A novel bacterial class V dye-decolourizing peroxidase from the extremophile Deinococcus radiodurans: cloning, expression optimization, purification, crystallization, initial characterization and X-ray diffraction analysis

The gene encoding a novel bacterial class V dye-decolourizing peroxidase (DyP) from the extremely radiation-resistant and desiccation-resistant bacterium Deinococcus radiodurans was cloned and optimum conditions for high-yield expression were identified. The optimized crystals diffracted to 2.2 Å resolution and the initial characterization and crystallographic analysis is promising with respect to providing highly desired insights into this largely unknown family of DyPs.

Abstract

Deinococcus radiodurans is a bacterium with extreme resistance to desiccation and radiation. The resistance mechanism is unknown, but an efficient reactive oxygen species (ROS) scavenging system and DNA-repair and DNA-protection mechanisms are believed to play important roles. Here, the cloning and small- and medium-scale expression tests of a novel dye-decolourizing peroxidase from D. radiodurans (DrDyP) using three different Escherichia coli strains and three different temperatures in order to identify the optimum conditions for the expression of recombinant DrDyP are presented. The best expression conditions were used for large-scale expression and yielded ∼10 mg recombinant DrDyP per litre of culture after purification. Initial characterization experiments demonstrated unusual features with regard to the haem spin state, which motivated the crystallization experiment. The obtained crystals were used for data collection and diffracted to 2.2 Å resolution. The crystals belonged to the trigonal space group P3₁ or P3₂, with unit-cell parameters a = b = 64.13, c = 111.32 Å, and are predicted to contain one DrDyP molecule per asymmetric unit. Structure determination by molecular replacement using previously determined structures of dye-decolourizing peroxidases with ∼30% sequence identity at ∼2 Å resolution as templates are ongoing.

1. Introduction  

Deinococcus radiodurans is a bacterium that has a remarkable capacity to overcome oxidative stress (Battista, 1997 ▸). Oxidative stress is caused by reactive oxygen species (ROS), which can damage cellular macromolecules (Slade & Radman, 2011 ▸). When oxidative modifications occur in humans, they can induce cancer, neurodegenerative diseases and premature aging (Birben et al., 2012 ▸; Cui et al., 2012 ▸). Given that D. radiodurans displays an outstanding resistance to oxidative stress and that reactive oxygen species are associated with these disorders, a comprehensive study of the strategies used by D. radiodurans to combat oxidative stress may open novel opportunities for anti-aging and anticancer treatments and for the treatment of degenerative disorders (Gerber et al., 2015 ▸; Slade & Radman, 2011 ▸).

To date, the mechanism behind the resistance of D. radiodurans to oxidative stress remains unclear, but it is believed to be caused by the combination of an efficient DNA-repair process and an antioxidant defence system, which enables proteins to retain catalytic activity and provides a swift response under stress conditions (Slade & Radman, 2011 ▸). The genome sequence of D. radiodurans shows that it possesses an expanded repertoire of genes encoding cellular antioxidant protection proteins (Makarova et al., 2001 ▸). It has three catalases (the katE catalases DR1998 and DRA0259, and the eukaryotic-type catalase DRA0146), four superoxide dismutases (the manganese-dependent superoxide dismutase DR1279, and three copper/zinc-dependent superoxide dismutases DR1546, DRA0202 and DR0644) and two peroxidases (the cytochrome c peroxidase DRA0301 and an iron-dependent peroxidase, the putative dye-decolourizing peroxidase DRA0145). It also possesses two DNA-binding proteins from starved cells (Dps) proteins Dps1 (DR2263) and Dps2 (DRB0092) (Romão et al., 2006 ▸; Cuypers et al., 2007 ▸).

Peroxidases are oxidoreductases that catalyze the oxidation of substrate molecules using hydrogen peroxide as an electron acceptor (Smulevich et al., 2010 ▸). The vast majority of peroxidases, although phylogenetically unrelated, contain a haem cofactor. They are present in both the eukaryotic and prokaryotic kingdoms and are classified into two superfamilies: the animal peroxidase and plant peroxidase superfamilies (Rahmanpour & Bugg, 2015 ▸). The DyP-type peroxidases are a newly identified superfamily of peroxidases which are unrelated to general peroxidases (Sugano et al., 2007 ▸), showing structural divergence and low sequence similarity with disparity in the haem pocket (which possesses a distal aspartic acid instead of histidine). DyP-type peroxidases also show different reaction characteristics compared with other peroxidases, such as low optimal pH values for activity and high affinity for structurally different substrates, including synthetic dyes (for example anthraquinone and azo dyes), β-carotene, aromatic sulfides, phenolic or nonphenolic lignin-compound units and metal ions (for example Mn²⁺) (Salvachúa et al., 2013 ▸; Sezer et al., 2013 ▸).

Initially, the DyP-type peroxidases were classified into four subfamilies, A, B, C and D, based on primary sequence. The bacterial enzymes were grouped into subfamilies A, B and C, and the fungal enzymes into subfamily D (Rahmanpour & Bugg, 2015 ▸). Subsequently, a reclassification into the classes P (primitive), I (intermediate) and V (advanced) was proposed based on tertiary structure (Yoshida & Sugano, 2015 ▸). Here, class P consists of the former class B, class I of the former class A and class V of the previous classes C and D.

Here, we present the expression optimization, purification, crystallization and initial characterization and X-ray diffraction analysis of a putative dye-decolourizing peroxidase (DRA0145) from D. radiodurans (DrDyP). Based on the initial classification and the PeroxiBase (Fawal et al., 2013 ▸), DrDyP belongs to the largely unexplored class C DyPs, which are organized within class V of the new categorization. Here, we will use the latter nomenclature when describing DrDyP. Our initial characterization experiments indicate unusual features of DrDyP arising from the presence of a low-spin haem state, suggesting that it lacks peroxidase activity at pH 7.5, which has not been observed previously for other DyPs. The structure determination of DrDyP could provide valuable information to explain these properties and provide further insight into this class of DyPs and their potential role in the mechanism of oxidative-stress resistance by D. radiodurans.

2. Materials and methods  

2.1. Macromolecule production  

The gene encoding a three-amino-acid N-terminally truncated version (Δ3) of DR_A0145 (accession code Q9RZ08), hereafter called DrDyP, from D. radiodurans was cloned into the pDEST14 expression vector from the Gateway system (GE Healthcare), as previously performed for endonuclease III (Sarre et al., 2014 ▸), using the primers described in Table 1 ▸. The primers contain nucleotides encoding an N-terminal His tag followed by a TEV cleavage site upstream of the gene encoding the protein. Insertion of the gene into the correct reading frame of the pDEST14 vector was confirmed using the GATC Biotech Sanger sequencing service (GATC Biotech).

Table 1. Macromolecule production.

In the primers, the nucleotides encoding the His tag are underlined, the TEV site nucleotides are in italics and the attB1 and attB2 sites (for recombination into the cloning vector) are in bold. The lowercase letters indicate the gene-encoding nucleotides.

Source organism	D. radiodurans
DNA source	D. radiodurans
Forward primer 1 (5′–3′)	CATCACCATCACCATCAC GAAAACCTGTATTTCCAGGGAGCAatgacgctgttcaaaaagctg
Reverse primer (5′–3′)	GGGGACCACTTTGTACAAGAAAGCTGGGTCttacttcagttcactcagcca
Forward primer 2 (5′–3′)	GGGGACAAGTTTGTACAAAAAAGCAGGCTTCGAAGGAGATAGAACCATG CATCACCATCACCATCAC
Cloning vector	pDONR221
Expression vector	pDEST14
Expression host	E. coli BL21(DE3)*, BL21(DE3)pLysS and BL21(DE3)*pRARE
Complete amino-acid sequence of the construct produced after TEV cleavage	GAMTLFKKLRELVHHNDKIDLDLDDIQATVLRERPEPYYGTHAMVRFDTAEGGRELLKLLPHIASAEKWWDVKYAWTAAAISYEGLKKLGVPQDSLDSFPESFKVGMAGRAEHLFDVGENDPKHWEKPFGTGQVHLALTIFAENEENWQKALVIAEHELEATKGVTLLMREDFGAQPDSRNSLGYKDMISNPAIEGSGIKPFPGQGPAIKPGEFVLGYPGEAGVPLGMPKPEVLGKNGTFVALRKYHTNAGSFNRYLKENAEYTGGDAELLAAKLVGRWRSGAPLTLAPKEDDPELGHDPNRNNDFTYKNDPEGLEVPLGSHIRRMNPRDTKLELLTDVNIHRIIRRATAYGPAYDPKADSLAEDKVERGLYFIFISAKAMDTTEFLQKEWINKANFIGQGSERDPIVGLQDEDLTFTLPKEPVRQRLRGMDTFNVLRGGEYLFMPSLSALKWLSELK
Theoretical molecular mass (kDa)	51.2

The expression plasmid containing the gene encoding DrDyP was transformed into three different Escherichia coli expression strains, BL21(DE3)*, BL21(DE3)pLysS and BL21(DE3)*pRARE, by heat shock. One transformant of each strain was used to inoculate 5 ml LB medium with ampicillin (200 µg ml⁻¹) and, in the case of pLysS, chloram­phenicol (34 µg ml⁻¹), and was grown at 37°C until an OD₆₀₀ of ∼0.6 was attained. At this point gene expression was induced by the addition of 0.5 mM IPTG. After 3 h of induction, the cells were harvested and dissolved in 200 µl extraction buffer (50 mM Tris–HCl pH 7.5, 150 mM NaCl, 5 mM MgCl₂, 0.1 mg ml⁻¹ lysozyme, 1 µg ml⁻¹ DNase), followed by three freeze–thaw cycles in liquid nitrogen and a water bath at 22°C. The resulting extract was centrifuged at 13 200 rev min⁻¹ in a tabletop centrifuge at 4°C and the supernatant (soluble fraction) was collected for SDS–PAGE analysis. The resulting pellet was dissolved in 50 µl 1% SDS and heated at 95°C for 5 min followed by centrifugation (insoluble fraction). The supernatant (15 µl) and the dissolved pellet (5 µl) were then loaded onto an SDS–PAGE gel for analysis after adding 5 µl SDS–PAGE sample-loading buffer [0.5 M Tris–HCl pH 6.8, 0.5%(w/v) bromophenol blue, 10%(w/v) SDS, 50% glycerol, 50 mM β-mercaptoethanol] to each and were heated to 95°C for 5 min. A non-induced expression culture of E. coli BL21(DE3)* cells was also included as a negative control, and samples from these cultures were treated the same as the cultures with induced expression.

One of the strains that overexpressed DrDyP was BL21(DE3)*, and this strain was used in a temperature-optimization expression test in order to determine the temperature at which the strain expressed the most protein. Three 1 ml pre-cultures of the strain were used to inoculate three 50 ml cultures of LB medium with 200 µg ml⁻¹ ampicillin at 18, 25 and 37°C. The cells were grown to an OD₆₀₀ of ∼0.6 and expression was induced by the addition of 0.5 mM IPTG as above. The length of induction varied depending on the growth temperature and lasted for 18, 6 and 3 h at 18, 25 and 37°C, respectively. After induction, the cells were harvested and the pellet was dissolved in 1 ml extraction buffer. Next, the samples were subjected to three freeze–thaw cycles as described for the small-scale expression test. The resulting extract was centrifuged at 13 200 rev min⁻¹ in a tabletop centrifuge at 4°C and the supernatant (soluble fraction) was collected for SDS–PAGE analysis. The resulting pellet was dissolved in 250 µl 1% SDS and heated at 95°C for 5 min followed by centrifugation (insoluble fraction). The supernatant (15 µl) and the dissolved pellet (5 µl) were then loaded onto an SDS–PAGE gel for analysis after adding 5 µl SDS–PAGE sample-loading buffer each and heated to 95°C for 5 min. A fourth culture with non-induced expression was also grown at 37°C and acted as a negative control.

Finally, DrDyP was expressed on a large scale in 1 l LB medium with 200 µg ml⁻¹ ampicillin in E. coli BL21(DE3)* cells at 37°C. The cells were harvested after 3 h of induction by the addition of 0.5 mM IPTG. The pellet was dissolved in 20 ml extraction buffer followed by four freeze–thaw cycles in liquid nitrogen and a water bath at 22°C. The resulting extract was centrifuged at 18 000 rev min⁻¹ at 4°C and the soluble fractions were applied onto a 1 ml HisTrap column (GE Healthcare) equilibrated with buffer A (50 mM Tris–HCl pH 7.5, 150 mM NaCl). The resin was washed with ten column volumes (CV) of buffer A and 10 CV of 5% buffer B (50 mM Tris–HCl pH 7.5, 150 mM NaCl, 500 mM imidazole). The protein was eluted with a gradient from 5 to 100% buffer B over 20 CV. The red/brown-coloured fractions containing DrDyP were pooled and dialyzed overnight (4°C) into buffer A to remove imidazole. His-tagged TEV protease (Kapust et al., 2001 ▸) was added to the sample in a 1:20 ratio before dialysis in order to facilitate cleavage of the His tag. After dialysis, the protein solution was applied onto a second 1 ml HisTap column in order to recover the protein without the His tag from the flowthrough. The pure red-coloured protein, with a molecular mass of 51.2 kDa after removal of the His tag, was concentrated to 9.1 mg ml⁻¹ and stored at 4°C. The purity was analysed by SDS–PAGE and was estimated to be greater than 95%.

The iron content in the sample was determined by inductively coupled plasma atomic emission spectrometry (ICP-AES) using an Ultima spectrometer from Horiba Jobin-Yvon.

Electronic absorption measurements were recorded at 18°C using a Shimadzu UV-1800 spectrophotometer. The protein concentration was ∼4 µM in 40 mM Britton–Robinson buffer solutions pH 3.0, 4.0, 5.0 and 7.5.

2.2. Crystallization  

The purified protein (9.1 mg ml⁻¹) in 50 mM Tris–HCl pH 7.5, 150 mM NaCl was used in sitting-drop crystallization experiments on a Honeybee Cartesian (Genomic Systems) crystallization robot with MDL3 plates (100 nl protein solution and 40 µl reservoir solution) and the JCSG+ crystallization screen (Molecular Dimensions). Small and fragile red crystals appeared in condition A7 with a reservoir solution consisting of 0.1 M CHES pH 9.5, 20% PEG 8000. With the purpose of improving the crystallization condition, hanging-drop optimization experiments were performed manually by varying the concentration of PEG 8000 from 18 to 23%(w/v) and the pH between 8.5 and 10. The best-looking crystals from the optimization experiment appeared in 18% PEG 8000, 0.1 M CHES pH 8.5. The crystallization conditions for the crystals used for diffraction data collection are described in Table 2 ▸.

Table 2. Crystallization conditions.

Method	Robot	Manual (hanging drop)
Plate type	Sitting-drop MDL3 plates	XRL hanging-drop 24-well MD3-11 plates
Temperature (K)	293.15	293.15
Protein concentration (mg ml−1)	9.10	9.10
Buffer composition of protein solution	50 mM Tris–HCl pH 7.5, 150 mM NaCl	50 mM Tris–HCl pH 7.5, 150 mM NaCl
Composition of reservoir solution	0.1 M CHES pH 9.5, 20% PEG 8000	0.1 M CHES pH 8.5, 18% PEG 8000
Volume and ratio of drop	0.2 µl, 1:1	2 µl, 1:1
Volume of reservoir (µl)	40	500

2.3. Data collection and processing  

The crystals from the JSCG+ screen were briefly soaked in 20% PEG 8000, 0.1 M CHES pH 9.5, 25% glycerol and the optimized crystals were soaked in 18% PEG 8000, 0.1 M CHES pH 8.5, 25% glycerol before being flash-cooled in liquid nitrogen and used for data collection on beamline ID30B at the European Synchrotron Facility (ESRF) for the crystals from the JCSG+ screen and beamline I03 at Diamond Light Source (DLS) for the optimized crystals. A crystal from the screen diffracted to 2.8 Å resolution, while an optimized crystal diffracted to 2.2 Å resolution. Here, we present the results for the best diffracting crystal. A total of 1420 images were collected using an oscillation width of 0.1° and the data were processed with XDS (Kabsch, 2010 ▸). The data-collection parameters and processing statistics for the optimized crystal are listed in Table 3 ▸.

Table 3. Data collection and processing.

Values in parentheses are for the outer shell.

Diffraction source	I03, DLS
Wavelength (Å)	0.97625
Temperature (K)	100
Detector	PILATUS3 6M
Crystal-to-detector distance (mm)	384.54
Rotation range per image (°)	0.10
Total rotation range (°)	142
Exposure time per image (s)	0.102
Space group	P31 or P32
a, b, c (Å)	64.13, 64.13, 111.32
Mosaicity (°)	0.117
Resolution range (Å)	55.54–2.20 (2.33–2.20)
Total No. of reflections	104559 (16905)
No. of unique reflections	25843 (4160)
Completeness (%)	99.5 (99.1)
Multiplicity	4.0 (4.1)
〈I/σ(I)〉†	11.0 (1.0)
R meas (%)	7.6 (124.4)
Overall B factor from Wilson plot (Å2)	62

^†Diffraction images were processed to 2.20 Å resolution, where CC_1/2 reaches 50.1% and they still contain significant data according to the authors of the data-processing program (Karplus & Diederichs, 2012 ▸). I/σ(I) falls below 2.0 at 2.4 Å resolution.

3. Results and discussion  

The gene encoding DrDyP was inserted into the pDEST14 expression vector containing nucleotides encoding an N-terminal His tag and TEV cleavage site and used for small-scale test expression (5 ml cultures) in three different E. coli strains [BL21(DE3)*, BL21(DE3)pLysS and BL21(DE3)*pRARE] at 37°C. DrDyP was overexpressed in soluble form in E. coli BL21(DE3)* and BL21(DE3)pLysS cells, and migrated as a single band at ∼50 kDa on a 12% SDS–PAGE gel, as expected from theoretical calculations (the theoretical mass is 53.2 kDa with the His and TEV tags); it could not be identified in the non-induced control cells [BL21(DE3)*] or in BL21(DE3)*pRARE (Fig. 1 ▸ a). BL21(DE3)* was selected for medium-scale test expressions at three different temperatures (18, 25 and 37°C) and three different induction times (18, 6 and 3 h, respectively). A high yield of overexpressed protein at ∼50 kDa was identified in all conditions and expression at 37°C for 3 h was chosen for the large-scale experiment (Fig. 1 ▸ b).

Figure 1.

Result of small- and medium-scale expression testing and purification. Protein bands showing overexpressed DrDyP are circled. (a) 12% SDS–PAGE gels from small-scale expression in E. coli BL21(DE3)*, BL21(DE3)pLysS and BL21(DE3)*pRARE. S and I indicate soluble and insoluble fractions, respectively. (b) 12% SDS–PAGE gels from medium-scale expression in E. coli BL21(DE3)* at 18, 25 and 37°C. (c) Purified DrDyP after concentration.

After large-scale expression, the protein was purified using two HisTrap steps with a dialysis and TEV cleavage step between them. The protein solution had a red colouration, as expected owing to the presence of the haem group. Above pH 4.0 the UV–Vis spectra of DrDyP displayed Soret bands around 416 nm and Q-bands in the range 500–580 nm, which are consistent with a low-spin haem iron (Fig. 2 ▸). This is very unusual for DyPs (and other peroxidases), which typically display heterogeneous spin populations mainly comprising high-spin haems, with the sixth coordination site either vacant for substrate binding or weakly bound to water (Smulevich et al., 2010 ▸). The high abundance of low-spin haem iron may indicate poor peroxidase activity, as previously observed for other DyPs (Sezer et al., 2013 ▸). Upon lowering the solution pH below 4.0, additional features were observed in the spectrum. The appearance of a 639 nm charge-transfer band in particular suggested the presence of a mixture of high-spin and low-spin states and could indicate that the protein is only activated at low pH.

Figure 2.

UV–Vis spectra of DrDyP at different pH values. From top to bottom: pH 3.0 (red), 4.0 (green), 5.0 (blue) and 7.5 (black). The inset shows the magnified 450–670 nm region. The protein concentration was ∼4 µM.

Large-scale expression of DrDyP was also performed in the presence of haemin, as this has been used as a supplement to growth media for the recombinant expression of some DyPs to obtain 100% incorporation of haem (Fernández-Fueyo et al., 2015 ▸; Santos et al., 2014 ▸). Here, the addition of haemin had a negative effect on the growth of the cells and resulted in a low yield of recombinant protein, and thus it was not used as a supplement to the growth medium for the expression of DrDyP. Quantification of the iron content in recombinant DrDyP also confirmed that the addition of haemin was not needed since it is mostly produced in the holoprotein form (0.99 ± 0.07 moles of iron per mole of protein) in LB medium. The Reinheitszahl value (A ₄₁₆/A ₂₈₀ ratio) was greater than 1.0, indicating good purity.

The purified DrDyP was used to screen crystallization conditions using the JCSG+ crystallization suite and a Cartesian Mini-Bee crystallization robot. Red-coloured crystals appeared in condition A7 [20%(w/v) PEG 8000, 0.1 M CHES pH 9.5] after two months (Fig. 3 ▸ a). The crystals were further optimized by hanging-drop vapour diffusion (manually), and after two weeks crystals that were suitable for data collection were obtained in 18%(w/v) PEG 8000, 0.1 M CHES pH 8.5 (Fig. 3 ▸ b). The crystals were flash-cooled in crystallization buffer with 25% glycerol and were used for data collection on beamline ID30B at the ESRF to a resolution of 2.8 Å for the crystals from JCSG+ screen condition A7 and on beamline I03 at DLS to a resolution of 2.2 Å for the optimized crystals. The data-collection statistics for the optimized crystal are shown in Table 3 ▸.

Figure 3.

Crystals obtained in (a) condition D7 of the JCSG+ crystallization screen (20% PEG 8000, 0.1 M CHES pH 9.5) and (b) a hanging-drop vapour-diffusion optimization experiment in 18% PEG 8000, 0.1 M CHES pH 8.5 (full size and enlarged view). (c) The crystal (∼100 µm) used for data collection on beamline I03 at DLS.

Preliminary X-ray diffraction analysis showed that the DrDyP crystals belonged to space group P3₁ or P3₂, with unit-cell parameters a = b = 64.46, c = 111.82 Å. Unit-cell volume considerations (Matthews, 1968 ▸; Kantardjieff & Rupp, 2003 ▸) indicated the presence of one DrDyP monomer in the asymmetric unit, with a calculated V _M of 2.61 ų Da⁻¹ and an estimated solvent content of 52%.

The crystal structures of DyPs from many organisms are already known (Rahmanpour & Bugg, 2015 ▸), but those with the highest sequence identity to DrDyP (∼30%) are from an Anabaena sp. (Yoshida et al., 2016 ▸), an Amycolatopsis sp. (Brown et al., 2012 ▸), Bjerkandera adusta (Yoshida et al., 2011 ▸) and Auricularia auricula-judae (Strittmatter et al., 2013 ▸). Attempts to solve the structure of DrDyp using these structures as templates for molecular replacement is ongoing. The crystal structure of DrDyp from D. radiodurans will provide crucial information regarding the observed unusual features of the spin state of the haem and the possible lack of activity at high pH, and thus provide novel information about the members of the class V DyPs.

Funding Statement

This work was funded by Fundação para a Ciência e a Tecnologia grants PTDC/BBB233 BEP/0561/2014, SFRH/BPD/94050/2013, and FRH/BPD/79566/2011. Seventh Framework Programme grant 283570. COMPETE2020 grant LISBOA-01-0145-FEDER-007660 .