Archaeal Protein Biogenesis: Posttranslational Modification and DegradationView this Special Issue
Research Article | Open Access
Matthew A. Humbard, Christopher J. Reuter, Kheir Zuobi-Hasona, Guangyin Zhou, Julie A. Maupin-Furlow, "Phosphorylation and Methylation of Proteasomal Proteins of the Haloarcheon Haloferax volcanii", Archaea, vol. 2010, Article ID 481725, 10 pages, 2010. https://doi.org/10.1155/2010/481725
Phosphorylation and Methylation of Proteasomal Proteins of the Haloarcheon Haloferax volcanii
Proteasomes are composed of 20S core particles (CPs) of - and -type subunits that associate with regulatory particle AAA ATPases such as the proteasome-activating nucleotidase (PAN) complexes of archaea. In this study, the roles and additional sites of post-translational modification of proteasomes were investigated using the archaeon Haloferax volcanii as a model. Indicative of phosphorylation, phosphatase-sensitive isoforms of and were detected by 2-DE immunoblot. To map these and other potential sites of post-translational modification, proteasomes were purified and analyzed by tandem mass spectrometry (MS/MS). Using this approach, several phosphosites were mapped including Thr147, 2 Thr13/Ser14 and PAN-A Ser340. Multiple methylation sites were also mapped to , thus, revealing a new type of proteasomal modification. Probing the biological role of and PAN-A phosphorylation by site-directed mutagenesis revealed dominant negative phenotypes for cell viability and/or pigmentation for variants including Thr147Ala, Thr158Ala and Ser58Ala. An H. volcanii Rio1p Ser/Thr kinase homolog was purified and shown to catalyze autophosphorylation and phosphotransfer to . The variants in Thr and Ser residues that displayed dominant negative phenotypes were significantly reduced in their ability to accept phosphoryl groups from Rio1p, thus, providing an important link between cell physiology and proteasomal phosphorylation.
Proteasomes are multicatalytic proteases found in all three domains of life and are essential for growth in many organisms, including haloarchaea and eukaryotes [1–5]. Accessory proteins, such as 19S cap and proteasome activating nucleotidase (PAN) complexes, and protein-modification pathways such as ubiquitination are important in regulating protein degradation by proteasomes. Proteasomal proteins are also subject to co- and posttranslational modification (PTM) including N-terminal acetylation [6–9], phosphorylation [10–27], S-glutathionation , N-myristoylation , O-linked glycosylation , and processing of N-terminal propeptides including exposure of the Thr active site residues of 20S core particles (CPs) .
Despite decades of study, the mechanisms and biological significance of many of the posttranslational modifications of proteasomes are poorly understood. In particular, our knowledge of the phosphorylation of proteasomes remains limited. Polo-kinase, casein kinase II and calcium/calmodulin-dependent protein kinase II are known to mediate the phosphorylation of 26S proteasomes in vitro [11, 13, 14, 16–19, 27, 32, 33]. There is also evidence that proteasomal phosphorylation is regulated by growth and/or signaling molecules such as -interferon [11, 12, 17, 22, 24, 26].
Our previous work demonstrated that archaeal proteasomes undergo a variety of PTMs and alterations in subunit composition. In H. volcanii, these PTMs include N-terminal acetylation of the CP 1 and 2, removal of the CP -propeptide and phosphorylation of the CP Ser129 [34, 35]. The subunit composition of H. volcanii proteasomal complexes also appears to change as cells enter stationary phase. H. volcanii synthesizes two different proteasome-activating nucleotidase proteins (PAN-A and PAN-B) and three different CP subunits (1, 2, and ) [36, 37]. Proteasomal CPs and PAN complexes of different subunit composition have been isolated from H. volcanii (, this study and unpublished results). While the transcript levels encoding these proteasomal proteins all increase as cells enter stationary phase, only the protein levels of 2 and PAN-B increase several-fold during this transition . Thus, the population of different proteasomal subtypes appears to be modulated by growth.
In this report, we elaborate on the PTM of H. volcanii proteasomal proteins including the mapping of additional sites of phosphorylation as well as a new type of modification for proteasomes, methyl-esterification. We also report a correlation between the phosphorylation state of the CP 1 protein and growth phase, provide evidence that Rio1p Ser/Thr kinase phosphorylates 1 and suggest through site-directed mutagenesis that the state of proteasomal phosphorylation influences cell physiology including viability and pigmentation.
2. Materials and Methods
Biochemicals were purchased from Sigma-Aldrich (St. Louis, MO). Other organic and inorganic analytical-grade chemicals were from Fisher Scientific (Atlanta, GA) and Bio-Rad (Hercules, CA). Radioactive chemicals were purchased from Perkin Elmer (Waltham, MA), and Strep chromatography resin was purchased from Qiagen (Valencia, CA). Desalted oligonucleotide primers were from Integrated DNA Technologies (Coralville, IA). Site-directed mutagenesis materials were from Invitrogen (Carlsbad, CA). Polyvinyl difluoride (PVDF) membranes for Western blots were from Amersham Biosciences (Piscataway, NJ). Alpha-casein was from ICN Biochemicals (Cleveland, OH).
2.2. Strains, Media, and Plasmids
Strains and plasmids used in this study are summarized in Supplementary Table available online at http://dx.doi.org/10.1155/2010/481725. H. volcanii strains H26 and GZ130 (psmA) and E. coli strain BL21(DE3) were used as hosts for protein production and purification. E. coli strains were grown in Luria-Bertani medium at 3C and 200 rpm, and H. volcanii strains were grown in ATCC 974 medium at 4C and 200 rpm. Media were supplemented with 100 g/mL of ampicillin, 50 g/mL of kanamycin, or 0.1 g/mL of novobiocin as needed. For salt stress tests, the concentration of NaCl was reduced from the typical 2.25 M NaCl of the “normal” ATCC 974 medium to 1, 1.125, 1.25, 1.375, and 1.5 M NaCl. For the salt stress test inoculum, cells were grown in “normal salt” ATCC 974 medium by streaking cells from 8C glycerol stocks onto fresh plates and growing these cells twice to log-phase in 2 mL medium. Each subculture was inoculated to a final OD600 of 0.01 to 0.02. Experiments were performed in biological triplicate and the mean S.D. was calculated.
2.3. Cloning and Site-Directed Mutagenesis
Site-directed mutagenesis (SDM) was performed using the Quikchange SDM kit as per manufacturer’s instructions (Stratagene, La Jolla, CA) with the primers used for mutagenesis listed in Supplementary Table . The fidelity of all PCR-amplified products was confirmed by DNA sequencing (Genomics Core, Interdisiplinary Center for Biotechnology Research, University of Florida, Gainesville, FL) using the dideoxy termination method with Applied Biosystems (Foster City, CA) Model 3130 Genetic analyzer.
2.4. Protein Purification
Proteasomal proteins 1, 2, and with C-terminal hexahistidine-(His6) tags were purified from recombinant E. coli BL21 (DE3) strains after induction isopropyl β-D-1-thiogalactopyranoside (IPTG) at 0.4 mM for 2 hours as previously described . In addition, proteasomal proteins (1, 2, , PAN-A, and PAN-B) with C-terminal His6-tags, a Rio1p kinase homolog (HVO_0135) with a C-terminal -WSHPQFEK (-StrepII) tag and CPs composed of 1-His6 and -StrepII were purified from recombinant H. volcanii. For protein purification, the various H. volcanii strains expressing these derivatives were grown to stationary phase (OD600 of 2.0–2.5), unless otherwise indicated. Phosphatase inhibitor cocktail I and II were included in all lysis buffers and diluted according to supplier (Sigma). For His6-tagged proteins, cells were lysed in 20 mM Tris pH 7.2 with 2 M NaCl (buffer A) supplemented with 5 mM imidazole. Cell lysate was loaded onto a nickel nitrilotriacetic acid (Ni-NTA) column (1.6 2.5 cm, Pharmacia) with a step gradient at 60 mM imidazole, and proteins were eluted from the column in buffer A with 500 mM imidazole. For StrepII-tagged proteins, cells were lysed and applied to a StrepTactin column (0.5 5 cm, Qiagen) in buffer A. Proteins were eluted from the column in buffer A with 2.5 mM desthiobiotin. For cells expressing both StrepII- and His6-tagged proteins, “two-step” purification was performed which incorporated sequential Ni-NTA and StrepTactin chromatography . Native molecular mass of Rio1p was determined by applying protein fractions eluted from a StrepTactin column to a calibrated Superose 200 HR 10/30 column as recommended by supplier (Pharmacia-GE Healthcare, Piscataway, NJ). Molecular mass standards for calibration included cytochrome c (12.4 kDa), carbonic anhydrase (29 kDa), bovine serum albumin (66 kDa), alcohol dehydrogenase (150 kDa), -amylase (200 kDa), and apoferritin (443 kDa) (Sigma). For molecular mass standards, the gel filtration column was equilibrated in 20 mM Tris buffer at pH 7.2 supplemented with 150 mM NaCl. Salt concentration was increased to 2 M NaCl for Rio1 kinase.
2.5. Two-Dimensional Gel Electrophoresis and Western Blot Analysis
TRIzol-extracted and benzonase-treated protein (up to 250 g) was applied to an 11 cm immobilized pH gradient (IPG) strip with a pH range of 3.9–5.1 (Bio-Rad). Proteins were separated according to their pI using manufacturer’s protocol and previously published methods . Proteins were further separated using either 10% or 12% SDS-PAGE Criterion gels (Bio-Rad) at 1C at 200 V for 50 or 70 minutes, respectively. Proteins were transferred from gels to PVDF membranes at 100 V and C for 90–100 minutes in 10 mM 2-(N-morpholino) ethanesulfonic acid (MES) buffer at pH 6.0 supplemented with 10% (vol/vol) methanol. Polyclonal antibodies raised in rabbit against PAN-A and 1 and phosphoamino monoclonal antibodies (clone 22a, Becton Dickinson Biosciences, San Jose, CA) were used for Western blot analysis as previously described .
2.6. Phosphatase Treatment
Protein (up to 80 g) was incubated (3C, 20 minutes) in the presence and absence of lambda phosphatase (400 U) in 100 L 50 mM Tris buffer at pH 7.5 supplemented with 0.1 mM EDTA, 5 mM dithiothreitol and 2 mM MnCl2 (lambda phosphatase buffer).
2.7. Peptidase Assay
Chymotrypsin-like peptide hydrolyzing activity was assayed by release of 7-amino-4-methylcoumarin from N-Suc-Leu-Leu-Val-Tyr-7-amido-4-methylcoumarin (Suc-LLVY-AMC; Sigma) at 6C for 10–30 minutes by an increase in fluorescence (excitation = 348 nm, emission = 440 nm) as previously described . The assay mixture (0.3 mL) contained 1–3 g protein and 20 M fluorogenic substrate in buffer A with 0.4% (v/v) dimethyl sulfoxide (DMSO).
2.8. Kinase Assay
A Rio1p kinase homolog (HVO_0135) was expressed in H. volcanii with a C-terminal StrepII tag (-WSHPQFEK) that included a GT linker (encoded by a KpnI site) and purified to electrophoretic homogeneity by StrepTactin chromatography as described above. The purified kinase was dialyzed into 20 mM Tris buffer (pH 7.2) with 2 M NaCl or 2 M KCl at C using a D-tube dialysis tube (EMD Chemicals, Gibbstown, NJ). Autophosphorylation was monitored by incubating purified Rio1p (1 g) with MnCl2 or MgCl2 (0–250 mM) for one hour at 3C in the presence of 200–400 Ci of  adenosine -triphosphate (ATP) (Perkin Elmer) in a reaction volume of 20 L. For in vitro phosphorylation, 10 g of 1 variants were included in the assay mixture. Kinase reactions were inactived by addition of an equal volume of reducing SDS-PAGE loading buffer and separated by 12.5% SDS-PAGE gels. Gels were stained with Coomassie brilliant blue and dried to Whatman paper. Blue sensitive X-ray film (Research Products International Corp., Mt. Prospect, IL) was exposed to the dried gel for 1–10 days at 8C.
2.9. Mass Spectrometry
Proteasomal fractions (3 to 6 g protein) from nickel affinity chromatography and/or two-step affinity chromatography were separated by reducing 12% SDS-PAGE. Gels were stained with Bio-Safe Coomassie (Bio-Rad). Protein bands corresponding to the molecular masses of the CP subunits were excised from the gel and destained with 100 mM NH4HCO3 in 50% (vol/vol) acetonitrile (C, overnight). Protein samples were reduced, alkylated in-gel, and digested with trypsin (Promega, Madison, WI). Capillary reverse-phase high-performance liquid chromatography (HPLC) separation of the protein digests was performed using a PepMap C18 column (75-m inside diameter, 15-cm length; LC Packings, San Francisco, CA) with an Ultimate capillary HPLC system (LC Packings). A gradient (90 or 120 minutes) from 5 to 50% acetonitrile in 0.1% acetic acid was used at a flow rate of 200 nL per min. Tandem mass spectrometry (MS/MS) analysis was performed using a hybrid quadrupole time-of-flight (QTOF) instrument (QSTAR) equipped with a nanoelectrospray source and operated with Analyst QS 1.1 data acquisition software (Applied Biosystems).
2.10. Database Searching
Tandem mass spectrometry (MS/MS) data generated by information-dependent acquisition with QSTAR were searched against the deduced proteome of the H. volcanii DS2 genome sequence  by using the Mascot (Matrix Science, Boston, MA) database search engine. Probability-based Mascot scores were determined by a comparison of search results against estimated random match population and are reported as 10 , where is the absolute probability. Individual Mascot ion scores greater than 32 were considered to indicate identity or extensive homology (). Scores below this default significant value were considered for protein identification in addition to validation by manual interpretation of MS/MS spectra. Carbamidomethylation of cysteine was used as a fixed modification based on sample preparation. Variable modifications that were implemented in the database search included deamidation of asparagine and glutamine, N-terminal acetylation, oxidation (single and double) of methionine, and phosphorylation of serine, threonine, and tyrosine.
3.1. Isoforms of 1 Modulated during Growth-Phase Transitions
Posttranslational modification of 1, one of the predominant proteasomal proteins in H. volcanii cells, was examined during various stages of growth from log- to late-stationary phase by 2-DE immunoblot of cell lysate. Two 1-specific isoforms of pI 3.0 and 4.4 were detected, both of which migrated at 37.5 kDa consistent with the mobility of the α1 protein in reducing SDS-PAGE gels . The 1-isoform of pI 4.4 was present at all stages of growth while the second more acidic 1-isoform of pI 3.0 was present in earlier phases of growth and disappeared in late-stationary phase (Figure 1(a)).
Affinity purification of CPs from early stationary-phase cells expressing a hexahistidine tagged (-His6) derivative of the catalytic β subunit revealed that both isoforms of 1 (pI 3.0 and 4.4) associate as CPs active in hydrolysis of the peptide reporter Suc-LLVY-AMC (Figure 1(b)). The 1 isoform of pI 4.4 was also detected in non-CP fractions. Treatment of these purified 20S CPs with phosphatase resulted in a basic shift in pI of both 1 and 2 proteins signifying both -type subunits were phosphorylated (Figure 1(c)).
Together these results suggest that the -type proteins of archaeal proteasomal CPs can be modified by phosphorylation. Our previous studies demonstrate that the subunits of H. volcanii CPs are also phosphorylated at Ser129 . Thus, phosphorylation appears to be a major modification of the proteasome system of this haloarchaeon. Based on the isoelectric focusing (IEF) migration and phosphatase sensitivity of both 1 isoforms of pI 3.0 and 4.4, we propose that 1 harbors at least two phosphosites. The least phosphorylated form of 1 (pI 4.4) was detected at all stages of growth and occurred in both free and CP-associated forms. In contrast, the more extensively phosphorylated form of 1 (pI 3.0) was only detected in association with CPs and was not detected in late-stationary phase. In addition to these modified forms of 1, CP subtypes appear to incorporate a phosphorylated form of 2. Thus, H. volcanii not only modulates the types of subunits present in the CP and PAN subtypes during the transition from log- to stationary-phase growth , but appears also to posttranslationally modify the CP proteasomal proteins during this transition.
3.2. 1, 2, and PAN-A Are Phosphorylated at Ser/Thr Residues
To further map the phosphosites of CPs and PAN complexes, the various subunits (1, 2, , PAN-A, and PAN-B) were enriched from H. volcanii by nickel chromatography using His6 epitope-tags and analyzed by MS after tryptic digest. Initially, CPs were purified from log-phase H. volcanii H26-pJAM204 cells expressing α1-His6 and separated by SDS-PAGE prior to RP-HPLC MS/MS-analysis. Using this approach, a high-probability phosphopeptide was reproducibly detected in the α1-specific protein band that mapped to amino acid residues 137 to 149 of 1 (ALLIGGVENGpTPR, where pT represents phospho-Thr147) (Supplementary Figure ). MS/MS-fragmentation of this peptide included a characteristic neutral loss (98) indicative of CID removal of the phosphoryl group and confirmed that Thr147 was a site of 1 phosphorylation. The Thr147 residue was conserved in eukaryotic α-type CP proteins and was predicted to be phosphorylated by NetPhos (with a score of 0.851). An 2-specific phosphopeptide was also detected by precursor ion scanning of 2-specific tryptic peptides purified from H. volcanii H26-pJAM205 cells expressing 2-His6. The 2-specific phosphopeptide was composed of amino acid residues 12 to 21 (GTSLFSPDGR) and again displayed a neutral loss characteristic of phosphorylated peptides (Supplementary Figure ). MS/MS, however, could not distinguish between Thr13 and Ser14 as the phosphosite due to the close vicinity of these residues within the peptide. Of these two residues, Ser14 has a higher probability of serving as a phosphosite compared to Thr13 based on modeling to known eukaryotic phosphoproteins with NetPhos scores of 0.957 and 0.243, respectively. In addition to the 1- and 2-specific phosphopeptides, a PAN-A-specific phosphopeptide was reproducibly detected by ESI-QTOF analysis of tryptic digestions of PAN complexes purified from H. volcanii expressing either PAN-A-His6 or PAN-B-His6 (DS70-pJAM650 or pJAM1012, resp.) (Supplementary Figure ). This phosphopeptide mapped to amino acid residues 337 to 361 of PAN-A (MNVpSDDVDFVELAEMADNASGADIK, where pS represents phospho-Ser340). MS/MS fragmentation confirmed that Ser340 was the phosphorylated amino acid including a neutral loss peak (98 for phosphorylation) and several internal fragmentation ions. It is important to note that the PAN-A-Ser340-specific phosphopeptide was detected in strains expressing either PAN-A-His6 or PAN-B-His6. This indicates that phosphorylated PAN-A associates with PAN-B in H. volcanii cells. Interestingly, the PAN-A Ser340 and amino acids surrounding this residue are highly conserved among eukaryotic 26S proteasome Rpt-like proteins including those of humans; however, the prediction of this as a phoshosite by NetPhos was low with a score of 0.464. Similar to the Ser129 phosphosite that was mapped to the CP subunit , analogous nonphosphorylated peptides also mapped to 1 Thr147, 2 Thr13/Ser14, and PAN-A Ser340 residues. These results suggest that the phosphosites of PAN and CP complexes are not fully occupied by phosphoryl groups and, thus, the phosphorylation of proteasomes may be regulated in the cell.
3.3. Variant 1 Proteins Affect Pigmentation and Cell Viability
Single and double knock-out of psmA (1) and panA (PAN-A) genes diminishes the ability of H. volcanii cells to adapt to low salt and other stressful conditions . These phenotypes are readily complemented by providing a copy of the corresponding wild type gene in trans and, thus, provide an ideal assay for examining how phosphorylation may influence in the function of α1 and PAN-A in the cell.
Initially, panA mutant cells were tested for their ability to be complemented for growth on low-salt media with panB, panA, or pan (PAN-A Ser340Ala) genes. For complementation, the genes were expressed in trans from a multicopy pHV2-based plasmid using a strong rRNA P2 promoter. While overexpression of panB from the plasmid did not rescue the sensitivity of the panA mutant to low salt conditions, expression of either panA or panA Ser340Ala restored growth to wild-type levels. This indicates that phosphorylation of Ser340 is not important for response to hypo-osmotic stress (data not shown) and suggests that panB cannot substitute for panA under these growth conditions.
Site-directed mutations in the psmA gene encoding single amino acid substitutions of Ser58Ala, Thr147Ala, Thr158Ala, and Tyr28Phe in 1 were tested (along with overexpression of psmC encoding 2) for their ability to complement the hypo-osmotic stress response phenotype of a psmA mutant. Similar to panA, the modified psmA and psmC genes were provided in trans using a pHV2-based plasmid and strong rRNA P2 promoter. The 1-residues selected for site-directed modification were based on the following: 1 Thr147 was detected in this study as a phosphosite by MS/MS (see above for details), 1 Tyr28 was homologous to the Tyr27 phosphosite identified in the 2-type subunit of eukaryotic CPs , and Ser58 and Thr158 are predicted to be phosphorylated based on on NetPhos 2.0 scores of 0.996 and 0.981, respectively, (significantly above the significance threshold of 0.500) . Overexpression of the psmC (2) gene did not complement the psmA mutant while the psmA and psmA-a83t (1 Tyr28Phe) genes complemented the reduced growth phenotype of the psmA mutant on low salt media to wild type levels. In contrast, the other substitutions to the psmA gene provided in trans severely inhibited overall growth of the psmA mutant strain. While each of the constructs was successfully transformed into the psmA knockout strain, growth of the transformed cells was drastically reduced after being transferred to liquid media or streaked onto a fresh plate (Figure 2(a)). Some of the substitutions in 1 including Ser58Ala and 1 Thr147Ala appeared to affect carotenoid biosynthesis as cells expressing these 1 variants were whiter than cells expressing wild-type 1 or no 1 at all (Figure 2(a)). Whole cell chymotrypsin-like (CL) peptidase activity of the different 1 substitution mutants expressed in trans was compared to wild type after initial transfer to liquid media. Only modest differences in specific activity were detected between the strains with the most pronounced difference in cells expressing 1 Thr158Ala, which had a 2-fold increase in CL activity compared to wild-type (data not shown). It is unclear at this time if these modest differences in CL activity can account for the phenotypes observed. Anti-1 immunoblot of cell lysate revealed all of the 1 variant proteins were produced at levels similar to wild type when the psmA genes were expressed in trans in the psmA knockout (Figure 2(b)). Both the hypo-osmotic stress and pigmentation phenotypes were recessive based on the observation that these 1 variants did not alter growth or pigmentation when expressed in trans in wild-type H. volcanii cells (data not shown).
3.4. Haloferax volcanii Encodes a Number of Putative Protein Kinases
Based on genome sequence, H. volcanii encodes for a number of putative serine/threonine kinases including two different Rio-type protein kinases, type-1 (HVO_0135) and type-2-like (HVO_0569) (Figure 3(a)) as well as homologs of the serine protein kinase PrkA, originally described in Bacillus subtilis . Rio-type kinases are a relatively new atypical protein kinase family with four unique types, Rio1, Rio2, Rio3, and RioB. Rio-type kinases are widespread throughout eukaryotes and archaea. Archaea typically have two different Rio-type kinases, Rio1 and Rio2 . Rio1 kinases are characterized by a consensus sequence STGKEA in the N-terminal domain of the protein and a second region of homology in the C-terminal domain, IDxxQ (where x represents any amino acid residue). While the Rio2 kinases can vary somewhat in domain structure, these kinases are characterized by a helix-turn-helix motif in the N-terminus followed by the amino acid sequence GxGKES and a C-terminal IDFPQ sequence . The Rio1p from H. volcanii contains the signature domain sequence but the Rio2p does not (Figure 3(a)).
3.5. H. volcanii Rio Type-I Kinase Catalyzes Autophosphorylation
To investigate the activity of the Rio1p of H. volcanii, the C-terminal residue of this kinase was fused to a StrepII tag and synthesized in recombinant H. volcanii (H26-pJAM2558). Rio1p was purified to electrophoretic homogeneity by StrepTactin chromatography (Figure 3(b)) and determined to be a 50-kDa monomer based on gel filtration chromatography compared to migration by reducing SDS-PAGE and deduced polypeptide sequence. Rio1p catalyzed its autophosphorylation in the presence of either Mg2+ or Mn2+ as the divalent cation (Figure 3(c)). The ideal concentration of cation was 10 to 50 mM Mg2+ in 20 mM Tris-buffer pH 7.2 supplemented with 2 M NaCl with 4 M enzyme. The levels of Rio1p autokinase activity were comparable when assayed in the presence of either 2 M KCl or NaCl (data not shown). All subsequent kinase reactions were carried out in 20 mM Tris buffer (pH 7.2) with 2 M NaCl and 50 mM MgCl2. Multiple bands were visible on autoradiographs after the Rio1p-mediated autophosphorylation reaction including a doublet at the apparent molecular mass of the Rio1p monomer in addition to a band running at approximately twice the mass. MS analysis of Rio1p (from solution) did not identify any contaminating proteins in the purified fraction (Supplementary Table ). The multiple phospho-bands observed after assay may represent a transient dimer that forms between two Rio1p proteins in the presence of ATP. It is also possible that there is a contaminating protein that is phosphorylated in the reaction mixture and is not detected by MS or SYPRO staining.
3.6. Rio1p Phosphorylates 1 Proteins In Vitro
Purified Rio1p was assayed for its ability to phosphorylate 1-His6 purified from recombinant E. coli, 20S CPs purified from H. volcanii and -casein from bovine milk. While the 1-proteins purified from E. coli () form a mixture of hexameric rings and monomers, these recombinant proteins are not anticipated to be phosphorylated in recombinant E. coli based on their migration as a single spot by 2-DE (data not shown) as well as the distant phylogenetic relationship between E. coli and H. volcanii. Rio1p phosphorylated the unboiled preparation but was unable to phosphorylate the CPs or boiled or -casein (Figure 4(a)). When equal amounts of and site-directed 1 variants all purified from E. coli were used in the kinase assay, varying levels of 1-phosphorylation were observed. Wild-type was the most heavily phosphorylated under these conditions. Both Ser58Ala and Thr147Ala were phosphorylated but to a lesser extent than wild-type (2.5-fold lower based on densitometry of autoradiograph). The variant Thr158Ala was not phosphorylated under these conditions (Figure 4(b)). Based on these results, coupled with our earlier findings by 2-DE, 1 is most likely phosphorylated at 2 or 3 different sites. It is unclear at this time why Thr158Ala was not phoshporylated by Rio1p in vitro. There may be drastic structural changes in this 1 protein that prevent meaningful interaction with the kinase.
3.7. 1 Is Methylesterified on Acidic Residues
In addition to phosphorylation, the 1 protein was also methylated. Five unique methylation sites were reproducibly detected with high probability by MS/MS for 1 (Asp20, Glu27, Glu62, Glu112, and Glu161) (Supplementary Figure ). Substantiating this finding, several of the MS/MS-fragmentations also showed neutral losses of methyl groups (14). Unmodified 1 peptides that mapped to the same residues as those that were methylated were also detected. The analogous peptide in its unmodified state was almost always present in the same preparation, suggesting that methylated and unmethylated 1 proteins exist simultaneously in the cell. O-linked methylesterification is a reversible posttranslational modification, and therefore 1 methylation may be regulated. However, there was no clear correlation to growth phase or environmental conditions examined that increased or decreased the frequency of the methylated state of 1 protein (data not shown).
The psmA gene, encoding for the CP 1 protein, is cotranscribed with a putative S-adenylosylmethionine-dependent methyltransferase (HVO_1093) . Whether this enzyme mediates 1 methylation remains to be determined; its in vitroα1-metylating activity could not be demonstrated (data not shown). Future work is needed to better understand how and why this modification takes place.
Our previous work established that the three CP subunits of H. volcanii proteasomes (1, 2, and ) are modified post- and/or co-translational [34, 35]. The subunit is phosphorylated and cleaved to expose the active site Thr, and the 1 and 2 subunits are -acetylated on the initiator methionine residue. However, additional modifications were suspected based on the detection of long isoform chains of CP proteins including at least four α1 isoforms that were separated by 2-DE .
In this study, the covalent modification of archaeal proteasomal proteins was further examined using H. volcanii as a model system. Additional sites of posttranslational modification were identified including the phosphorylation of 1, 2, and PAN-A as well as the methyl esterification of 1. The phosphorylation of 1 was temporally regulated with the number of phosphatase-sensitive 1 isoforms reduced during late-stationary phase. The most acidic, presumably most phosphorylated isoform of 1, was only detected in association with as 20S CPs and was not detected in late-stationary phase.
In addition to elaborating on the prevalence and temporal nature of archaeal proteasomal PTMs, a member of the Rio-kinase family, Rio1p, was demonstrated to catalyze the phosphorylation of 1 in vitro. Modification of Ser58, Thr147, or Thr158 residues of α1 to alanine significantly reduced this Rio1p-dependent phosphotransfer activity. Although the finding that Rio1p phosphorylates 1 in vitro does not imply that this interaction is physiological, preliminary data on this substrate/kinase pair may prove valuable in understanding the role phosphorylation plays in regulating 20S CPs or other proteins in archaea. Substrate/kinase pairs are rarely known in this domain of life .
In vivo expression of phosphosite mutant forms of 1 revealed a biological connection between the phosphorylation of proteasomes and the pigmentation and viability of archaeal cells. While the details of this link remain to be determined at the molecular level, cell division is known to be highly regulated by the ubiquitin (Ub)-proteasome system in eukaryotic cells . In addition, 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, a key enzyme in the mevalonate pathway for production of isoprenoids (including carotenoids) is heavily regulated by proteasomes , and the Ub ligase CrgA is required for proper regulation of carotenogenesis in fungi .
In addition to phosphorylation, the α1 subunit of 20S CPs was found to be heavily methylated in H. volcanii. Although methylation has not been previously reported as a PTM of proteasomes in any organism, there are examples of reversible methylation of proteins in bacteria and archaea. Like bacteria, the haloarchaea have several taxis proteins that are O-methylated resulting in altered protein-protein interactions that enable cells to respond to external stimuli [51–53]. Haloarchaea may regulate the activity of other systems, such as proteasomes, using a similar strategy. Alternatively, the methyl esterification of 1 may be used to reversibly modulate the overall charge of proteasomal CPs. Most haloarchaeal proteins, including 1, are highly acidic (low pI) with the acidic residues typically on the surface of the protein forming a hydration shell. This shell aids in “salting-in” the protein and maintaining the activity of the protein at the unusually high concentrations of salt found in the cytosol of these microorganisms. Although we did not detect quantitative differences in the methylation state of 1 under different growth conditions, 1 is required for H. volcanii to properly respond to low salt stress . Whether the methyl esterification of 1 modifies protein-protein interactions in regulatory mechanisms such as PAN : CP interactions, buffers 1 in the cytosol at different concentrations of salt and/or mediates other physiological processes remains to be determined.
The authors thank Scott McClung and Dr. S. Chen at the University of Florida Interdisciplinary Center for Biological Research for assistance with mass spectrometry experiments and Dr. S. Shanker for DNA sequencing. This work was funded in part by NIH R01 GM057498 and DOE DE-FG02-05ER15650.
Supplementary material includes a list of the strains and plasmids used in this study, oligonucleotide primer pairs used for site-directed mutagenesis, MS/MS-results that support the homogeneity of Rio1p kinase from H. volcanii, and mass spectra used to identify the sites of post-translational modification of H. volcanii proteasomes including the phosphosites α 1 Thr147, α 2 Thr13/Ser14 and PAN-A Ser340 as well as the methylesterification sites α 1 Asp20, Glu27, Glu62, Glu112 and Glu161.
- D. H. Lee, K. Tanaka, T. Tamura, C. H. Chung, and A. Ichihara, “PRS3 encoding an essential subunit of yeast proteasomes homologous to mammalian proteasome subunit C5,” Biochemical and Biophysical Research Communications, vol. 182, no. 2, pp. 452–460, 1992.
- D. H. Lee, T. Tamura, C. H. Chung, K. Tanaka, and A. Ichihara, “Molecular cloning of the yeast proteasome PRS2 gene identical to the suppressor gene scl1+,” Biochemistry International, vol. 23, no. 4, pp. 689–696, 1991.
- E. Georgatsou, T. Georgakopoulos, and G. Thireos, “Molecular cloning of an essential yeast gene encoding a proteasomal subunit,” FEBS Letters, vol. 299, no. 1, pp. 39–43, 1992.
- T. Fujiwara, K. Tanaka, E. Orino et al., “Proteasomes are essential for yeast proliferation. cDNA cloning and gene disruption of two major subunits,” Journal of Biological Chemistry, vol. 265, no. 27, pp. 16604–16613, 1990.
- G. Zhou, D. Kowalczyk, M. A. Humbard, S. Rohatgi, and J. A. Maupin-Furlow, “Proteasomal components required for cell growth and stress responses in the haloarchaeon Haloferax volcanii,” Journal of Bacteriology, vol. 190, no. 24, pp. 8096–8105, 2008.
- C. S. Arendt and M. Hochstrasser, “Eukaryotic 20S proteasome catalytic subunit propeptides prevent active site inactivation by N-terminal acetylation and promote particle assembly,” EMBO Journal, vol. 18, no. 13, pp. 3575–3585, 1999.
- Y. Kimura, Y. Saeki, H. Yokosawa, B. Polevoda, F. Sherman, and H. Hirano, “N-terminal modifications of the 19S regulatory particle subunits of the yeast proteasome,” Archives of Biochemistry and Biophysics, vol. 409, no. 2, pp. 341–348, 2003.
- Y. Kimura, M. Takaoka, S. Tanaka et al., “Nα-acetylation and proteolytic activity of the yeast 20 S proteasome,” Journal of Biological Chemistry, vol. 275, no. 7, pp. 4635–4639, 2000.
- F. Tokunaga, R. Aruga, S. Iwanaga et al., “The NH2-terminal residues of rat liver proteasome (multicatalytic proteinase complex) subunits, C2, C3 and C8, are Nα-acetylated,” FEBS Letters, vol. 263, no. 2, pp. 373–375, 1990.
- F. Bardag-Gorce, R. Venkatesh, J. Li, B. A. French, and S. W. French, “Hyperphosphorylation of rat liver proteasome subunits: the effects of ethanol and okadaic acid are compared,” Life Sciences, vol. 75, no. 5, pp. 585–597, 2004.
- S. Bose, F. L. L. Stratford, K. I. Broadfoot, G. G. F. Mason, and A. J. Rivett, “Phosphorylation of 20S proteasome alpha subunit C8 (α7) stabilizes the 26S proteasome and plays a role in the regulation of proteasome complexes by γ-interferon,” Biochemical Journal, vol. 378, no. 1, pp. 177–184, 2004.
- S. Bose, P. Brooks, G. G. F. Mason, and A. J. Rivett, “γ-interferon decreases the level of 26 S proteasomes and changes the pattern of phosphorylation,” Biochemical Journal, vol. 353, no. 2, pp. 291–297, 2001.
- S. Bose, G. G. F. Mason, and A. J. Rivett, “Phosphorylation of proteasomes in mammalian cells,” Molecular Biology Reports, vol. 26, no. 1-2, pp. 11–14, 1999.
- J. G. Castaño, E. Mahillo, P. Arizti, and J. Arribas, “Phosphorylation of C8 and C9 subunits of the multicatalytic proteinase by casein kinase II and identification of the C8 phosphorylation sites by direct mutagenesis,” Biochemistry, vol. 35, no. 12, pp. 3782–3789, 1996.
- S. Claverol, O. Burlet-Schiltz, E. Girbal-Neuhauser, J. E. Gairin, and B. Monsarrat, “Mapping and structural dissection of human 20 S proteasome using proteomic approaches,” Molecular and Cellular Proteomics, vol. 1, no. 8, pp. 567–578, 2002.
- P. Fernández Murray, P. S. Pardo, A. M. Zelada, and S. Passeron, “In vivo and in vitro phosphorylation of Candida albicans 20S proteasome,” Archives of Biochemistry and Biophysics, vol. 404, no. 1, pp. 116–125, 2002.
- R. Horiguchi, M. Yoshikuni, M. Tokumoto, Y. Nagahama, and T. Tokumoto, “Identification of a protein kinase which phosphorylates a subunit of the 26S proteasome and changes in its activity during meiotic cell cycle in goldfish oocytes,” Cellular Signalling, vol. 17, no. 2, pp. 205–215, 2005.
- Y. Iwafune, H. Kawasaki, and H. Hirano, “Identification of three phosphorylation sites in the α7 subunit of the yeast 20S proteasome in vivo using mass spectrometry,” Archives of Biochemistry and Biophysics, vol. 431, no. 1, pp. 9–15, 2004.
- G. G. F. Mason, K. B. Hendil, and A. J. Rivett, “Phosphorylation of proteasomes in mammalian cells Identification of two phosphorylated subunits and the effect of phosphorylation on activity,” European Journal of Biochemistry, vol. 238, no. 2, pp. 453–462, 1996.
- G. G. F. Mason, R. Z. Murray, D. Pappin, and A. J. Rivett, “Phosphorylation of ATPase subunits of the 26S proteasome,” FEBS Letters, vol. 430, no. 3, pp. 269–274, 1998.
- P. S. Pardo, P. F. Murray, K. Walz, L. Franco, and S. Passeron, “In vivo and in vitro phosphorylation of the α7/PRS1 subunit of Saccharomyces cerevisiae 20 S proteasome: in vitro phosphorylation by protein kinase CK2 is absolutely dependent on polylysine,” Archives of Biochemistry and Biophysics, vol. 349, no. 2, pp. 397–401, 1998.
- A. J. Rivett, S. Bose, P. Brooks, and K. I. Broadfoot, “Regulation of proteasome complexes by γ-interferon and phosphorylation,” Biochimie, vol. 83, no. 3-4, pp. 363–366, 2001.
- K. Satoh, H. Sasajima, K.-I. Nyoumura, H. Yokosawa, and H. Sawada, “Assembly of the 26S proteasome is regulated by phosphorylation of the p45/Rpt6 ATPase subunit,” Biochemistry, vol. 40, no. 2, pp. 314–319, 2001.
- M. Tokumoto, R. Horiguchi, Y. Nagahama, K. Ishikawa, and T. Tokumoto, “Two proteins, a goldfish 20S proteasome subunit and the protein interacting with 26S proteasome, change in the meiotic cell cycle,” European Journal of Biochemistry, vol. 267, no. 1, pp. 97–103, 2000.
- M. Umeda, Y. Manabe, and H. Uchimiya, “Phosphorylation of the C2 subunit of the proteasome in rice (Oryza sativa L.),” FEBS Letters, vol. 403, no. 3, pp. 313–317, 1997.
- Y. Wakata, M. Tokumoto, R. Horiguchi, K. Ishikawa, Y. Nagahama, and T. Tokumoto, “Identification of -type subunits of the Xenopus 20S proteasome and analysis of their changes during the meiotic cell cycle,” FEBS Letters, vol. 5, article 18, 2004.
- F. Zhang, Y. Hu, P. Huang, C. A. Toleman, A. J. Paterson, and J. E. Kudlow, “Proteasome function is regulated by cyclic AMP-dependent protein kinase through phosphorylation of Rpt6,” Journal of Biological Chemistry, vol. 282, no. 31, pp. 22460–22471, 2007.
- M. Demasi, G. M. Silva, and L. E. S. Netto, “20 S proteasome from Saccharomyces cerevisiae is responsive to redox modifications and is S-glutathionylated,” Journal of Biological Chemistry, vol. 278, no. 1, pp. 679–685, 2003.
- S. C. Lee and B. D. Shaw, “A novel interaction between N-myristoylation and the 26S proteasome during cell morphogenesis,” Molecular Microbiology, vol. 63, no. 4, pp. 1039–1053, 2007.
- N. E. Zachara and G. W. Hart, “O-GlcNAc modification: a nutritional sensor that modulates proteasome function,” Trends in Cell Biology, vol. 14, no. 5, pp. 218–221, 2004.
- E. Seemuller, A. Lupas, D. Stock, J. Lowe, R. Huber, and W. Baumeister, “Proteasome from Thermoplasma acidophilum: a threonine protease,” Science, vol. 268, no. 5210, pp. 579–582, 1995.
- Y. Feng, D. L. Longo, and D. K. Ferris, “Polo-like kinase interacts with proteasomes and regulates their activity,” Cell Growth and Differentiation, vol. 12, no. 1, pp. 29–37, 2001.
- S. N. Djakovic, L. A. Schwarz, B. Barylko, G. N. DeMartino, and G. N. Patrick, “Regulation of the proteasome by neuronal activity and calcium/calmodulin-dependent protein kinase II,” Journal of Biological Chemistry, vol. 284, no. 39, pp. 26655–26665, 2009.
- H. L. Wilson, H. C. Aldrich, and J. Maupin-Furlow, “Halophilic 20S proteasomes of the archaeon Haloferax volcanii: purification, characterization, and gene sequence analysis,” Journal of Bacteriology, vol. 181, no. 18, pp. 5814–5824, 1999.
- M. A. Humbard, S. M. Stevens Jr., and J. A. Maupin-Furlow, “Post-translational modication of the 20S proteasomal proteins of the archaeon Haloferax volcanii,” Journal of Bacteriology, vol. 188, no. 21, pp. 7521–7530, 2006.
- S. J. Kaczowka and J. A. Maupin-Furlow, “Subunit topology of two 20S proteasomes from Haloferax volcanii,” Journal of Bacteriology, vol. 185, no. 1, pp. 165–174, 2003.
- C. J. Reuter, S. J. Kaczowka, and J. A. Maupin-Furlow, “Differential regulation of the PanA and PanB proteasome-activating nucleotidase and 20S proteasomal proteins of the haloarchaeon Haloferax volcanii,” Journal of Bacteriology, vol. 186, no. 22, pp. 7763–7772, 2004.
- M. A. Humbard, G. Zhou, and J. A. Maupin-Furlow, “The N-terminal penultimate residue of 20S proteasome α1 influences its Nα acetylation and protein levels as well as growth rate and stress responses of Haloferax volcanii,” Journal of Bacteriology, vol. 191, no. 12, pp. 3794–3803, 2009.
- P. A. Kirkland, J. Busby, S. Stevens Jr., and J. A. Maupin-Furlow, “Trizol-based method for sample preparation and isoelectric focusing of halophilic proteins,” Analytical Biochemistry, vol. 351, no. 2, pp. 254–259, 2006.
- A. J. Barrett, “Fluorimetric assays for cathepsin B and cathepsin H with methylcoumarylamide substrates,” Biochemical Journal, vol. 187, no. 3, pp. 909–912, 1980.
- A. L. Hartman, C. Norais, J. H. Badger et al., “The complete genome sequence of Haloferax volcanii DS2, a model archaeon,” PLoS One, vol. 5, article e9605, 2010.
- N. Blom, S. Gammeltoft, and S. Brunak, “Sequence and structure-based prediction of eukaryotic protein phosphorylation sites,” Journal of Molecular Biology, vol. 294, no. 5, pp. 1351–1362, 1999.
- C. Fischer, C. Geourjon, C. Bourson, and J. Deutscher, “Cloning and characterization of the Bacillus subtilis prkA gene encoding a novel serine protein kinase,” Gene, vol. 168, no. 1, pp. 55–60, 1996.
- N. LaRonde-LeBlanc, T. Guszczynski, T. Copeland, and A. Wlodawer, “Autophosphorylation of Archaeoglobus fulgidus Rio2 and crystal structures of its nucleotide-metal ion complexes,” FEBS Journal, vol. 272, no. 11, pp. 2800–2810, 2005.
- N. LaRonde-LeBlanc and A. Wlodawer, “A family portrait of the RIO kinases,” Journal of Biological Chemistry, vol. 280, no. 45, pp. 37297–37300, 2005.
- M. A. Gil, K. E. Sherwood, and J. A. Maupin-Furlow, “Transcriptional linkage of Haloferax volcanii proteasomal genes with non-proteasomal gene neighbours including RNase P, MOSC domain and SAM-methyltransferase homologues,” Microbiology, vol. 153, no. 9, pp. 3009–3022, 2007.
- B. Wang, S. Yang, L. Zhang, and Z.-G. He, “Archaeal eukaryote-like serine/threonine protein kinase interacts with and phosphorylates a forkhead-associated-domain-containing protein,” Journal of Bacteriology, vol. 192, no. 7, pp. 1956–1964, 2010.
- C. López-Otín and T. Hunter, “The regulatory crosstalk between kinases and proteases in cancer,” Nature Reviews Cancer, vol. 10, no. 4, pp. 278–292, 2010.
- R. Y. Hampton and R. M. Garza, “Protein quality control as a strategy for cellular regulation: lessons from ubiqu it in-mediated regulation of the sterol pathway,” Chemical Reviews, vol. 109, no. 4, pp. 1561–1574, 2009.
- F. Silva, E. Navarro, A. Peñaranda, L. Murcia-Flores, S. Torres-Martínez, and V. Garre, “A RING-finger protein regulates carotenogenesis via proteolysis-independent ubiquitylation of a white collar-1-like activator,” Molecular Microbiology, vol. 70, no. 4, pp. 1026–1036, 2008.
- A. Brooun, W. Zhang, and M. Alam, “Primary structure and functional analysis of the soluble transducer protein HtrXI in the archaeon Halobacterium salinarium,” Journal of Bacteriology, vol. 179, no. 9, pp. 2963–2968, 1997.
- E. Hildebrand and A. Schimz, “The lifetime of photosensory signals in Halobacterium halobium and its dependence on protein methylation,” Biochimica et Biophysica Acta, vol. 1052, no. 1, pp. 96–105, 1990.
- S. Hou, A. Brooun, H. S. Yu, T. Freitas, and M. Alam, “Sensory rhodopsin II transducer HtrII is also responsible for serine chemotaxis in the archaeon Halobacterium salinarum,” Journal of Bacteriology, vol. 180, no. 6, pp. 1600–1602, 1998.
Copyright © 2010 Matthew A. Humbard et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.