Index INTRODUCTION MATERIALS AND METHODS Microbial techniques and DNA engineering. Construction of tandem ubiquitin genes. Construction of RN3 in vitro transcription vectors. Purification of the 26S proteasome (19). Preparation of Poly-Ub-Sic1PY. Western blotting. Expression of tUB in Xenopus embryos. RESULTS AND DISCUSSION Construction of the genes encoding polyubiquitin insensitive to ubiquitin hydrolases. tUb inhibits proteolysis by the ubiquitin proteasome system. tUb forms a complex with the 26S proteasome. tUb inhibits 26S proteasome activity in vitro. tUB inhibits Xenopus egg cleavage. References

INTRODUCTION

The 26S proteasome is a protease of a molecular size of about 2,000kDa and is well conserved among eukaryotes. The 26S proteasome consists of two subcomplexes: the 20S proteasome and the 19S regulatory particle (RP) (Coux et al. 1996). The 20S proteasome consists of 4 rings (2 α rings and 2 β rings in the order of α β β α) that are stacked to form a cylinder-like structure (Kopp et al. 1997). Each α and β ring is composed of 7 closely related subunits. The 20S proteasome is a catalytic subunit in which active sites for protease are sequestered inside the lumen (Groll et al. 1997). The RP attaches to both ends or one of the ends of the 20S proteasome to form the 26S proteasome. The functions of the RP are believed to be to recognize and bind ubiquitinylated proteins, to unfold the proteins, and to deliver them to the catalytic center. The subunit composition of the RP has been elu-cidated (Glickman et al. 1998a). The RP comprises two subcomplexes: the base consisting of six ATPases (Rpt1~Rpt6) and two non ATPase subunits (Rpn1 and Rpn2) and the lid consisting of non-ATPase subunits (Rpn3, Rpn5~Rpn12) (Glickman et al. 1998b).

The 26S proteasome acts as degradation-machinery in the ubiquitin-proteasome system (Hershko and Ciechanover 1992). In this system, proteins to be degraded are tagged with ubiquitin at a target lysine (K) residue by an isopeptidyl bond via three successive reactions catalyzed by E1, E2 and E3 enzymes. Further ubiquitins attach to a lysine residue (K) inside of the ubiquitin to produce a polyubiquitin chain that serves as a recognition signal for degradation by the 26S proteasome. There are seven lysine sites in ubiquitin at which ubiquitin forms a polyubiqutin chain by an isopeptide linkage with the C-terminal glycine (G) of ubiquitin (Peng et al. 2003). Among polyubiquitin chains, K48 polyubiquitin chains (polyubiquitinylated at the K48 of ubiquitin) serve as a signal for degradation by the 26S proteasome (Chau et al. 1989).

Pickart and her colleagues (Piotrowski et al. 1997) devised a method to produce a polyubiquitin chain (K48) with a defined length by exploiting E1 enzyme and E2-25K enzyme. They clearly showed that tetra ubiquitin is the shortest polyubiquitin chain that is targeted to the 26S proteasome. They found an inhibitory effect of 5xUb encoded by the UBI4 gene of Saccharomyces cerevisiae on the 26S proteasome activity; however, more rigorous experiments will be necessary to understand this phenomenon. The structure of 5xUb seems likely to be more like that of K63 polyubiquitin, which does not serve as a degradation signal in vivo

Proteolysis by the 26S proteasome regulates various biological phenomena, such as cell cycle progression, apo-ptosis, inflammatory reactions, etc. (Hershko and Ciechanover 1998; Glickman and Ciechanover 2001). To demonstrate the involvement of the 26S proteasome in a certain phenomenon, a specific inhibitor of proteasome activity is widely used (Rock et al. 1994). If the inhibitor blocks or increases development of the phenotype, then the proteasome is inferred to participate in the phenomenon.

Several types of low molecular weight inhibitors of the proteasome have been identified (Tsubuki et al. 1992; Lee et al. 1998). The most widely used are peptide aldehydes, such as CDZ-Leu-Leu-Leucinal (MG132). This inhibitor is incorporated into the lumen of the 20S proteasome and makes a transition state of the protease enzymatic activity. The inhibition is reversible. Lactacystin (Omura et al. 1991) inhibits the proteasome irreversibly by covalently linking with the hydroxyl groups of the active site threonine (Fenteany 1995). This inhibitor shows high specificity toward the proteasome. Another inhibitor functioning by a similar mechanism to that of lactacystin is vinylsufone (Bogyo et al. 1997). All these inhibitors are valuable tools for revealing the proteasome participation in various phenomena in cultured cells or tissues. However, it is impossible to deliver these drugs to a specific site or tissue or organ to inhibit the proteasome activity at a targeted location.

Here we devised gene constructs encoding ubiquitin-hydrolase-insensitive tandem ubiquitins consisting of 2~ 8 units of a head-to-tail ubiquitin fusion and found that these ubiquitin derivatives are strong inhibitors of the 26S proteasome.

MATERIALS AND METHODS

Microbial techniques and DNA engineering.

Yeast strains used in this study were isogenic to W303D (MATα/MATa/ leu2/leu2 his3/his3 trp1/trp1 ura3/ura3 ssd1/ssd1 ade2/ade2 can1/can1). Culture conditions and methods of genetic analysis were described previously (Sherman 1991). For affinity purification of the 26S proteasome, RPN11-FLAG epitope-tagged strain YYS40 (MATa leu2 his3 trp1 ura3 ssd1 ade2 can1 rpn11::RPN11-PFT-HIS3) was constructed as described previously (Saeki et al. 2003). For expression of hexahistidine tag (His₆)-fused tUb4 in E.coli, the tUB4 gene was inserted into the pQE vector (QIAGEN).

Construction of tandem ubiquitin genes.

Polymerase chain reaction (PCR) was carried out by using a set of primers (#1-EcoRI: 5’-GGAAGAATTCATGCAAATTTT-CGTCAAAACTCTAAAGGG-3’, #2-Xho1: 5’-GGAACTCGAGCAGCCCTCAACCTCAAGACAAGG-3’) and W303D genomic DNA as template. By this reaction, a series of ubiquitin multimer genes were amplified from the UBI4 gene. The DNA segment containing two tandem ubiq-uitin units was excised from a 1% agarose gel after electrophoresis. In this diubiquitin, the 75G 76G of the most downstream ubiquitin were changed to A’s. The nucleotide sequence of this diubiquitin gene was determined to confirm the correct PCR amplification. Next, the G residue in the 76G-1M linkage, the joint of the C-terminal glycine residue of the 1^st ubiquitin and 1^st methionine residue of the 2^nd ubiquitin, was changed to V by PCR mutagenesis using two sets of primers, (#1-EcoRI, #4: 5’-GACAAAAATTTGCATAACACCTCTCAGTC-3’) and (#2-XhoI, #3: 5’-GACTGAGAGGTGTTATGCAAA-TTTTTGTC-3’) and the diubiquitin gene as template. The mutant diubiquitin gene, designated tUB2, thus obtained was cloned into the EcoRI-XhoI gap of Bluescript KS+ (Stratagene). The tUB2 gene was amplified by PCR using a set of primers (#5-BamHI: 5’-GGAAG-GATCCATGCAAATTTTCGTCAAAACTCTAACAGGG-3’ and #6-MunI: 5’-GGAACAATTGAGCAGCCCTCAACCTCAAGACAAGG-3’) and the tUB2 gene as template. The resultant DNA segment was joined with the tUB2 gene to generate the tUB4 gene. The tUB2 gene and tUB4 genes were amplified by PCR using a set of primers (#1-EcoRI and #7-BglII: 5’-AAGGAGATCTAGCAGCCCTCAACCTCAAGACAAGG-3’) and the product was joined with tUB4 to generate the tUB6 gene and tUB8 gene, respectively. The terminal restriction sites of the tUB4 gene described above were changed to be compatible for the subsequent cloning experiments. Mono-ubiquitin and tUB3 genes were constructed by PCR using a set of primers (#1-EcoRI, #2-XhoI) and the tUB4 gene as template. DNA segments corresponding to mono- and tri-ubiquitin genes were excised from a 1% agarose gel. Each of the UB ~ tUB8 genes was inserted into the EcoRI-XhoI gap of pKT10(GAL1) (Kawamura et al. 1996). pEK206 containing tUB6-HA was constructed. tUB6-HA was found to be equivalent to tUBI6 with respect to the ability to inhibit growth.

Construction of RN3 in vitro transcription vectors.

To produce translatable tUB messenger RNA, tUB6 and tUB6-HA and monomer ubiquitin genes were separately inserted into an RN3 vector (Lemaire 1995) in an appropriate direction.

Purification of the 26S proteasome (19).

Strains carrying RPN11-FLAG were grown to OD600nm=1.0 in SRaf-Ura medium (raffinose was used as the sole carbon source), and then galactose was added and the cells were further cultured for 6 hr at 25°C. Cells were collected, washed twice with water and once with buffer A (50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 10% glycerol) and stored at –80°C until use. One gram of cells were resuspended in 1 ml of buffer B (buffer A containing 4 mM ATP, 10 mM MgCl₂, and 2 x ATP regenerating system) and lysed by mixing with glass beads. The extract was collected and supplemented with an equal volume of buffer B and centrifuged at 15,000 rpm (TOMY MX 100) for 20 min at 4°C. A slurry of M2-beads (anti-Flag antibody-coated beads) (80 μl) was added to the supernatant and the mixture was rotated for 1 hr at 4°C. Then, the beads were washed five times with 1 ml of proteasome buffer (buffer A containing 2 mM ATP and 5 mM MgCl₂) and two times with 1 ml of proteasome buffer containing 0.2% Triton X-100, and washed again two times with 1 ml of proteasome buffer. Then, the beads were collected into a mini-column and incubated with 100 μl of 3 x Flag peptide (200 μg/ml) in proteasome buffer for 30 min at 4°C. Purified 26S proteasomes were divided into small aliquots and stored at –80°C.

Preparation of Poly-Ub-Sic1^PY.

To prepare polyubiquitinylated Sic1, we developed a novel method using Rsp5 as E3 ligase. Details of this method will be published elsewhere (Y. S., manuscript in preparation). In brief, the PY motif was introduced into the N-terminal region of Sic1, designated as Sic1^PY, to allow Rsp5 binding. Sic1 and Sic1^PY were expressed as fusion proteins with T7-Sic1-HAT, a T7-tag at the N-terminus and a HAT-tag (Clontech) at the C-terminus, using the pET system (Novagen). Sic1 and Sic1^PY proteins were purified accor-ding to the method described by Verma et al. (2002). Ubc4 and Rsp5 were expressed as GST-fusion proteins in E. coli, and purified using glutathione-immobilized beads (Amersham). The GST-tag was removed using PreScission protease (Amersham) and untagged Ubc4 and Rsp5 were recovered. Uba1 was purified from S. cerevisiae by our unpublished method. Sic1 or Sic1^PY was incubated with Uba1, Ubc4, Rsp5, and ubiquitin in the presence of ATP. The reaction was monitored by Western blotting with anti-T7 antibody (Novagen). Sic1^PY was efficiently polyubiquitinylated in an Rsp5-dependent manner.

Western blotting.

Total cell extract was prepared by the mild-alkali method (Kushnirov 2000). Proteins were subjected to SDS-PAGE, transferred to PVDF membranes, and blotted with the following specific antibodies: rabbit anti-ubiquitin serum (SIGMA), mouse anti-HA antibody (12CA5, Roche), horseradish peroxidase (HRP)-conjugated mouse anti-T7 antibody (Novagen), mouse anti-β-galactosidase antibody (Promega), mouse anti-Flag antibody (SIGMA), or rabbit anti-Rpt5 antibody (AFFINITI). The blots, except those detected with anti-T7 antibody, were incubated with HRP-conjugated secondary antibodies (Amersham) and visualized by ECL (Amersham). Signals were detected by using the LAS3000 system (Fuji).

Expression of tUB in Xenopus embryos.

Full-length mRNAs for tUB6, tUB6-HA, and monomer ubiquitin genes were synthesized in vitro using an mMESSAGEmMACHINE T3 kit (Ambion). Synthesized mRNA was dissolved in RNase-free water, and 5 ng of mRNA was injected in a volume of 9.2 nl into a blastomere of a two-cell stage Xenopus embryo. Embryos were cultured in 0.2 x MMR (20 mM NaCl, 0.4 mM KCl, 0.4 mM CaCl₂, 0.2 mM MgCl₂, 1 mM HEPES pH 7.4) solution at 20°C, and photographs were taken at appropriate time points after fertilization. At the blastula stage, each embryo was individually harvested, crushed in extraction buffer (20 mM HEPES, 100 mM KCl, 5 mM MgCl₂, 5 mM EGTA, 10 mM N-ethylmaleimide, 100 mM MG132 at pH 7.5) with complete protease inhibitor cocktail (Roche), and centrifuged to collect the cytoplasmic fraction. Samples of this fraction were boiled in SDS buffer and subjected to Western blot analysis with anti-ubiquitin antibody or anti-HA antibody.

RESULTS AND DISCUSSION

Construction of the genes encoding polyubiquitin insensitive to ubiquitin hydrolases.

The yeast UBI4 gene encoding five ubiquitins, which are joined tandemly in the manner of a tail-to-head concatemer, is translated as a fusion product (Ozkaynak 1984). There are three other genes encoding ubiquitin, UBI1~UBI3, in each of which ubiquitin is produced as a fusion with a ribosomal protein (Ozkaynak 1987). Ubiquitin is produced from these primary translation products so efficiently by ubiquitin hydrolases that primary translation products of ubiquitin fusion proteins cannot be seen in cells (Sabin 1989). Therefore, it has not been tested whether linear polyubiquitin chains have any biological effects.

Here, we constructed genes encoding tandem polyubiquitin chains, designated tUBs, which are not cleaved by ubiquitin hydrolases (Fig. 1A). The tUB genes, and mono-ubiquitin gene as well, were placed downstream of the GAL1 promoter so that their expression could be induced by galactose. Plasmid containing one of the tandem ubiquitin genes was introduced into a wild-type strain W303-1B (MATα). When these transformants were grown in SGal-Ura (using galactose as sole carbon source), tUb of the expected size was produced in the respective transformants (Fig. 1B). Interestingly, tUb2 (dimer) was extensively polyubiquitinylated and monoubiquitinylation was seen in all the tUbs tested. To examine the effect of the production of tUb on cellular growth, ten-fold serial dilutions of representative transformant cultures were spotted on SC-Ura and SGal-Ura. The results clearly showed that all tUbs tested were toxic whereas the mono-ubiquitin was not (Fig. 1C). For unknown reasons, tUb3 (trimer) displayed a weaker inhibition.

View Details

Fig. 1. Characterization of tUbs. A, A schematic drawing of Ub derivatives. B, Induction of tUb. Each tUB gene was introduced into W303-1B and a representative transformant was grown in SGal-Ura for 3.5 hours. Cell extracts were prepared and analyzed by Western blotting using anti-ubiquitin antibody. C, Growth inhibition by the tUB genes. W303-1B carrying the indicated tUB gene was first grown in SC-URA medium until stationary phase. Cell density was adjusted to OD600nm=1.0 and 5 μl of serial 10-fold dilutions were spotted on SC-Ura and SGal-Ura and plates were incubated at 25°C for 3 days.

tUb inhibits proteolysis by the ubiquitin proteasome system.

To examine whether tUb inhibits the ubiquitin-proteasome pathway, we determined the steady state level of Gcn4, whose degradation by the ubiquitin-proteasome pathway has been well documented (Meimoun et al. 2000; Chi et al. 2001). Plasmid carrying the tUB6 (hexamer) gene and the empty vector were introduced separately into the GCN4-HA strain, in which the GCN4 gene had been replaced with the GCN4-HA gene, and each transformant was grown in SGal-Ura or SC-Ura for 3.5 hours. As a reference, a proteasome-defective strain, the nin1-1/rpn12-1 GCN4-HA strain, was incubated at 37°C for 3.5 hours in SC-Ura. Extracts were prepared and proteins were separated by SDS-PAGE and then subjected to Western blotting using anti-HA antibody. As shown in Fig. 2A, a lower level of Gcn4-HA was seen in the GCN4-HA strain without expression of the tUB6 gene, whereas Gcn4-HA accumulated in the cells expressing tUB6 and high molecular mass polyubiquitinylated bands were also seen. The nin1-1 GCN4-HA strain incubated at a restrictive temperature displayed the same pattern of accumulation of polyubiquitinylated Gcn4-HA as the cells expressing tUB6: Gcn4 was stabilized and polyubiquitinylated derivatives were accumulated. These data suggest that tUb inhibits degradation of Gcn4-HA by the 26S proteasome but not ubiquitinylation of Gcn4. Inhibition of the 26S proteasome was further examined by assessing the degradation of Ub-X-β-galactosidase (X = S, R, or P). Each ubiquitin fusion gene was introduced into the cells harboring vector or one of the tUB genes. After the cells were incubated for 4 hours in SGal-Ura-Trp, β-galactosidase activity was assayed. S-β-galactosidase is an inefficient substrate for the N-end rule pathway whereas R-β-galactosidase is degraded promptly by the N-end rule pathway (Bachmair et al. 1986), which was reflected in the steady state level of β-galactosidase activity in cells without tUb (Fig. 2B). The steady state level of β-galactosidase in the cells expressing S-β-galactosiase was similar in all the cells tested whereas β-galactosidase activity derived from R-β-galactosidase was low in the absence of tUb and higher in the cells producing tUb4 (tetramer), tUb6, or tUb8 (octamer). Production of tUb2 and tUb3 stabilized R-β-galactosidase, although to a lesser extent. Ub-P-β-galactosidase is a poor substrate for ubiquitin hydrolases and is degraded by the UFD pathway (Johnson et al. 1995). In contrast to R-β-galactosidase, Ub-P-β-galactosidase was highly stabilized in cells producing tUb2 as well as in cells producing tUb4, tUb6, and tUb8. Extracts prepared from cells expressing the indicated UB-X-lacZ and tUB genes were analyzed by Western blotting using anti-β-galactosidase antibody (Fig. 2C). The data confirmed the results shown in Fig. 2B. Ub-S- and Ub-R-β-galactosidases were processed by ubiquitin hydrolases to produce S-β-galactosidase and R-β-galactosidase, respectively, indicating that tUb does not inhibit ubiquitin hydrolase activities. Altogether, the results imply that tUb is likely to inhibit protease activity of the 26S proteasome.

View Details

Fig. 2. tUb inhibits proteolysis by the ubiquitin-proteasome systen. A, Stabilization of Gcn4 by tUb. The GCN4-HA strain containing the tUB6 gene was grown in SC-Ura or SGal-Ura medium for 6 hours. As a reference, the nin1-1 GCN4-HA strain incubated at 37°C for 3.5 hours was similarly analyzed. Proteins in extracts were separated by SDS-PAGE followed by Western blotting using anti-HA antibody. *; a product of a non-specific reaction. B, Stability of N-end rule substrates and UFD substrate. W303D strain containing two plasmids, one carrying a tUB gene and the other carrying a Ub-X-lacZ gene, was grown overnight in SRaf-Ura-Trp and reinoculated in the same medium. After 3 hours of incubation, 2% galactose was added and incubation was continued for another 4 hours. Then, extract was prepared and β-galactosidase was assayed as described (Miller 1972). Data shown are an average from three independent experiments. Enzyme activity corresponding to 100% was defined as that expressed by extract prepared from cells without tUb and is shown in the figure. Chain length of tUb was indicated as the number with x at the bottom. C, The steady state level of β-galactosidase analyzed by Western blotting. Samples of extract containing tUb2 and tUb6 in the previous experiment were analyzed by Western blotting using anti-β-galactosidase antibody.

tUb forms a complex with the 26S proteasome.

To obtain clues about the mechanism of action of tUb, inter-action between tUb and the 26S proteasome was examined. First, binding between tUb and the 26S proteasome was examined. To this end, we constructed RPN11-FLAG strains containing one of tUB genes. Test strains thus constructed were incubated in SGal-Ura medium to induce production of tUb and were employed in pull-down experiments using anti-Flag antibody followed by Western blotting with anti-ubiquitin antibody. As shown in Fig. 3A, tUb4 or larger tUbs, tUb6 and tUb8, coprecipitated with the 26S proteasome. This is comparable to the results shown by Piotrowski et al. (1997) that K48-linked isopeptidyltetraubiquitin is the smallest polyubiquitin chain that binds the 26S proteasome and inhibits its activity. Extra bands seen in the lane containing tUb6 or tUb8 were not identified but they are most probably ubiquitinylated forms or truncated forms of tUb6 or tUb8. tUb2 inhibits proteolysis but does not bind the 26S proteasome, suggesting that tUb2 does not inhibit the 26S proteasome activity directly; in this case, polyubiquitinylated tUb2 may bind and inhibit the 26S pro-teasome activity. The immunoprecipitated 26S proteasome containing tUb was visualized by Coomassie brilliant blue staining. In cases in which the migration of tUb does not overlap with that of a subunit of the proteasome, such as in case of tUb8, the band of tUb was seen on the gel. The lane containing tUb8 was scanned by densitometry, and the results indicated that approximately one molecule of tUb8 was contained in one molecule of the 19S RP (Fig. 3B). Binding of tUb with the 26S proteasome was examined by a gel shift assay. The wild-type strain possessing the tUB6 gene was grown under induced or non-induced conditions, and then cell extracts were prepared and analyzed by native polyacylamide gel electrophoresis as described previously (Elsasser 2002). Proteasomes were visualized by the in-gel assay of their peptidase activities using the fluorogenic substrate Suc-LLVY-MCA. The results indicated that tUb does not inhibit peptidases of the 20S proteasome. As shown in Fig. 3C, both 26S proteasomes (RP₂CP and RP₁CP) in extract containing tUb6 migrated slower than those in extract without tUb6. In contrast, the mobility of the 20S proteasome was not changed, indicating that tUb6 bound to the 26S proteasome but not to the 20S proteasome. It should be noted that almost all the 26S proteasome in the extract containing tUb6 shifted, indicating that all the 26S proteasome molecules likely contain tUb6.

View Details

Fig. 3. Biochemical characterization of tUb. A, Binding of tUb to the 26S proteasome. Plasmid carrying the indicated tUB gene was introduced into YYS40 whose RPN11 had been replaced with RPN11-FLAG. Each strain was grown in SGal-Ura for 6 hours and extract was prepared. The 26S proteasome was purified by the immuno-affinity purification method using anti-Flag antibody. The 26S proteasome thus purified was separated by SDS-PAGE. tUb coprecipitated with the 26S proteasome was detected with anti-ubiquitin antibody. Subunit composition was visualized by Coomassie brilliant blue (CBB) staining. An arrow indicates polyubiquitinylated tUb2. * indicates the respective tUb. The chain length of tUb is shown at the top of each lane. B, Densitometric tracing. Lanes “cont” and 7 “8x” (the right most) with CBB staining were scanned by LAS3000. The position of tUb8 is indicated by an arrow. C, Gel shift assay. W303-1A carrying vector or tUB6 was grown in SC-Ura or SGal-Ura for 6 hours and extracts were prepared. Extract containing 200 μg of protein or the purified 26S proteasome (0.5 pmol) was loaded on a 4% native polyacrylamide gel. After electrophoresis, the positions of proteasomes were visualized by in-gel assays of peptidase activity using 0.1 mM Suc-LLVY-MCA as substrate. To visualize the position of the 20S proteasome, 0.05% SDS was added to the reaction mixture.

We showed that tUbs larger than the tetramer are efficiently bound to the 26S proteasome. However, it is possible that tUbs inhibit some reactions involving ubiquitin other than proteolysis by the 26S proteasome. To explore this possibility, pull-down experiments using Flag-Ub or Flag-tUb4 as bait were conducted. Proteins pulled down by this procedure were resolved by SDS-PAGE (Fig. 4A). Few proteins were co-precipitated with Flag-Ub, whereas many proteins were co-precipitated with Flag-tUb4, most of which were identified as subunits of the 26S proteasome. To further confirm that proteins associated with tUb4 are from the 26S proteasome, precipitates were washed with salt and eluted proteins were separated by SDS-PAGE and stained with CBB (Fig. 4B, upper panel). Salt-eluted proteins separated by SDS-PAGE were analyzed by Western blotting using anti-Rpt5, anti-Rpn9, and anti-Ub antibodies (Fig. 4B, lower panel). The results indicated that the 26S proteasome was eluted completely from tUb4 by salt treatment. Proteins eluted with salt treatment did not react with anti-Ub antibody. It should be noted that Hsp70 (Ssa1) was identified among the proteins pulled down with Flag-tUb4. All together, these findings imply that it is likely that the primary target of tUb is the 26S proteasome

View Details

Fig. 4. Affinity purification of tUb-bound proteins. Wild-type cells carrying PGAL1-FLAG-tUB1 (1x), PGAL1-FLAG-tUB4 (4x), or a control plasmid, pYF3 (cont), were grown to OD600nm=1.0 in SRaf-Trp medium, and then galactose was added and the cells were further cultured for 6 hours at 25°C. The cells were harvested, washed and resuspended with buffer A containing protease inhibitor cocktail, and lysed with glass beads. After centrifugation at 15,000 rpm for 30 min at 4°C, the supernatant was recovered, mixed with M2 beads and rotated for 1 hour at 4°C. After washing extensively with buffer A, the beads were divided into two. Half of the beads were incubated with Flag peptide and the eluted materials were collected (no treatment). The other half were treated with 0.6 M NaCl, and the salt-eluted proteins were collected (eluted), and subsequently the bead-retained materials were eluted with Flag peptide (retained). A, The proteins were separated by SDS-PAGE and stained with CBB. Purified 26S proteasome was used as a control. The protein band at 70 kDa was identified as Ssa1 by peptide mass fingerprinting. B, The same samples as in A were analyzed by Western blotting with anti-Ub (upper panel) or anti-Rpt5 and anti-Rpn9 antibodies (lower panel).

tUb inhibits 26S proteasome activity in vitro.

The simplest explanation of the effect of tUb on proteolysis is that tUb inhibits the 26S proteasome. This should be tested directly by a more straightforward method. We attempted an in vitro assay of the 26S proteasome activity using polyubiquitinylated Sic1 (Poly-Ub-Sic1^PY) as substrate. Sic1^PY, in which the PY motif that serves as a binding site for Rsp5 had been introduced into the N-terminal region, was successfully polyubiquitinylated in vitro in the reaction mixture containing E1, E2, and Rsp5 as E3 enzyme (our unpublished results). Poly-Ub-Sic1^PY thus formed was found to be a good substrate for the in vitro assay of the 26S proteasome activity (Fig. 5A). To examine the effect of tUb on the 26S proteasome activity, the 26S proteasome and the 26S proteasome-tUb6 complex were purified by the immuno-affinity purification method. The purity of the samples was confirmed by SDS-PAGE (Fig. 5B). Equal amounts of two proteasome preparations was separately incubated with Poly-Ub-Sic1^PY. The 26S proteasome with tUb6 showed much lower degrading activity toward Poly-Ub-Sic1^PY than did the 26S proteasome. It should be noted that cleavage of polyubiquitin chains from Poly-Ub-Sic^PY was inhibited in the reaction mixture containing the 26S proteasome with tUb6 (Fig. 5C, left panel). Next, we purified tUb4 using the E. coli expression system. When tUb4 was added to the reaction mixture, the activity of the 26S proteasome toward Poly-Ub-Sic1^PY was inhibited (Fig. 5C, right panel).

View Details

Fig. 5. tUb inhibits the 26S proteasome in vitro. A, Degradation of Poly-Ub-Sic1PY in vitro. Purified 26S proteasome (1 pmol) was added to Poly-Ub-Sic1PY (2 pmol) or Sic1 (control; 2 pmol) in 10 μl of proteasome buffer and incubated at 30°C. The reaction was terminated by the addition of SDS-loading buffer. The samples were subjected to Western blotting with anti-T7 antibody. Usually, after incubation for 5 min under the above conditions, the signals corresponding to Poly-Ub-Sic1PY at the gel top were eliminated and a weak signal, about 6% of the starting material, was detected at the position of unmodified Sic1PY. B. The 26S proteasome and the 26S proteasome with tUb6 were purified by the immuno-affinity method. Each sample was subjected to SDS-PAGE followed by CBB staining (left panel), native PAGE followed by in-gel assays (middle panel), and incubation with Poly-Ub-Sic1PY for 0, 5, and 10 minutes followed by Western blotting with anti-T7 antibody (right panel). C. The purified 26S proteasome (final concentration 12 nM) was preincubated with or without tUb4 (1 μM) for 15 minutes at 30°C, and then Poly-Ub-Sic1PY (50 nM) was added and reaction products were analyzed by SDS-PAGE followed by Western blotting using anti-T7 antibody.

tUB inhibits Xenopus egg cleavage.

As shown above, tUbs were found to inhibit the 26S proteasome and cell growth. To examine whether tUb can be used more generally, we examined the effects of over-expression of tUB on mitosis in Xenopus embryos. We overexpressed tUB6, or tUB6-HA, by microinjecting translatable mRNAs (5 ng) into fertilized embryos of Xenopus laevis. Forced expression of tUB6 in a blastomere at the two-cell stage caused marked abnormalities of the injected cell and its progeny (Fig. 6A). The cleavage pattern of the injected blastomere became progressively more abnormal, so that at the equivalent of the blastula stage, cleavage furrows were either absent or incomplete in the tUB mRNA-injected blastomere (Fig. 6A-ii, indicated by an arrowhead). This effect was still observable when the amount of injected tUB mRNA was reduced from 5 ng to 1 ng. Injection of 5 ng of monoubiquitin mRNA did not appreciably affect cell-cycle progression or subsequent developmental processes as compared with those in non-injected control blastomeres (Fig. 6A-i). Western blot analysis showed that the tUb6-HA protein became detectable within 1 hour after the injection of mRNAs. As in yeast, tUb6-HA was produced as native and monoubiquitinylated forms (Fig. 6B). In tUb6-producing embryos, polyubiquitinylated proteins were accumulated. These results confirmed that tUb6 production interfered with the degradation of polyubiquitinylated proteins by the 26S proteasome in Xenopus embryos. The protein levels remained unchanged until mid-blastula stage.

View Details

Fig. 6. Over-expression of tUB6 leads to impaired mitosis in Xenopus embryos. A, Effects of over-expression of tUB6 on the cell division of Xenopus blastomeres. Monomer ubiquitin mRNA (i) and tUB6 mRNA(ii) were injected into a blastomere (indicated by an arrowhead) at the two-cell stage. Photographs were taken at 5.5 hours (blastula stage) after fertilization. B, Detection of tandem ubiquitin protein expressed in Xenopus embryos after mRNA injection. tUB6-HA mRNA was injected into a blastomere at the two-cell stage and Western blot analyses with anti-HA antibody and anti-polyubiquitin antibody were performed using the extracts prepared from the cells harvested at the indicated times. Asterisks indicate artifact bands.

tUb, when overproduced, inhibited cell division in yeast cells and Xenopus embryos, and tUb inhibited the activity of the proteasome, most probably by binding to it, indicating that tUb can be widely used as inhibitor of proteasomes. The fact that tUb is produced by the cellular protein synthesis system indicates that tUb can be delivered and expressed in a targeted tissue or organ by using an appropriate promoter and can inhibit proteasomes there. Such inhibition is impossible with currently available inhibitors. Thus, tUb is a promising tool for the study of the ubiquitin-proteasome pathway in a wide range of organisms.

A part of this study was supported by a Grant-in-aid for scientific research from MEXT.