Production and chemiluminescent free radical reactions of glyoxal in lipid peroxidation of linoleic acid by the ligninolytic enzyme, manganese peroxidase Takashi Watanabe 1 , Nobuaki Shirai 2 , Hitomi Okada 1 , Yoichi Honda 1 and Masaaki Kuwahara 1 1 Laboratory of Biomass Conversion, Wood Research Institute, Kyoto University, Gokasho, Uji, Japan; 2 Industrial Research Center of Shiga Prefecture, Ritto, Kamitoyama, Japan Glyoxal is a key compound involved in glyoxal oxidase (GLOX)-dependent production of glyoxylate, oxalate and H 2 O 2 by lignin-degrading basidiomycetes. In this paper, we report that glyoxal was produced from a metabolite of ligninolytic fungi, linoleic acid, by manganese peroxidase (MnP)-dependent lipid peroxidation. In the absence of the parent substrate of linoleic acid, the dialdehyde was oxidized by MnP and Mn(III) chelate to start free radical reactions with emission of chemiluminescence at 700– 710 nm. The spectroscopic profile of the light emission is distinguishable from (a) singlet oxygen, (b) triplet carbonyls from dioxetane and a-hydroxyperoxyl radicals, and (c) biacyl triplet formed by the coupling of two acyl radicals. The photon emission of glyoxal by MnP was activated by co-oxidation of tartrate. The MnP-dependent oxidation of glyoxal in tartrate buffers continued for 10 days without addition of exogenous H 2 O 2 . The importance of these results is discussed in relation to the free radical chemistry of lignin biodegradation by wood rot fungi. Keywords: Manganese peroxidase; lipid peroxidation; Ceriporiopsis subvermispora; acyl radical. Lignin biodegradation by white rot fungi is an extracellular chemical event generating free radicals. Lignin-degrading enzymes, lignin peroxidase (LiP), manganese peroxidase (MnP) and laccase (Lac), play a key role in generating free radicals from lignin and oxidizable fungal metabolites such as oxalate, glyoxylate, malonate, hydroquinones and aryl alcohols. Due to the participation of peroxidases in the lignin breakdown, a supply of hydrogen peroxide is essential to drive the extracellular enzymatic process. So far, several oxidases have been proposed as the enzymes which carry out this task. The finding that glyoxal and glyoxal oxidase (GLOX) are secreted by white rot fungi strongly suggests that the GLOX system plays a key role in the extracellular H 2 O 2 production [1–6]. As GLOX is activated by peroxidases, the peroxidase-dependent lignin-degradation can be controlled by the combination of GLOX and its substrate, glyoxal [2,7]. Thus, the importance of glyoxal oxidation in wood decay has been recognized. However, little is known about the biosynthetic route for the extra- cellular production of glyoxal by wood rot fungi. In this paper, we first report that a ligninolytic enzyme, MnP, is able to catalyze formation of glyoxal from a metabolite of wood rot fungi, linoleic acid [8], by lipid peroxidation. The glyoxal produced by MnP can be converted to glyoxylate and oxalate by GLOX [6] and these carboxylic acids are further oxidized by MnP or LiP/VA to yield O 2 † – and CO 2 † – , which in turn reduce free radicals and transition metals like Fe(III) [9–12]. Thus, the present result highlights the new roles of MnP-dependent lipid peroxi- dation in free radical chemistry of wood rot fungi. In lipid peroxidation of USFAs, it has been reported that Mn(II) reacts with a chain-carrying radical, peroxyl radical (LOO†), to terminate the chain reactions [13,14]. This raises the question of how the MnP-lipid system generates free radicals in the presence of antioxidant, Mn(II). Recently, we reported that the chain-braking antioxidative activity of Mn(II) is suppressed by regeneration of free radicals by breaking down of LOOH with MnP [15]. In this process, we found that acyl radicals were predominantly formed. This suggests that hydrogen abstraction from aldehydes is involved in the major chain propagation reactions of the MnP-dependent lipid peroxidation. The observation of acyl radicals in the MnP/lipid system prompted us to analyze whether MnP can directly oxidize the aldehyde intermediate in order to carry chain-reactions Correspondence to T. Watanabe, Laboratory of Biomass Conversion, Wood Research Institute, Kyoto University, Gokasho, Uji, Kyoto 611-0011, Japan, Fax: 181 744 38 3600, E-mail: twatanab@kuwri.kyoto-u.ac.jp Enzymes: manganese peroxidase (EC 1.11.1.13); lipoxygenase [linoleate:oxygenoxidoreductase (EC 1.11.13)]; glyoxal oxidase (EC 1.2.3 ). (Received 8 May 2001, revised 24 September 2001, accepted 27 September 2001) Abbreviations:O 2 † – , superoxide anion; CO 2 † – , formate anion radical; MnP, manganese peroxidase; LiP, lignin peroxidase; HRP, horseradish peroxidase; 13(S)-HPODE, 13(S)-hydroperoxy-9Z,11E-octadecadienoic acid; SFA, saturated fatty acid; USFA, unsaturated fatty acid; 2,6-DMP, 2,6-dimethoxyphenol; ESR, electron spin resonance; MDA, malondialdehyde; MSTFA, N-methyl-N-trimethylsilyltrifluroacetamide; DFB, decafluorobenzene; GLOX, glyoxal oxidase, TBARS, thiobarbituric acid reactive substances; PFBHA, O-(2,3,4,5,6-pentafluorobenzyl)hydroxylamine hydrochloride; PFBO, pentafluorobenzyl oxime; CH 3 CN, acetonitrile; MeOH, methyl alcohol; EtOH, ethyl alcohol; DM, n-dodecyl b-maltoside; DHMA, dihydroxymaleic acid; EI/GC/MS, electron ionization-gas chromatography-mass spectrometer; PAH, polycyclic aromatic hydrocarbon. Eur. J. Biochem. 268, 6114–6122 (2001) q FEBS 2001 without the aid of the other oxidizable substrates. We now report the formation and chemiluminescent chain reactions of glyoxal in MnP-dependent lipid peroxidation of linoleic acid. MATERIALS AND METHODS General methods Manganese (II) sulfate and 1,2,3-trimethoxybenzene, decafluorobenzene, 1-dodecanal, 1-decanal, 2,4-nonadienal, 1-hexanal, 1-nonanal, 1-pentanal, 1-octanal, 1-butanal, glyoxlic acid, glycol aldehyde, glyoxal were obtained from Wako Pure Chemical Industries (Tokyo, Japan). trans,trans-2,4-Decadienal, cis-4-decenal was obtained from Aldrich Chemical Company (Milwaukee, USA). trans-2- Hexenal, 1-undecanal, 1-heptanal, trans-2-nonenal, 1-tride- canal, 2-butanone, 3-buten-2-one, 3-pentanone, 2-pentanone, 2-heptanone, 2-hexanone, 2-octanone was obtained from Tokyo Kasei Kogyo (Tokyo, Japan). Linoleic acid was purchased from Nacalai Tescque (Kyoto, Japan). The linoleic acid was purified by passing through a Sep-Pak TM CN Light cartridge (Waters, Milford, MA, USA). After dissolving in n-hexane, the eluent from the cartridge was evaporated with a gentle stream of N 2 gas. Milli-Q TM water was used throughout the experiments. All of the chemicals used were of analytical reagent grade. 13(S)-Hydroperoxy-9Z,11E-oc- tadecadienoic acid [13(S)-HPODE] was prepared as described previously [15]. Malondialdehyde (MDA) was synthesized as described previously [16]. Enzyme preparation Crude MnP from Ceriporiopsis subvermispora FP-90031 was collected from 7-day-old cultures grown on a wood medium composed of beech wood (5 g), glucose (0.7 g) and peptone (0.7 g) at 28 8C. The culture filtrate was dialyzed against 20 m M sodium succinate buffer (pH 4.5). The dialyzate was concentrated by ultrafiltration, precipitated with (NH 4 ) 2 SO 4 and then purified by gel filtration on Superdex 75 PG (1.6  60 cm, Amersham Pharmacia Biotech, Sweden) using 20 m M sodium succinate buffer containing 0.1 M NaCl as an eluent. Fractions showing MnP activities were collected, and desalted with Centriprep YM-30 (cut off, 30,000, Millipore, USA). MnP was further purified by preparative IEF as described previously [15] [pI 3.40, Reinheitzahl (RZ, A at l max /A 280) value: 3.0, 1.0 U ¼ 8.75  10 211 mol]. Low molecular mass com- pounds were removed by successive washings with Milli- Q TM water with a Centricut N-10 ultrafiltration concentrator (cut off, 10 000, Kurabo, Japan) before use. For the time course experiments of aldehyde production, the enzyme purified on Superdex 75 PG was desalted with distilled water in Centricut N-10 and used without further purification (15 U : mL 21 ). Laccase activity in the partially purified fraction was below 0.02 U : mL 21 . Glyoxal oxidase activity was not found in all the enzyme preparations. Enzyme assay MnP activity was measured with 2,6-DMP. The reaction mixture contained 0. 2 m M 2,6-DMP, 0. 5 mM MnSO 4 , 0. 1 m M H 2 O 2 ,25mM sodium tartrate buffer (pH 3. 0) and the enzyme solution. Reactions were started by adding H 2 O 2 and were quantified by monitoring the initial rate of increase in absorbance at 470 nm in the presence and absence of manganese. One unit of enzyme activity is defined as the amount of enzyme that oxidizes 1 mmol of 2,6-DMP in 1 min. Laccase activity was measured with 2,6-DMP under the same conditions but without H 2 O 2 . Lipoxygenase activity was measured by O 2 uptake in a reaction system containing 1. 5 m M linoleic acid, 1 mM n-dodecyl b-maltoside (DM) and 20 mM Tris/HCl buffer (pH. 9. 0). One unit of lipoxygenase activity is defined as the amount of enzyme that absorbs 1 mmol of O 2 in 1 min. GLOX activity was measured by O 2 uptake in a reaction system containing 3 m M glyoxal in 20 mM sodium tartrate (pH 3. 0), acetate (pH 4. 5) or phosphate (pH 6. 0) buffers. Electron ionization/gas chromatography/mass spectrometetry (EI/GC/MS) analysis of oxidation products by MnP Linoleic acid and aldehydes were reacted with 250 mU of the purified MnP, 0.5 m M of Mn(II) and 50 mM of H 2 O 2 at 20 8C for 1–24 h in 10 m M acetate, formate, lactate and tartrate buffers (pH 4.5). After the reaction, 0.5 mL of aqueous PFBHA (0.05 M, 200 mL) was added and reacted at 35 8C for 0.5 h [17]. To this solution, 10 mL of a 10-m M methanol solution of decafluorobenzene (DFB) and a drop of 18-N-sulfuric acid were added and the mixture was partitioned between n-hexane and H 2 O twice. The hexane layer was dried over Na 2 SO 4 , evaporated with a gentle stream of N 2 gas and directly injected into an EI/GC/ MS system. The EI/GC/MS analysis was done with a Shimadzu QP-5050 A GC/MS with ionization energy of 70 eV on CP-Sil-8 (50 m  0.25 mm internal diemeter, Chrompack, Netherlands) using helium as a carrier gas. The column oven temperature was raised from 80 8C to 250 8C at 5 8C : min 21 , and maintained at 250 8C for 20 min. The time course of glyoxal production by MnP was analyzed as described above after the reaction with and without linoleic acid in formate and tartrate buffers. EI/GC/MS analyses of authentic aldehydes and ketones were carried out using a 0.6-m M methanol solution after derivatization with PFBHA under the conditions described above. Tetramethylsilation by N-methyl-N-trimethylsilyltrifluroacetamide (MSTFA) was carried out as described previously [18]. Chemiluminescence measurements Chemiluminescence was measured by an ultra-high sensi- tive photon counter (ARGUS-50/VIM, Hamamatsu Photo- nics, Hamamatsu, Japan) equipped with a charge-coupled device (CCD) camera connected with an image intensifier and ARGUS-50 image processor. The wavelength range of the detector was 350–650 nm, 512  483 pixels, and the noise count was 0.15 c.p.s. The reactions were carried out in a cuvette for a 96-well microplate reader. The conditions for each experiment are described in the figure legends. Inactivation of MnP was carried out by heating the MnP in a boiling water bath for 10 min The chemiluminescence spectra were measured by a simultaneous multiwavelength analyzer CLA-SP2 (Tohoku Electronic Industries Co. Ltd, Sendai, Japan) with an q FEBS 2001 Production of glyoxal in lipid peroxidation by MnP (Eur. J. Biochem. 268) 6115 incident slit width of 1.0 mm. The wavelength range of the spectrometer was 370–820 nm. Experimental conditions are described in the legend of each figure. RESULTS Formation of glyoxal in the reaction of linoleic acid with MnP Lipid peroxidation by MnP is a free radical process capable of decomposing recalcitrant PAH and nonphenolic lignin model compounds [19 – 21]. We previously reported that the oxidation of linoleic acid by MnP produced acyl radicals in both tartrate and acetate buffers [15]. The formation of acyl radicals strongly suggests that hydrogen abstraction from aldehyde [22,23] is involved in the oxidative process. To analyze the aldehydes formed by this reaction, linoleic acid was reacted with MnP for 19 h at 20 8C in sodium acetate, formate, lactate and tartrate buffers and the oxidation products were analyzed by EI/GC/MS after derivatization to pentafluorobenzyloxims (PFBO) with PFBHA [17]. EI/ GC/MS analysis of the reaction products and authentic 19 aldehydes and seven ketones demonstrated that glyoxal, 1-hexanal and 1-pentanal were formed from linoleic acid by MnP in any of the buffer systems (Fig. 1). Syn and anti-isomers of these PFBO derivatives were separated on the GC/MS column. The mass spectrum of PFBO derivatives of glyoxal formed from linoleic acid is shown in Fig. 2, together with that of authentic standard. MDA, a major peroxidation product derived from polyunsaturated fatty acids was not detected in the reaction products of MnP in contrast to the oxidation of linoleic acid by xantine/ xanthinoxidase/Fe(II) [24]. The mass fragments of PFBO derivatives characteristic to saturated aldehyde (m/z 239), 2-enal (m/z 250), 2,4-dienal (m/z 276) and saturated 2-ketones (m/z 72) [25], were not observed in the spectra of unidentified carbonyl compounds, indicating that the MnP/Mn(II)/lipid system proceeds by complex radical reactions involving the formation of unusual carbonyl species. Tetramethylsilation with MSTFA did not change the mass chromatogram at m/z 181 that originates from C–O bond cleavage products of pentafluorobenzyl oxime [26] (data not shown). The reactions of MnP in four different buffers clearly demonstrate that the formation of glyoxal was significantly stimulated by the presence of tartrate. Therefore, the reaction was carried out with and without linoleic acid in sodium formate and tartrate buffers (Figs 1 and 3). GC/MS analysis demonstrated that glyoxal was explosively produced after 6 h in tartrate buffer containing linoleic Fig. 1. Mass chromatograms of PFBO derivatives of products of lipid peroxidation by C. subvermipora MnP and soybean lipoxygenase at m/z 181. (A) Products of the oxidation of linoleic acid by MnP in sodium acetate buffer for 19 h. The reaction system (500 mL) contained 3 m M linoleic acid, 500 mM MnSO 4 ,50mM H 2 O 2 , 0.02% of Tween 20, 250 mU of purified MnP and 10 mM sodium acetate buffer (pH 4.5). (B) As (A) but 10 m M sodium formate buffer (pH 4.5) was used instead of acetate buffer. (C), As (A) but 10 mM sodium lactate buffer (pH 4.5) was used instead of acetate buffer. (D) As (A) but 10 m M sodium tartrate buffer (pH 4.5) was used instead of acetate buffer. (E) As (B) but the reaction was carried out without addition of linoleic acid. (F) As (D) but the reaction was carried out without addition of linoleic acid. (G) Products of the oxidation of linoleic acid with soybean lipoxygenase. Linoleic acid (3 m M) was reacted with soybean lipoxygenase (10 U) in 40 mM Tris/HCl buffer (pH 9.0) containing 0.02% of Tween 20 for 24 h at 20 8C. (H) Products of the oxidation of 13(S )HPODE by MnP in sodium lactate buffer (pH 4.5). for 3 h. The reaction system contained 3 m M 13(S )HPODE, 500 mM MnSO 4 ,50mM H 2 O 2 , 0.02% of Tween 20, MnP (250 mU) and 10 mM sodium lactate buffer (pH 4.5). 6116 T. Watanabe et al. (Eur. J. Biochem. 268) q FEBS 2001 acid. However, direct formation of glyoxal from tartrate was also observed. In formate buffer, the production of glyoxal was dependent on the presence of linoleic acid. The same results were obtained with lactate and acetate buffers (data not shown). In contrast to the oxidation of linoleic acid, oxidation of 13(S )HPODE with MnP selectively produced 1-hexanal for 1–3 h (Fig. 1H). No PFBO-derivatives were detected after the prolonged reaction of 13(S )HPODE. Thus it was found that the formation of glyoxal was not catalyzed by the direct oxidation of 13(S )HPODE with MnP. Oxidation of linoleic acid with soybean lipoxyegnase produced 1-hexanal and 1-pentanal (Fig. 1G). Emission of chemiluminescence in lipid peroxidation The chemiluminescence detector is a powerful tool for analyzing the oxidation of aldehydes due to its high sensitivity and emission spectra characteristic to chemically excited species. Therefore, oxidation of aldehydes and linoleic acid by MnP was analyzed by a chemiluminescence detector, in comparison with light emission by lipoxygenase and the Fenton reaction (Fig. 4). Lipoxygenase is an enzyme that abstracts hydrogen from the bis-allylic position of unsaturated fatty acids containing cis,cis-1,4-pentadienyl moiety. In the reaction with linoleic acid, the fatty acid is oxidized to yield a pentyl radical [27] and 12-oxododecyl- cis-9-enoic acid [28] via b-scission of hydroperoxide intermediates, leading to production of 1-hexanal [29] and 1-pentanal as shown in Fig. 1G. When linoleic acid was oxidized by soybean lipoxygenase, emission of chemilumi- nescence was close to the background level, both in the presence and absence of Fe(II). In the Fenton system, chemiluminescence was also below the background level, except for a weak emission of light from linoleic acid after 2 days (Fig. 4). In contrast to these oxidation systems, reactions of glyoxal with MnP in tartrate buffer emitted strong chemilumines- cence. As shown in Figs 5 and 6, intensive light emission was observed immediately after the reaction started. The photon emission reached a maximum (35 000 counts : h 21 ) within 30 min, and then decreased, but chemiluminescence of < 9000 counts : h 21 was observed even after 1 h. In lactate, formate and acetate buffers, the photon emission was also observed within 30 min but the intensity was much lower than that of the tartrate system. In the tartrate system, the photon emission continued for 10 days, both in the presence and absence of exogenous H 2 O 2 added initially (Fig. 6). The photon emission from glyoxal was dependent on the presence of Mn(II) and active enzyme. However, it was found that the emission of chemiluminescence continued for around 10 days when the reaction was started without addition of glyoxal. This can be explained by the in situ formation of glyoxal from tartrate with MnP (Fig. 1,3). In the MnP-catalyzed oxidation of linoleic acid and the other aldehydes, the light emission was not observed when Mn(II) was omitted from the reaction system (data not shown). When linoleic acid and seven different aldehydes were reacted with Mn(III)–tartrate complex, strong light emission was observed in the reaction system with glyoxal (Fig. 7). The maximum photon emission intensity from glyoxal reached 12 000 counts : h 21 . The photon emission was also Fig. 2. Mass spectra of PFBO derivatives of glyoxal formed by the oxidation of linoleic acid with MnP (A) and authentic standard (B). (A) Glyoxal formed by the oxidation of linoleic acid with MnP for 19 h. The reaction system (500 mL) contained 3 m M linoleic acid, 500 mM MnSO 4 ,50mM H 2 O 2 , 0.02% of Tween 20, MnP (250 mU) and 10 mM sodium acetate buffer (pH 4.5). (B) Authentic standard of glyoxal. * 1/10 of the original signal intensity. Fig. 3. Time course of glyoxal formation by MnP. (A) Glyoxal formed by the reaction of linoleic acid with MnP in sodium tartrate buffer. The reaction system (500 mL) contained 3 m M linoleic acid, 500 mM MnSO 4 ,50mM H 2 O 2 , 0.02% of Tween 20, MnP (250 mU) and 10 mM sodium tartrate buffer (pH 4.5). (B) As (A) but linoleic acid was omitted. (C) As (A) but 10 m M sodium formate buffer (pH 4.5) was used instead of sodium tartrate bufer. (D) As (C) but linoleic acid was omitted. q FEBS 2001 Production of glyoxal in lipid peroxidation by MnP (Eur. J. Biochem. 268) 6117 Fig. 4. Time course of light emission during oxidation of linoleic acid by soybean lipoxygenase (I) and the Fenton reaction (II). (I): (A) The reaction system (200 mL) contained 4m M linoleic acid, 10 U of lipoxygenase, 0.05% of Tween 20, 10 m M Tris/HCl buffer (pH 9.0). (B) As (A) but lipoxygenase was omitted. (C) As (A) but 0.5 m M FeSO 4 was added. II: (A) The reaction system (200 mL) contained 4 m M linoleic acid, 0.1 m M FeSO 4 , 0.2 mM H 2 O 2 , 0.05% of Tween 20. (B) As (A) but glyoxal was added instead of linoleic acid. (C) As (A) but trans-2-nonenal was added instead of linoleic acid. (D) As (A) but 1-dodecanal was added instead of linoleic acid. (E) As (A) but 1-hexanal was added instead of linoleic acid. (F) As (A) but 2,4-nonadienal was added instead of linoleic acid. (G) As (A) but MDA was added instead of linoleic acid. (H) As (A) but linoleic acid was omitted. Fig. 5. Chemiluminescence emitted by the oxidation of aldehydes and linoleic acid with MnP in sodium tartrate buffer. (A) The reaction system (200 mL) contained 4 m M linoleic acid, 250 mU of MnP, 500 m M MnSO 4 , 0.2 mM H 2 O 2 , 0.05% of Tween 20 and 10 m M sodium tartrate buffer (pH 4.5). (B) As (A) but glyoxal was added instead of linoleic acid. (C) As (A) but trans-2-nonenal was added instead of linoleic acid. (D) As (A) but 1-dodecanal was added instead of linoleic acid. (E) As (A) but 1-hexanal was added instead of linoleic acid. (F) As (A) but 2,4-nonadienal was added instead of linoleic acid. (G) As (A) but MDA was added instead of linoleic acid. (H) As (A) but linoleic acid was omitted. Inset shows the time course of the reactions (A–H) during 2.5 h. Fig. 6. Chemiluminescence emitted by the oxidation of glyoxal with MnP. (I): (A) The reaction system (200 mL) contained 4 m M glyoxal, 250 mU of MnP, 500 m M MnSO 4 , 0.2 mM H 2 O 2 , 0.05% of Tween 20 and 10 m M sodium acetate buffer (pH 4.5); (B) As in (A) but 10 m M sodium formate buffer (pH 4.5) was used instead of sodium acetate bufer. (C) As (A) but 10 m M sodium tartrate buffer (pH 4.5) was used instead of sodium acetate bufer. (D) As (A) but 10 m M sodium lactate buffer (pH 4.5) was used instead of sodium acetate buffer. (II): (A) The reaction system (200 mL) contained 4m M glyoxal, 250 mU of MnP, 500 mM MnSO 4 , 0.2 m M H 2 O 2 , 0.05% of Tween 20 and 10 mM sodium tartrate buffer (pH 4.5). (B) As (A) but MnSO 4 was omitted. (C) As (A) but H 2 O 2 was omitted. (D) As (A) but glyoxal was omitted. (E) As (A) but inactivated MnP was used instead of native MnP. 6118 T. Watanabe et al. (Eur. J. Biochem. 268) q FEBS 2001 observed with MDA, trans-2-nonenal and linoleic acid, but the intenisity was less than 1/10 of that observed in the oxidation of glyoxal. Light emissions from 1-dedecanal, 1-hexanal, 2,4-nonadienal were almost at the same level as that observed without addition of linoleic acid/aldehyde. No photon emission was observed in the reactions of these oxidizable substrates with Mn(II)–tartrate complex (data not shown). These results demonstrate that MnP catalyzes oxidation of Mn(II) to Mn(III), which in turn reacts with glyoxal to generate electronically excited species. Co-oxida- tion of tartrate is essential to carry the chemiluminescent chain reactions over several days. Figure 8 shows emission spectra obtained by glyoxal oxidation with (A) Mn(III)–lactate complex, and (B) MnP/ Mn(II)/H 2 O 2 in tartrate buffer. The spectra showed a broad single peak with emission maxima at 700 and 710 nm, respectively. Figure 2C is the chemiluminescence spectrum of singlet oxygen formed by the reaction of HClO – with H 2 O 2 . The spectrum showed two sharp emission maxima at 634 and 704 nm as reported [30]. In the spectra obtained by the glyoxal oxidation by MnP and Mn(III)–lactate, no shoulder peaks of the dimol emission of singlet oxygen was observed. DISCUSSION Formation and oxidation of glyoxal in lipid peroxidation of linoleic acid by MnP It has been widely recognized that selective white rot fungi such as Ceriporiopsis subvermispora can delignify wood without penetration of their extracellular enzymes into wood cell walls. Therefore, there is increasing interest in the roles of low molecular mass compounds that generate free radicals capable of decomposing lignin at a site far from enzymes. Lipid peroxidation by manganese peroxidase (MnP) is one candidate for this system because diffusible Mn(III) chelate can react with lipid and lipid hydroperoxides to generate free radicals [15]. When C. subvermispora was cultivated on wood meal medium, the fungus produced saturated fatty acids (SFAs) and unsaturated fatty acids (USFAs) including linoleic acid at an incipient stage of cultivation and consume them with concomitant formation of lipid hydroperoxide and TBARS during prolonged cultivation [8]. In the lipid peroxidation process, titres of MnP reached a maximum around day 4 and then gradually decreased, coincident with the peroxidation of the fatty acids. Thus, accumulated data supports the involvement of MnP-dependent lipid peroxidation in wood decay by white rot fungi. With regard to the radicals produced in the lipid peroxidation by MnP, we reported that MnP oxidized linoleic acid to generate acyl radicals in acetate and tartrate buffers in the presence of Mn(II) [15]. The formation of acyl radicals strongly suggests that hydrogen abstraction from aldehydes is involved in the chain propagation cycle of the MnP-dependent lipid peroxidation. Therefore, aldehydes formed by the MnP/linoleic acid/Mn(II)/H 2 O 2 reactions were analyzed by EI/GC/MS after derivatization to PFBO (Figs 1–3). The GC/MS analysis demonstrated that oxidation of linoleic acid with MnP produced glyoxal, 1-hexanal and 1-pentanal. Time course experiments of the MnP reactions showed that glyoxal was not formed on initiation of the lipid peroxidation but after 6 h (Fig. 3). In contrast to the oxidation of linoleic acid, the reaction of MnP with 13(S)-HPODE selectively produced 1-hexanal, indicating that the glyoxal formation is not catalyzed by the direct breakdown of lipid hydroperoxide with MnP and Mn(III) chelates. In wood decay, a supply of extracellular hydrogen peroxide is essential to initiate peroxidase-dependent free radical processes. An extracellular oxidase, GLOX, has been identified from wood decay fungi and is considered as an enzyme that catalyzes extracelular H 2 O 2 production. However, little is reported about the biosynthetic pathway of glyoxal in wood rot fungi. In 1994, Hammel et al. Fig. 7. Chemiluminescence emitted by the oxidation of aldehydes and linoleic acid with Mn(III)–tartrate complex. (A) The reaction system (200 mL) contained 4 m M linoeic acid, 0.05% of Tween 20, and 2.5 m M Mn(III)–tartrate. A Mn(III) – tartrate solution (10 m M) was prepared by dissolving 0.1 m mol of Mn(III)–acetate in 10 mL of 0.1 M sodium tartrate buffer (pH 4.5). The reaction was initiated by adding 50 mL of this solution. Therefore, the final concentration of tartrate in the reaction system was 25 m M. (B) As (A) but glyoxal was added instead of linoleic acid. (C) As (A) but trans-2-nonenal was added instead of linoleic acid. (D) As (A) but 1-dodecanal was added instead of linoleic acid. (E) As (A) but 1-hexanal was added instead of linoleic acid. (F) As (A) but 2,4-nonadienal was added instead of linoleic acid. (G) As (A) but MDA was added instead of linoleic acid. (H) As (A) but linoleic acid was omitted.Time course of the photon emission from glyoxal is shown separately from that of the other oxidizable compounds due to the difference of emission intensity. q FEBS 2001 Production of glyoxal in lipid peroxidation by MnP (Eur. J. Biochem. 268) 6119 reported that lignin peroxidase deomposed a b-O-4 lignin model compound with production of glycol aldehyde, a substrate of GLOX [6]. The glycol aldehyde formed by this process was converted to oxalate with the production of 2.8 equivalent of H 2 O 2 . Therefore, they proposed a pathway producing oxalate from the b-O-4 lignin model compound via glyoxal and glyoxylate. However, there has been no direct evidence for the glyoxal formation from lignin by the LiP/GLOX system. The finding of glyoxal formation by MnP-dependent lipid peroxidation indicates that MnP and Mn(III) chelate part in the formation of glyoxal, leading to the enzymatic production of glyoxylate, oxalate and H 2 O 2 by GLOX. Mn(III) stabilized by the former two carboxylic acids can diffuse into the wood cell wall region. At the same time, glyoxylate and oxalate are oxidized by Mn(III) to produce CO 2 † – and O 2 † – [11,12]. The same reaction was also catalyzed by LiP/VA [9,10]. Due to the high reduction potential of CO 2 † – , the radical catalyzes reduction of Fe(III) [12] and reductive dehalogenation of recalcitrant aromatic halides [31]. O 2 † – catalyzes oxidation of Mn(II) and reduction of Fe(III) in addition to disproportionation yielding H 2 O 2 . A combination of the iron reduction and H 2 O 2 formation generates hydroxyl radicals. Thus, MnP- dependent lipid peroxidation provides the substrate of GLOX to produce active oxygen species in combination with redox cycle of transition metals. Oxidation of glyoxal by MnP In lipid peroxidation involving aldehyde oxidation, it has been postulated that acyl radicals are formed from aldehydes by hydrogen abstraction with radicals (X†) [22] or transition metals [23] according to: X† 1 RVCHO ! XH 1 RCO† ð1Þ M 31 ðM 21 Þ 1 RCHO ! M 21 ðM 1 Þ 1 RCO† 1 H 1 ð2Þ With regard to the light emission from acyl radicals, four different pathways can be discussed (Fig. 9). As shown in Fig. 9 (pathway 2) two acyl radicals can recombine to produce a biacyl triplet [32]. The light emission reported in excited biacyl compounds like biacetyl (l max at 515 and 560 nm) [32] is different from the emission spectra observed in the MnP reactions. a-Oxidation of aldehydes via a dioxetane intermediate also produces excited triplet carbonyls (Fig. 9, pathway 1) but the absorption maximum of the light emission is in the range of l max 450–550 nm [28,32]. Singlet oxygen can be formed by disproportionation of two a-hydroxyperoxyl radicals (Fig. 9, reaction 4). However, there are no shoulder peaks of the dimol emission from singlet oxygen (l max 634 and 703 nm) [30] in the spectrum of glyoxal oxidation by MnP (Fig. 8), indicating that 1 O 2 is not the major excited species formed by the MnP/glyoxal system. The other possible route for Fig. 8. Chemiluminescence spectra of (A) oxidation of glyoxal by MnP in tartrate buffer (B) oxidation of glyoxal by Mn(III)–lactate complex and (C) singlet oxygen formed by the reaction of ClO – with H 2 O 2 . (A) The reaction system (500 mL) contained 3 mM glyoxal, 250 mU of MnP, 500 m M MnSO 4 ,50mM H 2 O 2 , 0.02% of Tween 20 and 10 m M sodium tartrate buffer (pH 4.5). (B) The reaction system (500 mL) contained 3 m M glyoxal, 0.02% of Tween 20, and 2 mM Mn(III)–lactate. A Mn(III)–lactate solution (10 mM) was prepared by dissolving 0.1 m mol of Mn(III)–acetate in 10 mL of 0.1 M sodium lactate buffer (pH 4.5). The reaction was initiated by adding 100 mLof this solution. Therefore, the final concentration of lactate in the reaction system was 20 m M. (C) The reaction was started by adding 1 mL of 30% H 2 O 2 and 3 mL of 10% NaClO solution. Scanning time for (A) (B) (C) were 20, 15, and 5 min, respectively. 6120 T. Watanabe et al. (Eur. J. Biochem. 268) q FEBS 2001 chemiluminescence emission from acyl radical is a forma- tion of triplet carbonyls from a-hydroxyperoxyl radicals (Fig. 9- [3]) that has been reported in the oxidation of aetaldehyde with xantine oxidase [33]. However, the emission maximum of the chemiluminescence by this mechanism was lower than 500 nm [33] (Fig. 9). Thus, the MnP-dependent light emission from glyoxal at 700 nm is a new chemical event difficult to explain by the excited species from acyl radicals reported before. As shown in Fig. 8, the reaction of Mn(III)–lactate with glyoxal emitted the chemiluminescence similar to that observed in the reactions of MnP. Therefore, it can be concluded that Mn(III) chelate is capable of abstracting hydrogen from glyoxal to form the electronically excited species, which spontaneously decay to emit the chemilumi- nescence. As shown in Fig. 1, peroxidation of linoleic acid with MnP produced glyoxal. However, the chemiclumines- cence from linoleic acid was much less intensive than that of the direct oxidation of glyoxal with MnP. This suggests that the excited species from glyoxal react with peroxidation intermediates from linoleic acid, leading to quenching of the excited compound [34]. Oxidation of tartrate by MnP In this study, we also found that tartrate is oxidized by MnP and Mn(III) chelates in aqueous solutions to produce glyoxal with emission of chemiluminescence. Although tartrate is not a metabolite of ligninolytic fungi, this carboxylic acid is widely used in studies on MnP. For tartrate, unlike oxalate and malonate, there has been no report of MnP-catalyzed degradation. For instance, the ratio of H 2 O 2 consumption vs. oxidation of Mn(II) by MnP in tartrate buffer is reported to be nonstoichiometric due to reduction of H 2 O 2 to O 2 by Mn(III)–tartrate complex [35,36]. However, no special attention was paid to the oxidation of tartrate itself in these studies. This may be due to the understanding that the reactivity of tartrate is too low to be involved in the free radical reactions by Mn(III). For instance, Perez reported that veratryl alcohol oxidation by LiP is not affected by the presence of tartrate in the presence or absence of Mn(II) and Mn(III) [37]. More recently, Collins reported that the rate of 2,2 0 -azinobiz(3-ethylbenzo-6-thiazolinesulfonic acide) (ABTS† 1 ) reduction is enhanced by the presence of malonate, glyoxylate and oxalate but no stimulating effects of tartrate on the ABTS† 1 reduction was observed [38]. In contrast, the results obtained in the present study clearly indicate that tartrate itself was oxidized by MnP to produce glyoxal (Figs 1–3), thereby assisting chain reactions of the aldehyde accompanied by photon emission (Fig. 7,8). In lipid peroxidation of linoleic acid by MnP, the consecutive formation of aldehydes and acyl radicals was observed in acetate and formate buffers as well as in tartrate buffer. Therefore, we propose that the enzymatic process produces counterpart compounds like tartrate to assist the chain propagation reactions of acyl radicals in combination with redox cycle of Mn(II)/Mn(III). In conclusion, the first evidence for the production of glyoxal from linoleic acid in MnP-dependent lipid peroxi- dation has been presented. Glyoxal formed by this process can be used as a substrate of GLOX and MnP to participate in the extracellular free radical reactions of wood rot fungi. In addition to the interest on lignin biodegradation, the analysis of manganese-dependent glyoxal oxidation associ- ated with tartrate oxidation will lead to the understanding of cellular injury caused by the carcinogenic aldehyde in the presence of catalytic amount of manganese. ACKNOWLEDGEMENTS This work was supported by a Grant-in-Aid for Scientific Research (B) from the Ministry of Education, Science, Sports and Culture, Japan. We are grateful to Ms M. Nakagawa for technical assistance in the analysis of aldehydes. We also thank Dr Rie Yamada, Tohoku Electoric Co. Ltd, for the measurement of chemiluminescence spectra. REFERENCES 1. Kersten, P.J. & Kirk, K.T. (1987) Involvment of a new enzyme, glyoxal oxidase, in extracellular H 2 O 2 production by Phanero- chaete chrysosporium. J. Bacteriol. 169, 2195–2201. 2. Kersten, P.J. (1990) Glyoxal oxidase of Phanerochaete chryso- sporium: its characterization and activation by lignin peroxidase. Proc. Natl Acad. Sci. USA 87, 2936–2940. 3. Kersten, P.J., Witek, C., vanden Wymelenberg, A. & Cullen, D. (1995) Phanerochaete chrysosporium glyoxal oxidase is encoded by two allelic variants: structure, genomic organization, and heterologous expression of glx1 and glx2. J. Bacteriol. 177, 6106–6110. Fig. 9. Formation of electronically excited species from acyl radicals reported before [28,30–32]. Aldehydes are oxidized by carbon- centered radicals and metal salts to form acyl radicals [22,23]. Triplet carbonyls, biacyl triplet, and singlet oxygen are formed from the acyl radicals [28,30–32]. Their chemiluminescence spectra are distinguishable from the electronically excited species formed by the reactions of glyoxal with MnP/Mn(II)/H 2 O 2 or Mn(III) chelate. q FEBS 2001 Production of glyoxal in lipid peroxidation by MnP (Eur. J. Biochem. 268) 6121 4. Whittaker, M.M., Kersten, P.J., Nakamura, N., Sanders-Loehr, J., Schweizer, E.S. & Whittaker, J.W. (1996) Glyoxal oxidase from Phanerochaete chrysosporium is a new radical-copper oxidase. J. Biol. Chem. 27, 681–687. 5. Whittaker, M.M., Kersten, P.J., Cullen, D. & Whittaker, J.W. (1999) Identification of catalytic residues in glyoxal oxidase by targeted mutagenesis. J. Biol. Chem. 274, 36226–36232. 6. Hammel, K.E., Mozuch, M.D., Jensen, K.A. Jr & Kersten, P.J. (1994) H 2 O 2 recycling during oxidation of the aryl glycerol b-aryl ether lignin structure by lignin peroxidase and glyoxal oxidase. Biochemistry 33, 13349–13354. 7. Kurek, B. & Kersten, P.J. (1995) Physiological reguration of glyoxal oxidase from Phanerochaete chrysosporium by peroxidase systems. Enzyme Microbial. Technol. 17, 751–756. 8. Enoki, M., Watanabe, T., Nakagame, S., Koller, K., Messner, K., Honda, Y. & Kuwahara, M. (1999) Extracellular lipid peroxidation of selective white-rot fungus, Ceriporiopsis subvermispora. FEMS Microbiol. Lett. 180, 205–211. 9. Popp, J.L., Kalyanaraman, B. & Kirk, T.K. (1990) Lignin peroxi- dase oxidation of Mn 21 in the presence of veratryl alcohol, malonic or oxalic acid, and oxygen. Biochemistry 29, 10475–10480. 10. Akamatsu, Y., Ma, D.B., Higuchi, T. & Shimada, M. (1990) A novel enzymatic decarboxylation of oxalic acid by the lignin peroxidase system of white-rot fungus Phanerochaete chryso- sporium. FEBS Lett. 269, 261–263. 11. Kuan, I.C. & Tien, M. (1993) Glyoxylate-supported reactions catalyzed by Mn peroxidase of Phanerochaete chrysosporium: activity in the absence of added hydrogen peroxide. Arch. Biochem. Biophys. 302, 447 –454. 12. Khindaria, A., Grover, T.A. & Aust., S.D. (1994) Oxalate- dependent reductive activity of manganese peroxidase from Phanerochaete chrysosporium. Arch. Biochem. Biophys. 314, 301–306. 13. Szire ´ ki, I., Mohanakumar, K.P., Rauhara, P., Kim, H.G., Yeh, K.J. & Chiueh, C.C. (1998) Manganese: a transition metal protects nigrostriatal neurons from oxidative stress in the iron-induced animal model of parkinsonism. Neuroscience 85, 1101–1111. 14. Coassin, M., Fulvio, Ursini, F. & Bindoli, A. (1992) Antioxidant effect of manganese. Arch. Biochem. Biophys. 299, 330–333. 15. Watanabe, T., Katayama, S., Enoki, M., Honda, Y. & Kuwahara, M. (2000) Formation of acyl radical in lipid peroxidation of linoleic acid by manganese-dependent peroxidase from Ceriporiopsis subvermispora and Bjerkandera adusta. Eur. J. Biochem. 13, 4222–4231. 16. Yeo, H.C., Helbock, H.J., Chyu, D.W., Ames, B.N. & (1994) Assay of malondialdehyde in biological fluids by gas chromatography- mass spectrometry. Anal. Biochem. 220, 391–396. 17. Rauli, S., Puppo, M.D., Magni, F. & Kienle, M.G. (1998) Validation of malondialdehyde and 4-hydroxy-2-trans-nonenal measurement in plasma by NICI-GC-MS. J. Biochem. 123, 918–923. 18. Spiteller, P. & Spiteller, G. (1998) Strong dependence of the lipid peroxidation product specturm whether Fe 21 /O 2 or Fe 31 /O 2 is used as oxidant Biochim. Biophys. Acta 1392, 23–40. 19. Bao, W., Fukushima, Y., Jensen, K.A. Jr, Moen, M.A. & Hammel, K.E. (1994) Oxidative degradation of non-phenolic lignin during lipid peroxidation by fungal manganese peroxidase. FEBS Lett. 354, 297–300. 20. Moen, M.A. & Hammel, K.E. (1994) Lipid peroxidation by the manganese peroxidase of Phanerochaete chrysosporium is the basis for phenanthrene oxidation by the intact fungus. Appl. Environ. Microbiol. 60, 1956–1961. 21. Bogan, B.W., Lamar, R.T. & Hammel, K.E. (1996) Fluorene oxidation in vivo by Phanerochaete chrysosporium and in vitro during manganese peroxidase-dependent lipid peroxidation. Appl. Environ. Microbiol. 62, 1788–1792. 22. Greenley, T.L. & Davis, M.J. (1992) Detection of radicals produced by reaction of hydroperoxides with rat river microsomal fractions. Biochim. Biophys. Acta 1116, 192–203. 23. Bawn, C.E.H., Hobin, T.P. & Raphael, L. (1956) The metal-salt- catalyzed oxidation of aldehyde. Proc. Roy. Soc. A 237, 313–324. 24. Liu, J., Yeo, H.C., Doniger, S.J. & Ames, B.N. (1997) Assay of aldehydes from lipid peroxidation: gas chromatography-mass spectrometry compared to thiobarbituric acid. Anal. Biochem. 245, 161–166. 25. Glaze, W.H., Koga, M. & Cancilla, D. (1989) Ozonation byproducts. 2. Improvement of an aqueous-phase derivatization method for the detection of formaldehyde, other carbonyl compounds formed by the ozonation of drinking water. Environ. Sci. Technol. 23, 838–847. 26. Hsu, F.F., Hazen, S.L., Giblin, D., Turk, J., Heinecke, J.W. & Gross, M.L. (1999) Mass spectrometric analysis of pentafluorobenzyl oxime derivatives of reactive biological aldehydes. Intern. J. Mass Spectrometry 185/186/187, 795–812. 27. Iwahashi, H., Albro, P.W., McGown, S.R., Tomer, K.B. & Mason, R.P. (1991) Isolation and identification of a-(4-pyridyl-1-oxide)- N-tert-butylnitrone radical adducts formed by the decomposition of the hydroperoxides of linoleic acid, linolenic acid and arachidonic acid by soybean lipoxigenase. Arch. Biochem. Biophys. 285, 172–180. 28. Oliveria, O.M.M., Haun, M., Duran, N., O’Brien, P.J., O’Brien, C.R., Bechara, E.J.H. & Cilento, G. (1978) Enzyme-generated electronically excited carbonyl compounds. J. Biol. Chem. 253, 4707–4712. 29. Bisakowski, B., Perraud, X. & Kermasha, S. (1997) Characteriz- ation of hydroperoxides and carbonyl compounds obtained by lipoxygenase extracts of selected microorganisms. Biosci. Biotech. Biochem. 61, 1262–1269. 30. Di Mascio, P., Catalani, L.H. & Bechara, J.H. (1992) Are dioxetanes chmeiluminescent intermediates in lipoperoxidations? Free Rad. Biol. Med. 12, 471–478. 31. Cameron, M.D. & Aust., S.D. (1999) Degradation of chemicals by reactive radicals produced by cellobiose dehydrogenase from Phanerochaete chrysosporium. Arch. Biochem. Biophys. 367, 115–121. 32. Haun, M., Du ´ ran, N., Augusto, O. & Cilento, G. (1980) Model studies of the a-peroxidase system: formation of an electronically excited product. Arch. Biochem. Biophys. 200, 245–252. 33. Puntarulo, S. & Cederbaum, A.I. (1989) Chemiluminescence from acetaldehyde oxidation by xanthine oxidase involves generation of and interactions with hydroxyl radicals. Clin. Exp. Res. 13, 84–90. 34. De la Fuente, U.J.R. & Lissi, G.E.A. (1992) Excited carbonyl formation in the combination and disproprtionation of free radicals. J. Biolumin. Chemilumin. 7, 27 –35. 35. Wariishi, H., Valli, K. & Gold, M.H. (1992) Manganese (II) oxidation by manganese peroxidase from the basidiomycete Phanerochaete chrysosporium. J. Biol. Chem. 267, 23688 –23695. 36. Aitken, M.D. & Irvine, R.L. (1990) Characterization of reactions catalyzed by manganese peroxidase from Phanerochaete chryso- sporium. Arch. Biochem. Biophys. 276, 405 –414. 37. Perez, J. & Jeffries, T.W. (1993) Role of organic acid chelators in manganese regulation of lignin degradation by Phanerochaete chrysosporium. Appl. Biochem. Biotechnol. 39/40, 227–238. 38. Collins, P.J., Dobson, A.D.W. & Field, J.A. (1998) Reduction of the 2,2 0 -azinobis(3-ethylbenzthiazoline-6-sulfonate) cation radical by physiological organic acids in the absence and presence of manganese. Appl. Environ. Microbiol. 64, 2026–2031. 6122 T. Watanabe et al. (Eur. J. Biochem. 268) q FEBS 2001 . Production and chemiluminescent free radical reactions of glyoxal in lipid peroxidation of linoleic acid by the ligninolytic enzyme, manganese peroxidase Takashi. explained by the in situ formation of glyoxal from tartrate with MnP (Fig. 1,3). In the MnP-catalyzed oxidation of linoleic acid and the other aldehydes, the