation lacks

RpnlO, and said MCBP(s) comprise: (a) a Rad23 polypeptide or a Dsk2 polypeptide; and (b) a mutant RpnlO lacking the VWA domain, or a mutant RpnlO lacking the UIM domain. In certain embodiments, the system is adapted for assaying a specific subset of target proteins. certain embodiments, the MCBP(s) are recombinantly produced. In certain such embodiments, the recombinant MCBP is a fusion protein, including, for example, an epitope tagged protein, such as GST-, MBP-, 6-His, HA-, Ig-, or

FLAG-tagged. In certain embodiments, the system further comprises a facilitator, such as the VWA domain of an RpnlO protein. Another aspect of the invention provides a method for monitoring the ability of at least one multiubiquitin chain binding protein (MCBPs) to promote degradation of at least one ubiquitin-conjugated target protein by a 26S proteasome comprising: using any of the subject assay systems described above, and determining the degree of degradation of said target proteins. Another aspect of the invention provides a method for screening for an agent that inhibits the degradation of a ubiquitinated target protein by a 26S proteasome, the method comprising: (a) incubating the ubiquitinated target protein in any one of the subject systems described above; (b) determining and comparing the degree of degradation of said target protein in the presence or absence of a test agent, wherein more complete degradation of said target protein in the absence of the test agent than the presence of the test agent is indicative that the test agent inhibits the degradation of said target protein. In certain embodiments, the method further comprises determining the binding, if any, of the test agent to the 26S proteasome and/or the MCBP(s). Another aspect of the invention provides a method to screen for agents that can modulate (e.g. enhance or inhibit) the function of a MCBP-mediated 26S proteasome activity, the method comprising: (1) isolating, from among a plurality of candidate agents, one or more agent(s), if any, that can interact with the MCBP protein; (2) determining the MCBP-mediated 26S proteasome activity in the presence or absence of said candidate agents identified in (1), wherein an enhanced or inhibited proteasome activity, respectively, in the presence of the candidate agent(s) is indicative that the candidate agent(s) is/are enhancer or inhibitor, respectively, of the function of the MCBP-mediated 26S proteasome activity. In some embodiments, the MCBP is Rad23, RpnlO, Dsk2, Rpt2, or Cdc48/Ufdl complex. In some embodiment, the candidate agents are from a library, such as a polynucleotide library (which may be able to express proteins), a polypeptide library, a small chemical compound library (preferably with max. molecular weight of about 5000 Da.), an organic compound library, an inorganic compound library, a library of chemicals synthesized by split-pool methods, or a library of compounds with unknown identity. In some embodiments, step (1) is carried out by high throughput screening.

For example, more than about 10, 50, 100, 400, 1000, 1500, 5000 or more compounds maybe screening in parellel (such as in 96-well, 384-well, 1536-well plates, etc.) or nearly parallel. Agents to be screened may be added to individual compartments via, for example, pin-transfer robot or other device that transfers a pre-determined amount of test agents. Results may be recorded, stored, and analyzed by computer devices and software. Initial / primary screens may be backed up by further re-screening to verify the results. Pools of compounds may be screened together initially, and positive pools may be split and tested further to increase overall throughput. hi some embodiments, step (1) is effectuated by direct in vitro binding assay between the MCBP and the candidate agents. For example, BIACORE can be used to directly measure the binding affinity between the MCBP and the candidate agents. In a preferred embodiment, step (1) may be effectuated by a gel-shift binding assay. In some embodiments, step (1) is effectuated by two-hybrid binding assay using the MCBP as a bait protein. The two-hybrid assay may be performed in a bacterial, a yeast, or mammalian system. Reverse two-hybrid assay may also be used to isolate agents that bind the MCBP protein. Another aspect of the invention provides method of screening for 26S proteasome modulators, comprising: (a) incubating a plurality of candidate agents, each individually or in combination, with a recombinant MCBP protein and a ubiquitinated target protein; (b) measuring binding of the MCBP protein to the ubiquitinated target protein in the presence or absence of the candidate agent(s); wherein significant change in binding in the presence of the candidate agent(s) is indicative that the candidate agent(s) is a proteasome modulator. In some embodiments, the modulator is an inhibitor. In other embodiments, the modulator is an enhancer / stimulator of 26S proteasome function. In some embodiments, the method further comprises determining the activity of a 26S proteasome preparation comprising the MCBP protein, in the presence of the agent(s) identified in (b) as modulator(s), wherein a modulated activity of the 26S proteasome verifies that the candidate agent(s) is/are 26S proteasome modulator(s). In some embodiments, the activity of the 26S proteasome is determined by using the ubiquitinated target protein as a substrate. In some embodiments, the 26S proteasome is isolated from wild-type cells. In some embodiments, the 26S proteasome is reconstituted from a defective 26S proteasome substantially incapable of degrading the ubiquitinated target protein, and a recombinantly produced MCBP protein. Another aspect of the invention provides a method for screening for an agent that inhibits a ubiquitinated target protein's entry into the proteasome, the method comprising: (a) incubating, in the presence of a test agent, the ubiquitinated target protein and a sufficient amount of a recombinant MCBP protein that would restore degradation activity of a proteasome preparation lacking a corresponding functional MCBP; (b) adding a 26S proteasome preparation, wherein the MCBP corresponding to the recombinant MCBP was absent or defective in said 26S proteasome preparation; (c) determining the effect of the test agent, wherein substantial reduction of target protein degradation is indicative that the test agent inhibits the ubiquitinated target protein's entry into the proteasome. In some embodiments, the test agent inhibits at least about 50% of target protein degradation. In some embodiments, the test agent inhibits nearly 100%) of target protein degradation. It is contemplated that all embodiments described above are applicable to all different aspects of the invention. It is also contemplated that any of the above embodiments can be freely combined with one or more other such embodiments whenever appropriate. Specific embodiments of the invention are described in more detail below. However, these are illustrative embodiments, and should not be construed as limiting in any respect.

Brief Description of the Figures

Figure 1. Structural and functional characterization of 26S proteasomes isolated from rpnlOΔ and rad23Δ mutants by affinity chromatography. Extracts from wild-type and mutant yeast strains expressing PRE1^FH were incubated with anti-FLAG M2 resin. Bound proteins were eluted with FLAG peptide and analyzed by (A) SDS- PAGE and Coomassie blue staining; (B) native gel (nondenaturing) electrophoresis and Coomassie blue staining; or (C) nondenaturing electrophoresis and incubation with a fluorogenic peptide substrate (Verma et al, Mol. Biol. Cell 11: 3425-3439, 2000). (D) rpnlOΔ 26S are completely defective in the degradation and deubiquitination of UbMbpSicl. UbMbpSicl was incubated at 30°C with 26S proteasomes isolated from either wild-type or rpnlOΔ cells. Degradation reactions (lanes 2 and 5) were set up and analyzed by SDS-PAGE followed by immunoblotting with anti-Sicl polyclonal antibody as described in Experimental Procedures. For assessing deubiquitination (lanes 3 and 4), the 26S proteasome preparations were pre-incubated with 100 μM epoxomicin for 45 min at 30°C before incubation with UbMbpSicl . 26S proteasomes isolated from rad23Δ mutants were partially defective in (E) degradation and (F) deubiquitination of UbMbpSicl. Analysis was performed as described for rpnlOΔ proteasomes in (D). Figure 2. The Degradation and deubiquitination defects of rpnlOΔ and rad23Δ 26S proteasomes can be rescued by recombinant proteins. (A-D) GST-fusion proteins were isolated from E. coli by glutathione sepharose cliromatography, and various amounts of purified protein (indicated on top of each figure) were pre-incubated with wild-type and mutant 26S proteasomes on ice for 15 min. Degradation was initiated by the addition of UbMbpSic 1 , and reactions were incubated at 30°C for 5 min. DUB assays included a 45 min pre-incubation of 26S proteasomes with epoxomicin subsequent to addition of recombinant protein. Reactions were analyzed by SDS-PAGE and immuno-b lotting for Sicl as in Figure ID. Figure 3. Complementation of rad23Δ proteasomes requires both the Ub binding UBA domains and the proteasome binding UbL domain of Rad23. (A) The UBA domains bind UbMbpSicl. Purified GST and GST fusion proteins (1 μg each) bound to glutathione beads were incubated with UbMbpSicl, after which the input (20%) of total) and bound material (33%> of total) were fractionated by SDS-PAGE and visualized by immunoblotting with anti-Sicl serum. Note that GST- UBA lacks the UbL domain but contains both UBA domains found in Rad23, whereas GST-UbL is the reciprocal molecule lacking both UBA domains (Rao and Sastry, J. Biol. Chem. 277: 11691-11695, 2002). (B) Rescue of rad23Δ 26S proteasomes by Rad23. Deubiquitination reactions were set up using rad23Δ 26S proteasomes and UbMbpSicl in the presence or absence of GST- Rad23 (80 nM), GST-UBA (80 and 40 nM respectively), or GST-Ubl (80 and 40 nM), respectively, as described in the legend to Figure ID. (C) Rescue of rpnlOΔ 26S DUB defect by full-length Rad23 and GST- VWA. Deubiquitination reactions were assayed by incubation of UbMbpSicl with rpnlOΔ 26S proteasomes in the presence or absence of various GST-fusion proteins as described above. Figure 4. 26S proteasomes from rpnlOΔ are defective in binding UbMbpSicl . (A and B) The binding defect of rpnlOΔ 26S proteasomes can be rescued by either recombinant RpnlO or Rad23. Extracts from wild- type (WT), rpnlOΔ, and rpnlOΔrad23Δ cells expressing PRE1^FH or untagged PRE1 (UT) were bound to anti-Flag M2 resin in the presence of ATP and washed with buffer containing ATP as described for 26S purification. Resin-immobilized 26S proteasomes were then incubated with 1 mM phenanthroline, 2.5 μM Ub aldehyde, 100 μM MG132, 1 mM ATP, and 5 mM MgCl₂ in the absence or presence of the various GST-fusion proteins on ice for 60 min. UbMbpSicl was then added, and, after 90 min incubation at 4°C, the bound fraction was washed and analyzed by SDS-PAGE and immunoblotting for Sicl. In (A), 5% of input and 25% of the bound fractions were loaded.

Figure 5. RpnlO UTM domain and Rad23 serve redundant roles in Sicl turnover in vivo. (A-F) Wild-type and mutant cells expressing a GAL1 -dήven, epitope-tagged (HaHis₆) allele of SIC 1 in addition to endogenous untagged SIC1 were arrested with factor and released synchronously into the cell cycle at 25°C (except rpnlOΔrad23Δ, which were released at 30°C because they grew poorly at 25°C). Extracts were prepared at the indicated time points and analyzed by SDS-PAGE followed by immunoblotting with anti-Sicl serum that detects both the endogenous and the epitope-tagged versions of Sicl. (G) Wild-type, rpnl 0 VWA rad23Δ , and rpnl 0Δ rad23Δ cells collected at the indicated time points were evaluated for cell cycle distribution by flow cytometry. Figure 6. UPS substrates have differential requirements for multiubiquitin chain receptors in vivo. For experiments shown in panels (A)-(D), aliquots of cells of the indicated genotypes were withdrawn at various times after initiation of chase (min), and whole cell lysates were fractionated by SDS-PAGE and immunoblotted with the indicated antibodies. (A) Wild-type and mutant cells expressing HA epitope- tagged Cln2 from the GAL1 promoter were grown in YP raffinose at 30°C, and expression of Cln2-HA was induced with 2% galactose at 25°C for 90 min. Induction was terminated and chase was initiated by transfer of cells to YP-2%> dextrose. (B) To monitor turnover of Far 1, wild-type and mutant cells were arrested with factor for 3 hr at 25 °C, and the chase period was initiated by release into fresh medium in the absence of factor, which results in rapid down-regulation of Far 1 message. (C) The stability of CPY*HA was monitored upon initiating a chase period by adding 100 μg/ml cycloheximide to wild-type and mutant cultures at 25°C. (D) Cycloheximide chase was done as described in (C) to monitor turnover of Degl-GFP. Figure 7. Model for physiological targeting pathways that deliver ubiquitinated substrates to the 26S proteasome. The schematic shows the 20S proteolytic core capped by the base, which comprises a hexameric ring of the AAA ATPases (Rptl-Rpt6) and the PC repeat containing proteins Rpnl and Rpn2 (collectively depicted as a oval). Rad23 and RpnlO associate with the proteasome via the Rpnl/Rpn2 subunits to deliver substrates tethered to their Ub binding domains (UBD), including Farl, Sicl, Gic2, and Clb2. Deubiquitination and degradation of substrates delivered by Rad23 requires a facilitator activity (FA) encoded within the VWA domain of RpnlO. Dsk2, a UBA domain containing protein like Rad23, is postulated to also deliver substrates to the same entry port used by Rad23, but the identity of these substrates remains unknown. Ufd 1 -containing complexes that contain Cdc48 are proposed to deliver ERAD and non-ERAD substrates such as CPY*, Degl, and Cdc5 to the proteasome, but the putative proteasome binding domain (PBD) and docking site employed by this complex remain unknown. Ubiquitinated Cln2 is targeted for degradation by a pathway that remains unknown but does not require the activity of RpnlO, Rad23, Dsk2, or Ufdl. It is possible that Cln2 gains access to the proteasome via the putative Rpt5 gateway or an unknown receptor or utilizes multiple receptor pathways in a highly redundant manner.

Detailed Description of the Invention I. Overview One aspect of the invention provides a system for assaying the activity of a ubiquitin-proteasome pathway, said system comprising: (1) a defective 26S proteasome preparation substantially unable to degrade a ubiquitin-conjugated target protein, (2) one or more functional multiubiquitin chain binding protein(s) (MCBP(s)) in an amount that substantially restores the ubiquitin-mediated proteasome degradation of said target protein to a wild-type level, with the proviso that said system is not a wild-type 26S proteasome preparation. The "26S proteasome" is composed of a core protease, known as the 20S proteasome, capped at one or both ends by the 19S regulatory complex (RC). The RC is composed of at least 18 different subunits in two subcomplexes, the base and the lid, which form the portions proximal and distal to the 20S proteolytic core, respectively. The subject assay system is partly based on the surprising discovery that proteasome-targeting pathways downstream of the ubiquitin ligases exhibit a surprising degree of substrate specificity, such that certain specific proteins targeted for 26S proteasome degradation via polyubiquitination bind specifically to selected multiubiquitin chain binding proteins (MCBPs). Therefore, only certain (but not all) MCBPs are required for 26 S proteasome-mediated degradation of some target proteins, although one target protein may be directed to the 26S proteasome redundantly via more than one MCBP proteins / pathways. The subject assay system is also partly based on the surprising discovery that only an appropriate amount of MCBP can effectively reconstitute an assay system of the subject invention. In other words, excessive or insufficient amount of MCBP may not enhance, or may even inhibit the activity of the subject assay system, even though the same MCBP, when present in appropriate amounts, will greatly facilitate the subject assay. In a preferred embodiment, the amount of the functional MCBP(s) is no more than about 2-fold, preferably no more than about 1.5-fold in molar excess over 26S proteasomes in the system, which preferably is at the wild-type level. In certain embodiments, the subject assay system is an in vitro system. For example, the defective 26S proteasome preparation may be prepared from cell or tissue extracts, wherein the cell or tissue is defective in 26S proteasome function. To illustrate, the cell may be a low eukaryotic cell, such as budding yeast S. cerevisiae or fission yeast S. pombe. The yeast cell may have defective 26S proteasome for lack of (or have diminished level of) one or more of the MCBPs such as Rpnl Op, Rad23p, or Desk2p, due to a deletion of genes encoding such MCBPs. Alternatively, such MCBPs in wild-type yeast cell extracts may be depleted by, for example, immuno-depletion with antibodies specific for the MCBPs, selective chemical extraction of certain MCBPs, or affinity depletion using immobilized ligands known to bind such MCBPs. In addition, the expression of such MCBPs may be inhibited / abolished by RNAi or antisense technology. For example, wild- type yeast cells may be transfected by constructs (transiently or stably) expressing antisense ohgonucleotides which prevent / inhibit the expression of MCBPs. Similarly, wild-type yeast cells may be transfected by constructs (transiently or stably) expressing various RNAi constructs (e.g., encoding short hairpin RNA, or complementary strands of siRNA) which prevent / inhibit the expression of MCBPs. Antisense ohgonucleotides (either unmodified, or modified by unnatural nucleotides to enhance solubility, stability, and/or cellular uptake) or various siRNA, shRNA maybe directly added to in vitro culture of cells to suppress the expression of certain MCBPs or other proteins. 26S proteasome preparation made from such cells may be defective in degrading ubiquitinated target proteins. Such defective 26S proteasome preparations may contain no more than about 30%, 20%, 10%o, 5%>, or 1% of wild- type level of activity. These methods may also be used in higher eukaryotic cells, such as human cells / tissues or other non-human high eukaryotes. In another embodiment, the assay system of the invention is an in vivo system. For example, a yeast cell may be engineered to have a null allele of the gene rpnlO, and a null allele of the gene rad23. A transgene of rpnlO or rad23 under the control of a heterologous promoter, such as an inducible promoter may be introduced into the genome of the yeast cell. When the inducible promoter is not activated, the yeast 26S proteasome is defective in that the cell lacks Rpnl Op and Rad23p. Upon induction of the inducible promoter operatively linked to the transgenic rpnlO or rad23 gene, an artificial amount of RpnlOp or Rad23p may be expressed in the cell, thereby generating a non- wild-type 26S proteasome preparation in a living cell, which is suitable for assaying at least a subset of ubiquitin-conjugated target proteins, such as Sicl. A similar system may be implemented in higher eukaryotic cells, such as in a mammalian cell. The mammalian RpnlO and/or Rad23 genes may be knocked out, and heterologous genes under the control of inducible promoters may be introduced into such cells using, for example, selectable markers (e.g., neomycin or hygromycin resistant genes, etc.)Recombinant MCBPs may also be introduced into the intact cell by various means. In some embodiments, the MCBPs are introduced into the cell via electroporation. In other embodiments, the MCBPs are introduced via direct microinjection of the recombinantly produced (and optionally purified) MCBPs or polynucleotides encoding such MCBPs into the cell. In yet another embodiment, the MCBPs are introduced via transfecting or transforming the cell with polynucleotides which encode such MCBPs. In some embodiments, incubating the cells with MCBPs directly may allow entry of the MCBPs into the cell. Optionally, the direct entry of MCBPs may be enhanced by fusing the

MCBPs to certain polypeptides that mediate trans-plasma membrane translocation of polypeptides. Several proteins and small peptides have the ability to transduce or travel through biological membranes independent of classical receptor- or endocytosis-mediated pathways. Examples of these proteins include the HIV-1 TAT protein, the herpes simplex virus 1 (HSV-1) DNA-binding protein VP22, and the Drosophila Antennapedia (Antp) homeotic transcription factor. The small protein transduction domains (PTDs) from these proteins can be fused to other macromolecules, peptides or proteins to successfully transport them into a cell. Sequence alignments of the transduction domains from these proteins show a high basic amino acid content (Lys and Arg) which may facilitate interaction of these regions with negatively charged lipids in the membrane. Secondary structure analyses show no consistent structure between all three domains. The advantages of using fusions of these transduction domains is that protein entry is rapid, concentration-dependent and appears to work with difficult cell types. However, each of the three commonly used PTDs has its own unique considerations. While fusion proteins of > 1,000 residues are possible with the TAT and VP22 approaches, the size of the protein being transduced as an Antp fusion is limited to <100 residues. In certain embodiments, the VP22 strategy is used as an indirect method in that the vector bearing the fusion construct is transfected into cells where the fusion protein is made and the resulting protein then transduced into surrounding cells, hi some embodiments, the above-described PTD-mediated protein delivery systems covalently attach the transduction domains to the protein being delivered, either by creating a DNA construct in a specially designed vector, or by chemically cross-linking the protein and PTD via functional groups on each molecule. In certain embodiments, cationic lipids, such as liposomes may be used to introduce the subject MCBPs into the cells. Liposomes have been rigorously investigated as vehicles to deliver ohgonucleotides, DNA (gene) constructs, proteins, and small drug molecules into cells. Certain lipids, when placed in an aqueous solution and sonicated, form closed vesicles of a circularized lipid bilayer surrounding an aqueous compartment. These vesicles or liposomes can be formed in a solution containing the molecule to be delivered. The exact composition and/or mixture of cationic lipids used can be altered, depending upon the macromolecule of interest and the cell type used. For example, Pierce has introduced Pro-Ject Protein Transfection Reagent (Product # 89850), which utilizes a unique cationic lipid formulation that is non-cytotoxic and is capable of delivering a variety of proteins into numerous cell types. The protein to be delivered is mixed with the liposome reagent and is overlay onto cultured cells. The liposome:protein complex fuses with the cell membrane or is internalized via an endosome. The protein or macromolecule of interest is released from the complex into the cytoplasm free of lipids and escaping lysosomal degradation. The non-covalent nature of these complexes is a major advantage of the liposome strategy as the delivered protein is not modified and therefore maintains its activity. As with all direct protein delivery systems, the time saved is significant over the indirect DNA transfection procedures. Certain test agents may be delivered together with the MCBPs into the cell via the liposome preparation. In certain embodiments, the MCBPs may comprise one or more of: an RpnlO polypeptide, an Rad23 polypeptide, a Dsk2 polypeptide, a Cdc48/Ufdl/Npl4 complex, or an Rpt5 polypeptide. For each of these MCBP proteins, homologs from different species may be interchangeable functionally, such that a human RpnlO protein may be supplemented to a yeast 26S proteasome preparation lacking a yeast RpnlO protein to reconstitute a functional assay system, and vice versa. Therefore, "an RpnlO protein / polypeptide" encompasses all functionally equivalent / interchangeable RpnlO proteins, even proteins from distantly-related species, that can substitute one another in at least one defective 26S proteasome preparation. The same applies to all other MCBP proteins, including Rad23, Desk2, Cdc48, Ufdl, Nρl4, Parkin, Ufd4, and Hul5. Thus in certain embodiments, at least one of said MCBP(s) may be from a species different from the species of the defective 26S proteasome preparation. For example, the defective 26S proteasome preparation may be from a non-human eukaryote, such as a yeast. In some embodiments, the MCBP(s) may comprise one or more human Rad23 homologs selected from hHR23A and hHR23B, one or more human Dsk2 homologs selected from hPLIC-1 and hPLIC-2, or human RpnlO homolog S5A. In other embodiments, the defective 26S proteasome preparation is from a human. In certain embodiments, the presence of an appropriate amount of one or more functional MCBP(s) in the defective 26S proteasome preparation restores at least about 30%, 50%, 60%, 70%, 80%, 90%, or nearly 100% of the wild-type level of ubiquitin-mediated proteasome degradation of the target protein. In certain embodiments, the amount of one or more functional MCBP(s) in the subject system is an optimum amount determined by titration. For example, in an in vitro assay system embodiment, the titration may be performed by providing and testing a range of different concentrations of MCBP(s) added to the defective 26S proteasome preparation, or intentionally choosing an MCBP from a different species. In certain other embodiments, the system is an in vivo assay system, and the titration may be performed by operatively linking the coding sequence of the MCBP(s) with a range of promoters with different transcription / translation strength, or different activity under different inducing signals. Such promoters are well-known in the art. In certain embodiments, the amount of one or more functional MCBP(s) in the subject assay system is no more than 2-fold, preferably no more than 1.5-fold molar excess over wild-type proteasomes. The assay system of the invention may be used for a wide range of ubiquitinated target proteins. In some embodiments, the target protein is one or more of: Sicl, Farl, Clb2, Gic2, CPY*, Cln2, or a homolog thereof. In some embodiments, the assay system of the invention lacks one or more MCBP(s). For example, the lacking MCBP(s) may be different from those functional MCBP(s) added to the assay system. To illustrate, in certain embodiments of the assay system, a yeast 26S proteasome preparation may lack RpnlOp, but the assay system contains: (a) a Rad23 polypeptide or a Dsk2 polypeptide; and (b) a mutant RpnlO lacking the VWA domain, or a mutant RpnlO lacking the UIM domain. In some embodiments, the subject assay system may be adapted to the assay of only a specific subset of target proteins, but not all other target proteins. The MCBP proteins may be obtained and provided directly to the subject system via recombinant technology. Alternatively, genes encoding the MCBP proteins may be introduced into the system, and MCBP proteins expressed from such genes. In certain embodiments, the assay system of the invention further comprises, if it does not have one already, a facilitator, such as the VWA domain of an RpnlO protein. Another aspect of the invention provides a method for monitoring the ability of at least one multiubiquitin chain binding protein (MCBPs) to promote degradation of at least one ubiquitin-conjugated target protein by a 26S proteasome comprising: using the system of any of the above-described assay systems, and determining the degree of degradation of the target proteins. h certain embodiments, the MCBPs of the system may include one or more of: an RpnlO protein, a Dsk2 protein, an Rad23 protein, a Ufd/Cdc48 protein complex, and an Rpt5 protein. In a preferred embodiment, the MCBPs are no more than about 2-fold, preferably no more than about 1.5-fold in molar excess over 26S proteasomes in the system. The 26S proteasomes in the system are preferably at the wild-type level. Yet another aspect of the invention provides a method for screening for an agent that inhibits the degradation of a ubiquitinated target protein by a 26S proteasome, the method comprising: (a) incubating the ubiquitinated target protein in any of the above-described assay systems; (b) determining and comparing the degree of degradation of said target protein in the presence or absence of a test agent, wherein more complete degradation of said target protein in the absence of the test agent than the presence of the test agent is indicative that the test agent inhibits the degradation of said target protein. The candidate agents to be screened can be any molecule, including nucleosides, nucleotides or polynucleotides; amino acids, polypeptides; mono- or polysaccharides; various lipids; steroids; ions; etc. hi certain embodiments, the agents may be tissue extracts or compositions with unclear ingredients, including pools of molecules with unknown relative ratio, which might be further analyzed if the extract or pool is initially found to be an effective inhibitor. The candidate agents may be present in a natural or synthetic library, which can be screened in a high- throughput fashion using the in vitro and/or in vivo assay systems of the invention. In certain embodiments, the same assay may be used to identify agents that enhance (rather than inhibit) the function of 26 S proteasome in the assay system of the invention. Another aspect of the invention provides method of screening for 26S proteasome modulators, comprising: (a) incubating a plurality of candidate agents, each individually or in combination, with a recombinant MCBP protein and a ubiquitinated target protein; (b) measuring binding of the MCBP protein to the ubiquitinated target protein in the presence or absence of the candidate agent(s); wherein significant change in binding in the presence of the candidate agent(s) is indicative that the candidate agent(s) is a proteasome modulator. In some embodiments, the modulator is an inhibitor. In other embodiments, the modulator is an enhancer / stimulator of 26S proteasome function. In some embodiments, the method further comprises determining the activity of a 26S proteasome preparation comprising the MCBP protein, in the presence of the agent(s) identified in (b) as modulator(s), wherein a modulated activity of the 26S proteasome verifies that the candidate agent(s) is/are 26S proteasome modulator(s). In some embodiments, the activity of the 26S proteasome is determined by using the ubiquitinated target protein as a substrate. In some embodiments, the 26S proteasome is isolated from wild-type cells. In some embodiments, the 26S proteasome is reconstituted from a defective 26S proteasome substantially incapable of degrading the ubiquitinated target protein, and a recombinantly produced MCBP protein. In some embodiment, the candidate agents are from a library, such as a polynucleotide library (which may be able to express proteins), a polypeptide library, a small chemical compound library (preferably with max. molecular weight of about 5000 Da.), an organic compound library, an inorganic compound library, a library of chemicals synthesized by split-pool methods, or a library of compounds with unknown identity. In some embodiments, step (1) is carried ou by high throughput screening.

For example, more than about 10, 50, 100, 400, 1000, 1500, 5000 or more compounds may be screening in parallel (such as in 96-well, 384-well, 1536-well plates, etc.) or nearly parallel. In some embodiments, step (1) is effectuated by direct in vitro binding assay between the MCBP and the candidate agents. For example, BIOCORE can be used to directly measure the binding affinity between the MCBP and the candidate agents. In a preferred embodiment, step (1) may be effectuated by a gel-shift binding assay. In some embodiments, step (1) is effectuated by two-hybrid binding assay using the MCBP as a bait protein. The two-hybrid assay may be performed in a bacterial, a yeast, or mammalian system. Reverse two-hybrid assay may also be used to isolate agents that bind the MCBP protein. Yet a further aspect of the invention provides a method for screening for an agent that inhibits a ubiquitinated target protein's entry into the proteasome, the method comprising: (a) incubating, in the presence of a test agent, the ubiquitinated target protein and a sufficient amount of a recombinant MCBP protein that would restore degradation activity of a proteasome preparation lacking a corresponding functional MCBP; (b) adding a 26S proteasome preparation, wherein the MCBP corresponding to the recombinant MCBP was absent or defective in said 26S proteasome preparation; (c) determining the effect of the test agent, wherein substantial reduction of target protein degradation is indicative that the test agent inhibits the ubiquitinated target protein's entry into the proteasome. In some embodiments, the test agent inhibits at least about 50% of target protein degradation. In some embodiments, the test agent inhibits nearly 100% of target protein degradation. Another aspect of the invention provides a method to screen for agents that can modulate (e.g. enhance or inhibit) the function of a MCBP-mediated 26S proteasome activity, the method comprising: (1) isolating, from among a plurality of candidate agents, one or more agent(s), if any, that can interact with the MCBP protein; (2) determining the MCBP-mediated 26S proteasome activity in the presence or absence of said candidate agents identified in (1), wherein an enhanced or inhibited proteasome activity, respectively, in the presence of the candidate agent(s) is indicative that the candidate agent(s) is/are enhancer or inhibitor, respectively, of the function of the MCBP-mediated 26S proteasome activity. In some embodiments, the MCBP is Rad23, RpnlO, Dsk2, Rpt2, or Cdc48/Ufdl complex.

It is contemplated that all embodiments described above are applicable to all different aspects of the invention. It is also contemplated that any of the above embodiments can be freely combined with one or more other such embodiments whenever appropriate. Specific embodiments of the invention are described in more detail below. However, these are illustrative embodiments, and should not be construed as limiting in any respect.

MCBPs The multiubiquitin chain binding proteins (MCBPs) of the subject invention are proteins or polypeptides that can bind selectively to the polyubiquitin chains (polyUb) of the ubiquitinated target proteins. These proteins typically contain a domain (e.g. polyUb binding domain) that binds to the polyUb moiety, especially the polyUb with at least 2, preferably 4 ubiquitin moieties (e.g., between 2-7 ubiquitin moieties). Examples of such domains include the ULM domain of Rpn 10, and the UBA domain of Rad23, which are distinct domains that bind the polyUb chain. Some MCBPs may also contain at least one domain that binds to the 26S proteasome via at least one of its component proteins. See, for example, the UbL domain of the Rad23 and Dsk2 proteins. The MCBPs may contain additional domains with additional functions, such as the VWA domain of the RpnlO protein, which may function independently of the polyUb binding domain as a "facilitator." These MCBPs may be obtained from various species, and may be used in defective 26S proteasome preparation from a different species, to assay the degradation of a target protein, including target protein of a third species. The sequences of some of the MCBPs are listed below, while other homologs, derivatives, chimeras, or functional equivalents thereof (such as those produced by random mutagenesis coupled with functional screening, or produced by recombinant DNA technology) are within the scope of the invention.

RpnlO: "An Rpnl 0 protein / polypeptide" includes various homologs or functional equivalents of the RpnlO family of proteins, as represented by the human and budding yeast RpnlO proteins. The human RpnlO homolog, or 26S proteinase chain 5a protein (Ferrell et al, FEBSLett. 381(1-2): 143-8, 1996), is listed below (swissprot: locus PSD4_HUMAN, accession P55036):

1 mvlestmvcv dnseymrngd flptrlqaqq davnivchsk trsnpennvg litlandcev 61 lttltpdtgr ilsklhtvqp kgkitfctgi rvahlalkhr qgknhkrarii afvgspvedn 121 ekdlvklakr lkkekvnvdi infgeeevnt ekltafvntl ngkdgtgshl vtvppgpsla 181 dalisspila geggamlglg asdfefgvdp sadpelalal rvsmeeqrqr qeeearraaa 241 asaaeagiat tgtedsddal lkmtisqqef grtglpdlss mteeeqiaya mqmslqgaef 301 gqaesadida ssamdtsepa keeddydvmq dpeflqsvle nlpgvdpnne airnamgsla 361 sqatkdgkkd kkeedkk

The S. cerevisiae RpnlO homolog, or RpnlOp (Johnston et al, Science 265(5181): 2077-82, 1994), is listed below (NP_012070): 1 mvleatvlvi dnseysrngd fprtrfeaqi dsvefifqak rnsnpentvg lisgaganpr 61 vlstftaefg kilaglhdtq iegklhmata lqiaqltlkh rqnkvqhqri vafvcspisd 121 srdelirlak tlkknnvavd iinfgeieqn telldefiaa vnnpqeetsh lltvtpgprl 181 lyeniasspi ileegssgmg afggsggdsd angtfmdfgv dpsmdpelam alrlsmeeeq 241 qrqerlrqqq qqqdqpeqse qpeqhqdk

BLASTp search of the available non-redundant databases (GenBank CDS translations + PDB + SwissProt + PER. + PRF, excluding environmental samples) revealed numerous homologs, including those from other species, such as: human proteasome 26S non-ATPase subunit 4 isoform 1 (NP_002801) and isoform 2

(NP_722544); rat Psmd4 protein (NP_112621); mouse proteasome 26S non-ATPase subunit 4 (NP_032977); Xenopus PSMD4 protein (AAH43989); zebra fish homolog NP_001002112; Drosophila melanogaster AT14053p protein (AAL90071); etc. A number of lower eukaryote homologs are also available, include Saccharomyces cerevisiae (NP_012070); Schizosaccharomyces pombe

(CAA22589); Candida albicans SC5314 protein (EAK95200); Dictyostelium discoideum (AAO53083); etc. Plant RpnlO include: Arabidopsis thaliana 26S proteasome regulatory subunit S5A (RPN10) (NP_195575). All these and other unlisted homologs may be used in the assay system of the invention.

Rad23: "An Rad23 protein / polypeptide" includes various homologs or functional equivalents of the Rad23 family of proteins, as represented by the human and yeast Rad23 proteins. The human Rad23 homolog NP_004619 (see, for example, Masutani et al, EMBO J. 13(8): 1831-43, 1994), is listed below: 1 markraagge prgrelrsqk skakskarre eeeedafede kppkksllsk vsqgkrkrgc 61 s pggsadgp akkkvakvtv ksenlkvikd ealsdgddlr dfpsdlkkah hlkrgatmne 121 dsneeeeese ndweeveels epvlgdvres tafsrsllpv kpveieietp eqaktrerse 181 kiklefetyl rramkrfnkg vhedthkvhl lcllangfyr nnicsqpdlh aiglsiipar 241 ftrvlprdvd tyylsnlvkw figtftvnae lsaseqdnlq ttlerrfaiy sarddeelvh 301 ifllilralq lltrlvlslq piplksatak gkkpskerlt adpggssets sqvlenhtkp 361 ktskgtkqee tfakgtcrps akgkrnkggr kkrskpssse edegpgdkqe katqrrphgr 421 errvasrvsy keesgsdeag sgsdfelssg easdpsdeds epgppkqrka papqrtkags 481 ksasrt rgs hrkdpslpaa ssssssskrg kkmcsdgeka ekrsiagidq wlevfceqee 541 kwvcvdcv g wgqpltcyk yatkpmtyw gidsdgwvrd vtqrydpvw tvtrkcrvda 601 ewwaetlrpy qspfmdrekk edlefqakhm dqplptaigl yknhplyalk rhllkyeaiy 661 petaailgyc rgeavysrdc vhtlhsrdtw lkkarwrlg evpykmvkgf snrarkarla 721 epqlreendl glfgywqtee yqppvavdgk vprnefgnvy lflpsmmpig cvqlnlpnlh 781 rvarkldidc vqaitgfdfh ggyshpvtdg yivceefkdv lltaweneqa vierkekekk 841 ekralgnwkl lakgllirer lkrrygpkse aaaphtdagg glssdeeegt ssqaeaaril 901 aaswpqnred eekqklkggp kktkrekkaa ashlfpfekl

The yeast Rad23 homolog Rad23ρ (NP_010877), (see, for example, Dietrich et al, Nature 387(6632 Suppl): 78-81, 1997), is listed below: 1 mvsltfknfk kekvpldlep sntiletktk laqsiscees qikliysgkv lqdsktvsec 61 glkdgdqwf mvsqkkstkt kvteppiape sattpgrens teaspstdas aapaatapeg 121 sqpqeeqtat tertesastp gfwgterne tierimemgy qreeveralr aafnnpdrav 181 eyllmgipen lrqpepqqqt aaaaeqpsta attaeqpaed dlfaqaaqgg nassgalgtt 241 ggatdaaqgg ppgsigltve dllslrqws gnpealapll enisarypql rehimanpev 301 fvsmlleavg dnmqdvmega ddmvegedie vtgeaaaagl gqgegegsfq vdytpeddqa 361 isrlcelgfe rdlviqvyfa cdkneeaaan ilfsdhad

BLASTp search of the available non-redundant databases (GenBank CDS translations + PDB + SwissProt + PIR + PRF, excluding environmental samples) revealed numerous homologs, including those from other species, such as: human (NP_004984); human (NP_109376); human UV excision repair protein RAD23 homolog B (hHR23B) (P54727); human UV excision repair protein RAD23 homolog A (hHR23A) (P54725); rat (XP_232194); mouse (NP_033557, NP_033036, and NP_033037); Drosophila (NP_476862); Saccharomyces cerevisiae (NP_010877); etc. All these and other unlisted homologs may be used in the assay system of the invention.

Dsk2: "An Dsk2 protein / polypeptide" includes various homologs or functional equivalents of the Dsk2 family of proteins, as represented by the human and yeast Dsk2 proteins. The human Dsk2 homolog Ubiquilin 1 (Protein linking IAP with cytoskeleton-1, or liPLIC-1), see, for example, Kleijnen et al, Mol Cell. 6(2): 409- 19, 2000, is listed below as Q9UMX0 or AAG02473:

1 maesgesggp pgsqdsaaga egagapaaaa saepkimkvt vktpkekeef avpenssvqq 61 fkeeiskrfk shtdqlvlif agkilkdqdt lsqhgihdgl tvhlviktqn rpqdhsaqqt 121 ntagsnvtts stpnsnstsg satsnpfglg glgglaglss lglnttnfse lqsqmqrqll 181 snpemmvqim enpfvqsmls npdlmrqlim anpqmqqliq rnpeish ln npdimrqtle 241 larnpammqe mmrnqdrals nlesipggyn alrrmytdiq epmlsaaqeq fggnpfaslv 301 sntssgegsq psrtenrdpl pnpwapqtsq sssassgtas tvggttgsta sgtsgqstta 361 pnlvpgvgas mfntpgmqsl Iqqitenpql mqnmlsapym rsmmqslsqn pdlaaqramln 421 nplfagnpql qeqmrqqlpt flqqmqnpdt lsamsnpram qallqiqqgl qtlateapgl 481 ipgftpglga lgstggssgt ngsnatpsen tsptagttep ghqqfiqqml qalagvnpql 541 qnpevrfqqq leqlsamgfl nreanlqali atggdinaai erllgsqps

The yeast Dsk2 homolog Dsk2p (NP_014003), (see, for example, Bowman et al, Nature 387(6632 Suppl): 90-3, 1997), is listed below: 1 mslnihiksg qdkwevnvap estvlqfkea inkangipva nqrliysgki lkddqtvesy 61 hiqdghsvhl vksqpkpqta saagannata tgaaagtgat pnmssgqsag fnpladltsa 121 ryagylnmps admfgpdgga lnndsnnqde llrmmenpif qsqmnemlsn pqmldfmiqs 181 npqlqamgpq arqmlqspmf rqmltnpdmi rqsmqfarmm dpnagmgsag gaasafpapg 241 gdapeegsnt nttsssntgn nagtnagtna gantaanpfa sllnpalnpf anagnaastg 301 mpafdpalla smfqppvqas qaedtrppee ryehqlrqln dmgffdfdrn vaalrrsggs 361 vqgaldslln gdv

BLASTp search of the available non-redundant databases (GenBank CDS translations + PDB + SwissProt + PER. + PRF, excluding environmental samples) revealed numerous homologs, including those from other species, such as: bovine ubiquilin 1 (NP_777053); human ubiquilin 1 isoform 2 (NP_444295); rat (NP_446199); mouse (AAH26847); Saccharomyces cerevisiae (NP_010877); Schizosaccharomycespom.be (NP_596231); etc. Ufdl "A Ufdl protein / polypeptide" includes various homologs or functional equivalents of the Ufdl family of proteins, as represented by the human and yeast Ufdl proteins. The human Ufdl homologs, ubiquitin fusion degradation 1 proteins, include those sequences disclosed in XP_496272, CAC20414, and AAD28788, the last of which is listed below: 1 mfsfnmfdhp iprvfqnrfs tqyrcfsvsm lagpndrsdv ekggkiimpp saldqlsrln 61 itypmlfklt nknsdrmthc gvlefvadeg icylphwmmq nllleegglv qvesvnlqva 121 tyskfqpqsp dflditnpka vlenalrnfa clttgdviai nynekiyelr vmetkpdkav 181 siiecdmnvd fdaplgykep erqvqheest egeadhsgya gelgfrafsg sgnrldgkkk 241 gvepspspik pgdikrgipn yefklgkitf irnsrplvkk veedeaggrf vafsgegqsl 301 rkkgrkp

The yeast Ufdl homolog Ufdlp (NP_011562), (see, for example, Tettelin et al, Nature 387(6632 Suppl): 81-4, 1997), is listed below:

1 mfsgfssfgg gngfvnmpqt feeffrcypi ammndrirkd danfggkifl ppsalsklsm 61 Inirypmlfk Itanetgrvt hggvlefiae egrvylpq m metlgiqpgs llqisstdvp 121 Igqfvklepq svdfldisdp kavlenvlrn fstltvddvi eisyngktfk ikilevkpes 181 ssksicviet dlvtdfappv gyvepdykal kaqqdkekkn sfgkgqvldp svlgqgsmst 241 ridyagians srnklskfvg qgqnisgkap kaepkqdikd mkitfdgepa kldlpegqlf 301 fgfpmvlpke deesaagsks seqnfqgqgi slrksnkrkt ksdhdssksk apkspeviei 361 d

Other Ufdl -like proteins include: rat (NP_445870); mouse (NP_035802 or AAH06630); Drosophila melanogaster (Q9VTF9 or AAK00731); Caenorhabditis elegans (NP_502348 or NP_502349); Schizosaccharomyces pombe (NP_596780 or CAB59876); Arabidopsis thaliana (NP_974709, NP_973557, NP_180471, NP_565504, or NP_568048); Neurospora crassa (CAD70384), etc. Rpt5: "A Rpt5 protein / polypeptide" includes various homologs or functional equivalents of the Rpt5 family of proteins, as represented by the human and yeast Rpt5 proteins. The human Rpt5 homologs, PSMC3 protein, include AAH73165, which is listed below (other sequences include NP_002795):

1 agagrvksrd lwgevfhwsr egfkkdginl plrlppgpif cpllqemnll pniespvtrq 61 ekmatvwdea eqdgigeevl kmsteeiiqr trlldseiki mksevlrvth elqamkdkik 121 ensekikvnk tlpylvsnvi elldvdpndq eedganidld sqrkgkcavi ktstrqtyfl 181 pviglvdaek lkpgdlvgvn kdsyliletl pteydsrvka mevderpteq ysdiggldkq 241 iqelveaivl pmnhkekfen lgiqppkgvl ygppgtgkt llaracaaqt katflklagp 301 qlvqmfigdg aklvrdafal akekapsiif ideldaigtk rfdsekagdr evqrtmlell 361 nqldgfqpnt qvkviaatnr vdildpallr sgrldrkief pmpneearar imqihsrkmn 421 vspdvnyeel arctddfnga qckavcveag mialrrgate lthedymegi levqakkkan 481 lqyya

The yeast Rpt5 homolog, Rpt5p (NP_014760), (see, for example, Dujon et al, Nature 387(6632 Suppl): 98-102, 1997), is listed below: 1 matleeldaq tlpgddeldq eilnlstqel qtraklldne irifrselqr lshennvmle 61 kikdnkekik nnrqlpylva nwevmdmne iedkensest tqggnvnldn tavgkaawk 121 tssrqtvflp mvglvdpdkl kpndlvgvnk dsylildtlp sefdsrvkam evdekptety 181 sdvggldkqi eelveaivlp mkradkfkdm girapkgalm ygppgtgktl laracaaqtn 241 atflklaapq lvqmyigega klvrdafala kekaptiifi deldaigtkr fdseksgdre 301 vqrtmlelln qldgfssddr vkvlaatnrv dvldpallrs grldrkiefp lpsedsraqi 361 lqihsrk tt dddinwqela rstdefngaq lkavtveagm ialrngqssv khedfvegis 421 evqarksksv sfya

Other Rpt5 -like proteins include: rat (Q63569); mouse (AAH05783); Xenopus laevis Psmc3-prov protein (AAH54164); Candida albicans (EAK96478); Schizosaccharomyces pombe (T11634); Neurospora crassa (CAF06032), etc.

Exemplary 26S proteasome preparation The following protocol may be used for preparing 26S proteasome preparations (wild-type or defective) from yeast cells, such as wild-type strain or mutant strain with one or more MCBP proteins deleted. Similar methods can be used to prepare other tissue / cell extracts for isolation of 26S proteasome. Wild-type and/or mutant cells are grown to an O.D. of about 0.2-0.3. Cells can be harvested by centrifugation and drop frozen in liquid nitrogen before later use. Extracts from wild-type or mutant cells may be passed through an affinity column that specifically bind one component of the 26S proteasome, such as a component protein tagged by the FLAG epitope (use anti-Flag M2 resin in the presence of ATP in this case). The column can then be washed with buffer containing ATP as described for 26S purification. Resin-immobilized 26S proteasomes can then be incubated with 1 mM phenanthroline, 2.5 μM Ub aldehyde, 100 μM MG132, 1 mM ATP, and 5 mM MgCl₂ in the absence or presence of the various MCBPs on ice for 60 min. before the ubiquitinated substrate is added. The assay can be carried out by 90 min incubation at 4°C. The bound fraction can be washed and analyzed by SDS-PAGE and immunoblotting for the target protein.

Degradation and Deubiquitination Assays Ubiquitinated substrates such as MbpSicl (see Seol et al, Genes Dev. 13: 1614-1626, 1999, incorporated herein by reference) and affinity-purified 26S proteasomes (Verma et al, Mol. Biol. Cell 11: 3425-3439, 2000, incorporated herein by reference) may be prepared essentially as described in these two references (see below). Degradation and deubiquitination assays (-300 nM substrate, -100 nM proteasome, incubated at 30°C for 5 min) may be conducted as described previously (see, for example, Verma et al, Science 298: 611-615, 2002, incorporated herein by reference). Briefly, to make 26S proteasomes preparations by affinity purification, tagged or untagged cells (e.g., yeast strains) are grown to an optical density of

2.0, typically in 9 liters of synthetic medium containing 0.67% yeast nitrogen base minus amino acids, 2% dextrose, 0.5%> casamino acids, and 20 mg/1 adenine and tryptophan. Cells are harvested and washed once with ice cold water. The cell pellet is drop frozen in liquid N₂, placed inside a mortar (which in turn is nestled inside an ice bucket filled with dry ice), and manually ground with the pestle to a fine powder (typically 15-30 min, depending on the amount being ground). The pellet being ground is kept frozen by scooping liquid N₂ into the mortar every 2 min. The ground powder is collected in a 50-ml screw-cap tube and drop frozen in liquid N₂. The powder is thawed in one pellet volume of 50 mM Tris, pH 7.5, 150 mM NaCl, 10% glycerol, 5 mM MgCl₂ (buffer A). ATP and 1 O ATP-regenerating mix (ARS) (Verma et al, 1997) are added to final concentrations of 5 mM and lx, respectively. Where indicated, ATP and ARS are substituted with 5 mM ATP-γ-S. The thawed cell lysate is centrifuged in an SS34 rotor (Sorvall, Newtown, CT) for 20 min at 17,000 rpm, and the pellet is discarded. A 13 -ml aliquot of the supernatant (-130 mg of protein), is supplemented again with 5 mM ATP (or ATP-γ-S) and 1 x ARS and is mixed with 300 μl of anti-Flag M2 agarose beads (Sigma, St. Louis, MO) for 90 min on a rotating wheel at 5°C. The beads are then collected, transferred to 2-ml micro fuge tubes, and washed with 50 volumes total of buffer A containing 2 mM ATP plus 0.2% Triton. The beads are next washed twice with buffer A containing 2 mM ATP, and specifically bound proteins are eluted for 3 h at 5°C with three bead volumes of elution buffer containing 25 mM Tris, pH 7.5, 150 mM NaCl, 15% glycerol, 5 mM MgCl₂, 2 mM ATP, and 100 μg/mlFlag peptide. Typically, the yield is -250 μg of purified 26S from 130 mg of lysate. To purify 20S proteasomes, the entire procedure described above may be carried out in the absence of ATP and 1* ARS. To purify 19S caps, the entire protocol described above may be carried out using the RPT1^FH strain. Purification of the 19S cap may be carried out in the absence or presence of ATP and 1 ARS, or in the presence of ATP-γ-S. The above methods are merely illustrative embodiments, and are by no means limiting. Proteasomes isolated with similar or different methods may also be used in the subject invention. Sicl ubiquitination reactions are performed as described in Feldman et al,

Cell 91: 221-230, 1997, incorporated herein by reference). Briefly, in an exemplary embodiment, all ubiquitination reactions contained: 4 μg ubiquitin, 60 ng Cdc34p, 25 ng ^His6Ubalp, 1 μL of a 10x ATP-regenerating system (20 mM HEPES [pH 7.2], 10 mM ATP, 10 mM MgOAc, 300 mM creatinine phosphate, 0.5 mg/niL creatinine phosphokinase), 1 μL of 10x reaction buffer (40 mM MgOAc, 10 mM DTT, 1 mM PMSF), and -10,000 cpm [³⁵S]-MBP-Siclp^MH6 (-20 ng). Reactions performed with crude lysates included: 1 μL each of Sf9 cell lysate containing Cdc4p, Cdc53p, or Skplp; and 2 μl of affinity-purified Cln2p/Cdc28p^HA/Ckslp inase. Reactions utilizing pure protein components contained 25 ng of either GST-Skplp or Skplp^His6, plus 0.5 μL each of eluted ^PHCdc4p and Cdc53p^PH (from a 50 μL immunopurification as described above). Reactions are brought to a final volume of 10 μL with 20 mM HEPES [pH 7.4], 100 mM KOAc, 1 mM DTT. All components are mixed and incubated at 25°C for 90 minutes (unless otherwise indicated). Reactions are terminated by addition of Laemmeli sample buffer, resolved by SDS- PAGE, and visualized by autoradiography. In another exemplary embodiment, ubiquitinated target proteins, such as Sicl, are provided at -300 nM, and were incubated with purified 26S proteasomes (-100 nM) at 30°C for 5 min.

Preparation of GST-Fusion Proteins GST-fusion proteins may be expressed in a suitable host cell, such as BL21/pLysS, according to standard procedures. Proteins maybe eluted from glutathione resin with 50 mM Tris (pH 8.8), 50 mM NaCl, 5 mM DTT, 1 mM EDTA, and 40 mM glutathione at 4°C for 3 hr and then dialyzed against buffer containing 25 mM Tris (pH 7.5), 100 mM NaCl, and 15% glycerol. Aliquots may be drop frozen in liquid N and stored at -70°C.

FACS Analysis Yeast cells may be processed for flow cytometry as described, for example, in Verma et al, Science 278: 455-460, 1997, incorporated herein by reference.

Identifying MCBP Modulators (Inhibitors or Enhancers) One aspect of the invention provides a method to screen for agents that can modulate (e.g. enhance or inhibit) the function of a MCBP-mediated 26S proteasome activity, the method comprising: (1) isolating, from among a plurality of candidate agents, one or more agent(s), if any, that can interact with the MCBP protein; (2) determining the MCBP-mediated 26S proteasome activity in the presence or absence of said candidate agents identified in (1), wherein an enhanced or inhibited proteasome activity, respectively, in the presence of the candidate agent(s) is indicative that the candidate agent(s) is/are enhancer or inhibitor, respectively, of the function of the MCBP-mediated 26S proteasome activity. In some embodiments, step (1) is effectuated by direct in vitro binding assay between the MCBP and the candidate agents. For example, the MCBPs (or epitope tagged versions, such as GST-, MBP-, 6-His, HA-, Ig-, FLAG-tagged, etc.) may be immobilized on a solid support, so that the candidate agents may contact the immobilized MCBPs for direct binding assay. For example, BIACORE's SPR-based technology can be used to directly measure the binding affinity between the MCBP and the candidate agents. As is known in the art, surface plasmon resonance (SPR) allows sensitive detection of molecular interactions in real time, without the use of labels. In brief, SPR-based biosensors monitor interactions by measuring the mass concentration of biomolecules close to a surface. The surface is made specific by attaching one of the interacting partners, such as the MCBP protein. Sample containing the other partner(s), such as the library of candidate agents, either sequencially or simultaneously, flows over the surface. When molecules from the sample bind to the interactant attached to the surface, the local concentration changes and an SPR response is measured. The response is directly proportional to the mass of molecules that bind to the surface. The sensor chips for immobilizing binding partners (e.g. MCBPs), as well as the instrument for measuring the binding events are all commercially available from BIACORE (Neuchatel, Switzerland). Specifically, Biacore provides series of novel technology solutions that enhance the efficiency of the drug discovery and development process, from basic research through to manufacturing. Such systems include Biacore 3000, Biacore S51, Biacore 3000GxP, Biacore C, etc. hi some embodiments, various two-hybrid assays may be used to identify candidate agents that bind a target MCBP protein. Traditional yeast two-hybrid screening is described in many scientific and patent literatures, including, for example, U.S. Pat. No. 6,562,576. It is a method for detecting protein-protein interaction, in which two fusion proteins are prepared and allowed to interact with each other in yeast cells. The interaction between the two fusion proteins leads to protein trans-splicing, generating an active and detectable reporter. So MCBP may be used as a bait to screen for any prey / binding partner. WO0071743A1 describes a mammalian two-hybrid system for detecting an interaction between a first protein and a second protein in a mammalian cell, which comprises, in a mammalian cell having a DNA carrying a reporter gene ligated thereto in the downstream of a base sequence binding to a DNA-binding region, expressing a fused protein of the first protein with two or more transcriptional activation regions which are the same or different, and another fused protein of the second protein with the above-described DNA-binding region, and then detecting the expression of the reporter gene. U.S. Pat. No. 6,251,676 and US20010024794A1 also describe a mammalian two-hybrid system. WO0126022A1 describes an in silico two-hybrid system, which is a process for the determination of interacting biomolecules, wherein a) a first group is provided comprising sequences representing homologous biomolecules, b) at least one second group is provided comprising sequences representing homologous biomolecules, c) group correlation values between the sequences of the first group and the sequences of at least one second group are determined, and d) the probability of the interaction of the sequence represented biomolecules is determined on the basis of the group correlation values. EP0963376A1 and U.S. Pat. No. 6,051,381 describe a prokaryotic two- hybrid system that can detect homo- and heterodimeric protein interactions in E. coli and other cells. This system is useful for the same applications as a yeast two-hybrid system, i.e. interaction cloning, mapping protein interaction domains, analysing protein interactions, detecting protein interactions and detecting modulators thereof. The invention concerns a prokaryotic host cell comprising: a) a fusion protein having (i) a first DNA-binding domain and (ii) a first interacting domain; b) a fusion protein having (i) a second DNA-binding domain and (ii) a second interacting domain capable of binding to the first interacting domain; and c) a nucleic acid molecule having a reporter gene operatively linked to (i) a promoter, (ii) a first operator site capable of binding to the first DNA-binding domain, located upstream of the promoter, and (iii) a second operator site capable of binding the second DNA- binding domain, located downstream of the promoter of the reporter gene; wherein binding of the first interacting domain to the second interacting domain is signaled by altered expression of the reporter gene. Other bacterial two-hybrid system is described in, for example, WOO 173108 A2, US20020045237A1. WO9526400A1 describes a reverse two-hybrid assay to isolate agents that bind a protein bait. Most existing two-hybrid systems involve reconstitution in yeast of a transcriptional activator that drives expression of a "reporter" gene such as HIS3 or lacZ. Attempts to utilize these existing systems for drug discovery would necessarily involve screening for molecules that interfere with the transcriptional read-out, and would be subject to detecting any compound that non-specifically interfered with transcription, hi addition, since currently used reporter genes encode long-lived proteins, the assay would have to be performed over a lengthy time period to allow for decay of the preexisting reporter proteins. Any compound that would be toxic to yeast over this time period would also score as a "hit". The reverse two-hybrid interaction will avoid both of these pitfalls by driving the expression of a relay gene, such as the GAL80 gene, which encodes a protein that binds to and masks the activation domain of a transcriptional activator, such as Gal4. The reporter genes, which will provide the transcriptional read-out (HIS 3 or lasZ), are dependent upon functional Gal4 for expression. Only when the level of Gal80 masking protein is reduced by interfering with the two-hybrid interaction will Gal4 function as a transcriptional activator, providing a positive transcriptional read-out for molecules that inhibit the two-hybrid protein-protein interaction. An important feature of the reverse two-hybrid system is that the basal level and half-life of the relay protein, Gal80, can be fine-tuned to provide maximum sensitivity. The reverse two-hybrid system described in WO9632503A1 is suitable for identifying various molecular interactions (e.g., protein protein, protein/DNA, protein/RNA, or RNA RNA interactions). Similar methods are also disclosed in U.S. Pat. Nos. 5,965,368, 5,955,280, 5,948,620, all incorporated herein by reference. Phage display is another class of methods that can be used to isolate binding partners for a given protein, such as MCBP. Briefly, the library of candidate agents (e.g. proteins) may be expressed as a library of fusions with the coat protein of a phage (preferably a filamentous phage) or a replicable genetic particle. Such library of candidates can then be passed through a column of MCBP, or any configurations of immobilized MCBP, such that phage particles expressing an agent that can bind the MCBP are enriched. This process can be repeated several times until a desirably strong binding between a subset of phage / agents and MCBP are selected. Details and improvements of this method are described, for example, in EP0627488A1, US5702892, US5821047, WO9907842A1, WO0006717A2, WO0061790A1, WO0142454A2, WO0212513A3, etc. All incorporated herein by reference. Any other methods in the art that can detect, isolate, or purify agents that bind to a target MCBP are all within the scope of the invention. The candidate agents used in the invention may be pharmacologic agents already known in the art or may be agents previously unknown to have any pharmacological activity. The agents may be naturally arising or designed in the laboratory. They may be isolated from microorganisms, animals, or plants, or may be produced recombinantly, or synthesized by chemical methods known in the art. They may be small molecules, nucleic acids, proteins, peptides or peptidomimetics. In certain embodiments, candidate agents are small organic compounds having a molecular weight of more than 50 and less than about 5,000 Da, or less than about 2,500 daltons. Candidate agents comprise functional groups necessary for structural interaction with proteins, particularly hydrogen bonding, and 5 typically include at least an amine, carbonyl, hydroxyl or carboxyl group, preferably at least two of the functional chemical groups. The candidate agents often comprise cyclical carbon or heterocyclic structures and/or aromatic or polyaromatic structures substituted with one or more of the above functional groups. Candidate agents are also found among biomolecules including, but not limited to: peptides, saccharides,0 fatty acids, steroids, purines, pyrimidines, derivatives, structural analogs or combinations thereof. Candidate agents are obtained from a wide variety of sources including libraries of synthetic or natural compounds. For example, numerous means are available for random and directed synthesis of a wide variety of organic compounds5 and biomolecules, including expression of randomized ohgonucleotides and oligopeptides. Alternatively, libraries of natural compounds in the form of bacterial, fungal, plant and animal extracts are available or readily produced. Additionally, natural or synthetically produced libraries and compounds are readily modified through conventional chemical, physical and biochemical means, and may be used0 to produce combinatorial libraries, hi certain embodiments, the candidate agents can ) be obtained using any of the numerous approaches in combinatorial library methods known in the art, including: biological libraries; spatially addressable parallel solid phase or solution phase libraries; synthetic library methods requiring deconvolution; the "one-bead one-compound" library method; and synthetic library methods using5 affinity chromatography selection. (Lam, K. S. (1997) Anticancer Drug Des. 12:145). For example, many compound collections are available from public and commercial sources, including Chembridge Corporation (San Diego, CA; Diverset E); the NCI Diversity set, and the NCI open collection. The identity of any0 compounds of interest, if not already known, can be characterized by, for example, LC-MS. Examples of methods for the synthesis of molecular libraries can be found in the art, for example in: DeWitt et al. (1993) Proc. Natl. Acad. Sci. U.S.A. 90:6909; Erb et al. (1994) Proc. Natl. Acad. Sci. USA 91:11422; Zuckermann et al. (1994). J. Med. Chem. 37:2678; Cho et al. (1993) Science 261:1303; Carrell et al. (1994) Angew. Chem. hit. Ed. Engl. 33:2059; Carell et al. (1994) Angew. Chem. Int. Ed. Engl. 33:2061; Kuruvilla, et al., 2002. Nature, 416: 653-657; and in Gallop et al. (1994) J. Med. Chem. 37:1233.

Examples Recruitment of ubiquitinated proteins to the 26S proteasome lies at the heart of the ubiquitin-proteasome system (UPS). Genetic studies suggest a role for the multiubiquitin chain binding proteins (MCBPs) Rad23 and RpnlO in recruitment, but biochemical studies implicate the Rpt5 ATPase. We addressed this issue by analyzing degradation of the ubiquitinated Cdk inhibitor Sicl (UbSicl) in vitro. Mutant rpnlOΔ and rad23Δ proteasomes failed to bind or degrade UbSicl. Although RpnlO or Rad23 restored UbSicl recruitment to either mutant, rescue of degradation by Rad23 uncovered a requirement for the VWA domain of RpnlO. In vivo analyses confirmed that Rad23 and the multiubiquitin binding domain of RpnlO contribute to Sicl degradation. Turnover studies of multiple UPS substrates uncovered an unexpected degree of specificity in their requirements for MCBPs. Thus the data is consistent with the theory that recruitment of substrates to the proteasome by MCBPs provides an additional layer of substrate selectivity in the UPS.

Example I. Isolation of Intact 26S Proteasomes from rpnlOΔ and rad23Δ Mutants To address the molecular basis for substrate recruitment by the 26S proteasome, we employed a system that recapitulates the selective ubiquitination and degradation of budding yeast S-Cdk inhibitor Sicl using purified components (Verma et al, Mol. Cell 8: 439-448, 2001, entire contents incorporated herein by reference). The chromosomal locus that encodes PRE1, a subunit of the 20S core, was tagged with the FLAG epitope in wild-type, rpnlOΔ, and rad23Δ mutant cells. 26S proteasomes were purified by single-step affinity chromatography on anti- FLAG beads as described (Verma et al, Mol. Biol. Cell 11: 3425-3439, 2000, entire contents incorporated herein by reference.). The data in Figure 1 A demonstrate that subunit composition, as visualized by SDS-PAGE, was essentially the same for 26S proteasomes purified from wild-type and mutant cells. This result was corroborated by MudPIT mass spec analysis (Link et al, Nat. Biotechnol 17: 676-682, 1999). Assembly was also normal as determined by Coomassie blue staining (Figure IB) and in-gel peptidase assay of purified proteasomes separated on native gels (Figure 1C). Some decrease in the doubly capped particle (R2C) with concomitant increase in 20S was seen for the mutants, particularly rad23Δ.

Example II. rpnlOΔ and rad23Δ 26S Proteasomes Are Defective at Degrading Ubiquitinated Sicl The protein degradation activity of the wild-type and mutant 26S proteasomes was assessed by incubation with a ubiquitinated maltose binding protein-Sicl chimera (UbMbpSicl), which was prepared as described (Seol et al, Genes Dev. 13: 1614-1626, 1999, entire content incorporated herein by reference). Degradation was monitored by loss of high molecular weight Sicl, which typically migrates at the top of a 7.5% gel and is also observed in the stacker (Verma et al, Mol. Biol. Cell 11: 3425-3439, 2000; and Verma et al, Mol. Cell 8: 439-448, 2001). Whereas wild-type 26S proteasomes degraded UbMbpSicl rapidly, rpnlOΔ 26S proteasomes were completely defective (compare lanes 2 and 5 with lane 1, Figure ID), and rad23Δ proteasomes were largely but not completely defective (Figure IE). The strength of these defects was surprising given the reported mild phenotype of rpnlOΔ mutants (Fu et al, J. Biol. Chem. 273: 1970-1981, 1998; and van Nocker et al, Mol. Cell. Biol. 16: 6020-6028, 1996). To confimi these unexpected results by a different method, we also evaluated whether rpnlOΔ and rad23Δ proteasomes were deficient in Rpnl 1 -dependent substrate deubiquitination (DUB) activity (Verma et al, Science 298: 611-615, 2002; and Yao and Cohen, Nαtwre 419: 403-407, 2002). A block in Rpnl 1 DUB activity leads to a block in degradation. Rpnl 1 activity is assayed in the presence of the 20S core protease inhibitor epoxomicin, which results in conversion of ubiquitinated substrate to an unmodified protein (MbpSicl ; lane 4, > Figure ID) (Verma et al, Science 298: 611-615, 2002). While not wishing to be bound by any theory, it was presumed that, concomitant with its deubiquitination by Rpnl 1, MbpSicl was translocated into the lumen of the 20S core but was not degraded due to the presence of epoxomicin. This is supported by the observation that MbpSicl formed upon incubation with proteasomes in vitro - but not naive MbpSic - was specifically coprecipitated with 20S subunits. As was observed in the degradation assay, rpnlOΔ proteasomes were completely deficient in deubiquitination of MbpSicl (Figure ID, lanes 3 and 4), whereas rad23Δ proteasomes were largely but not completely defective (Figure IF). Because it is easier to visualize the accumulation of deubiquitinated Sicl as opposed to the disappearance of ubiquitinated Sicl to evaluate proteasome function, the DUB assay was sometimes used in lieu of the degradation assay in subsequent experiments.

Example III. Restoration of Activity by Recombinant RpnlO and Rad23 Although rpnlOΔ and rad23Δ proteasomes appeared to be fairly normal by multiple physical and functional criteria (Figure 1), it remained possible that they were indirectly and/or irreversibly compromised by the absence of either of these proteins. To address this possibility, we performed add-back experiments using recombinant GST-RpnlO and GST-Rad23 purified fromE. coli. Strikingly, deubiquitination (Figure 2B) and degradation (Figure 2A) activities comparable to wild-type levels were obtained upon adding back GST-RpnlO to rpnlOΔ proteasomes. The effect of GST-RpnlO was exquisitely dosage sensitive. Very low levels (30-60 nM) were sufficient to rescue rpnlOΔ proteasomes but had little effect on wild-type proteasomes. However, at a concentration (120 nM) just -1.5- to 2-fold in molar excess over wild-type proteasomes, inhibition was observed, and at -3- to 4-fold molar excess (300 nM), inhibition was complete. Essentially the same effect was seen if GST-RpnlO was cleaved with thrombin to remove GST (data not shown). The ability of GST-RpnlO to rescue rpnlOΔ proteasomes allowed us to map the domains of RpnlO required for complementation. Mutational analysis of RPN10 in prior studies has demonstrated that the N-terminal domain of RpnlO (also called the von Willebrand A or VWA domain) (Whittaker and Hynes, Mol. Biol. Cell 13: 3369-3387, 2002) is required for conferring resistance to amino acid analogs and Ub-Pro-β-gal degradation (Fu et al, J. Biol. Chem. 273: 1970-1981, 1998). The C terminus contains the conserved LAMALRL multiUb chain recognition motif that constitutes part of the UIM domain and that is also required for binding UbMbpSicl. No phenotype has ever been linked to this domain, even though it constitutes the multiUb chain recognition domain of RpnlO. As shown in Figure 2D, either point mutation (first five amino acids of the recognition motif mutated; GST-N5rpnl0) or deletion of the UIM domain (GST-VWARpnlO or UIM^") destroyed RpnlO activity, underscoring the requirement for the UIM domain of RpnlO for UbMbpSicl degradation. To our knowledge, this is the first functional assay in which a direct requirement for the UIM has been demonstrated. We next investigated the ability of recombinant Rad23 to complement the partial defect in DUB activity observed with rad23Δ 26S proteasomes. The results in Figure 2C demonstrate that bacterially expressed GST-Rad23 was functional and rescued the DUB defect. As observed for RpnlO, optimal rescue by GST-Rad23 was highly concentration dependent. Efficient restoration of activity was observed at 40 nM, but high concentrations of GST-Rad23 actually inhibited the basal activity of rad23Δ proteasomes. A recent study using wild-type 26S proteasomes supplemented with a 3 -fold molar excess of Rad23 concluded that Rad23 has an inhibitory function in proteolysis (Raasi and Pickart, J. Biol. Chem. 278: 8951-8959, 2003). Likewise, previous reports documented an inhibitory role for RpnlO in vitro (Deveraux et al, J. Biol. Chem. 270: 29660-29663, 1995). However, our observations indicate that both Rad23 and RpnlO actually promote protein degradation by the proteasome - at least when the substrate is UbSicl - but that for both proteins it is advantageous to use mutant proteasome preparations to identify the optimal dose, preferably through titration, because these proteins inhibit degradation even when present in only modest stoichiometric excess over the 26S proteasome. One simple explanation of why Rad23 present in rpnlOΔ proteasomes and RpnlO present in rad23Δ proteasomes did not provide sufficient activity to sustain normal rates of UbMbpSicl turnover is that Rad23 is normally present at only substoichiometric levels in 26S proteasome preparations, such that there was not enough to sustain UbMbpSicl turnover in the absence of RpnlO. This contention is consistent with SDS-PAGE/microsequence analysis of purified yeast proteasomes (Glickman et al, Mol. Cell. Biol. 18: 3149-3162, 1998), immunoblot analysis of purified mammalian proteasomes (Raasi and Pickart, J. Biol. Chem. 278: 8951-8959, 2003), and the very low sequence coverage observed for Rad23 in our MudPIT experiments. Likewise, immunoblotting experiments revealed that RpnlO was present in rad23Δ proteasomes at one-third to one-half the levels observed in wild- type 26S proteasomes (data not shown). Significantly, addition of just 30 nM RpnlO rescued the defective DUB activity of rad23Δ 26S proteasomes (Figure 2C), arguing that RpnlO and Rad23 can act redundantly to sustain UbMbpSicl deubiquitination and turnover, and the action of Rpnl 0 was not dependent upon Rad23.

Example IV. Redundant Roles for Rad23 and the UIM Domain of RpnlO in Sustaining UbSicl Degradation Cross-rescue of rad23Δ 26S proteasomes by RpnlO encouraged us to investigate if the reverse was true, i.e., could addition of Rad23 restore activity to rpnlOΔ 26S proteasomes? Surprisingly, although recombinant GST-Rad23 was fully functional in restoring activity to rad23Δ 26S proteasomes (Figure 2C), it rescued rpnlOΔ 26S proteasomes weakly (Figure 2D). Because the requirement for RpnlO function for in vivo turnover of the synthetic reporter substrate Ub-Pro-β-gal mapped to the N-terminal VWA domain of RpnlO (Fu et al, J. Biol. Chem. 273: 1970-1981, 1998), we wondered whether Rad23 would rescue rpnlOΔ proteasomes in the presence of the VWA domain of RpnlO. Remarkably, although GST-VWARpnlO (UIM domain deleted) and GST-N5rpnl0 (mutant UIM) by themselves were inactive, the combination of either protein with GSTRad23 restored full activity to rpnlOΔ proteasomes (Figure 2D). Taken together, these observations support two important conclusions about the functions of RpnlO and Rad23. First, the Ub binding domains of RpnlO and Rad23 do not need to act sequentially. Instead, there exists a functional redundancy between Rad23 (see below) and the RpnlO UIM domain, suggesting that they function in parallel pathways to sustain degradation of Sicl . Second, the VWA domain of RpnlO was required for Rad23 to promote optimal rates of UbSicl proteolysis. This was also observed with Dsk2, another UbL-UBA domain protein like Rad23 (Funakoshi et al, Proc. Natl. Acad. Sci. USA 99: 745-750, 2002). Although rescue was weak, there was clearly an enhancement in activity when the RpnlO VWA domain and Dsk2 were added together ( Figure 2D, lanes 11 and 14). It could be that Dsk2 is less potent than Rad23 because it has only one UBA domain, and Rad23 has two. Indeed, Dsk2 bound less UbMbpSicl than Rad23 (data not shown). Since RpnlO functions to enhance the weak complementation by Rad23 (and Dsk2), we propose the term "facilitator" for RpnlO.

Example V. Both the UBA and the UbL Regions of Rad23 Are Required for Function Rescue of rad23 Δ 26S proteasomes by recombinant Rad23 allowed us to assess the relative contributions of both its Ub chain binding (UBA) and proteasome binding (UbL) regions. Consistent with prior studies (Schauber et al, Nature 391: 715-718, 1998; and Wilkinson et al, Nat. Cell Biol. 3: 939-943, 2001), a mutant protein (not shown) lacking the UbL but containing both UBA domains bound UbMbpSicl (Figure 3 A), whereas the reciprocal construct that contains the UbL domain but lacks both UBA domains selectively bound 26S proteasomes (data not shown). However, neither the UbL nor UBA segments sustained robust rescue of rad23Δ (Figure 3B) or rpnlOΔ (Figure 3C) 26S proteasomes.

Example VI. Rad23 and the UIM Domain of Rpnl 0 Link UbSicl to the Proteasome The ability of the UBA domain of Rad23 and the UIM domain of RpnlO to bind multiUb chains (Figures 3 A) suggested that the redundant function provided by these elements is to target UbSicl to the proteasome for degradation. To address this hypothesis, the substrate binding capacities of wild-type and rpnlOΔ 26S proteasomes were investigated by incubating UbMbpSicl (in the presence of inhibitors of deubiquitination and degradation) with 26S proteasomes immobilized on anti-Flag-beads (Figure 4A). Wild-type 26S proteasomes bound UbMbpSicl whereas rpnlOΔ 26S proteasomes displayed little or no binding activity. GST-RpnlO efficiently rescued the substrate binding defect of rpnl 0Δ proteasomes (Figure 4), but GST-VWARpnlO and GST-N5rpnl0 did not (Figure 4B), underscoring that this recruitment activity required the UEVI domain. GST-Rad23 bound rpnlOΔ proteasomes in a UbL-dependent manner (data not shown) and endowed them with enhanced substrate binding activity (Figure 4).

Example VII. RpnlO VWA Domain Facilitates the Degradation- Promoting Activity of Rad23 Surprisingly, although the VWA domain of RpnlO was required for optimal proteolysis-promoting activity of Rad23 (Figure 2D), it was not required for Rad23- dependent tethering of UbMbpSicl to the proteasome (Figure 4). Thus, binding is not a reliable surrogate assay for degradation. We conclude that the VWA domain acts downstream of Rad23 and enables proteasome-bound, ubiqutinated substrate to engage productively with the degradation machinery. Owing to its additional facilitator function encoded within the VWA domain, we suggest that the term facilitator be applied to RpnlO to distinguish it from substrate receptors such as Rad23. A widespread role for Rpnl 0 as a substrate receptor facilitator is suggested by the findings that deletion of RPN10 in Drosophila results in pupal lethality (Szianka et al, J. Cell Sci. 116: 1023-1033, 2003), and its down-regulation by RNAi causes G2/M phase arrest in Trypanosoma brucei (Li and Wang, J. Biol. Chem. 277: 42686-42693, 2002). Given that yeast rpnlOΔ mutants are viable, we surmise that either Rad23, Dsk2, or other substrate receptors retain sufficient function to sustain life (note the weak albeit detectable activity of Rad23 in the absence of RpnlOVWA; Figure 2D, lane 10), or other proteins provide a facilitator function in vivo that is redundant with that of RpnlO's VWA domain.

Example VIII. Both RPN10 and RAD23 Contribute to Sicl Turnover In vivo The in vitro assays indicate important roles for RpnlO and Rad23 in Sicl turnover. To date, all studies on these mutants in vivo have relied either on artificial substrates (vanNocker et al, Mol. Cell. Biol. 16: 6020-6028, 1996); indirect readouts for degradation, such as steady state analysis (Wilkinson et al, Nat. Cell Biol. 3: 939-943, 2001); or a substrate (Clb2) whose degradation is subject to indirect regulation via cell cycle checkpoints (Lambertson et al, Genetics 153: 69- 79, 1999). Thus, to monitor Sicl degradation in vivo, we evaluated turnover during the appropriate cell cycle phase. Wild-type and mutant cells were arrested in Gl with factor and then released synchronously into the cell cycle (Figure 5). Both GAL1 -expressed and endogenous Sicl are normally degraded at the Gl/S boundary (Verma et al, Science 278: 455-460, 1997). As shown in Figure 5, both GAL1- expressed and endogenous Sicl tapered off by 45 min as cells entered S phase. Based on our reconstitution experiments, we reasoned that Sicl might be targeted for degradation in vivo by either Rad23 or the UEM domain of RpnlO. Indeed, whereas Sicl was degraded with normal kinetics in rad23Δ and in a mutant lacking the UIM domain of RpnlO (rpnl0VWA+), significant stabilization was observed in an rpnl0VWA+ rad23Δ double mutant. As expected from the facilitator role played by the VWA domain in the operation of other receptor pathways in vitro, Sicl was significantly more stable in rpnlOΔ than in rpnl0VWA+ cells. Additionally, failure to promptly degrade Sicl correlated with a reduced rate of entry into S phase, as shown for the rpnlOΔ rad23Δ mutant (Figure 5), which remained in Gl phase 75 min after release from factor. Degradation of Sicl is essential for entry into S phase (Verma et al, Science 278: 455-460, 1997). Delayed entry into S phase and residual turnover of Sicl in rpnlOΔ rad23Δ cells indicate that there must exist a third receptor pathway (possibly Dsk2, Figure 2D) by which Sicl can engage the proteasome and be degraded, albeit at a greatly reduced rate. Since the rpnlOΔrad23Δ double mutant displayed unexpectedly strong stabilization of Sicl, the growth phenotype of this mutant was reassessed. It has been reported that these mutants are cold sensitive at 13°C (Lambertson et al, Genetics 153: 69-79, 1999). However, we observed a severe growth defect even at 25°C (data not shown), which was exacerbated in synthetic medium. Consistent with the in vitro and in vivo data presented here and elsewhere (Fu et al, J. Biol. Chem. 273: 1970-1981, 1998), the slow growth phenotypes of the double mutant were linked to the absence of the VWA domain of RPN10 (data not shown). Example IX. Specificity in the Requirement for Different MCBPs for In vivo Turnover of UPS Substrates To address the generality of our observations, we next tested whether the relative contributions of Rad23 and RpnlO to Sicl degradation would hold true for another physiological substrate of the UPS — the Gl cyclin Cln2 (Deshaies et al, EMBO J. 14: 303-312, 1995). HA-tagged Cln2 expressed from the GAL1 promoter was rapidly degraded in Gl phase cells and unlike Sicl was not stabilized in rpnlOΔ, rad23Δ, or rpnlOΔrad23Δ mutants. This prompted us to look at its turnover in additional MCBP mutants. As shown by the data in Figure 6A, mutations in the genes encoding the UBA domain-containing putative targeting factors Ddil, Dsk2 (Saeki et al, Biochem. Biophys. Res. Commun. 293: 986-992a, 2002), and the UT3 domain-containing Ufdl (Ye et al, J. Cell Biol. 162: 71-84, 2003) had no effect on Chι2 turnover. From this analysis, we conclude that an as yet unknown receptor or set of receptors, possibly including Rpt5, functions to link ubiquitinated Cln2 to the proteasome. Whereas Sicl is a substrate of the E3 Ub ligase SCF^Cdc4 (Seol et al, Genes Dev. 13: 1614-1626, 1999), Cln2 is an SCF⁰"^"1 substrate (Seol et al, Genes Dev. 13: 1614-1626, 1999; and Skowyra et al, Science 284: 662-665, 1999). To determine if the identity of the ubiquitin ligase influenced the different receptor dependencies exhibited by Sicl and Cln2, we examined the turnover of the SCF^Cdc substrate Farl (Henchoz et al, Genes Dev. 11: 3046-3060, 1997) and the SCF^Glτl substrate Gic2 (Jaquenoud et al, EMBO J. 17: 5360-5373, 1998). Farl is a Gl cyclin-Cdk inhibitor, and Gic2 is an effector of the Cdc42 cell polarity regulator. In both cases, turnover of the endogenous protein was examined during Gl phase, when Farl and Gic2 are normally degraded (Jaquenoud et al, EMBO J. 17: 5360-5373, 1998). In contrast to Sicl, Farl degradation was impeded more in rad23Δ than in rpnlOΔ mutants (Figure 6B). Meanwhile, Gic2 mimicked Sicl and not Cln2 in that it was strongly stabilized in rpnlOΔ cells (data not shown). Additionally, Clb2, an APC substrate (Harper et al, Genes Dev. 16: 2179-2206, 2002), also mimicked Sicl (Supplemental Figure S5). In addition to proteolysis of regulatory proteins, the UPS is also required for the degradation of misfolded proteins. Secretory pathway proteins that fail to fold properly in the ER are retrotranslocated into the cytosol and degraded by the 26S proteasome in a process called ER-associated degradation (ERAD) (Tsai et al, Nat. Rev. Mol. Cell Biol. 3: 246-255, 2002). The Cdc48/Ufdl/Npl4 complex is required for ERAD and recognizes membrane-associated Ub conjugates via the UT3 domains of Ufdl/Cdc48 (Ye et al, J. Cell Biol. 162: 71-84, 2003). The ERAD substrate CPY* is stabilized in mutants defective in individual subunits of the Cdc48/Ufdl/Npl4 complex (Jarosch et al, Nat. Cell Biol. 4: 134-139, 2002) (Figure 6C). To determine if ERAD substrates are "handed off to proteasomal receptors following their extraction from the membrane by Cdc48/Ufdl/Npl4 (Flierman et al, J. Biol. Chem. 278 : 34774-34782, 2003), we evaluated the turnover of CPY* in rpnlOΔ and rad23Δ mutants. Surprisingly, no stabilization was observed (Figure 6C). These data suggest that Cdc48/Ufdl/Npl4 may shepherd the extracted CPY* directly to the proteasome or deliver it to Rpt5 or an as yet unknown receptor. The Cdc48/Ufdl complex binds specifically to K48-linked polyUb chains via the UT3 domain (Ye et al, J. Cell Biol. 162: 71-84, 2003) and also participates in degradation of non-ERAD substrates such as cytosolic UbV⁷⁶-V-β-galactosidase (Johnson et al, J. Biol. Chem. 270: 17442-17456, 1995) and spindle disassembly factors Cdc5 and Asel (Cao et al, Cell 115: 355-367, 2003). We monitored the turnover of the cytoplasmic Degl-Gfp, which contains the degradation signal from the transcriptional repressor MATc.2. This fusion substrate is interesting because, although it is soluble, it is ubiquitinated by enzymes resident in the ER membrane (Swanson et al, Genes Dev. 15: 2660-2674, 2001). As shown in Figure 6D, Degl- Gfp was stabilized in ufdl-1. However, like the ERAD substrate CPY*, Degl-Gfp was not stabilized in rpnlOΔ mutants.

From the data presented above, an important principle emerges from considering the targeting requirements observed for physiological versus synthetic substrates. Reporter substrates such as Ub-Pro-β-gal, Ub -V-β-gal, and Ub -V- DHFR exhibit simultaneous dependence on multiple putative receptor pathways, including RpnlO, Rad23, and Cdc48/Ufdl (Johnson et al, J. Biol. Chem. 270:

17442-17456, 1995; Rao and Sastry, J. Biol. Chem. 211: 11691-11695, 2002; and Xie and Varshavsky, Nat. Cell Biol. 4: 1003-1007, 2002). This simultaneous dependence suggests that these factors typically serve nonredundant, possibly even sequential (Chen and Madura, Mol. Cell. Biol. 22: 4902-4913, 2002) roles in degradation. By contrast, none of the physiological substrates examined in this study (including Farl, Sicl, Gic2, Cln2, CPY*, and Clb2) exhibited an equivalently broad dependence on multiple putative receptor pathways. Thus, although synthetic substrates have proved very useful for defining components of the UPS system, we caution that their turnover may not be reflective of typical physiologic mechanisms, and, thus, general conclusions about the mechanism/specificity of the UPS should preferably be rooted in the study of physiological substrates. It was commonly thought that specificity in substrate turnover by the UPS lies at the level of ubiquitin chain assembly controlled by E2, E3, and isopeptidase enzymes. Our findings, however, lead to the unexpected conclusion that proteasome- targeting pathways downstream of the ubiquitin ligases exhibit a surprising degree of substrate specificity. A scheme that graphically summarizes the key points is illustrated in Figure 7. In Figure 7, RpnlO, Rad23, Dsk2, and possibly Ufdl/Cdc48 and Rpt5 comprise distinct receptor pathways that link ubiquitinated substrates to the proteasome. Although there are no actual functional data indicating that either Ufdl/Cdc48 or Rpt5 recruits ubiquitinated substrates to the proteasome, others have suggested a receptor function for Rpt5 based on cross-linking data (Lam et al, Nature 416: 763-767, 2002), and we provide a receptor activity for Ufdl/Cdc48 in light of data reported here and elsewhere (Flierman et al, J. Biol. Chem. 278: 34774-34782, 2003; and Ye et al, J. Cell Biol. 162: 71-84, 2003). Some substrates, like Sicl and Clb2, are recruited to the proteasome and degraded in a manner that depends strongly on the MCBP receptor and/or facilitator (FA) functions of the proteasome subunit RpnlO, whereas others, such as Farl, show a weaker dependence on RpnlO and a correspondingly stronger dependence on Rad23. Yet other substrates such as CPY* and Degl-Gfp appear to bypass RpnlO entirely but depend on a complex containing Ufdl and Cdc48. (It has been reported that Farl degradation also depends upon Cdc48 using a novel Gl -specific td allele (Fu et al, J. Cell Biol. 163: 21-26, 2003), but we have not observed a defect in Farl turnover in cdc48-3 or ufdl-1 mutants; data not shown). Finally, at least one substrate, Cln2, does not depend upon any known receptor pathway. However, our data on Sicl underscore that it is important to distinguish "dependency" from "involvement." Rad23 can be involved in Sicl turnover (as evidenced by the fact that Sicl was unstable in rpnlO^VWΛ but was stabilized in rpnlO^VWΛ rad23Δ), even though Sicl turnover does not normally depend upon Rad23 (as evidenced by rapid Sicl turnover in a rad23Δ mutant). Thus, Cln2 may not depend upon the known receptors, because it can be targeted by multiple receptors in a highly redundant manner, or because it arrives at the proteasome by a distinct route involving Rpt5 or an unknown receptor. Yet other targeting strategies are likely to exist, given that ubiquitin ligases such as Parkin, Ufd4, and Hul5 can bind directly to the proteasome (Demand et al, Curr. Biol. 11: 1569-1577, 2001; Sakata et al, EMBO Rep. 4: 301- 306, 2003; Xie and Varshavsky, Nat. Cell Biol. 4: 1003-1007, 2002; and Leggett et al, Mol. Cell 10: 495-507, 2002). Our data indicate that a putative receptor activity intrinsic to Rpt5 (Lam et al, Nature 416: 763-767, 2002) by itself is insufficient to target UbSicl for degradation in a defined in vitro system. Moreover, our in vivo analysis implies that an Rpt5-mediated targeting mechanism would appear to be insufficient to sustain normal rates of degradation in vivo for seven of eight UPS substrates characterized in this study. While not wishing to be bound by particular theory, it is possible that Rpt5 serves as the primary conduit by which a subset of unstable proteins poorly represented in this study (but possibly including Cln2) gains access to the proteasome. On the other hand, it is also possible that Rpt5 serves as a central conduit that gathers together substrates delivered by different receptor pathways (RpnlO, Rad23, and Cdc48/Ufdl) and positions them for subsequent unfolding, deubiquitination, and translocation. This latter possibility calls to mind translocation of secretory precursors cross the ER membrane, where the primary signal peptide- mediated targeting step is carried out by upstream receptors such as Signal Recognition Particle, following which the signal peptide is transferred to the Sec61 channel to enable precursor translocation across the membrane. An analogous two- step recognition system may operate in E. coli, where the SspB protein functions as a specificity factor for the AAA ATPase ClpX, enhancing degradation of ssrA- tagged substrates (Levchenko et al, Science 289: 2354-2356, 2000). Amerik, A., Swaminathan, S., Krantz, B.A., Wilkinson, K.D. and Hochstrasser, M., 1997. In vivo disassembly of free polyubiquitin chains by yeast Ubpl4 modulates rates of protein degradation by the proteasome. EMBO J. 16, pp. 4826-4838. Cao, K., Nakajima, R, Meyer, H.H. and Zheng, Y., 2003. The AAA-ATPase Cdc48/p97 regulates spindle disassembly at the end of mitosis. Cell 115, pp. 355-367. Chau, V., Tobias, J.W., Bachmair, A., Marriott, D., Ecker, D.J., Gonda, D.K. and Varshavsky, A., 1989. A multiubiquitin chain is confined to specific lysine in a targeted short-lived protein. Science 243, pp. 1576-1583. Chen, L. and Madura, K., 2002. Rad23 promotes the targeting of proteolytic substrates to the proteasome. Mol. Cell. Biol. 22, pp. 4902-4913. Dai, R.M. and Li, C.C., 2001. Valosin-containing protein is a multi-ubiquitin chain- targeting factor required in ubiquitin-proteasome degradation. Nat. Cell Biol. 3, pp. 740-744. Demand, J., Alberti, S., Patterson, C. and Hohfeld, J., 2001. Cooperation of a ubiquitin domain protein and an E3 ubiquitin ligase during chaperone/proteasome coupling. Curr. Biol. 11, pp. 1569-1577.

Deshaies, R.J., Chau, V. and Kirschner, M., 1995. Ubiquitination of the Gl cyclin Cln2p by a Cdc34p-dependent pathway. EMBO J. 14, pp. 303-312. Deveraux, Q., Ustrell, V., Pickart, C. and Rechsteiner, M., 1994. A 26S protease subunit that binds ubiquitin conjugates. J. Biol. Chem. 269, pp. 7059-7061. Deveraux, Q., van Nocker, S., Maliaffey, D., Vierstra, R. and Rechsteiner, M., 1995. Inhibition of ubiquitin-mediated proteolysis by the Arabidopsis 26 S protease subunit S5a. J. Biol. Chem. 270, pp. 29660-29663. Elsasser, S., Gali, R.R., Schwickart, M., Larsen, C.N., Leggett, D.S., Muller, B., Feng, M.T., Tubing, F., Dittmar, G.A. and Finley, D., 2002. Proteasome subunit Rpnl binds ubiquitin-like protein domains. Nat. Cell Biol. 4, pp. 725-730.

Flierman, D., Ye, Y., Dai, M., Chau, V. and Rapoport, T.A., 2003. Polyubiquitin serves as a recognition signal, rather than a ratcheting molecule, during retro translocation of proteins across the endoplasmic reticulum membrane. J. Biol. Chem. 278, pp. 34774-34782. Flynn, J.M., Neher, S.B., Kim, Y.I., Sauer, R.T. and Baker, T.A., 2003. Proteomic discovery of cellular substrates of the ClpXP protease reveals five classes of ClpX-recognition signals. Mol. Cell 11, pp. 671-683.

Fu, H., Sadis, S., Rubin, D.M., Glickman, M., van Nocker, S., Finley, D. and Vierstra, R.D., 1998. Multiubiquitin chain binding and protein degradation are mediated by distinct domains within the 26 S proteasome subunit Mcbl . Fu, X., Ng, C, Feng, D. and Liang, C, 2003. Cdc48p is required for the cell cycle commitment point at Start via degradation of the Gl-CDK inhibitor Farlp. J. Cell Biol. 163, pp. 21-26. Funakoshi, M., Sasaki, T., Nishimoto, T. and Kobayashi, H., 2002. Budding yeast Dsk2p is a polyubiquitin-binding protein that can interact with the proteasome. Proc. Natl. Acad. Sci. USA 99, pp. 745-750. Glickman, M.H., Rubin, D.M., Fried, V.A. and Finley, D., 1998. The regulatory particle of the Saccharomyces cerevisiae proteasome. Mol. Cell. Biol. 18, pp. 3149-3162. Harper, J.W., Burton, J.L. and Solomon, M.J., 2002. The anaphase-promoting complex: it's not just for mitosis any more. Genes Dev. 16, pp. 2179-2206. Hartmann-Petersen, R, Hendil, K.B. and Gordon, C, 2003. Ubiquitin binding proteins protect ubiquitin conjugates from disassembly. FEBS Lett. 535, pp. 77— 81. Henchoz, S., Chi, Y., Catarin, B., Herskowitz, I., Deshaies, R.J. and Peter, M., 1997. Phosphorylation- and ubiquitin-dependent degradation of the cyclin-dependent kinase inhibitor Farlp in budding yeast. Genes Dev. 11, pp. 3046-3060.

Hershko, A. and Ciechanover, A., 1998. The ubiquitin system. Annu. Rev. Biochem. 67, pp. 425-479. Hofmann, K. and Bucher, P., 1996. The UBA domain: a sequence motif present in multiple enzyme classes of the ubiquitination pathway. Trends Biochem. Sci. 21, pp. 172-173.

Jaquenoud, M., Gulli, M.P., Peter, K. and Peter, M., 1998. The Cdc42ρ effector Gic2p is targeted for ubiquitin-dependent degradation by the SCFGrrl complex. EMBO J. 17, pp. 5360-5373.

Jarosch, E., Taxis, C, Volkwein, C, Bordallo, J., Finley, D., Wolf, D.H. and Sommer, T., 2002. Protein dislocation from the ER requires polyubiquitination and the AAA-ATPase Cdc48. Nat. Cell Biol. 4, pp. 134-139. Johnson, E.S., Ma, P.C, Ota, I.M. and Varshavsky, A., 1995. A proteolytic pathway that recognizes ubiquitin as a degradation signal. J. Biol. Chem. 270, pp. 17442- 17456.

Kleijnen, M.F., Shih, A.H., Zhou, P., Kumar, S., Soccio, R.E., Kedersha, N.L., Gill, G. and Howley, P.M., 2000. The hPLIC proteins may provide a link between the ubiquitination machinery and the proteasome. Mol. Cell 6, pp. 409-419.

Lam, Y.A., Lawson, T.G., Velayutham, M., Zweier, J.L. and Pickart, CM., 2002. A proteasomal ATPase subunit recognizes the polyubiquitin degradation signal. Nature 416, pp. 763-767.

Lambertson, D., Chen, L. and Madura, K., 1999. Pleiotropic defects caused by loss of the proteasome-interacting factors Rad23 and Rpnl 0 of Saccharomyces cerevisiae. Genetics 153, pp. 69-79.

Leggett, D.S., Hanna, J., Borodovsky, A., Crosas, B., Schmidt, M., Baker, R.T., Walz, T., Ploegh, H. and Finley, D., 2002. Multiple associated proteins regulate proteasome structure and function. Mol. Cell 10, pp. 495-507. Levchenko, I., Seidel, M., Sauer, R.T. and Baker, T.A., 2000. A specificity- enhancing factor for the ClpXP degradation machine. Science 289, pp. 2354- 2356.

Li, Z. and Wang, C.C., 2002. Functional characterization of the 11 non-ATPase subunit proteins in the trypanosome 19 S proteasomal regulatory complex. J. Biol. Chem. 277, pp. 42686-42693.

Link, A.J., Eng, J., Schieltz, D.M., Carmack, E., Mize, G.J., Morris, D.R., Garvik, B.M. and Yates III, J.R., 1999. Direct analysis of protein complexes using mass spectrometry. Nat. Biotechnol. 17, pp. 676-682.

Ortolan, T.G., Tongaonkar, P., Lambertson, D., Chen, L., Schauber, C. and Madura, K., 2000. The DNA repair protein rad23 is a negative regulator of multiubiquitin chain assembly. Nat. Cell Biol. 2, pp. 601-608.

Pickart, CM. and Cohen, R.E., 2004. Proteasomes and their kin: proteases in the machine age. Nat. Rev. Mol. Cell Biol. 5, pp. 177-187.

Raasi, S. and Pickart, CM., 2003. Rad23 ubiquitin-associated domains (UBA) inhibit 26 S proteasome-catalyzed proteolysis by sequestering lysine 48-linked polyubiquitin chains. J. Biol. Chem. 278, pp. 8951-8959. Rao, H. and Sastry, A., 2002. Recognition of specific ubiquitin conjugates is important for the proteolytic functions of the ubiquitin-associated domain proteins Dsk2 and Rad23. J. Biol. Chem. 277, pp. 11691-11695.

Saeki, Y., Saitoh, A., Toh-e, A. and Yokosawa, H., 2002. Ubiquitin-like proteins and RpnlO play cooperative roles in ubiquitin-dependent proteolysis. Biochem. Biophys. Res. Commun. 293, pp. 986-992 a .

Saeki, Y., Sone, T., Toh-e_;, A. and Yokosawa, H., 2002. Identification of ubiquitin- like protein-binding subunits of the 26S proteasome. Biochem. Biophys. Res. Commun. 296, pp. 813-819 b .

Sakata, E., Yamaguchi, Y., Kurimoto, E., Kikuchi, J., Yokoyama, S., Yamada, S., Kawahara, H., Yokosawa, H., Hattori, N., Mizuno, Y. et al, 2003. Parkin binds the RpnlO subunit of 26S proteasomes through its ubiquitin-like domain. EMBO Rep. 4, pp. 301-306.

Schauber, C, Chen, L., Tongaonkar, P., Vega, I., Lambertson, D., Potts, W. and Madura, K., 1998. Rad23 links DNA repair to the ubiquitin/proteasome pathway. Nature 391, pp. 715-718.

Seol, J., Feldman, R., Zachariae, W., Shevchenko, A., Correll, C, Lyapina, S., Chi, Y., Galova, M., Claypool, J., Sandmeyer, S. et al, 1999. Cdc53/cullin and the essential Hrtl RING-H2 subunit of SCF define a ubiquitin ligase module that activates the E2 enzyme Cdc34. Genes Dev. 13, pp. 1614-1626. Skowyra, D., Koepp, D.M., Kamura, T., Conrad, M.N., Conaway, R.C, Conaway, J.W., Elledge, S.J. and Harper, J.W., 1999. Reconstitution of Gl cyclin ubiquitination with complexes containing SCFGrrl and rbxl. Science 284, pp. 662-665.

Swanson, R., Locher, M. and Hochstrasser, M., 2001. A conserved ubiquitin ligase of the nuclear envelope/endoplasmic reticulum that fiinctions in both ER- associated and Matalpha2 repressor degradation. Genes Dev. 15, pp. 2660-2674.

Szlanka, T., Haracska, L., Kiss, I., Deak, P., Kurucz, E., Ando, L, Viragh, E. and Udvardy, A., 2003. Deletion of proteasomal subunit S5a/Rρnl0/p54 causes lethality, multiple mitotic defects and overexpression of proteasomal genes in Drosophila melanogaster. J. Cell Sci. 116, pp. 1023-1033.

Thrower, J., Hoffman, L., Rechsteiner, M. and Pickart, C, 2000. Recognition of the polyubiquitin proteolytic signal. EMBO J. 19, pp. 94-102.

Tsai, B., Ye, Y. and Rapoport, T.A., 2002. Retro-translocation of proteins from the endoplasmic reticulum into the cytosol. Nat. Rev. Mol. Cell Biol. 3, pp. 246- 255. van Nocker, S., Sadis, S., Rubin, D.M., Glickman, M., Fu, H., Coux, O., Wefes, I., Finley, D. and Vierstra, R.D., 1996. The multiubiquitin-chain-binding protein Mcbl is a component of the 26S proteasome in Saccharomyces cerevisiae and plays a nonessential, substrate-specific role in protein turnover. Mol. Cell. Biol. 16, pp. 6020-6028.

Verma, R., Annan, R.S., Huddleston, M.J., Carr, S.A., Reynard, G. and Deshaies, R.J., 1997. Phosphorylation of Siclp by Gl Cdk required for its degradation and entry into S phase. Science 278, pp. 455-460.

Veπna, R., Chen, S., Feldman, R., Schieltz, D., Yates, J., Dohmen, R.J. and Deshaies, R.J., 2000. Proteasomal proteomics: identification of nucleotide- sensitive proteasome-interacting proteins by mass spectrometric analysis of affinity-purified proteins. Mol. Biol. Cell 11, pp. 3425-3439.

Verma, R., McDonald, H., Yates III, J.R. and Deshaies, R.J., 2001. Selective degradation of ubiquitinated Sicl by purified 26S proteasome yields active S phase cyclin-Cdk. Mol. Cell 8, pp. 439-448.

Veπna, R., Aravind, L., Oania, R., McDonald, W.H., Yates III, J.R., Koonin, E.V. and Deshaies, R. J., 2002. Role of Rpnl 1 metalloprotease in deubiquitination and degradation by the 26S proteasome. Science 298, pp. 611-615.

Whittaker, CA. and Hynes, R.O., 2002. Distribution and evolution of von Willebrand/integrin A domains: widely dispersed domains with roles in cell adhesion and elsewhere. Mol. Biol. Cell 13, pp. 3369-3387. Wilkinson, C.R., Seeger, M., Hartmann-Petersen, R., Stone, M., Wallace, M., Semple, C and Gordon, C, 2001. Proteins containing the UBA domain are able to bind to multi-ubiquitin chains. Nat. Cell Biol. 3, pp. 939-943. Xie, Y. and Varshavsky, A., 2002. UFD4 lacking the proteasome-binding region catalyses ubiquitination but is impaired in proteolysis. Nat. Cell Biol. 4, pp. 1003-1007.

Yao, T. and Cohen, R.E., 2002. A cryptic protease couples deubiquitination and degradation by the proteasome. Nature 419, pp. 403-407.

Ye, Y., Meyer, H.H. and Rapoport, T.A., 2003. Function of the p97-Ufdl-Npl4 complex in retrotranslocation from the ER to the cytosol: dual recognition of nonubiquitinated polypeptide segments and polyubiquitin chains. J. Cell Biol. 162, pp. 71-84.

All publications and patents mentioned herein are hereby incorporated by reference in their entirety as if each individual publication or patent was specifically and individually indicated to be incorporated by reference. In case of conflict, the present application, including any definitions herein, will control.

Equivalency: While specific embodiments of the subject inventions are explicitly disclosed herein, the above specification is illustrative and not restrictive. Many variations of the inventions will become apparent to those skilled in the art upon review of this specification and the claims below. The full scope of the inventions should be determined by reference to the claims, along with their full scope of equivalents, and the specification, along with such variations.

Claims

Claims:

1. A method for monitoring the ability of at least one multiubiquitin chain binding protein (MCBPs) to promote degradation of at least one ubiquitin- conjugated target protein by a 26S proteasome, the method comprising determining the degree of degradation of said target proteins in a system comprising:

(1) a defective 26S proteasome preparation substantially unable to degrade a ubiquitin-conjugated target protein,

(2) one or more functional multiubiquitin chain binding protein(s) (MCBP(s)) in an amount that substantially restores the ubiquitin-mediated proteasome degradation of said target protein to a wild-type level, with the proviso that said system is not a wild-type 26S proteasome preparation.

2. The method of claim 1, wherein said system is an in vitro system.

3. The method of claim 1, wherein said system is an in vivo system.

4. The method of claim 1, wherein said MCBP(s) comprise one or more of: an RpnlO polypeptide, an Rad23 polypeptide, a Dsk2 polypeptide, a Cdc48/Ufdl/Npl4 complex, or an Rpt5 polypeptide.

5. The method of claim 1, wherein said MCBP(s) comprise one or more of: a Parkin polypeptide, a Ufd4 polypeptide, or an Hul5 polypeptide.

6. The method of claim 1 , wherein the presence of said amount of one or more functional MCBP(s) in said defective 26S proteasome preparation restores at least about 30%, 50%, 60%, 70%, 80%, 90%, or nearly 100%, of the wild-type level of ubiquitin-mediated proteasome degradation of said target protein.

7. The method of claim 1, wherein at least one of said MCBP(s) is from a species different from the species of said defective 26S proteasome preparation.

8. The method of claim 7, wherein said defective 26S proteasome preparation is from a non-human eukaryote.

9. The method of claim 8, wherein said eukaryote is yeast.

10. The method of claim 9, wherein said MCBP(s) comprise one or more human Rad23 homologs selected from hHR23 A and hHR23B, one or more human Dsk2 homologs selected from hPLIC-l and hPLIC-2, or human RpnlO homolog S5A.

11. The method of claim 7, wherein said defective 26S proteasome preparation is from human.

12. The method of claim 7 or 11, wherein said defective 26S proteasome preparation has a diminished level of one or more MCBP(s).

13. The method of claim 12, wherein said diminished level of MCBP(s) is effectuated by selective chemical extraction of said MCBP(s).

14. The method of claim 12, wherein said diminished level of MCBP(s) is effectuated by gene knock-out of said MCBP(s).

15. The method of claim 12, wherein said diminished level of MCBP(s) is effectuated by RNAi of said MCBP(s).

16. The method of claim 15, wherein said RNAi utilizes small inhibitor RNA (siRNA) or short hairpin RNA (shRNA).

17. The method of claim 12, wherein said diminished level of MCBP(s) is effectuated by immunodepletion or affinity depletion of said MCBP(s).

18. The method of claim 12, wherein said diminished level is no more than about 30%, 20%, 10%, 5%, or 1% of wild-type level.

19. The method of claim 1 , wherein the amount of one or more functional MCBP(s) in said system is an optimum amount determined by titration.

20. The method of claim 19, wherein the system is an in vitro system, and said titration is performed by providing a range of different concentrations of MCBP(s).

21. The method of claim 19, wherein the system is an in vivo system, and said titration is performed by operatively linking the coding sequence of said MCBP(s) with a range of promoters with different transcription / translation strength.

I

22. The method of claim 1 , wherein the amount of one or more functional

MCBP(s) in said system is no more than 2-fold, preferably no more than 1.5-fold molar excess over the 26S proteasomes.

23. The method of claim 1, wherein said target protein is one or more of: Sicl, Farl , Clb2, Gic2, CPY*, Cln2, or a homolog thereof.

24. The method of claim 1, wherein the system lacks one or more MCBP(s).

25. The method of claim 24, wherein the lacking MCBP(s) are different from said MCBP(s) in (2).

26. The method of claim 25, wherein said defective 26S proteasome preparation lacks RpnlO, and said MCBP(s) comprise:

(a) a Rad23 polypeptide or a Dsk2 polypeptide; and

(b) a mutant Rpnl 0 lacking the VWA domain, or a mutant Rpnl 0 lacking the UIM domain.

27. The method of claim 1, which is adapted for assaying a specific subset of target proteins.

28. The method of claim 1 , wherein said MCBP(s) are recombinantly produced.

29. The method of claim 1 , wherein the system further comprises a facilitator, such as the VWA domain of an RpnlO protein.

30. A system for assaying the activity of a ubiquitin-proteasome pathway, said system comprising:

(1) a defective 26S proteasome preparation substantially unable to degrade a ubiquitin-conjugated target protein,

(2) one or more functional multiubiquitin chain binding protein(s) (MCBP(s)) in an amount that substantially restores the ubiquitin-mediated proteasome degradation of said target protein to a wild-type level, with the proviso that said system is not a wild-type 26S proteasome preparation.

31. A method for screening for an agent that inhibits the degradation of a ubiquitinated target protein by a 26S proteasome, the method comprising: (a) incubating the ubiquitinated target protein in a system comprising: (i) a defective 26S proteasome preparation substantially unable to degrade a ubiquitin-conjugated target protein, and (ii) one or more functional multiubiquitin chain binding protein(s) (MCBP(s)) in an amount that substantially restores the ubiquitin- mediated proteasome degradation of said target protein to a wild-type level, with the proviso that said system is not a wild-type 26S proteasome preparation; (b) determining and comparing the degree of degradation of said target protein in the presence or absence of a test agent, wherein more complete degradation of said target protein in the absence of the test agent than in the presence of the test agent is indicative that the test agent inhibits the degradation of said target protein.

32. The method of claim 31 , further comprising determining the binding, if any, of the test agent to the 26S proteasome and/or the MCBP(s).

33. A method to screen for agents that can inhibit (or enhance) the function of a MCBP-mediated 26S proteasome activity, the method comprising:

(1) isolating, from among a plurality of candidate agents, one or more agent(s), if any, that can interact with the MCBP protein;

(2) determining the MCBP-mediated 26S proteasome activity in the presence or absence of said candidate agents identified in (1), wherein an inhibited (or enhanced) proteasome activity in the presence of the candidate agent(s) is indicative that the candidate agent(s) is/are enhancer (or inhibitor) of the function of the MCBP-mediated 26S proteasome activity.

34. The method of claim 33, wherein the MCBP protein is a Rad23 protein, a RpnlO protein, a Dsk2 protem, a Rpt2 protein, or a Cdc48/Ufdl protein complex.

35. The method of claim 33, wherein the candidate agents are from a library.

36. The method of claim 35, wherein the library is a polymicleotide library, a polypeptide library, a small chemical compound library, an organic compound library, an inorganic compound library, a library of chemicals synthesized by split- pool methods, or a library of compounds with unknown identity.

37. The method of claim 33, wherein step (1) is carried out by high throughput screening.

38. The method of claim 37, wherein the high throughput screening screens more than about 10, 50, 100, 400, 1000, 1500, 5000 or more compounds in parallel.

39. The method of claim 33, wherein step (1) is effectuated by direct in vitro binding assay between the MCBP and the candidate agents.

40. The method of claim 33, wherein step (1) is effectuated by a gel-shift binding assay.

41. The method of claim 33, wherein step (1) is effectuated by two-hybrid binding assay using the MCBP as a bait protein.

42. The method of claim 33, wherein the two-hybrid binding assay is performed in a bacterial, a yeast, or a mammalian system.

43. A method of screening for 26S proteasome modulators, comprising: (a) incubating a plurality of candidate agents, each individually or in combination, with a recombinant MCBP protein and a ubiquitinated target protein; (b) measuring binding of the MCBP protein to the ubiquitinated target protein in the presence or absence of the candidate agent(s); wherein significant change in binding in the presence of the candidate agent(s) is indicative that the candidate agent(s) is a proteasome modulator.

44. The method of claim 43, wherein the modulator is an inhibitor.

45. The method of claim 43, further comprising determining the activity of a 26S proteasome preparation comprising the MCBP protein, in the presence of the agent(s) identified in (b) as modulator(s), wherein a modulated activity of the 26S proteasome verifies that the candidate agent(s) is/are 26S proteasome modulator(s).

46. The method of claim 45, wherein the activity of the 26S proteasome is determined by using the ubiquitinated target protein as a substrate.

47. The method of claim 45, wherein said 26S proteasome is isolated from wild- type cells.

48. The method of claim 45, wherein the 26S proteasome is reconstituted from a defective 26S proteasome substantially incapable of degrading the ubiquitinated target protein, and a recombinantly produced MCBP protein.

49. The method of claim 45, further comprising determining the effect of the modulator agent(s) identified in (b) on the proteasome 20S peptidase activity, wherein a substantially unchanged 20S peptidase activity is indicative that the modulator agent(s) affects proteasome entry of a ubiquitinated target protein.

50. A method for screening for an agent that inhibits a ubiquitinated target protein's entry into the proteasome, the method comprising: (a) incubating, in the presence of a test agent, the ubiquinated target protein and a sufficient amount of a recombinant MCBP protein that would restore degradation activity of a proteasome preparation lacking a corresponding functional MCBP;

(b) adding a 26S proteasome preparation, wherein the MCBP corresponding to the recombinant MCBP was absent or defective in said 26S proteasome preparation;

(c) determining the effect of the test agent, wherein substantial reduction of target protein degradation is indicative that the test agent inliibits the ubiquitinated target protein's entry into the proteasome.

51. The method of claim 50, wherein the test agent inliibits at least about 50%> of target protein degradation.

52. The method of claim 50, wherein the test agent inliibits nearly 100%> of target protein degradation.