TEXT

Top Abstract Text References

The ligninolytic enzymes lignin peroxidase (LiP) and manganese peroxidase (or manganese-dependent peroxidase) (MnP) were described in 1983 and 1984 in the fungus Phanerochaete chrysosporium (6, 9, 24). Their corresponding cDNA were cloned, and recombinant enzymes were obtained in 1987 to 1989 (11, 15, 18, 25), and the molecular structure of both proteins was reported in 1993 and 1994 (16, 17, 23). A third type of ligninolytic peroxidase has been described in several species of the genera Pleurotus and Bjerkandera, characterized by sharing catalytic properties of MnP (efficient oxidation of Mn²⁺ to Mn³⁺) and LiP (oxidation of veratryl alcohol, p-dimethoxybenzene, and non-phenolic lignin model dimers) and a high affinity for substituted hydroquinones (2, 7, 8, 12, 13, 20). Recently, two peroxidases produced in solid-state fermentation and liquid cultures of Pleurotus eryngii have been cloned, the latter appearing as two 99% identical variants (MnPL1 and MnPL2) encoded by two alleles of the same gene (3, 19). These versatile enzymes have higher sequence and structural homology with P. chrysosporium LiP than with MnP, but their molecular models showed a putative Mn²⁺ interaction site near the internal propionate of heme, accounting for their ability to oxidize low concentrations of this cation. Expression of the gene encoding the peroxidase isolated from P. eryngii liquid cultures, allelic variant MnPL2 (19), in Escherichia coli (by using the T7 lac promoter and terminator) resulted in a recombinant protein detected by immunoblotting (unpublished results). However, all attempts to obtain active peroxidase by protocols described previously (21) for refolding denatured proteins were unsuccessful. In order to obtain an active recombinant enzyme, the expression of P. eryngii peroxidase in Aspergillus nidulans, as a model organism for fungal genetics studies, was attempted.

A cDNA fragment encoding the whole protein with its signal peptide was amplified, cloned, and used to transform an A. nidulans arg mutant strain. For this purpose, two PCR primers (5'-CGggatccCCCATGTCTTTCAAGACGC-3' and 5'-GgaattcTTACGATCCAGGGACGGG-3') were synthesized (restriction sites shown in lowercase), and a BamHI-EcoRI fragment corresponding to MnPL2 cDNA (GenBank accession no. AF007222) was amplified by using MnPL2 cDNA cloned into pBSK+/ as a template (19). The PCR products were separated on 0.8% agarose and purified, and the BamHI-EcoRI fragment was cloned into palcA1 (5) (containing the alcA promoter inducible by ethanol or threonine and repressed by glucose [10], the trpC terminator, and the argB gene encoding ornithine carbamoyltransferase as a selection marker), yielding pALMP2. Automatic sequencing of the PCR fragment, with synthetic oligonucleotides as primers, confirmed that the sequence matched that of MnPL2 cDNA exactly. Mycelia of A. nidulans (biA1 methG1 argB2) (IJFM A729, derived from FGSC A89 and A219) were grown at 28°C and 180 rpm in a minimal medium (4) containing 10 µg of D-biotin per liter, 74.5 mg of L-methionine per liter, and 0.53 g of L-arginine per liter. Protoplasts were obtained by incubating washed mycelia (1 g [wet weight]) with Novozym 234 (20 mg) in 1.2 M MgSO₄ (buffered at pH 5.8) containing serum albumin (24 mg) and separated by centrifugation at 4,000 × g with an overlay of 0.6 M sorbitol in 0.1 M Tris-HCl (pH 7.5). They were transformed (1 to 2 µg of pALMP2) in 10 mM Tris-HCl (pH 7.5) containing polyethylene glycol 6000, sorbitol, and CaCl₂ (26), inoculated in a selective minimal medium with 10 µg of D-biotin per liter, 74.5 mg of L-methionine per liter, and 1 M sucrose, in a soft agar overlay, and incubated at 37°C. Southern blot analysis after DNA digestion (with the argB gene as a probe) showed that the plasmid containing mnpl2 cDNA was integrated in the genome of A. nidulans, occupying a position different from that of argB of the host fungus, and no evidence for multiple integration was obtained.

Different growth conditions for the recombinant A. nidulans strain containing mnpl2 DNA were investigated for peroxidase production. Since peroxidase expression was under the alcA promoter, it was induced by substituting threonine for glucose. The fungus was grown in minimal or complete media (4) containing 10 µg of D-biotin per liter and 74.5 mg of L-methionine per liter at 28°C and 180 rpm for 24 h. Then, washed mycelia were transferred to induction media with the same composition but containing 0.05% glucose and 100 mM threonine, with or without 0.5 g of hemin per liter. Peroxidase activity in culture samples collected during incubation (96 h) was estimated by the formation of Mn³⁺-tartrate complex (12). The addition of hemin to the medium was required for peroxidase production in both minimal and complete media, the latter yielding the highest peroxidase activity (up to 150 U/liter). It had already been reported that the addition of hemin to media increased the activities of other peroxidases expressed in ascomycetes (1, 22).

The recombinant peroxidase MnPL2* from A. nidulans was purified from cultures in a hemin-containing complete medium. Maximal peroxidase activity was observed 43 h after induction, and protein MnPL2* was purified from concentrated and dialyzed (against 10 mM sodium tartrate, pH 4.5) culture liquid by using the protocol described for P. eryngii MnPL2 (12) with an additional step. As shown in Fig. 1A, Mono-Q chromatography yielded a single peak with high absorbance at 410 nm and peroxidase activity corresponding to protein MnPL2*. This peak showed the same elution volume as allelic variant MnPL2 from P. eryngii CBS 613.91 (IJFM A169) grown in glucose-peptone medium (12) (Fig. 1A). Proteins MnPL2 and MnPL2* were N deglycosylated with endo--N-acetylglucosaminidase, and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) of native and deglycosylated proteins was performed in 12% polyacrylamide gels, which were stained with AgNO₃. As shown in Fig. 1B, both proteins showed the same molecular mass before (43 kDa) and after (41 kDa) deglycosylation. In the case of protein MnPL2*, a contaminant protein representing 13% of the peak collected from Mono-Q was detected. Since this protein did not seem to be N glycosylated, a purification step was performed based on affinity for concanavalin A (ConA). The peroxidase peak from Mono-Q chromatography was concentrated, dialyzed against 10 mM sodium tartrate (pH 5) containing 0.5 M NaCl, 1 mM MnCl₂, and 1 mM CaCl₂, and applied to a ConA-Sepharose column. The retained protein MnPL2* was eluted with 0.2 M -D-methylglucoside, dialyzed, and stored at 20°C. In this way, electrophoretically homogeneous MnPL2* protein was obtained, as shown in Fig. 1C. N-terminal sequencing of the recombinant protein by automated Edman degradation revealed the same sequence obtained for mature protein MnPL2 from P. eryngii. The whole process used to purify peroxidase MnPL2* is summarized in Table 1. The purification yield and factor are relative to peroxidase activity after ultrafiltration, but they should be higher if referred to activity in the culture liquid, where an unknown reaction seemed to partially interfere with the activity assay. Despite the fact that recombinant peroxidase activity in cultures of A. nidulans was at a lower level than peroxidase activity in cultures of P. eryngii (12), the specific activity of purified protein MnPL2* was similar to that of protein MnPL2 from P. eryngii. Expression in other fungal systems, such as Aspergillus oryzae or Pichia pastoris (14, 22), will be attempted in the future when high peroxidase levels are required.

View larger version (45K): [in this window] [in a new window]   FIG. 1.   Purification of peroxidase MnPL2* from recombinant A. nidulans. (A) Superimposition of Mono-Q profiles (410 nm) of protein MnPL2* from A. nidulans (dashed line) and proteins MnPL1 and MnPL2 from P. eryngii (solid line) showing the same elution volume for wild-type and recombinant MnPL2 (NaCl gradient shown as a dotted line). (B) SDS-PAGE of proteins in peaks MnPL2* and MnPL2, showing similar molecular masses before (G) and after (D) deglycosylation; lane S, low-molecular-mass standards from Bio-Rad. (C) SDS-PAGE of glycosylated protein MnPL2* after a final purification step with ConA, which removed any contaminating protein.

Finally, steady-state kinetic constants for oxidation of Mn²⁺, veratryl alcohol, and methoxyhydroquinone by peroxidases MnPL2* and MnPL2 were compared, under the conditions described for P. eryngii peroxidase (8). The apparent K_m and turnover numbers (t) are shown in Table 2, revealing that both peroxidases have similar enzymatic activities oxidizing Mn²⁺ to Mn³⁺ as well as phenolic and nonphenolic aromatic substrates with high affinity on both Mn²⁺ and substituted hydroquinones. The recombinant peroxidase also oxidized 2,6-dimethoxyphenol, although no kinetic constants were calculated. The K_m for H₂O₂ (around 10 µm) was similar to that of the P. eryngii enzyme.

In summary, the P. eryngii peroxidase (allelic variant MnPL2) could be successfully expressed and secreted in A. nidulans under the alcA promoter with the mnpl2 cDNA including its signal sequence, and electrophoretically homogeneous peroxidase MnPL2* was isolated with a final purification factor of 115. This recombinant protein exhibited the same molecular mass (determined by SDS-PAGE) of peroxidase MnPL2 before and after deglycosylation, as well as the same catalytic properties of the P. eryngii enzyme. The latter finding confirms that the enzyme produced by P. eryngii, the biochemical and molecular characteristics of which have been recently reported (19), represents a new type of versatile peroxidase sharing catalytic properties of both LiP and MnP of P. chrysosporium. Site-directed mutagenesis studies are in progress, to confirm the involvement of several key residues in the catalytic properties of this enzyme, by using the heterologous expression system described here.