Reactive Oxygen Species and Induction of Lignin Peroxidase in Phanerochaete chrysosporium

Abstract

We studied oxidative stress in lignin peroxidase (LIP)-producing cultures (cultures flushed with pure O₂) of Phanerochaete chrysosporium by comparing levels of reactive oxygen species (ROS), cumulative oxidative damage, and antioxidant enzymes with those found in non-LIP-producing cultures (cultures grown with free exchange of atmospheric air [control cultures]). A significant increase in the intracellular peroxide concentration and the degree of oxidative damage to macromolecules, e.g., DNA, lipids, and proteins, was observed when the fungus was exposed to pure O₂ gas. The specific activities of manganese superoxide dismutase, catalase, glutathione reductase, and glutathione peroxidase and the consumption of glutathione were all higher in cultures exposed to pure O₂ (oxygenated cultures) than in cultures grown with atmospheric air. Significantly higher gene expression of the LIP-H2 isozyme occurred in the oxygenated cultures. A hydroxyl radical scavenger, dimethyl sulfoxide (50 mM), added to the culture every 12 h, completely abolished LIP expression at the mRNA and protein levels. This effect was confirmed by in situ generation of hydroxyl radicals via the Fenton reaction, which significantly enhanced LIP expression. The level of intracellular cyclic AMP (cAMP) was correlated with the starvation conditions regardless of the oxygenation regimen applied, and similar cAMP levels were obtained at high O₂ concentrations and in cultures grown with atmospheric air. These results suggest that even though cAMP is a prerequisite for LIP expression, high levels of ROS, preferentially hydroxyl radicals, are required to trigger LIP synthesis. Thus, the induction of LIP expression by O₂ is at least partially mediated by the intracellular ROS.

The white rot fungus Phanerochaete chrysosporium can degrade and metabolize lignin and a broad range of recalcitrant organopollutants (14, 34). Lignin depolymerization is achieved primarily by one-electron oxidation reactions catalyzed by extracellular oxidases and peroxidases in the presence of extracellular hydrogen peroxide (H₂O₂). Lignin peroxidase (LIP) is an enzyme commonly associated with the extracellular degradation of lignin by P. chrysosporium. It oxidizes phenols to phenoxy radicals and nonphenolic aromatics to radical cations (6, 21). Liquid cultures of P. chrysosporium must be starved (nitrogen or carbon depletion) and exposed to a pure O₂ atmosphere to trigger LIP expression (4, 11, 26). Boominathan and Reddy (7) showed that transcription of genes encoding some LIP enzymes is controlled by cyclic AMP (cAMP) levels.

Increasing oxygen availability to liquid cultures of P. chrysosporium increases LIP levels (35, 36). The high O₂ concentration presumably leads to increased production of reactive oxygen species (ROS) relative to that during normal metabolism, subjecting the fungus to a ROS-rich environment, in addition to the toxic radical intermediates produced in the ligninolytic phase. Zacchi et al. (44-46) suggested that cultures of P. chrysosporium exposed to O₂ to trigger LIP synthesis are subjected to oxygen toxicity, which leads to disorganization of the cellular ultrastructure and chlamydospore development, probably in response to accumulating ROS. LIP activities were similar in cultures of P. chrysosporium exposed to pure O₂ gas (oxygenated cultures) and in manganese-deficient cultures grown with atmospheric air (36). No manganese-containing superoxide dismutase (MnSOD) activity was detected in the latter cultures, suggesting that both cultures are saturated by the ROS needed to trigger LIP expression. The level of transcription of manganese peroxidase in cultures of P. chrysosporium was reported to correlate with the concentration of exogenous hydrogen peroxide added in the absence or presence of Mn²⁺ (28). The transcription of versatile peroxidase in liquid cultures of Pleurotus eryngii was also found to correlate with the addition of exogenous hydrogen peroxide or hydroxyl radicals (38).

Protection against the deleterious effects of ROS has been analyzed in depth in yeasts, including Saccharomyces cerevisiae (19), Schizosaccharomyces pombe (24), Candida albicans (20), and Hansenula mrakii (42). The filamentous fungus Penicillium chrysogenum is also highly resistant to oxidative stress caused by high concentrations of peroxides and superoxide anions, which are attributed to high levels of catalase and glutathione peroxidase activity (12, 13). Catalase activity was examined in different strains of P. chrysosporium under various growth conditions (23, 29). In low-nitrogen cultures (i.e., excess glucose), production of catalase preceded that of LIP by approximately 3 days, whereas in low-carbon cultures, catalase was produced at an even higher level, but LIP activity was not detected during the incubation period (23). The specific activities of both catalase and SOD decreased at the end of the primary growth phase before idiophase in liquid cultures of P. chrysosporium (CMI 174727) (30).

LIP production in liquid cultures of P. chrysosporium occurs only when the cultures are flushed with pure O₂. The objective of this study was to determine the mode of action by which this high O₂ requirement triggers LIP synthesis. Our working hypothesis was that ROS induce LIP gene expression. The results of this work provide new insights in the understanding of the signal transduction mechanisms triggering LIP synthesis in response to high oxygen levels in P. chrysosporium in particular and in white rot fungi in general.

MATERIALS AND METHODS

Reagents.

Hematoxylin, 5,5′-dithiobis-2-nitrobenzoic acid (DTNB), NADP (NADPH), veratryl alcohol, phenylmethanesulfonyl fluoride (PMSF), reduced glutathione (GSH), oxidized glutathione (GSSG), glutathione reductase, 2,4-dinitrophenylhydrazine (DNPH), p-nitrophenylphosphate, dimethyl sulfoxide (DMSO), mannitol, MnSOD from Escherichia coli, calf thymus DNA, N-methyl-2-phenylindole, and bovine serum albumin were purchased from Sigma Chemical Co. (St. Louis, Mo.). 1-Methyl-2-vinylpyridinium trifluoromethanesulfonate was purchased from Bioxytech (Oxis International Inc., Portland, Oreg.). Aldehyde reactive probe (ARP)-biotinylated alkoxyamine, and 2′,7′-dichlorodihydrofluorescein diacetate were from Molecular Probes (Eugene, Oreg.). Vectastain ABC-alkaline phosphatase reagent was from Vector Laboratories (Burlingame, Calif.), and dihydroethidium was from Fluka (Buchs, Switzerland).

Fungal strain and culture conditions.

The filamentous fungus P. chrysosporium Burds BKM-F-1767 (ATCC 24725) was used for this study. The fungus was maintained at 4°C on 2% (wt/vol) malt extract agar slants and inoculated by the method of Tien and Kirk (41). The growth medium was prepared as previously described (35, 36) with initial concentrations of glucose, diammonium tartrate, and MnSO₄ · H₂O of 56, 2.4, and 0.225 mM, respectively. The fungus was grown in submerged liquid cultures (90 ml) at 175 rpm and at 37°C in 250-ml flasks sealed with rubber stoppers, and the headspace was flushed twice a day with O₂ for 2 min at a flow rate of 1 liter/min (oxygenated cultures). Oxygen gas was of medical-grade purity. Cultures grown with free exchange of atmospheric air (flasks sealed with dense paper plugs) (aerated cultures) served as controls. Unless otherwise specified, the cultures were incubated for 5 days. For determination of LIP activity, ROS concentration, macromolecule damage, and antioxidant enzyme activities, the entire contents of flasks were taken after 24, 48, 72, 96, and 120 h of culture.

To determine the effect of hydroxyl radical (OH^· ) on LIP synthesis, the fungus was grown under the same conditions as described above, in oxygenated or aerated cultures with or without the addition of 50 mM DMSO every 12 h for 120 h (27). The effect of DMSO on cell viability was measured by inoculating mycelial plugs onto agar plates containing the growth medium supplemented with various concentrations of DMSO and comparing the amount of radial growth. For determination of LIP expression (activity, transcription, and heme protein analyses), the entire contents of flasks were taken after 96 and 120 h of culture.

The direct effect of OH^· on LIP expression was studied through the in situ generation of OH^· via the Fenton reaction with Fe(II)-EDTA (0.1 mM) and H₂O₂ (0.5 mM). The Fenton reagents were added separately or together to cultures grown for 96 h with free exchange of atmospheric air, and the cultures were incubated for an additional hour. Cultures that were incubated without Fenton reagents or OH^· scavengers served as controls. Cultures incubated with Fenton reagents in the presence of either 50 mM DMSO or 50 mM mannitol served as negative controls. After the additional incubation, the sample flasks were harvested, and LIP transcription was then determined.

ROS determination.

Intracellular ROS concentrations were determined in liquid cultures of P. chrysosporium at tropophase and idiophase. H₂O₂ and superoxide anion (O₂^·−) levels were detected by adding 2′,7′-dichlorodihydrofluorescein diacetate (13, 37) and dihydroethidium (9, 13), respectively, to the cultures. These compounds were added to the cultures at a final concentration of 30 μM and incubated for 30 min at 37°C and 250 rpm. The mycelia were then harvested, washed with double-distilled water, and treated with 2 ml of 5% 5-sulfosalicylic acid (vol/vol) for 20 min at 4°C. The treated mycelia were centrifuged at 20,000 × g for 15 min at 4°C. The production of 2′,7′-dichlorofluorescein (a fluorescent indicator of peroxide) and ethidium (a fluorescent indicator of superoxide) was measured in the extracellular medium and in the cellular extract with a fluorescence spectrophotometer (F-2000; Hitachi, Tokyo, Japan). To detect 2′,7′-dichlorofluorescein, the extinction and emission wavelengths were 520 ± 6 and 590 ± 10 nm, respectively. To detect ethidium, the extinction and emission wavelengths were 501 and 521 nm, respectively.

Cell extract preparations.

Pellets were separated from extracellular fluid by filtration through 200-μm-pore-size nylon mesh (Amiad Filtration Systems, Amiad, Israel). Extracellular fluid was used for LIP activity measurements and the fractionation of heme proteins.

Cell extracts were prepared differently for the four different aims.

(i) Determination of oxidative damage of macromolecules, intracellular enzymatic activities, and protein content.

Samples were homogenized in the cold for 2 min in 50 mM phosphate buffer, pH 7, with an Ultra-Turrax T25 homogenizer (IKA, Staufen, Germany). The homogenate was centrifuged at 20,000 × g for 20 min at 4°C. PMSF (1 mM) was added to each sample at the time of homogenization.

(ii) Measurement of intracellular cAMP.

EDTA (4 mM) in 2 ml of 5% (vol/vol) perchloric acid was added to the samples before homogenization to prevent enzymatic degradation of cAMP. Homogenization and centrifugation were performed as described above.

(iii) Measurement of GSH and GSSG content.

Two milliliters of 5% (vol/vol) perchloric acid was added to each sample before homogenization. The homogenates were centrifuged at 20,000 × g for 15 min at 4°C.

(iv) RNA and DNA isolation.

Frozen biomass (frozen by treatment with liquid nitrogen) was ground with a mortar and pestle. The cell extract was separated from cell debris by centrifugation at 20,000 × g for 20 min at 4°C.

Measurements of oxidative damage. (i) Oxidized lipids.

The levels of oxidized lipids were detected with a lipid peroxidation assay kit (catalog no. 437634; Calbiochem, San Diego, Calif.) following the manufacturer's instructions. This kit detects a chromogen that absorbs 586-nm-wavelength light and is formed by the reaction between malondialdehyde (MDA), an end product of the peroxidation of polyunsaturated fatty acids, and N-methyl-2-phenylindole.

(ii) Oxidized proteins.

The levels of oxidized proteins were monitored by measuring carbonyl accumulation, which serves as an early marker for metal-catalyzed protein oxidation (40). Redox cycling cations, e.g., Fe²⁺, can bind to cation-binding sites on proteins and, when combined with H₂O₂, transform side chain amine groups on some amino acids into carbonyl groups. Protein carbonyl content (PCC) in the cellular extracts was measured in an enzyme-linked immunosorbent assay (ELISA), a modified version of the method described by Buss et al. (8). The hydrazones obtained from derivatization of proteins with DNPH, were reacted with biotinylated anti-DNPH and quantified using Vectastain ABC-alkaline phosphatase reagent instead of streptavidin-biotinylated horseradish peroxidase. A standard curve representing 0 to 4 nmol of PCC/mg of protein was prepared with oxidized bovine serum albumin (8).

(iii) Oxidized DNA.

The principal damage to DNA produced by ROS is the loss of a base and the production of abasic (AP) sites. Oxidative damage of DNA molecules was measured as the number of AP sites/10⁴ bp by an ELISA-like assay (22). A biotinylated aldehyde-specific reagent (ARP-biotinylated alkoxyamine) reacted specifically with the aldehyde group present in the open ring in AP sites, and the amount of biotinylated AP sites was quantified using Vectastain ABC-alkaline phosphatase reagent. Calf thymus DNA containing 1 to 5 AP sites/10⁴ bp, was used for the standard calibration curve. DNA was extracted by a modified version of the rapid method of Raeder and Broda (33).

GSH/GSSG ratio.

The GSH/GSSG molar ratio was determined by using the Bioxytech GSH:GSSG-412 assay kit (catalog no. 21040) according to the manufacturer's instructions with slight modifications. This method employs DTNB, which reacts with GSH to form 2-nitro-5-thiobenzoate anion (TNB²⁻), a product which can be detected spectrophotometrically with 412-nm-wavelength light with an extinction coefficient of 14,150 M⁻¹ cm⁻¹. The concentration of GSH was calculated from a calibration curve (0 to 3 μM GSH). The assay was conducted on two parallel sets of samples. One set of samples, preincubated with gluthathione reductase, gave total glutathione (reduced and oxidized) as GSH. The second set, used for GSSG determination, was first treated with the thiol-scavenging reagent 1-methyl-2-vinylpyridinium trifluoromethanesulfonate to scavenge GSH. The remaining GSSG was then reduced to GSH with NADPH catalyzed by glutathione reductase, and the GSH formed was measured. GSSG at concentrations from 0 to 1.5 μM was used as the standard to calibrate the curves. The GSH/GSSG ratio was calculated as follows: GSH/GSSG = [GSH − (2 GSSG)]/GSSG.

Determination of antioxidant enzyme activities. (i) MnSOD activity.

The SOD activity assay (5) was based on the SOD-mediated increase in the rate of auto-oxidation of hematoxylin in aqueous alkaline solution, which yields a chromophore with maximum absorbance at 560 nm (29). The change in the absorbance of the mixture at 560 nm, which measured the auto-oxidation of hematoxylin, was monitored for 1 min.

The activity of the enzyme in each sample was calculated as the ratio between the reaction rates of the auto-oxidation of hematoxylin in the presence of sample (V_sample) and in the absence of sample (V_control). V_sample/V_control was translated to units of activity (units per milliliter) by means of a calibration curve prepared with 0 to 400 U of purified MnSOD from E. coli per ml. Actual MnSOD activity was expressed as relative units.

CuZnSOD activity is sensitive to both H₂O₂ and cyanide. In contrast, MnSOD activity is not inhibited by H₂O₂ or cyanide. To distinguish MnSOD from CuZnSOD, MnSOD activity was assayed in the presence of 5 mM KCN in all cases.

(ii) Catalase activity.

Catalase activity was assayed by measuring the degradation of H₂O₂. The rate of disappearance of H₂O₂ was monitored by measuring A₂₄₀ (25). One unit of catalase decomposed 1 μmol of H₂O₂ in 1 min (ɛH₂O₂ = 39.4 M⁻¹ cm⁻¹).

(iii) Glutathione reductase activity.

Glutathione reductase activity was determined by monitoring the rate of production of TNB²⁻ at 412 nm. TNB²⁻ is formed by the reduction of DTNB by GSH, which in turn is formed by the reduction of GSSG by glutathione reductase (25). One unit of enzyme activity was defined as the production of 1 μmol of TNB²⁻ per min (ɛTNB2⁻ = 14,150 M⁻¹ cm⁻¹).

(iv) Glutathione peroxidase activity.

Glutathione peroxidase activity was assayed by the method of Chiu et al. (10) with slight modifications. One unit of enzyme activity was defined as the oxidation of 1 μmol of NADPH per min (ɛ_NADPH = 6,220 M⁻¹ cm⁻¹).

cAMP assay.

Intracellular cAMP was measured with the ³H-labeled cAMP assay kit (catalog no.TRK 432; Amersham, Little Chalfont, Buckinghamshire, United Kingdom). The cAMP assay is based on the competition between unlabeled cAMP and a fixed quantity of the tritium-labeled compound for binding to a protein which has a high specificity and affinity for cAMP. The amount of labeled protein-cAMP complex formed is inversely related to the amount of unlabeled cAMP present in the assay sample.

Expression of LIP. (i) LIP activity.

Enzyme activity was measured in the extracellular fluid by the method of Tien and Kirk (41). One unit of activity was defined as 1 μmol of veratryl alcohol oxidized to veratryl aldehyde per min (ɛ_{veratryl aldehyde} = 9,300 M⁻¹ cm⁻¹).

(ii) LIP transcription.

Total RNA was isolated from P. chrysosporium cell cultures with the Tri reagent (Sigma), according to the manufacturer's instructions. LIP mRNA levels were measured by reverse transcription-PCR (RT-PCR) and compared to P. chrysosporium 18S rRNA levels (1). We measured only the mRNA level of isozyme LIP-H2. First-strand cDNA synthesis and the amplification reaction were performed in the same tube (Access RT-PCR System; Promega, Madison, Wis.). The 50-μl reaction mixture contained avian myeloblastosis virus (AMV) RT/Tfl reaction buffer, 100 μg of total RNA, 0.2 mM concentration (total) of a mixture of the four deoxynucleoside triphosphates, 5 U of AMV reverse transcriptase, and 5 U of the thermostable Tfl DNA polymerase. The mixture was incubated at 48°C for 45 min for first-strand cDNA synthesis and then heated to 94°C for 2 min to inactivate the reverse transcriptase. The primer 5′-AGAGCGCCGTGAAGATGAACTGGA-3′ was used for the first-strand cDNA synthesis of LIP-H2, and the primer 5′-CAACTACGAGCTTTTTAACTGC-3′ was used for the first-strand cDNA synthesis of 18S rRNA.

Second-strand cDNA synthesis and PCR amplification with primers specific for LIP-H2 (encoded by the lip gene deposited in GenBank under accession number X15599) (antisense [5′-TCAGGTTCTCGCTCGCATGCT-3′] and sense [5′-AGAGCGCCGTGAAGATGAACTGGA-3′]) were performed in a minicycler (MJ Research, Watertown, Mass.) under the following conditions: 40 cycles of PCR amplification, with 1 cycle consisting of 30 s at 94°C, 1 min at 62°C, and 2 min at 68°C. 18S cDNA was amplified in a similar manner, except that 10 cycles and a temperature of 55°C for annealing were used. Two 18S rRNA-specific primers (antisense [5′-CAAATTACCCAATCCCGACAC-3′] and sense [5′-CAACTACGAGCTTTTTAACTGC-3′]) were used. Forty cycles of amplification of the LIP cDNA and 10 cycles of amplification of the 18S cDNA produced enough PCR products for semiquantitative analysis, while maintaining the linear relationship between the concentrations of both transcripts. LIP-H2 RNA and 18S rRNA levels were precalibrated and diluted to levels below the saturation levels after PCR analysis. The short cycling of the rRNA amplicons was not at saturation levels. The precalibration identified plateau cycle levels. The PCR products (5 μl) were electrophoresed on a 2% agarose gel at 100 V for about 45 min. All reaction mixtures were electrophoresed on the same gel. DNA was stained with 30 μg of ethidium bromide (Sigma) in 100 ml of distilled water and then imaged and analyzed with a Fluor-S MultiImager (Bio-Rad, Hercules, Calif.).

(iii) LIP heme protein analysis.

LIP heme proteins were analyzed as previously reported (36). Heme proteins (isozymes H1 to H10) were identified on the basis of elution properties and the results of activity tests.

RESULTS

ROS level.

Mycelia produced in cultures treated with oxygen grew and utilized glucose at a rate similar to that of cultures grown with atmospheric air (not shown).

In cultures flushed with pure O₂, intracellular H₂O₂ concentration increased gradually with culture age, reaching a maximum during idiophase (almost 10-fold higher than the initial level), but remained stable during the entire incubation period in the aerated control cultures, with only a slight increase of less than twofold (Fig. 1). In contrast, O₂^·− levels were highest at tropophase and decreased during the transition to idiophase in aerated and oxygenated cultures (Fig. 1).

FIG. 1.

Levels of hydrogen peroxide and superoxide anion in oxygenated and aerated cultures of P. chrysosporium as a function of culture age. Oxygenation was performed by flushing the headspace of the culture flasks with pure oxygen gas for 2 min at a flow rate of 1 liter/min. Aerated cultures are cultures exposed to free exchange of air. The levels of hydrogen peroxide in oxygenated cultures (•) and aerated cultures (○) and the levels of O₂^·− in oxygenated cultures (▴) and aerated cultures (▵) were determined by fluorometric detection of the oxidation products of 2′,7′-dichlorodihydrofluorescein diacetate and dihydroethidium, respectively. FU₅₂₁, fluorescence units at 521 nm; FU₆₀₀, fluorescence units at 600 nm. The means ± standard deviations (error bars) of eight replicates are shown.

Antioxidative enzyme activity.

ROS measurements alone are insufficient to determine the existence of oxidative stress, probably because of the short half-lives of ROS and the fact that the measurements taken are of steady-state concentrations. Coupling ROS level with the change in activity of antioxidant enzymes and cumulative oxidative damage of macromolecules gives a better indication of the existence of stress. MnSOD, but not CuZnSOD, activity was detected in fungal cell extracts, since no change in activity was observed in the presence of KCN as described by Belinky et al. (5). MnSOD activity increased at similar rates in oxygenated and aerated cultures during the first 96 h of incubation (Fig. 2). Thereafter, it increased approximately 1 order of magnitude in the oxygenated cultures but remained constant in the aerated cultures. Catalase activity peaked at 96 and 72 h in the oxygenated and aerated cultures, respectively, although the peak was twice as high for oxygenated cultures (Fig. 2).

FIG. 2.

MnSOD and catalase activities in cell extracts of P. chrysosporium as a function of culture age from oxygenated and aerated cultures. SOD activity (measured in relative units [rU]) in oxygenated cultures (▴) and aerated cultures (▵) was assayed by the auto-oxidation of hematoxylin in the presence of 5 mM KCN. Catalase activity in oxygenated cultures (•) and aerated cultures (○) was determined by monitoring the degradation of H₂O₂ as measured by the absorbance at 240 nm. The means ± standard deviations (error bars) of eight replicates are shown.

Glutathione peroxidase activities were similar during tropophase in oxygenated and aerated cultures (2 to 3 U/g) but increased to 15 U/g at idiophase in the oxygenated cultures compared to approximately 6 U/g in the aerated cultures (Table 1). In contrast, glutathione reductase activities were similar in both cultures during tropophase and idiophase (Table 1). Although glutathione reductase activity hardly changed over time, its level was 10-to 85-fold higher than that of glutathione peroxidase. The GSH/GSSG ratio was significantly lower in oxygenated cultures than in aerated cultures and decreased from tropophase to idiophase in both cultures but exhibited a more pronounced decline in oxygenated cultures (Table 1).

TABLE 1.

Glutathione-related functions in cell extracts of P. chrysosporium^a

Culture conditions	Enzyme activity (U/g of protein)		GSH/GSSG ratio
	Glutathione peroxidase	Glutathione reductase
Incubated for 24 h
Aerated (control)	3.3 ± 0.8	173.2 ± 11.6	70.5 ± 9.5
Oxygenated	2.1 ± 0.7	179.8 ± 10.7	49.0 ± 1.1
Incubated for 120 h
Aerated (control)	5.8 ± 0.9	164.7 ± 38.0	25.3 ± 5.8
Oxygenated	15.3 ± 2.2	154.2 ± 39.5	2.71 ± 1.8

^aGlutathione peroxidase activity was determined by monitoring the oxygenation of NADPH. Glutathione reductase activity and the concentrations of GSSG and GSH were determined by monitoring the oxidation production of TNB²⁻. The values are means ± standard deviations of eight replicates.

Cumulative oxidative damage.

Lipid oxidation and protein carbonylation (Fig. 3) were significantly higher (roughly fourfold higher) in oxygenated cultures approaching and during idiophase. Some protein and lipid damage was also detected in aerated (control) cultures but at much lower levels. DNA AP site production measured after 24 and 120 h of growth were 3 ± 0.45 and 7 ± 0.97 AP sites/10⁴ bp, respectively, in oxygenated cultures and 2 ± 0.42 and 3 ± 0.48 AP sites/10⁴ bp, respectively, in aerated cultures. These results indicate that the overall response of the antioxidant system was insufficient to prevent macromolecule damage. Although the fungus survived when flushed with pure O₂, the high levels of oxidized products of lipids, proteins, and DNA compared to those in cultures grown with atmospheric air confirmed the development of severe oxidative stress in oxygenated, i.e., LIP-producing, cultures.

FIG. 3.

Oxidative damage in oxygenated and aerated cultures of P. chrysosporium as a function of culture age. Lipid oxidative damage was determined in oxygenated cultures (▴) and aerated cultures (▵) by the MDA assay. Protein oxidative damage was determined in oxygenated cultures (•) and aerated cultures (○) by measuring PCC using an ELISA-like method. The means ± standard deviations (error bars) of eight replicates are shown.

LIP activity and transcription.

The pattern of extracellular LIP activity was similar to the patterns of ROS accumulation and macromolecule damage in oxygenated cultures, peaking at 120 h (Fig. 4). Negligible LIP activity was observed in aerated cultures. The correspondence between LIP activity and oxidative damage suggests a role for ROS in inducing LIP synthesis. Glucose consumption and biomass development in liquid cultures and radial growth on agar plates containing 50 mM DMSO, an OH^· scavenger, were similar to the values for cultures and agar plates lacking DMSO. Thus, 50 mM DMSO does not affect the availability of carbon and the viability of fungal cells (not shown).

FIG. 4.

LIP activity in the extracellular fluid of P. chrysosporium cultures as a function of culture age. LIP activity in oxygenated cultures (□) and aerated cultures (○) was assayed by monitoring the oxidation of veratryl alcohol to veratryl aldehyde. The means ± standard deviations (error bars) of eight replicates are shown.

No LIP activity was detected in cultures supplemented with DMSO (i.e., grown in the absence of intracellular OH^·) under either aerated or oxygenated conditions (not shown). No significant levels of heme protein or LIP isozyme activity were detected in cultures grown with DMSO, in contrast to aerated and oxygenated cultures grown without DMSO (Fig. 5). The LIP-H8 isoenzyme might be an exception to this general pattern. The level of LIP-H8 isoenzyme in the experimental cultures was very low (perhaps within the baseline drift), which makes it difficult to determine whether the H8 isoenzyme is regulated differently, like the other heme proteins. The levels of the other heme proteins decreased significantly.

FIG. 5.

Production of LIP isozymes by P. chrysosporium. LIP isozymes in oxygenated (solid line) and aerated (broken) cultures in the absence of the hydroxyl radical scavenger DMSO (controls) were examined. No detection signals were obtained in the presence of DMSO in any culture. Fractionation was performed by anion-exchange high-performance liquid chromatography analysis, with a MonoQ column with monitoring at 409 nm. mAU, milli absorbance units.

These results indicated that OH^· might differentially influence transcription. We tested this hypothesis by measuring mRNA levels of LIP-H2 isozyme and its dephosphorylated form, LIP-H1, the predominant LIP isoenzyme expressed. Significantly higher LIP-H2 transcription occurred in oxygenated cultures than in aerated cultures, and no LIP transcripts were obtained in the presence of 50 mM DMSO (Fig. 6A and B). The level of 18S mRNA was constant and did not vary with the different treatments (Fig. 6C and D). Similar results were obtained with Northern blots (data not shown). The absence of LIP-H2 activity and transcription in cultures in which hydroxyl radicals were scavenged suggests the involvement of this radical as a second messenger in LIP gene induction. Additional support for the involvement of OH^· in LIP gene induction was obtained by measuring LIP-H2 mRNA after in situ generation of OH^· via the Fenton reaction (Fig. 7). Higher expression of LIP-H2 occurred in cultures with Fenton reagents than in regular cultures (with no Fenton reagents), while either DMSO or mannitol repressed LIP-H2 expression. Slightly higher LIP-H2 expression occurred when Fe(II) alone was added, while when H₂O₂ alone was added, LIP-H2 expression was increased, although its intensity was considerably lower than when both Fenton reagents were added together (data not shown). The more pronounced repression observed with mannitol may be due to a higher OH^· concentration scavenged for mannitol than for DMSO. The level of 18S mRNA remained constant across the treatments.

FIG. 6.

LIP transcription in cell extracts of P. chrysosporium from oxygenated and aerated cultures in the presence and absence of the hydroxyl radical scavenger DMSO. (A and C) Aerated cultures; (B and D) oxygenated cultures. LIP-H2 and 18S mRNA levels were determined by RT-PCR analysis after 96 and 120 h of culture. Lanes M, DNA markers.

FIG. 7.

Effect of in situ generation of OH^· in LIP transcription in cell extracts of 96-h-old cultures of P. chrysosporium. Cultures that were grown with free exchange of atmospheric air for 96 h were incubated for 1 h at 37°C and at 175 rpm with or without Fenton reagents (0.1 mM Fe-EDTA and 0.5 mM H₂O₂) in the absence or presence of DMSO or mannitol (50 mM). LIP-H2 and 18S mRNA levels were determined by RT-PCR. Lane M, DNA markers.

Intracellular cAMP level.

Intracellular cAMP levels responded directly to starvation and were almost identical in both aerated and oxygenated cultures whether or not LIP synthesis occurred (Fig. 8). In all cases, it peaked near the transition to the idiophase around day 2 as a function of nitrogen starvation. This trend corresponded to a complete depletion of nitrogen during the first 30 h (not shown). These results strongly suggest that even though cAMP is a prerequisite for LIP expression, a second factor, possibly induced by ROS, may also be involved. Indeed, only cultures exposed to pure O₂ had significant LIP activity.

FIG. 8.

Intracellular cAMP level in oxygenated (□) and aerated (○) cultures of P. chrysosporium, as a function of culture age. The means ± standard deviations (error bars) of five replicates are shown.

DISCUSSION

We found that P. chrysosporium cells produce a high concentration of ROS when stimulated by pure O₂, which seems to be involved in modulating LIP expression. Oxidative stress was confirmed by measuring ROS concentration, the enhancement of the antioxidant defense system, and oxidative damage of major macromolecules, lipids, proteins, and DNA. The low GSH/GSGG ratio and the high activities of SOD, catalase, and glutathione peroxidase in oxygenated cultures of P. chrysosporium could be associated with the neutralization of ROS, including lipid peroxyl radicals. Resistance of P. chrysogenum to high concentrations of peroxide (12) was explained by high catalase and glutathione peroxidase activities. The levels of GSSG were higher than those of GSH levels when this fungus was exposed to the superoxide-generating agent menadione.

Collectively, our results are consistent with recent hypotheses in which oxidative stress conditions affect LIP synthesis in P. chrysosporium (35, 36, 44-46). Rothschild et al. (36) suggested that Mn deficiency increases the level of oxygen radicals generated within the cell. Zacchi et al. (44-46) reported a major loss in organization of cellular structure, possibly due to oxygen toxicity, in carbon-limited cultures of P. chrysosporium exposed to an atmosphere of pure O_2, which implies that LIP may be synthesized in response to oxidative stress.

Oxygen free radicals or ROS mediate cellular activity and regulate gene expression. Superoxide mediates cell growth and motility in tumor cells and endothelial cells (31, 32, 39), and oxygen free radicals and their derivatives help activate mitogen-activated protein kinases, which induce phosphorylation and dephosphorylation cascades, leading to the conversion of extracellular effects into intracellular actions (2). Oxidants mediate the mitogenic signaling induced by the Ras oncogene (18). The increased production of oxygen free radicals induced by high oxygen levels in liquid cultures of P. chrysosporium may act as second messengers for the regulation of LIP gene expression.

It is difficult to directly identify ROS in any biological sample due to their very short half-lives. A common approach is to add radical scavengers that undergo characteristic preferential reactions with ROS over an extended period and then measure the accumulation of the diagnostic products in the culture (16). The hydroxyl radical, which has the shortest half-life, is the most reactive and toxic oxygen radical, and it is nonspecific to tissues and macromolecules. Scavenging of OH^· appeared to be the most effective way of proving its influence in vivo. DMSO is a well-known scavenger of hydroxyl radicals (3, 15, 17, 27, 43). At 50 mM, DMSO did not affect the viability of the fungus, as evidenced by biomass, glucose consumption, and protein levels; however, it might directly or indirectly alter nutrient availability and ultimately lip expression. On the basis of our results, we hypothesize that ROS, and in particular hydroxyl radicals, mediate LIP expression. When these radicals are depleted by DMSO, LIP expression is not induced at either the mRNA or protein level, regardless of the type of oxygenation. The lower levels of oxygen in aerated cultures may produce a low concentration of hydroxyl radicals, resulting in a slight induction of LIP expression. This hypothesis is also supported by the fact that in situ generation of OH^· enhances LIP-H2 expression and that this expression was repressed by the OH^· scavengers DMSO and mannitol.

The induction of LIP by oxygen may result from reactions of ROS (hydroxyl radicals) with the cell surface or from the entry of these species into the cells. Boominathan and Reddy (7) suggested that transcription from genes encoding certain LIP enzymes is controlled by cAMP levels, a starvation signal. We found that intracellular cAMP levels responded to starvation conditions and that similar cAMP levels occurred with or without oxygenation. However, LIP expression was affected only when pure O₂ was flushed, i.e., under oxidative stress, suggesting that cAMP is necessary, but not sufficient, for LIP expression, and that high levels of ROS, preferably hydroxyl radicals, are needed to trigger LIP synthesis in P. chrysosporium.

Taken together, our findings demonstrate that the conditions used to promote LIP synthesis in P. chrysosporium result in oxidative stress, indicating the involvement of ROS in the induction of LIP gene expression. Although discerning the role for ROS in the signal transduction pathway of LIP requires further work, the results presented here provide a firm framework for future studies on the regulation of fungal peroxidases.