Characterization of a novel manganese peroxidase-lignin peroxidase hybrid isozyme produced by Bjerkandera species strain BOS55 in the absence of manganese.

A novel manganese-dependent peroxidase (MnP) isozyme produced in manganese-free cultures of Bjerkandera sp. strain BOS55 was purified and characterized. The production of the enzyme was greatly stimulated by the exogenous addition of various physiological organic acids such as glycolate, glyoxylate, and oxalate. The physical properties of the enzyme are similar to those of MnP isozymes from different white rot fungi (Mr = 43,000, pI 3.88, and epsilon407 nm = 123 mM-1 cm-1). The Bjerkandera MnP was efficient in the oxidation of Mn(II), as indicated by the kinetic constants (low Km of 51 microM and turnover number of 59 s-1). Furthermore, the isozyme was able to oxidize various substrates in the absence of manganese, such as 2,6-dimethoxyphenol, guaiacol, ABTS, 3-hydroxyanthranilic acid, and o- and p-anisidine. An interesting characteristic of the isozyme was its ability to oxidize nonphenolic substrates, veratryl alcohol and 1,4-dimethoxybenzene, without manganese addition. The affinity for veratryl alcohol (Km = 116 microM) and its turnover number (2.8 s-1) are comparable to those of lignin peroxidase (LiP) isozymes from other white rot fungi. Manganese at concentrations greater than 0.1 mM severely inhibited the oxidation of veratryl alcohol. The results suggest that this single isozyme is a hybrid between MnP and LiP found in other white rot fungi. The N-terminal amino acid sequence showed a very high homology to those of both MnP and LiP isozymes from Trametes versicolor.

Lignin is an abundant natural aromatic polymer occurring in the woody tissue of higher plants. Due to its hydrophobicity and complex random structure lacking regular hydrolyzable bonds, lignin is poorly biodegraded by most microorganisms. The best degraders of lignin are white rot fungi that produce extracellular peroxidases (1). These enzymes are involved in the initial attack of the lignin polymer (2). Manganese-dependent peroxidase (MnP) 1 is one of the most frequently encountered peroxidases among white rot fungi (3).
MnP is a glycosylated hemeprotein occurring mostly as a number of isozymes (1). MnP has the same catalytic cycle as other peroxidases, involving a 2-electron oxidation of the heme by H 2 O 2 and subsequent reduction of compound I via compound II in two 1-electron steps to the native enzyme (4,5). A distinctive characteristic of this enzyme is that the best reducing substrate for compounds I and II is Mn(II), a metal naturally present in wood. The Mn(III) formed oxidizes other substrates (6). Crystallographic and site-directed mutant studies confirmed the presence of a unique manganese-binding site in MnP from the best studied white rot fungus, Phanerochaete chrysosporium (7,8).
Organic acids such as oxalate, glyoxylate, and lactate were shown to have an important role in the mechanism of MnP and lignin degradation (6,9,10). Mn(III) is stripped from the enzyme by organic acids, and the formed Mn(III)-organic acid complex acts as a diffusible mediator in the oxidation of lignin by MnP (6,9). Mn(III) can also oxidize organic acids, yielding radicals (11). There are two possible sources of organic acids during lignin degradation. The fungi are able to produce de novo aliphatic organic acids, mainly oxalate (12). Moreover, organic acids are formed from the degradation of lignin (13,14).
The ligninolytic enzymes are produced during the secondary metabolism triggered by nutrient limitation (1,15). Manganese is an absolute requirement for expression of mnp genes in several white rot fungi (16,17). Consistent with manganese requirement, putative metal response elements were found in the promoter regions (15,18). Furthermore, enhanced MnP production is associated with increased manganese concentrations in many white rot fungi (17,19,20).
Bjerkandera sp. strain BOS55 is a good MnP producer (20). Previous studies have shown that this fungus is nitrogenunregulated, and MnP production is greatly increased by nutrient nitrogen sufficiency and excess (21). MnP production is also enhanced by supplementing the culture medium with simple organic acids (20). Following the general trend of most white rot fungi, manganese is stimulatory for MnP in this fungus (20). However, Bjerkandera sp. strain BOS55 is atypical since it produces considerable MnP in manganese-free media (20,22). In this study, we characterized a MnP isozyme produced in the absence of manganese. Furthermore, the manganese dependence for the oxidation of various substrates was tested.

MATERIALS AND METHODS
Microorganism-Bjerkandera sp. strain BOS55 (ATCC 90940) was maintained in glucose/peptone/yeast extract slants as described earlier (23). Prior to the experiment, the fungus was transferred to glucose/ malt extract plates (20), which were incubated for 4 -6 days at 30°C. As inocula, 6-mm diameter agar plugs from the leading mycelial edge were used in the experiments.
Media-The standard basal medium used contained 2.2 mM nitrogen as diammonium tartrate, 56 mM glucose, 2 mg/liter thiamin, and manganese-free BIII mineral medium (pH 4.5), and the standard buffer (2,2-dimethyl succinate) was omitted (24). All glassware was previously washed with 5 M HNO 3 to remove contaminating manganese. The measured concentration in the manganese-free medium was always Ͻ0.01 M. In some experiments, the basal medium was also supplemented with manganese nutrients, providing a final concentration of 0.033 mM. Glycolic acid, glyoxylic acid, and oxalic acids (5 mM) were individually added (neutralized to pH 4.5) to the cultures at the time of incubation in order to study their stimulatory effect on MnP production.
Culture Conditions-Filter-sterilized 25-ml aliquots were placed in 250-ml cylindrical flasks, which were previously autoclaved with a drop of demineralized water (22). Each flask was inoculated with three agar plugs. Cultures were incubated statically under an air atmosphere at 27°C.
Enzyme Purification with Fast Protein Liquid Chromatography (FPLC)-The collected supernatant was separated from the mycelial mat by centrifugation (20,000 ϫ g). The extracellular fluid was washed with 10 mM sodium acetate (pH 6.0) and concentrated by ultrafiltration with a PM-10 membrane (10 kDa; Amicon, Rotterdam, The Netherlands). The concentrated supernatants were loaded in 10 mM sodium acetate onto a MonoQ anion-exchange column (Pharmacia, Uppsala, Sweden), and heme-containing proteins were monitored with a wavelength of 405 nm. Proteins were eluted with a linear gradient of sodium acetate up to 450 mM (pH 6.0) in 45 min, and 1-ml fractions were collected (1 ml/min) for 45 min. The fractions, 33-35, that contained the most MnP activity were collected and then washed and concentrated by ultrafiltration. The concentrate was eluted by creating a decreasing pH gradient in the pH range of 4.5-2.7 on a MonoP chromatofocusing column (Pharmacia) as described earlier (25), and 1-ml fractions were collected (1 ml/min) for 40 min. The MnP-containing fraction was washed with demineralized water and concentrated by ultrafiltration. The concentrate was used for enzyme characterization, kinetic, and substrate oxidation studies.
The MonoP enzyme concentrate was brought through two additional FPLC steps to confirm its purity. First, gel filtration was carried out using a Superdex 75 HR 10/30 column (10 ϫ 300 mm, volume of 24 ml; Pharmacia). The enzyme was eluted in 10 mM sodium acetate buffer (pH 6.0) with a flow rate of 0.5 ml/min for 50 min, and 0.5-ml fractions were collected. A single symmetrical peak was observed, and the enzyme-containing fractions, 22 and 23, were pooled and concentrated by ultrafiltration. The enzyme was washed with 10 mM potassium phosphate buffer (pH 6.0). Diammonium sulfate was added to the protein, reaching a final concentration of 1 M, and loaded onto a 1-ml hydrophobic interaction chromatography column (phenyl-Sepharose HP, Pharmacia). Proteins were eluted in a decreasing linear gradient of 1 M diammonium sulfate in 10 mM potassium phosphate buffer (pH 6.0) in 45 min (1 ml/min). Fractions (1 ml) were also collected every minute. Again, a single symmetrical peak was observed eluting from 19 to 22 min.
Enzyme Characterization-Protein concentration was determined according to Bradford (26) using bovine serum albumin as a standard. The concentration of purified MnP was also measured at a wavelength of 406 nm and calculated using ⑀ 406 nm ϭ 129 mM Ϫ1 cm Ϫ1 (6), which gave approximately the same result. The molecular mass of the enzyme was determined by SDS-polyacrylamide gel electrophoresis (12% polyacrylamide gel) in a Phastsystem (Pharmacia) using marker proteins (low molecular mass calibration kit from Pharmacia) as standards. The isoelectric point was estimated by chromatofocusing by measuring the pH of the collected fraction containing MnP isozyme. The enzyme absorbance spectrum was determined with 86 mg/liter MnP in the presence of 50 mM malonate buffer (pH 4.5) by scanning the absorbance of the enzyme at a wavelength range of 350 -700 nm using a Perkin-Elmer Lambda1 UV-visible spectrophotometer. The oxidized enzyme was scanned after addition of 0.2 mM H 2 O 2 .
Manganese Inhibition of Veratryl Alcohol Oxidation-In some experiments, the effect of Mn(II) on veratryl alcohol oxidation was elucidated. The oxidation of 0.1 mM veratryl alcohol was tested during a 10-min incubation period with 0.4 mM H 2 O 2 in 50 mM sodium succinate buffer (pH 3.8) with 2 mM sodium pyrophosphate and varying Mn(II) concentrations. For comparison, a similar experiment was conducted with purified LiP-5 isozyme from Bjerkandera sp. strain BOS55 (28) supplied at the same units in terms of veratryl alcohol oxidation. Formation of veratraldehyde was analyzed by HPLC.
Determination of Background Manganese Content in Culture Media-Manganese content of the culture media was measured by a inductivity-coupled plasma mass spectrometer by the Department of Soil Science and Plant Nutrition, Wageningen Agricultural University (Wageningen, The Netherlands). The detection limit for manganese is 0.002 M.
N-terminal Sequence Determination-The N-terminal amino acid sequence determination was carried out at the Rijks Universiteit Leiden, Faculteit der Geneeskunde, Vakgroep Medische Biochemie (Leiden, The Netherlands).
Chemicals-All chemicals were commercially available and used without further purification.

MnP Production in the Cultures of Bjerkandera sp. Strain BOS55 and Enzyme
Purification-Bjerkandera sp. strain BOS55 was cultivated in manganese-free nitrogen-limited glucose medium in the presence and absence of simple organic acids. The FPLC studies using the MonoQ column demonstrated the increased production of hemeprotein with MnP activity as a result of organic acid addition to the cultures. As an example, the FPLC profiles and enzyme activities measured in the collected fractions of extracellular fluid from a 7-day-old culture grown in the absence or presence of 5 mM glycolate are shown in Fig. 1. Similar results were obtained with oxalate and glyoxylate. In the cultures without the organic acids, very little hemeprotein and limited MnP activities were detected. The addition of glycolate stimulated the production of several heme peaks. However, most of the MnP activity was attributed to one peak (fractions [33][34][35]. In other studies with glycolate and other organic acids (data not shown), the MnP activity was sometimes distributed between two major peaks: fractions 33-35 and 38 -39, respectively. Since MnP activity was consistently present in fractions 33-35, this isozyme was isolated and used for further characterization. The purification is summarized in Table I.
Physical Properties and N-terminal Amino Acid Sequence-As the first step in the enzyme characterization, the molecular mass was measured by the SDS-polyacrylamide gel electrophoresis method and was found to be 43,000 Da (Fig. 2). This gel electrophoresis resulted in only one band, indicating a successful purification. The pI was 3.88 as estimated by the chromatofocusing method. The spectrum of the enzyme showed a typical peak for native enzyme at 407 nm with a molar extinction coefficient of 123 mM Ϫ1 cm Ϫ1 , and the addition of excess hydrogen peroxide resulted in a shift to 420 nm (Fig. 3). The N-terminal amino acid sequence of the protein was also determined up to 25 amino acids: VACPDGVNTATNAACCALF-AVRDDI.
Catalytic Properties and Manganese Dependence of Substrate Oxidation-Different substrates were used to evaluate the manganese dependence of their oxidation at pH 4.5 in 50 mM malonate buffer. The results are summarized in Table II. The substrate 2,6-dimethoxyphenol (DMP) was oxidized in the presence and absence of manganese. The oxidation rate of DMP without manganese was 18% of that in the presence of manganese. This trend was observed in the case of other substrates such as guaiacol, ABTS, 3-hydroxyanthranilate, o-anisidine, and p-anisidine. The enzyme was also able to directly oxidize veratryl alcohol at pH 4.5 without any manganese (Table III).
The kinetics of the oxidizing substrate (H 2 O 2 ) and of various direct reducing substrates of the enzyme were compared in terms of K m and turnover number (Table III). At a physiological pH of 4.5, Mn(II) was clearly the best reducing substrate, with a turnover of 59 s Ϫ1 . Nonetheless, veratryl alcohol, DMP, and ABTS were directly oxidized by the enzyme at this pH, with a turnover of 2.3 to 1.3 s Ϫ1 in the absence of Mn(II). The enzyme had a high affinity for Mn(II), DMP, and ABTS, whereas the affinity for veratryl alcohol at pH 4.5 was an order of magnitude less.
The pH dependence of the oxidation of several substrates was also studied. Fig. 4 shows the pH dependence of Mn(II) and DMP oxidation. The highest rate of Mn(II) oxidation was observed at pH 4.5. At pH 3.0, the Mn(II) oxidizing activity was negligible. The direct oxidation of DMP was the highest at pH 3.0, and at this pH, the rate was comparable in the presence and absence of manganese. Likewise, the oxidation of ABTS and veratryl alcohol in the absence of manganese was 2-3-fold higher at pH 3.0 compared with pH 4.5. At pH 3.0, the affinity    of the Bjerkandera MnP for veratryl alcohol was ϳ5-fold greater than at pH 4.5 (Table III).
To verify that the manganese and veratryl alcohol oxidation activities were due to one isozyme, the enzyme was purified further by two additional FPLC steps: gel filtration and hydrophobic interaction. After each step, only one symmetrical peak was observed, and the molar ratio of 0.1 mM Mn(II) oxidation rate to 0.1 mM veratryl alcohol oxidation rate at pH 4.0 remained constant at 23 (data not shown).
Product Identification of Nonphenolic Compound Oxidation-Several nonphenolic compounds were incubated in the absence of manganese for 4 hours with the Bjerkandera MnP at pH 3.0 in 50 mM succinate buffer. The products of the oxidation were identified by HPLC diode array detector. The results are summarized in Table IV. The main product of veratryl alcohol oxidation was veratraldehyde, recovered with a molar yield of 46%. 1,4-Benzoquinone was also identified as a product of 1,4-dimethoxybenzene oxidation, with a molar yield of 41%. Even 2-chloro-1,4-dimethoxybenzene was oxidized to a small extent by the isozyme. Benzyl alcohol was not significantly oxidized by this enzyme.
Effect of Manganese on Veratryl Alcohol Oxidation-The veratryl alcohol oxidation to veratraldehyde by Bjerkandera MnP and LiP was compared in the presence of manganese (Table V).
LiP was unaffected by the presence of either 0.3 mM Mn(II) or Mn(III). However, Bjerkandera MnP behaved differently. Mn(II) in the concentration range of 0.1-1 mM inhibited the oxidation of veratryl alcohol. Interestingly, Mn(II) at a low concentration (0.033 mM) stimulated the veratryl alcohol oxidation to veratraldehyde. DISCUSSION Bjerkandera sp. strain BOS55 produces MnP in high quantities under different culture conditions. An interesting characteristic of this fungus is the ability to produce MnP in manganese-free media (20,22). Recently, the production of MnP under manganese deficiency was reported for two Pleurotus species, P. eryngii and P. ostreatus, and the isozymes were characterized (29,30). In the Pleurotus strains, MnP production was inhibited by even trace amounts of manganese. In contrast, Bjerkandera sp. strain BOS55 MnP production is stimulated by manganese nutrients. The same MnP isozymes produced under manganese deficiency were also detected in manganese-sufficient cultures (data not shown). A previous study showed that the addition of certain organic acids enhanced MnP production under manganese-excess conditions (20). In this study, glycolate, glyoxylate, and oxalate highly stimulated MnP production under manganese-deficient conditions. These acids are physiological metabolites of white rot fungi (6,12,14,31). The role of the organic acids in MnP production still requires further studies.
The physical characteristics in terms of molecular mass, isoelectric point, and heme absorbance of the purified MnP do not differ from those of other MnP isozymes described from other white rot fungi (32)(33)(34)(35). The turnover number of Mn(II) oxidation by MnP from Bjerkandera sp. strain BOS55 shows that this isozyme is efficient and comparable to other MnP isozymes described in the literature (28, 29, 36 -38). However, the Bjerkandera MnP appears to be unique since it has a broader substrate specificity compared with most MnP isozymes reported. This isozyme is able to directly oxidize various aromatic amines as well as guaiacol and DMP in the absence of manganese. Although the oxidation rate was lower in the absence of manganese than in its presence at pH 4.5, the oxidation of these compounds was completely independent of manganese at pH 3.0.
Remarkable was the observation that Bjerkandera MnP was even able to oxidize veratryl alcohol and 1,4-dimethoxybenzene, which are traditional LiP substrates, in the absence of manganese. Since veratryl alcohol was oxidized, but not benzyl alcohol, the oxidation of veratryl alcohol most likely proceeds via a cationic radical mechanism rather than a benzylic radical mechanism. The observed oxidation of 1,4-dimethoxybenzene to 1,4-benzoquinone also supports a cation radical mechanism. The K m and V max values for veratryl alcohol oxidation (Table  III) are comparable to the results reported for LiP isozymes from other white rot fungi (39,40). However, unlike classic LiP isozymes (41,42), Mn(II) concentrations greater than 0.1 mM  were found to severely inhibit the oxidation of veratryl alcohol by the Bjerkandera MnP. This observation would be expected since both compounds as substrates of the enzyme would have to compete for the same oxidized heme group. The fact that a low rate-limiting (noncompetitive) concentration (0.033 mM) of Mn(II) stimulated veratryl alcohol oxidation indicates that Mn(II) helped to protect the enzyme from H 2 O 2 inactivation, probably by completing the catalytic cycle. It was also observed that veratryl alcohol (0.1 mM) incubated with the enzyme without Mn(II) could delay the H 2 O 2 inactivation of the enzyme as measured by Mn(II)-oxidizing ability (data not shown). These phenomena can only be explained by a single enzyme utilizing both compounds as a substrate.
Veratryl alcohol is a de novo secondary metabolite produced by many white rot fungi, including Bjerkandera sp. strain BOS55 (3). Concentrations ranging from 0.01 to 0.80 mM occur in liquid cultures as well as during solid-state fermentation of wood (23,43). The oxidation of veratryl alcohol has generally been attributed to LiP as opposed to MnP due to the high ionization potential of the compound, which is not directly oxidized by Mn(III). Veratryl alcohol has been implicated in several roles in the catalysis of LiP. This metabolite mediates the oxidation of aromatic compounds with lower redox potential than that of veratryl alcohol cation radical (44, 45). As the best reducing substrate for LiP compound II, veratryl alcohol can reduce the enzyme to the native state, completing the catalytic cycle (46). Consequently, the presence of veratryl alcohol prevents the inactivation of LiP by H 2 O 2 (47). The fact that the Bjerkandera MnP isozyme oxidizes veratryl alcohol may indicate that this metabolite can have a role in the catalysis of MnP under manganese-free conditions. Previously, only a few MnP isozymes have been shown to carry out manganese-independent oxidation. Several aromatic amines were directly oxidized by MnP from Ceriporiopsis subvermispora (34); however, manganese was essential for the oxidation of guaiacol. The MnP from Pleurotus strains was shown to oxidize DMP in the absence of manganese (29,30). The MnP from Lentinus edodes is reported to oxidize veratryl alcohol; however, unlike the Bjerkandera MnP, manganese was an essential requirement (48). Although it is known that purified MnP from P. eryngii and P. ostreatus directly oxidizes veratryl alcohol without manganese (29,30), the K m value of the Bjerkandera MnP was ϳ25-35-fold lower. Thus, only the Bjerkandera MnP would truly be functional under physiological conditions.
All the kinetic data suggest that the Bjerkandera MnP isozyme described here has characteristics of both MnP and LiP. The enzyme with the most comparable substrate spectrum reported in the literature to that of the Bjerkandera MnP is the Pleurotus MnP (29). However, the N-terminal amino acid sequences of the Pleurotus MnP isozymes have 8 -9 amino acids that do not match in the first 20 amino acids, indicating that the Bjerkandera MnP is a distinct gene product. The N-terminal amino acid sequences of MnP from other white rot fungi such as P. chrysosporium, C. subvermispora, Phanerochaete sordida, L. edodes, and Dichomitus squalens were also not very homologous to that of Bjerkandera MnP (33, 35, 49 -52). The highest homology of the N-terminal amino acid sequence of Bjerkandera MnP is with those of both MnP-2 and LiP-7 from Trametes versicolor (53), with only 1 and 2 nonmatching amino acids in the first 25 residues, respectively. However, the manganese-independent substrate specificity of Trametes MnP isozymes has yet to be studied.
In conclusion, the MnP from Bjerkandera sp. strain BOS55 is a unique enzyme that can best be described as a hybrid between manganese peroxidase and lignin peroxidase. The catalytic efficiencies in Mn(II) and veratryl alcohol oxidation by this single isozyme are comparable to those of MnP and LiP, respectively, from other white rot fungi.