METHODS AND COMPOSITIONS FOR ENHANCING PROTEASOME ACTIVITY

RELATED APPLICATION

[0001] This patent application claims the benefit of U.S. Provisional Patent

Application Serial No. 60/931,745, filed May 24, 2007, the contents of which are herein incorporated by reference in their entirety.

TECHNICAL FIELD

[0002] This invention relates to methods and compositions for enhancing proteasome activity in a cell.

STATEMENT AS TO FEDERALLY SPONSORED RESEARCH

[0003] This invention was made with U.S. Government support under National

Institutes of Health Grant Nos. GM065592 and GM066492. The government has certain rights in the invention.

BACKGROUND OF THE INVENTION

[0004] The ubiquitin-proteasome system is the major pathway for intracellular protein degradation in eukaryotes. Substrates of this pathway acquire a covalently linked polyubiquitin chain through the action of a cascade of ubiquitin activating and conjugating enzymes. This ubiquitin chain serves as the recognition motif for a large multi-subunit protease known as the proteasome, which processively degrades the substrate into small polypeptides while releasing ubiquitin for reuse (Pickart and Cohen. (2004). Proteasomes and their kin: proteases in the machine age. Nat. Rev. MoI. Cell. Biol. 5, 177-187).

SUMMARY OF THE INVENTION

[0005] The present invention provides methods and compositions for enhancing the activity of the proteasome in cells. When a ubiquitinated protein is degraded, the metabolic fates of ubiquitin and its conjugative target are uncoupled at the proteasome by deubiquitination. Proteasomes of budding yeast contain two deubiquitinating enzymes, Ubp6 and Rpnl 1. Rpnl 1 promotes protein breakdown through its degradation-coupled activity. In contrast, Ubp6 has a potent and evolutionarily conserved capacity to inhibit the proteasome in vitro and in vivo. However, inhibitory capacity is retained in a catalytically inactive point mutant of Ubp6, indicating that Ubp6 has both deubiquitinating activity and proteasome inhibitor activity. These functions cooperate in that inhibition of degradation by Ubp6 is

accompanied by progressive trimming of the substrate's ubiquitin chain. Rpnl 1 catalyzes substrate-proximal chain cleavage, and Ubp6 inhibits degradation at or upstream of this step, so that inhibition of degradation by Ubp6 results in a switch in the mode of ubiquitin chain processing. Thus, Ubp6 regulates both the nature and magnitude of proteasome activity. [0006] The methods and compositions for enhancing the activity of the proteasome in cells modulate the activity of a major component of the proteasome, known as Ubp6 (yeast) or Uspl4 (human), an endogenous inhibitor of the proteasome. For example, the methods and compositions described herein reduce, e.g., partially or completely, the inhibitory activity of Uspl4 on a proteasome, thereby specifically enhancing the protein-degradation activity of the proteasome. The invention also provides methods of screening to identify inhibitors of Ubp6, Uspl4, and/or both Ubp6 and Uspl4.

[0007] Uspl4 is a deubiquitinating enzyme that binds tightly to the proteasome. The full-length human Uspl4 contains 494 amino acids, with a 9 kDa UbI domain (ubiquitin-like domain) at the N-terminus followed by a 45 kDa catalytic domain. (See Hu, et al. (2005). Structure and mechanisms of the proteasome-associated deubiquitinating enzyme Uspl4. EMBO J. 24, 3747-3756). The Uspl4 catalytic domain comprises residues 91-494 of the amino acid sequence shown, for example, in GenBank Accession No. P54578, hereby incorporated by reference in its entirety.

[0008] Understanding of this Uspl4 protein has been achieved mainly through study of its yeast ortholog, Ubp6. Most proteasomes in the cell are complexed to Ubp6. Ubp6 is a strong inhibitor of proteasome function, both in vivo and in vitro. Thus, the proteasome, and the ubiquitin-proteasome pathway as a whole, operate in a partially inhibited state. This is an evolutionarily conserved property, because it has been found that Uspl4 also has the capacity to inhibit yeast proteasomes.

[0009] As described herein, Ubp6 exerts a two-fold inhibitory influence on the proteasome: it inhibits proteasomes both catalytically and non-catalytically. Catalytic inhibition is mediated by the removal of ubiquitin groups from the ubiquitinated substrate protein. This is inhibitory because ubiquitin targets the substrate protein to the proteasome and is necessary for substrate degradation. Noncatalytic inhibition is defined as the component of proteasome inhibition observed when inhibition is accomplished using an active-site mutant of Ubp6.

[0010] The invention also provides methods of treating a proteinopathy (or protein- folding disease), including, by way of non-limiting example, neurodegenerative disorders and other diseases such as certain cancers. The methods of treating a proteinopathy include

methods of reducing the severity of a proteinopathy, methods of reducing at least one symptom of a proteinopathy, and methods of stabilizing at least one symptom associated with a proteinopathy. The proteinopathy, or a symptom thereof, is, e.g., partially or completely alleviated using the methods described herein.

[0011] A major insight into human disease arising from studies of the last decade and more is that they are frequently caused by abnormal or misfolded proteins. These diseases are known collectively as proteinopathies and appear to be caused by the accumulation of aberrant proteins. Proteinopathies include, for example, many neurodegenerative diseases, such as, e.g., Alzheimer's Disease, Parkinson's Disease, Lewy Body Dementia, ALS, Huntington's Disease, Spinocerebellar Ataxias, and Spinobulbar Muscular Atrophy, and other diseases, including some cancers. In general, misfolded proteins are eliminated from the body by the proteasome. But in the case of such diseases, the proteasome cannot eliminate all of these harmful proteins. Thus, the methods of the present invention allow for the treatment or prevention of a proteinopathy by enhancing the activity of a proteasome through the administration of an inhibitor of Ubp6/Uspl4. The enhanced activity of the proteasome, in turn, is used to remove the harmful proteins, thereby ameliorating such diseases. Currently, there is no recognized strategy for enhancing the activity of the proteasome pathway. As described herein, the proteasome pathway does not ordinarily operate at full efficiency, but rather it operates below its true capacity because the pathway is partially inhibited by Ubp6/Uspl4. Thus, the inhibition of Ubp6/Uspl4 of the methods and compositions described herein provide a simple approach to enhancing proteasome activity in the cell.

[0012] For example, it has been found that the absence of Ubp6 alleviated the effects of toxic, expanded forms of the Huntington gene product, when Huntington was expressed in yeast (Fig. 13). This phenomenon was independent of any effect of Ubp6 on ubiquitin levels, suggesting that the noncatalytic inhibitory effect of Ubp6 is most likely responsible. [0013] The methods and compositions described herein are administered to a subject suffering from, or suspected of suffering from, a proteinopathy. A subject suffering from a proteinopathy is identified by methods known in the art. For example, a subject is suffering from a proteinopathy such as a neurological disorder is identified using any of a variety of clinical and/or laboratory tests such as, physical examination, radiologic examination and blood and/or urine analysis.

[0014] Administration of a Uspl4 inhibitor to a patient suffering from, predisposed to, or suspected of suffering from, a proteinopathy is considered successful if any of a variety

of laboratory or clinical results is achieved. For example, administration of a Uspl4 inhibitor to a patient suffering from a proteinopathy is considered successful one or more of the symptoms associated with the disorder is alleviated, reduced, inhibited or does not progress to a further, i.e., worse, state. Administration of a Uspl4 inhibitor to a patient suffering from a proteinopathy is considered successful if the disorder, e.g., a neurological disorder or certain cancers, enters remission or does not progress to a further, i.e., worse, state. [0015] The invention described herein provides a method of enhancing proteasome activity in a cell by contacting, treating or otherwise exposing the cell with a Uspl4 inhibitor in an amount sufficient to reduce endogenous inhibition of proteasome-mediated protein degradation. For example, the Uspl4 inhibitor is a small-molecule inhibitor. A "small molecule", as used herein, is meant to refer to a composition that has a molecular weight of less than about 5 kD and most preferably less than about 4 kD. Small molecules can be, e.g., nucleic acids, peptides, polypeptides, peptidomimetics, carbohydrates, lipids or other organic or inorganic molecules. Libraries of chemical and/or biological mixtures, such as fungal, bacterial, or algal extracts, are known in the art and can be screened with any of the assays of the invention. For example, the Uspl4 inhibitor interacts with, e.g., binds to or otherwise interferes with, at least a portion of the catalytic region of Uspl4, e.g., residues 91-49 of the amino acid sequence shown in GenBank Accession No. P54578. In some embodiments, the Uspl4 inhibitor interacts with, e.g., binds to or otherwise interferes with the binding surface, i.e., surface groove, between the Palm and Thumb region of Uspl4. (Hu, et al. (2005). Structure and mechanisms of the proteasome-associated deubiquitinating enzyme Uspl4. EMBO J. 24, 3747-3756). In some embodiments, the Uspl4 inhibitor interacts with, e.g., binds to or otherwise interferes with a portion of the Uspl4 polypeptide sequence of GenBank Accession No. P54578 selected from α 1 helix (residues 114-124), α 2 helix (residues 127-133), α 3 helix (residues 147-165), α 4 helix (residues 173-182), α 5 helix (residues 200-214), α 6 helix (residues 244-247), α 7 helix (residues 287-291), α 8 helix (residues 361-376), α 9 helix (residues 460-464), βl strand (residues 169-171), β2 strand (residues 249-251), β3 strand (residues 252-257), β4 strand (residues 269-271), β5 strand (residues 273-275), β6 strand (residues 296-301), β7 strand (residues 308-316), β8 strand (residues 325-329), β 9 strand (residues 352-354), βlO strand (residues 418-428), βl 1 strand (residues 434-441), βl2 strand (residues 447-449), βl3 strand (residues 477-480), BLl loop (residues 329-351), BL2 loop (residues 429-433), and any combinations thereof. [0016] In some embodiments, the Uspl4 inhibitor used in the methods of enhancing proteasome activity impedes, e.g., partially or completely, a catalytic activity of Uspl4. For

example, the Uspl4 inhibitor interacts with the catalytic site of Uspl4. In some embodiments of the methods of enhancing proteasome activity described herein, the Uspl4 inhibitor impedes, e.g., partially or completely, a non-catalytic inhibitory activity of Uspl4. For example, the Uspl4 inhibitor increases the rate of proteasome-mediated degradation of ubiquitin or a ubiquitinated polypeptide as compared to the rate of proteasome-mediated degradation in the absence of the Uspl4 inhibitor.

[0017] The invention also provides a method of treating a proteinopathy in a subject, or alleviating a symptom associated with a proteinopathy in a subject, by administering to the subject a Uspl4 inhibitor in an amount effective to reduce inhibition of proteasome-mediated protein degradation by Uspl4. As used herein, the term patient includes human and veterinary subjects. The term subject includes humans and other mammals. Proteinopathies to be treated by the methods described herein include, by way of non-limiting example, a neurodegenerative disorder or a cancer. For example, the proteinopathy is Huntington's disease. Alleviation of a symptom associated with a proteinopathy is, e.g., complete or partial.

[0018] In some embodiments, the Uspl4 inhibitor is a small-molecule inhibitor. For example, the Uspl4 inhibitor interacts with, e.g., binds to or otherwise interferes with, at least a portion of the catalytic region of Uspl4, e.g., residues 91-49 of the amino acid sequence shown in GenBank Accession No. P54578. In some embodiments, the Uspl4 inhibitor interacts with, e.g., binds to or otherwise interferes with the binding surface, i.e., surface groove, between the Palm and Thumb region of Uspl4. (Hu, et al. (2005). Structure and mechanisms of the proteasome-associated deubiquitinating enzyme Uspl4. EMBO J. 24, 3747-3756). In some embodiments, the Uspl4 inhibitor interacts with, e.g., binds to or otherwise interferes with a portion of the Uspl4 polypeptide sequence of GenBank Accession No. P54578 selected from α 1 helix (residues 114-124), α 2 helix (residues 127-133), α 3 helix (residues 147-165), α 4 helix (residues 173-182), α 5 helix (residues 200-214), α 6 helix (residues 244-247), α 7 helix (residues 287-291), α 8 helix (residues 361-376), α 9 helix (residues 460-464), βl strand (residues 169-171), β2 strand (residues 249-251), β3 strand (residues 252-257), β4 strand (residues 269-271), β5 strand (residues 273-275), β6 strand (residues 296-301), β7 strand (residues 308-316), β8 strand (residues 325-329), β 9 strand (residues 352-354), βlO strand (residues 418-428), βl 1 strand (residues 434-441), βl2 strand (residues 447-449), βl3 strand (residues 477-480), BLl loop (residues 329-351), BL2 loop (residues 429-433), and any combinations thereof.

[0019] In some embodiments of the methods of treating a proteinopathy described herein, the Uspl4 inhibitor impedes, e.g., partially or completely, a catalytic activity of Uspl4. For example, the Uspl4 inhibitor interacts with the catalytic site of Uspl4. In some embodiments, the Uspl4 inhibitor used in the methods of treating a proteinopathy impedes, e.g., partially or completely, a non-catalytic inhibitory activity of Uspl4. For example, the Uspl4 inhibitor increases the rate of proteasome-mediated degradation of ubiquitin or a ubiquitinated polypeptide as compared to the rate of proteasome-mediated degradation in the absence of the Uspl4 inhibitor.

[0020] The invention also provides a variety of screening assays useful to identify

Uspl4 inhibitors. For example, the invention described herein provides a method of screening for a Uspl4 inhibitor by (i) providing a Uspl4 polypeptide and a proteasome under conditions sufficient to allow the Uspl4 polypeptide and proteasome to form a complex; (ii) contacting, treating or otherwise exposing the Uspl4 polypeptide/proteasome complex with a test compound; and (iii) determining whether the test compound dissociates the Uspl4 polypeptide/proteasome complex, whereby dissociation of the Uspl4 polypeptide/proteasome complex indicates that the test compound is a Uspl4 inhibitor. The test compound is, for example, a small molecule, and the proteasome is selected from a yeast proteasome, a murine proteasome and a human proteasome. The test compound is optionally detectably labeled. [0021] In some embodiments of the methods of screening for a Uspl4 inhibitor, the

Uspl4 inhibitor to be identified impedes, e.g., partially or completely, a catalytic activity of Uspl4. For example, the Uspl4 inhibitor interacts with the catalytic site of Uspl4. In some embodiments, the Uspl4 inhibitor used in this screening method impedes, e.g., partially or completely, a non-catalytic inhibitory activity of Uspl4. For example, the Uspl4 inhibitor increases the rate of proteasome-mediated degradation of ubiquitin or a ubiquitinated polypeptide as compared to the rate of proteasome-mediated degradation in the absence of the Uspl4 inhibitor.

[0022] The invention also provides a method of screening for a Uspl4 inhibitor that interacts with the catalytic site of Uspl4 by (i) providing a test solution comprising Uspl4 polypeptide and a proteasome under conditions sufficient to allow the Uspl4 polypeptide and proteasome to form a complex; (ii) contacting, treating or otherwise exposing the test solution with a test compound and a substrate of the Uspl4 polypeptide/proteasome, wherein the substrate is coupled to a reporter that is detectable after cleavage by a deubiquitinating enzyme; (iii) determining whether the substrate is deubiquitinated in the presence of the test compound and the Uspl4 polypeptide/proteasome complex, whereby an absence of

deubiquitination of the substrate in the presence of the test compound and the Uspl4 polypeptide/proteasome complex indicates that the test compound interacts with the Uspl4 catalytic site. In some embodiments, the substrate is ubiquitin-AMC. The test compound is, for example, a small molecule. In some embodiments, the reporter is fluorescently labeled. [0023] The invention also provides a method of screening for Uspl4 inhibitors that impede, e.g., partially or completely, non-catalytic proteasome inhibition by Uspl4 by (i) providing a test solution comprising Uspl4 polypeptide and a proteasome under conditions sufficient to allow the Uspl4 polypeptide and proteasome to form a complex; (ii) contacting, treating or otherwise exposing the test solution with a test compound and a ubiquitinated cyclin B polypeptide; (iii) determining the rate of degradation of the ubiquitinated cyclin B polypeptide in the presence of the test compound; and (iv) comparing the rate of degradation of the ubiquitinated cyclin B polypeptide in the absence of the test compound; whereby an increased rate of degradation of ubiquitinated cyclin B in the presence of the test compound as compared to the rate of degradation in the absence of the test compound indicates that the test compound is an inhibitor of non-catalytic proteasome inhibition by Uspl4. The Uspl4 polypeptide is, for example, a catalytically inactivated form of Uspl4. For example, the Uspl4 polypeptide contains a mutation, e.g., amino acid substitution, deletion, and/or insertion, at one or more residues in the catalytic domain, e.g., residues 91-49 of the amino acid sequence shown in GenBank Accession No. P54578. For example, one or more catalytically active residues is mutated, such as for example, Cysl 14, His435 and Asp451 residues of the amino acid sequence shown in GenBank Accession No. P54578. In some embodiments, the test compound is a small molecule. The proteasome is, for example, selected from a yeast proteasome, a murine proteasome and a human proteasome. In some embodiments, the ubiquitinated cyclin B polypeptide is a ubiquitinated form of a cyclin B- luciferase fusion protein.

[0024] Other features, objects, and advantages of the invention will be apparent from the description and drawings. All publications and patent documents cited herein are incorporated herein by reference as if each such publication or document was specifically and individually indicated to be incorporated herein by reference. Citation of publications and patent documents is not intended as an admission that any is pertinent prior art, nor does it constitute any admission as to the contents or date of the same.

BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figures IA- IE are a series of illustrations and graphs depicting the ability of

Upb6 to inhibit degradation of ubiquitinated cyclin B by purified yeast proteasomes. [0026] Figures IA, IB and 1C depict the processing of ub-cyclin B in vitro by wild- type or ubpόA proteasomes, visualized by immunoblot with antibodies to cyclin Bl (Figure IA), Ubp6 (Figure IB), or proteasome subunit Rpn5 (Figure 1C).

[0027] Figure ID compares wild-type and ubpόA proteasomes (10 μg), as determined by SDS-PAGE followed by Coomassie staining.

[0028] Figure IE is a graph depicting proteasome-dependent hydrolysis of suc-

LLVY-AMC, a ubiquitin-independent fluorogenic proteasome substrate. Error bars reflect standard deviations.

[0029] Figure IF depicts processing of ub-cyclin B by proteasomes in the presence of purified recombinant Ubp6 (* represents a cyclin B species generated by residual thrombin activity deriving from recombinant Ubp6 purifications).

[0030] Figure IG depicts O-phenanthroline (o-PA) inhibition of cyclin B degradation by ubpόδ proteasomes, implicating Rpn 1 l's deubiquitinating activity in this process. o-PA, dissolved in DMSO, was added at 10 mM, 10 min prior to cyclin B.

[0031] Figures 2A-2C are a series of illustrations depicting the inhibition of cyclin B degradation by epoxomicin.

[0032] Figure 2 A depicts the effect of purified Ubp6 on proteasome-mediated ub- cyclin B processing in the presence or absence of epoxomicin (100 μM). [0033] Figures 2B and 2C depicts processing of ub-cyclin B by wild-type and ubpόδ proteasomes in the presence or absence of epoxomicin. Ub-cyclin B was synthesized with wild-type or K48R ubiquitin, as indicated.

[0034] Figures 3 A and 3B are a series of illustrations depicting that deletion of the

UBP6 gene accelerates degradation of a proteasome substrate in vivo. [0035] Figure 3A depicts the growth of wild-type (SYT303), ubpόδ (SJHl 70), and ubrlδ (SYT304) yeast strains on media containing or lacking tryptophan, as indicated. [0036] Figure 3B depicts cycloheximide-chase reactions examining turnover of the

Ura3 and Trp 1 reporter proteins in wild-type (SYT318), ubpόδ (SJH 126), and ubrlδ (SYT342) strains. Reactions were visualized with immunoblots using antibodies against the HA-epitope or proteasome subunit, Rpn5, as indicated. The far right represents shorter exposure of the ubrlδ lanes.

[0037] Figures 4A-4F are a series of illustrations and graphs depicting a non-catalytic function of Ubp6 in proteasome inhibition.

[0038] Figure 4A is a schematic representation of the various Ubp6 mutants, MBP, maltose-binding protein.

[0039] Figures 4B and 4E depict Ub-cyclin B processing by ubpόδ proteasomes, in the presence or absence of the indicated Ubp6 species.

[0040] Figure 4C includes an illustration and a graph depicting a comparison of the extent of proteasome inhibition by wild-type Ubp6 and Ubp6-C 11 8 A by quantitative mass spectrometry. The upper panels of Figure 4C depict immunoblots using anti-cyclin B, -Ubp6, and -Rpnl2 antibodies as indicated. The lower panel of Figure 4C depicts quantitation results. Error bars represent the standard deviation of two independent quantitations.

[0041] Figure 4D depicts the failure of Ubp6 to deubiquitinate cyclin B when proteasomes are absent.

[0042] Figure 4F depicts the electrophoretic mobility shift of proteasome bound to ubiquitinated Cdc34. Proteasomes were visualized using the fluorogenic substrate suc-

LLVY-AMC.

[0043] Figures 5A-5E are a series of illustrations depicting in vivo differentiation of ubpόδ and ubp6-CH8A mutants.

[0044] Figures 5A and 5B depict the growth of wild-type (SJH30), wild-type overexpressing ubiquitin (SJH34), ubpόδ (SJH31), and ubpόδ overexpressing ubiquitin

(SJH35) yeast stains on selective plates containing copper sulfate (lOOμM) in the presence or absence of canavanine (1.5 μg/ml) or rapamycin (200 ng/ml), and grown at 30 ⁰ C for 3-7 days.

[0045] Figure 5C depicts the growth of wild-type (SJH 152), ubpόδ (SJH 153), ubpόδ expressing wild-type Ubp6 (SJH154), and ubpόδ expressing Ubp6-Cl 18A (SJH155) yeast strains on selective plates containing no drug, canavanine (1.5 μg/ml), or rapamycin (200 ng/ml) as indicated, and grown at 3O ⁰ C for 3-7 days. For SJH154 and SJH155, duplicates represent two independent transformants.

[0046] Figure 5D depicts cycloheximide-chase analyses of free ubiquitin turnover conducted in wild-type (SJH 120), ubpόδ overexpressing wild-type Ubp6 (SJH20), and ubpόδ overexpressing Ubp6-Cl 18A (SJH22) strains, as indicated.

[0047] Figure 5E depicts the growth of wild-type (SJHl 71), ubpόδ (SJH 172), ubpόδ expressing wild-type Ubp6 (SJH 173), and ubpόδ expressing Ubp6-Cl 18A (SJH 174) yeast strains on selective media containing or lacking tryptophan, as indicated.

[0048] Figures 6A-6C are a series of illustrations depicting degradation independence and evolutionary conservation of Ubp6 function.

[0049] Figures 6 A and 6B depict the processing of ub-cyclin B by RP (ubpόA) in the presence or absence of purified Ubp6. Reactions were visualized by immunoblotting with anti-cyclin B antibody or anti-ubiquitin antibody, as indicated.

[0050] Figure 6C depicts the effect of Ubp6 or its human homolog, Uspl4, on the processing of ub-cyclin B by yeast ubpόδ proteasomes.

[0051] Figure 7 is an illustration depicting cycloheximide-chase analysis of Gcn4 turnover in wild-type and ubpόδ strains. Logarithmically growing cells were treated with cycloheximide (20 μg/ml) and aliquots were taken at the indicated times. Cells were immediately resuspended in IX Laemmli loading buffer, and endogenous Gcn4 was detected by immunoblot using anti-Gcn4 antibody. The bottom panel is a lower exposure of the blot above.

[0052] Figure 8 is a graph depicting that the Ubp6-C 118 A mutant lacks deubiquitinating activity. Ub-AMC hydrolyzing activity of purified recombinant wild-type Ubp6 or Ubp6-Cl 18A (0.5 uM). Ub-AMC, no enzyme control. [0053] Figures 9 A and 9B are a series of illustrations and a graph depicting that ubiquitin-vinyl sulfone treated Ubp6 retains non-catalytic inhibition of degradation. [0054] Figure 9A depicts the result of an in vitro ub-cyclin B assay carried out with ubpόδ proteasomes (135 nM) supplemented with wild-type Ubp6 (540 nM) and ubiquitin- vinyl sulfone (2.5 uM), as indicated. The lower panel is anti-Ubp6 immunoblot showing modification of Ubp6 by Ub-VS. Ubp6 was present in four-fold molar excess over proteasome, and thus only a fraction of Ubp6 protein was modified. [0055] Figure 9B depicts the results of Ub-AMC (0.5 uM) hydroylsis assay.

[0056] Figures 1OA and 1OB are illustrations depicting the results of the Ub-Cyclin B binding assay. Resins bearing GST, GST-Rpn 10, and GST-Ubp6 were incubated with Ub- cyclin B. Excess glutathione was used to elute complexes, and eluates (10% of total) were visualized by immunoblot with anti-Cyclin B antibody (Figure 10A) or Coomassie staining (Figure 10B). No detectable binding of Ub-Cyclin B conjugates was observed with GST- Ubp6. Rpn 10, a known ubiquitin receptor, served as a positive control. Input lane represents 2.5% of the total Ub-cyclin B per reaction.

[0057] Figure 11 is an illustration depicting that the rapamycin-related functions of

Ubp6 require its ubiquitin-like domain. Wild-type (SUB62) and ubpόδ (SY255c) strains were transformed with empty vector (WTD72) or plasmids encoding Ubp6 (WTD72), Ubp6-

ublδ (WTD73), or Ubp6-Cl 18S (WTD74) expressed from the GAPl promoter. Cultures were spotted in three-fold serial dilutions onto selective plates containing canavanine (1.5 mg/ml) or rapamycin (200 ng/ml), and grown for 2-7 days at 30 ⁰ C. [0058] Figure 12 is an illustration depicting the RP-mediated disassembly of Ub- cyclin B conjugates by Ubp6. Proteasome regulatory particle (135 nM) purified from a ubpόδ strain was combined with a four-fold molar excess of Ubp6, Ubp6-ublδ, or Ubp6- Cl 18A, and Ub-Cyclin B, and incubated for the indicated times. Reactions were visualized by immunoblot with anti-ubiquitin antibody. * represents a non-ubiquitin immunoreactive band.

[0059] Figure 13 is an illustration depicting that loss of Ubp6 function protects cells against Huntington-mediated toxicity. Wild-type and ubpόδ cells were transformed with plasmids encoding non-pathogenic and pathogenic versions of the Huntington protein, expressed from a GAL promoter. Cells were spotted in 3 -fold serial dilutions on media containing glucose (weak expression of Htt genes) or galactose (strong expression of Htt genes).

[0060] Figure 14 is an illustration depicting the inhibition of proteasome by Uspl4.

[0061] Figure 15 is an illustration depicting a sequence alignment of Uspl4, its yeast ortholog Ubp6 and another member of the UBP family, HAUSP (also known as Usp7) taken from Hu, et al. (2005). EMBO J. 24, 3747-3756. Conserved residues are shaded whereas the catalytic triad residues are highlighted by a vertical shaded box. The secondary structural elements above the sequences are indicated for free USP 14 (lower) and HAUSP (upper). The four black arrowheads indicate the positions where Cys residues are supposed to be located in a zinc ribbon (Krishna SS, and Grishin NV (2004) The finger domain of the human deubiquitinating enzyme HAUSP is a zinc ribbon. Cell Cycle 3: 1046-1049). Sequence alignment employed the program ClustalW. Entries shown are from the SwissProt Database: HAUSP (Human; W:Q93009); USP14 (Human; SW:P54578); and UBP6 (S. cerevisiae; SW:P35127).

DETAILED DESCRIPTION

[0062] The ubiquitin-proteasome system is the major pathway for intracellular protein degradation in eukaryotes. Substrates of this pathway acquire a covalently linked polyubiquitin chain through the action of a cascade of ubiquitin activating and conjugating enzymes. This ubiquitin chain serves as the recognition motif for a large multi-subunit protease known as the proteasome, which processively degrades the substrate into small

polypeptides while releasing ubiquitin for reuse (Pickart and Cohen. (2004).Proteasomes and their kin: proteases in the machine age. Nat. Rev. MoI. Cell. Biol. 5, 177-187). [0063] The proteasome is an approximately 2.5-MDa protein complex consisting of at least 33 distinct subunits in yeast (Pickart and Cohen, 2004). Its proteolytic active sites are housed at the center of the structure in a barrel-shaped subcomplex known as the core particle (CP, or 2OS complex) (Groll, et al. (1997). Structure of 2OS proteasome from yeast at 2.4 A resolution. Nature 386, 463-471). At either axial end of the CP, a second subcomplex known as the regulatory particle (RP, or 19S complex or PA700) may bind. When one or both ends of the CP are occupied by the RP, the structure is referred to as the 26S proteasome. The RP comprises two subcomplexes of its own: the base and the lid (Glickman, et al. (1998). A subcomplex of the proteasome regulatory particle required for ubiquitin-conjugate degradation and related to the COP9-signalosome and eIF3. Cell 94, 615-623). The base is proximal to the CP and contains, among other proteins, six ATPases that are thought to form a ring structure that abuts the CP. The lid is distal to the CP and its subunit Rpnl 1 is thought to possess a metalloprotease-based deubiquitinating activity (Verma, et al. (2002). Role of Rpnl 1 metalloprotease in deubiquitination and degradation by the 26S proteasome. Science 298, 611-6 15; Yao and Cohen. (2002). A cryptic protease couples deubiquitination and degradation by the proteasome. Nature 419, 403-407; Maytal-Kivity, et al. (2002). MPN+, a putative catalytic motif found in a subset of MPN domain proteins from eukaryotes and prokaryotes, is critical for Rpn 11 function. BMC Biochem. 3, 28). [0064] The proteasome displays a number of characteristics unusual for a protease.

First, its proteolytic active sites are sequestered within the hollow cylindrical chamber of the CP (Groll, et al. (1997). Structure of 2OS proteasome from yeast at 2.4 A resolution. Nature 386, 463-471). Providing access to this internal chamber are narrow gates at either end of the CP, and gating appears to be modulated by the Rpt2 subunit of the base (Pickart and Cohen. (2004). Proteasomes and their kin: proteases in the machine age. Nat. Rev. MoI. Cell. Biol. 5, 177-187). These features presumably serve to prevent the unregulated destruction of intracellular proteins, and also impose a requirement for substrate unfolding, as most folded proteins are too large to pass through the open translocation channel leading to the CP. Protein unfolding is apparently mediated by the six ATPases of the base (Braun, et al. (1999). The base of the proteasome regulatory particle exhibits chaperone-like activity. Nat. Cell Biol. 1, 221-226), which may explain, at least in part, a second unusual feature of the proteasome, its ATP dependence. Finally, substrate recognition by the proteasome is, with some exceptions, dependent on the presence of a ubiquitin chain. Multiple ubiquitin

receptors have been identified, some of which are core proteasomal subunits while others are substoichiometric proteasome-associating factors (Elsasser and Finley. (2005). Delivery of ubiquitinated substrates to protein-unfolding machines. Nat Cell Biol. 7, 742-9). At some point after recognition, the ubiquitin chain is removed from the substrate to facilitate substrate degradation and to minimize degradation of ubiquitin. This function is carried out by deubiquitinating enzymes that reside in the proteasome.

[0065] In budding yeast S. cerevisiae, at least two deubiquitinating enzymes are thought to contribute to deubiquitination by the proteasome: Rpnl 1 and Ubp6 (Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell. 10, 495-507; Verma, et al. (2002). Role of Rpnl 1 metalloprotease in deubiquitination and degradation by the 26S proteasome. Science 298, 611-6 15; Yao and Cohen. (2002). A cryptic protease couples deubiquitination and degradation by the proteasome. Nature 419, 403-407; Maytal-Kivity, et al. (2002). MPN+, a putative catalytic motif found in a subset of MPN domain proteins from eukaryotes and prokaryotes, is critical for Rpn 11 function. BMC Biochem. 3, 28; Chemova, et al. (2003). Pleiotropic effects of Ubp6 loss on drug sensitivities and yeast prion are due to depletion of the free ubiquitin pool. J. Biol. Chem. 278, 52102-521 15; Guterman and Glickman. (2004). Complementary roles for Rpnl 1 and Ubp6 in deubiquitination and proteolysis by the proteasome. J. Biol. Chem. 279, 1729-1738). Rpnl 1 is a core structural component of the lid, and whereas most deubiquitinating enzymes utilize a cysteine-based proteolytic mechanism, Rpnl 1 is thought to be a metalloprotease. Point mutations in its metal-coordinating site impair substrate degradation in vitro and vivo, indicating a positive role for Rpnl 1 in protein degradation (Verma, et al. (2002). Role of Rpnl 1 metalloprotease in deubiquitination and degradation by the 26S proteasome. Science 298, 611-6 15; Yao and Cohen. (2002). A cryptic protease couples deubiquitination and degradation by the proteasome. Nature 419, 403-407).

[0066] The second proteasomal deubiquitinating enzyme, Ubp6, has been classified as a proteasome-associated protein, largely on the basis of its ready dissociation from the proteasome in the presence of high salt concentrations (Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell. 10, 495-507). Ubp6, a cysteine protease, is an abundant component of proteasomes (Verma, et al. (2000). Proteasomal proteomics: identification of nucleotide-sensitive proteasome-interacting proteins by mass spectrometric analysis of affinity-purified proteasomes. MoI. Biol. Cell. 11, 3425-3439; Leggett et al., 2002), and in contrast to Rpnl 1, Ubp6 associates with the base.

An N-terminal ubiquitin-like domain (UbI) of Ubp6 mediates this interaction. Binding of

Ubp6 to the proteasome activates Ubpό's catalytic activity over 300-fold (Leggett et al., 2002), indicating an intimate functional relationship between Ubp6 and the proteasome. However, the precise role of Ubp6 in proteasome function has remained uncertain. Some reports have assigned Ubp6 little or no role in proteasome-mediated deubiquitination or degradation, while others have argued that Ubp6, like Rpnl 1, facilitates proteasome- mediated degradation.

[0067] The data and examples provided herein present the first detailed analysis of how Ubp6 affects the degradation of a physiological substrate of the proteasome. Ubp6 inhibits protein breakdown by the proteasome, that a major component of its inhibitory effect is non-catalyticin nature. The inhibitory effect was observed in vivo and in vitro, and on different test substrates. During the degradation delay that Ubp6 imposes, substrate deubiquitination proceeds on proteasomes, but the mode of deubiquitination is altered from that seen in the absence of Ubp6: degradation-linked "en bloc" chain removal by Rpnl 1 is replaced by progressive chain trimming by Ubp6. Thus, Ubp6 uses catalytic and non- catalytic mechanisms to modulate proteasome function, and the coordinated activity of multiple proteasomal deubiquitinating enzymes controls substrate fate. [0068] Ubp6, a proteasomal deubiquitinating enzyme, is an endogenous proteasome inhibitor. Proteasomes regenerate ubiquitin by separating it from substrate prior to substrate degradation. However, it seems unlikely that this function alone could account for the existence of three distinct deubiquitinating enzymes in the proteasomes of most eukaryotes. Indeed, the mammalian proteasome has as many distinct active sites for deubiquitination as it has proteolytic active sites for substrate hydrolysis.

[0069] The activity of one proteasomal deubiquitinating enzyme, Rpnl 1 , is coupled to and facilitates degradation (Verma, et al. (2002). Role of Rpnl 1 metalloprotease in deubiquitination and degradation by the 26S proteasome. Science 298, 611-6 15; Yao and Cohen. (2002). A cryptic protease couples deubiquitination and degradation by the proteasome. Nature 419, 403-407). The studies, data and examples provided herein demonstrate that another proteasomal deubiquitinating enzyme, Ubp6, has the capacity to inhibit degradation. Studies with Uch37 have shown that a proteasomal deubiquitinating enzyme can potentially antagonize degradation by premature deubiquitination (Lam, et al. (1997). Editing of ubiquitin conjugates by an isopeptidase in the 26S proteasome. Nature 385, 737-740). [0070] However, Ubp6 antagonizes degradation by a novel mechanism that does not require deubiquitination of the degradative substrate: it inhibits the proteasome directly.

Proteasomes operate on ub-cyclin B in vitro with substantially reduced efficiency when they are bound to Ubp6, and Ubp6 shows a comparable in vivo effect in reducing flux of at least some specific substrates through the proteasome. Thus, by virtue of Ubp6 the proteasome is under strong inhibitory control under standard growth conditions.

[0071] Substrate deubiquitination by Rpnl 1 is negatively controlled by Ubp6. One effect of Ubpό's inhibitory action is to prevent en bloc deubiquitination of the substrate by Rpnl 1 (Fig. 2). During the degradation delay imposed by Ubp6, its own deubiquitinating activity is operational, as can be seen in a gradual reduction in the number of substrate-bound ubiquitin groups (Figs. 4 and 6). Thus, Ubp6 disassembles the ubiquitin chains and inhibits the degradation of the same substrate. In this way, the mode of substrate deubiquitination is significantly altered from that seen in the absence of Ubp6. These results show that the proteasome has distinct modes of deubiquitination, and that the en bloc mode is subject to negative control, principally by anon-catalytic function of Ubp6.

[0072] If Ubp6 were to inhibit degradation downstream of Rpnl 1, the substrate would lack ubiquitin modification during degradation delay. In this scenario, release from Ubp6- mediated inhibition is unproductive, as substrates lacking a ubiquitin modification dissociate from the proteasome rather than degrade. Thus, inhibition of the proteasome by Ubp6 delays the decision of whether the substrate will be degraded, and during the delay, attrition of the ubiquitin chain on the substrate proceeds gradually through the activity of Ubp6. The length of time allowed for delay of degradation is an important parameter. During this time period, shortening of the chain beyond a critical length leads to dissociation of the substrate from the proteasome. This model indicates the presence of two distinct activities in Ubp6, catalytic and non-catalytic in nature.

[0073] The catalytic activity of Ubp6 also inhibits degradation, and in a manner that is strongly dependent on the duration of non-catalytic inhibition. The in vivo data (Fig. 5E) provide support for this mechanism. However, the extent of cyclin B ubiquitination achieved in the in vitro reaction may be high enough that limited chain attrition does not compromise binding of the conjugate to the proteasome, while substrates bearing shorter chains, or exposed to longer time courses, would display such an inhibitory capacity of Ubp6 more clearly.

[0074] The Mechanism of Proteasome Inhibition by Ubp6: The studies provided herein in the Examples elucidate the mechanism of proteasome inhibition by Ubp6. These studies provide the first demonstration that a deubiquitinating enzyme utilize non-catalytic functions to modulate the activity of another protein or complex. No interference with

substrate binding on the part of Ubp6 was observed, and an interaction between Ubp6 and the proteasome mediates inhibition. Ubp6 is tethered to the proteasome through its UbI domain, which binds subunit Rpnl of the base (Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell. 10, 495-5072002; Stone, et al. (2004). Uch2/Uch37 is the major deubiquitinating enzyme associated with the 26S proteasome in fission yeast. J. MoI. Biol. 344, 697-706). However, proteasome binding by the UbI domain itself has no inhibitory effect (Fig. 4E). Indeed, both the UbI and C-terminal domains are required for inhibition. A simple model is that the UbI domain serves only a proteasome- tethering role and that the C-terminal domain contains the specific sequence information for inhibition. Crystal structures of Ubp6 and its mammalian homolog, USP 14, indicate several regions within the C-terminal domain of Ubp6, including a notable surface-exposed alpha helix, that are not conserved with the unrelated deubiquitinating enzyme HAUSP, and are unlikely to contribute to enzymatic activity (Hu, et al. (2005). Structure and mechanisms of the proteasome-associated deubiquitinating enzyme USP14. EMBO J. 24, 3747-3756). In contrast, these regions are highly conserved among Ubp6 homologs, and it may be that one of these regions mediates the non-catalytic activity of the protein.

[0075] The studies presented herein in the Examples were designed to rule out roles for Ubp6 in substrate targeting (Fig. 4E and F), gate opening (Fig. IE), modulation of CP active sites (Fig. IE), as well as modulation of overall proteasomal ATPase activity. Previous studies suggested that the C-terminal domain of Ubp6 has lid-binding activity (Leggett et al., 2002); this interaction might effect inhibition. This model is particularly interesting since the lid is required for full activation of Ubp6 upon complex formation with the proteasome (Leggett et al., 2002). Thus, the same Ubp6-lid binding event could conceivably inhibit the proteasome and assist in the activation of Ubp6. Since Rpnl 1 is a component of the lid, the lid interaction model also indicates that Ubp6 inhibits degradation by directly inhibiting Rpnl 1 rather than an unknown step upstream of Rpnl 1.

[0076] Proteasome inhibition by Ubp6 is not complete. Although this is not surprising

(complete inhibition of the proteasome would be detrimental), the mechanism by which inhibition is relieved is unclear. It is also unclear whether release of the proteasome from Ubp6-mediated inhibition is itself a regulated step. Since Ubp6 is not an integral subunit of the proteasome, the fraction of proteasomes associated with Ubp6 can be freely varied. The extent of proteasome inhibition is optionally also regulated by altering cellular Ubp6 levels, or alternatively via post-translational modulation of Ubp6 activity.

[0077] Studies are also designed to determine the scope of proteasome inhibition by

Ubp6, i.e., the generality of the degradation-inhibitory effect of Ubp6. To determine how substrate-specific the inhibitory effects of Ubp6 are, studies are designed to determine the effect of ubpό mutations on the rates of degradation of a wider variety of substrates. If substrates are affected to very different degrees by Ubp6, it would suggest that Ubp6 may slow degradation rates to enhance the selectivity of protein degradation by the proteasome. Studies of nucleotide polymerases (Joyce and Benkovic. (2004). DNA polymerase fidelity: kinetics, structure, and checkpoints. Biochemistry 43, 143 17-14324) as well as protein synthesis (Cochella and Green. (2005). Fidelity in protein synthesis. Curr. Biol. 15, R536- 540) have shown that that speed and fidelity cannot be optimized simultaneously, but are competing functional criteria. The definition of fidelity for a protease cannot be as clear as for a polymerase, but fidelity may be more easily defined from the standpoint of the fate of ubiquitin groups than of substrates. Ubp6 clearly enhances ubiquitin sparing by the proteasome both in vitro and in vivo, and thus enhances a significant aspect of its fidelity. [0078] Several previous studies have reported Ubp6 to play positive roles in the degradation of specific proteins (Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell. 10, 495-507; Miura and Abe. (2004). Multiple ubiquitin-specific protease genes are involved in degradation of yeast tryptophan permease Tat2 at high pressure. FEMS Microbiol. Lett. 239, 171-179), in contrast to the data and results presented in the Examples herein. These observations do not appear to reflect the scope of the proteasome-inhibitory effect described here, but rather the critical role of Ubpό in maintaining cellular ubiquitin levels, and the importance of ubiquitin levels in determining the degradation rates of these particular proteins. For example, yeast Tat2 can be degraded in a £/βP<J-dependent manner, but Tat2 is a membrane protein and its degradation proceeds through the vacuole rather than the proteasome. In addition, Tat2 degradation is sensitive to mutations in other genes that affect cellular ubiquitin levels (Miura and Abe, 2004). The other known example is ubiquitin-proline β-galactosidase (Leggett et al., 2002), whose degradation is highly sensitive to ubiquitin levels (Johnson, et al. (1995). A proteolytic pathway that recognizes ubiquitin as a degradation signal. J. Biol. Chem. 270, 17442-17456). [0079] The substrate-stabilizing effects of ubiquitin deficiency could obscure the destabilizing effects of release from proteasome inhibition in the ubpόδ mutant. Studies are designed to uncouple these opposing effects in ataxia mice, whose biochemical basis is loss- of-function mutation in Ubp6/Uspl4. These mice display a severe neurologic phenotype culminating in widespread paralysis and premature death (Wilson, et al. (2002). Synaptic

defects in ataxia mice result from a mutation in Uspl4, encoding a ubiquitin-specific protease. Nat. Genet. 32, 420-425). Studies are designed to determine whether these phenotypes are a result of ubiquitin depletion or enhanced proteasomal degradation, or both, and whether either of these cellular phenomena have relevance for human neurologic or neurodegenerative diseases.

[0080] In recent years, pharmacologic inhibition of the proteasome's proteolytic active sites has emerged as an effective anti-cancer treatment in such clinical contexts as multiple myeloma (Adams, J. (2004). The development of proteasome inhibitors as anticancer drugs. Cancer Cell. 5, 417-421). The efficacy of such drugs may reflect an increased requirement for proteasome function in some cancer cells. On the other hand, a deficit of proteasome function can contribute significantly to pathophysiology. Deficient proteasome function has been suggested, for example, to underlie various proteinopathies and neurodegenerative diseases (Goldberg. (2003). Protein degradation and protection against misfolded or damaged proteins. Nature 426, 895-899). For diseases characterized by proteasome deficiency, drugs that inhibit a broadly acting proteasome inhibitor such as Ubp6 provide an effective therapy.

[0081] Structure and Function ofUspl4: The full-length human Uspl4 contains 494 amino acids, with a 9 kDa UbI domain (ubiquitin-like domain) at the N-terminus followed by a 45 kDa catalytic domain. (See Hu, et al. (2005). Structure and mechanisms of the proteasome-associated deubiquitinating enzyme Uspl4. EMBO J. 24, 3747-3756, hereby incorporated by reference in its entirety). The Uspl4 catalytic domain comprises residues 91- 494 of the amino acid sequence shown, for example, in GenBank Accession No. P54578, hereby incorporated by reference in its entirety.

[0082] The family of ubiquitin-specific processing protease (UBP) are cysteine proteases that contain highly divergent sequences and exhibit strong homology mainly in two regions that surround the catalytic Cys and His residues; these are the so-called Cys Box (~19 amino acids) and the His Box (60-90 amino acids) (See e.g., Hu et al.; Papa FR, and Hochstrasser M (1993) The yeast DOA4 gene encodes a deubiquitinating enzyme related to a product of the human tre-2 oncogene. Nature 366: 313-319; D'Andrea A, and Pellman D (1998) Deubiquitinating enzymes: a new class of biological regulators. Crit Rev Biochem MoI Biol 33: 337-352).

[0083] The catalytic domain of Uspl4 resembles an extended right hand comprised of three domains: Fingers, Palm, and Thumb (see e.g., Hu et al. at Figures 1 and 2). The three- domain organization creates a prominent binding surface between the Fingers and the PaIm-

Thumb scaffold, which is predicted to bind to ubiquitin. Uspl4 binds to ubiquitin using the Fingers domain and the surface groove between the Palm and the Thumb. (Hu et al. 2005). The UbI domain of Uspl4 is responsible for binding to the 19S regulatory particle (RP) of the 26S proteasome.

[0084] In Uspl4, the Thumb contains 6 α helices (αl- α 6) and one short β strand

(βl), with the N-terminal Cys Box adopting an extended conformation. The Palm consists of a six-stranded (β5, β8, βlO— βl3) central β sheet, three α helices (al- α9), one short β strand (β9), and several surface loops. Notably, two surface loops hover above and partially fill the predicted binding pocket for the C-terminus of ubiquitin. These two loops are named blocking loops 1 and 2 (BLl and BL2). The Fingers comprise five β strands (β2-β4, β6, and β7). Packing interactions between the central β sheet in the Palm and the globular Thumb do not give rise to an interdomain cleft between the Palm and the Thumb, which is needed for the accommodation of ubiquitin C-terminus.

[0085] The Cysl 14, His435 and Asp451 residues of Uspl4 form a catalytic triad in the active site of free (i.e., unbound) Uspl4. Although the catalytic triad residues already adopt a productive conformation, access to these residues by ubiquitin is restricted. (Hu et al. 2005) Right above the active site of USP 14, the two surface loops BL2 and BLl are positioned very close to the predicted binding groove for the C-terminus of ubiquitin. Loops BL2 and BLl would likely block access of the C-terminus of ubiquitin to the active site of USP 14. Thus, the blockade of the ubiquitin C-terminus binding groove by loops BL2 and BLl must be removed in order for USP 14 to catalyze deubiquitination. Upon binding to ubiquitin, the active site of Uspl4 undergoes conformational changes, for example at residues Tyr333, Phe 331 and Ser 432, which act in concert to accommodate the C-terminus of ubiquitin in the cleft between the Palm and Thumb domains formed by the conformational changes. (Hu et al. 2005). The conformational changes in the BLl and BL2 loops are facilitated by interactions between conserved residues in these loops and residues on ubiquitin. (Ibid.).

[0086] As described above, Ubp6 is the functional homolog of Uspl4 in

Saccharomyces cerevisiae and shares 31% sequence identity with Uspl4 in the catalytic core domain. A comparison of the sequences of Uspl4 and Ubp6 is shown in Figure 15, taken from Hu et al. (2005). It has been shown that the structure of the Uspl4 catalytic domain is also very similar to that of the catalytic core domain from Ubp6 (see e.g., RCSB Protein Data Bank, Accession No. IVJV). Similar to Uspl4, Ubp6 contains two surface loops that are located above and partially block the predicted binding pocket for the C-terminus of ubiquitin

(see e.g., Figure 1C of Hu et al.). These two surface loops exhibit nearly identical topology as BLl and BL2 in Uspl4 (Ibid.). The local structural differences between Uspl4 and Ubp6 concentrate in surface regions whereas the core structural elements are nearly identical to each other (Ibid.). The active site of Ubp6 adopts a highly similar conformation to that of Uspl4 and includes the catalytic triad residues of Cysl 18, His447, and Asn465. (Hu et al. 2005).

[0087] The details of one or more embodiments of the invention are set forth in the accompanying description herein. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred methods and materials are now described. Other features, objects and advantages of the invention will be apparent from the description. In the specification, the singular forms also include the plural unless the context clearly dictates otherwise. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. In the case of conflict, the present specification will control.

EXAMPLES

Example 1: Materials and Methods

[0088] Yeast Strains: The yeast strains used herein are presented below in Table 1.

Standard techniques were used for strain constructions and transformations. Yeast were cultured at 30 ⁰ C. YPD medium consisted of 1% yeast extract, 2% Bacto-peptone, and 2% dextrose. Synthetic medium consisted of 0.7% Difco Yeast Nitrogen Base supplemented with amino acids, adenine, uracil, and 2% dextrose.

[0089] Recombinant Proteins: The recombinant proteins used herein are presented below in Table 1. Recombinant proteins were expressed as GST-fusion proteins, purified by glutathione-Sepharose affinity chromatography, and eluted by thrombin cleavage, as described (Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell. 10, 495-507). Eluates were treated with benzamidine or benzamidine-Sepharose to inactivate thrombin.

[0090] Ubp6-C 118 A mutants were generated by site-directed mutagenesis using the

Quikchange System™ (Stratagene), and verified by DNA sequencing. The UbI-MBP construct was generated by fusing the first 83 codons from UBP6 to the 5'-end of the complete coding sequence of the maltose binding protein from E. coli.

Finley et al. (1987). The yeast polyubiquitin gene is essential for resistance to high temperatures, starvation, and other stresses. Cell 48, 1035-1046.

* Amerik, et al. (2000). Analysis of the deubiquitinating enzymes of the yeast Saccharomyces cerevisiae. Biol. Chem. 381, 981-992).

[0091] Proteasome Purification: A previously described procedure for purification of the proteasome (Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell. 10, 495-507) was modified by adding 1 mM ATP and 5 mM MgCl ₂ to all purification buffers, and by washing resins with 100 bed vol. of buffer containing 50 mM NaCl, as opposed to 50 bed vol. of buffer containing 100 mM NaCl. RP was purified in the presence of ATP and MgCl ₂ as described (Leggett et al., 2002). [0092] In Vitro Deubiquitination/Degradation Assays: Proteasomes (135 nM) were incubated with ub-cyclin B in the presence of 1 mM ATP and buffer (50 mM Tris [pH 7.51, 1 mM EDTA, 5 mM MgCl ₂ ) for the indicated times. Reactions were terminated by the addition of 5X SDS Laemmli loading buffer (Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell. 10, 495-507) and boiled for 5 min. Where indicated, recombinant, purified Ubp6 or Uspl4 species (540 nM) were

preincubated with proteasomes for 5 min prior to the reaction. UbI and UbI-MBP species were used at 1.35 μM to compensate for lower affinity of these species for proteasome binding compared to full length Ubp6 species (Leggett et al., 2002). Epoxomicin (100 μM) was used to inhibit proteolytic activity of proteasomes. Unless otherwise noted, immunoblots were visualized with anti-cyclin Bl polyclonal antibodies. Polyclonal antibodies recognizing Ubp6, Rpn5, Rpnl2, and ubiquitin were used as indicated. Ub-cyclin B was added in molar excess over proteasome, generally on the order of 25-fold, except for Figs. 4C, 6A and 6B, where the substrate: enzyme ratio was increased by a factor of five. [0093] In the experiments of Figs. 1, 4 (except Fig. 4C), and 6C, time points were sampled after the preparation of a single reaction mixture. The zero time points therefore reflect a lag which represents the amount of time required to sample the first time point after having prepared the reaction mixture. This time lag was consistent between samples, and less than 20 seconds. However, the activity of Ubp6 is rapid enough to reduce cyclin B ubiquitination by time zero. In the experiments of Figs. 2, 4C, and 6 (except 6C), each time point was carried out as an independent reaction. The zero time points were prepared in the presence of 5X LLB, and accordingly, no processing could occur before time zero. [0094] In Vivo Degradation Assays: For plating assays, three-fold serial dilutions of yeast cultures grown in YPD were spotted onto plates containing or lacking tryptophan. For cycloheximide chase assays, exponential cultures in YPD supplemented with 100 μM copper sulfate (to further induce expression of the reporter proteins) were normalized to an equivalent cell density and cycloheximide (50 μg/mL) was added. At the indicated time points, an aliquot of each culture was removed, and cells were pelleted and resuspended in IX LLB. Samples were immediately frozen on dry ice for the remainder of the time course, and subsequently boiled for 5 min. Visualization was by immunoblotting with anti-HA antibody (12CA5) or anti-Rpn5 antibody, as indicated.

[0095] Ubiquitination of Cyclin B: Ubiquitination of cyclin B was carried out.

Briefly, for each 30 μl reaction, Anaphase Promoting Complex (APC) was immunopurified from 600 μ 1 of mitotic Xenopus egg extract with 12 μg of anti-Cdc27 antibodies (AF3.1, Santa Cruz Biotechnologies, Santa Cruz, CA) bound to 30 μl of Affiprep Protein A beads (Bio- Rad, Hercules, CA). Beads were washed three times with XB high salt (10 mM potassium HEPES pH 7.7, 500 mM KCl, 0.1 mM CaCl ₂ , 1 mM MgCl ₂ ), two times with XB (same but with 100 mM KCl), and then three times into the reaction buffer (20 mM Tris pH 7.5, 100 mM KCl, 2.5 mM MgCl ₂ , 2 mM ATP). Ubiquitination reactions contained immunoprecipitated APC on 30 μl of beads, and 30 μl of a mix containing recombinant

MBP-human El at 200 μg/ml, histidine-tagged Ubc4 at 66 μg/ml, ubiquitin at 1.25 mg/ml, and 50 μg/ml of histidine-tagged cyclin Bl/cdc2 complex expressed in baculovirus. Reactions were incubated for 90 minutes at 22 ⁰ C with agitation on a shaker at 1,000 RPM. Reaction supernatants were combined with the eluate from the wash of beads with 20 μl of reaction buffer. Reactions were either used fresh after preparation or frozen at -80 ⁰ C until use with purified proteasomes.

[0096] Construction ofN-End Rule Reporter Proteins: The CUPl promoter and the ubiquitin-Lys-LacI extension were amplified from the plasmid pUbl66 by polymerase chain reaction (PCR) using the primers YT 146 and YT 147. The resulting fragment was digested with the restriction endonucleases Sphl and HindIII and inserted into pUCl 19 to generate pYT602. A 2x-HA epitope was constructed using synthetic oligonucleotides bearing HindIII and Xhol restriction sites, and inserted by blunt-end ligation into the Smal site of pUC119 to generate pYT598. The Pstl/Hindlll fragment from pYT602 and the Hindlll/Xhol fragment from pYT598 were inserted into pBluescript-KS to yield pYT614. A ClonNAT cassette from pAG25 (Goldstein and McCusker. (1999). Three new dominant drug resistance cassettes for gene disruption in Saccharomyces cerevisiae. Yeast 15, 154 1-53) was then inserted N- terminally at the Notl site yielding pYT616. This plasmid contains the sequence ClonNAT- [P-Cupl]-Ub-K-LacI-2xHA.

[0097] For the TRPl fusion construct, a 100-base pair (bp) region from 5'- untranslated region of the gene immediately adjacent to the open reading frame start site was amplified from YDp-W (Berben, et al. (1991). The YDp plasmids: a uniform set of vectors bearing versatile gene disruption cassettes for Saccharomyces cerevisiae. Yeast 7, 475-7) by PCR using the primers YT169 and YT180; a 520 bp region from the 5' end of the TRPl coding region was amplified from the plasmid YDp-W using YT 164 and YTl 65. Utilizing PCR with the primers YTl 69 and YTl 65, these two fragments were co-amplified with the ClonNAT-[P-Cupl]-Ub-K-LacI- 2xHA fragment, positioning the 5'-untranslated region of TRPl at the 5' end of the fragment, and the TRPl ORF region immediately downstream of and in frame with the HA epitope. The URA3 fusion construct was assembled analogously, using the primers YTl 58 and YTl 79 (for the 5' untranslated region of URAS) and YTl 60 and YT161 (for the ORF of the URAS gene), and YDp-U (Berben, et al. (1991). The YDp plasmids: a uniform set of vectors bearing versatile gene disruption cassettes for Saccharomyces cerevisiae. Yeast 7, 475-7) as the source of the URA3 DNA. [0098] Strains sYT234 and sYT235 were constructed by transforming SUB62 with wildtype DNA from URA3 and TRPl, respectively, and selecting for colonies that were

either TRP ⁺ or URA ⁺ . Strains bearing the Trpl or Ura3 reporter proteins were generated by homologous recombination at the endogenous locus by transformation with the appropriate PCR fragment. Positive transformants were verified by PCR.

[0099] Quantitation ofProteasome Inhibition: A standard substrate degradation assay was carried out, and analyzed in parallel by SDS-PAGE followed by Coomassie blue staining and immunoblot with anti-cyclin B antibody. Immunoblots were used to identify and excise the cyclin B-containing regions of the Coomassie-stained gel, and the lanes were analyzed by mass spectrometry.

[00100] Ub-Cdc34 Binding Assay: Binding of ubiquitinated-Cdc34 (8 pmoles) to proteasome (4 pmoles) was carried by native gel electrophoresis followed by overlay of the fluorogenic activity-based probe, suc-LL VY-AMC, as previously described (Elsasser, et al. (2004). Rad23 and RpnlO serve as alternative ubiquitin receptors for the proteasome. J. Biol. Chem. 279, 268 17-26822). Ubp6 or Ubp6-Cl 18A (8 pmoles) were preincubated with proteasomes prior to substrate addition.

[00101] In Vivo Assays of Drug Sensitivity: Cultures were grown in synthetic media lacking tryptophan, normalized to an OD _6O o of 0.2, spotted in three-fold serial dilutions onto plates consisting of synthetic media supplemented with the appropriate drug as indicated, and incubated at 30 ⁰ C for 3-7 days. For ubiquitin overexpression, plates were supplemented with copper sulfate (100 μM) to augment plasmid expression.

[00102] Analysis of In Vivo Ubiquitin Turnover: Exponential phase cultures of the indicated strains bearing GAL-inducible UBP6 genes were grown in selective media containing 2% raffinose and 2% galactose. Cultures were normalized by OD _6O o and cycloheximide was added to a final concentration of 50 μg/ml. Ubiquitin depletion in the presence of UBP6 is minimized at this drug concentration, but increases at higher levels of cycloheximide (Hanna, et al. (2003). Ubiquitin depletion as a key mediator of toxicity by translational inhibitors. MoI. Cell. Biol. 23, 925 1-9261). Aliquots were taken at the indicated time points after cycloheximide addition and OD _6O o was measured to ensure that an equal number of cells was taken at each time point. Aliquots were processed and analyzed as previously described (Hanna et al., 2003).

[00103] Recombinant Uspl4: Recombinant GST-Uspl4 was expressed in E. coli

Rosetta cells (Novagen) and purified by glutathione-Sepharose affinity resins. The GST moiety was removed prior to the degradation assay, using the protease thrombin. [00104] Human proteasome: A stable 293 cell line expressing hRpnl 1-HTBH was used. Human proteasomes were purified as previously described with some modifications

(Wang X and Huang L (2007) Identifying dynamic interactors of protein complexes by quantitative mass spectrometry. MoI Cell Proteomics. 2008 Jan; 7(l):46-57. Epub 2007 Oct 12.

[00105] In vitro Ubn-cyclin Bl degradation assay: Assays were performed in proteasome preparation buffer (50 mM Tris [pH 7.5], 5 mM MgCl ₂ , and 5 mM ATP) at ambient temperature for the indicated times. The 26S human proteasomes (4 nM) were incubated with approximately 30 nM of Ub-conjugated cyclin Bl as substrates. Cyclin was ubiquitinated using purified APC ubiquitin ligase. Where indicated, a 15-fold molar access of recombinant Uspl4 over proteasomes were preincubated prior to adding Ubn-cyclin Bl.

Example 2: Ubp6 Inhibition of the Degradation of Ubiquitinated Cyclin B [00106] To study the potential role of Ubp6 in proteasome function, an in vitro system for proteasome-mediated degradation of a model substrate was developed. Ubiquitination of the short-lived cell cycle regulator cyclin B was achieved via a reaction requiring El, Ubc4 (E2), immunopurified APC (E3), ubiquitin, and ATP. Upon incubation of ubiquitinated cyclin B with proteasomes purified from ubpόδ mutants, it was observed that the rate of cyclin B degradation was much greater than that of wild-type proteasomes (Fig. IA). Significantly, the slower degradation of cyclin B exhibited by wild-type proteasomes was accompanied by progressive removal of ubiquitin groups from the substrate. The absence of Ubp6 in proteasomes purified from the ubpόδ strain (Fig. IB) was verified, and it was also verified that the amounts of proteasome were comparable between the two preparations (Fig. 1C).

[00107] The next studies were designed to determine whether the observed difference in cyclin B degradation was due specifically to Ubp6. Proteasomes were examined by Coomassie blue staining (Fig. ID) and native gel electrophoresis, and no evidence for either an unexpected compositional difference or a gross structural abnormality in mutant proteasomes was found (see also Guterman and Glickman. (2004). Complementary roles for Rpnl 1 and Ubp6 in deubiquitination and proteolysis by the proteasome. J. Biol. Chem. 279, 1729-1738). Furthermore, the activity of mutant proteasomes against the peptide substrate suc-LLVY-AMC, which is hydrolyzed in an RP-dependent but ubiquitin-independent manner, was comparable to wild-type (Fig. IE), indicating that inhibition of cyclin B degradation by Ubp6 was not due to suppression of the core proteolytic activity of the proteasome, nor to closing the gate into the CP. Comparable Ubp6 add-back results were obtained with ubpόδ proteasomes purified using a CP affinity tag. Finally, when bacterially-

expressed purified Ubp6 was added back to proteasomes lacking Ubp6, a marked inhibition of cyclin B degradation (Fig. IF) was observed, indicating that Ubp6 itself is the inhibitory component of wild-type proteasomes. Inhibition of degradation by recombinant Ubp6 (Fig. IF) was accompanied by progressive ubiquitin chain shortening, as seen with wild-type proteasomes (Fig. IA).

[00108] Rpnl 1 is known to promote deubiquitination and degradation by the proteasome (Verma, et al. (2002). Role of Rpnl 1 metalloprotease in deubiquitination and degradation by the 26S proteasome. Science 298, 611-6 15; Yao and Cohen. (2002). A cryptic protease couples deubiquitination and degradation by the proteasome. Nature 419, 403-407). To verify that Rpnl 1 was active in the system used herein, ubpόδ proteasomes were used with the metal chelator o-phenanthroline (o-PA), an inhibitor of Rpnl 1. o-PA strongly inhibited deubiquitination and degradation of cyclin B (Fig. 1 G). In contrast, o-PA treatment had no effect on LLVY-hydrolysis by proteasomes. Thus, in contrast to Ubp6, Rpnl 1 promotes cyclin B degradation.

Example 3: In vitro Confirmation of Proteasome Inhibition by Ubp6 [00109] Chemical inhibitors of the proteasome were used to confirm that the rapid disappearance of cyclin B in the absence of Ubp6 represents cyclin B degradation by proteasomes as opposed to a possible unknown component of the samples. For these experiments, epoxomicin, an inhibitor of the proteolytic active sites of the CP, was used. Epoxomicin preferentially targets the chymotrypsin-like activity of the CP; at high concentrations, such as used here, epoxomicin can inhibit all three proteolytic sites, although complete proteasome inhibition is generally not observed (Kisselev, et al. (2006) Importance of the proteasome's different proteolytic sites and the efficacy of inhibitors varies with the protein substrate. J. Biol. Chem. 257,8582-90. Epub 2006 Feb 2). The yield of deubiquitinated cyclin B reaction products from ubpόδ proteasomes after a ten-minute incubation was greatly increased by epoxomicin, verifying that disappearance of cyclin B involved proteasome-mediated degradation (Fig. 2A). A diminished effect of epoxomicin was observed when the same experiment was carried out using proteasomes containing Ubp6, reflecting a reduced rate of cyclin B breakdown in the absence of epoxomicin (Fig. 2A). These Ubp6-dependent differences could be recapitulated by endogenous Ubp6 present in wild-type proteasomes (Fig. 2B). [00110] The cyclin B used in the in vitro assay described herein may be linked through not only lys-48 (K48) of ubiquitin but also Kl 1 and K63. K48 of ubiquitin was not required

for in vitro degradation of ub-cyclin B. A side-by-side comparison of wild-type and ubpόδ proteasomes; in both cases is provided herein, proteasomes deubiquitinate and degrade cyclin B without regard for the presence of K48 ubiquitin linkages in its attached ubiquitin chains (Fig. 2C). Thus, although K48-linked ubiquitin chains may be the dominant chain type promoting degradation in vivo, this linkage is not required for rapid proteasomal degradation in vitro. For many substrates, the requirement for K48 in protein degradation may reflect properties of the E3 enzymes involved in chain synthesis.

[00111] Finally, for reactions lacking Ubp6, epoxomicin stabilized, in addition to unmodified cyclin B, a second major species, which is apparently mono-ubiquitinated cyclin B (Fig. 2A-C). These results suggest that in the absence of Ubp6, while a majority of ub- cyclin B is fully deubiquitinated by Rpnl 1 prior to degradation, a significant fraction is left incompletely deubiquitinated, and the resulting mono-ubiquitinated species is apparently rapidly degraded. In vivo, ubiquitin undergoes accelerated degradation by the proteasome in the absence of Ubp6 (Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell. 10, 495-507; Hanna, et al. (2003). Ubiquitin depletion as a key mediator of toxicity by translational inhibitors. MoI. Cell. Biol. 23, 925 1-9261; Chernova, et al. (2003). Pleiotropic effects of Ubp6 loss on drug sensitivities and yeast prion are due to depletion of the free ubiquitin pool. J. Biol. Chem. 275, 52102-521 15; Fig. 5D). The assignment of mono-ubiquitinated cyclin B is consistent with previous studies. It was proposed that when Ubp6 is absent, one or more substrate-bound ubiquitin groups are translocated into the CP and degraded along with their substrate (see Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell. 10, 495- 507), but until now there had been no biochemical basis to account for accelerated ubiquitin turnover. It was proposed that increased turnover of mono-ubiquitinated degradative intermediates as suggested by Fig. 2 may contribute significantly to the rapid degradation of ubiquitin in vivo in the absence of Ubp6.

Example 4: Ubp6 Impairs Proteasome-Mediated Degradation in vivo [00112] The in vitro data predicted accelerated degradation of at least some proteasomal substrates in vivo in the absence of Ubp6. To test this idea, chromosomal integration was used to render the biosynthetic enzymes Trp 1 and Ura3 unstable by appending an N-terminal segment that directs these proteins to the N-end rule pathway for degradation (Varshavsky. (2005). Regulated protein degradation. Trends Biochem. Sci. 30,

283-286.). Accordingly, growth of such strains in the absence of the relevant metabolite

should reflect the rate of degradation of the respective reporter protein. Strains harboring proteolytic defects should stabilize the reporters, and thus display increased growth relative to wild-type. Ubrl, the E3 of the N-end rule pathway, governs ubiquitination of the reporter proteins. As expected, the ubrlδ mutant displayed a strong growth advantage over wild-type when cultured on media lacking tryptophan (Fig. 3A) or uracil.

[00113] Although such assays have typically been used to characterize degradation defects, it was reasoned that if significant growth were detectable in wild-type strains, the assay could also be used to identify mutants with increased degradation rates. Indeed, when ubpόδ mutants were tested, they displayed a dramatic growth defect relative to wild-type (Fig. 3A), consistent with a proteasome hypermorphic effect. Additionally, a ubrlδ ubpόδ double mutant retained the robust growth phenotype of the ubrlδ single mutant, ruling out a protein synthesis defect as the cause of the poor growth of the ubpόδ strain. Ura3 reporter strains that are wild-type for UBP6 showed no growth in the absence of uracil, precluding the corresponding comparison.

[00114] The results from Fig. 3A suggested that the reporter proteins were turned over more rapidly in the ubpόδ strain. To test this idea directly, cycloheximidechase analyses were conducted in strains harboring both reporter constructs. In the wild-type strain rapid degradation of both reporters was observed, with half-lives on the order of several minutes (Fig. 3B). In the ubrlδ mutant, both reporters were highly stabilized (Fig. 3B). In contrast, in the ubpόδ mutant, the Ura3 reporter disappeared more rapidly than in wild-type, and showed a decreased steady-state level (Fig. 3B). Steady-state levels of the Trpl reporter were also depressed in the ubpόδ mutant (Fig. 3B), falling below the level of detection. The level of a control protein, proteasome subunit Rpn5, remained constant throughout, indicating the specificity of these effects.

[00115] In addition to N-end Rule substrates, the short-lived transcription factor Gcn4, expressed from its own promoter in untagged form, was evaluated. Again, accelerated degradation in the ubpόδ mutant was observed, with a concomitant decrease in steady-state levels (Fig. 7). Thus, multiple short-lived proteins are turned over more rapidly in vivo in the absence of Ubp6.

Example 5: The Deubiquitinating Activity of Ubp6 is not Required for Proteasome Inhibition

[00116] To determine which functional elements of Ubp6 are required for inhibiting proteasome function, a number of Ubp6 mutants were generated and tested (Fig. 4A). An N-

terminal ubiquitin-like domain (UbI) is necessary and sufficient to mediate binding to the proteasome, but is not required for the catalytic activity of purified Ubp6 (Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell. 10, 495-507). Deletion of the UbI domain rendered Ubp6 non-inhibitory in the cyclin B degradation assay (Fig. 4B), suggesting that Ubp6 must be bound to the proteasome to inhibit cyclin B degradation. The UbI domain by itself also had no effect on cyclin B degradation (Fig. 4E). Thus, the ability of Ubp6 to inhibit degradation of cyclin B jointly requires the UbI and C-terminal domains of the protein.

[00117] Although efficient degradation of ubiquitin conjugates by the proteasome is thought to require removal of ubiquitin groups prior to translocation of the substrate into the CP, premature removal of ubiquitin chains could result in dissociation of the substrate and antagonize its degradation (Lam, et al. (1997). Editing of ubiquitin conjugates by an isopeptidase in the 26S proteasome. Nature 355, 737-740). Such a scenario could provide a straightforward means by which deubiquitinating activity at the proteasome could inhibit protein breakdown. The joint requirements of the UbI and C-terminal domains of Ubp6 for inhibiting cyclin B degradation are consistent with this view. However, a catalytically inactive point mutant of Ubp6 (Ubp6-C 118A) proved competent for inhibition (Fig. 4B). Inhibition of degradation by Ubp6-C 118A was also observed using proteasomes purified via a CP affinity tag. Thus, Ubp6 does not prevent cyclin B degradation simply through premature deubiquitination. To substantiate this conclusion, it was verified that Ubp6-C 118A is completely devoid of deubiquitinating activity against the model substrate Ub-AMC (Fig. 8), but retains wild-type binding affinities for the proteasome (Chernova, et al. (2003). Pleiotropic effects of Ubp6 loss on drug sensitivities and yeast prion are due to depletion of the free ubiquitin pool. J. Biol. Chem. 278, 52102-521 15; Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell. 10, 495-507). [00118] As an alternate approach to assay non-catalytic inhibition, ubiquitin-vinyl sulfone (Ub-VS), an irreversible active site inhibitor of Ubp6, was used. Residual Ub-AMC hyrdolyzing activity in the presence of Ubp6-Ub-VS is at the level of background in this assay. Although it completely eliminated catalytic activity of wild-type Ubp6 as measured by Ub-AMC hydrolysis, inhibition of cyclin B degradation by Ubp6 persisted (Fig. 9). This finding also indicates that inhibition of degradation by Ubp6-Cl 18A cannot be attributed to nonproductive ubiquitin chain binding at the active site of the mutant enzyme. Indeed the large excess of substrate over Ubp6 (see Methods described in Example 1) is inconsistent with a model in which Ubp6 inhibits degradation by sequestering substrate. Taken together,

these data show that Ubp6 has two distinct functions, one catalytic and one non-catalytic. Although these functions are distinct, each requires both the UbI domain and the C-terminal domain. The UbI domain serves in both cases to target Ubp6 to the proteasome, and that both functions are obligatorily executed on the proteasome.

[00119] Although Ubp6-C 118A was, like wild-type Ubp6, competent to inhibit cyclin

B degradation, the fate of cyclin B over the time course of the incubation was affected by the loss of deubiquitinating activity in the mutant enzyme. When wild-type Ubp6 was used for reconstitution, a progressive reduction in electrophoretic mobility of cyclin B immunoreactive material was seen, whereas these mobility shifts were not observed with Ubp6-Cl 18A (Fig. 4B). These data imply that the mobility shifts represent trimming of ubiquitin or ubiquitin chains bound to cyclin B, and that the trimming reaction is mediated predominantly or exclusively by the deubiquitinating activity of Ubp6. [00120] To quantify the extent of proteasome inhibition by Ubp6, a method of quantitative mass spectrometry was utilized. After a 20-minute incubation, approximately half of the total cyclin B was degraded by ubpόδ proteasomes (Fig. 4C). High molecular weight ubiquitin-cyclin B conjugates were almost completely eliminated by this time, but degradation would not be expected to go to completion because a fraction of the input cyclin B had not been ubiquitinated. Addition of purified Ubp6 or Ubp6-C 118A resulted in approximately 70% inhibition of degradation at this time point (Fig. 4C), which may underestimate the Ubp6 inhibitory effect, since the course of degradation seemed to be nearing completion in the control sample.

Example 6: Ubp6 Prevents Rpnll -Dependent Ubiquitin Chain Removal

[00121] The linkage of Rpnl 1 activity to substrate degradation can be abrogated by proteasome inhibitors, presumably because they act downstream of Rpnl 1 (Verma, et al. (2002). Role of Rpnl 1 metalloprotease in deubiquitination and degradation by the 26S proteasome. Science 298, 611-6 15; Yao and Cohen. (2002). A cryptic protease couples deubiquitination and degradation by the proteasome. Nature 419, 403-407). Thus, to visualize the activity of Rpnl 1 on cyclin B, proteasome inhibitor treated ubpόδ proteasomes were used, as in Fig. 2. This figure showed, as described above, that the effect of Ubp6 on cyclin B disappearance is mediated by an altered rate of proteasome-mediated degradation. However, since the results of Fig. 4 indicate that the deubiquitination seen during degradation delay is mediated by Ubp6, and not by Rpnl 1 , Ubp6 inhibits the proteasome at a point in the reaction pathway such that Rpnl 1 -mediated chain removal is prevented.

[00122] In the presence of epoxomicin, Rpnl 1 can be seen to cleave substrate-linked chains at or near their base, resulting in the production of a prominent band of unmodified substrate protein (Verma et al., 2002; Yao et al., 2002; Fig. 2). When the reaction products from wild-type and ubpόδ proteasomes pre-treated with epoxomicin were compared, a greater amount of unmodified cyclin B was observed in ubpόδ reactions, reflecting substrate deubiquitinated but not degraded by the proteasome (Fig. 2A-C). In the presence of Ubp6, a lesser amount of unmodified cyclin B was stabilized by the inhibitor; instead, the majority of the cyclin B remained as higher molecular weight species, most likely reflecting partial deubiquitination. The increased yield of unmodified cyclin B in the presence of ubpόδ proteasomes as compared to wild-type indicates that Rpnl 1 -dependent chain removal from substrate is suppressed in the presence of Ubp6. However, it remains unclear whether Ubp6 directly inhibits Rpn 11 or some other activity that may function upstream of Rpnl 1.

Example 7: Interference with Substrate-Proteasome Interaction Cannot Explain Proteasome Inhibition by Ubp6

[00123] One explanation for the degradation-inhibitory effect of Ubp6 is that binding of Ubp6 to the proteasome is competitive with that of substrate. Several lines of evidence indicate that Ubp6 does not inhibit the proteasome by this mechanism. As seen in Fig. 4B, the deubiquitination of cyclin B that occurs in the complete reaction mixture is mediated by Ubp6. However, when Ubp6 is incubated in the presence of cyclin B, but without proteasomes, no deubiquitination of cyclin B is observed (Fig. 4D). Thus, Ubp6 requires the proteasome for its deubiquitinating activity on cyclin B. These data suggest that both Ubp6 and cyclin B are bound to proteasomes so long as cyclin B deubiquitination proceeds, which can be an extended period, since the deubiquitination reaction is progressive over the time courses shown in Fig. 4.

[00124] Ubp6 could conceivably inhibit cyclin B degradation via a substrate-titration mechanism. However, direct interaction between Ubp6 and ub-cyclin B were not able to be observed using a GST pull down assay that sensitively detects binding of ub-cyclin B to ubiquitin receptors such as RpnlO, Rad23, and Dsk2 (Fig. 10). These findings are consistent with a previous report that the KM of Ubp6 for ubiquitin ethyl ester is so high as to be unmeasurable (Chernova, et al. (2003). Pleiotropic effects of Ubp6 loss on drug sensitivities and yeast prion are due to depletion of the free ubiquitin pool. J. Biol. Chem. 278, 52102-521 15).

[00125] Ubp6 is large enough, at 57 kDa, that its presence on the proteasome could conceivably impede binding of substrates or substrate receptors. Also, the pathway of substrate translocation through the proteasome might be blocked by non-specific steric interference. To address this possibility a fusion protein in which the UbI of Ubp6 was fused N-terminally to the maltose binding protein of E. coli was constructed (Fig. 4A), resulting in a fusion protein of nearly the same size as Ubp6. Despite binding the proteasome, this construct, as for the UbI alone, failed to inhibit cyclin B degradation (Fig. 4E). [00126] Finally, the effect of Ubp6 on conjugate binding to the proteasome by using auto-ubiquitinated Cdc34, a substrate that binds proteasomes but is not degraded, was directly evaluated (Elsasser, et al. (2004). Rad23 and RpnlO serve as alternative ubiquitin receptors for the proteasome. J. Biol. Chem. 279, 268 17-26822). Using an activity-based electrophoretic mobility shift assay (Elsasser et al., 2004), comparable conjugate binding was observed in the presence and absence of Ubp6- CI l 8A (Fig. 4F). Conjugate binding was observed in the presence of wild-type Ubp6, but was also accompanied by deubiquitination, as determined by immunoblot. Taken together, the data of Figs. 4E and 4F indicate that non- catalytic proteasome inhibition by Ubp6 is unlikely to be achieved through simple sterically- based mechanisms, but instead may involve specific functionalities within the C-terminal domain of Ubp6.

Example 8: In vivo Differentiation of ubpόA and ubp6-C118A mutants [00127] The next studies were designed to generate in vivo evidence for the non- equivalence of the null and catalytic point mutants of Ubp6. A number of ubpόδ phenotypes that can be suppressed by ubiquitin overexpression have been described (Chemova et al., 2003; Hanna, et al. (2003). Ubiquitin depletion as a key mediator of toxicity by translational inhibitors. MoI. Cell. Biol. 23, 925 1-9261), including hypersensitivity to the amino acid analog canavanine (Fig. 5A) and to several translational inhibitors. Intriguingly, it was found that not all ubpόδ phenotypes are ubiquitin dependent. For instance, a strong resistance of ubpόδ to rapamycin was observed, which was not suppressed by ubiquitin overexpression (Fig. 5B). This result suggests functions of Ubp6 outside of ubiquitin regeneration, possibly reflecting noncatalytic functions of Ubp6. To test this hypothesis, wild-type Ubp6 and Ubp6- Cl 18A from its endogenous promoter were expressed in a ubpόδ strain. Wild-type Ubp6 complemented both the canavanine and rapamycin phenotypes (Fig. 5C). As expected, expression of Ubp6-Cl 18A failed to rescue canavanine hypersensitivity (Fig. 5C; Leggett, et al. (2002). Multiple associated proteins regulate proteasome structure and function. MoI. Cell.

10, 495-507), but in contrast largely complemented rapamycin resistance (Fig. 5C). Ubp6- ublA recapitulated the null phenotype, suggesting that the rapamycin-related non-catalytic functions of Ubp6 occur on the proteasome (Fig. 11). These results provide in vivo evidence that the ubpόA and ubp6-Cl 18A mutants are not equivalent, and support the hypothesis that Ubp6 performs non-catalytic functions, as observed in vitro.

[00128] In the absence of Ubp6, free ubiquitin is rapidly turned over (Fig. 5D; Leggett et al., 2002). The failure of Ubp6-Cl 18A to complement ubiquitin-dependent phenotypes such as canavanine hypersensitivity suggested aberrant ubiquitin turnover in ubp6-C118A mutants. Overexpression of wild-type Ubp6, but not Ubp6-Cl 18A, restored ubiquitin stability (Fig. 5D), indicating that the ubiquitin recycling function of Ubp6 is catalytic in nature. These results provided a mechanistic basis for understanding why the Ubp6-Cl 18A mutation affects ubiquitin-dependent but not ubiquitin-independent phenotypes, and indicated that the non-catalytic function of Ubp6 is distinct from its function in ubiquitin homeostasis. [00129] Finally, the Ub-K-Tφ 1 substrate described in Figure 3 was used to test whether non-catalytic inhibition of degradation by Ubp6 was also operational in vivo. As shown in Figure 5E, expression of Ubp6-Cl 18A restored growth on media lacking tryptophan, indicating stabilization of the Ub-K-Trpl protein. Importantly, wild-type Ubp6 was more effective in stabilizing Ub-K-Trpl than was Ubp6-Cl 18A. Thus both catalytic and non-catalytic features of Ubp6 mediate proteasome inhibition in vivo.

Example 9: Regulatory Particle is Sufficient For Deubiquitination by Ubp6 but not Rpnll

[00130] Rpnl 1 has been reported to function in the context of the 26S proteasome, but not in the context of the RP or the lid (Verma, et al. (2002). Role of Rpnl 1 metalloprotease in deubiquitination and degradation by the 26S proteasome. Science 298, 611-6 15), although there have been differing reports (Yao and Cohen. (2002). A cryptic protease couples deubiquitination and degradation by the proteasome. Nature 419, 403-407; Guterman and Glickman. (2004). Complementary roles for Rpnl 1 and Ubp6 in deubiquitination and proteolysis by the proteasome. J. Biol. Chem. 279, 1729-1738). In the assay used herein, RP purified from a ubpόδ mutant produced no detectable deubiquitination of eye Hn B conjugates (Fig. 6A). In contrast, addition of purified Ubp6 to RP resulted in chain shortening, although to a lesser extent than with 26S proteasomes (Fig. 6A). Addition of Ubp6-Cl 18 A to RP had no effect on substrate processing (Fig. 12), further verifying that ubiquitin chain shortening seen with wild-type Ubp6 reflects the catalytic activity of Ubp6 itself. The cyclin B

immunoreactivity before and after Ubp6 processing was of approximately equal intensity, consistent with the view that Ubp6 shortens ubiquitin chains, but rarely produces unmodified substrate as does Rpnl 1.

[00131] The fate of ubiquitin in the presence of RP-bound Ubp6 was examined by anti-ubiquitin immunoblots. A decrease in the overall size and intensity of the high molecular weight material was found, consistent with the data from Fig. 6A, but reaction products were also observed co-migrating with di-ubiquitin, tri-ubiquitin, and larger ubiquitin polymers (Fig. 6B). Presumably, mono-ubiquitin was also generated, although this was difficult to discern because of the high levels of free ubiquitin present in the reactions. In reactions utilizing lower amounts of substrate, Ubp6-dependent generation of free ubiquitin was readily observed (Fig. 12). These results suggest a preference of Ubp6 for the distal end of ubiquitin chains (see also Hu, et al. (2005). Structure and mechanisms of the proteasome- associated deubiquitinating enzyme USP14. EMBO J. 24, 3747-3756); alternately, Ubp6 may cleave proximally to remove mono-ubiquitin and small ubiquitin chains. A third deubiquitinating enzyme, Uch37, which is absent from budding yeast, has been reported to possess a distal-end-directed activity on the proteasome (Lam, et al. (1997). Editing of ubiquitin conjugates by an isopeptidase in the 26S proteasome. Nature 355, 737-740). However, whereas Uch37 appears to remove ubiquitins one at a time, shortening of chains by Ubp6 is not limited to the trimming of single ubiquitins.

Example 10: The In vitro Functions of Ubp6 are Evolutionarily Conserved

[00132] Ubp6 and its human homolog Uspl4 share 32% sequence identity (Chernova, et al. (2003). Pleiotropic effects of Ubp6 loss on drug sensitivities and yeast prion are due to depletion of the free ubiquitin pool. J. Biol. Chem. 278, 52102-521 15). Several features of Ubp6, including proteasome binding and activation by proteasome binding, are shared by Uspl4 (Borodovsky, et al. (2001). A novel active site-directed probe specific for deubiquitylating enzymes reveals proteasome association of USP 14. EMBO J. 20, 5187- 5196). Indeed, overexpressed Uspl4 complements several in vivo phenotypes oϊubpόδ in yeast (Chernova et al., 2003). It was therefore tested whether purified recombinant Uspl4 could substitute for Ubp6 in the cyclin B degradation assay, using ubpόδ proteasomes from yeast. Uspl4 largely recapitulated the effects of Ubp6 (Fig. 6C): ub-cyclin B was not only stabilized against degradation by Uspl4, but progressively deubiquitinated to lower molecular weight forms. Uspl4 is slightly less efficient than Ubp6 in this assay, perhaps reflecting a

decreased affinity of Uspl4 for yeast proteasomes compared to Ubp6. These results, in combination with the reported in vivo complementation of ubpόA by Uspl4, indicate that the observed effects of Ubp6 in this reconstituted system are likely to represent major functions of Ubp6 in vivo.

Example 11: Inhibition of human proteasome by Uspl4: dominant catalytic component [00133] In the study shown in Figure 14, 4 nM of human proteasomes were incubated with approximately 30 nM of ubiquitinated cyclin B in buffer (50 mM Tris [pH 7.5], 5 mM MgC12, and 5 mM ATP). Where indicated, 15-fold molar access of Uspl4 over proteasome (60 nM) was preincubated for 5 min before initiating degradation reactions. Ubiquitinated cyclin B was added at time zero. SDS-sample buffer was added to stop the reaction at the indicated time. Uspl4 CA is a variant of Uspl4 with a single amino acid substitution that eliminates the active site nucleophile of the enzyme (a cysteine to alanine replacement). Uspl4 WT is the wild-type or unmutated, catalytically active form of the enzyme. In the three lanes at left, no Uspl4 was added. The data show that the ability of Uspl4 to remove ubiquitin chains from the substrate is critical to its suppression of cyclin B degradation, which applies to other proteasome substrates. These findings indicate that small-molecule inhibitors of the catalytic activity of Uspl4 accelerate proteasome action on ubiquitinated substrates.

[00134] The invention having now been described by way of written description and example, those of skill in the art will recognize that the invention can be practiced in a variety of embodiments and that the description and examples above are for purposes of illustration and not limitation of the following claims.