- Email: [email protected]
Properties of the Beta Subunit of the Proteasome Activator PA28 (11S REG)
Archives of Biochemistry and Biophysics Vol. 384, No. 1, December 1, pp. 174 –180, 2000 doi:10.1006/abbi.2000.2112, available online at http://www.idealibrary.com on
Properties of the Beta Subunit of the Proteasome Activator PA28 (11S REG) Sherwin Wilk, 1 Wei-Er Chen, and Ronald P. Magnusson Department of Pharmacology, Mount Sinai School of Medicine, One Gustave L. Levy place, New York, New York 10029
Received July 25, 2000, and in revised form August 29, 2000
The proteasome activator PA28 (11S REG) is composed of two homologous subunits termed ␣ and . The properties of the recombinant -subunit were explored and compared to the properties of the recombinant ␣-subunit. PA28 produced in an Escherichia coli expression system migrates on a calibrated gel filtration column as an apparent heptamer (M r ⴝ 250,000). Low concentrations of SDS (0.005%), dissociate the protein to a monomer (M r ⴝ 33,000). PA28 has a complex effect on proteasome activity. At concentrations which favor oligomerization (> 2 M), PA28 is a strong proteasome activator although its affinity for the proteasome is about 10-fold less than recombinant PA28␣. The catalytic properties of the PA28␣ and PA28-activated proteasome are similar. At low concentrations, PA28 is a monomer and a potent allosteric proteasome inhibitor. These studies show that oligomerization of PA28 is required for proteasome activation and that PA28 monomers are potent proteasome inhibitors. © 2000 Academic Press Key Words: proteasome; multicatalytic proteinase complex; proteasome activator; PA28; 11S regulator; proteasome inhibitor.
The 20S proteasome is the catalytic core of the major extralysosomal proteolytic system of the cell. It plays an essential role in many vital cellular processes, including regulation of the cell cycle, degradation of transcription factors and regulatory molecules, and processing of peptides for presentation by the MHC class-I system (1, 2). Early studies demonstrated that distinct active sites of the 20S proteasome cleave peptide bonds after acidic, basic, and hydrophobic amino acids characterizing this macromolecule as a multicatalytic pro1 To whom correspondence and reprint requests should be addressed. Fax: (212) 831-0114. E-mail: [email protected].
174
teinase complex (3). Electron microscopic analysis of the proteasome from Thermoplasma acidophilum revealed a cylindrical structure made up of four stacked heptameric subunit rings (4). X-ray crystallographic analysis of the yeast enzyme demonstrated that the three distinct active sites reside on -type subunits that form the inner rings (5). The active sites are sequestered within a central chamber with no apparent substrate entry channel. This structure serves to protect the cell against uncontrolled proteolysis. It is also evident that regulatory molecules must exist to facilitate substrate entry. Two major types of proteasome regulatory molecules have been described (6). A 19S macromolecular assembly of proteins can cap the 20S proteasome on both ends to form the ubiquitin-protein degrading 26S proteasome (2). The 19S caps presumably recognize the polyubiquitin tag, bind, and unfold the target protein and open an entry port for the protein substrate in an ATP-dependent process (7). A distinct regulator termed 11S REG (8) or PA28 (9) markedly activates the degradation of peptides but not proteins in an ATPindependent fashion. PA28 is composed of two homologous 28-kDa subunits termed ␣ and  (see 10 for a recent review). Recombinant PA28␣ is a heptamer (11). Although endogenous PA28 was proposed to be a hexamer made up of alternating ␣- and -subunits (12, 13), analysis of recombinant PA28␣, by mass spectroscopy is consistent with a heptameric structure (14). The exact subunit arrangement of the ␣,-heptamer is not clear. Apparently only the hetero-oligomer exists in vivo since PA28-deficient mice also do not have detectable PA28␣ (15). However, at present, the existence of free PA28 cannot be ruled out. A third protein member of the PA28 family was identified by a homology search as the previously cloned K i autoantigen (16) and renamed as PA28␥ or REG␥ (17). Immunohistochemical analysis conducted with antibodies directed against ␣-, -, and ␥⫺ proteins show 0003-9861/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
PROTEASOME ACTIVATOR SUBUNIT BETA
that ␣- and -subunits have an identical cellular localization and are predominantly found in the cytosol and nucleolus (18). PA28␥ is found in the nucleus but not nucleolus and in cytoplasmic microtubular-like extensions and inclusion bodies (18). Although there is general agreement concerning the ability of recombinant PA28␣ to activate the proteasome, there is no consensus concerning the properties of PA28. Realini et al. report good activation by PA28 at concentrations higher than needed to see activation by PA28␣ (19). They were not able to isolate an oligomer stable to gel filtration, although they proposed that oligomerization may occur at high concentrations. Two other groups could neither detect formation of a homooligomer nor activation (20, 21). We report that recombinant PA28 can be produced in high concentrations in an E. coli expression system and that the molecule can be isolated as an apparent heptamer by gel filtration. The effect of PA28 on proteasome activity is complex. At low concentrations (⬃0.18 M), PA28 exists as a monomer and is a strong proteasome inhibitor. At relatively high concentrations (⬎ 2 M), oligomerization is favored and PA28 is a strong activator of the proteasome. The catalytic profile of the PA28-activated proteasome is similar to that of the PA28␣⫺activated proteasome although PA28 has a much lower affinity for the proteasome than PA28␣. EXPERIMENTAL PROCEDURES
Materials Bovine pituitary 20S proteasome was purified by a modification of the method of Orlowski and Michaud (22), in which a phenyl Sepharose chromatographic step was substituted for their second anion exchange chromatographic step (23). Succinyl-Leu-Leu-Val-Tyr-7amido-4-methylcoumarin (suc-LLVY-AMC) 2 was purchased from Novabiochem (San Diego, CA). Z-LLE-NA was synthesized as described (24). Z-D-ALR-NA was prepared by conventional solution phase synthesis. All molecular weight markers for gel filtration chromatography were from Sigma, Inc. (St. Louis, MO). The QuikChange Site-directed mutagenesis kit and XL2 Ultracompetent cells were obtained from Stratagene, Inc. (La Jolla, CA). E. coli BL21(DE3) cells and the pET16b vector were from Novagen (Madison, WI). Oligonucleotides were synthesized by Integrated DNA Technologies, Inc. (Coralville, IA). DNA sequencing was performed by the Utah State Biotechnology Center (Logan, UT). Apparently homogeneous recombinant PA28␣ expressed in a baculovirus system was obtained as previously described (25). Hsp-90 was purified to apparent homogeneity from the supernatant fraction of a bovine pituitary homogenate. Hsp-90 copurifies with the 20S proteasome and can be separated after the second anion exchange chromatographic step of the method of Orlowski and Michaud (22). Hsp90 elutes after the proteasome in an essentially homogeneous form.
2 Abbreviations used: AMC, 7-amido-4-methylcoumarin; NA, 2-naphthylamide; Z, N-benzyloxycarbonyl; PGPH, peptidylglutamyl peptide bond hydrolyzing; suc, succinyl; TCA, trichloroacetic acid.
175
Methods Preparation of a pET16b-PA28 expression plasmid. The sequence of the human beta subunit of the proteasome activator PA28 (26) was used to search the NCBI expressed sequence tag (est) data base (dbest). This search revealed a human est (ATCC 1072203). The full-length sequence cloned in the vector pT7T3D-Pac was identical to the input sequence with the following exceptions: Thr 229 in the input sequence (codon ACC) was Asn in the ATCC clone (codon AAC), and the codon for Arg 21 (AGG in the input sequence) was AGA in the ATCC clone. The substitution of Asn for Thr is in agreement with the PA28 sequence obtained by Realini et al. (19). The ATCC clone contains Ala 2 in agreement with the sequence of Ahn et al. (26), whereas the clone described by Realini et al. contains Ser in this position. To subclone the full length insert into the pET-16b prokaryotic expression vector, site-directed mutagenesis was first performed to introduce a NcoI site at the 5⬘ end. Site-directed mutagenesis was performed with the aid of a Stratagene Quik-change kit. Oligonucleotides were designed in which the sequence around the start codon GCATGG was mutated to CCATGG. The oligonucleotides synthesized were for the forward direction 5⬘ CACGGCTTGGCCATGGTGCTTCAGTCGCTAG 3⬘ and for the reverse direction 5⬘ CTAGCGACTGAAGCACCATGGCCAAGCCGTG 3⬘. Since the PA28 insert contains an internal NcoI site, the mutated plasmid was partially digested with NcoI and fully digested with NotI and the full-length insert was purified by gel electrophoresis. The pET-16b vector was digested with NcoI and BamHI. This digest removes the His tag and the factor Xa site. The insert was ligated into the cut vector with the aid of a NotI–BamHI linker. Expression and purification of PA28. E. Coli strain BL21(DE3) was transformed with the pET16b-PA28 plasmid. A single colony was added to 25 ml LB medium and incubated for 4.5 h at 37°C. The culture was then transferred to 3 liters LB medium and grown overnight at room temperature. IPTG was then added (0.2 mg/ml) and the material collected after 7 h incubation at room temperature. After centrifugation, the pellet was extracted with 10 mM Tris–HCl, pH 7.5. Greater than 50% of the expressed protein was present in the soluble fraction of the cell. PA28 was purified by sequential FPLC on mono-Q, phenyl Superose, and Superose 6 columns as described for PA28␣ (25). Purification was monitored by SDS–PAGE analysis of the column fractions. The resulting protein was judged to be more than 95% pure by SDS–PAGE (Fig. 1). Measurement of enzymatic activity. Enzymatic activity was measured with the substrates suc-LLVY-AMC, Z-D-ALR-NA, and Z-LLENA. Enzyme and substrate in the presence or absence of PA28 proteins at the concentrations designated in the figure legends were incubated in a 0.05 M Tris–HCl buffer, pH 7.5, at a final volume of 100 l. Reactions were terminated with 100 l 10% TCA and the chromogen released was measured spectrophotometrically at 560 nm (AMC) or 580 nm (NA) by a diazotization procedure (24). HPLC analysis. HPLC was performed on a Waters Model 600E Instrument fitted with a 4.6 mm ⫻ 25 cm Supelcosil 5 m C8 column (Supelco). The column was equilibrated with 15% acetonitrile/ 0.05%TFA at a flow rate of 1.0 ml/min. Elution was carried out by linearly increasing the acetonitrile concentration to 55% over a 34min period. Eluting peaks were detected by absorbance at 210 nm at a sensitivity of 0.1 AUFS. Determination of relative molecular mass (M r): (a) Superose 6 chromatography. FPLC was conducted on a Pharmacia system. A Superose 6 column (Pharmacia) was equilibrated with 0.05 M Tris– HCl/0.1 M NaCl. The column was calibrated with thyroglobin (669,000), ferritin (440,000), catalase (232,000), and aldolase (158,000). Blue dextran was used to measure the void volume. M r was determined from a plot of log M r vs K where K ⫽ V-V o/V t-V o, V o
176
WILK, CHEN, AND MAGNUSSON
PA28 Has a Biphasic Effect on the Chymotrypsinlike and PGPH Activities of the Proteasome
FIG. 1. SDS–PAGE (12.5% gel) analysis of the purification of expressed PA28. Lanes 1 and 6, molecular mass markers in kilodaltons; lane 2, whole cell extract; lane 3, active fraction after mono Q chromatography; lane 4, active fraction after phenyl Superose chromatography; lane 5 active fraction after Superose 6 chromatography. Proteins stained with Coomassie blue.
The bovine pituitary 20S proteasome was incubated with various concentrations of recombinant human PA28 and the hydrolysis of suc-LLVY-AMC was determined. A biphasic effect was observed. At concentrations of PA28 less than 2 M inhibition occurred whereas at higher concentrations the chymotrypsinlike activity was stimulated (Fig. 2). Maximal stimulation in different experiments varied from three to fivefold. A similar effect was noted for the PGPH activity although somewhat higher concentrations of PA28 were required for stimulation (Fig. 2). Stimulation of the trypsinlike activity was more modest and inhibition was not readily apparent at low concentrations of PA28 (Fig. 2). Since inhibition at low concentrations of PA28 has previously not been described, this phenomenon was analyzed in greater detail. Maximal inhibition of the chymotrypsinlike activity occurred at a PA28 concentration of 180 nM with 50% inhibition at 25 nM (Fig. 3). A molecular weight of 28,000 was used to calculate molar concentrations. As a control we evaluated inhibition of the proteasome by BSA and obtained an EC50 of 165 nM approximately sevenfold greater than the PA28 monomer (Fig. 3).
is the column void volume, V t the total volume, and V the elution volume of the protein of interest. Proteins were detected at 254 nm. (b) Superose 12 chromatography. A Superose 12 column (Pharmacia) was equilibrated with 0.05 M Tris–HCl/0.1 M NaCl. The column was calibrated with bovine serum albumin (66,000), ovalbumin (45,000), carbonic anhydrase (29,000), ␣-lactalbumin (14,300), and the oxidized B chain of insulin (3500). M r was determined as described above for Superose 6 chromatography.
RESULTS
Recombinant PA28 Is Highly Expressed in E. coli and Readily Purified In an earlier study we were able to obtain very high expression of the alpha subunit of PA28 in a baculovirus system (25). However, we were not able to obtain a similar expression level for the beta subunit in baculovirus infected Sf-9 cells. We therefore explored the expression of PA28 in E. coli. Induction by IPTG at room temperature as described under Experimental Procedures gave high expression with at least half of the expressed protein present in the soluble fraction. We then applied the same FPLC procedure used to purify PA28␣ (25) and obtained the beta subunit in greater than 95% purity (Fig. 1). To obtain active PA28 it was essential to conduct the sequential steps continuously or if necessary to store active fractions at ⫺80°C between steps.
FIG. 2. Effect of PA28 on the catalytic activities of the 20S proteasome. Incubation mixtures in a total volume of 100 l contained 1.1 g proteasome, PA28 at the concentrations shown on the abscissa, a substrate concentration of 0.4 mM and 0.05 M Tris–HCl, pH 7.5. The chymotrypsinlike activity (ChT-L) was measured with sucLLVY-AMC, the PGPH activity with Z-LLE-NA and the trypsinlike activity (T-L) with Z-D-ALR-NA. Samples were incubated at 37°C for 30 min and the chromogen released determined as described under Experimental Procedures. Velocity is given on the ordinate as absorbance units at 560 nm for AMC and 580 nm for NA.
PROTEASOME ACTIVATOR SUBUNIT BETA
FIG. 3. Comparison of the inhibition of the 20S proteasome-catalyzed hydrolysis of suc-LLVY-AMC by low concentrations of PA28 and BSA. Incubation conditions as described in the legend to Fig. 2. Concentrations of protein inhibitors shown on the abscissa. PA28 (E); BSA (F).
Low Concentrations of PA28 Act as a Negative Allosteric Effector The mechanism of inhibition of the chymotrypsinlike activity of the proteasome by PA28 was studied by determining inhibition as a function of substrate concentration. In this experiment, incubations were conducted in the presence of 170 nM PA28, 16 nM proteasome and substrate concentrations of suc-LLVYAMC ranging from 0 to 50 mM. In the absence of PA28, the degradation of suc-LLVY-AMC displayed normal Michaelis–Menten kinetics (Fig. 4). It should be noted however that some preparations of the proteasome assayed with this substrate exhibit cooperativity. These differences may be due to conformational changes in the protein during the purification proce-
FIG. 4. Inhibition of the chymotrypsinlike activity of the 20S proteasome by low concentrations of PA28 as a function of substrate concentration. Incubation mixtures in a total volume of 100 l contained proteasome at a concentration of 16 nM, suc-LLVY-AMC at concentrations shown on the abscissa, 0.05 M Tris–HCl, pH 7.5, and either 170 nM PA28 (F) or no PA28 (E). Velocity is given on the ordinate as absorbance units at 560 nm.
177
FIG. 5. Elution of PA28 on a calibrated Superose 6 column. The elution volume of PA28 was slightly less than that of carbonic anhydrase (232 kDa; 15.3 ml). The arrow designates the elution volume of PA28 monomer.
dure. Low concentrations of PA28 greatly increased cooperativity resulting in a large increase in Km but an unchanged Vmax (Fig. 4). The data were analyzed by a Hill plot. The Hill coefficient increased from unity in the absence of PA28 to 2.4 in its presence. Stimulation or Inhibition Depends on the Oligomeric State of PA28 PA28 is reported in the literature to exist as a monomer (21). It is also reported that the recombinant -subunit does not form a hexamer or heptamer stable to gel filtration (19). It was therefore surprising to observe that PA28 after purification on a calibrated Superose 6 column migrated predominantly as a species with an elution volume nearly identical to that of heptameric PA28␣. The M r was calculated as 250,000. A minor component with an elution volume consistent with that of the monomer was seen as was a small peak between the two which may represent an intermediate form (Fig. 5). The ability of PA28 to form a homooligomer and stimulate the proteasome is undoubtedly a concentration-dependent effect as suggested by Realini et al. (19). On the other hand we found strong inhibition of the chymotrypsinlike activity of the proteasome at low concentrations of PA28. We previously reported that heptameric PA28␣ could be dissociated by very low concentrations of SDS (0.005%) and that the dissociated protein could not stimulate the proteasome (25). To determine that monomeric PA28 is the inhibitory species, we incubated the proteasome with a relatively high concentration of PA28 (5.4 M) in the presence and absence of 0.005% SDS. In the absence of SDS this concentration of PA28 stimulated the chymotrypsinlike activity 3.7-fold. When the incubation was conducted in the presence of 0.005% SDS, activity
178
WILK, CHEN, AND MAGNUSSON
Monomeric PA28 Is a Stronger Proteasome Inhibitor Than Hsp90
FIG. 6. Comparison of the stimulation of the chymotrypsinlike activity of the 20 S proteasome by PA28␣ and PA28 as a function of activator concentration. Incubation mixtures in a total volume of 100 l contained 0.38 g proteasome, 0.4 mM suc-LLVY-AMC, PA28 ␣ or  at the concentrations shown on the abscissa, and 0.05 M Tris–HCl, pH 7.5. Molar concentrations of PA28␣ and PA28 were calculated using the molecular weight of the monomer (28,000). Samples were incubated at 37°C for 30 min and the AMC released determined as described under Experimental Procedures. Velocity is given on the ordinate as absorbance units at 560 nm.
Other proteins have been reported to inhibit the proteasome. One of the most potent inhibitors is heat shock protein 90 (Hsp90) (28). It was therefore of interest to compare the inhibition of the proteasome produced by monomeric PA28 to that of Hsp90. Bovine pituitary Hsp90 was purified to apparent homogeneity as described under Experimental Procedures. Hsp90 strongly inhibited the proteasome with an IC 50 of 50 nM measured at a proteasome concentration of 12.5 nM (Fig. 8). The EC50 is greater than that found for PA28 (Fig. 3). Unlike other inhibitory proteins such as PI31 (29, 30) or HIV-tat (31), Hsp-90 did not antagonize the activation of the proteasome by PA28␣ or PA28 (not shown). Geldanamycin, a benzoquinone ansamycin antagonist of Hsp90 (32) tested at a concentration of 50 M only slightly attenuated inhibition of the proteasome by Hsp-90 (not shown).
was totally inhibited. The detergent alone had no effect on catalytic activity. The M r of the 0.005%-treated protein on a calibrated Superose 12 column was 33,000. PA28 Is a Strong Proteasome Activator But Is Less Potent Than PA28␣ Stimulation of the chymotrypsinlike activity of the proteasome by the purified recombinant PA28 ␣ and  proteins was compared. At the substrate concentration used, PA28␣ maximally stimulated the chymotrypsinlike activity approximately six-fold. PA28 stimulated this activity approximately five fold (Fig. 6). The EC50 of PA28 was about ten-fold higher than the EC50 of PA28␣. Unlike the beta subunit, low concentrations of PA28␣ did not inhibit the proteasome. We have previously used the ovalbumin immunodominant epitope SIINFEKL to study proteasome activation since this peptide is only slowly cleaved by the latent proteasome. In the presence of PA28␣ it is cleaved after both hydrophobic and acidic amino acids (27). After a 1 h incubation of SIINFEKL with the latent proteasome, there was no detectable hydrolysis. Addition of either PA28 or PA28␣ markedly stimulated degradation to SIINFE and SIINF (Fig. 7). It should be noted that in these experiments the concentration of PA28 exceeded that of PA28␣ by eightfold.
FIG. 7. HPLC analysis of the degradation of SIINFEKL by the latent proteasome, the PA28 activated proteasome, and the PA28␣ activated proteasome. Incubation mixtures in a total volume of 125 l contained 1.52 g 20S proteasome, SIINFEKL at a final concentration of 200 M, 0.05 M Tris–HCl buffer, pH 7.5, and either no activator (A) or 40 g PA28 (B) or 5 g PA28␣ (C). Samples were incubated for 1 h at 37°C. HPLC was carried out as described under Experimental Procedures. Peak 1, SIINFE; peak II, SIINF; peak III, SIINFEKL. Inj designates point of injection.
PROTEASOME ACTIVATOR SUBUNIT BETA
FIG. 8. Inhibition of the chymotrypsinlike activity of the 20S proteasome by Hsp90. Incubation mixtures in a total volume of 100 l contained 1.34 g proteasome, 0.4 mM suc-LLVY-AMC, Hsp90 in the concentrations shown on the abscissa and 0.05 M Tris–HCl, pH 7.5. Samples were incubated at 37°C for 30 min and the AMC released determined as described under Experimental Procedures. Velocity is given on the ordinate as absorbance units at 560 nm.
DISCUSSION
There is disagreement in the literature as to whether PA28 can stimulate the proteasome (19 –21). We clearly show that this protein at relatively high concentrations is an effective activator. It is further evident from these studies that the oligomeric state of PA28 determines its effect on the catalytic activities of the proteasome. Activation is associated with formation of an oligomer, apparently a heptamer. In agreement with the findings of Realini et al. (19) and in contrast to other reports (20, 21) we observe strong stimulation by high concentrations of PA28 although the magnitude of stimulation does not reach that produced by PA28␣. Moreover equivalent proteasome activation by the recombinant alpha subunit occurs at about a 10-fold lower molar concentration than the beta subunit (Fig. 6). The association of PA28 monomers is weak and concentration dependent. By contrast, the monomer-heptamer equilibrium of PA28␣ is strongly shifted toward the heptamer (25). Other investigators have been unable to obtain active PA28 or if active have not been able to isolate an oligomer stable to gel filtration (19, 21). This is explained by the relative instability of PA28 and also by the concentration dependence of oligomerization. We found active PA28 to be less stable than active PA28␣. For this reason PA28 must be rapidly purified and we have successfully used an FPLC procedure. Moreover if the protein is not immediately taken from one step to the
179
next it is imperative to store the solution at ⫺80°C to preserve activity. In addition good expression is necessary to obtain a preparation of PA28 stable to gel filtration (Fig. 1). The stimulation of the three catalytic activities of the 20 S proteasome by PA28 differs. The strongest effect is exerted on the chymotrypsinlike activity. The PGPH activity is more moderately enhanced and the trypsinlike activity is least affected (Fig. 2). Similar results were obtained by Realini et al. (19). If the effect of binding PA28 to the outer ␣-rings of the proteasome were only to open channels for the access of substrate, one might expect a similar activation of the three activities. Since the active sites responsible for the three activities reside on distinct subunits, the differential effect is most likely due to alterations of the geometries of the active sites although one cannot rule out facilitation of substrate access as also contributing to the overall stimulation. Low concentrations of PA28 strongly inhibit the proteasome. It is unclear why others have not noted such inhibition. Realini et al. (19) also report a biphasic effect of PA28 but they observe modest stimulation of suc-LLVY-AMC hydrolysis at low concentrations and stimulation equivalent to PA28␣ at high concentrations. By contrast we find marked inhibition of the chymotrypsinlike activity at low concentrations of PA28. We have shown that very low concentrations of SDS (0.005%) dissociate PA28 and that this dissociated and inhibitory protein is a monomer. Since oligomer formation is necessary for stimulation, it is difficult to account for the reported stimulation by low concentrations of this protein. Our results demonstrate that monomeric PA28 can bind to the proteasome but that the proteasome does not serve as a template for assembly of the oligomer. Binding of monomeric PA28 to the proteasome inhibits catalytic activity and inhibition is clearly allosteric (Fig. 4). This is expected since this protein should not directly contact the active sites of the latent 20S proteasome. PA28 inhibited the hydrolysis of sucLLVY-AMC by 50% at a concentration of 25 nM, with maximal inhibition at a concentration of 185 nM. In these experiments the proteasome concentration was 15.7 nM and since each proteasome molecule contains 14 alpha subunits, the concentration of proteasome alpha subunits was 220 nM. Assuming that PA28 monomers can bind to all alpha subunits, maximal inhibition occurs at approximately a 1:1 molar ratio. Several other proteins have been reported to strongly inhibit the proteasome (Table I). We compared the inhibition by PA28 monomers to that produced by Hsp90 (26) and found the IC50 of PA28 to be lower than the IC50 of Hsp 90 (Table I). The proteasome inhibitor PI31 was recently characterized. Zaiss et al. (29) report that at a proteasome concentration of 3 nM,
180
WILK, CHEN, AND MAGNUSSON TABLE I
Comparison of the Potencies of Protein Inhibitors of the Proteasome
Inhibitor PA28 monomer Hsp90 PI31 Hiv-Tat ␦-Aminolevulinate dehydratase
I.C.50 (nM) 25 55 55 50 50 ⬃75
Proteasome Conc. (nM)
Ref.
15.7 12.5 3 41 1.4
This study This study 29 30 31
43
33
50% inhibition occurred at a concentration of PI31 of about 55 nM. McCutchen-Maloney et al. report 50% inhibition at about a 1:1 molar ratio of PI31 to proteasome (30). Therefore the PA28 monomer appears to be an even more potent inhibitor than PI31. HIV-1 Tat inhibits hydrolysis of suc-LLVY-AMC by 50% at a concentration of 50 nM inhibitor and 1.4 nM proteasome (31). ␦-aminolevulinate dehydratase another characterized proteasome inhibitor produces 50% inhibition at a 2:1 molar ratio of inhibitor to proteasome (33). These inhibitory proteins differ with respect to their antagonism of the activation of PA28. Both PI31 and HIV-Tat antagonize activation of the proteasome by PA28 (29 –31), whereas PA28 monomer and Hsp90 do not (data not shown). In summary, we have shown that the beta subunit of PA28 has a complex effect on the chymotrypsinlike (and PGPH) activities of the proteasome. At high concentrations this subunit oligomerizes and stimulates the proteasome. At low concentrations where the subunit exists as a monomer, the proteasome is strongly inhibited. The inhibition is as strong or stronger than other described protein inhibitors of the proteasome. Whether this inhibition represents a control mechanism in vivo remains speculative since it is not known whether free PA28 exists in the cell. ACKNOWLEDGMENT These studies were supported by NIH Grant NS 29936.
REFERENCES 1. Tanaka, K. (1998) J. Biochem. 123, 195–204. 2. Coux, O., Tanaka, K., and Goldberg, A. L. (1996) Annu. Rev. Biochem. 65, 801– 847. 3. Wilk, S., and Orlowski, M. (1983) J. Neurochem. 40, 842– 849. 4. Pu¨hler, G., Weinkauf, S., Bachman, L., Muller, S., Engel, A., Hegerl, R., and Baumeister, W. (1992) EMBO J. 11, 1607–1616. 5. Groll, M., Ditzel, L., Lowe, J., Stock, D., Bochtler, M., Bartunik, H. D., and Huber, R. (1997) Nature 386, 463– 471. 6. DeMartino, G. N., and Slaughter, C. A. (1999) J. Biol. Chem. 274, 22123–22126.
7. Gottesman, S., Maurizi, M. R., and Wickner, S. (1997) Cell 91, 435– 438. 8. Dubiel, W., Pratt, G., Ferrell, K., and Rechsteiner, M. (1992) J. Biol. Chem. 267, 22369 –22377. 9. Chu-Ping, M. Slaughter, C. A., and DeMartino, G. N. (1992) J. Biol. Chem. 267, 10515–10523. 10. Rechsteiner M., Realini, C., and Ustrell, V. (2000) Biochem. J. 345, 1–15. 11. Johnston, S. C., Whitby, F. G., Realini, C., Rechsteiner, M., and Hill, C. P. (1997) Protein Sci. 6, 2469 –2473. 12. Ahn, K., Erlander, M., Letureq, D., Peterson, P. A., Fru¨h, K., and Yang, Y. (1996) J. Biol. Chem. 271, 18237–18242. 13. Song, X., Mott, J. D., von Kampen, J., Pramanik, B., Tanaka, K., Slaughter, C. A., and DeMartino, G. D. (1996) J. Biol. Chem. 271, 26410 –26417. 14. Zhang, Z., Krutchinsky, A., Endicott, S., Realini, C., Rechsteiner, M., and Standing, M. (1999) Biochemistry 38, 5651–5658. 15. Preckel, T., Fung-Leung, W-P., Cai, Z., Vitiello, A., Salter-Cid, L., Winqvist, O., Wolfe, T. G., Von Herrath, M., Angulo, A., Ghazal, P., Lee, J-D., Fourie, A. M., Wu, Y., Pang, J., Ngo, K., Peterrson, P. A., Fru¨h, K., and Yang, Y. (1999) Science 286, 2162–2165. 16. Nikaido, T., Shimada, K., Shibata, M., Hata, M., Sakamoto, M., Takasaki, Y., Sato, C., Takahashi, T., and Nishida, Y. (1990) Clin Exp Immunol. 79, 209 –214. 17. Tanahashi, N., Yokota, K. Y., Ahn, J. Y ., Chung, C. H., Fujiwara, T., Takahashi, E-I, DeMartino, G. D., Slaughter, C. A., Toyonaga, T., Yamamura, K-I., Shimbara, N., and Tanaka, K. (1997) Genes Cells 2, 195–211. 18. Wojcik, C., Tanaka, K., Paweletz, N., Naab, U., and Wilk, S. (1998) Eur. J. Cell. Biol. 77, 151–160. 19. Realini, C. Jensen, C. C., Zhang, Z., Johnston, S. C., Knowlton, J. R., Hill, C. P., and Rechsteiner, M. (1997) J. Biol. Chem. 272, 25483–25492. 20. Kuehn, L., and Dahlmann, B. (1996) FEBS Lett. 394, 183–186. 21. Song, X., von Kampen, J. Slaughter, C. A., and DeMartino, G. N. (1997) J. Biol. Chem. 272, 27994 –28000. 22. Orlowski, M., and Michaud, C. (1989) Biochemistry 28, 9270 – 9278. 23. Wilk, S., and Chen, W-E. Current Protocols in Protein Science, Proteolytic Enzymes (B. Dunn, Ed.), Wiley, in press. 24. Wilk, S., and Orlowski, M. (1980) J. Neurochem. 35, 1172–1182. 25. Wilk, S., Chen, W-E., and Magnusson, R. P. (1998) Arch. Biochem. Biophys. 359, 283–290. 26. Ahn, J. Y., Tanahashi, N., Akiyama, K-Y, Hisamatsu, H., Noda, C., Tanaka, K., Chung, C. H., Shibmara, N., Willy, P. J., Mott, J. D., Slaughter, C. A., and DeMartino, G. N. (1995) FEBS Lett. 366, 37– 42. 27. Wilk, S., Chen, W-E., and Magnusson, R. P. (1999) Arch. Biochem. Biophys. 362, 283–290. 28. Tsubuki, S., Saito, Y., and Kawashima, S. (1994) FEBS Lett. 344, 229 –233. 29. Zaiss, D. M. W., Standera, S., Holzhutter, H., Kloetzel, P-M., and Sijts, A. J. A. M. (1999) FEBS Lett. 457, 333–338. 30. McCutchen-Maloney, S. L., Matsuda, K., Shimbara, N., Binns, D. D., Tanaka, K., Slaughter, C. A., and DeMartino, G. N. (2000) J. Biol. Chem. 275, 18557–18565. 31. Seeger, M., Ferrell, K., Frank, R., and Dubiel, W. (1997) J. Biol. Chem. 272, 8145– 8148. 32. Whitesell, L., Mimnaugh, E. G., De Costa, B., Myers, C. E., and Neckers, L. M. (1994) Proc. Natl. Acad. Sci. USA 91, 8324 – 8328. 33. Guo, G. G., Gui, M., and Etlinger, J. D. (1994) J. Biol. Chem. 269, 12399 –12402.
Properties of the Beta Subunit of the Proteasome Activator PA28 (11S REG) Sherwin Wilk, 1 Wei-Er Chen, and Ronald P. Magnusson Department of Pharmacology, Mount Sinai School of Medicine, One Gustave L. Levy place, New York, New York 10029
Received July 25, 2000, and in revised form August 29, 2000
The proteasome activator PA28 (11S REG) is composed of two homologous subunits termed ␣ and . The properties of the recombinant -subunit were explored and compared to the properties of the recombinant ␣-subunit. PA28 produced in an Escherichia coli expression system migrates on a calibrated gel filtration column as an apparent heptamer (M r ⴝ 250,000). Low concentrations of SDS (0.005%), dissociate the protein to a monomer (M r ⴝ 33,000). PA28 has a complex effect on proteasome activity. At concentrations which favor oligomerization (> 2 M), PA28 is a strong proteasome activator although its affinity for the proteasome is about 10-fold less than recombinant PA28␣. The catalytic properties of the PA28␣ and PA28-activated proteasome are similar. At low concentrations, PA28 is a monomer and a potent allosteric proteasome inhibitor. These studies show that oligomerization of PA28 is required for proteasome activation and that PA28 monomers are potent proteasome inhibitors. © 2000 Academic Press Key Words: proteasome; multicatalytic proteinase complex; proteasome activator; PA28; 11S regulator; proteasome inhibitor.
The 20S proteasome is the catalytic core of the major extralysosomal proteolytic system of the cell. It plays an essential role in many vital cellular processes, including regulation of the cell cycle, degradation of transcription factors and regulatory molecules, and processing of peptides for presentation by the MHC class-I system (1, 2). Early studies demonstrated that distinct active sites of the 20S proteasome cleave peptide bonds after acidic, basic, and hydrophobic amino acids characterizing this macromolecule as a multicatalytic pro1 To whom correspondence and reprint requests should be addressed. Fax: (212) 831-0114. E-mail: [email protected].
174
teinase complex (3). Electron microscopic analysis of the proteasome from Thermoplasma acidophilum revealed a cylindrical structure made up of four stacked heptameric subunit rings (4). X-ray crystallographic analysis of the yeast enzyme demonstrated that the three distinct active sites reside on -type subunits that form the inner rings (5). The active sites are sequestered within a central chamber with no apparent substrate entry channel. This structure serves to protect the cell against uncontrolled proteolysis. It is also evident that regulatory molecules must exist to facilitate substrate entry. Two major types of proteasome regulatory molecules have been described (6). A 19S macromolecular assembly of proteins can cap the 20S proteasome on both ends to form the ubiquitin-protein degrading 26S proteasome (2). The 19S caps presumably recognize the polyubiquitin tag, bind, and unfold the target protein and open an entry port for the protein substrate in an ATP-dependent process (7). A distinct regulator termed 11S REG (8) or PA28 (9) markedly activates the degradation of peptides but not proteins in an ATPindependent fashion. PA28 is composed of two homologous 28-kDa subunits termed ␣ and  (see 10 for a recent review). Recombinant PA28␣ is a heptamer (11). Although endogenous PA28 was proposed to be a hexamer made up of alternating ␣- and -subunits (12, 13), analysis of recombinant PA28␣, by mass spectroscopy is consistent with a heptameric structure (14). The exact subunit arrangement of the ␣,-heptamer is not clear. Apparently only the hetero-oligomer exists in vivo since PA28-deficient mice also do not have detectable PA28␣ (15). However, at present, the existence of free PA28 cannot be ruled out. A third protein member of the PA28 family was identified by a homology search as the previously cloned K i autoantigen (16) and renamed as PA28␥ or REG␥ (17). Immunohistochemical analysis conducted with antibodies directed against ␣-, -, and ␥⫺ proteins show 0003-9861/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
PROTEASOME ACTIVATOR SUBUNIT BETA
that ␣- and -subunits have an identical cellular localization and are predominantly found in the cytosol and nucleolus (18). PA28␥ is found in the nucleus but not nucleolus and in cytoplasmic microtubular-like extensions and inclusion bodies (18). Although there is general agreement concerning the ability of recombinant PA28␣ to activate the proteasome, there is no consensus concerning the properties of PA28. Realini et al. report good activation by PA28 at concentrations higher than needed to see activation by PA28␣ (19). They were not able to isolate an oligomer stable to gel filtration, although they proposed that oligomerization may occur at high concentrations. Two other groups could neither detect formation of a homooligomer nor activation (20, 21). We report that recombinant PA28 can be produced in high concentrations in an E. coli expression system and that the molecule can be isolated as an apparent heptamer by gel filtration. The effect of PA28 on proteasome activity is complex. At low concentrations (⬃0.18 M), PA28 exists as a monomer and is a strong proteasome inhibitor. At relatively high concentrations (⬎ 2 M), oligomerization is favored and PA28 is a strong activator of the proteasome. The catalytic profile of the PA28-activated proteasome is similar to that of the PA28␣⫺activated proteasome although PA28 has a much lower affinity for the proteasome than PA28␣. EXPERIMENTAL PROCEDURES
Materials Bovine pituitary 20S proteasome was purified by a modification of the method of Orlowski and Michaud (22), in which a phenyl Sepharose chromatographic step was substituted for their second anion exchange chromatographic step (23). Succinyl-Leu-Leu-Val-Tyr-7amido-4-methylcoumarin (suc-LLVY-AMC) 2 was purchased from Novabiochem (San Diego, CA). Z-LLE-NA was synthesized as described (24). Z-D-ALR-NA was prepared by conventional solution phase synthesis. All molecular weight markers for gel filtration chromatography were from Sigma, Inc. (St. Louis, MO). The QuikChange Site-directed mutagenesis kit and XL2 Ultracompetent cells were obtained from Stratagene, Inc. (La Jolla, CA). E. coli BL21(DE3) cells and the pET16b vector were from Novagen (Madison, WI). Oligonucleotides were synthesized by Integrated DNA Technologies, Inc. (Coralville, IA). DNA sequencing was performed by the Utah State Biotechnology Center (Logan, UT). Apparently homogeneous recombinant PA28␣ expressed in a baculovirus system was obtained as previously described (25). Hsp-90 was purified to apparent homogeneity from the supernatant fraction of a bovine pituitary homogenate. Hsp-90 copurifies with the 20S proteasome and can be separated after the second anion exchange chromatographic step of the method of Orlowski and Michaud (22). Hsp90 elutes after the proteasome in an essentially homogeneous form.
2 Abbreviations used: AMC, 7-amido-4-methylcoumarin; NA, 2-naphthylamide; Z, N-benzyloxycarbonyl; PGPH, peptidylglutamyl peptide bond hydrolyzing; suc, succinyl; TCA, trichloroacetic acid.
175
Methods Preparation of a pET16b-PA28 expression plasmid. The sequence of the human beta subunit of the proteasome activator PA28 (26) was used to search the NCBI expressed sequence tag (est) data base (dbest). This search revealed a human est (ATCC 1072203). The full-length sequence cloned in the vector pT7T3D-Pac was identical to the input sequence with the following exceptions: Thr 229 in the input sequence (codon ACC) was Asn in the ATCC clone (codon AAC), and the codon for Arg 21 (AGG in the input sequence) was AGA in the ATCC clone. The substitution of Asn for Thr is in agreement with the PA28 sequence obtained by Realini et al. (19). The ATCC clone contains Ala 2 in agreement with the sequence of Ahn et al. (26), whereas the clone described by Realini et al. contains Ser in this position. To subclone the full length insert into the pET-16b prokaryotic expression vector, site-directed mutagenesis was first performed to introduce a NcoI site at the 5⬘ end. Site-directed mutagenesis was performed with the aid of a Stratagene Quik-change kit. Oligonucleotides were designed in which the sequence around the start codon GCATGG was mutated to CCATGG. The oligonucleotides synthesized were for the forward direction 5⬘ CACGGCTTGGCCATGGTGCTTCAGTCGCTAG 3⬘ and for the reverse direction 5⬘ CTAGCGACTGAAGCACCATGGCCAAGCCGTG 3⬘. Since the PA28 insert contains an internal NcoI site, the mutated plasmid was partially digested with NcoI and fully digested with NotI and the full-length insert was purified by gel electrophoresis. The pET-16b vector was digested with NcoI and BamHI. This digest removes the His tag and the factor Xa site. The insert was ligated into the cut vector with the aid of a NotI–BamHI linker. Expression and purification of PA28. E. Coli strain BL21(DE3) was transformed with the pET16b-PA28 plasmid. A single colony was added to 25 ml LB medium and incubated for 4.5 h at 37°C. The culture was then transferred to 3 liters LB medium and grown overnight at room temperature. IPTG was then added (0.2 mg/ml) and the material collected after 7 h incubation at room temperature. After centrifugation, the pellet was extracted with 10 mM Tris–HCl, pH 7.5. Greater than 50% of the expressed protein was present in the soluble fraction of the cell. PA28 was purified by sequential FPLC on mono-Q, phenyl Superose, and Superose 6 columns as described for PA28␣ (25). Purification was monitored by SDS–PAGE analysis of the column fractions. The resulting protein was judged to be more than 95% pure by SDS–PAGE (Fig. 1). Measurement of enzymatic activity. Enzymatic activity was measured with the substrates suc-LLVY-AMC, Z-D-ALR-NA, and Z-LLENA. Enzyme and substrate in the presence or absence of PA28 proteins at the concentrations designated in the figure legends were incubated in a 0.05 M Tris–HCl buffer, pH 7.5, at a final volume of 100 l. Reactions were terminated with 100 l 10% TCA and the chromogen released was measured spectrophotometrically at 560 nm (AMC) or 580 nm (NA) by a diazotization procedure (24). HPLC analysis. HPLC was performed on a Waters Model 600E Instrument fitted with a 4.6 mm ⫻ 25 cm Supelcosil 5 m C8 column (Supelco). The column was equilibrated with 15% acetonitrile/ 0.05%TFA at a flow rate of 1.0 ml/min. Elution was carried out by linearly increasing the acetonitrile concentration to 55% over a 34min period. Eluting peaks were detected by absorbance at 210 nm at a sensitivity of 0.1 AUFS. Determination of relative molecular mass (M r): (a) Superose 6 chromatography. FPLC was conducted on a Pharmacia system. A Superose 6 column (Pharmacia) was equilibrated with 0.05 M Tris– HCl/0.1 M NaCl. The column was calibrated with thyroglobin (669,000), ferritin (440,000), catalase (232,000), and aldolase (158,000). Blue dextran was used to measure the void volume. M r was determined from a plot of log M r vs K where K ⫽ V-V o/V t-V o, V o
176
WILK, CHEN, AND MAGNUSSON
PA28 Has a Biphasic Effect on the Chymotrypsinlike and PGPH Activities of the Proteasome
FIG. 1. SDS–PAGE (12.5% gel) analysis of the purification of expressed PA28. Lanes 1 and 6, molecular mass markers in kilodaltons; lane 2, whole cell extract; lane 3, active fraction after mono Q chromatography; lane 4, active fraction after phenyl Superose chromatography; lane 5 active fraction after Superose 6 chromatography. Proteins stained with Coomassie blue.
The bovine pituitary 20S proteasome was incubated with various concentrations of recombinant human PA28 and the hydrolysis of suc-LLVY-AMC was determined. A biphasic effect was observed. At concentrations of PA28 less than 2 M inhibition occurred whereas at higher concentrations the chymotrypsinlike activity was stimulated (Fig. 2). Maximal stimulation in different experiments varied from three to fivefold. A similar effect was noted for the PGPH activity although somewhat higher concentrations of PA28 were required for stimulation (Fig. 2). Stimulation of the trypsinlike activity was more modest and inhibition was not readily apparent at low concentrations of PA28 (Fig. 2). Since inhibition at low concentrations of PA28 has previously not been described, this phenomenon was analyzed in greater detail. Maximal inhibition of the chymotrypsinlike activity occurred at a PA28 concentration of 180 nM with 50% inhibition at 25 nM (Fig. 3). A molecular weight of 28,000 was used to calculate molar concentrations. As a control we evaluated inhibition of the proteasome by BSA and obtained an EC50 of 165 nM approximately sevenfold greater than the PA28 monomer (Fig. 3).
is the column void volume, V t the total volume, and V the elution volume of the protein of interest. Proteins were detected at 254 nm. (b) Superose 12 chromatography. A Superose 12 column (Pharmacia) was equilibrated with 0.05 M Tris–HCl/0.1 M NaCl. The column was calibrated with bovine serum albumin (66,000), ovalbumin (45,000), carbonic anhydrase (29,000), ␣-lactalbumin (14,300), and the oxidized B chain of insulin (3500). M r was determined as described above for Superose 6 chromatography.
RESULTS
Recombinant PA28 Is Highly Expressed in E. coli and Readily Purified In an earlier study we were able to obtain very high expression of the alpha subunit of PA28 in a baculovirus system (25). However, we were not able to obtain a similar expression level for the beta subunit in baculovirus infected Sf-9 cells. We therefore explored the expression of PA28 in E. coli. Induction by IPTG at room temperature as described under Experimental Procedures gave high expression with at least half of the expressed protein present in the soluble fraction. We then applied the same FPLC procedure used to purify PA28␣ (25) and obtained the beta subunit in greater than 95% purity (Fig. 1). To obtain active PA28 it was essential to conduct the sequential steps continuously or if necessary to store active fractions at ⫺80°C between steps.
FIG. 2. Effect of PA28 on the catalytic activities of the 20S proteasome. Incubation mixtures in a total volume of 100 l contained 1.1 g proteasome, PA28 at the concentrations shown on the abscissa, a substrate concentration of 0.4 mM and 0.05 M Tris–HCl, pH 7.5. The chymotrypsinlike activity (ChT-L) was measured with sucLLVY-AMC, the PGPH activity with Z-LLE-NA and the trypsinlike activity (T-L) with Z-D-ALR-NA. Samples were incubated at 37°C for 30 min and the chromogen released determined as described under Experimental Procedures. Velocity is given on the ordinate as absorbance units at 560 nm for AMC and 580 nm for NA.
PROTEASOME ACTIVATOR SUBUNIT BETA
FIG. 3. Comparison of the inhibition of the 20S proteasome-catalyzed hydrolysis of suc-LLVY-AMC by low concentrations of PA28 and BSA. Incubation conditions as described in the legend to Fig. 2. Concentrations of protein inhibitors shown on the abscissa. PA28 (E); BSA (F).
Low Concentrations of PA28 Act as a Negative Allosteric Effector The mechanism of inhibition of the chymotrypsinlike activity of the proteasome by PA28 was studied by determining inhibition as a function of substrate concentration. In this experiment, incubations were conducted in the presence of 170 nM PA28, 16 nM proteasome and substrate concentrations of suc-LLVYAMC ranging from 0 to 50 mM. In the absence of PA28, the degradation of suc-LLVY-AMC displayed normal Michaelis–Menten kinetics (Fig. 4). It should be noted however that some preparations of the proteasome assayed with this substrate exhibit cooperativity. These differences may be due to conformational changes in the protein during the purification proce-
FIG. 4. Inhibition of the chymotrypsinlike activity of the 20S proteasome by low concentrations of PA28 as a function of substrate concentration. Incubation mixtures in a total volume of 100 l contained proteasome at a concentration of 16 nM, suc-LLVY-AMC at concentrations shown on the abscissa, 0.05 M Tris–HCl, pH 7.5, and either 170 nM PA28 (F) or no PA28 (E). Velocity is given on the ordinate as absorbance units at 560 nm.
177
FIG. 5. Elution of PA28 on a calibrated Superose 6 column. The elution volume of PA28 was slightly less than that of carbonic anhydrase (232 kDa; 15.3 ml). The arrow designates the elution volume of PA28 monomer.
dure. Low concentrations of PA28 greatly increased cooperativity resulting in a large increase in Km but an unchanged Vmax (Fig. 4). The data were analyzed by a Hill plot. The Hill coefficient increased from unity in the absence of PA28 to 2.4 in its presence. Stimulation or Inhibition Depends on the Oligomeric State of PA28 PA28 is reported in the literature to exist as a monomer (21). It is also reported that the recombinant -subunit does not form a hexamer or heptamer stable to gel filtration (19). It was therefore surprising to observe that PA28 after purification on a calibrated Superose 6 column migrated predominantly as a species with an elution volume nearly identical to that of heptameric PA28␣. The M r was calculated as 250,000. A minor component with an elution volume consistent with that of the monomer was seen as was a small peak between the two which may represent an intermediate form (Fig. 5). The ability of PA28 to form a homooligomer and stimulate the proteasome is undoubtedly a concentration-dependent effect as suggested by Realini et al. (19). On the other hand we found strong inhibition of the chymotrypsinlike activity of the proteasome at low concentrations of PA28. We previously reported that heptameric PA28␣ could be dissociated by very low concentrations of SDS (0.005%) and that the dissociated protein could not stimulate the proteasome (25). To determine that monomeric PA28 is the inhibitory species, we incubated the proteasome with a relatively high concentration of PA28 (5.4 M) in the presence and absence of 0.005% SDS. In the absence of SDS this concentration of PA28 stimulated the chymotrypsinlike activity 3.7-fold. When the incubation was conducted in the presence of 0.005% SDS, activity
178
WILK, CHEN, AND MAGNUSSON
Monomeric PA28 Is a Stronger Proteasome Inhibitor Than Hsp90
FIG. 6. Comparison of the stimulation of the chymotrypsinlike activity of the 20 S proteasome by PA28␣ and PA28 as a function of activator concentration. Incubation mixtures in a total volume of 100 l contained 0.38 g proteasome, 0.4 mM suc-LLVY-AMC, PA28 ␣ or  at the concentrations shown on the abscissa, and 0.05 M Tris–HCl, pH 7.5. Molar concentrations of PA28␣ and PA28 were calculated using the molecular weight of the monomer (28,000). Samples were incubated at 37°C for 30 min and the AMC released determined as described under Experimental Procedures. Velocity is given on the ordinate as absorbance units at 560 nm.
Other proteins have been reported to inhibit the proteasome. One of the most potent inhibitors is heat shock protein 90 (Hsp90) (28). It was therefore of interest to compare the inhibition of the proteasome produced by monomeric PA28 to that of Hsp90. Bovine pituitary Hsp90 was purified to apparent homogeneity as described under Experimental Procedures. Hsp90 strongly inhibited the proteasome with an IC 50 of 50 nM measured at a proteasome concentration of 12.5 nM (Fig. 8). The EC50 is greater than that found for PA28 (Fig. 3). Unlike other inhibitory proteins such as PI31 (29, 30) or HIV-tat (31), Hsp-90 did not antagonize the activation of the proteasome by PA28␣ or PA28 (not shown). Geldanamycin, a benzoquinone ansamycin antagonist of Hsp90 (32) tested at a concentration of 50 M only slightly attenuated inhibition of the proteasome by Hsp-90 (not shown).
was totally inhibited. The detergent alone had no effect on catalytic activity. The M r of the 0.005%-treated protein on a calibrated Superose 12 column was 33,000. PA28 Is a Strong Proteasome Activator But Is Less Potent Than PA28␣ Stimulation of the chymotrypsinlike activity of the proteasome by the purified recombinant PA28 ␣ and  proteins was compared. At the substrate concentration used, PA28␣ maximally stimulated the chymotrypsinlike activity approximately six-fold. PA28 stimulated this activity approximately five fold (Fig. 6). The EC50 of PA28 was about ten-fold higher than the EC50 of PA28␣. Unlike the beta subunit, low concentrations of PA28␣ did not inhibit the proteasome. We have previously used the ovalbumin immunodominant epitope SIINFEKL to study proteasome activation since this peptide is only slowly cleaved by the latent proteasome. In the presence of PA28␣ it is cleaved after both hydrophobic and acidic amino acids (27). After a 1 h incubation of SIINFEKL with the latent proteasome, there was no detectable hydrolysis. Addition of either PA28 or PA28␣ markedly stimulated degradation to SIINFE and SIINF (Fig. 7). It should be noted that in these experiments the concentration of PA28 exceeded that of PA28␣ by eightfold.
FIG. 7. HPLC analysis of the degradation of SIINFEKL by the latent proteasome, the PA28 activated proteasome, and the PA28␣ activated proteasome. Incubation mixtures in a total volume of 125 l contained 1.52 g 20S proteasome, SIINFEKL at a final concentration of 200 M, 0.05 M Tris–HCl buffer, pH 7.5, and either no activator (A) or 40 g PA28 (B) or 5 g PA28␣ (C). Samples were incubated for 1 h at 37°C. HPLC was carried out as described under Experimental Procedures. Peak 1, SIINFE; peak II, SIINF; peak III, SIINFEKL. Inj designates point of injection.
PROTEASOME ACTIVATOR SUBUNIT BETA
FIG. 8. Inhibition of the chymotrypsinlike activity of the 20S proteasome by Hsp90. Incubation mixtures in a total volume of 100 l contained 1.34 g proteasome, 0.4 mM suc-LLVY-AMC, Hsp90 in the concentrations shown on the abscissa and 0.05 M Tris–HCl, pH 7.5. Samples were incubated at 37°C for 30 min and the AMC released determined as described under Experimental Procedures. Velocity is given on the ordinate as absorbance units at 560 nm.
DISCUSSION
There is disagreement in the literature as to whether PA28 can stimulate the proteasome (19 –21). We clearly show that this protein at relatively high concentrations is an effective activator. It is further evident from these studies that the oligomeric state of PA28 determines its effect on the catalytic activities of the proteasome. Activation is associated with formation of an oligomer, apparently a heptamer. In agreement with the findings of Realini et al. (19) and in contrast to other reports (20, 21) we observe strong stimulation by high concentrations of PA28 although the magnitude of stimulation does not reach that produced by PA28␣. Moreover equivalent proteasome activation by the recombinant alpha subunit occurs at about a 10-fold lower molar concentration than the beta subunit (Fig. 6). The association of PA28 monomers is weak and concentration dependent. By contrast, the monomer-heptamer equilibrium of PA28␣ is strongly shifted toward the heptamer (25). Other investigators have been unable to obtain active PA28 or if active have not been able to isolate an oligomer stable to gel filtration (19, 21). This is explained by the relative instability of PA28 and also by the concentration dependence of oligomerization. We found active PA28 to be less stable than active PA28␣. For this reason PA28 must be rapidly purified and we have successfully used an FPLC procedure. Moreover if the protein is not immediately taken from one step to the
179
next it is imperative to store the solution at ⫺80°C to preserve activity. In addition good expression is necessary to obtain a preparation of PA28 stable to gel filtration (Fig. 1). The stimulation of the three catalytic activities of the 20 S proteasome by PA28 differs. The strongest effect is exerted on the chymotrypsinlike activity. The PGPH activity is more moderately enhanced and the trypsinlike activity is least affected (Fig. 2). Similar results were obtained by Realini et al. (19). If the effect of binding PA28 to the outer ␣-rings of the proteasome were only to open channels for the access of substrate, one might expect a similar activation of the three activities. Since the active sites responsible for the three activities reside on distinct subunits, the differential effect is most likely due to alterations of the geometries of the active sites although one cannot rule out facilitation of substrate access as also contributing to the overall stimulation. Low concentrations of PA28 strongly inhibit the proteasome. It is unclear why others have not noted such inhibition. Realini et al. (19) also report a biphasic effect of PA28 but they observe modest stimulation of suc-LLVY-AMC hydrolysis at low concentrations and stimulation equivalent to PA28␣ at high concentrations. By contrast we find marked inhibition of the chymotrypsinlike activity at low concentrations of PA28. We have shown that very low concentrations of SDS (0.005%) dissociate PA28 and that this dissociated and inhibitory protein is a monomer. Since oligomer formation is necessary for stimulation, it is difficult to account for the reported stimulation by low concentrations of this protein. Our results demonstrate that monomeric PA28 can bind to the proteasome but that the proteasome does not serve as a template for assembly of the oligomer. Binding of monomeric PA28 to the proteasome inhibits catalytic activity and inhibition is clearly allosteric (Fig. 4). This is expected since this protein should not directly contact the active sites of the latent 20S proteasome. PA28 inhibited the hydrolysis of sucLLVY-AMC by 50% at a concentration of 25 nM, with maximal inhibition at a concentration of 185 nM. In these experiments the proteasome concentration was 15.7 nM and since each proteasome molecule contains 14 alpha subunits, the concentration of proteasome alpha subunits was 220 nM. Assuming that PA28 monomers can bind to all alpha subunits, maximal inhibition occurs at approximately a 1:1 molar ratio. Several other proteins have been reported to strongly inhibit the proteasome (Table I). We compared the inhibition by PA28 monomers to that produced by Hsp90 (26) and found the IC50 of PA28 to be lower than the IC50 of Hsp 90 (Table I). The proteasome inhibitor PI31 was recently characterized. Zaiss et al. (29) report that at a proteasome concentration of 3 nM,
180
WILK, CHEN, AND MAGNUSSON TABLE I
Comparison of the Potencies of Protein Inhibitors of the Proteasome
Inhibitor PA28 monomer Hsp90 PI31 Hiv-Tat ␦-Aminolevulinate dehydratase
I.C.50 (nM) 25 55 55 50 50 ⬃75
Proteasome Conc. (nM)
Ref.
15.7 12.5 3 41 1.4
This study This study 29 30 31
43
33
50% inhibition occurred at a concentration of PI31 of about 55 nM. McCutchen-Maloney et al. report 50% inhibition at about a 1:1 molar ratio of PI31 to proteasome (30). Therefore the PA28 monomer appears to be an even more potent inhibitor than PI31. HIV-1 Tat inhibits hydrolysis of suc-LLVY-AMC by 50% at a concentration of 50 nM inhibitor and 1.4 nM proteasome (31). ␦-aminolevulinate dehydratase another characterized proteasome inhibitor produces 50% inhibition at a 2:1 molar ratio of inhibitor to proteasome (33). These inhibitory proteins differ with respect to their antagonism of the activation of PA28. Both PI31 and HIV-Tat antagonize activation of the proteasome by PA28 (29 –31), whereas PA28 monomer and Hsp90 do not (data not shown). In summary, we have shown that the beta subunit of PA28 has a complex effect on the chymotrypsinlike (and PGPH) activities of the proteasome. At high concentrations this subunit oligomerizes and stimulates the proteasome. At low concentrations where the subunit exists as a monomer, the proteasome is strongly inhibited. The inhibition is as strong or stronger than other described protein inhibitors of the proteasome. Whether this inhibition represents a control mechanism in vivo remains speculative since it is not known whether free PA28 exists in the cell. ACKNOWLEDGMENT These studies were supported by NIH Grant NS 29936.
REFERENCES 1. Tanaka, K. (1998) J. Biochem. 123, 195–204. 2. Coux, O., Tanaka, K., and Goldberg, A. L. (1996) Annu. Rev. Biochem. 65, 801– 847. 3. Wilk, S., and Orlowski, M. (1983) J. Neurochem. 40, 842– 849. 4. Pu¨hler, G., Weinkauf, S., Bachman, L., Muller, S., Engel, A., Hegerl, R., and Baumeister, W. (1992) EMBO J. 11, 1607–1616. 5. Groll, M., Ditzel, L., Lowe, J., Stock, D., Bochtler, M., Bartunik, H. D., and Huber, R. (1997) Nature 386, 463– 471. 6. DeMartino, G. N., and Slaughter, C. A. (1999) J. Biol. Chem. 274, 22123–22126.
7. Gottesman, S., Maurizi, M. R., and Wickner, S. (1997) Cell 91, 435– 438. 8. Dubiel, W., Pratt, G., Ferrell, K., and Rechsteiner, M. (1992) J. Biol. Chem. 267, 22369 –22377. 9. Chu-Ping, M. Slaughter, C. A., and DeMartino, G. N. (1992) J. Biol. Chem. 267, 10515–10523. 10. Rechsteiner M., Realini, C., and Ustrell, V. (2000) Biochem. J. 345, 1–15. 11. Johnston, S. C., Whitby, F. G., Realini, C., Rechsteiner, M., and Hill, C. P. (1997) Protein Sci. 6, 2469 –2473. 12. Ahn, K., Erlander, M., Letureq, D., Peterson, P. A., Fru¨h, K., and Yang, Y. (1996) J. Biol. Chem. 271, 18237–18242. 13. Song, X., Mott, J. D., von Kampen, J., Pramanik, B., Tanaka, K., Slaughter, C. A., and DeMartino, G. D. (1996) J. Biol. Chem. 271, 26410 –26417. 14. Zhang, Z., Krutchinsky, A., Endicott, S., Realini, C., Rechsteiner, M., and Standing, M. (1999) Biochemistry 38, 5651–5658. 15. Preckel, T., Fung-Leung, W-P., Cai, Z., Vitiello, A., Salter-Cid, L., Winqvist, O., Wolfe, T. G., Von Herrath, M., Angulo, A., Ghazal, P., Lee, J-D., Fourie, A. M., Wu, Y., Pang, J., Ngo, K., Peterrson, P. A., Fru¨h, K., and Yang, Y. (1999) Science 286, 2162–2165. 16. Nikaido, T., Shimada, K., Shibata, M., Hata, M., Sakamoto, M., Takasaki, Y., Sato, C., Takahashi, T., and Nishida, Y. (1990) Clin Exp Immunol. 79, 209 –214. 17. Tanahashi, N., Yokota, K. Y., Ahn, J. Y ., Chung, C. H., Fujiwara, T., Takahashi, E-I, DeMartino, G. D., Slaughter, C. A., Toyonaga, T., Yamamura, K-I., Shimbara, N., and Tanaka, K. (1997) Genes Cells 2, 195–211. 18. Wojcik, C., Tanaka, K., Paweletz, N., Naab, U., and Wilk, S. (1998) Eur. J. Cell. Biol. 77, 151–160. 19. Realini, C. Jensen, C. C., Zhang, Z., Johnston, S. C., Knowlton, J. R., Hill, C. P., and Rechsteiner, M. (1997) J. Biol. Chem. 272, 25483–25492. 20. Kuehn, L., and Dahlmann, B. (1996) FEBS Lett. 394, 183–186. 21. Song, X., von Kampen, J. Slaughter, C. A., and DeMartino, G. N. (1997) J. Biol. Chem. 272, 27994 –28000. 22. Orlowski, M., and Michaud, C. (1989) Biochemistry 28, 9270 – 9278. 23. Wilk, S., and Chen, W-E. Current Protocols in Protein Science, Proteolytic Enzymes (B. Dunn, Ed.), Wiley, in press. 24. Wilk, S., and Orlowski, M. (1980) J. Neurochem. 35, 1172–1182. 25. Wilk, S., Chen, W-E., and Magnusson, R. P. (1998) Arch. Biochem. Biophys. 359, 283–290. 26. Ahn, J. Y., Tanahashi, N., Akiyama, K-Y, Hisamatsu, H., Noda, C., Tanaka, K., Chung, C. H., Shibmara, N., Willy, P. J., Mott, J. D., Slaughter, C. A., and DeMartino, G. N. (1995) FEBS Lett. 366, 37– 42. 27. Wilk, S., Chen, W-E., and Magnusson, R. P. (1999) Arch. Biochem. Biophys. 362, 283–290. 28. Tsubuki, S., Saito, Y., and Kawashima, S. (1994) FEBS Lett. 344, 229 –233. 29. Zaiss, D. M. W., Standera, S., Holzhutter, H., Kloetzel, P-M., and Sijts, A. J. A. M. (1999) FEBS Lett. 457, 333–338. 30. McCutchen-Maloney, S. L., Matsuda, K., Shimbara, N., Binns, D. D., Tanaka, K., Slaughter, C. A., and DeMartino, G. N. (2000) J. Biol. Chem. 275, 18557–18565. 31. Seeger, M., Ferrell, K., Frank, R., and Dubiel, W. (1997) J. Biol. Chem. 272, 8145– 8148. 32. Whitesell, L., Mimnaugh, E. G., De Costa, B., Myers, C. E., and Neckers, L. M. (1994) Proc. Natl. Acad. Sci. USA 91, 8324 – 8328. 33. Guo, G. G., Gui, M., and Etlinger, J. D. (1994) J. Biol. Chem. 269, 12399 –12402.