The DJ-1/ThiJ/PfpI superfamily is a group of proteins found in diverse organisms. This superfamily includes versatile proteins, such as proteases, chaperones, heat-shock proteins and human Parkinson's disease protein. Most members of the DJ-1/ThiJ/PfpI superfamily are oligomers and are classified into subfamilies depending on discriminating quaternary structures (DJ-1, YhbO and Hsp types). SAV1875, a conserved protein from Staphylococcus aureus, is a member of the YhbO-type subfamily. However, its structure and function remain unknown. Thus, to understand the function and activity mechanism of this protein, the crystal structure of SAV1875 from S. aureus was determined. The overall fold of SAV1875 is similar to that observed for the DJ-1/ThiJ/PfpI superfamily. The cysteine residue located in the dimeric interface (Cys105) forms a catalytic triad with His106 and Asp77, and it is spontaneously oxidized to Cys105-SO2H in the crystal structure. To study the oxidative propensity of Cys105 and the corresponding functional differences with changes in cysteine oxidation state, the crystal structures of SAV1875 variants E17N, E17D and C105D, and over-oxidized SAV1875 were determined. We identified SAV1875 as a novel member of the YhbO-type subfamily exhibiting chaperone function. However, if SAV1875 is over-oxidized further with H2O2, its chaperone activity is eliminated. On the basis of our study, we suggest that SAV1875 functions as a chaperone and the redox state of Cys105 may play an important role.

INTRODUCTION

Staphylococcus aureus is one of the most common pathogens, causing various diseases ranging from mild infections, such as skin infections and food poisoning, to life-threatening infections, such as sepsis, endocarditis and toxic shock syndrome [1]. S. aureus has adapted to circumvent therapeutic strategies by developing resistance to antibiotics. These resistant strains were initially observed in hospital settings, but are now prevalent in the community, infecting otherwise healthy individuals [2]. Most strains of S. aureus are resistant not only to penicillin but also to methicillin [3,4]. Moreover, there are newly emerging strains such as VRSA (vancomycin-resistant S. aureus). More specifically, vancomycin-resistant S. aureus bacteria are classified as VISA (vancomycin-intermediate S. aureus) if the MIC (minimum inhibitory concentration) against vancomycin is 4–8 μg/ml and classified as VRSA if the MIC is more than 16 μg/ml [5]. For example, S. aureus Mu50 (ATCC 700699), which was first isolated from a sternal abscess in a child who had received a prolonged course of vancomycin treatment [6], is classified as VISA because the MIC against vancomycin is 8 μg/ml [7,8]. Because of this bacterial resistance, the prognosis for a S. aureus infection is still poor despite early diagnosis and appropriate treatment [9]. Currently, the spread of resistance in bacterial populations is a major health challenge. Therefore understanding the detailed mechanism of how bacteria adapt to environmental changes is essential for developing better therapeutics. Thus research on the bacterial system under environmental stress conditions is emerging as being important.

In the present study, the structural and functional identification of wild-type and SAV1875 mutants was conducted. SAV1875, a conserved hypothetical protein from S. aureus Mu50, is a member of the DJ-1/ThiJ/PfpI superfamily. However, its structure and function are unknown. Because the functionally known members of the DJ-1/ThiJ/PfpI superfamily (denoted DJ-1 superfamily hereinafter) are typically involved in the general stress response system, SAV1875 is predicted to have a role in cellular protection against environmental stress. Several examples of DJ-1 superfamily proteins include the human DJ-1 protein that protects cells from oxidative stress [10], Escherichia coli Hsp31, which is involved in thermal stress protection, acid resistance and glyoxalase activity [1113], PhpI and PfpI, which are proteases from Pyrococcus horikoshii [14] and Pyrococcus furious [15] respectively, Deinococcus radiodurans DR1199, which is a general stress-resistance protein [16], and E. coli YhbO, a protein highly sensitive to oxidative, thermal, UV and pH stresses [17].

The crystal structures of several members of the DJ-1 superfamily have been solved, and all members of the family have common features [18]. Most structurally known proteins in the DJ-1 superfamily are oligomers, and each monomer consists of an α/β sandwich fold which includes a cysteine nucleophile at the ‘nucleophilic elbow’, between a β-strand and an α-helix. This cysteine residue on the nucleophilic elbow forms a catalytic dyad/triad. Despite these similarities, DJ-1 superfamily proteins can be sorted into distinct subfamilies according to their quaternary structure: DJ-1, YhbO and Hsp (heat-shock protein) type [18,19]. Each subfamily has different characteristics in the oligomeric state and a catalytic dyad/triad conformation. Consequently, these commonalities and distinctive features suggest that the reactive cysteine and oligomeric state of DJ-1 superfamily proteins appear to be crucial in the functions of these proteins [20,21].

In the present study, we elucidated the crystal structures of the wild-type, mutant and over-oxidized SAV1875. To gain further insight into SAV1875 structure and function, chaperone and protease activity tests were conducted. These functional tests revealed that SAV1875 has chaperone activity. The three mutants E17N, E17D and C105D and over-oxidized SAV1875 containing Cys105-SO3H (cysteine sulfonic acid) were designed to identify the oxidative propensity of Cys105 as well as the corresponding structural and functional differences. The over-oxidized SAV1875 containing Cys105-SO3H lost its chaperone activity. The implication of these results is that SAV1875 functions as a molecular chaperone and the redox state of SAV1875 is related to the activity. The structural and functional research regarding oxidative stress will promote a better understanding of bacterial physiology and how bacteria defend against oxidative stress.

EXPERIMENTAL

Cloning, protein expression and purification

The ORF encoding SAV1875 was amplified from S. aureus Mu50 genomic DNA by PCR using 5′-GACTG-CATATGACTAAAAAAGTAGCAATTATTC-3′ as the forward primer and 5′-GTACACTCGAGTTGTAATTGTTTAACGAT-TTCT-3′ as the reverse primer. The NdeI and XhoI restriction sites are underlined and were used for cloning into the pET-21a(+) vector (Novagen). The resulting construct has eight additional residues (LEHHHHHH) that encode a C-terminal hexahistidine tag. For removal of the hexahistidine tag, the SAV1875 gene was inserted into a pET-28a(+) (Novagen) vector. The resulting construct in pET-28a(+) comprised residues 1–171 with an additional 20 residues (MGSSHHHHHHSSGLVPRGSH) containing a thrombin-cleavage site. The sequence of the cloned gene was confirmed by DNA sequencing (results not shown).

To prepare mutants, an EZchange Site-Directed Mutagenesis kit (Enzynomics) was used to generate point mutations in the SAV1875 recombinant pET-21a(+) plasmid. The point mutations resulted in separate multiple recombinant plasmids, specifically E17N, E17D and C105D. The sequences of the reconstructed mutants were confirmed by DNA sequencing (results not shown).

Wild-type and SAV1875 mutants (E17N, E17D and C105D) in pET-21a(+) were overexpressed in E. coli BL21(DE3) cells (Novagen) and grown at 37°C in LB medium supplemented with ampicillin (50 μg/ml) until the D600 reached 0.5. Recombinant protein expression was induced by the addition of 0.5 mM IPTG, and the cells were allowed to grow for an additional 4 h at 37°C. The cells were harvested by centrifugation at 4°C. The cell pellet was resuspended in lysis buffer (50 mM Tris/HCl, pH 7.5, and 500 mM NaCl) and disrupted at 4°C using an Ultrasonic processor (Cole-Parmer). The cell lysate was centrifuged at 30000 g for 1 h at 4°C. The cleared supernatant was purified by binding to an Ni-NTA (Ni2+-nitrilotriacetate) affinity column (Qiagen; 3 ml of resin per litre of cell lysate) and eluted with binding buffer containing 100 mM imidazole. Further purification and buffer exchange were achieved by size-exclusion chromatography using a Superdex 75 (10/300 GL) column (GE Healthcare Life Sciences) that was equilibrated previously with buffer (50 mM Tris/HCl, pH 7.5, and 200 mM NaCl). The purities of the hexahistidine-tagged wild-type and SAV1875 mutants were estimated to be over 95% by SDS/PAGE. The purified proteins were concentrated to 10 mg/ml by ultrafiltration in 10000 Da molecular-mass cut-off spin columns (Millipore). The absorbance at 280 nm was measured, and the calculated molar absorption coefficient of 5960 M−1·cm−1 (Swiss-Prot; http://www.expasy.org) was employed to determine the protein concentration.

Crystallization and data collection

Crystallization was performed at 293 K by the hanging-drop vapour-diffusion method using 24-well VDX plates (Hampton Research). Initial crystallization conditions were established using screening kits from Hampton Research (Crystal Screens I and II, Index, PEG/Ion and MembFac) and from Emerald BioSystems (Wizard I, II, III and IV). For the optimal growth of the SAV1875 crystals, each hanging drop was prepared on a siliconized coverslip by mixing 1 μl of protein solution and 1 μl of precipitant solution [29% (w/v) PEG monomethyl ether 2000 and 100 mM Bis-Tris, pH 6.5], and this drop was equilibrated against a 1 ml reservoir of precipitant solution. The SAV1875 mutants (E17D, E17N and C105D) and over-oxidized SAV1875 crystals were prepared in the same manner, but with different precipitant solutions: for E17D, 44% (w/v) PPG (polypropylene glycol) 400 and 100 mM Bis-Tris (pH 6.0); for E17N, 47.5% (w/v) PPG400, 0.2 M guanidinium chloride and 100 mM Bis-Tris (pH 6.0); for C105D, 37.5% (w/v) PPG400 and 100 mM Bis-Tris (pH 6.0); and for over-oxidized SAV1875, 47.5% (w/v) PPG400 and 100 mM Bis-Tris (pH 5.5). These conditions yielded needle-shaped crystals for each protein that grew to dimensions of 1.0 mm×0.4 mm×0.4 mm in 3 days. All crystals belonged to space group P21212 and contain two molecules per asymmetric unit.

For crystal freezing, the crystals were transferred to a cryoprotectant solution with 30% (v/v) glycerol in the crystallization condition for several minutes before being flash-frozen in a stream of nitrogen gas at 100 K. Diffraction data were collected on beamlines 6C and 5C at the Pohang Light Source, Pohang, Republic of Korea, and BL-17A at the Photon Factory, Japan. The raw data were processed and scaled using the HKL2000 program suite [22]. Further data analysis was carried out using the CCP4 suite [23]. Data collection statistics are summarized in Table 1. 

Table 1
Crystallographic data collection and refinement statistics

Values in parentheses indicate the statistics for the highest resolution shell. Rsym=Σ(|Ihkl−<Ihkl>|/Σ<Ihkl>, where Ihkl=single value of measured intensity of hkl reflection, and <Ihkl>=mean of all measured values of intensity of hkl reflection. Rwork=Σ(|FobsFcalc|/ΣFobs, where Fobs=observed structure factor amplitude, and Fcalc=structure factor calculated from the model. Rfree is computed in the same manner as Rwork, but from a test set containing 5% of data excluded from the refinement calculation.

 Wild-type E17D E17N C105D Over-oxidized SAV1875 
Data collection 
Beamline PAL-6C PAL-5C PAL-5C PF-17A PAL-5C 
 Wavelength (Å) 1.24 0.98 0.98 0.97 0.98 
 Resolution range (Å) 40.00–2.10 40.00–1.80 40.00–1.90 40.00–1.80 40.00–1.65 
 Space group P2121P2121P2121P2121P2121
 Unit cell parameters (Å) a=81.455 a=81.313 a=82.353 a=82.055 a=81.429 
 b=95.004 b=94.150 b=95.647 b=93.946 b=94.715 
 c=42.648 c=42.198 c=41.569 c=42.843 c=42.418 
 Observations (total/unique) 129585/19995 251449/30762 251442/26532 422814/31130 470168/39845 
 Completeness (%) 99.5 (93.0) 99.3 (100.0) 99.8 (100.0) 99.0 (98.2) 98.2 (83.1) 
 Redundancy 6.5 (4.7) 7.2 (8.2) 9.5 (10.2) 13.6 (13.9) 11.8 (7.6) 
 Rsym 8.2 (22.6) 6.8 (16.3) 8.7 (41.8) 8.5 (39.5) 5.7 (23.6) 
 I/σ 39.2 (9.0) 58.2 (26.3) 55.1 (11.3) 66.9 (13.6) 72.3 (10.2) 
Refinement 
 Rwork (%) 19.11 19.39 18.75 18.46 19.15 
 Rfree (%) 22.65 22.44 21.64 21.76 21.72 
 Protein atoms 2,729 2,633 2,668 2,647 2,632 
 Water molecules 52 68 46 136 114 
 Average B value (Å221.35 20.53 24.08 21.09 17.24 
 RMSD bond (Å) 0.007 0.009 0.007 0.009 0.009 
 RMSD angle (°) 1.052 1.330 1.178 1.320 1.379 
Ramachandran analysis (%) 
 Favoured region 98.29 99.11 99.12 98.53 99.41 
 Allowed region 1.71 0.89 0.88 1.17 0.59 
 Outliers 0.29 
 Wild-type E17D E17N C105D Over-oxidized SAV1875 
Data collection 
Beamline PAL-6C PAL-5C PAL-5C PF-17A PAL-5C 
 Wavelength (Å) 1.24 0.98 0.98 0.97 0.98 
 Resolution range (Å) 40.00–2.10 40.00–1.80 40.00–1.90 40.00–1.80 40.00–1.65 
 Space group P2121P2121P2121P2121P2121
 Unit cell parameters (Å) a=81.455 a=81.313 a=82.353 a=82.055 a=81.429 
 b=95.004 b=94.150 b=95.647 b=93.946 b=94.715 
 c=42.648 c=42.198 c=41.569 c=42.843 c=42.418 
 Observations (total/unique) 129585/19995 251449/30762 251442/26532 422814/31130 470168/39845 
 Completeness (%) 99.5 (93.0) 99.3 (100.0) 99.8 (100.0) 99.0 (98.2) 98.2 (83.1) 
 Redundancy 6.5 (4.7) 7.2 (8.2) 9.5 (10.2) 13.6 (13.9) 11.8 (7.6) 
 Rsym 8.2 (22.6) 6.8 (16.3) 8.7 (41.8) 8.5 (39.5) 5.7 (23.6) 
 I/σ 39.2 (9.0) 58.2 (26.3) 55.1 (11.3) 66.9 (13.6) 72.3 (10.2) 
Refinement 
 Rwork (%) 19.11 19.39 18.75 18.46 19.15 
 Rfree (%) 22.65 22.44 21.64 21.76 21.72 
 Protein atoms 2,729 2,633 2,668 2,647 2,632 
 Water molecules 52 68 46 136 114 
 Average B value (Å221.35 20.53 24.08 21.09 17.24 
 RMSD bond (Å) 0.007 0.009 0.007 0.009 0.009 
 RMSD angle (°) 1.052 1.330 1.178 1.320 1.379 
Ramachandran analysis (%) 
 Favoured region 98.29 99.11 99.12 98.53 99.41 
 Allowed region 1.71 0.89 0.88 1.17 0.59 
 Outliers 0.29 

Structure determination and refinement

To determine the structure of wild-type SAV1875, molecular replacement was used with the program MolRep [24] within the CCP4 suite [23] using the homologous structure of YhbO from E. coli (PDB code 1OI4) as a search model. To determine structures of the SAV1875 mutants (E17D, E17N and C105D) and over-oxidized SAV1875, molecular replacement was performed using the wild-type SAV1875 structure as the search model. Refinement of each crystal structure was done through iterative cycles of model building using COOT [25], followed by refinement of the models with Refmac5 [25,26]. A 5% portion of the data was set aside before refinement for the Rfree calculations for each dataset [27]. Solvent molecules became apparent in the later stages of refinement and were added into the model. Further refinement was pursued until no further decrease in Rfree was observed. The final models exhibited good stereochemical geometry (Table 1) when the overall geometry was validated with PROCHECK [28]. Refinement statistics are summarized in Table 1. Structural alignments were carried out using the program PyMOL (http://www.pymol.org) and UCSF Chimera (http://www.cgl.ucsf.edu/chimera) [29], which were then used for the construction and generation of all Figures. Protein interfaces, surfaces and assemblies were calculated using the PISA server at the European Bioinformatics Institute (http://www.ebi.ac.uk/pdbe/prot_int/pistart.html) [30].

PAGE

SDS/PAGE was conducted according to the Laemmli method [31] using a 12% (w/v) polyacrylamide gel. The samples were treated with 1% (w/v) SDS and 5% (v/v) 2-mercaptoethanol at 100°C for 5 min before electrophoresis in a vertical Mini Gel system (Bio-Rad Laboratories). The proteins were stained with Coomassie Brilliant Blue R250 (Thermo Scientific). Additionally, for the separation of SAV1875 depending on the m/z ratio between oxidized and reduced state, native PAGE was performed and analysis was conducted using a 10% (w/v) polyacrylamide gel without either SDS or 2-mercaptoethanol. Native PAGE was performed in a buffer (25 mM Tris/HCl, pH 8.3, and 192 mM glycine). The staining was performed as described above for SDS/PAGE.

Mass spectrometry

Mass analyses were performed on a nano-HPLC system (Dionex Ultimate 3000 RSLCnano System, Thermo Scientific) coupled with a hybrid quadrupole-obitrap mass spectrometer (Q-Exactive, Thermo Scientific) at the NICEM (National Instrumentation Center for Environmental Management, Seoul National University, Seoul, Republic of Korea). Protein samples (10 μl) were loaded on to a C8 reverse-phase column (INNO5, Young Jin Biochrom). HPLC was used for room temperature gradient elution at a flow rate of 150 μl/min by using a linear gradient from 0.1% formic acid in water (solvent A) to 0.1% formic acid in acetonitrile (solvent B). The total run time for each sample was 20 min. The molecular mass of protein was generated from several multiply charged peaks using the Xcalibur 2.2 Software (Thermo Scientific).

Protein oxidation

For the complete oxidation of cysteine, the wild-type and SAV1875 mutants at 20 mg/ml (1 mM) were incubated for 30 min at room temperature in H2O2 (Sigma–Aldrich) with a molar ratio of protein/H2O2 of 1:50. After treatment, excess H2O2 was removed by extensive dialysis with buffer (50 mM Tris/HCl, pH 7.5). The control consisted of the same protein with water, which was incubated under the same conditions. The oxidation states of Cys105 were determined by MS and native PAGE.

Protease activity assay and zymogram

The protease activity assay of wild-type, mutant and over-oxidized SAV1875 was performed using the MGT protease assay kit (Marker Gene Technologies). FITC-labelled casein was cleaved into smaller fragments, and highly fluorescently labelled peptides were released [32]. Fluorescence increase is proportional to protease activity, and this fluorescence was measured in a continuous assay format using a Multi-Mode microplate reader (SpectraMax M5e) with excitation and emission wavelengths of 485 and 528 nm respectively. The FITC–casein was diluted to 1 mg/ml with reaction buffer (100 mM sodium phosphate, pH 7.6, and 150 mM NaCl). By mixing FITC–casein with a gradient of protein concentrations (final protein concentration from 1 μM to 15 μM, separately), the protease activity was monitored. The proteins were diluted into the reaction buffer immediately before conducting the test. The mixture of FITC–casein and protein (wild-type, mutant and over-oxidized SAV1875) were sealed and incubated for 2 h at 37°C under protection from UV light.

For zymogram analysis, 0.1% gelatin co-polymerized with the acrylamide gels were used. Electrophoresis was performed at 4°C at a constant voltage of 100 V. The gels were washed in 2.5% (w/v) Triton X-100 solution at room temperature for 40 min. Following washing, gels were incubated for 20 h at 37°C in a buffer (50 mM Tris/HCl, pH 8.0, 5 mM CaCl2 and 0.02% NaN3) for proteolytic activity. Staining with Coomassie Brilliant Blue shows the proteolytically cleaved sites as a clear band on a dark background [33].

Chaperone assay

To monitor the chaperone activity of wild-type, mutant and over-oxidized SAV1875, citrate synthase was employed as a substrate [34,35]. Initially, to identify chaperone activity, 75 μg of citrate synthase (Sigma–Aldrich) was mixed with a solution of 100 mM Tris/HCl, pH 8.0, 20 mM DTT and 6 M guanidinium chloride. The citrate synthase mixture (75 μg of citrate synthase, 100 mM Tris/HCl, pH 8.0, 6 M guanidinium chloride and 20 mM DTT) was incubated for 1 h at 25°C; consequently, the citrate synthase in this solution was denatured. After incubation, refolding of citrate synthase was achieved by 100-fold dilution with a solution of 100 mM Tris/HCl (pH 8.0) containing 5 μM wild-type, mutant or over-oxidized SAV1875. The diluted solution was mixed with acetyl-CoA, oxaloacetate, MnCl2 and DTNB [5,5′-dithiobis-(2-nitrobenzoic acid)] to detect the activity of citrate synthase (100 mM Tris/HCl, pH 8.0, 1 mM DTNB, 0.2 mM MnCl2, 0.4 mM oxaloacetic acid and 0.3 mM acetyl-CoA). After mixing, only the active refolded enzyme will break acetyl-CoA into the acetyl group and CoA. The CoA reacts with DTNB, which acts as a colouring agent, and this produces a yellow TNB (5-thio-2-nitrobenzoic acid)–CoASH compound that is detectable at 412 nm using a Multi-Mode microplate reader (SpectraMax M5e) [36]. After a 70 min reaction period, the highest specific activity value obtained for TNB–CoASH was considered to be 100% and the lowest was 0%. The calculated specific activities of the wild-type, mutant and over-oxidized SAV1875 were expressed as a percentage of this value.

PDB codes

Protein co-ordinates and structure factors have been deposited in the RCSB PDB under codes 4Y0N for wild-type SAV1875, 4Y1F for E17D SAV1875, 4Y1G for E17N SAV1875, 4Y1E for C105D SAV1875 and 4Y1R for over-oxidized SAV1875.

RESULTS

Structure of SAV1875

The 2.1 Å (1 Å=0.1 nm) crystal structure of SAV1875 has clear electron density for 171 amino acids including the additional C-terminal tag and lacks the methionine residue at position 1. The protein consists of an α/β sandwich fold with eight α-helices and eight β-strands. Six β-strands, β2 (residues 31–35), β1 (residues 4–8), β5 (residues 67–70), β6 (residues 101–104), β8 (residues 150–153) and β7 (residues 145–147), are aligned in the centre, wherein the β7 strand is antiparallel to the central β-strands. Eight α-helices surround the core of β-strands. Both sides of the β-strands are covered with α-helices [α1 (residues 15–27), α7 (residues 156–158) and α8 (residues 160–171) are located on one side, and α2 (residues 62–64), α3 (residues 75–81), α4 (residues 86–96), α5 (residues 107–113) and α6 (residues 126–134) are on the other side]. Additionally, antiparallel strands β3 (residues 42–44) and β4 (residues 50–52) are located on top of the sandwich structure (Figure 1A).

Crystal structure of SAV1875

Figure 1
Crystal structure of SAV1875

(A) Ribbon representation of the SAV1875 monomer. SAV1875 shows a sandwich structure with a β1-α1-β2-β3-β4-α2-β5-α3-α4-β6-α5-α6-β7-β8-α7-α8 topology. (B) The SAV1875 dimer is shown in a ribbon representation from side view. Chain A is coloured cyan and chain B is coloured green. Three helices (α3, α5 and α6) are involved in the dimeric interface. The nucleophilic elbow (strand–nucleophile–helix motif) in each domain is shown in a darker colour. Cysteine is positioned in the turn of the nucleophilic elbow, and the catalytic triad is shown as sticks (CSD105, His106 and Asp77 from the adjacent monomer). The oxidized Cys105 is denoted CSD105. (C) Potential surface charge of the crystal structure of SAV1875, calculated with UCSF Chimera [29], where surfaces are coloured between −10 kcal/mol·e (red) and +10 kcal/mol·e (blue) (1 kcal=4.184 kJ). Most of the surface is negatively charged. The canyon is indicated.

Figure 1
Crystal structure of SAV1875

(A) Ribbon representation of the SAV1875 monomer. SAV1875 shows a sandwich structure with a β1-α1-β2-β3-β4-α2-β5-α3-α4-β6-α5-α6-β7-β8-α7-α8 topology. (B) The SAV1875 dimer is shown in a ribbon representation from side view. Chain A is coloured cyan and chain B is coloured green. Three helices (α3, α5 and α6) are involved in the dimeric interface. The nucleophilic elbow (strand–nucleophile–helix motif) in each domain is shown in a darker colour. Cysteine is positioned in the turn of the nucleophilic elbow, and the catalytic triad is shown as sticks (CSD105, His106 and Asp77 from the adjacent monomer). The oxidized Cys105 is denoted CSD105. (C) Potential surface charge of the crystal structure of SAV1875, calculated with UCSF Chimera [29], where surfaces are coloured between −10 kcal/mol·e (red) and +10 kcal/mol·e (blue) (1 kcal=4.184 kJ). Most of the surface is negatively charged. The canyon is indicated.

Oligomeric state of SAV1875

The crystal structure revealed that SAV1875 exists as a compact homodimer with an interface that buries ∼828 Å2 per subunit (12.3% of the subunit surface). The dimer involves contacts between three helices (α3, α5 and α6) on each monomer (Figure 1B). On its dimeric interface, there are three salt bridge pairs between Asp77 and His106, Asp115 and Lys130, and Asp131 and Arg80. Numerous hydrophobic residues are also found in the dimeric interface, including Gly73, Phe74, His78, Gly81, Ile112, Leu126, Val128 and Leu132. On the basis of its quaternary structure, SAV1875 was classified as a new member of the YhbO-type subfamily.

The DJ-1 superfamily proteins can be classified into subfamilies on the basis of their quaternary structure: DJ-1-type, YhbO-type and Hsp-type. The DJ-1-type proteins (DJ-1 and YajL) form a dimeric interface consisting of α-helices, β-strands and loops. The YhbO-type proteins (YhbO, PhpI, DR1199 and Ton1285) interact with the other subunits via three helices. The Hsp-type proteins (Hsp31) form a unique dimerization surface consisting of N-terminal β-strands and loops [18,19].

The dimer crystal structure of SAV1875 is consistent with the results of size-exclusion chromatography, which indicates a dimeric state of SAV1875 with an apparent molecular mass of 39 kDa in solution. However, the YhbO-type dimer proteins can form a ring-like hexamer in solution by trimerization. With the exception of DR1199, which has a long protrudent loop towards the centre, other YhbO-type proteins, including PhpI, YhbO and Ton1285, can exist as hexamers in a solution [18,37]. Because an additional terminal amino acid extension alters the oligomerization modes of YhbO, we prepared a variant of SAV1875: a histidine-tag-removed protein. Both forms of histidine-tagged and tag-removed SAV1875 proteins form only dimers in a solution. This may be because of the wider dimerization mode of the two subunits in SAV1875 compared with PhpI. If SAV1875 were to make the same trimerization as PhpI, the helices and loops, which form a DJ-1-type dimeric interface, would sterically inhibit each other (Supplementary Figure S1).

When viewing the surface area, the electrostatic distribution of SAV1875 reveals several characteristics. The majority of the surface is negatively charged with several exceptions that indicate hydrophobic patches on the surface area which are dominated by Ile16, Ala28, Gly29, Ala39, Val44, His47, Gly48, Ala58, His78, Gly85, Gly119, Leu126, Arg129, His138, Val146, Ala148 and Pro156. The dimeric interface of SAV1875 shows a deep canyon and this spans the dimer (Figure 1C). These characteristics are similar to those of Hsp31 [11], known chaperone proteins in the DJ-1 superfamily. Therefore these findings suggest that SAV1875 may function as a chaperone.

Comparison of SAV1875 with DJ-1 superfamily proteins

The overall fold of SAV1875 shows structural similarity to other members of DJ-1 superfamily proteins, as predicted by sequence homology: YhbO (49% sequence identity), PhpI (42% sequence identity), DR1199 (40% sequence identity), Ton1285 (39% sequence identity) and DJ-1 (29% sequence identity). A structure-based sequence alignment was acquired using the ClustalW web server tool [38] and viewed using the ESPript web server tool (http://espript.ibcp.fr/ESPript/ESPript/) [39,40]. The sequence alignment and superposition of SAV1875 with the structures of YhbO, PhpI, DR1199, Ton1285 and DJ-1 are shown in Figures 2(A) and 2(B) respectively.

Comparison of SAV1875 with other members of the DJ-1/ThiJ/PfpI superfamily

Figure 2
Comparison of SAV1875 with other members of the DJ-1/ThiJ/PfpI superfamily

(A) Comparison of SAV1875 with other members of the DJ-1/ThiJ/PfpI superfamily. Sequence alignment of SAV1875 with YhbO (49% sequence identity), PhpI (42% sequence identity), DR1199 (40% sequence identity), Ton1285 (39% sequence identity) and DJ-1 (29% sequence identity) proteins. Identical residues are coloured white on a red background and similar residues are red on a white background. Secondary-structure elements (springs, α-helices; arrows, β-strands) are represented above the sequences and are numbered. The Figure was constructed using ESPript [40]. (B) The superposition of SAV1875 (cyan) with the structures of YhbO (pink, 0.8 Å RMSD, PDB code 1OI4), PhpI (blue, 0.9 Å RMSD, PDB code 1G2I), DR1199 (orange, 1.1 Å RMSD, PDB code 2VRN), Ton1285 (green, 0.9 Å RMSD, PDB code 3L18) and DJ-1 (grey, 1.2 Å RMSD, PDB code 3SF8). The overall fold of the core domain of SAV1875 is similar to that of other members of the DJ-1/ThiJ/PfpI superfamily. The Figure was constructed using UCSF Chimera [29].

Figure 2
Comparison of SAV1875 with other members of the DJ-1/ThiJ/PfpI superfamily

(A) Comparison of SAV1875 with other members of the DJ-1/ThiJ/PfpI superfamily. Sequence alignment of SAV1875 with YhbO (49% sequence identity), PhpI (42% sequence identity), DR1199 (40% sequence identity), Ton1285 (39% sequence identity) and DJ-1 (29% sequence identity) proteins. Identical residues are coloured white on a red background and similar residues are red on a white background. Secondary-structure elements (springs, α-helices; arrows, β-strands) are represented above the sequences and are numbered. The Figure was constructed using ESPript [40]. (B) The superposition of SAV1875 (cyan) with the structures of YhbO (pink, 0.8 Å RMSD, PDB code 1OI4), PhpI (blue, 0.9 Å RMSD, PDB code 1G2I), DR1199 (orange, 1.1 Å RMSD, PDB code 2VRN), Ton1285 (green, 0.9 Å RMSD, PDB code 3L18) and DJ-1 (grey, 1.2 Å RMSD, PDB code 3SF8). The overall fold of the core domain of SAV1875 is similar to that of other members of the DJ-1/ThiJ/PfpI superfamily. The Figure was constructed using UCSF Chimera [29].

In addition to the similarity of the overall fold within the DJ-1 superfamily proteins, all of the DJ-1 superfamily proteins share the reactive cysteine residue at a sharp turn between a β-strand and α-helix. This strand–nucleophile–helix motif is defined as a ‘nucleophilic elbow’. Residues 101–113 (α5 and β6) of SAV1875 form the nucleophilic elbow. Cys105 lies at the sharp turn between the α5 helix and the β6 strand (Figure 1B). In nearly all the DJ-1 superfamily proteins, except the DJ-1-type subfamily, the cysteine forms a catalytic triad with a neighbouring histidine and an acidic residue. However, the structural features of the catalytic triad vary between the different DJ-1 subfamilies. The YhbO-type proteins constitute a catalytic triad with cysteine, the histidine next to cysteine, and an acidic residue from the other subunit [14,16]. The Hsp-type proteins form a catalytic triad using cysteine, the histidine next to cysteine and an acidic residue from an intrasubunit from the other domain (P domain) [11,41].

The catalytic triad that was found in SAV1875 consists of Cys105, His106 and Asp77 (from the adjacent monomer) and was located in its dimeric interface, specifically on the depressed area (Figure 1B). The catalytic triad shares the same handedness with the YhbO-type proteins. Cys105 interacts with Nδ His106, and Asp77 interacts with Nε of His106. However, Cys105 is oxidized to Cys105-SO2H in the crystal structure of SAV1875 and is denoted CSD105. Similar results were observed previously in the homologous cysteine residues from the DJ-1 superfamily proteins, in which the cysteine residue involved was very sensitive to oxidation, forming either Cys-SOH or Cys-SO2H. In the crystal structure of DJ-1 (DJ-1-type), the cysteine is oxidized to Cys-SO2H. YajL (DJ-1-type) exhibits both Cys-SOH and Cys-SO2H modification on each subunit. The crystal structure of DR1199 (YhbO-type) shows a Cys-SOH modification, and the cysteine residue in YDR533Cp (Hsp-type) exists as Cys-SO2H [16,4143]. These oxidized cysteine residues are stabilized by specific surrounding residues near the cysteine residue, including an acidic residue which is analogous to Glu17 in SAV1875.

Stabilization of oxidized Cys105

The cysteine residue at the nucleophilic elbow is absolutely conserved in the DJ-1 superfamily proteins, and is considered to be a critical residue for their catalytic function [44]. In addition to the cysteine residue, another highly conserved acidic residue in the DJ-1 superfamily is glutamate, which donates hydrogen bonds to the oxidized cysteine residue. SAV1875 has a conserved glutamate residue, Glu17. In the crystal structure of SAV1875, oxidized Cys105 makes hydrogen bonds with Glu17 (2.71 Å), His106 (2.69 Å) and the Gly73/His106 amide pocket (3.02 Å and 2.96 Å respectively; Figure 3A). Because the Oδ of oxidized Cys105 and the Oε of Glu17 form hydrogen bonds, we speculated that a glutamate substitution at position 17 would alter the cysteine oxidation status without disturbing the catalytic triad. Thus aspartate and asparagine mutants (E17D and E17N respectively) were designed because these amino acids have shorter carbon chains (one carbon) than glutamate. In addition, Cys105 was mutated to aspartate, which mimics Cys-SO2H, to generate the only oxidized form of SAV1875. The crystal structure of each mutant was identified and showed distinct differences in their cysteine oxidation states. In the crystal structure of wild-type and E17D mutant, Cys105 is oxidized to Cys105-SO2H, and Cys-SOH is observed in the E17N mutant. In the crystal structure of the E17D mutant, the hydrogen bond between oxidized Cys105 and Asp17 is retained, showing a distance of 2.89 Å (Figure 3B). The E17N mutant crystal structure showed a disruption of the hydrogen bonds between Cys105 and Asn17 (Figure 3C). However, the oxidized cysteine residues of the SAV1875 mutants still maintain hydrogen bonds with His106 and with the Gly73/His106 amide pocket.

Oxidation states of cysteine at residue 105

Figure 3
Oxidation states of cysteine at residue 105

The electron density map of Cys105 before the introduction of oxygen atoms and interacting hydrogen bonds after the introduction. The 2FoFc electron density map contoured at a level of 1.0σ is shown in grey and the FoFc electron density map contoured at a level of 3.0σ is illustrated in green. Distances are given in Å. Left panels, chain A; middle panels, chain B; right panels, hydrogen bonds. (A) Cys105 in the wild-type is oxidized to Cys105-SO2H. (B) Cys105 in E17D shows Cys105-SO2H (magenta) and the corresponding region in wild-type (light green). (C) Cys105 in E17N shows Cys105-SOH (blue) and the corresponding region in wild-type (light green). (D) Cys105 in over-oxidized SAV1875 is oxidized to Cys105-SO3H, which shows additional density for the sulfonic group in chain A (left panel), but Cys105-SO2H in chain B (middle panel). (E) In C105D, Asp105 mimics Cys105-SO2H, but the geometry of the oxygen atoms is different from that of Cys105-SO2H. However, Glu17 and Asp105 are stabilized by hydrogen bonds with the introduction of water molecules. Figures were created with PyMOL (http://www.pymol.org).

Figure 3
Oxidation states of cysteine at residue 105

The electron density map of Cys105 before the introduction of oxygen atoms and interacting hydrogen bonds after the introduction. The 2FoFc electron density map contoured at a level of 1.0σ is shown in grey and the FoFc electron density map contoured at a level of 3.0σ is illustrated in green. Distances are given in Å. Left panels, chain A; middle panels, chain B; right panels, hydrogen bonds. (A) Cys105 in the wild-type is oxidized to Cys105-SO2H. (B) Cys105 in E17D shows Cys105-SO2H (magenta) and the corresponding region in wild-type (light green). (C) Cys105 in E17N shows Cys105-SOH (blue) and the corresponding region in wild-type (light green). (D) Cys105 in over-oxidized SAV1875 is oxidized to Cys105-SO3H, which shows additional density for the sulfonic group in chain A (left panel), but Cys105-SO2H in chain B (middle panel). (E) In C105D, Asp105 mimics Cys105-SO2H, but the geometry of the oxygen atoms is different from that of Cys105-SO2H. However, Glu17 and Asp105 are stabilized by hydrogen bonds with the introduction of water molecules. Figures were created with PyMOL (http://www.pymol.org).

These results are not consistent with a similar study on the DJ-1 protein [42,45]. To stabilize the oxidation state of Cys106 in the DJ-1 protein, Glu18 and the Gly75/Ala107 amide pocket are required. In the crystal structures, Cys106 of the DJ-1 E18N mutant is oxidized to Cys106-SO2H, and Cys106 of the DJ-1 E18D mutant is modified to Cys106-SOH, which indicates minor oxidation toward the Gly75/Ala107 amide pocket. The opposite behaviour was observed in the SAV1875 mutants and is due to the local structural difference around Cys105. In detail, Gly73 and His106 are shown to form an amide pocket as Gly75/Ala107 in DJ-1. However, in SAV1875, the His106 side chain provides additional strong hydrogen-bonding to oxidized Cys105. Therefore the oxygen atom of Cys105-SOH in the E17N mutant is towards the amide pocket. The occupancy of the oxygen atoms of Cys106-SO2H in the E17D mutant is greater towards the amide pocket because of this hydrogen bond. Moreover, because there is a greater distance between Cys105 and Asn17 of the E17N mutant than Cys105 and Asp17 of the E17D mutant, cysteine oxidation to Cys105-SO2H in E17N SAV1875 remains incomplete (Figures 3B and 3C). Cys105-SO2H was observed as a main peak in both the E17D and E17N mutants in a mass spectrum that were incubated under H2O2 conditions. Therefore we propose that the E17N mutant is oxidation-impaired, but not oxidation-deficient.

Cys105 in wild-type SAV1875 under H2O2 conditions can form Cys105-SO3H in chain A. Each oxygen atom is stabilized by a hydrogen bond (Figure 3D). Because the molecular geometries of aspartate and cysteine sulfinic acid are different, the C105D mutant is not topologically the same as wild-type SAV1875. However, the introduction of water molecules, which form hydrogen bonds with Glu17 and Asp105, yields a stable C105D mutant structure (Figure 3E).

Oxidation propensity of Cys105 in SAV1875

Among all amino acids in a protein, the most oxidation-susceptible residues are the sulfur-containing ones, i.e. cysteine and methionine [4648]. Because recombinant SAV1875 lacks the only methionine residue at position 1 and possesses a single cysteine residue at position 105, the oxidation–reduction state in solution could be conveniently monitored by MS and PAGE [47]. Cysteine undergoes oxidation and reduction reactions and results in four different forms: Cys-SH2, Cys-SOH, Cys-SO2H and Cys-SO3H [49]. The Cys105 in SAV1875 exists as both Cys105-SH2 and Cys105-SO2H in solution. One single intense peak with a molecular mass of 39 kDa in size-exclusion chromatography indicates the presence of the only dimer molecules of SAV1875 (results not shown). SDS/PAGE shows a single band as a monomer with a molecular mass of 19 kDa. In contrast, the mass spectrum contained two peaks with molecular masses of 19565 and 19597 Da, which is +32 Da compared with the first peak. The two main peaks corresponded to SAV1875-Cys105-SH2 and SAV1875-Cys105-SO2H respectively (calculated mass of 19565.7 Da without the methionine residue at position 1). Native PAGE separates SAV1875 into two main bands. Because the m/z ratio is preserved during native PAGE, SAV1875 can migrate towards the positively charged electrode depending on its overall negative surface charge without dissociation of the structure. This result corroborates the MS, which indicated that SAV1875 exists in both reduced (SAV1875-Cys105-SH2) and oxidized (SAV1875-Cys105-SO2H) states. The C105D mutant supports this finding further because the C105D mutant, which lacks the only cysteine residue that can be oxidized (this recombinant C105D mutant lacks the only methionine residue at position 1), has only one band on native PAGE and only one peak in MS (Figure 4A). Furthermore, the following methods were used to reduce oxidized Cys105: (i) 5 mM DTT was added to all buffers during purification and the protein was handled quickly on ice; (ii) 10 mM DTT was introduced to the final buffer (50 mM Tris/HCl, pH 7.5, 200 mM NaCl and 10 mM DTT); and (iii) 1 mM PMSF was employed during all purification steps and ablated by DTT. Despite these efforts to reduce the oxidized cysteine residue, MS and native PAGE indicated that SAV1875 retained the two distinctive main bands, SAV1875-SH2 and SAV1875-SO2H. These data therefore imply that the oxidative modification of Cys105 in SAV1875 is very robust.

Mass spectrometry of SAV1875

Figure 4
Mass spectrometry of SAV1875

MS was used to monitor the oxidation state of SAV1875 in solution. (A) MS analysis of SAV1875. Two major peaks are observed at 19565.0 and 19597.0. The calculated mass of SAV1875 is 19565 Da if the only cysteine residue, Cys105, is reduced (this recombinant SAV1875 lacks the only methionine residue at position 1). In addition to the first peak, there is another main peak at 19597 Da corresponding to two additional oxygen atoms. The right panel corresponds to native PAGE (top) and SDS/PAGE (bottom) analysis. Before exposure to H2O2, there were two separate SAV1875 bands in native PAGE. The oxidized band was located in the same region as the C105D. (B) MS analysis of over-oxidized SAV1875. Two main peaks were observed at 19.596.9 and 19612.9 Da, corresponding to SAV1875-Cys105-SO2H and SAV1875-Cys105-SO3H respectively. The right panel corresponds to native PAGE (top) and SDS/PAGE (bottom). For over-oxidized SAV1875, there were three separate bands in native PAGE. One oxidized band is shown at the same region as the C105D and another oxidized band is located underneath the oxidized band. These native PAGE results coincide with the MS analysis. WT, wild-type.

Figure 4
Mass spectrometry of SAV1875

MS was used to monitor the oxidation state of SAV1875 in solution. (A) MS analysis of SAV1875. Two major peaks are observed at 19565.0 and 19597.0. The calculated mass of SAV1875 is 19565 Da if the only cysteine residue, Cys105, is reduced (this recombinant SAV1875 lacks the only methionine residue at position 1). In addition to the first peak, there is another main peak at 19597 Da corresponding to two additional oxygen atoms. The right panel corresponds to native PAGE (top) and SDS/PAGE (bottom) analysis. Before exposure to H2O2, there were two separate SAV1875 bands in native PAGE. The oxidized band was located in the same region as the C105D. (B) MS analysis of over-oxidized SAV1875. Two main peaks were observed at 19.596.9 and 19612.9 Da, corresponding to SAV1875-Cys105-SO2H and SAV1875-Cys105-SO3H respectively. The right panel corresponds to native PAGE (top) and SDS/PAGE (bottom). For over-oxidized SAV1875, there were three separate bands in native PAGE. One oxidized band is shown at the same region as the C105D and another oxidized band is located underneath the oxidized band. These native PAGE results coincide with the MS analysis. WT, wild-type.

MS and native PAGE confirmed that both SAV1875-Cys105-SH2 and SAV1875-Cys105-SO2H are oxidized further to SAV1875-Cys105-SO3H when the protein was exposed to H2O2 with a molar ratio of SAV1875/H2O2 of 1:50. The MS showed an enlarged peak at a molecular mass of 19612.9 Da (SAV1875-Cys105-SO3H) after exposure to H2O2, which is +48 Da compared with SAV1875-Cys105-SH2 (Figure 4B). This corresponds with the native PAGE result, which has an additional single band below SAV1875-Cys105-SO2H, indicating SAV1875-Cys105-SO3H. CD data showed that the secondary structure of SAV1875 is not altered, even if SAV1875-Cys105-SO2H is oxidized further to the over-oxidized state of SAV1875-Cys105-SO3H (Supplementary Figure S2). The crystal structure of over-oxidized SAV1875 was determined to a resolution of 1.65 Å. The electron density distribution surrounding Cys105 suggested over-oxidation to Cys105-SO3H. The Sγ atom of Cys105 is oxidized to Cys105-SO3H in subunit A and Cys105-SO2H in subunit B (Figure 3D).

Protease activity of SAV1875

SAV1875 has been reported to harbour a PhpI endopeptidase domain, which is found in PhpI from Pyrococcus horikoshii (http://www.uniprot.org/uniprot/P0A0K0) [50]. PhpI is an intracellular protease that belongs to the YhbO-type subfamily, and it shows the highest activity in a multimeric complex [14]. Ton1285 is another YhbO-type protein which has proteolytic activity [18]. PhpI and Ton1285 exist as hexamers in solution, converging their catalytic triad towards the solvent-accessible centre area of the ring-like structure. However, the dimer proteins in the YhbO-type subfamily, recombinant YhbO and DR1199, do not show proteolytic activity towards peptides or casein [16,37]. To determine the proteolytic activity, if any, wild-type, mutant and over-oxidized SAV1875 were used. SAV1875 did not display any significant proteolytic activity towards FITC–casein even after a long incubation time, at a high temperature (more than 37°C) or SAV1875 concentration (up to 50 μM), or in a reducing environment (presence of 1 mM 2-mercaptoethanol). A gelatin-overlay assay did not show any cleared zones. Therefore dimeric SAV1875 is surmised not to have a proteolytic function against casein or gelatin (results not shown). If present, the catalytic activity is very low. We speculate that the existence of a catalytic triad and ring-like multimeric structure signify proteolytic function for YhbO-type proteins. Moreover, further research is needed into factors that may influence the proteolytic function of DJ-1 superfamily proteins.

Chaperone activity of SAV1875

Chaperones are part of a functionally related group of proteins assisting in protein folding under stress conditions or preventing folding defects by binding to unstructured and hydrophobic regions of target proteins. Chaperones in the DJ-1 superfamily, such as Hsp31, work as dimers by forming a canyon and bowl on their dimeric interface [11]. When the surfaces of Hsp31 were viewed, hydrophobic patches were detected around the canyon for the binding of unstructured proteins [51]. DJ-1 and YajL, which are categorized as members of the DJ-1-type subfamily, have also shown chaperone activity. Interestingly, DJ-1 exhibited a greater chaperone activity towards α-synuclein when Cys106 was oxidized to Cys106-SO2H and lost activity when the cysteine residue was oxidized further to Cys106-SO3H, which correlated with a loss of some of its secondary structure [52,53]. SAV1875 is defined as a member of the YhbO-type subfamily, displaying structural differences from the DJ-1-type or Hsp-type subfamily. The YhbO-type proteins with a chaperone function had not yet been identified. However, SAV1875 expresses similar surface patterns of a canyon to that of Hsp31, DJ-1 and YajL. SAV1875 has a canyon that winds from the dimeric interface to each side of the subunit, and hydrophobic patches are detected around the canyon (Figure 1C). These are characteristics of SAV1875 that are distinctive from the YhbO-type proteins. Other YhbO-type proteins are hexamers (PhpI and Ton1285) or dimers (recombinant YhbO and DR1199) without apparent canyon structure. Furthermore, Cys105 in SAV1875 favours oxidation to Cys105-SO2H, similar to a DJ-1 protein. Considering these characteristics, SAV1875 is expected to hold an unfolded protein, and the oxidation state of Cys105 may play a crucial role in chaperone function.

The wild-type, mutant and the over-oxidized SAV1875 were tested for chaperone activity using citrate synthase. The data show a chaperone-facilitated renaturation of citrate synthase in the wild-type, E17D, E17N and C105D mutants. There are no significant differences in the chaperone activity between the wild-type and the SAV1875 mutants. However, negative results were obtained with over-oxidized SAV1875, which was previously exposed to a 1:50 molar ratio of SAV1875/H2O2. We therefore determined that SAV1875 functions as a molecular chaperone with oxidized Cys105, but it loses chaperone activity when this cysteine is oxidized further to Cys105-SO3H, even though over-oxidized SAV1875 is structurally stable (Figure 5). CD revealed that the secondary structure is well conserved (Supplementary Figure S2). The native PAGE result showed that the surface charge is preserved in over-oxidized SAV1875 with Cys105-SO3H (Figure 4) [54]. From this study, we identified SAV1875 as a novel chaperone protein in the YhbO-type subfamily and that the canyon surface structure along with cysteine redox state are the key elements for the chaperone function.

Chaperone activity of SAV1875 depending on cysteine oxidation state

Figure 5
Chaperone activity of SAV1875 depending on cysteine oxidation state

SAV1875 exerts a chaperone function. The chaperone activity was assayed by monitoring an increase in absorbance at 412 nm. (A) The control (■) contained 50 μl of the reaction mixture only, which was 1 mM DTNB, 0.2 mM MnCl2, 0.4 mM oxaloacetic acid, 0.3 mM acetyl-CoA and 100 mM Tris/HCl (pH 8.0). The control with citrate synthase (□) contained an additional 0.75 μg of denatured citrate synthase. A final concentration of 5 μM was used for the wild-type (●), E17N (○). E17D (▲), C105D (Δ) and over-oxidized SAV1875 (♦) and was added to a citrate synthase reaction mixture. (B) The calculation of chaperone activity at the 70 min time point. The data from three scans were averaged. WT, wild-type.

Figure 5
Chaperone activity of SAV1875 depending on cysteine oxidation state

SAV1875 exerts a chaperone function. The chaperone activity was assayed by monitoring an increase in absorbance at 412 nm. (A) The control (■) contained 50 μl of the reaction mixture only, which was 1 mM DTNB, 0.2 mM MnCl2, 0.4 mM oxaloacetic acid, 0.3 mM acetyl-CoA and 100 mM Tris/HCl (pH 8.0). The control with citrate synthase (□) contained an additional 0.75 μg of denatured citrate synthase. A final concentration of 5 μM was used for the wild-type (●), E17N (○). E17D (▲), C105D (Δ) and over-oxidized SAV1875 (♦) and was added to a citrate synthase reaction mixture. (B) The calculation of chaperone activity at the 70 min time point. The data from three scans were averaged. WT, wild-type.

DISCUSSION

Our structural studies confirmed SAV1875 as a member of the DJ-1 superfamily because it had all the pertinent traits of this group, including the characteristic sandwich fold and nucleophilic arm. It is specifically categorized in the YhbO-type subfamily because of its aspects of dimeric interface and catalytic triad containing a reactive cysteine residue. In our crystal structure, Cys105 was oxidized to Cys105-SO2H. This modification is common in DJ-1 superfamily proteins, such as DR1199, YDR533Cp and YhbO, even though it was not intentionally produced. As is customary among members of this family, cysteine exhibited the highest specialization. Therefore the oxidation status of cysteine and the corresponding structural differences may account for the function of SAV1875. The oxidized Cys105 accepts hydrogen bonds from the Gly73/His106 amide pocket, the His106 side chain and the protonated Oε atom of Glu17. The residues that are structurally equivalent to Cys105 and Glu17 in SAV1875 are highly conserved in other members of the DJ-1 superfamily, indicating that cysteine and glutamate may play a crucial role in the function of the protein. The influence of Glu17 on the oxidation of Cys105 was verified with the crystal structures of the E17D, E17N and C105D mutants. In the crystal structures, the hydrogen bond between Cys105 and the residue at position 17 was weakened or eliminated in the glutamate mutants (E17D and E17N). C105D, which mimics cysteine sulfinic acid, had a different configuration of oxygen atoms (Oε) compared with cysteine sulfinic acid, but the hydrogen bonds were maintained due to the introduction of water molecules. Moreover, the cysteine residue can be oxidized further to Cys105-SO3H using excess H2O2. The crystal structure of over-oxidized SAV1875 revealed stable formation of Cys105-SO3H.

Because the DJ-1 superfamily includes proteins spanning many functions, such as proteases, chaperones and general stress-response proteins, SAV1875 is also predicted to perform similar biological functions. Thus several functional tests were conducted to determine the function of SAV1875 and the relationship between the oxidation states of cysteine and the differences in protein activity. PhpI, a homologue of SAV1875 in the YhbO-type subfamily, has been identified as a protease [14]. Another YhbO-type protein, Ton1285, also has a proteolytic role [18]. From a structural standpoint, SAV1875 has a sandwich fold and catalytic triad, but has different oligomerization states. Unlike dimeric SAV1875, PhpI and Ton1285 exist as hexamers in solution. Proteolytic activity was not detected in SAV1875. This analysis indicates that other factors along with oligomeric state are biologically relevant and still require precise characterization.

Hsp31 displays chaperone activity by holding unfolded protein with its canyon and hydrophobic surface. DJ-1 and YajL have shown chaperone activity as well, showing similar surface structure. Interestingly, when Cys106 of DJ-1 was oxidized to Cys106-SO2H, chaperone activity increased. When the DJ-1 with Cys106-SO2H was oxidized further, it lost its chaperone activity owing to structural perturbation [53]. SAV1875 has a similar canyon-shaped surface and oxidized cysteine residue, suggesting that SAV1875 may work as a chaperone. We discovered that the wild-type and SAV1875 mutants assisted in the folding of citrate synthase but over-oxidized SAV1875 did not, even though its structure was maintained. Further detailed studies on diverse substrates and in vivo tests are needed to verify the exact mechanism of the SAV1875 chaperone activity.

In summary, the crystal structures of YhbO-type protein SAV1875, its mutants and over-oxidized SAV1875 were determined. Although the overall sandwich fold and dimerization mode are similar to YhbO-type proteins, SAV1875 has distinctive characteristics that include its oligomeric state and surface canyon structure. On the basis of these similarities and differences, we predicted the possible functions of SAV1875 and determined that it functions as a chaperone. The characterization of SAV1875 led to the discovery of a new putative chaperone protein in the YhbO-type subfamily. Further research into the DJ-1 superfamily proteins and in vivo studies would advance our understanding of the fundamentals of cysteine oxidation and functional events.

AUTHOR CONTRIBUTION

Hyo Jung Kim, Ae-Ran Kwon, and Bong-Jin Lee designed and carried out experiments, analysed data and wrote this manuscript.

We thank the beamline staff at Pohang Light Source (beamlines 5C and 6C) and Photon Factory (BL-17A) for assistance during X-ray experiments.

FUNDING

This study was supported by the Korea Healthcare Technology R&D Project, Ministry for Health Welfare, Republic of Korea [grant number A092006] and by the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (MEST) of the Korean government [grant numbers 2007-0056817 and 2012-R1A2A1A01003569]. This study was also supported by the 2014 BK21 Plus Project for Medicine, Dentistry, and Pharmacy and by the Pohang Accelerator Laboratory (PAL) through the abroad beamtime programme of the Synchrotron Radiation Facility Project under MEST and has been performed under the approval of the Photon Factory Program Advisory Committee (proposal number 2011G237).

Abbreviations

     
  • DTNB

    5,5′-dithiobis-(2-nitrobenzoic acid)

  •  
  • Hsp

    heat-shock protein

  •  
  • MIC

    minimum inhibitory concentration

  •  
  • PPG

    polypropylene glycol

  •  
  • TNB

    5-thio-2-nitrobenzoic acid

  •  
  • VISA

    vancomycin-intermediate Staphylococcus aureus

  •  
  • VRSA

    vancomycin-resistant Staphylococcus aureus

References

References
1
Archer
G.L.
Staphylococcus aureus: a well-armed pathogen
Clin. Infect. Dis.
1998
, vol. 
26
 (pg. 
1179
-
1181
)
[PubMed]
2
Klevens
R.M.
Morrison
M.A.
Nadle
J.
Petit
S.
Gershman
K.
Ray
S.
Harrison
L.H.
Lynfield
R.
Dumyati
G.
Townes
J.M.
, et al. 
Invasive methicillin-resistant Staphylococcus aureus infections in the United States
JAMA
2007
, vol. 
298
 (pg. 
1763
-
1771
)
[PubMed]
3
Hiramatsu
K.
Vancomycin-resistant Staphylococcus aureus: a new model of antibiotic resistance
Lancet Infect. Dis.
2001
, vol. 
1
 (pg. 
147
-
155
)
[PubMed]
4
Tenover
F.C.
Biddle
J.W.
Lancaster
M.V.
Increasing resistance to vancomycin and other glycopeptides in Staphylococcus aureus
Emerg. Infect. Dis.
2001
, vol. 
7
 (pg. 
327
-
332
)
[PubMed]
5
Sieradzki
K.
Tomasz
A.
A highly vancomycin-resistant laboratory mutant of Staphylococcus aureus
FEMS Microbiol. Lett.
1996
, vol. 
142
 (pg. 
161
-
166
)
[PubMed]
6
Hiramatsu
K.
The emergence of Staphylococcus aureus with reduced susceptibility to vancomycin in Japan
Am. J. Med.
1998
, vol. 
104
 (pg. 
7S
-
10S
)
[PubMed]
7
Hanaki
H.
Labischinski
H.
Sasaki
K.
Kuwahara-Arai
K.
Inaba
Y.
Hiramatsu
K.
[Mechanism of vancomycin resistance in MRSA strain Mu50]
Jpn. J. Antibiot.
1998
, vol. 
51
 (pg. 
237
-
247
)
[PubMed]
8
Turco
T.F.
Melko
G.P.
Williams
J.R.
Vancomycin intermediate-resistant Staphylococcus aureus
Ann. Pharmacother.
1998
, vol. 
32
 (pg. 
758
-
760
)
[PubMed]
9
Woods
C.
Colice
G.
Methicillin-resistant Staphylococcus aureus pneumonia in adults
Expert Rev. Respir. Med.
2014
, vol. 
8
 (pg. 
641
-
651
)
[PubMed]
10
Honbou
K.
Suzuki
N.N.
Horiuchi
M.
Niki
T.
Taira
T.
Ariga
H.
Inagaki
F.
The crystal structure of DJ-1, a protein related to male fertility and Parkinson's disease
J. Biol. Chem.
2003
, vol. 
278
 (pg. 
31380
-
31384
)
[PubMed]
11
Quigley
P.M.
Korotkov
K.
Baneyx
F.
Hol
W.G.
The 1.6-Å crystal structure of the class of chaperones represented by Escherichia coli Hsp31 reveals a putative catalytic triad
Proc. Natl. Acad. Sci. U.S.A.
2003
, vol. 
100
 (pg. 
3137
-
3142
)
[PubMed]
12
Subedi
K.P.
Choi
D.
Kim
I.
Min
B.
Park
C.
Hsp31 of Escherichia coli K-12 is glyoxalase III
Mol. Microbiol.
2011
, vol. 
81
 (pg. 
926
-
936
)
[PubMed]
13
Mujacic
M.
Baneyx
F.
Chaperone Hsp31 contributes to acid resistance in stationary-phase Escherichia coli
Appl. Environ. Microbiol.
2007
, vol. 
73
 (pg. 
1014
-
1018
)
[PubMed]
14
Du
X.
Choi
I.G.
Kim
R.
Wang
W.
Jancarik
J.
Yokota
H.
Kim
S.H.
Crystal structure of an intracellular protease from Pyrococcus horikoshii at 2-Å resolution
Proc. Natl. Acad. Sci. U.S.A.
2000
, vol. 
97
 (pg. 
14079
-
14084
)
[PubMed]
15
Rawlings
N.D.
Morton
F.R.
Barrett
A.J.
MEROPS: the peptidase database
Nucleic Acids Res.
2006
, vol. 
34
 (pg. 
D270
-
D272
)
[PubMed]
16
Fioravanti
E.
Dura
M.A.
Lascoux
D.
Micossi
E.
Franzetti
B.
McSweeney
S.
Structure of the stress response protein DR1199 from Deinococcus radiodurans: a member of the DJ-1 superfamily
Biochemistry
2008
, vol. 
47
 (pg. 
11581
-
11589
)
[PubMed]
17
Abdallah
J.
Caldas
T.
Kthiri
F.
Kern
R.
Richarme
G.
YhbO protects cells against multiple stresses
J. Bacteriol.
2007
, vol. 
189
 (pg. 
9140
-
9144
)
[PubMed]
18
Jung
H.J.
Kim
S.
Kim
Y.J.
Kim
M.K.
Kang
S.G.
Lee
J.H.
Kim
W.
Cha
S.S.
Dissection of the dimerization modes in the DJ-1 superfamily
Mol. Cells
2012
, vol. 
33
 (pg. 
163
-
171
)
[PubMed]
19
Wei
Y.
Ringe
D.
Wilson
M.A.
Ondrechen
M.J.
Identification of functional subclasses in the DJ-1 superfamily proteins
PLoS Comput. Biol.
2007
, vol. 
3
 pg. 
e10
 
[PubMed]
20
Herrera
F.E.
Zucchelli
S.
Jezierska
A.
Lavina
Z.S.
Gustincich
S.
Carloni
P.
On the oligomeric state of DJ-1 protein and its mutants associated with Parkinson disease: a combined computational and in vitro study
J. Biol. Chem.
2007
, vol. 
282
 (pg. 
24905
-
24914
)
[PubMed]
21
Ito
G.
Ariga
H.
Nakagawa
Y.
Iwatsubo
T.
Roles of distinct cysteine residues in S-nitrosylation and dimerization of DJ-1
Biochem. Biophys. Res. Commun.
2006
, vol. 
339
 (pg. 
667
-
672
)
[PubMed]
22
Collaborative Computational Project, Number 4
The CCP4 suite: programs for protein crystallography
Acta Crystallogr. D Biol. Crystallogr.
1994
, vol. 
50
 (pg. 
760
-
763
)
[PubMed]
23
Dodson
E.J.
Winn
M.
Ralph
A.
Collaborative Computational Project, Number 4: providing programs for protein crystallography
Methods Enzymol.
1997
, vol. 
277
 (pg. 
620
-
633
)
[PubMed]
24
Vagin
A.
Teplyakov
A.
Molecular replacement with MOLREP
Acta Crystallogr. D Biol. Crystallogr.
2010
, vol. 
66
 (pg. 
22
-
25
)
[PubMed]
25
Emsley
P.
Cowtan
K.
Coot: model-building tools for molecular graphics
Acta. Crystallogr. D Biol. Crystallogr.
2004
, vol. 
60
 (pg. 
2126
-
2132
)
[PubMed]
26
Murshudov
G.N.
Skubak
P.
Lebedev
A.A.
Pannu
N.S.
Steiner
R.A.
Nicholls
R.A.
Winn
M.D.
Long
F.
Vagin
A.A.
REFMAC5 for the refinement of macromolecular crystal structures
Acta. Crystallogr. D Biol. Crystallogr.
2011
, vol. 
67
 (pg. 
355
-
367
)
[PubMed]
27
Brunger
A.T.
Free R value: a novel statistical quantity for assessing the accuracy of crystal structures
Nature
1992
, vol. 
355
 (pg. 
472
-
475
)
[PubMed]
28
Laskowski
R.A.
Rullmannn
J.A.
MacArthur
M.W.
Kaptein
R.
Thornton
J.M.
AQUA and PROCHECK-NMR: programs for checking the quality of protein structures solved by NMR
J. Biomol. NMR
1996
, vol. 
8
 (pg. 
477
-
486
)
[PubMed]
29
Pettersen
E.F.
Goddard
T.D.
Huang
C.C.
Couch
G.S.
Greenblatt
D.M.
Meng
E.C.
Ferrin
T.E.
UCSF Chimera: a visualization system for exploratory research and analysis
J. Comput. Chem.
2004
, vol. 
25
 (pg. 
1605
-
1612
)
[PubMed]
30
Krissinel
E.
Henrick
K.
Inference of macromolecular assemblies from crystalline state
J. Mol. Biol.
2007
, vol. 
372
 (pg. 
774
-
797
)
[PubMed]
31
Laemmli
U.K.
Cleavage of structural proteins during the assembly of the head of bacteriophage T4
Nature
1970
, vol. 
227
 (pg. 
680
-
685
)
[PubMed]
32
Twining
S.S.
Fluorescein isothiocyanate-labeled casein assay for proteolytic enzymes
Anal. Biochem.
1984
, vol. 
143
 (pg. 
30
-
34
)
[PubMed]
33
Vandooren
J.
Geurts
N.
Martens
E.
Van den Steen
P.E.
Opdenakker
G.
Zymography methods for visualizing hydrolytic enzymes
Nat. Methods
2013
, vol. 
10
 (pg. 
211
-
220
)
[PubMed]
34
Lee
G.J.
Assaying proteins for molecular chaperone activity
Methods Cell Biol.
1995
, vol. 
50
 (pg. 
325
-
334
)
[PubMed]
35
Zhi
W.
Landry
S.J.
Gierasch
L.M.
Srere
P.A.
Renaturation of citrate synthase: influence of denaturant and folding assistants
Protein Sci.
1992
, vol. 
1
 (pg. 
522
-
529
)
[PubMed]
36
Morgunov
I.
Srere
P.A.
Interaction between citrate synthase and malate dehydrogenase: substrate channeling of oxaloacetate
J. Biol. Chem.
1998
, vol. 
273
 (pg. 
29540
-
29544
)
[PubMed]
37
Abdallah
J.
Kern
R.
Malki
A.
Eckey
V.
Richarme
G.
Cloning, expression, and purification of the general stress protein YhbO from Escherichia coli
Protein Expr. Purif.
2006
, vol. 
47
 (pg. 
455
-
460
)
[PubMed]
38
Larkin
M.A.
Blackshields
G.
Brown
N.P.
Chenna
R.
McGettigan
P.A.
McWilliam
H.
Valentin
F.
Wallace
I.M.
Wilm
A.
Lopez
R.
, et al. 
Clustal W and Clustal X version 2.0
Bioinformatics
2007
, vol. 
23
 (pg. 
2947
-
2948
)
[PubMed]
39
Robert
X.
Gouet
P.
Deciphering key features in protein structures with the new ENDscript server
Nucleic Acids Res.
2014
, vol. 
42
 (pg. 
W320
-
W324
)
[PubMed]
40
Gouet
P.
Courcelle
E.
Stuart
D.I.
Metoz
F.
ESPript: analysis of multiple sequence alignments in PostScript
Bioinformatics
1999
, vol. 
15
 (pg. 
305
-
308
)
[PubMed]
41
Wilson
M.A.
St Amour
C.V.
Collins
J.L.
Ringe
D.
Petsko
G.A.
The 1.8-Å resolution crystal structure of YDR533Cp from Saccharomyces cerevisiae: a member of the DJ-1/ThiJ/PfpI superfamily
Proc. Natl. Acad. Sci. U.S.A.
2004
, vol. 
101
 (pg. 
1531
-
1536
)
[PubMed]
42
Blackinton
J.
Lakshminarasimhan
M.
Thomas
K.J.
Ahmad
R.
Greggio
E.
Raza
A.S.
Cookson
M.R.
Wilson
M.A.
Formation of a stabilized cysteine sulfinic acid is critical for the mitochondrial function of the parkinsonism protein DJ-1
J. Biol. Chem.
2009
, vol. 
284
 (pg. 
6476
-
6485
)
[PubMed]
43
Wilson
M.A.
Ringe
D.
Petsko
G.A.
The atomic resolution crystal structure of the YajL (ThiJ) protein from Escherichia coli: a close prokaryotic homologue of the Parkinsonism-associated protein DJ-1
J. Mol. Biol.
2005
, vol. 
353
 (pg. 
678
-
691
)
[PubMed]
44
Bandyopadhyay
S.
Cookson
M.R.
Evolutionary and functional relationships within the DJ-1 superfamily
BMC Evol. Biol.
2004
, vol. 
4
 pg. 
6
 
[PubMed]
45
Witt
A.C.
Lakshminarasimhan
M.
Remington
B.C.
Hasim
S.
Pozharski
E.
Wilson
M.A.
Cysteine pKa depression by a protonated glutamic acid in human DJ-1
Biochemistry
2008
, vol. 
47
 (pg. 
7430
-
7440
)
[PubMed]
46
Rinalducci
S.
Murgiano
L.
Zolla
L.
Redox proteomics: basic principles and future perspectives for the detection of protein oxidation in plants
J. Exp. Bot.
2008
, vol. 
59
 (pg. 
3781
-
3801
)
[PubMed]
47
Roeser
J.
Bischoff
R.
Bruins
A.P.
Permentier
H.P.
Oxidative protein labeling in mass-spectrometry-based proteomics
Anal. Bioanal. Chem.
2010
, vol. 
397
 (pg. 
3441
-
3455
)
[PubMed]
48
Stadtman
E.R.
Levine
R.L.
Free radical-mediated oxidation of free amino acids and amino acid residues in proteins
Amino Acids
2003
, vol. 
25
 (pg. 
207
-
218
)
[PubMed]
49
Wani
R.
Nagata
A.
Murray
B.W.
Protein redox chemistry: post-translational cysteine modifications that regulate signal transduction and drug pharmacology
Front. Pharmacol.
2014
, vol. 
5
 pg. 
224
 
[PubMed]
50
Kuroda
M.
Ohta
T.
Uchiyama
I.
Baba
T.
Yuzawa
H.
Kobayashi
I.
Cui
L.
Oguchi
A.
Aoki
K.
Nagai
Y.
, et al. 
Whole genome sequencing of methicillin-resistant Staphylococcus aureus
Lancet
2001
, vol. 
357
 (pg. 
1225
-
1240
)
[PubMed]
51
Sastry
M.S.
Quigley
P.M.
Hol
W.G.
Baneyx
F.
The linker–loop region of Escherichia coli chaperone Hsp31 functions as a gate that modulates high-affinity substrate binding at elevated temperatures
Proc. Natl. Acad. Sci. U.S.A.
2004
, vol. 
101
 (pg. 
8587
-
8592
)
[PubMed]
52
Shendelman
S.
Jonason
A.
Martinat
C.
Leete
T.
Abeliovich
A.
DJ-1 is a redox-dependent molecular chaperone that inhibits alpha-synuclein aggregate formation
PLoS Biol.
2004
, vol. 
2
 pg. 
e362
 
[PubMed]
53
Zhou
W.
Zhu
M.
Wilson
M.A.
Petsko
G.A.
Fink
A.L.
The oxidation state of DJ-1 regulates its chaperone activity toward alpha-synuclein
J. Mol. Biol.
2006
, vol. 
356
 (pg. 
1036
-
1048
)
[PubMed]
54
Greenfield
N.J.
Using circular dichroism spectra to estimate protein secondary structure
Nat. Protoc.
2006
, vol. 
1
 (pg. 
2876
-
2890
)
[PubMed]

Supplementary data