Expression of thermophilic two-domain laccase from Catenuloplanes japonicus in Escherichia coli and its activity against triarylmethane and azo dyes

Microorganism, and cloning of cjsl gene

Strain Catenuloplanes japonicus VKM Ac-875 was obtained from the All-Russian collection of microorganisms (http://www.vkm.ru/Collections.htm). The strain was grown on peptone yeast agar media, and its genomic DNA was purified from the biomass using diaGene kit for genomic DNA isolation from bacterial cell cultures (Dia-M, Moscow, Russia). Primers for PCR were designed based on the predicted multicopper oxidase sequence from the genome of Catenuloplanes japonicus NRRL B-16061 (NCBI Reference Sequence of protein: WP_033344226.1). Using the designed primers (875F 5′-ATGGACGACAACGTTGACAAACC-3′ and 875R 5′-TCATCCGGTGTGCCCTCC-3′), a corresponding PCR product was amplified. Correct amplification of the target gene was verified by sequencing. To clone the gene into expression vector, primers 875Fe (5′-AGTGGATCCGCGGGAGCCACCCGGAAG-3′) and 875Re (5′-AGTAAGCTTTCATCCGGTGTGCCCTCC-3′) were designed (cleavage sites for restriction endonucleases are underscored). The first primer introduced a BamHI site, and the second one inserted a HindIII site after the stop codon. The DNA fragment encoding cjsl, without the region for the signal sequence, was generated by PCR with 875Fe and 875Re primers and genomic DNA as template. The BamHI/HindIII-digested amplicon was cloned into pQE-30 (Qiagen, Hilden, Germany). Plasmid pQE::cjsl was used to transform E. coli M15 (pREP4) expression strain. Transformants were selected on LB plates containing 100 µg ampicillin/ml and 25 µg kanamycin/ml.

Recombinant expression and CjSL purification

For the production of CjSL, strain E. coli M15 (pREP4) transformed with pQE::cjsl plasmid was grown at 37 °C with agitation at 250 rpm to a cell density of 0.2–0.25 (A₆₀₀). Then, 0.2 mM isopropyl-β-D-thiogalactopyranoside and 1 mM CuSO₄ were added to culture medium. After that, the cells were first incubated for 18 h at 20 °C with agitation at 100 rpm, then they were incubated for 24 h at 25 °C without shaking. Cells were collected by centrifugation at 4,000 g for 30 min, suspended in 20 ml of 20 mM Tris-HCl buffer, pH 8.0, containing 0.5 M NaCl and 1 mM imidazole (buffer A) and disrupted by sonication. Cell debris was removed by centrifugation (40 min at 8.000 g). The protein was purified by affinity chromatography on a HisTrap 5-ml column (GE Healthcare, Chicago, IL, USA). Cell extract was loaded onto a HisTrap column equilibrated with buffer A. After loading, the column was washed with four volumes of the buffer A and then washed with four volumes of the buffer B (20 mM Tris-HCl, 0.5 M NaCl, 50 mM imidazole, pH 8.0). Active fractions containing the enzyme were eluted with buffer C (20 mM Tris-HCl, 0.5 M NaCl, 300 mM imidazole, pH 8.0). After the chromatography stage, the protein was dialyzed against 20 mM Tris-HCl buffer (pH 7.5) with 0.1 M NaCl, and then against 20 mM Tris-HCl buffer pH 7.5 without NaCl.

Enzyme characterization

The concentration of the protein was determined using the molar extinction at 280 nm (ε = 43,930 M⁻¹ × cm⁻¹) calculated from the protein sequence using Vector NTI Program (Life Technologies, Carlsbad, CA, USA). Presence of cupredoxin domains in the enzyme structure was identified using BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and InterProScan (https://www.ebi.ac.uk/interpro/) services. Presence of signal peptide was also identified in InterProScan. The UV-Vis absorption spectrum and molecular weight of native and denaturated protein were determined as described earlier (Trubitsina et al., 2015).

Laccase activity was determined at room temperature by measuring the oxidation of 1 mM 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonate) (ABTS) and 1 mM 2,6-dimethoxyphenol (2,6-DMP) in 50 mM Britton–Robinson buffer (Britton & Robinson, 1931) (at рН 3.6 and рН 9.2, respectively). The 50 mM Britton–Robinson buffer was prepared by mixing equal amounts of 0.05 M boric acid, 0.05 M orthophosphoric acid and 0.05 M acetic acid, and adjusting pH to the required value using 1 M NaOH. The oxidation of the substrates was detected by measuring the absorbance at 420 nm for ABTS (ɛ = 36,000 M⁻¹ × cm⁻¹) (Heinfling et al., 1998) and at 469 nm for 2,6-DMP (ɛ = 49,600 M⁻¹ × cm⁻¹) (Wariishi, Valli & Gold, 1992)

Optimal pH value was determined with ABTS (in a pH range from 3 to 5.5) and 2,6-DMP (in pH range from 8.0 to 10.5) in 50 mM Britton-Robinson buffer. The temperature dependence of the activity was determined in 50 mM Britton-Robinson buffer (pH 3.6) at temperatures from 30 °C to 90 °C using 1 mM 2,6-DMP as substrate. pH stability was estimated by incubation of the enzyme at room temperature for 15 days in the same buffer at pH values 3, 5, 7, 9, 11. Residual activity assay was determined in 50 mM Britton-Robinson buffer (pH 3.6) using 1 mM ABTS as substrate. Thermal stability of the enzyme was measured in 50 mM Britton-Robinson buffer (pH 3.6) using 1 mM ABTS at 80 °C, 90 °C and at water boiling temperature, by incubating the enzyme in thin-wall microtubes for 1 h (protein aliquots were taken every 10 min to determine the activity loss during the experiment).

The substrate specificity of the recombinant laccase was examined against 26 different putative substrates by detecting the changes in their absorption spectra after 24 h incubation. The substrate specificity was assayed at pH 3.6 or pH 9.2 for non-phenolic and phenolic compounds, respectively, in 50 mM Britton-Robinson buffer. The compounds assayed were: tyrosine, gallic acid (3,4,5-trihydroxybenzoic acid), ferulic acid (3-methoxy-4-hydroxycinnamic acid), vanillic acid (4-hydroxy-3-methoxybenzoic acid), tannic acid, caffeic acid (3,4-dihydroxycinnamic acid), gentisic acid (2,5-dihydroxybenzoic acid), syringic acid (4-hydroxy-3,5-dimethoxybenzoic acid), o-coumaric acid, 2-thiobarbituric acid, 3,4-dihydroxybenzoic acid, 4-hydroxybenzoic acid, K₄[Fe(CN)₆], ABTS, hydroquinone, methylhydroquinone, syringaldehyde (4-hydroxy-3,5-dimethoxybenzaldehyde), pyrogallol (01,2,3-trihydroxybenzene), guaiacol (2-methoxyphenol), catechol (1,2-dihydroxybenzene), vanillin (4-hydroxy-3-methoxybenzaldehyde), syringaldazine, 2,6-dimethoxyphenol, 3,5-dimethoxyphenol, 3,4,5-trimethoxyphenol, 2-aminophenol. The substrate concentration used was 1 mM.

Sodium azide (NaN₃), sodium fluoride (NaF), sodium chloride (NaCl), ethylenediaminetetraacetic acid (EDTA) and 1,10-phenanthroline were used as inhibitors. Inhibitor concentrations were 1, 10 and 100 mM in the reaction mixture with 1 mM ABTS or 1 mM 2,6-DMP as substrates in 50 mM Britton–Robinson buffer (pH 3.6 or 9.2 respectively).

The steady-state kinetic constants were obtained for the substrates 2,6-DMP and ABTS at the substrates’ pH optima at 30 °C. Calculation of the apparent kinetic constants was performed by a nonlinear regression of the data using Sigma Plot 11.0 software.

Dye decolorization

The decolorization of the following triarylmethane dyes: Malachite Green (λ_max = 617 nm (pH 4.0)), Brilliant Green (λ_max = 623 nm (pH 4.0)); azo dyes: Methyl Orange (λ_max = 478 nm (pH 4.0); λ_max = 464 nm (pH 6.5; 9.2)), Methyl Red (λ_max = 524 nm (pH 4.0; 6.5); λ_max = 432 nm (pH 9.2)), Congo Red (λ_max = 570 nm (pH 4.0); λ_max = 488 nm (pH 6.5; 9.2)), by laccase was tested. The reaction mixture (1 ml) contained 50 mM Britton–Robinson buffer (pH 4.0, 6.5 or pH 9.2), dye (final concentration of 50 µM) and purified enzyme 200 mU. A total of 50 µM ABTS and 50 µM syringaldehyde (SA) were used as potential redox mediators. Reactions mixture was incubated at 30 °C for 24 h. Control samples without enzyme were run in parallel under the same conditions. The dye degradation was judged by change of absorption spectrum of the oxidized compound and expressed in decolorization rate. The calculation formula: D = (A0−A1)/A0 × 100%, where D represents the decolorization rate (%/day), A0 is the initial absorbance of the dye solution at the maximum absorption wavelength, A1 results from the initial absorbance of the dye solution at the maximum absorption wavelength after the reaction.

Homology modeling

An initial model of Ac-875 laccase was created using SwissModel automatic protein model builder (Waterhouse et al., 2018). A large set of models of Ac-875 laccase was built in MODELLER (Webb & Sali, 2016). An experimentally determined structure of another trimeric two-domain laccase from Streptomyces coelicolor (PDB ID: 3CG8) was used as a template for model building. Parts of the protein that were not included into the alignment were not included into the models. A trimeric alignment was used to generate trimeric models. The best models were sorted according to their z-DOPE score (result of assess_normalized_dope command) (Shen & Sali, 2006) taken on average for three subunit models. Model refinement was achieved by internal resources of MODELLER package, including simulated annealing, steepest descent and in vacuo molecular dynamics.

Statistical analysis

Mean values and standard deviations were calculated for at least three replicates.

Protein cloning, expression and purification

Two-domain laccase gene was identified in the genome of actinobacterium Catenuloplanes japonicus NRRLB-16061 using BLAST program. Analysis of amino acid sequence in InterPro Scan revealed two cupredoxin domains (74–191, 195–307) and a signal peptide facilitating enzyme translocation via TAT-pathway (1–45). Conserved copper-binding residues, ten histidines and one cysteine, were all present in the protein sequence. The structural laccase gene designated as cjsl was cloned into pQE-30 expression plasmid without the TAT-leader sequence. pQE::cjsl construct was used for transformation of competent cells of E. coli M15 (pREP4). Laccase was expressed as a mature enzyme, without a signal peptide, with a 6 × His-tag at the N-terminus instead. Thankfully to one-step purification on Ni-sepharose column, laccase was obtained in electrophoretically homogenous state. Presence of copper in the medium during heterologous expression and a step of microaerobic cultivation of the strain for saturation of the culture by copper ions (Durao et al., 2008) were mandatory conditions for obtaining blue active enzyme. The yield of the protein was low, about 10 mg from one liter of culture.

Enzyme characterization

Based on amino acid sequence of CjSL, the calculated molecular weight of the protein was 33.8 kDa. CjSL boiled with β-mercaptoethanol and SDS migrated at 37 kDa in SDS-PAGE, CjSL with β-mercaptoethanol and SDS without boiling migrated 70 kDa (Fig. 1A). Gradient native gel electrophoresis showed that the molecular weight of the native active protein was around 200 kDa (Fig. 1B). All this points on multimeric state of the native laccase. The absorption spectrum of the enzyme had a maximum at 600 nm due to the presence of T1-copper center conferring deep blue color, and the shoulder at 340 nm can be attributed to the presence T3-copper center (Fig. 2).

The enzyme catalyzed oxidation of an electron donor ABTS at acidic pH values. Maximal ABTS oxidation rate was observed at рН 3.6. Oxidation of an electron and proton donor phenolic compound 2,6-DMP by CjSL was observed at alkaline pH values with an optimum of pH 9.2 (Fig. 3A). The optimum reaction temperature for the recombinant laccase was 70 °C (Fig. 3B). In terms of thermal denaturation, CjSL turned out to be a very stable enzyme. After an hour of incubation at 80 °C, the laccase preserved 63% of its initial activity, at 90 °C, 53% of the initial activity, and 26% of the initial activity was conserved after an hour of boiling (Fig. 4). CjSL was most stable at pH 5 and 7. It retained around 67% and 60% of activity after 15 days of incubation at pH 5.0 and 7.0, respectively. At more acidic conditions, it was less stable. At рН 3.0, the residual activity of the enzyme by the end of 24 h of incubation was around 12% (Fig. 5).

The effect of some conventional laccase inhibitors (NaN₃, NaF, NaCl, EDTA and 1,10-phenanthroline) was studied. Table 1 summarizes the results. Inhibition effect was measured at acidic and alkaline conditions. It was established that only 1,10-phenanthroline acts as a good inhibitor in both the acidic and alkaline pH value (at 10 mM, the residual activity was 2.5% and 5.7% at рН 3.6 and 9.2, respectively). NaN₃, NaCl and NaF reduced laccase activity only at acidic conditions (Table 1), and NaN₃ was the most effective laccase inhibitor (at 10 mM, the residual activity of the laccase was less than 2%). At pH 9.2, NaN₃ decreased the laccase activity only at 100 mM, NaF did not display inhibitory effect, and NaCl had a low inhibitory effect (by 5% at 100 mM). EDTA dramatically reduced laccase activity at alkaline conditions (residual activity less than 9% at 10 mM). At рН 3.6, EDTA inhibited the laccase to a much lesser extent (50% of residual activity at 10 mM).

Investigation of the enzyme’s substrate specificity showed that CjSL was able to oxidize 2,6-dimethoxyphenol, 3,4,5-trimethoxyphenol, 2-aminophenol, catechol, guaiacol, hydroquinone, pyrogallol, methylhydroquinone, syringaldehyde, syringaldazine, caffeic acid, o-coumaric acid, ferulic acid, gallic acid, gentisic acid, syringic acid, tannic acid, vanillic acid, 3,4-dihydroxybenzoicacid, ABTS and K₄[Fe(CN)₆]. The enzyme did not oxidize L-tyrosine, vanillin, 2-thiobarbituric acid, 3,5-dimethoxyphenol and 4-hydroxybenzoic acid. Table 2 summarizes the results.

The kinetic constants of CjSL were determined with ABTS (pH 3.6) and 2,6-DMP (pH 9.2), For ABTS, the K_m was 0.39 mM; V_max = 13.09 nmol/min. For 2,6-DMP, the K_m was 1.86 mM; V_max = 2.93 nmol/min.

Dye Decolorization

The purified laccase alone could efficiently decolorize only azo dye Congo Red at pH 4.0 (Table 3). Decolorization rate of Congo Red by laccase was 36%/day. Other dyes, namely Мalachite Green, Brilliant Green, Methyl Orange, and Methyl Red, were unoxidizable by laccase without a mediator or oxidized with a low rate (4%/day and less) (Table 3). Laccase could decolorize triphenylmethane dyes, Мalachite Green and Brilliant Green, only in presence of redox mediator ABTS. As shown in Table 3, 22% and 94% decolorization for Malachite Green and Brilliant Green (50 μM), respectively, was obtained within 24 h incubation with 200 mU ml⁻¹ of laccase in 50 mM Britton and Robinson buffer at 30 °C. Azo dye Methyl Orange was bleached with the highest rate at рН 4.0 in presence of ABTS (decolorization rate was 57%/day), whereas in presence of syringaldehyde the rate was quite low (6%) (Table 3). The highest decolorization rate for azo dye Methyl Red was observed at рН 6.5 in presence of ABTS and comprised 46%/day. At рН 4.0 and 9.2, the rate of Methyl Red decolorization in presence of ABTS was lower, comprising 12%/day and 10%/day, respectively. Congo Red with syringaldehyde and ABTS as redox mediators was most intensely decolorized at рН 6.5 (decolorization rate comprised 31% and 22%, respectively).