Hydroxylated Metabolites of 2,4-Dichlorophenol Imply a Fenton-Type Reaction in Gloeophyllum striatum

Abstract

While degrading 2,4-dichlorophenol, two strains of Gloeophyllum striatum, a basidiomycetous fungus causing brown rot decay of wood, simultaneously produced 4-chlorocatechol and 3,5-dichlorocatechol. These metabolites were identified by comparing high-performance liquid chromatography retention times and mass spectral data with those of chemically synthesized standards. Under similar conditions, 3-hydroxyphthalic hydrazide was generated from phthalic hydrazide, a reaction assumed to indicate hydroxyl radical formation. Accordingly, during chemical degradation of 2,4-dichlorophenol by Fenton's reagent, identical metabolites were formed. Both activities, the conversion of 2,4-[U-¹⁴C]dichlorophenol into ¹⁴CO₂ and the generation of 3-hydroxyphthalic hydrazide, were strongly inhibited by the hydroxyl radical scavenger mannitol and in the absence of iron. These results provide new evidence in favor of a Fenton-type degradation mechanism operative in Gloeophyllum.

Basidiomycetous fungi causing brown rot decay of wood play an important role in the carbon cycle. They depolymerize the cellulose component of wood while lignin remains as an amorphous brown residue. In this process, an extracellular Fenton-type mechanism, providing hydroxyl radicals via Fe²⁺ and hydrogen peroxide, has long been implicated (15). However, the potential of species like Gloeophyllum striatum and Gloeophyllum trabeum in the degradation of xenobiotics was explored only recently: two fluoroquinolone antibacterial drugs, polyethylene glycol, and two chlorophenols have been shown to be decomposed (9, 14, 23, 24). G. striatum produced three remarkable hydroxylated congeners of enrofloxacin, which could also be generated via Fenton's reaction (23). An extracellular 2,5-dimethoxyhydroquinone-driven Fenton reaction in G. trabeum, employed to depolymerize polyethylene glycol, represents the most recent mechanistic evidence (14).

Wheat straw cultures of G. striatum catalyzed CO₂ production from the environmental pollutant 2,4-dichlorophenol (2,4-DCP). Moreover, in a defined mineral medium lacking carbon, nitrogen, and phosphate, G. striatum expressed a degradation capacity similar to that shown on wheat straw (9, 23). Here, our aim was to provide further evidence for a Fenton-type reaction in G. striatum by (i) identifying primary metabolites of 2,4-DCP formed by the fungus and in Fenton's reaction as well; (ii) testing for hydroxylation of phthalic hydrazide (PH) giving 3-hydroxyphthalic hydrazide (3-OHPH), which has been postulated to specifically indicate hydroxyl radical formation (1, 2, 19); and (iii) inhibiting the conversion of 2,4-[U-¹⁴C]DCP to ¹⁴CO₂ as well as hydroxylation of PH under conditions antagonizing hydroxyl radical activity. Such data are intended to help to firmly establish the capability of some higher fungi to employ a Fenton-type reaction in the degradation of xenobiotics. This reaction may even take different forms in the few genera investigated so far (13, 14).

MATERIALS AND METHODS

Organisms.

The source and maintenance of G. striatum DSM 9592 and G. striatum DSM 10335 have been described in reference 23.

Degradation of 2,4-[U-¹⁴C]DCP by G. striatum or Fenton's reagent.

2,4-[U-¹⁴C]DCP (Sigma-Aldrich Chemie, Steinheim, Germany) had a specific activity of 9.3 mCi per mmol and a radiochemical purity of >95%, assessed as described previously (9). G. striatum, pregrown on malt medium, was transferred into a mineral medium (23) containing 20 μM FeSO₄. After addition of 20 μM 2,4-DCP (labeled with 0.3 kBq), culture vessels were incubated at 100 rpm and 24°C in the dark. Uninoculated flasks served as controls. ¹⁴CO₂ and ¹⁴C-labeled volatile organic compounds were quantified as described in reference 9. Radioactivity bound to mycelia was extracted three times with 5 ml of methanol. Aliquots of 1 ml of combined extracts and of culture supernatants were subjected to liquid scintillation counting (9). Dry weight was determined as described in reference 11. Mycelia were combusted in an OX 500 biological oxidizer (Zinsser Analytik, Frankfurt, Germany). Fenton's reagent (method A) consisted of 0.5 mM FeSO₄ and 1% hydrogen peroxide in 10 ml of 5 mM sodium acetate buffer, pH 4.5, 2,4-DCP was applied as described above.

Hydroxylation of PH by G. striatum or in Fenton's reaction.

The formation of 3-OHPH was detected by chemiluminescence (1), recorded on an RF-5001 PC fluorophotometer (Shimadzu, Duisburg, Germany) at a λ_em of 300 to 700 nm (λ_max = 415 nm [2]). Fenton's reagent was modified to contain 50 mM acetate buffer (method B). PH was applied at a final concentration of 0.1 g/liter (1). Fungal cultures or Fenton's reagent without PH served as a blank. Chemically prepared 3-OHPH (6) (purity, >98%, as checked by high-performance liquid chromatography [HPLC] at 310 nm) was employed to calibrate the chemiluminescence signal (Fig. 1) and to identify, by HPLC, the 3-OHPH produced, using a Merck-Hitachi HPLC system (method I) described before (11). PH and 3-OHPH were separated at an isocratic flow rate of 0.5 ml/min, using methanol and 0.1% phosphoric acid (45:55 [vol/vol]).

FIG. 1.

Dependence of chemiluminescence on the concentration of 3-OHPH. Its intensity is expressed as integration units (IU) over the emission range of 350 to 550 nm. Values given are the means ± standard deviations for triplicate measurements. The inset shows the emission spectrum of chemiluminescence in a typical sample.

Isolation of metabolites.

Cultures of G. striatum in mineral medium (23) as well as Fenton's reagent (method C; now consisting of 5 μM FeSO₄ and 0.01% hydrogen peroxide in 100 ml of 0.5 mM acetate buffer, pH 4.5) were supplemented with 250 μM 2,4-DCP and incubated for 48 h. Thereafter, fungal mycelia were removed by filtration. Culture supernatants and Fenton assay mixtures were adjusted to pH 2.0 with H₃PO₄. After three extractions with ethyl acetate, combined extracts were dried over anhydrous sodium sulfate. Following evaporation of the solvent at 40°C, dry matter was redissolved in methanol. To detect metabolites of 2,4-DCP, analytical HPLC was performed on an HPLC system, HP 1050 (method II), as described in reference 23. Metabolites were eluted by employing a linear gradient over 30 min.

HPLC-electrospray ionization mass spectrometry (HPLC-MS).

HPLC-MS was performed on HPLC system 140B (Applied Biosystems Inc., Foster City, Calif.) equipped with a Kromasil 100 C₁₈ column and linked to a MAT 900 MS as described in reference 24. The elution solvent consisted of 0.01% formic acid and acetonitrile. Starting at 100%, formic acid was linearly decreased to 0% over 30 min. The effluent of 200 μl/min was passed on to the MS interface without splitting. The range of m/z, 138 to 1,400, was scanned in the negative mode within 3 s.

Chemicals.

3,5-Dichlorocatechol, synthesized by Schwien et al. (20), was kindly provided by S. Kaschabeck (Stuttgart, Germany). 4-Chlorocatechol is available from Chemos GmbH (Regenstauf, Germany). All other chemicals were purchased from Sigma-Aldrich Chemie or Merck (Darmstadt, Germany).

RESULTS

Metabolites of 2,4-DCP formed by G. striatum or in Fenton's reaction.

HPLC analysis of supernatants of 48-h-old cultures of G. striatum DSM 9592 and G. striatum DSM 10335 provided profiles of metabolites of 2,4-DCP (Fig. 2). A profile obtained from Fenton's reaction after 48 h was also included; structures of primary congeners have not yet been reported (21). Two metabolites (2 and 3) were present in all profiles and had retention times identical to those of the standards, 4-chlorocatechol and 3,5-dichlorocatechol. MS analysis of these metabolites revealed molecular weights of 144 and 178, deduced from characteristic patterns of pseudomolecular ions due to the isotopy of chlorine and in accordance with its natural abundance. 4-Chlorocatechol could clearly be distinguished from a potential structural alternative, 2-chlorohydroquinone, by its retention time (Fig. 2). The structures of other metabolites produced by G. striatum remain to be elucidated. Apparently, some of those were also generated in Fenton's reaction or were already present as impurities in the mixture of standards, most likely indicating its oxidative decomposition.

FIG. 2.

HPLC elution profiles indicating metabolites of 2,4-DCP produced either by chemical degradation with Fenton's reagent (A) or by G. striatum DSM 10335 (B) or G. striatum DSM 9592 (C). A mixture of standard compounds (D) consisted of 2-chlorohydroquinone (1), 4-chlorocatechol (2), 3,5-dichlorocatechol (3), and 2,4-DCP (4), detected at retention times of 10.8, 15.0, 18.0, and 20.1 min, respectively.

¹⁴CO₂ formation from 2,4-[U-¹⁴C]DCP by G. striatum.

The time course of ¹⁴CO₂ production from 2,4-[U-¹⁴C]DCP by liquid cultures of G. striatum DSM 9592 is depicted in Fig. 3A. In mineral medium containing 20 μM FeSO₄, 15.8% of the applied radioactivity was liberated as ¹⁴CO₂ during 22 days. When mannitol was present at 5.6 or 56 mM, ¹⁴CO₂ production was inhibited by 28 and 78%, respectively. In contrast, 56 mM glucose only delayed ¹⁴CO₂ formation. When FeSO₄ was omitted from the medium, ¹⁴CO₂ formation was inhibited by 83%; residual activity was likely due to contaminating iron.

FIG. 3.

Total ¹⁴CO₂ production from 2,4-[U-¹⁴C]DCP (A) and chemiluminescence indicating the formation of 3-OHPH from PH (B) by liquid cultures of G. striatum DSM 9592 in a mineral medium (■) and in mineral medium supplemented with either 56 mM glucose (○), 5.6 mM mannitol (▴), or 56 mM mannitol (▾). Activities resulting upon omission of FeSO₄ from mineral medium (□) are shown. Values shown are the means ± standard deviations for triplicate cultures.

Total amounts of ¹⁴CO₂ produced, mycelial dry weight determined after 22 days of incubation, and dry weight-based ¹⁴CO₂ production are summarized in Table 1. Although similar absolute amounts of ¹⁴CO₂ were obtained in mineral medium and in mineral medium containing 56 mM glucose, in the latter fungal dry weight had almost tripled, indicating extensive growth of G. striatum. Hence, dry weight-based ¹⁴CO₂ production was markedly reduced. Mannitol, supporting growth to a lesser extent and, most likely, acting as hydroxyl radical scavenger (7), caused massive inhibition of dry weight-based ¹⁴CO₂ production. A similarly low value was observed when iron was omitted from the medium.

TABLE 1.

¹⁴CO₂ produced from 2,4-[U-¹⁴C]DCP by G. striatum DSM 9592 after 22 days^a

Medium	Total 14CO2 (nmol)	Fungal dry wt (mg)	14CO2/fungal dry wt (nmol/mg)
Mineral medium	189.6 ± 16.8	23.4 ± 2.1	8.1 ± 3.7
Glucose added at 56 mM	193.2 ± 13.2	68.3 ± 2.3b	2.8 ± 1.7
Mannitol added at 56 mM	42.0 ± 3.6b	44.4 ± 1.5b	0.9 ± 0.6
FeSO4 omitted	32.4 ± 8.4b	30.8 ± 1.7	1.0 ± 1.5

^aValues given are the means ± standard deviations for triplicate cultures. Standard deviations for dry weight-based production of ¹⁴CO₂ were calculated according to Gauss' law of errors. 

^bSignificantly different from values obtained in mineral medium (P < 0.02), as obtained by Tukey's honestly significant difference test. 

Balance of radioactivity.

The distribution of ¹⁴C label in the compartments of cultures of G. striatum DSM 9592 is shown in Table 2. ¹⁴CO₂ formation was highest in mineral medium, but it represented only a small fraction when mannitol was present or when iron was omitted. Compared to cultures in mineral medium, radioactivity collected in the traps for volatile compounds was approximately doubled in the presence of 56 mM mannitol or in the absence of FeSO₄. This indicated conversion of 2,4-DCP into congeners having a higher volatility than the parent compound. About 40% of the applied radioactivity remained in the supernatants of cultures in unsupplemented and glucose-amended mineral medium. This amount was reduced to 30%, if the medium was devoid of FeSO₄. Negligible radioactivity was recovered by methanolic extraction of mycelia, while between 5.0 and 8.6% of ¹⁴C was detected upon their combustion.

TABLE 2.

Distribution of radioactivity (percent) within cultures of G. striatum DSM 9592 after 22 days of incubation in mineral medium^a

Fraction	Uninoculated control	No modification	Mannitol added at 56 mM	FeSO4 omitted
14CO2	0.2 ± 0.1	15.8 ± 1.4b	3.5 ± 0.3b	2.7 ± 0.7b
Volatile organic compounds	9.2 ± 1.5	18.6 ± 1.0b	27.7 ± 1.8bc	41.8 ± 2.3bc
Culture supernatant	81.9 ± 0.5	40.3 ± 1.7b	39.0 ± 3.4b	30.1 ± 0.5bc
Methanol extract		0.9 ± 0.2	2.0 ± 0.3	1.3 ± 0.2
Mycelium		8.6 ± 0.4	7.1 ± 1.7	5.0 ± 0.1
Sum	91.3 ± 3.6	84.2 ± 2.5	79.3 ± 4.2	80.9 ± 2.5

^aValues given are the means ± standard deviations for triplicate cultures. Standard deviations for the sum of radioactivity were calculated according to Gauss' law of errors. 

^bSignificantly different from the uninoculated control (P < 0.02), as obtained by Tukey's honestly significant difference test. 

^cSignificantly different from mineral medium without modification (P < 0.02). 

Hydroxylation of PH by G. striatum.

When PH (instead of 2,4-DCP) was present in mineral medium, G. striatum DSM 9592 formed 3-OHPH; its oxidation under alkaline conditions gives rise to chemiluminescence (1). This was detected throughout the incubation period of 22 days (Fig. 3B). The identity of 3-OHPH was confirmed by HPLC analysis: its retention time of 9.8 min (compared to 6.7 min for PH) and its UV spectrum could be verified using chemically prepared 3-OHPH. In agreement with earlier reports, 3-OHPH showed one major and two minor absorbance maxima at 321, 334, and 277 nm, respectively (16), whereas PH exhibited a major maximum at 292 nm and a minor one at 262 nm (8). The addition of 56 mM (and even of 5.6 mM) mannitol strongly inhibited chemiluminescence (Fig. 3B). Glucose at 56 mM caused only a slight reduction in its intensity during the first days (transient quenching), and chemiluminescence tended to be highest from day 10 onward. Probably, this was due to consumption of glucose and an increased biomass. The omission of FeSO₄ from the medium resulted in drastically reduced chemiluminescence.

Decomposition of 2,4-DCP and PH by Fenton's reagent.

Cumulative ¹⁴CO₂ production from 2,4-[U-¹⁴C]DCP caused by Fenton's reagent is depicted in Fig. 4. After 7 days, 48.8% of the applied radioactivity could be attributed to ¹⁴CO₂ (9). No ¹⁴CO₂ was liberated when FeSO₄ was omitted. Chemiluminescence was most prominent until day 2. Again, the formation of 3-OHPH was confirmed by HPLC analysis. Upon exposure of 3-OHPH (instead of PH) to Fenton's reagent, essentially no chemiluminescence was detected. Instead, the reaction mixture turned brownish, indicating decomposition of 3-OHPH. No chemiluminescence was detected in the absence of FeSO₄. Glucose had been shown before to strongly inhibit the formation of 3-OHPH (19).

FIG. 4.

Total ¹⁴CO₂ production from 2,4-[U-¹⁴C]DCP (squares) and chemiluminescence indicating the formation of 3-OHPH (triangles) by Fenton's reagent in the presence (closed symbols) and in the absence (open symbols) of FeSO₄. Values shown are the means ± standard deviations for triplicate experiments.

DISCUSSION

During the degradation of 2,4-DCP by either G. striatum or Fenton's reagent, two dihydroxylated metabolites, 4-chlorocatechol and 3,5-dichlorocatechol, were formed simultaneously (Fig. 5). In aerobic bacteria and imperfect fungi, decomposition of chlorophenols is initiated by hydroxylating enzymes which have been located intracellularly. In these organisms, 3,5-dichlorocatechol is a known metabolite of 2,4-DCP (12, 18, 20), while 4-chlorocatechol was detected in Penicillium frequentans as a congener of 4-chlorophenol (12). Intracellular hydroxylation was also observed in white rot fungi: Phanerochaete chrysosporium produced 4,5-dichlorocatechol from 3,4-dichlorophenol (5).

FIG. 5.

Primary metabolites of 2,4-DCP (I) produced by the brown rot basidiomycete G. striatum. 3,5-Dichlorocatechol (II) and 4-chlorocatechol (III) are proposed to be formed simultaneously in a hydroxyl radical-driven reaction resulting in hydroxylation of the positions ortho to the phenolic group of 2,4-DCP.

Aromatic hydroxylation involving chemically generated hydroxyl radicals has been studied with a variety of compounds (10, 21, 22). From salicylic acid, congeners hydroxylated in ortho and para positions were formed: the generation of 2,3-dihydroxybenzoate and 2,5-dihydroxybenzoate has been used to detect hydroxyl radicals (7, 13). If exposed to hydroxyl radicals generated by gamma radiolysis of water, para-substituted phenols were hydroxylated predominantly in the ortho position. In addition, the amount of catechols thus formed was increased by electron-withdrawing substituents such as cyano or nitro groups in the para position (22). Because of the chlorine at C-4, the formation of 3,5-dichlorocatechol from 2,4-DCP could be expected in Fenton's reaction. However, concurrent formation of 4-chlorocatechol, indicating hydroxyl radical-driven dechlorination, appears to fully reflect the ortho-specific action of hydroxyl radicals. In Gloeophyllum, this activity is most likely located extracellularly (14). The formation of 3,5-dichlorocatechol and 4-chlorocatechol from 2,4-DCP parallels the hydroxylation of enrofloxacin by G. striatum and in Fenton's reaction in the ortho position to the piperazine moiety, i.e., at C-6, eliminating fluorine, and at C-8 (23).

The production of ¹⁴CO₂ from 2,4-[U-¹⁴C]DCP by G. striatum (and Fenton's reagent) indicates ring cleavage of the catechol intermediates. Alternatively, chlorocatechols can be degraded further by intracellular catechol-1,2-dioxygenases (12, 18, 20). However, no such enzymes were detected in the brown rot species Daedala quercina, Fomes pinicola, and Lenzites trabea, whereas all of these strains catalyzed an intradiol cleavage of 1,2,4-trihydroxybenzene (3), another potential metabolite of 2,4-DCP formed by G. striatum. The presence of a Fenton-type reaction mechanism in G. striatum is further supported by the effect of the hydroxyl radical scavenger mannitol (7), which strongly inhibited ¹⁴CO₂ production from 2,4-[U-¹⁴C]DCP as well as hydroxylation of PH. Moreover, Fe²⁺ is an essential component of Fenton's reagent in serving as an electron donor for hydrogen peroxide, thus causing hydroxyl radical formation (7, 13). Its addition to cultures of G. striatum greatly enhanced mineralization of 2,4-DCP as well as hydroxylation of PH. Recently, studies have shown that the brown rot fungus G. trabeum produced 2,5-dimethoxyhydroquinone and 4,5-dimethoxycatechol, which have been proposed to mediate Fe³⁺ reduction and hydrogen peroxide production by redox cycling (14, 17). All of these findings are in support of an extracellular Fenton-type reaction mechanism in Gloeophyllum.

Finally, the formation of metabolites of higher volatility than 2,4-DCP was enhanced by the addition of the hydroxyl radical quencher mannitol or by omitting Fe²⁺. The structures of such metabolites remain to be elucidated. The production of 3,4-dichloroanisole from 3,4-dichlorophenol by Phanerochaete chrysosporium, e.g., was attributed to cell-associated methylation (3), a process frequently observed in basidiomycetes (4). In conclusion, the simultaneous occurrence of 4-chlorocatechol and 3,5-dichlorocatechol, the dependence on iron and the effect of mannitol on CO₂ production from 2,4-DCP, and the generation of 3-OHPH from PH are in support of an extracellular Fenton-type degradation mechanism employed by G. striatum. However, straw cultures of G. striatum did not show ¹⁴CO₂ formation from 2,4,6-[U-¹⁴C]trinitrotoluene or [U-¹⁴C]pyrene (K. Fahr and D. Schlosser, unpublished data), which may indicate mechanistic limitations. Hence, the general role of Fenton chemistry-based reactions in the environmental fate of xenobiotics remains to be defined.