- Email: [email protected]
Identification of three phosphorylation sites in the α7 subunit of the yeast 20S proteasome in vivo using mass spectrometry
ABB Archives of Biochemistry and Biophysics 431 (2004) 9–15 www.elsevier.com/locate/yabbi
Identification of three phosphorylation sites in the a7 subunit of the yeast 20S proteasome in vivo using mass spectrometry Yuko Iwafune, Hiroshi Kawasaki*, Hisashi Hirano Kihara Institute for Biological Research/Graduate School of Integrated Science, Yokohama City University, Maioka 641-12, Totsuka-ku, Yokohama 244-0813, Japan Received 26 March 2004, and in revised form 20 July 2004 Available online 20 August 2004
Abstract The 26S proteasome complex, which consists of a 20S proteasome and a pair of 19S regulatory particles, plays important roles in the degradation of ubiquitinated proteins in eukaryotic cells. The a7 subunit of the budding yeast 20S proteasome is a major phosphorylatable subunit; serine residue(s) in its C-terminal region are phosphorylated in vitro by CKII. However, the exact in vivo phosphorylation sites have not been identified. In this study, using electrospray ionization quadrupole time-of-flight mass spectrometry analysis, we detected a mixture of singly, doubly, and triply phosphorylated C-terminal peptides isolated from a His-tagged construct of the a7 subunit by nickel-immobilized metal affinity chromatography. In addition, we identified three phosphorylation sites in the C-terminal region using MS/MS analysis and site-directed mutagenesis: Ser258, Ser263, and Ser264 residues. The MS/MS analysis of singly phosphorylated peptides showed that phosphorylation at these sites did not occur successively. Ó 2004 Elsevier Inc. All rights reserved. Keywords: Proteasome; 20S proteasome; a7; Phosphorylation; Mass spectrometry; Yeast; Casein kinase II; Identification; Modification; Post-translational modification
Selective protein degradation in cells is important for regulating cellular functions such as cell cycle progression, signal transduction, and protein quality control. The ubiquitin/26S proteasome pathway is involved in this degradation [1]. The 26S proteasome complex consists of a 20S proteasome and a pair of 19S regulatory particles [2,3]. The 20S proteasome is a hollow cylinder composed of four stacked heptamer rings of a- and btype subunits [2]. The eukaryotic 20S proteasome has three protease activities: caspase-, trypsin-, and chymotrypsin-like activities [4,5]. The 19S regulatory particle consists of six ATPase subunits and at least 11 nonATPase subunits, and it associates with the 20S proteasome in an ATP-dependent manner to form the 26S proteasome [3]. *
Corresponding author. Fax: +81 45 820 1901. E-mail address: [email protected] (H. Kawasaki).
0003-9861/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2004.07.020
The 26S proteasome is post-translationally modified by phosphorylation, N-acetylation, and N-myristoylation in various species [6–20]. Of these post-translational modifications, the phosphorylation of serine, threonine, or tyrosine residues is a key event for signal transduction, cell cycle progression, and protease activity. Phosphorylation plays an important role in the ubiquitin/26S proteasome pathway. Phosphorylation often serves to target substrates to their cognate E3 enzymes [1]. In addition, some subunits of the 26S proteasome are phosphorylated, affecting proteasome functions such as the association between the 20S proteasome and 19S regulatory particle and the protease activity of the 20S proteasome. Phosphorylation regulates the formation and function of proteasome complexes. Some subunits of the 20S proteasome and 19S regulatory particle are phosphorylated in yeast and mammals [6–9,14,20]. The a7 subunit in the 20S proteasome is
10
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
the major phosphorylated subunit in both. The a7 subunit is a substrate of CKII in vivo and in vitro [6,7,14]. Protein kinase CKII (formerly known as casein kinase II) is a ubiquitous, highly conserved Ser/Thr kinase found in all eukaryotes. In budding yeast and mammals, the CKII phosphorylation sites have been mapped to Ser/Thr residues in the C-terminal region. The sequence of the C-terminal region of the a7 subunit is divergent in budding yeast and mammals; nevertheless, both have putative phosphorylation sites for CKII. Castano et al. [7] reported that two Ser residues of the mammalian a7 subunit, Ser243 and Ser250, possess phosphorylation motifs for CKII [7]. Mutagenesis experiments revealed that these residues were phosphorylation sites in vitro [7]. By contrast, in vivo phosphorylation of the human a7 subunit is reported to occur at Ser250 only [8], suggesting the presence of a mechanism regulating phosphorylation. The in vivo phosphorylation sites of the yeast a7 subunit have not yet been identified. It is important to identify the in vivo phosphorylation sites to understand the physiological roles of the phosphorylation of the a7 subunit. In this paper, we identified Ser258, Ser263, and Ser264 as in vivo phosphorylation sites of the yeast a7 subunit using mass spectrometry and revealed that phosphorylation at these three sites does not occur successively.
Material and methods Yeast strains and medium Yeast cells were grown at 30 °C in YPD medium containing 1% yeast extract, 2% Bacto–peptone, 2% glucose, and 0.5 lg ml1 aureobasidin A (Takara, Otsu, Japan). Yeast cells were harvested at late log phase (OD at 600 nm, 2.5–3.5) for proteasome purification. The a7 deletion mutant Saccharomyces cerevisiae BY4743-YOR362C (ATCC No. 4021659) was purchased from the American Type Culture Collection. Plasmid pYC1 carrying the a7 gene of yeast 20S proteasome was a kind gift from Dr. K. Tanaka (The Tokyo Metropolitan Institute of Medical Science, Tokyo) [21]. Construction of an expression vector for the His6-tagged a7 subunit of proteasome A His6-tag was added at the C-terminus of the a7 subunit. The a7 subunit gene with the His6-tag was amplified by PCR using pYC1 plasmid as the template and the following primers: 5 0 -CGCCCGGGTCTTCAGC AATGACATCAATTGG-3 0 and 5 0 -GCGAGCTCTCA GTGGTGGTGGTGGTGGTGTTCTAGGTGAATAT CACCCTCTTGG-3 0 . The PCR fragment was cloned
into the SacI and SmaI sites of pBluescript II SK () (Stratagene, CA, USA). After confirming the sequence, the plasmid was digested with SacI and SmaI, and the fragment was transferred to the shuttle vector pAUR123 (Takara, Otsu, Japan). Yeast cells (BY4743-YOR362C) were transformed using the aureobasidin A yeast transformation system (Takara, Otsu, Japan) according to the manufacturerÕs instructions.
Purification of the 20S proteasome including the His6tagged a7 subunit The 20S proteasome including the a7 subunit with the His6-tag was purified using nickel-immobilized metal affinity chromatography (IMAC) with TOYOPEARL AF-Chelate-650M (Tosoh, Tokyo, Japan) and anion exchange chromatography using Poros HQ/L (Applied Biosystems, Uppsala, Sweden). The purified 20S proteasome including the His6-tagged a7 subunit was separated by SDS–PAGE. The a7 subunit with the His6-tag was detected by Western blotting using anti-penta His monoclonal antibody (Qiagen, Hilden, Germany). The proteins were stained using a Negative Gel Stain MS kit (Wako Pure Chemical Industries, Osaka, Japan). The gel pieces containing protein bands were cut out and destained with the destaining solution provided in this kit. After they were washed with MilliQ water three times and with acetonitrile once, they were dried using a vacuum centrifuge. The gel pieces were rehydrated in 100 mM Tris–HCl, pH 9.0, containing 1 ng ll1 lysylendopeptidase (Wako Pure Chemical Industries, Osaka, Japan) for 16 h at 37 °C. The digests were concentrated to 20 ll using a vacuum centrifuge. Isolation of the C-terminal peptide and analysis using ESI/Q-TOF MS The C-terminal peptide tagged with the His6-tag was isolated from the digest using IMAC chromatography. The peptides were concentrated to 20 ll using a vacuum centrifuge and desalted using a ZipTipl-C18 column (Millipore, Bedford, MA, USA) according to the manufacturerÕs instructions. The peptides were loaded into a borosilicate nanoflow tip (Micromass, Manchester, UK) and subjected to electrospray ionization quadrupole time-of-flight mass spectrometry (ESI/Q-TOF MS)1 (Micromass, Manchester, UK) with positive ion detection mode. The MS and MS/MS spectra were ana-
1
Abbreviations used: CBB, Coomassie brilliant blue; ESI, electrospray ionization; MS, mass spectrometry; Q-TOF, quadrupole time-offlight mass spectrometer; SDS–PAGE, SDS–polyacrylamide gel electrophoresis; IMAC, immobilized metal affinity chromatography; OD, optical density
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
lyzed using MassLynx v3.4 software (Micromass, Manchester, UK). The MS/MS spectrum was processed with MaxEnt3 program in MassLynx package. The deconvoluted single-charge spectrum was used for amino acid sequencing. Statistical analysis All data are expressed as means ± SE and represent three experiments. The significance of differences was analyzed using the t test, and P < 0.05 was taken as statistically significant.
11
purified 20S proteasome by SDS–PAGE. The His6tagged a7 subunit was the largest subunit in the yeast 20S proteasome; it was detected with anti-penta-His monoclonal antibody and identified using peptide mass fingerprinting. All His6-tagged a7 subunits were associated with the 20S proteasome in yeast cells; no free form of the His6-tagged a7 subunit was observed. The purified 20S proteasome including the His6-tagged a7 subunit had protease activity toward the Suc-LLVY-MCA substrate equal to that of the native 20S proteasome. Mass spectrometric analysis of the C-terminal peptides of the His6-tagged a7 subunit
Site-directed mutagenesis The mutation of Ser263 or Ser264 to Ala263 or Ala264 was carried out by PCR using the previously constructed vector encoding the His6-tagged a7 subunit as template and the following primers: S263A_F (AACGT CATGG CTAGTGATGATGAA), S263A_R (TTCAT CATCAC TAGCCATGACGTT), S264A_F (GTCATGTCCGC TGATGATGAAAAT), and S264A_R (ATTT TCAT CAT CAGCGGACATGAC) [22].
Results Purification of the 20S proteasome The 20S proteasome including the His6-tagged a7 subunit was purified using IMAC and anion exchange chromatography. Fig. 1 shows the subunit separation of the
Fig. 1. Purification of the 20S proteasome containing the His6-tagged a7 subunit. Purified 20S proteasome including the His6-tagged a7 subunit was separated by 12% SDS–PAGE. 1, Subunits were stained with CBB. 2, The His6-tagged a7 subunit was detected using antipenta-His monoclonal antibody (Qiagen).
A lysylendopeptidase digest of the a7 subunit was subjected to IMAC to isolate the His6-tagged C-terminal peptides. The eluant from IMAC was analyzed using ESI/Q-TOF MS. The mass spectrum of the C-terminal peptide of the a7 subunit, 246EINGDDDEDEDDSDN VMSSDDENAPVATNANATTDQEGDIHLEHHHH HH294, is shown in Fig. 2. The three major peaks (m/ z = 1138.2 [M + 5H], 1422.5 [M + 4H], and 1896.7 [M + 3H]) correspond to the mass of the His6-tagged C-terminal peptide phosphorylated at three sites (theoretical mass = 5686 Da). The minor peaks (m/z = 1402.3 [M + 4H] and 1870.3 [M + 3H]; m/z = 1382.6 [M + 4H]) also correspond to the His6-tagged C-terminal peptide phosphorylated at two (theoretical mass = 5606 Da) or one (theoretical mass = 5526 Da) sites, respectively. The peaks m/z = 1426.8 [M + 4H] and 1902.0 [M + 3H] correspond to the triply phosphorylated peptide with a mass increase of 16 Da by oxidation of the Met residue. No unphosphorylated C-terminal peptide was observed. When the relative amount of the triply phosphorylated peptide was defined as 100, the amount of the doubly phosphorylated peptide was less than 40 ± 10 and the
Fig. 2. MS analysis of the C-terminal peptides. The MS spectrum of the m/z range between 1000 and 2000 is shown. Peak identification is as follows. Peaks m/z = 1138.2 [M + 5H], 1422.5 [M + 4H], and 1896.7 [M + 3H] represent three phosphate groups. Peaks m/z = 1402.3 [M + 4H] and 1870.3 [M + 3H] indicate two phosphate groups. Peak m/z = 1382.6 [M + 4H] identifies one phosphate group. Peaks m/ z = 1426.8 [M + 4H] and 1902.0 [M + 3H] signify three phosphate groups and an oxidized Met residue.
12
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
amount of the singly phosphorylated peptide was less than 10 ± 7. The formation of positive ions from the peptide phosphorylated at three sites should be suppressed more than in the peptide phosphorylated at two sites or one site due to the strongly negative charge of the phosphate group. However, the signal of the peptide phosphorylated at three sites was the strongest. These results suggest that the three sites in the C-terminal region are phosphorylated in nearly all a7 subunits in yeast cells in vivo. We observed the neutral loss of H3PO4 (decrease of 98 Da per phosphate group) in the MS/MS analysis of these peptides. However, the MS/MS spectra could not identify the phosphorylation sites because of the strongly negative charge of the phosphate group. We replaced each Ser residue at the putative phosphorylation sites in the C-terminal region of the a7 subunit with an Ala residue using site-directed mutagenesis to resolve the difficulty. Identification of in vivo phosphorylation sites using MS/ MS analysis Each C-terminal peptide of the a7 subunit from two mutants, S263A and S264A, was analyzed using ESI/QTOF MS. Fig. 3 shows the MS spectra of the C-terminal peptides from the wild type and the S263A and S264A mutants. Three major peaks from the wild type (m/ z = 1138.2 [M + 5H], 1122.2 [M + 5H], and 1106.2 [M + 5H]) were assigned to peptides phosphorylated at three (theoretical mass = 5686 Da), two (theoretical mass = 5606 Da), or one (theoretical mass = 5526 Da) site(s). The two major peaks observed in the spectra of S263A and S264A (m/z = 1119.0 [M + 5H] and 1103.0 [M + 5H]) were assigned to peptides phosphorylated at two (theoretical mass = 5590.4 Da) or one (theoretical mass = 5510.4 Da) site(s). These results show that the wild type has three phosphorylation sites and the S263A and S264A mutants have two phosphorylation sites. In the MS/MS analysis of the S263A mutant using the peak at m/z = 1119.0, the sequence of the C-terminal peptide of S263A and the phosphorylations at Ser258 and Ser264 were confirmed (Fig. 4). Moreover, the ion at m/z = 1103.3 was assigned to a monophosphorylated C-terminal peptide; this peak contained two peptides phosphorylated at either Ser258 or Ser264, which was confirmed with MS/MS analysis (Fig. 5). These results indicate the heterogeneity of phosphorylation in the Cterminal region of the a7 subunit. We analyzed the C-terminal peptide of the S264A mutant using MS/MS analysis and confirmed that only Ser258 and Ser263 were phosphorylated. We have also performed the mutagenesis experiments at Ser258 and confirmed the phosphorylation of the site by the MS/ MS analysis to identify the phosphorylation both at Ser263 and Ser264 (data not shown).
Fig. 3. The MS analysis of the C-terminal peptides of the a7 subunit from the wild type and S263A and S264A mutants. The MS spectrum of the m/z range between 1090 and 1150 is shown. Peak identification is as follows. Peak m/z = 1138.2 [M + 5H] identifies three phosphate groups. Peaks m/z = 1122.2 [M + 5H] and 1119.0 [M + 5H] indicate two phosphate groups. Peaks m/z = 1106.2 [M + 5H] and 1103.0 [M + 5H] signify one phosphate group.
Discussion In this study, we identified three phosphorylation sites of the budding yeast a7 proteasome subunit in vivo using mass spectrometry. First, we used ESI/Q-TOF MS to estimate the number of phosphate groups attached to the C-terminal peptide isolated from an in-gel digest with lysylendopeptidase. Then, we determined the identity of the three phosphorylation sites in the C-terminal region (Ser258, Ser263, and Ser264 residues) with an MS/MS analysis and site-directed mutagenesis. No unphosphorylated C-terminal peptides were observed, suggesting that most of the C-terminal region of the a7 subunit is phosphorylated in yeast cells. In yeast and mammals, the a7 subunit is the main proteasome component phosphorylated in vivo, and some Ser residues of the C-terminal region can be phosphorylated by CKII in vitro [7,14]. The phosphorylation of the a7 subunit in budding yeast occurs in the presence of polylysine in vitro; however, the a7 subunit in mammals is phosphorylated by CKII under basal conditions without polylysine [7,14,20]. This manner of phosphorylation distinguishes the budding yeast proteasome from
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
Fig. 4. The MS/MS analysis of the phosphopeptides (m/z = 1119.0) from the S263A mutant. (A) Fragments of the y- and b-series identified in the MS/MS spectra from the phosphopeptide of Ser258 and Ser264 are shown. P indicates a phosphorylated amino acid. (B) MS/MS spectrum of the precursors at m/z = 1119.0 (shown in Fig. 3). (C) Expansion of the m/z range between 950 and 1900. The phosphorylation of Ser258 was confirmed by b-series fragment ions (from b9 to b16). Some a-series fragments were also observed. S*, Phosphoserine. (D) Expansion of the m/z range between 2750 and 3750. The phosphorylation of Ser264 was confirmed by y-series fragment ions (from y25 to y33). S*, Phosphoserine.
the mammalian proteasome. CKII is responsible for the constitutive phosphorylation of proteins and is a key player in signal transduction and amplification in eukaryotic cells. A study of the phosphorylation of calmodulin reported that CKII activity is regulated by polybasic peptides, suggesting that the polylysine-dependent phosphorylation requires specific regional and structural features [23,24]. The multiple alignment of the C-terminal region of the a7 subunits shows a low sequence similarity between budding yeast and other
13
species (Fig. 6). The Ser/Thr residues, which are potential phosphorylation sites, are not conserved in even yeast such as Schizosaccharomyces pombe and Candida albicans, indicating that the C-terminal region of the a7 subunit is highly divergent across species. However, the C-terminal region of the a7 subunit is the main target of phosphorylation by CKII in mammals and budding yeast, suggesting that phosphorylation of the a7 subunit is a widespread mechanism conserved from budding yeast to mammals. The C-terminal region in budding yeast has three putative phosphorylation sites for CKII: Ser263, Ser264, and Thr279. Our analysis revealed that three Ser residues, Ser258, Ser263, and Ser264, were phosphorylated and that Thr279 was not phosphorylated in vivo. Although there is low similarity phosphorylation motif in the sequence around Ser258, it is rich in acidic residues, and CKII often targets amino acid residues in an acidic region. Therefore, it is likely that all three Ser residues in the C-terminal region in the budding yeast a7 subunit are phosphorylated by CKII in vivo. Recently, Claverol et al. [8] observed phosphorylation only on Ser250, and not on Ser243, in the mammalian a7 subunit in vivo, even though both sites can be phosphorylated by CKII in vitro [7,8]. Both Ser243 and Ser250 were in the same recovered tryptic peptide (242ESLKEEDESDDDNM255), but these authors only recovered the peptide phosphorylated at Ser250 using IMAC. This suggests that either Ser243 is phosphorylated by CKII only in vitro or it is phosphorylated only after Ser250 has been phosphorylated. Some Ser/Thr kinases, such as CKII, are known to phosphorylate multiple substrate sites only after the prerequisite phosphorylation of one particular site [25]. In our study, however, in vivo phosphorylation of the a7 subunit C-terminal region was not successive; we observed the phosphorylation of two residues (Ser258 and Ser264 or Ser258 and Ser263) in the doubly phosphorylated peptides from S263A and S264A mutants, while the phosphorylation of Ser258 or Ser264 and Ser258 or Ser263 was observed in the singly phosphorylated peptides from the mutants. These results suggest that the phosphorylations at these sites do not occur successively. The C-terminal region of a7 subunit is located near the surface interacting with 19S regulatory particle [2]. Bose et al. [6] recently reported that the phosphorylation of a7 subunit is not essential for the interaction between 20S proteasome and 19S regulatory particle, but stabilizes it. The phosphorylation of a subunits of 20S proteasome also affects the proteolytic activities of 20S proteasome [10,20]. Our preliminary results suggest that the phosphorylation at the C-terminal region of a7 subunit does not affect the proteolytic activity of 20S proteasome. Other a subunits are also phosphorylated in yeast, in addition to the a7 subunit, but the b subunits are not phosphorylated [6–10,13,14,16–20]. It is possible
14
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
Fig. 5. The MS/MS analysis of the phosphopeptides (m/z = 1103.0) from the S263A mutant. (A) Fragments of the y-, y 0 -, b-, and b 0 -series identified in the MS/MS spectra from a monophosphopeptide on Ser258 are shown. P indicates a phosphorylated amino acid. (B) Fragments of y-, y00 -, b-, and b00 -series identified in the MS/MS spectra from the monophosphopeptide on Ser264 are shown. P indicates a phosphorylated amino acid. (C) MS/MS spectrum of the precursors at m/z = 1103.0 (shown in Fig. 3). (D) Expansion of the m/z range between 950 and 1900. The phosphorylation of Ser258 was confirmed using b-series fragment ions (from b9 to b16). Some a-series fragments were also observed. S*, Phosphoserine. (E) Expansion of the m/ z range between 2950 and 3750. The phosphorylation of Ser264 was confirmed using y-series fragment ions (from y27 to y33). S*, Phosphoserine.
Fig. 6. Comparison of the amino acid sequence of the C-terminal region of the a7 subunit in different species. The underlined Ser/Thr residues are potential sites of phosphorylation by CKII. Phosphorylation sites reported in previous studies or in this study are in bold [7,8]. GenBank Accession Nos. for the sequences: human BAA00659, mouse O70435, Xenopus laevis S38529, Spinacia oleracea O24362, Arabidopsis thaliana NP_180270, Oryza sativa BAB89648, Drosophila melanogaster NP_724834, Caenorhabditis elegans NP_496177, S. cerevisiae P21242, S. pombe NP_588040. The sequence of C. albicans is from http://genolist.pasteur.fr/CandidaDB/ (Accession No. CA4890). The consensus phosphorylation motif for CKII was predicted from [26].
that the phosphorylation of a subunits other than the a7 affects the proteolytic activity. The phosphorylation at Tyr residue of a7 subunit is also observed [10]. The phosphorylation of the C-terminal region of the a7 subunit might be responsible for regulating the association of other complexes, such as the 19S regulatory particle.
The physiological function of a subunit phosphorylation remains to be investigated. The roles of the three phosphorylation sites identified in this study are still not clear at present; however, the determination of the phosphorylation sites in vivo is an important step toward understanding their possible significance.
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
Acknowledgment We thank Dr. K. Tanaka (The Tokyo Metropolitan Institute of Medical Science) for providing the pYC1 vector. References [1] M.H. Glickman, A. Ciechanover, Physiol. Rev. 82 (2002) 373– 428. [2] M. Groll, L. Ditzel, J. Lowe, D. Stock, M. Bochtler, H.D. Bartunik, R. Huber, Nature 386 (1997) 463–471. [3] M.H. Glickman, D.M. Rubin, V.A. Fried, D. Finley, Mol. Cell. Biol. 18 (1998) 3149–3162. [4] W. Heinemeyer, M. Fischer, T. Krimmer, U. Stachon, D.H. Wolf, J. Biol. Chem. 272 (1997) 25200–25209. [5] O. Coux, K. Tanaka, A.L. Goldberg, Annu. Rev. Biochem. 65 (1996) 801–847. [6] S. Bose, F.L. Stratford, K.I. Broadfoot, G.G. Mason, A.J. Rivett, Biochem. J. 378 (2004) 177–184. [7] J.G. Castano, E. Mahillo, P. Arizti, J. Arribas, Biochemistry 35 (1996) 3782–3789. [8] S. Claverol, O. Burlet-Schiltz, E. Girbal-Neuhauser, J.E. Gairin, B. Monsarrat, Mol. Cell. Proteomics 1 (2002) 567–578. [9] P. Fernandez Murray, P.S. Pardo, A.M. Zelada, S. Passeron, Arch. Biochem. Biophys. 404 (2002) 116–125. [10] Y. Iwafune, H. Kawasaki, H. Hirano, Electrophoresis 23 (2002) 329–338. [11] Y. Kimura, M. Takaoka, S. Tanaka, H. Sassa, K. Tanaka, B. Polevoda, F. Sherman, H. Hirano, J. Biol. Chem. 275 (2000) 4635–4639.
15
[12] Y. Kimura, Y. Saeki, H. Yokosawa, B. Polevoda, F. Sherman, H. Hirano, Arch. Biochem. Biophys. 409 (2003) 341– 348. [13] G.G. Mason, R.Z. Murray, D. Pappin, A.J. Rivett, FEBS Lett. 430 (1998) 269–274. [14] P.S. Pardo, P.F. Murray, K. Walz, L. Franco, S. Passeron, Arch. Biochem. Biophys. 349 (1998) 397–401. [15] A.J. Rivett, S. Bose, P. Brooks, K.I. Broadfoot, Biochimie 83 (2001) 363–366. [16] S. Bose, P. Brooks, G.G. Mason, A.J. Rivett, Biochem. J. 353 (2001) 291–297. [17] M. Tokumoto, R. Horiguchi, Y. Nagahama, T. Tokumoto, Gene 239 (1999) 301–308. [18] M. Umeda, Y. Manabe, H. Uchimiya, FEBS Lett. 403 (1997) 313–317. [19] A. Yamagata, D.B. Kristensen, Y. Takeda, Y. Miyamoto, K. Okada, M. Inamatsu, K. Yoshizato, Proteomics 2 (2002) 1267– 1276. [20] G.G. Mason, K.B. Hendil, A.J. Rivett, Eur. J. Biochem. 238 (1996) 453–462. [21] T. Fujiwara, K. Tanaka, E. Orino, T. Yoshimura, A. Kumatori, T. Tamura, C.H. Chung, T. Nakai, K. Yamaguchi, S. Shin, et al., J. Biol. Chem. 265 (1990) 16604–16613. [22] A.N. Vallejo, R.J. Pogulis, L.R. Pease, Mutagenesis and Synthesis of Novel Recombinant Genes Using PCR, Cold Spring Harbor Laboratory Press, NY, 1995. [23] A.P. Bidwai, J.C. Reed, C.V. Glover, Arch. Biochem. Biophys. 300 (1993) 265–270. [24] F. Meggio, B. Boldyreff, O.G. Issinger, L.A. Pinna, Biochemistry 33 (1994) 4336–4342. [25] H. Flotow, P.J. Roach, J. Biol. Chem. 264 (1989) 9126– 9128. [26] L.A. Pinna, Biochim. Biophys. Acta 1054 (1990) 267–284.
Identification of three phosphorylation sites in the a7 subunit of the yeast 20S proteasome in vivo using mass spectrometry Yuko Iwafune, Hiroshi Kawasaki*, Hisashi Hirano Kihara Institute for Biological Research/Graduate School of Integrated Science, Yokohama City University, Maioka 641-12, Totsuka-ku, Yokohama 244-0813, Japan Received 26 March 2004, and in revised form 20 July 2004 Available online 20 August 2004
Abstract The 26S proteasome complex, which consists of a 20S proteasome and a pair of 19S regulatory particles, plays important roles in the degradation of ubiquitinated proteins in eukaryotic cells. The a7 subunit of the budding yeast 20S proteasome is a major phosphorylatable subunit; serine residue(s) in its C-terminal region are phosphorylated in vitro by CKII. However, the exact in vivo phosphorylation sites have not been identified. In this study, using electrospray ionization quadrupole time-of-flight mass spectrometry analysis, we detected a mixture of singly, doubly, and triply phosphorylated C-terminal peptides isolated from a His-tagged construct of the a7 subunit by nickel-immobilized metal affinity chromatography. In addition, we identified three phosphorylation sites in the C-terminal region using MS/MS analysis and site-directed mutagenesis: Ser258, Ser263, and Ser264 residues. The MS/MS analysis of singly phosphorylated peptides showed that phosphorylation at these sites did not occur successively. Ó 2004 Elsevier Inc. All rights reserved. Keywords: Proteasome; 20S proteasome; a7; Phosphorylation; Mass spectrometry; Yeast; Casein kinase II; Identification; Modification; Post-translational modification
Selective protein degradation in cells is important for regulating cellular functions such as cell cycle progression, signal transduction, and protein quality control. The ubiquitin/26S proteasome pathway is involved in this degradation [1]. The 26S proteasome complex consists of a 20S proteasome and a pair of 19S regulatory particles [2,3]. The 20S proteasome is a hollow cylinder composed of four stacked heptamer rings of a- and btype subunits [2]. The eukaryotic 20S proteasome has three protease activities: caspase-, trypsin-, and chymotrypsin-like activities [4,5]. The 19S regulatory particle consists of six ATPase subunits and at least 11 nonATPase subunits, and it associates with the 20S proteasome in an ATP-dependent manner to form the 26S proteasome [3]. *
Corresponding author. Fax: +81 45 820 1901. E-mail address: [email protected] (H. Kawasaki).
0003-9861/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2004.07.020
The 26S proteasome is post-translationally modified by phosphorylation, N-acetylation, and N-myristoylation in various species [6–20]. Of these post-translational modifications, the phosphorylation of serine, threonine, or tyrosine residues is a key event for signal transduction, cell cycle progression, and protease activity. Phosphorylation plays an important role in the ubiquitin/26S proteasome pathway. Phosphorylation often serves to target substrates to their cognate E3 enzymes [1]. In addition, some subunits of the 26S proteasome are phosphorylated, affecting proteasome functions such as the association between the 20S proteasome and 19S regulatory particle and the protease activity of the 20S proteasome. Phosphorylation regulates the formation and function of proteasome complexes. Some subunits of the 20S proteasome and 19S regulatory particle are phosphorylated in yeast and mammals [6–9,14,20]. The a7 subunit in the 20S proteasome is
10
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
the major phosphorylated subunit in both. The a7 subunit is a substrate of CKII in vivo and in vitro [6,7,14]. Protein kinase CKII (formerly known as casein kinase II) is a ubiquitous, highly conserved Ser/Thr kinase found in all eukaryotes. In budding yeast and mammals, the CKII phosphorylation sites have been mapped to Ser/Thr residues in the C-terminal region. The sequence of the C-terminal region of the a7 subunit is divergent in budding yeast and mammals; nevertheless, both have putative phosphorylation sites for CKII. Castano et al. [7] reported that two Ser residues of the mammalian a7 subunit, Ser243 and Ser250, possess phosphorylation motifs for CKII [7]. Mutagenesis experiments revealed that these residues were phosphorylation sites in vitro [7]. By contrast, in vivo phosphorylation of the human a7 subunit is reported to occur at Ser250 only [8], suggesting the presence of a mechanism regulating phosphorylation. The in vivo phosphorylation sites of the yeast a7 subunit have not yet been identified. It is important to identify the in vivo phosphorylation sites to understand the physiological roles of the phosphorylation of the a7 subunit. In this paper, we identified Ser258, Ser263, and Ser264 as in vivo phosphorylation sites of the yeast a7 subunit using mass spectrometry and revealed that phosphorylation at these three sites does not occur successively.
Material and methods Yeast strains and medium Yeast cells were grown at 30 °C in YPD medium containing 1% yeast extract, 2% Bacto–peptone, 2% glucose, and 0.5 lg ml1 aureobasidin A (Takara, Otsu, Japan). Yeast cells were harvested at late log phase (OD at 600 nm, 2.5–3.5) for proteasome purification. The a7 deletion mutant Saccharomyces cerevisiae BY4743-YOR362C (ATCC No. 4021659) was purchased from the American Type Culture Collection. Plasmid pYC1 carrying the a7 gene of yeast 20S proteasome was a kind gift from Dr. K. Tanaka (The Tokyo Metropolitan Institute of Medical Science, Tokyo) [21]. Construction of an expression vector for the His6-tagged a7 subunit of proteasome A His6-tag was added at the C-terminus of the a7 subunit. The a7 subunit gene with the His6-tag was amplified by PCR using pYC1 plasmid as the template and the following primers: 5 0 -CGCCCGGGTCTTCAGC AATGACATCAATTGG-3 0 and 5 0 -GCGAGCTCTCA GTGGTGGTGGTGGTGGTGTTCTAGGTGAATAT CACCCTCTTGG-3 0 . The PCR fragment was cloned
into the SacI and SmaI sites of pBluescript II SK () (Stratagene, CA, USA). After confirming the sequence, the plasmid was digested with SacI and SmaI, and the fragment was transferred to the shuttle vector pAUR123 (Takara, Otsu, Japan). Yeast cells (BY4743-YOR362C) were transformed using the aureobasidin A yeast transformation system (Takara, Otsu, Japan) according to the manufacturerÕs instructions.
Purification of the 20S proteasome including the His6tagged a7 subunit The 20S proteasome including the a7 subunit with the His6-tag was purified using nickel-immobilized metal affinity chromatography (IMAC) with TOYOPEARL AF-Chelate-650M (Tosoh, Tokyo, Japan) and anion exchange chromatography using Poros HQ/L (Applied Biosystems, Uppsala, Sweden). The purified 20S proteasome including the His6-tagged a7 subunit was separated by SDS–PAGE. The a7 subunit with the His6-tag was detected by Western blotting using anti-penta His monoclonal antibody (Qiagen, Hilden, Germany). The proteins were stained using a Negative Gel Stain MS kit (Wako Pure Chemical Industries, Osaka, Japan). The gel pieces containing protein bands were cut out and destained with the destaining solution provided in this kit. After they were washed with MilliQ water three times and with acetonitrile once, they were dried using a vacuum centrifuge. The gel pieces were rehydrated in 100 mM Tris–HCl, pH 9.0, containing 1 ng ll1 lysylendopeptidase (Wako Pure Chemical Industries, Osaka, Japan) for 16 h at 37 °C. The digests were concentrated to 20 ll using a vacuum centrifuge. Isolation of the C-terminal peptide and analysis using ESI/Q-TOF MS The C-terminal peptide tagged with the His6-tag was isolated from the digest using IMAC chromatography. The peptides were concentrated to 20 ll using a vacuum centrifuge and desalted using a ZipTipl-C18 column (Millipore, Bedford, MA, USA) according to the manufacturerÕs instructions. The peptides were loaded into a borosilicate nanoflow tip (Micromass, Manchester, UK) and subjected to electrospray ionization quadrupole time-of-flight mass spectrometry (ESI/Q-TOF MS)1 (Micromass, Manchester, UK) with positive ion detection mode. The MS and MS/MS spectra were ana-
1
Abbreviations used: CBB, Coomassie brilliant blue; ESI, electrospray ionization; MS, mass spectrometry; Q-TOF, quadrupole time-offlight mass spectrometer; SDS–PAGE, SDS–polyacrylamide gel electrophoresis; IMAC, immobilized metal affinity chromatography; OD, optical density
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
lyzed using MassLynx v3.4 software (Micromass, Manchester, UK). The MS/MS spectrum was processed with MaxEnt3 program in MassLynx package. The deconvoluted single-charge spectrum was used for amino acid sequencing. Statistical analysis All data are expressed as means ± SE and represent three experiments. The significance of differences was analyzed using the t test, and P < 0.05 was taken as statistically significant.
11
purified 20S proteasome by SDS–PAGE. The His6tagged a7 subunit was the largest subunit in the yeast 20S proteasome; it was detected with anti-penta-His monoclonal antibody and identified using peptide mass fingerprinting. All His6-tagged a7 subunits were associated with the 20S proteasome in yeast cells; no free form of the His6-tagged a7 subunit was observed. The purified 20S proteasome including the His6-tagged a7 subunit had protease activity toward the Suc-LLVY-MCA substrate equal to that of the native 20S proteasome. Mass spectrometric analysis of the C-terminal peptides of the His6-tagged a7 subunit
Site-directed mutagenesis The mutation of Ser263 or Ser264 to Ala263 or Ala264 was carried out by PCR using the previously constructed vector encoding the His6-tagged a7 subunit as template and the following primers: S263A_F (AACGT CATGG CTAGTGATGATGAA), S263A_R (TTCAT CATCAC TAGCCATGACGTT), S264A_F (GTCATGTCCGC TGATGATGAAAAT), and S264A_R (ATTT TCAT CAT CAGCGGACATGAC) [22].
Results Purification of the 20S proteasome The 20S proteasome including the His6-tagged a7 subunit was purified using IMAC and anion exchange chromatography. Fig. 1 shows the subunit separation of the
Fig. 1. Purification of the 20S proteasome containing the His6-tagged a7 subunit. Purified 20S proteasome including the His6-tagged a7 subunit was separated by 12% SDS–PAGE. 1, Subunits were stained with CBB. 2, The His6-tagged a7 subunit was detected using antipenta-His monoclonal antibody (Qiagen).
A lysylendopeptidase digest of the a7 subunit was subjected to IMAC to isolate the His6-tagged C-terminal peptides. The eluant from IMAC was analyzed using ESI/Q-TOF MS. The mass spectrum of the C-terminal peptide of the a7 subunit, 246EINGDDDEDEDDSDN VMSSDDENAPVATNANATTDQEGDIHLEHHHH HH294, is shown in Fig. 2. The three major peaks (m/ z = 1138.2 [M + 5H], 1422.5 [M + 4H], and 1896.7 [M + 3H]) correspond to the mass of the His6-tagged C-terminal peptide phosphorylated at three sites (theoretical mass = 5686 Da). The minor peaks (m/z = 1402.3 [M + 4H] and 1870.3 [M + 3H]; m/z = 1382.6 [M + 4H]) also correspond to the His6-tagged C-terminal peptide phosphorylated at two (theoretical mass = 5606 Da) or one (theoretical mass = 5526 Da) sites, respectively. The peaks m/z = 1426.8 [M + 4H] and 1902.0 [M + 3H] correspond to the triply phosphorylated peptide with a mass increase of 16 Da by oxidation of the Met residue. No unphosphorylated C-terminal peptide was observed. When the relative amount of the triply phosphorylated peptide was defined as 100, the amount of the doubly phosphorylated peptide was less than 40 ± 10 and the
Fig. 2. MS analysis of the C-terminal peptides. The MS spectrum of the m/z range between 1000 and 2000 is shown. Peak identification is as follows. Peaks m/z = 1138.2 [M + 5H], 1422.5 [M + 4H], and 1896.7 [M + 3H] represent three phosphate groups. Peaks m/z = 1402.3 [M + 4H] and 1870.3 [M + 3H] indicate two phosphate groups. Peak m/z = 1382.6 [M + 4H] identifies one phosphate group. Peaks m/ z = 1426.8 [M + 4H] and 1902.0 [M + 3H] signify three phosphate groups and an oxidized Met residue.
12
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
amount of the singly phosphorylated peptide was less than 10 ± 7. The formation of positive ions from the peptide phosphorylated at three sites should be suppressed more than in the peptide phosphorylated at two sites or one site due to the strongly negative charge of the phosphate group. However, the signal of the peptide phosphorylated at three sites was the strongest. These results suggest that the three sites in the C-terminal region are phosphorylated in nearly all a7 subunits in yeast cells in vivo. We observed the neutral loss of H3PO4 (decrease of 98 Da per phosphate group) in the MS/MS analysis of these peptides. However, the MS/MS spectra could not identify the phosphorylation sites because of the strongly negative charge of the phosphate group. We replaced each Ser residue at the putative phosphorylation sites in the C-terminal region of the a7 subunit with an Ala residue using site-directed mutagenesis to resolve the difficulty. Identification of in vivo phosphorylation sites using MS/ MS analysis Each C-terminal peptide of the a7 subunit from two mutants, S263A and S264A, was analyzed using ESI/QTOF MS. Fig. 3 shows the MS spectra of the C-terminal peptides from the wild type and the S263A and S264A mutants. Three major peaks from the wild type (m/ z = 1138.2 [M + 5H], 1122.2 [M + 5H], and 1106.2 [M + 5H]) were assigned to peptides phosphorylated at three (theoretical mass = 5686 Da), two (theoretical mass = 5606 Da), or one (theoretical mass = 5526 Da) site(s). The two major peaks observed in the spectra of S263A and S264A (m/z = 1119.0 [M + 5H] and 1103.0 [M + 5H]) were assigned to peptides phosphorylated at two (theoretical mass = 5590.4 Da) or one (theoretical mass = 5510.4 Da) site(s). These results show that the wild type has three phosphorylation sites and the S263A and S264A mutants have two phosphorylation sites. In the MS/MS analysis of the S263A mutant using the peak at m/z = 1119.0, the sequence of the C-terminal peptide of S263A and the phosphorylations at Ser258 and Ser264 were confirmed (Fig. 4). Moreover, the ion at m/z = 1103.3 was assigned to a monophosphorylated C-terminal peptide; this peak contained two peptides phosphorylated at either Ser258 or Ser264, which was confirmed with MS/MS analysis (Fig. 5). These results indicate the heterogeneity of phosphorylation in the Cterminal region of the a7 subunit. We analyzed the C-terminal peptide of the S264A mutant using MS/MS analysis and confirmed that only Ser258 and Ser263 were phosphorylated. We have also performed the mutagenesis experiments at Ser258 and confirmed the phosphorylation of the site by the MS/ MS analysis to identify the phosphorylation both at Ser263 and Ser264 (data not shown).
Fig. 3. The MS analysis of the C-terminal peptides of the a7 subunit from the wild type and S263A and S264A mutants. The MS spectrum of the m/z range between 1090 and 1150 is shown. Peak identification is as follows. Peak m/z = 1138.2 [M + 5H] identifies three phosphate groups. Peaks m/z = 1122.2 [M + 5H] and 1119.0 [M + 5H] indicate two phosphate groups. Peaks m/z = 1106.2 [M + 5H] and 1103.0 [M + 5H] signify one phosphate group.
Discussion In this study, we identified three phosphorylation sites of the budding yeast a7 proteasome subunit in vivo using mass spectrometry. First, we used ESI/Q-TOF MS to estimate the number of phosphate groups attached to the C-terminal peptide isolated from an in-gel digest with lysylendopeptidase. Then, we determined the identity of the three phosphorylation sites in the C-terminal region (Ser258, Ser263, and Ser264 residues) with an MS/MS analysis and site-directed mutagenesis. No unphosphorylated C-terminal peptides were observed, suggesting that most of the C-terminal region of the a7 subunit is phosphorylated in yeast cells. In yeast and mammals, the a7 subunit is the main proteasome component phosphorylated in vivo, and some Ser residues of the C-terminal region can be phosphorylated by CKII in vitro [7,14]. The phosphorylation of the a7 subunit in budding yeast occurs in the presence of polylysine in vitro; however, the a7 subunit in mammals is phosphorylated by CKII under basal conditions without polylysine [7,14,20]. This manner of phosphorylation distinguishes the budding yeast proteasome from
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
Fig. 4. The MS/MS analysis of the phosphopeptides (m/z = 1119.0) from the S263A mutant. (A) Fragments of the y- and b-series identified in the MS/MS spectra from the phosphopeptide of Ser258 and Ser264 are shown. P indicates a phosphorylated amino acid. (B) MS/MS spectrum of the precursors at m/z = 1119.0 (shown in Fig. 3). (C) Expansion of the m/z range between 950 and 1900. The phosphorylation of Ser258 was confirmed by b-series fragment ions (from b9 to b16). Some a-series fragments were also observed. S*, Phosphoserine. (D) Expansion of the m/z range between 2750 and 3750. The phosphorylation of Ser264 was confirmed by y-series fragment ions (from y25 to y33). S*, Phosphoserine.
the mammalian proteasome. CKII is responsible for the constitutive phosphorylation of proteins and is a key player in signal transduction and amplification in eukaryotic cells. A study of the phosphorylation of calmodulin reported that CKII activity is regulated by polybasic peptides, suggesting that the polylysine-dependent phosphorylation requires specific regional and structural features [23,24]. The multiple alignment of the C-terminal region of the a7 subunits shows a low sequence similarity between budding yeast and other
13
species (Fig. 6). The Ser/Thr residues, which are potential phosphorylation sites, are not conserved in even yeast such as Schizosaccharomyces pombe and Candida albicans, indicating that the C-terminal region of the a7 subunit is highly divergent across species. However, the C-terminal region of the a7 subunit is the main target of phosphorylation by CKII in mammals and budding yeast, suggesting that phosphorylation of the a7 subunit is a widespread mechanism conserved from budding yeast to mammals. The C-terminal region in budding yeast has three putative phosphorylation sites for CKII: Ser263, Ser264, and Thr279. Our analysis revealed that three Ser residues, Ser258, Ser263, and Ser264, were phosphorylated and that Thr279 was not phosphorylated in vivo. Although there is low similarity phosphorylation motif in the sequence around Ser258, it is rich in acidic residues, and CKII often targets amino acid residues in an acidic region. Therefore, it is likely that all three Ser residues in the C-terminal region in the budding yeast a7 subunit are phosphorylated by CKII in vivo. Recently, Claverol et al. [8] observed phosphorylation only on Ser250, and not on Ser243, in the mammalian a7 subunit in vivo, even though both sites can be phosphorylated by CKII in vitro [7,8]. Both Ser243 and Ser250 were in the same recovered tryptic peptide (242ESLKEEDESDDDNM255), but these authors only recovered the peptide phosphorylated at Ser250 using IMAC. This suggests that either Ser243 is phosphorylated by CKII only in vitro or it is phosphorylated only after Ser250 has been phosphorylated. Some Ser/Thr kinases, such as CKII, are known to phosphorylate multiple substrate sites only after the prerequisite phosphorylation of one particular site [25]. In our study, however, in vivo phosphorylation of the a7 subunit C-terminal region was not successive; we observed the phosphorylation of two residues (Ser258 and Ser264 or Ser258 and Ser263) in the doubly phosphorylated peptides from S263A and S264A mutants, while the phosphorylation of Ser258 or Ser264 and Ser258 or Ser263 was observed in the singly phosphorylated peptides from the mutants. These results suggest that the phosphorylations at these sites do not occur successively. The C-terminal region of a7 subunit is located near the surface interacting with 19S regulatory particle [2]. Bose et al. [6] recently reported that the phosphorylation of a7 subunit is not essential for the interaction between 20S proteasome and 19S regulatory particle, but stabilizes it. The phosphorylation of a subunits of 20S proteasome also affects the proteolytic activities of 20S proteasome [10,20]. Our preliminary results suggest that the phosphorylation at the C-terminal region of a7 subunit does not affect the proteolytic activity of 20S proteasome. Other a subunits are also phosphorylated in yeast, in addition to the a7 subunit, but the b subunits are not phosphorylated [6–10,13,14,16–20]. It is possible
14
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
Fig. 5. The MS/MS analysis of the phosphopeptides (m/z = 1103.0) from the S263A mutant. (A) Fragments of the y-, y 0 -, b-, and b 0 -series identified in the MS/MS spectra from a monophosphopeptide on Ser258 are shown. P indicates a phosphorylated amino acid. (B) Fragments of y-, y00 -, b-, and b00 -series identified in the MS/MS spectra from the monophosphopeptide on Ser264 are shown. P indicates a phosphorylated amino acid. (C) MS/MS spectrum of the precursors at m/z = 1103.0 (shown in Fig. 3). (D) Expansion of the m/z range between 950 and 1900. The phosphorylation of Ser258 was confirmed using b-series fragment ions (from b9 to b16). Some a-series fragments were also observed. S*, Phosphoserine. (E) Expansion of the m/ z range between 2950 and 3750. The phosphorylation of Ser264 was confirmed using y-series fragment ions (from y27 to y33). S*, Phosphoserine.
Fig. 6. Comparison of the amino acid sequence of the C-terminal region of the a7 subunit in different species. The underlined Ser/Thr residues are potential sites of phosphorylation by CKII. Phosphorylation sites reported in previous studies or in this study are in bold [7,8]. GenBank Accession Nos. for the sequences: human BAA00659, mouse O70435, Xenopus laevis S38529, Spinacia oleracea O24362, Arabidopsis thaliana NP_180270, Oryza sativa BAB89648, Drosophila melanogaster NP_724834, Caenorhabditis elegans NP_496177, S. cerevisiae P21242, S. pombe NP_588040. The sequence of C. albicans is from http://genolist.pasteur.fr/CandidaDB/ (Accession No. CA4890). The consensus phosphorylation motif for CKII was predicted from [26].
that the phosphorylation of a subunits other than the a7 affects the proteolytic activity. The phosphorylation at Tyr residue of a7 subunit is also observed [10]. The phosphorylation of the C-terminal region of the a7 subunit might be responsible for regulating the association of other complexes, such as the 19S regulatory particle.
The physiological function of a subunit phosphorylation remains to be investigated. The roles of the three phosphorylation sites identified in this study are still not clear at present; however, the determination of the phosphorylation sites in vivo is an important step toward understanding their possible significance.
Y. Iwafune et al. / Archives of Biochemistry and Biophysics 431 (2004) 9–15
Acknowledgment We thank Dr. K. Tanaka (The Tokyo Metropolitan Institute of Medical Science) for providing the pYC1 vector. References [1] M.H. Glickman, A. Ciechanover, Physiol. Rev. 82 (2002) 373– 428. [2] M. Groll, L. Ditzel, J. Lowe, D. Stock, M. Bochtler, H.D. Bartunik, R. Huber, Nature 386 (1997) 463–471. [3] M.H. Glickman, D.M. Rubin, V.A. Fried, D. Finley, Mol. Cell. Biol. 18 (1998) 3149–3162. [4] W. Heinemeyer, M. Fischer, T. Krimmer, U. Stachon, D.H. Wolf, J. Biol. Chem. 272 (1997) 25200–25209. [5] O. Coux, K. Tanaka, A.L. Goldberg, Annu. Rev. Biochem. 65 (1996) 801–847. [6] S. Bose, F.L. Stratford, K.I. Broadfoot, G.G. Mason, A.J. Rivett, Biochem. J. 378 (2004) 177–184. [7] J.G. Castano, E. Mahillo, P. Arizti, J. Arribas, Biochemistry 35 (1996) 3782–3789. [8] S. Claverol, O. Burlet-Schiltz, E. Girbal-Neuhauser, J.E. Gairin, B. Monsarrat, Mol. Cell. Proteomics 1 (2002) 567–578. [9] P. Fernandez Murray, P.S. Pardo, A.M. Zelada, S. Passeron, Arch. Biochem. Biophys. 404 (2002) 116–125. [10] Y. Iwafune, H. Kawasaki, H. Hirano, Electrophoresis 23 (2002) 329–338. [11] Y. Kimura, M. Takaoka, S. Tanaka, H. Sassa, K. Tanaka, B. Polevoda, F. Sherman, H. Hirano, J. Biol. Chem. 275 (2000) 4635–4639.
15
[12] Y. Kimura, Y. Saeki, H. Yokosawa, B. Polevoda, F. Sherman, H. Hirano, Arch. Biochem. Biophys. 409 (2003) 341– 348. [13] G.G. Mason, R.Z. Murray, D. Pappin, A.J. Rivett, FEBS Lett. 430 (1998) 269–274. [14] P.S. Pardo, P.F. Murray, K. Walz, L. Franco, S. Passeron, Arch. Biochem. Biophys. 349 (1998) 397–401. [15] A.J. Rivett, S. Bose, P. Brooks, K.I. Broadfoot, Biochimie 83 (2001) 363–366. [16] S. Bose, P. Brooks, G.G. Mason, A.J. Rivett, Biochem. J. 353 (2001) 291–297. [17] M. Tokumoto, R. Horiguchi, Y. Nagahama, T. Tokumoto, Gene 239 (1999) 301–308. [18] M. Umeda, Y. Manabe, H. Uchimiya, FEBS Lett. 403 (1997) 313–317. [19] A. Yamagata, D.B. Kristensen, Y. Takeda, Y. Miyamoto, K. Okada, M. Inamatsu, K. Yoshizato, Proteomics 2 (2002) 1267– 1276. [20] G.G. Mason, K.B. Hendil, A.J. Rivett, Eur. J. Biochem. 238 (1996) 453–462. [21] T. Fujiwara, K. Tanaka, E. Orino, T. Yoshimura, A. Kumatori, T. Tamura, C.H. Chung, T. Nakai, K. Yamaguchi, S. Shin, et al., J. Biol. Chem. 265 (1990) 16604–16613. [22] A.N. Vallejo, R.J. Pogulis, L.R. Pease, Mutagenesis and Synthesis of Novel Recombinant Genes Using PCR, Cold Spring Harbor Laboratory Press, NY, 1995. [23] A.P. Bidwai, J.C. Reed, C.V. Glover, Arch. Biochem. Biophys. 300 (1993) 265–270. [24] F. Meggio, B. Boldyreff, O.G. Issinger, L.A. Pinna, Biochemistry 33 (1994) 4336–4342. [25] H. Flotow, P.J. Roach, J. Biol. Chem. 264 (1989) 9126– 9128. [26] L.A. Pinna, Biochim. Biophys. Acta 1054 (1990) 267–284.