INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Except for cellulose, lignin is the most abundant biological compound found in nature, yet it is degraded by only a small number of microorganisms, primarily basidiomycetes (white rot fungi) (5, 12, 33). Lignin biodegradation by these fungi has obvious ecological significance and also has promising biotechnological applications (e.g., biopulping and wastewater treatment) (3, 38, 46). White rot fungi produce a variety of extracellular enzymes that are thought to be involved in lignin degradation, the best characterized of which are laccase, lignin peroxidase (LiP), and manganese peroxidase (MnP) (22). These enzymes are capable of forming radicals inside the lignin polymer, which results in destabilization of bonds and finally in the breakdown of the macromolecule (35). In many fungi, MnP is thought to play the crucial role in the primary attack on lignin, because it generates the strong oxidant Mn³⁺; this oxidant acts as a diffusible redox mediator which attacks certain aromatic moieties of the lignin polymer (6, 22, 61, 62). It has been proposed that another possible mechanism for primary attack on lignin is the mechanism of Pycnoporus cinnabarinus, which lacks MnP and uses laccase-mediated reactions to cleave lignin (11).

Most previous studies of the ligninolytic enzymes of white rot fungi have been carried out with liquid media (22, 47, 60), and there have been only a few studies in which the production and action of these enzymes have been investigated during solid-state fermentation (SSF). MnP and/or laccase was found during SSF of sawdust with Lentinus edodes, Ceriporiopsis subvermispora, or Phanerochaete chrysosporium and also in straw cultures of Pleurotus eryngii and Pleurotus ostreatus (7, 9, 15, 39, 41). Production of LiP during SSF has been observed only during growth of Phlebia radiata and P. chrysosporium on wheat straw and wood pulp, respectively (9, 56). Even less information is available about the formation of organic acids by white rot fungi during growth on lignocellulose. Dicarboxylic or -hydroxycarboxylic acids are required for MnP activity; these acids act as chelators both for Mn²⁺ and Mn³⁺ ions, and in addition, they serve as buffers (36, 62). For the part, the formation of such acids has been investigated with liquid cultures, and oxalic acid has been found to be the main organic acid produced by white rot fungi (37, 55). A number of white rot fungi were recently tested for organic acid production during SSF of wheat straw, and oxalic acid again was the main fungal metabolite of this type (17).

We have recently reported that MnP from the white rot fungi Nematoloma frowardii and P. radiata is capable of depolymerizing synthetic and natural lignin in a cell-free, malonate-buffered reaction system, which results in the formation of water-soluble lignin fragments and CO₂ (27, 28). Moreover, MnP from N. frowardii has also been found to mineralize a number of other substances, including humic acids and xenobiotic compounds (24, 25, 49). In the present study, we demonstrated that MnP is the predominant ligninolytic enzyme during SSF of wheat straw with N. frowardii and that this fungus produces sufficient amounts of organic acids so that effective MnP activity can occur.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Fungus. The agaric white rot fungus N. frowardii b19 (= DSM 11239 = ATCC 201144) was isolated from fruiting bodies on decaying Nothofagus wood in Bariloche, Argentina (23). Master cultures were subcultured on malt extract agar slants and maintained at 4°C until they were used.

Culture conditions. SSF was carried out in 250-ml flasks containing 15 g of chopped wheat straw, which was obtained from J. M. Pelayo (SAICA, Zaragoza, Spain). The manganese content of the straw was 11.4 mg kg¹ (204 µM), 70% of which was extractable with water (41a). The straw was sterilized twice by heating it at 121°C for 20 min, and 22.5 mg of glucose in 45 ml of deionized (filter-sterilized) water was added (3 ml of H₂O per g of straw). The flasks were inoculated with five agar plugs (diameter, 0.7 cm), closed with stoppers fitted with inlet and outlet tubes for aeration, and incubated at 24°C with a constant flow of water-saturated air (86 ml min¹). Control flasks were incubated without fungus under the same conditions. Fermented straw from three flasks was harvested every 2 days starting 6 days after inoculation and ending 25 days after inoculation.

Enzyme assays. MnP activity was determined by a modified method as described by Wariishi et al. (62). Each 1-ml (final volume) reaction mixture contained 50 mM sodium malonate (pH 4.5), 0.5 mM MnCl₂, 0.2 mM H₂O₂, and 5 to 50 µl of straw extract or purified enzyme preparation. The reaction was initiated at 25°C by adding H₂O₂, and the rate of Mn³⁺-malonate complex formation was monitored by measuring the increase in absorbance at 270 nm (₂₇₀ = 11,590 M cm¹). MnP activity was also measured by using 2,2'-azino-bis(3-ethylthiazoline-6-sulfonate) (ABTS) as the substrate under the conditions described above (26).

1

1

1

1

1

Enzyme purification. After straw extracts (ca. 330 ml from one culture flask) were thawed, they were centrifuged at 12,000 × g for 30 min to remove the precipitates. Then, they were concentrated to ca. 15 ml by ultrafiltration with a 250-ml filter unit equipped with a 10-kDa cutoff filter (Amicon, Beverly, Mass.). Subsequently, the samples were dialyzed by repeated washing with 25 mM acetate buffer (pH 5.5).

1

1

MnP characterization. Purified MnP was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (PAGE) and native isoelectric focusing (IEF)-PAGE (56). Discontinuous gel electrophoresis was carried out by using 10% gels and a Protean II apparatus (Bio-Rad, Richmond, Calif.). Proteins were visualized by silver staining with a Bio-Rad kit, and GIBCO/BRL low-range molecular weight standards were used to calibrate the gel.

1

Radioactive experiments. ¹⁴C-ring-labelled synthetic lignin (¹⁴C-DHP) with a molecular mass of 4 to 10 kDa was polymerized from ¹⁴C-ring-labelled coniferyl alcohol. The properties of this preparation, including the gel permeation chromatography properties, have been described previously (12, 58). The ¹⁴C-DHP was added as a DMF-water suspension (1:20) to 2.5 g of sterilized wheat straw in 100-ml culture flasks, and the final radioactivity was approximately 22,000 dpm. The flasks were each inoculated with three agar plugs, tightly closed with a rubber septum, and incubated at 24°C in the dark. The ¹⁴C-labelled volatile organic compounds and ¹⁴CO₂ that evolved were trapped weekly by bubbling any gas released through two sequential flasks containing Opti-Fluor and Carbosorb/Opti-Fluor (Packard Instrument B.V., Groningen, The Netherlands) (50). Pure oxygen was used for flushing. Radioactivity was measured by liquid scintillation counting with a model 1411 counter (Wallac Oy, Turku, Finland). At the end of cultivation, the straw cultures were each extracted with 20 ml of deionized water and then with 20 ml of dioxane, and radioactivity was detected in the supernatants. The residual straw, including the fungal mycelium, was combusted in a combustion chamber (Junitek Oy, Turku, Finland), and the trapped ¹⁴CO₂ was quantified. All of the results were expressed as means ± standard deviations based on three replicates.

High-performance liquid chromatography (HPLC). Organic acids in the straw extracts were analyzed by using a model HP 1090 liquid chromatograph (Hewlett-Packard, Waldbronn, Germany) equipped with an UltraSep ES FS column (Knauer, Gross-Umstadt, Germany) (24, 26). Phosphoric acid (10 mM) was used as the solvent at a flow rate of 0.55 ml min¹, and chromatograms were recorded at 210 nm. Authentic standards consisting of various organic acids were used for calibration. The concentrations of organic acids were expressed in relation to the initial water contents of the straw cultures.

CZE. Capillary zone electrophoresis (CZE) analysis was used to quantify oxalic acid in the straw extracts (17). This procedure was performed with a model HP 3D CE system equipped with a diode array detector (Hewlett-Packard). Indirect detection at 300 nm was used, and the reference wavelength was 200 nm. A fused silica capillary column (inside diameter, 50 µm; outside diameter, 360 µm; length, 80 cm) was purchased from Composite Metal Services Ltd. (Worcester, United Kingdom). The voltage applied was 25 kV, and the capillary temperature was maintained at 10°C. Samples were injected by applying 50 × 10⁵ mPa of pressure for 4 s. The buffer solution used was 5 mM potassium hydrogen phthalate supplemented with 0.5 mM cetyltrimethylammonium bromide (pH 6) (Hewlett-Packard). Peaks were identified by adding commercially available oxalic acid (17).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Enzyme activities. High levels of MnP activity and lower levels of laccase activity were detected in extracts of wheat straw fermented with N. frowardii. Figure 1 shows the time courses for laccase and MnP activities during SSF. Laccase activity appeared first, and the maximum level of activity was observed on day 10, after which the level of activity decreased rapidly, although low levels of laccase activity were detected until the end of incubation. MnP was first detected in the straw extracts after 10 days of fermentation, and a local maximum level of activity occurred on day 12. After this, the MnP activity stagnated, but the level of activity increased again after 16 days of incubation until the end of the experiment. When the laccase substrate ABTS was used as an additional substrate in the MnP assay, the levels of activity were always about 30% lower than the levels of activity obtained in the Mn²⁺ assay. In each case, however, the maximum level of MnP activity was considerably higher than the maximum level of laccase activity in the straw extracts; the maximum level of MnP activity, which was detected by monitoring the formation of Mn(III)-malonate complexes, was 5,600 U liter¹; the respective MnP activity level for the oxidation of ABTS was 3,700 U liter¹. Laccase reached a maximum activity level of 450 U liter¹ with ABTS as the substrate. LiP activity was not detected in the straw extracts by the veratryl alcohol method either because of the presence of inhibitors or because of color interference by aromatic straw depolymerization products or both. The same phenomenon was observed in previous studies in which lignocelluloses were used as growth substrates for white rot fungi (9, 56).

View larger version (14K): [in this window] [in a new window]   FIG. 1.   Activities of MnP () and laccase () during SSF of wheat straw with N. frowardii. Enzyme activities are expressed in relation to the water content of the straw (in units per liter). Each data point is the mean enzyme activity value for three extracted flasks; the bars indicate standard deviations.

Purification of MnP. The 23-day-old straw extracts were used to purify MnP, because they contained the highest levels of enzyme activity. Although colored degradation products from straw were partially removed by the ultrafiltration procedure, the concentrated samples had still an intense brownish color. A high level of MnP activity and a low level of laccase activity were observed in the concentrate, but LiP activity was not detected. During the purification procedure, most of the colored compounds were bound to the Sepharose-Q, but they were partially eluted from the column as the concentration of salt in the gradient increased. These compounds interfered with monitoring absorbance at both 280 and 409 nm, which resulted in a constant increase in absorbance during elution (Fig. 2). Thus, the protein profile had only one distinct peak (peak P1) at 280 nm, a corresponding heme peak (peak H1), and another small heme peak (peak H2). Our determination of the enzyme activities in the fractions revealed that MnP, laccase, and LiP were present (Fig. 2). No activity was found to be associated with peaks P1 and H1, but the small heme peak, peak H2, corresponded to a high level of MnP activity. In contrast to the results obtained with the concentrated crude extract, LiP activity was detected after purification on Sepharose-Q at a level similar to the laccase activity level; nevertheless, the level of LiP activity was low compared with the level of MnP activity. Removing the colored straw degradation products during anion-exchange chromatography probably made detection of LiP by the veratryl alcohol method possible.

View larger version (23K): [in this window] [in a new window]   FIG. 2.   Separation of extracellular proteins from 23-day-old cultures of N. frowardii grown on wheat straw by anion-exchange chromatography on Sepharose-Q. (A) ---, absorbance at 280 nm (OD [280 nm]); , absorbance at 409 nm (OD [409 nm]); ..., salt gradient (0.05 to 0.30 M NaCl). (B) Enzyme activities. Lacc, laccase.

View larger version (24K): [in this window] [in a new window]   FIG. 3.   Purification of pooled ligninolytic activities on a Mono-Q Sepharose column. (A) ---, absorbance at 409 nm (OD [409 nm]); , salt gradient. (B) Enzyme activities. Lacc, laccase.

Enzyme characterization. Purified SSF MnP was analyzed by several gel techniques, and its molecular mass and pI were determined. The enzyme produced a single band on the sodium dodecyl sulfate-PAGE gel at a molecular mass of 50 kDa, a value that is higher than the molecular masses of MnP from liquid cultures (42 to 44 kDa). Analysis of IEF-PAGE gels revealed a single, relatively broad band for SSF MnP when both ABTS and phenol red staining were used (Fig. 4). The pI of SSF MnP (3.0 to 3.2) was nearly identical to the pI of MnP2 purified from liquid cultures of N. frowardii (3.1 to 3.3), whereas the pI of MnP1 was slightly higher (3.8 to 4.0) (51).

View larger version (58K): [in this window] [in a new window]   FIG. 4.   IEF analysis of MnP separated from N. frowardii during SSF of wheat straw (lanes 1 and 2) and IEF analysis of MnP2 (lane 3) and MnP1 (lane 4) obtained from liquid cultures (51). MnP activity was stained with phenol red (lane 1) or ABTS (lanes 2 through 4) in the presence of Mn(II) and H2O2.

Production of organic acids during SSF. During fermentation of wheat straw, the pH of the extracts decreased from 5.5 to 4.0, indicating that organic acids were produced by the fungus. HPLC analysis revealed the presence of high levels of malate and fumarate. Figure 5 shows the time courses for malate, fumarate, and oxalate concentrations in relation to the water content of the fermented wheat straw. Malate and fumarate were detected after 8 days of incubation prior to production of MnP, and the maximum malate and fumarate concentrations (45 mM malate, 3.5 mM fumarate) were detected on day 10. The concentrations of these compounds decreased rapidly after MnP appeared, but they were detected until the end of the experiment on day 25. The decreases in concentrations were probably due to MnP-Mn³⁺-catalyzed decomposition of malate and fumarate, similar to the decarboxylation reactions that have been described for oxalate, glyoxylate, and malonate (26, 55). The HPLC analyses revealed that oxalate was formed in addition to malate and fumarate; however, the presence of other compounds, probably originating from straw, prevented a quantitative analysis. Therefore, CZE was used to analyze oxalic acid in the straw samples. Oxalic acid appeared later than malate and fumarate, on day 12, and the maximum concentration of oxalic acid (10 mM) was observed on day 16 (Fig. 5). After this, the oxalate concentration decreased rapidly, probably due to the high level of MnP activity.

View larger version (18K): [in this window] [in a new window]   FIG. 5.   Accumulation of malate (), fumarate (), and oxalate () during SSF of wheat straw with N. frowardii. Millimolar concentrations are expressed in relation to the water content of straw. Each data point is the mean organic acid level for three extracted culture flasks; the bars indicate standard deviations.

Mineralization of ¹⁴C-DHP during growth of N. frowardii on wheat straw. ¹⁴C-DHP was effectively mineralized during SSF of wheat straw with N. frowardii. Figure 6 shows the time course of ¹⁴CO₂ evolution over a period of 12 weeks. Mineralization started after 1 week, the maximum rate of mineralization (1.7% ¹⁴CO₂ per day) was reached rapidly after 2 weeks of incubation, and then the rate was nearly constant for the next 3 weeks, after which the rate decreased slowly. Interestingly, the rate of ¹⁴C-DHP mineralization increased simultaneously with the production of MnP and the decrease in accumulated organic acid contents (Fig. 1, 5, and 6), indicating that MnP may be involved in the mineralization process. The balance of radioactivity at the end of incubation showed that ca. 75% of the ¹⁴C-DHP was converted into ¹⁴CO₂ and 13% was converted into ¹⁴C-labelled water-soluble compounds; 4% of the radioactivity was extractable with dioxane representing nonconverted ¹⁴C-DHP, and 6% was detected as ¹⁴CO₂ after combustion of the residual straw, including the well-developed fungal biomass (Table 1).

View larger version (13K): [in this window] [in a new window]   FIG. 6.   Mineralization of 14C-ring-labelled synthetic lignin (22,000 dpm) by N. frowardii during growth on wheat straw (). , noninoculated control. The bars indicate standard deviations (n = 3).

Mineralization and solubilization of ¹⁴C-DHP by MnP. An in vitro reaction system was designed to simulate the conditions in the straw cultures of N. frowardii on the 12th day of cultivation. Using comparable concentrations of malate and fumarate, which were detected during SSF, we examined the ability of purified MnP to mineralize and solubilize ¹⁴C-DHP in a cell-free system. Figure 7 shows the time course of ¹⁴CO₂ evolution from ¹⁴C-DHP due to purified MnP in malate- or malate-fumarate-containing reaction systems in the absence of external H₂O₂. About 7.5% of the initial radioactivity was released as ¹⁴CO₂ in the malate-fumarate-containing reaction mixture within 16 days, whereas when malate was used alone, only 6% of the ¹⁴C-DHP was converted into ¹⁴CO₂. The double bond in the fumarate molecule probably made the formation of radicals possible, which stimulated lignin degradation in a way similar to the way that has been postulated for radicals derived from unsaturated fatty acids (29). The extent of mineralization was relatively low during the first 2 days of incubation but then increased considerably, and mineralization occurred until the end of the experiment. Controls released less than 0.5% ¹⁴CO₂. The maximum rate of in vitro mineralization (ca. 1.1% ¹⁴CO₂ per day) was in the same range as the maximum rate in the fungal straw cultures (1.7% ¹⁴CO₂ per day).

View larger version (14K): [in this window] [in a new window]   FIG. 7.   Production of 14CO2 from 14C-DHP by purified N. frowardii MnP in a cell-free system. Each reaction mixture contained 2 U of MnP, 30 mM malate, 5 mM fumarate, 1 mM MnCl2, 100 mg of unlabelled DHP per liter, and 22,000 dpm of 14C-DHP and was incubated on a rotary shaker at 37°C in the dark. The data points are means of three replicates, and the bars indicate standard deviations. Symbols: , complete reaction mixture; , reaction mixture without fumarate; , control without MnP.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The white rot fungus N. frowardii produced MnP as the predominant ligninolytic enzyme during SSF of wheat straw, and by using comparable concentrations of organic acids excreted by the fungus, purified MnP was able to mineralize and solubilize ¹⁴C-DHP in a cell-free system. Thus, we demonstrated for the first time that production of MnP and organic acids is directly connected with mineralization of lignin.

Lignocellulose contains high levels of manganese (Mn), which, after calcium, potassium, and magnesium, is the most abundant metal. Mn concentrations up to 150 ppm have been detected in several soft- and hardwoods, and up to 50 ppm of Mn has been detected in wheat straw (13, 44). White rot fungi were found to accumulate Mn as MnO₂ during growth on lignocellulose in black regions and flecks that contained more than 100-fold more Mn than sound lignocellulose (4).

MnP activities have been found during SSF of different lignocelluloses (wood, pulp, straw) with white rot fungi. Thus, the corticioid and polyporoid white rot fungi P. chrysosporium, Rigidoporus lignosus, and C. subvermispora grown on wood chips or sawdust produced multiple forms of MnP (9, 16, 41). MnP was also found to be the main ligninolytic enzyme during treatment of kraft pulp with Trametes versicolor, although laccase activity was also present (46, 48). Other authors investigated the activities of ligninolytic enzymes of three white rot fungi (P. chrysosporium, T. versicolor, Coriolopsis polyzona) during growth on wheat straw (59). MnP activity was the dominant enzyme activity during the initial phase of incubation, and the activity profiles of MnP were similar in all three fungi. Five species of the genus Pleurotus produced MnP and laccase as the major lignin-degrading enzymes when they were grown on straw under SSF conditions, and the maximum level of MnP activity was 4 to 10 times higher than the maximum level of laccase activity (7). Later it was demonstrated that in addition to MnP, LiP is secreted during SSF of wheat straw (56), an observation which we also made with N. frowardii, although the level of LiP activity was comparatively low. In their study, Vares et al. used P. radiata, a wood-rotting fungus that belongs to the same family of basidiomycetes (Meruliaceae) as the most investigated white rot fungus, P. chrysosporium (19). The P. radiata MnP isozymes MnPb and MnPa purified from wheat straw cultures each had a molecular mass of 50 kDa and had pI values of 3.4 to 3.9 and 4.9 to 5.3, respectively (56). Unlike P. radiata, our agaric fungus, N. frowardii (a member of the family Strophariaceae), produced only one MnP isozyme, which had a molecular mass of 50 kDa and a relatively low pI (3.0 to 3.2).

Production of organic acids, which are thought to be mediators of ligninolytic enzymes (in particular, chelators of Mn³⁺ generated by MnP), by wood-rotting basidiomycetes has been investigated previously, but most previous studies have been performed with liquid cultures (2, 10, 54) and information about the conditions in lignocellulose is limited (53). In all cases, accumulation of oxalic acid was reported, but only Takao also observed the formation of substantial amounts of other organic acids (malate, fumarate, succinate) by using CaCO₃-containing shake cultures of a number of white and brown rot fungi (54). As far as we know, only one report of production of organic acids in a natural substrate of ligninolytic fungi has been published (17). Glakin et al. (17) demonstrated that oxalate was produced during SSF of wheat straw with different white rot fungi (e.g., P. radiata, P. chrysosporium, and C. subvermispora). Due to analytical difficulties (only CZE was used), other organic acids were not detected in the straw extracts. In the present study, this problem was overcome by using an HPLC method for detection of organic acids other than oxalate (24, 26). Using this method, we showed for the first time that high levels of malate and fumarate are produced by a white rot fungus during SSF.

Effective mineralization and solubilization of lignin by white-rot fungi have been demonstrated by using both natural and synthetic ¹⁴C-labelled lignins and lignocelluloses (6, 18, 21, 31, 32, 57). So far, the highest level of mineralization of a ¹⁴C-labelled lignin was observed with P. radiata, which released up to 71% ¹⁴CO₂ from ¹⁴C-DHP when it was grown in a liquid medium (20). Interestingly, only 3% of the radioactivity was associated with the fungal biomass after 40 days of cultivation. Similar results were obtained with natural ¹⁴C-labelled lignins (e.g., lignins from fir or oak; 58 to 61% mineralization; 12 to 13% ¹⁴C in the mycelium) (20). Our results obtained with N. frowardii confirmed these results and revealed an even higher level of mineralization (75% of the ¹⁴C-DHP) during SSF, while also only a small percentage of the initial radioactivity (6%) was incorporated into the residual straw and the fungal biomass. The rate of lignin mineralization during SSF slowed down later than in liquid cultures, and substantial ¹⁴CO₂ evolution was observed until the end of cultivation on day 80.

Given the assumption that ¹⁴C-labelled organic substances are normally converted intracellularly into ¹⁴CO₂, the incorporation of such a small amount of radioactivity into the biomass is remarkable. In connection with our other findings, the label distribution provides an additional indication that at least some extracellular mineralization of lignin occurs. A low but significant level of mineralization of water-soluble lignin fragments by filter-sterilized P. chrysosporium culture fluids was reported by Boyle et. al. (6). These authors concluded that certain extracellular enzymes might be responsible for this mineralization. We recently demonstrated that crude MnP and purified MnP from N. frowardii are in fact capable of converting lignin and other aromatic and aliphatic compounds to CO₂ in a cell-free reaction system (24-28). The cell-free reaction system routinely contained malonic acid, which is known to be the optimal chelator for Mn³⁺ formed by MnP (1, 62). Malonic acid, however, was not detected in the cultures of N. frowardii and was secreted only in trace amounts by other white rot fungi (62). Our present results show that malonate can be successfully replaced by the fungal metabolite malate (or malate-fumarate) during direct mineralization of lignin by MnP.

The formation of carboxylic groups or related structures from aromatic rings and the subsequent decarboxylation of these structures by Mn³⁺ are probably the basis for the MnP-catalyzed mineralization of aromatic compounds (24, 52). It has been reported that phenanthrene and veratryl alcohol are converted to a biphenyl dicarboxylic acid and a lactone, respectively, by MnP (8, 43). Moreover, reactive radicals (e.g., superoxide, carbon-centered radicals, and peroxyl radical), which are formed from organic acids by MnP, might also be involved in the mineralization process (26, 30). Furthermore, we propose that LiP may support the whole degradation process by cleaving bonds of recalcitrant lignin structures (e.g., ether bridges between nonphenolic lignin moieties).

On the basis of the present results, we propose that certain white rot fungi are able to mineralize lignin extracellularly; this proposal does not rule out the possibility that a substantial amount of lignin is also converted intracellularly into CO₂. Future investigations will have to clarify to what extent extracellular mineralization occurs under natural conditions and whether similar systems have developed in other wood-rotting fungi.