Báo cáo khoa học: Irreversible cross-linking of heme to the distal tryptophan of stromal ascorbate peroxidase in response to rapid inactivation by H2O2 ppt

Irreversible cross-linking of heme to the distal tryptophan of stromal ascorbate peroxidase in response to rapid inactivation by H 2 O 2 Sakihito Kitajima 1 , Taise Shimaoka 2 , Miyo Kurioka 1 and Akiho Yokota 3 1 Graduate School of Science and Technology, Kyoto Institute of Technology, Japan 2 Research Institute of Innovative Technology for the Earth (RITE), Kyoto, Japan 3 Graduate School of Biological Science, Nara Institute of Science and Technology (NAIST), Nara, Japan Ascorbate peroxidase (APX; EC 1.11.1.11) isoforms of chloroplasts play a central role in scavenging reactive oxygen species such as O 2 –Æ and H 2 O 2 , which are gen- erated in large amounts by photosystems when there is an energy surplus. Chloroplasts of higher plants have two APX isoforms, one localized in the stroma and the other bound to the stromal side of the thylakoid membrane [1]. APX first reacts with one molecule of H 2 O 2 and forms a porphyrin-based (compound I) and then a protein-based (decay product of compound I [2]) radical intermediate. The intermediates are then reduced back to the resting state through compound II by interaction with two molecules of ascorbate. Paradoxically, if an excess of H 2 O 2 is produced in chloroplasts as a result of oxidative stress such as drought or intense light, they are rapidly inactivated because the reaction intermediate of the APX is attacked by excess H 2 O 2 instead of being reduced by ascorbate [1,3,4]. As a result of the inactivation of APX, H 2 O 2 con- tinues to accumulate, resulting in cell damage. In vitro experiments reveal that, under conditions of ascorbate depletion, inacti vation of APX occurs w ithin m inutes [5,6]. Other APX isoforms are localized in the cytosol [7] and microbodies [8] of plants. These other isoforms and cytochrome c peroxidase (CCP) [9], a yeast homo- log of APX, are much more tolerant to H 2 O 2 than chloroplast APXs, although they have similar amino- acid sequences and structures [10]. The aim of this study was to determine why chloro- plast APXs are more sensitive to H 2 O 2 than other per- oxidases. Using a chimera of stromal APX and red algal H 2 O 2 -tolerant APX, we previously showed that a unique loop structure near the catalytic site is involved in the rapid inactivation of stromal APX [6]. In the present study, we examined the structural change asso- ciated with inactivation by H 2 O 2 . Keywords ascorbate peroxidase; chloroplast; cross-link; hydrogen peroxide; inactivation Correspondence S. Kitajima, Graduate School of Science and Technology, Kyoto Institute of Technology, Sakyo-ku, Kyoto 606-8585, Japan Fax: +81 75 724 7762 Tel: +81 75 724 7791 E-mail: sakito@kit.ac.jp (Received 1 February 2007, revised 22 March 2007, accepted 16 April 2007) doi:10.1111/j.1742-4658.2007.05829.x Ascorbate peroxidase (APX) isoforms localized in the stroma and thyla- koid membrane of chloroplasts play a central role in scavenging reactive oxygen species generated by photosystems. These enzymes are inactivated within minutes by H 2 O 2 when the reducing substrate, ascorbate, is deple- ted. We found that, when the enzyme is inactivated by H 2 O 2 , a heme at the catalytic site of a stromal APX isoform is irreversibly cross-linked to a tryptophan residue facing the distal cavity. Mutation of this tryptophan to phenylalanine abolished the cross-linking and increased the half-time for inactivation from < 10 to 62 s. In contrast with H 2 O 2 -tolerant peroxidases, rapid formation of the cross-link in APXs suggests that a radical in the reaction intermediate tends to be located in the distal tryptophan so that heme is easily cross-linked to it. This is the first report of a mutation that improves the tolerance of chloroplast APXs to H 2 O 2 . Abbreviations APX, ascorbate peroxidase; CCP, cytochrome c peroxidase; tsAPX, stromal APX from tobacco. FEBS Journal 274 (2007) 3013–3020 ª 2007 The Authors Journal compilation ª 2007 FEBS 3013 Results Irreversible cross-linking of heme to apoprotein in H 2 O 2 -inactivated stromal APX We produced recombinant stromal APX from tobacco (Nicotiana tabacum) (tsAPX) in Escherichia coli and examined its elution on RP-HPLC before or after treatment for 25 s with 20 mol H 2 O 2 , which reduced the activity by  80% (Fig. 1). Separate peaks for heme (18.2 min) and apoprotein (25.4 min) were observed for untreated tsAPX (Fig. 2A), whereas signi- ficant amounts of heme were coeluted with the apo- protein (broad peaks at 26.4 and 27.0 min) when tsAPX was treated with H 2 O 2 (Fig. 2B). MALDI- TOF-MS of the H 2 O 2 -treated enzyme revealed a peak for the apoprotein ( 32 720 m ⁄ z) and a second peak ( 33 360 m ⁄ z) roughly corresponding to apoprotein plus heme (616.48 Da) and oxygen (Fig. 2E). The mass measurement error in these experiments was less than 250 p.p.m. Although part of tsAPX was polymerized by treatment with H 2 O 2 , detectable by SDS ⁄ PAGE (data not shown) as indicated in CCP [11], the poly- mers were not detectable by MALDI-TOF-MS because of low sensitivity in the higher mass range. These results indicate that, when inactivated by H 2 O 2 , heme is irreversibly cross-linked to the APX apoprotein. Identification of the heme-binding amino-acid residue in H 2 O 2 -inactivated stromal APX To determine which amino-acid residue in the apopro- tein was cross-linked to heme, the inactivated tsAPX was digested with trypsin and subjected to RP-HPLC. We found at least four peaks that absorbed at 400 nm (38.0, 38.9, 40.5 and 50.9 min; Fig. 3A). Because the retention time for the product eluted at 50.9 min was 0 20 40 60 80 100 remaining activity (%) 0 30 60 90 120 150 180 time (sec) tsAPXW35F, – H 2 O 2 tsAPXW35F, + H 2 O 2 tsAPX, – H 2 O 2 tsAPX, + H 2 O 2 Fig. 1. Remaining activities of tsAPX and tsAPXW35F treated with H 2 O 2 . APXs were treated with or without 20 mol H 2 O 2 in O 2 -free 50 m M sodium phosphate, pH 7.0, at 25 °C. The concentration of treated and untreated APXs was 1.9 l M. Results are mean ± SD from five measurements. Fig. 2. Inactivation and cross-linking of APXs by H 2 O 2 . APXs were untreated (A,C) or treated (B,D) with 20 mol H 2 O 2 in O 2 -free 50 mM sodium phosphate, pH 7.0, at 25 °C. Enzymes were separated by HPLC on a C4 reversed-phase column. Protein and heme were detected at 220 nm (thin line) and 400 nm (thick line), respectively. (A) 2.0 l M untreated tsAPX; (B) 1.9 lM tsAPX treated with H 2 O 2 for 25 s; (C) 2.1 l M untreated tsAPXW35F; (D) 2.1 lM tsAPXW35F treated with H 2 O 2 for 120 s; (E) MALDI-TOF-MS spectra of tsAPX treated with or without 5 mol H 2 O 2 for 2 min. The mass measure- ment error was less than 250 p.p.m. mAU, Milliabsorbance units. H 2 O 2 -mediated inactivation of stromal APX S. Kitajima et al. 3014 FEBS Journal 274 (2007) 3013–3020 ª 2007 The Authors Journal compilation ª 2007 FEBS identical with that for a commercial sample of hemin (data not shown), we concluded that it represented a free heme species. The mass spectrum of the most abundant peak (38.0 min; Fig. 3A, inset) had a [M +H] + ion at 1892.8 m ⁄ z, and its MS ⁄ MS spec- trum indicated that it was a peptide with the sequence HDAGTYNK (Fig. 3B). This matches a predicted tryptic peptide corresponding to residues 33–43 (LGWHDAGTYNK) (Fig. 4A), although Leu33, Gly34, and Trp35 could not be assigned in the MS ⁄ MS spectrum. Of these three residues, Trp35 is the only reactive residue and therefore the most likely to cross- link to the heme. The 1892.8 m ⁄ z value obtained was 15.7 higher than the sum of the calculated masses for the protonated peptide LGWHDAGTYNK (1260.67) plus heme (616.48), suggesting incorporation of an oxy- gen atom. We could not identify the two other minor peaks (38.9 and 40.5 min). The side chain of Trp35 faces the distal cavity formed by the heme and N-ter- minal half of the apoprotein of the catalytic site (Fig. 4B). Because the indole ring of Trp35 is 3.18 A ˚ from the porphyrin ring in tsAPX (the distance between nitrogen of the indole ring and C6 of the por- phyrin ring) [10], the heme must move toward Trp35 to form a covalent bond. In active cytosolic APX, ascor- bate binds to the c-meso edge of heme (a propionated side of heme) [12]. The loss of ascorbate-oxidizing 0 100 200 300 400 500 600 700 800 15 20 25 30 35 40 45 50 55 time (min) mAU (220 nm) 0 100 200 300 400 500 600 mAU (400 nm) A 38.0 min 38.9 min 40.5 min 50.9 min 0 100 200 300 400 500 600 700 240 280 320 360 400 440 nm mAU 38.0 min LGW(O-heme) (987.3)LGW(O-heme) (987.3)LGW(O-heme) (987.3) B LGW(O-heme) (987.3) Fig. 3. Identification of the amino-acid resi- due cross-linked to heme in H 2 O 2 -treated tsAPX. tsAPX was treated with 20 mol H 2 O 2 at 25 °C for 20 s and then digested with trypsin. (A) C18 RP-HPLC. Peptide and heme were detected at 220 nm (thin line) and 400 nm (thick line), respectively. Inset, UV ⁄ Vis spectra of the product eluted at 38.0 min. (B) MS ⁄ MS spectrum of the product eluted at 38.0 min. The peptide sequences obtained from b and y fragment ions are indicated. mAU, Milliabsorbance units. Distal side Proximal side Trp35 Distal side Proximal side Trp35 A B Fig. 4. Deduced amino-acid sequence of tsAPX (A) and structure of its catalytic site [10] (B). The sequence of the tryptic peptide identi- fied from the MS ⁄ MS spectrum in Fig. 3B is underlined. S. Kitajima et al. H 2 O 2 -mediated inactivation of stromal APX FEBS Journal 274 (2007) 3013–3020 ª 2007 The Authors Journal compilation ª 2007 FEBS 3015 activity in the cross-linked form of tsAPX may there- fore be due to the repositioning of heme, preventing it from interacting with ascorbate. Effect of Trp35 mutation To investigate the role of Trp35 in the cross-link and the inactivation of APX by H 2 O 2 , we created a mutant form of tsAPX in which Trp35 was changed to phenyl- alanine (tsAPXW35F). We found that tsAPX in O 2 -free 50 mm sodium phosphate, pH 7.0, had a Soret band at 404 nm (e 404 ¼ 105 mm )1 Æcm )1 ) and a shoulder around 380 nm (Fig. 5A). This is similar to the spectrum for yeast CCP, which has a five- coordinated high-spin ferric heme [13]. The spectrum of tsAPXW35F, however, lacked the shoulder around 380 nm (Fig. 5B) and had a more intense Soret band (e 405 ¼ 122 mm )1 Æcm )1 ), which is typical of a six- coordinated ferric heme, as found in the CCP mutant [13,14]. Spectra for these two APXs in the visible region were similar, and both had two charge-transfer bands (Fig. 5A,B), which is characteristic of high-spin heme species. This is slightly different from a similar mutant of cytosolic APX from soybean reported by Badyal et al. [15]; specifically, when the corresponding tryptophan was changed to alanine in soybean cytoso- lic APX, a peak appeared at 564 nm, which is charac- teristic of low-spin heme species. Furthermore, the spectra for the six-coordinated low-spin ferric forms, prepared by treatment with KCN, were similar for tsAPX and tsAPXW35F (Fig. 5A,B). Finally, the K m and k cat values for tsAPXW35F were only slightly dif- ferent from those for tsAPX (Table 1). These results suggest that, except for a difference in the coordination of the distal side of the heme ferric atom, the W35F mutation did not cause a significant change in the structure of tsAPX. Next, we examined the effects of the mutation on the interaction of tsAPX with excess H 2 O 2 . In con- trast with tsAPX, a cross-link was not observed in tsAPXW35F, even when it was treated with 20 mol H 2 O 2 for 120 s (Fig. 2C,D). This agrees with the MS ⁄ MS results showing that Trp35 is the most likely site for cross-linking to heme. This also indicates that the tryptic peptides of two unidentified minor peaks eluted at 38.9 and 40.5 m (Fig. 3A) are probably parti- ally digested products containing Trp35 cross-linked to heme. In the presence of 20 mol H 2 O 2 , tsAPXW35F had a half-time of inactivation of 62 s, which is more than 6.2-fold longer than for tsAPX (< 10 s) (Fig. 1). We obtained similar results when we included excess BSA in the reaction to exclude the possible effect of con- taminating apoprotein (data not shown). These results strongly support the idea that the formation of the cross-link is at least part of the reason for the rapid inactivation of tsAPX. In these experiments, tsAPXW35F was inactivated in 3 min by H 2 O 2 . Whether this reflects other aspects of the inactivation mechanism for tsAPX is not clear; it is possible that tsAPXW35F is inactivated by a distinct mechanism because the heme can no longer be cross-linked to the enzyme. The possibility of a distinct mechanism of inactivation is supported by the difference in the spectral changes for H 2 O 2 -treated tsAPXW35F and tsAPX (Fig. 6A,B). Discussion In these studies, we showed that heme cross-links to the distal tryptophan in tsAPX within minutes when Fig. 5. Absorption spectra of tsAPX (A) and tsAPXW35F (B) treated with or without 0.2 m M KCN in O 2 -free 50 mM sodium phosphate, pH 7.0. Table 1. Steady-state kinetic parameters of APXs. K m(Asc) a (lM) K mðH 2 O 2 Þ b (lM) k cat (s )1 Æheme )1 ) tsAPX c 395 ± 27 21.7 ± 1 2510 ± 90 tsAPXW35F 170 ± 7 93.9 ± 0.9 3410 ± 130 a K m for ascorbate; b K m for H 2 O 2 ; c Kitajima et al.[6]. H 2 O 2 -mediated inactivation of stromal APX S. Kitajima et al. 3016 FEBS Journal 274 (2007) 3013–3020 ª 2007 The Authors Journal compilation ª 2007 FEBS treated with H 2 O 2 in the absence of ascorbate. On the basis of these results, we propose that the rapid inacti- vation of tsAPX is at least partly due to repositioning of heme caused by cross-linking between heme and the distal tryptophan resulting from reaction with H 2 O 2 . Recently, Pipirou et al. [16] reported that part of the heme molecule is cross-linked to the distal tryptophan in the cytosolic APX isoform when it reacts with excess H 2 O 2 in the absence of ascorbate. They pro- posed that a vinyl group of heme is bound to C1 of the distal tryptophan and hydroxylated. The cross-link between heme and tryptophan may occur in a similar way in tsAPX. In a H 2 O 2 -tolerant APX isoform from red algae [6,17], excess H 2 O 2 also causes cross-linking between heme and the apoprotein, but the ratio of cross-linked heme to total heme was lower than in tsAPX (data not shown). In CCP of yeast, the distal tryptophan is also conserved, but it has not been reported to cross-link to heme. Thus, H 2 O 2 -mediated cross-linking in perxoxidases other than tsAPX, if it occurs, may be much slower. Why the heme rapidly forms a cross-link in tsAPX is uncertain. In CCP, a proximal tryptophan residue distant from the porphyrin is a major radical site in its reaction intermediate [18]. In addition, when its reducing substrate, cytochrome c, is absent, the rad- ical is transferred to, oxidizes, and disrupts trypto- phan and tyrosine residues distant from the heme [19–21]. Also, in cytosolic APX of pea [2] and in fungal lignin peroxidase [22], when the reducing sub- strate is absent, the radical is thought to transfer from porphyrin to tryptophans far from heme, resulting in their hydroxylation. The relocation of the radical means that these amino-acid residues directly or indirectly donate electrons as endogenous reducing substrates to the porphyrin radical, protect- ing the enzyme from over-oxidation by excess H 2 O 2 . The cross-linking of Trp35 to heme of tsAPX thus suggests that the radical in the reaction intermediate is located in porphyrin and Trp35, but that reloca- tion to residues distant from heme, if it occurs, is much slower than in other peroxidases. As a result, the cross-linking of heme may occur readily in tsAPX. In bifunctional catalase–peroxidase, a bacterial homolog of APX, a radical is transferred from por- phyrin to another tryptophan residue that is connec- ted to a propionate side chain of porphyrin by a hydrogen-bonding network through two water mole- cules [23]. The interaction of the propionate side chain with the protein is different for tsAPX and cytosolic APX, because of a unique 16-amino-acid stretch [10] that confers higher sensitivity to H 2 O 2 [6]. A change in the interaction of the propionate with amino-acid residues may therefore influence transfer of the radical in the reaction intermediate of tsAPX. Theoretically, the propionate side chain is also involved in electron transfer from amino-acid residues to the porphyrin [24]. The tendency of the radical to remain near the heme may allow a more rapid catalytic turnover, although at the expense of tolerance to H 2 O 2 . This might have been evolutionary pressure on the chloroplast APXs. In fact, the specific activities of chloroplast APXs reported to date are much higher than those of the cytosolic APXs [25–30]. In conclusion, we have shown that the rapid inacti- vation of tsAPX is at least partly due to cross-linking between heme and the distal tryptophan as a result of reaction with H 2 O 2 . Given the amino-acid sequence similarity between stromal and thylakoid-bound APXs (reviewed in [31]), the inactivation mechanism pro- posed here should also be relevant for thylakoid-bound APX. Absorbance Absorbance Absorbance Absorbance 0.005 0.01 0.015 0.02 0.025 0.03 0.05 0.1 0.15 0.2 0.25 A 0 0.05 0.1 0.15 0.2 0 0.005 0.01 0.015 0.02 00 B untreated untreated 300 350 400 450 500 550 600 650 700 nm untreated untreated Fig. 6. Spectral change in tsAPX and tsAPXW35F treated with 20 mol H 2 O 2 . APXs were solubilized in O 2 -free 50 mM sodium phosphate, pH 7.0, and then treated with H 2 O 2 . (A) tsAPX (2.1 lM) before treatment and 3, 9, 18, 60, and 120 s after addition of H 2 O 2 . (B) tsAPXW35F (1.9 lM) before treatment and 3, 60, 120, 180, and 300 s after the addition of H 2 O 2 . Spectral changes were monitored at 22 °C using a photodiode array spectrophotometer. S. Kitajima et al. H 2 O 2 -mediated inactivation of stromal APX FEBS Journal 274 (2007) 3013–3020 ª 2007 The Authors Journal compilation ª 2007 FEBS 3017 Experimental procedures Preparation of recombinant APXs Expression plasmids for tsAPX were constructed as des- cribed previously [6]. Trp35 was mutated by PCR-mediated site-directed mutagenesis. The expression plasmids encoded APXs corresponding to residues 92–386 of accession num- ber AB022274, with methionine and glycine residues at the N-terminus, which are not present in the native enzyme. Recombinant APXs produced in E. coli BL21(DE3) were purified by sequential steps of chromatography on HiPrep 16 ⁄ 10 DEAE FF (Amersham Bioscience, Piscataway, NJ, USA), HiLoad 16 ⁄ 10 Phenyl Sepharose HP (Amersham Bioscience), and HiLoad 16 ⁄ 60 Superdex 75 pg (Amersham Bioscience) as previously described [6]. Purified tsAPX and tsAPXW35F appeared as single bands when separated by SDS ⁄ PAGE (data not shown). O 2 -free APX solution was prepared by passing APX in 10 mm potassium phosphate, pH 7.0, 1 mm EDTA, 1 mm ascorbate, and 0.15 m KCl through two Sephadex G25 col- umns (NAP5 and PD10 columns; Amersham Bioscience) and elution with 50 mm sodium phosphate, pH 7.0, that had been degassed by bubbling with N 2 gas. Before analy- sis, the concentration of APX was determined from the absorption of heme. The absorption coefficients of the Soret peak for tsAPX and tsAPXW35F were 105 [6] and 122 mm )1 Æcm )2 , respect- ively. The value for tsAPXW35 was determined according to the heme content and UV ⁄ Vis absorption spectra. Heme contents were determined by the pyridine hemochromogen method [32] with horseradish peroxidase (Nacalai tesque, Kyoto, Japan) as a standard (e ¼ 100 mm )1 Æcm )1 at 403 nm [33]). The heme contents per tsAPX and tsAPXW35F molecule were  70% and 80%, respectively, indicating that  30% and 20% were the apoenzyme. Enzyme assay APX activity was measured as described previously [6], except that the reaction mixture was supplemented with 0.01 mgÆmL )1 BSA for the experiment in Fig. 1E. The K m values for ascorbate and H 2 O 2 and the k cat values were determined as described previously [6]. HPLC and MS HPLC analysis was performed using an LC-VP HPLC system (Shimadzu, Kyoto, Japan) equipped with a SPD- M10AVP photodiode array UV-Vis detector (Shimadzu). The column was maintained at 40 °C. For treatment of APXs with H 2 O 2 , 20 mol H 2 O 2 was manually added to 3.5 mL O 2 -free APX solution with stirring at 25 °C. The reaction was terminated by adding 0.5 mm ascorbic acid. H 2 O 2 and ascorbate were removed by passing the sample through an Econopack 10DG column (Bio-Rad, Hercules, CA, USA) that had been equilibrated with 50 mm sodium phosphate, pH 7.0. For analysis of the heme–apoprotein cross-link, APX was denatured by adding a half volume of 8 : 3 HCl ⁄ acetic acid before injection on to a C4 reversed- phase column (4.6 · 250 mm; 5 lm; Vydac, Hesperia, CA, USA). Protein and heme were separated by delivery of 35% acetonitrile and 0.1% trifluoroacetic acid for 14 min, followed by a linear gradient of 35–45% acetonitrile over 15 min. For MS of undigested APX, acid-denatured APX sample was washed with a ZipTip C4 microcolumn (Millipore, Bedford, MA, USA), and MALDI-TOF analysis was per- formed on a Reflex III mass spectrometer (Bruker Dalto- nics, Bremen, Germany) in linear mode using sinapic acid (Fluka, Buchs, Switzerland) as the matrix. MALDI spectra were externally calibrated using Protein Calibration Stand- ard II (Bruker Daltonics). For MS of trypsin-digested samples, inactivated APX in 50 mm sodium phosphate, pH 7.0, was precipitated with acetone and dissolved in 4 m urea and 50 mm ammonium bicarbonate, pH 8.0. Sequencing-grade trypsin (Promega, Madison, WI, USA) was added to the solution at a molar ratio of 1 : 50 and incubated at 37 °C for 6 h. The resulting reaction mixture was separated on a C18 reversed-phase column (4.6 · 150 mm; TSK gel ODS-100S; Toso, Tokyo, Japan). Peptide and heme were separated by delivery of 0.1% trifluoroacetic acid for 10 min, followed by a linear gradient of 0–10% acetonitrile over 15 min, a linear gradi- ent of 10–40% acetonitrile over 30 min, and a linear gradi- ent of 40–100% acetonitrile over 5 min. The peptide fraction showing both heme and peptide absorbance was isolated and concentrated by evaporation. The sample was loaded on to a ZipTip C18 microcolumn (Millipore) and eluted with 60% acetonitrile and 0.1% formic acid for ana- lysis using a Q-TOF Ultima mass spectrometer (Waters Co., Milford, MA, USA). MS and MS ⁄ MS data were acquired and processed automatically using MassLynx 4.0 software (Waters Co.). Acknowledgements We thank Ms. Yuki Shinzaki for technical assistance. This study was supported in part by the Research Association for Biotechnology, which is subsidized by the Ministry of Economy, Trade and Industry of Japan. References 1 Asada K (1999) The water-water cycle in chloroplasts: scavenging of active oxygens and dissipation of excess photons. Annu Rev Plant Physiol Plant Mol Biol 50, 601–639. H 2 O 2 -mediated inactivation of stromal APX S. Kitajima et al. 3018 FEBS Journal 274 (2007) 3013–3020 ª 2007 The Authors Journal compilation ª 2007 FEBS 2 Hiner AN, Martinez JI, Arnao MB, Acosta M, Turner DD, Lloyd Raven E & Rodriguez-Lopez JN (2001) Detection of a tryptophan radical in the reaction of ascorbate peroxidase with hydrogen peroxide. Eur J Biochem 268, 3091–3098. 3 Shikanai T, Takeda T, Yamauchi H, Sano S, Tomizawa K, Yokota A & Shigeoka S (1998) Inhibition of ascor- bate peroxidase under oxidative stress in tobacco having bacterial catalase in chlorolasts. FEBS Lett 428, 47–51. 4 Yoshimura K, Yabuta Y, Ishikawa T & Shigeoka S (2000) Expression of spinach ascorbate peroxidase iso- enzymes in response to oxidative stresses. Plant Physiol 123, 223–233. 5 Miyake C & Asada K (1996) Inactivation mechanism of ascorbate peroxidase at low concentrations of ascorbate: hydrogen peroxide decomposes compound I of ascor- bate peroxidase. Plant Cell Physiol 37, 423–430. 6 Kitajima S, Tomizawa K, Shigeoka S & Yokota A (2006) An inserted loop region of stromal ascorbate per- oxidase is involved in its hydrogen peroxide-mediated inactivation. FEBS J 273, 2704–2710. 7 Asada K (1992) Ascorbate peroxidase: a hydrogen per- oxidase scavenging system in plants. Physiol Plant 85, 235–241. 8 Ishikawa T, Yoshimura K, Sakai K, Tamoi M, Takeda T & Shigeoka S (1998) Molecular characterization and physiological role of a glyoxysome-bound ascorbate per- oxidase from spinach. Plant Cell Physiol 39, 23–34. 9 Fox T, Tsaprailis G & English AM (1994) Fluorescence investigation of yeast cytochrome c peroxidase oxidation by hydrogen peroxide and enzyme activities of the oxi- dized enzyme. Biochemistry 33, 186–191. 10 Wada K, Tada T, Nakamura Y, Ishikawa T, Yabuta Y, Yoshimura K, Shigeoka S & Nishimura K (2003) Crys- tal structure of chloroplastic ascorbate peroxidase from tobacco plants and structural insights into its instability. J Biochem (Tokyo) 134, 239–244. 11 Spangler BD & Erman JE (1986) Cytochrome c peroxi- dase compound I: formation of covalent protein cross- links during the endogenous reduction of the active site. Biochim Biophys Acta 872, 155–157. 12 Sharp KH, Moody PC, Brown KA & Lloyd Raven E (2004) Crystal structure of the ascorbate peroxidase- salicylhydroxamic acid complex. Biochemistry 43, 8644–8651. 13 Smulevich G, Mauro JM, Fishel LA, English AM, Kraut J & Spiro TG (1988) Heme pocket interactions in cytochrome c peroxidase studied by site-directed muta- genesis and resonance Raman spectroscopy. Biochemis- try 26, 5477–5485. 14 Goodin DB, Davidson MG, Roe JA, Mauk AG & Smith M (1991) Amino acid substitutions at trypto- phan-51 of cytochrome c peroxidase: effects on coordi- nation, species preference for cytochrome c, and electron transfer. Biochemistry 30, 4953–4962. 15 Badyal SK, Joyce MG, Sharp KH, Seward HE, Mewies M, Basran J, Macdonald IK, Moody PC & Raven EL (2006) Conformational mobility in the active site of a heme peroxidase. J Biol Chem 281, 24512–24520. 16 Pipirou Z, Bottrill AR, Metcalfe CM, Mistry SC, Badyal SK, Rawlings BJ & Raven EL (2007) Autocata- lytic formation of a covalent link between tryptophan 41 and the heme in ascorbate peroxidase. Biochemistry 46, 2174–2180. 17 Kitajima S, Ueda M, Sano S, Miyake C, Kohchi T, Tomizawa K, Shigeoka S & Yokota A (2002) Stable form of ascorbate peroxidase from the red alga Gal- dieria partita similar to both chloroplastic and cytosolic isoforms of higher plants. Biosci Biotechnol Biochem 66, 2367–2375. 18 Dunford HB (1999) In Heme Peroxidase (Dunford HB, ed), pp. 219–251. John Wiley, New York, NY. 19 Pfister TD, Gengenbach AJ, Syn S & Lu Y (2001) The role of redox-active amino acids on compound I stabi- lity, substrate oxidation, and protein crosslinking in yeast cytochrome c peroxidase. Biochemistry 40, 14942– 14951. 20 Zhang H, He S & Mauk AG (2002) Radical formation at Tyr39 and Tyr153 following reaction of yeast cyto- chrome c peroxidase with hydrogen peroxide. Biochem- istry 41, 13507–13513. 21 Wright PJ & English AM (2003) Scavenging with TEMPO* to identify peptide- and protein-based radicals by mass spectrometry: advantages of spin scavenging over spin trapping. J Am Chem Soc 125, 8655–8665. 22 Piontek K, Smith AT & Blodig W (2001) Lignin peroxi- dase structure and function. Biochem Soc Trans 29, 111–116. 23 Jakopitsch C, Obinger C, Un S & Ivancich A (2006) Identification of Trp106 as the tryptophanyl radical intermediate in Synechocystis PCC6803 catalase-peroxi- dase by multifrequency electron paramagnetic resonance spectroscopy. J Inorg Biochem 100, 1091–1099. 24 Guallar V & Olsen B (2006) The role of the heme pro- pionates in heme biochemistry. J Inorg Biochem 100, 755–760. 25 Dalton DA, Hanus FJ, Russell SA & Evans HJ (1987) Purification, properties, and distribution of ascorbate peroxidase in legume root nodules. Plant Physol 83, 789–794. 26 Chen GX & Asada K (1989) Ascorbate peroxidase in tea leaves: Occurrence of two isozymes and the differ- ences in their enzymatic and molecular properties. Plant Cell Physiol 30, 987–998. 27 Mittler R & Zilinskas BA (1991) Purification and char- acterization of pea cytosolic ascorbate peroxidase. Plant Physiol 97, 962–968. 28 Miyake C, Cao WH & Asada K (1993) Purification and molecular properties of the thylakoid-bound ascorbate S. Kitajima et al. H 2 O 2 -mediated inactivation of stromal APX FEBS Journal 274 (2007) 3013–3020 ª 2007 The Authors Journal compilation ª 2007 FEBS 3019 peroxidase in spinach chloroplasts. Plant Cell Physiol 34, 881–889. 29 Kvaratskhelia M, George SJ & Thorneley RN (1997) Salicylic acid is a reducing substrate and not an effective inhibitor of ascorbate peroxidase. J Biol Chem 272, 20998–21001. 30 Yoshimura K, Ishikawa T, Nakamura Y, Tamoi M, Takeda T, Tada T, Nishimura K & Shigeoka S (1998) Comparative study on recombinant chloroplastic and cytosolic ascorbate peroxidase isozymes of spinach. Arch Biochem Biophys 353, 55–63. 31 Shigeoka S, Ishikawa T, Tamoi M, Miyagawa Y, Takeda T, Yabuta Y & Yoshimura K (2002) Regulation and function of ascorbate peroxidase isoenzymes. J Exp Bot 53, 1305–1319. 32 Tomita T, Tsuyama S, Imai Y & Kitagawa T (1997) Purification of bovine soluble guanylate cyclase and ADP-ribosylation on its small subunit by bacterial toxins. J Biochem (Tokyo) 199, 531–536. 33 Nakajima R & Yamazaki I (1979) The mechanism of indole-3-acetic acid oxidation by horseradish peroxidase. J Biol Chem 254, 872–878. H 2 O 2 -mediated inactivation of stromal APX S. Kitajima et al. 3020 FEBS Journal 274 (2007) 3013–3020 ª 2007 The Authors Journal compilation ª 2007 FEBS . Irreversible cross-linking of heme to the distal tryptophan of stromal ascorbate peroxidase in response to rapid inactivation by H 2 O 2 Sakihito Kitajima 1 , Taise Shimaoka 2 ,. isoform is irreversibly cross-linked to a tryptophan residue facing the distal cavity. Mutation of this tryptophan to phenylalanine abolished the cross-linking and increased the half-time for inactivation. the basis of these results, we propose that the rapid inacti- vation of tsAPX is at least partly due to repositioning of heme caused by cross-linking between heme and the distal tryptophan resulting