Thioredoxin Ch1 of Chlamydomonas reinhardtii displays an unusual resistance toward one-electron oxidation Ce ´ cile Sicard-Roselli 1 , Ste ´ phane Lemaire 2 , Jean-Pierre Jacquot 3 , Vincent Favaudon 4 , Christophe Marchand 5 and Chantal Houe ´ e-Levin 1 1 Laboratoire de Chimie Physique and 2 Institut de Biotechnologie des Plantes, Universite ´ Paris XI, Orsay, France; 3 UMR 1136 Interaction Arbres Microorganismes INRA UHP, Universite ´ de Nancy I, Vandoeuvre, France; 4 U 612 INSERM, Institut Curie, Centre Universitaire, Orsay, France; 5 Institut de Biochimie et Biophysique Mole ´ culaire et Cellulaire, CNRS UMR8619 and IFR46, Universite ´ Paris XI, Orsay, France To test thioredoxin resistance to oxidizing free radicals, we have studied the one-electron oxidation of wild-type thio- redoxin and of two forms with the point mutations D30A and W35A, using azide radicals generated by c-ray or pulse radiolysis. The oxidation patterns of wild-type thioredoxin and D30A are similar. In these forms, Trp35 is the primary target and is ÔrepairedÕ by one-electron reduction; first by intramolecular electron transfer from tyrosine, and then from other residues. Conversely, during oxidation of W35A, Trp13 is poorly reactive. For all proteins, activity is con- served showing an unusual resistance toward oxidation. Keywords: t hioredoxin; one-electron oxidation; radiolysis; tryptophan35 oxidation. Thioredoxins (Trx) are ubiquitous small proteins (100–120 amino acids) found in all living organisms from bacteria to vertebrates [1]. T hese proteins, whose active site contains the amino acid sequence -Cys-Gly-Pro-Cys-, exist either in an oxidized form with an intramolecu lar disulfide bond (Trx- S 2 ) o r in a reduced form with two thiol functions [Trx- (SH) 2 ]. They are involved in the reduction of disufide bonds and play a major role in the control of intracellular reduction potential and defense against oxidative stress. In addition, these proteins control t he release o f transcription factors NFKB and AP-1, and thus their oxidation state i s important in gene expression. During aerobic life, amino acid residues in proteins are subject to one-electron oxidation by reactive oxygen species, in suc h a way that the efficiency of cell defense against oxidative stress relies on the resistance of Trx to oxidation. Recently, Watson and Jones [2] showed that in cells both nuclear and cytoplasmic type 1 thioredoxins (Trx1) are relatively protected against oxidation and t hat the redox state of the cysteine residues in Trx1 was a good marker of oxidative stress. However, in addition to the cysteine residues o f t he active site, o ther amino acids can be oxidized by free radical p rocesses, which may induce modifications of the enzymatic properties of Trx. In proteins, one-electron oxidation is known to affect primarily Met, Tyr and Trp residues [3]. T he major d egradation products r esulting from such radical attack are dityrosine for tyrosine, N-formyl- kynurenin for tryptophan and methionine sulfoxide for methionine. Any of these transformations may affect the function of the enzyme and thus the redox homeostasy. The aim of this work was to determine the sensitivity of Trx toward one-electron oxidation. Therefore we studied the effect of overoxidation on Trx in its disulfide oxidized form (Trx-S 2 ) b y a zide radicals (N 3 _ ) u sing pulse and gamma radiolysis, and by measuring its enzymatic activity. Pulse and gam ma ra diolysis are c omplementary techniques. The first allows identification of transient radicals formed with their absorption spectra, and the second is used to oxidize protein solutions in greater quantity to perform analysis of the degradation products. With radiolysis, very specific radicals are generated quantitatively. Azide radicals are powerful one-electron acceptors formed by the reaction of N 3 – with the OH  radicals produced during irradiation of water solutions under N 2 O atmosphere [4]: N 2 O þ e À aq ! OH  þ OH À þ N 2 ð1Þ N À 3 þ OH ! N  3 þ OH À ð2Þ The reduction potential of N 3 _ is lower than that of OH • ,and the values are 1.3 V and 1.8 V vs. normal hydrogen electrode at neutral pH, respectively. N evetheless, N 3 _ is more selective and provides a simpler model of oxidation than OH  radicals allowing the determination of the main process of o xidation of OH • radicals, without all of the side- effects. They are known to r eact first with aromatic residues and m ost rapidly with tryptophan ( Reaction 3) [5,6]. Thus a well-known kinetic scheme is expected for the one-electron oxidation of proteins containing aromatic residues: N  3 þ HTrp-XX-TyrOH ! Trp  -XX-TyrOH þ N À 3 þ H þ ð3Þ Trp  -XX-TyrOH ! HTrp-XX-TyrO  ð4Þ Correspondence to C. Sicard-Roselli, Laboratoire de Chimie Physique, CNRS UMR 8000, Baˆ t. 350, Universite ´ Paris XI, F-91405 Orsay Cedex, France. Fax: +33 1 69 15 3 0 53, Tel.: +33 1 69 15 55 49, E-mail: cecile.sicard@lcp.u-psud.fr Abbreviations: Trx, thioredoxin; Trx1, type 1 thioredoxin; TyrOH, tyrosine phenolic group; Ty rO • , tyrosinyl radical; WT, wild-type. (Received 6 May 2004, revised 24 June 2004, accepted7July2004) Eur. J. Biochem. 271, 3481–3487 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04279.x Trp  -XX-TyrOH ! HTrp-XX  -TyrOH ð5Þ The tryptophan free radical Trp • resulting from Reaction 3 is then reduced by a t yrosine residue (Reaction 4) or by various amino acids (Reac tion 5) by a n intramolecular reaction. Thus, t he reaction ends up with dityrosine, among other final compounds. Trx possesses two tryptophan residues (positions 13 and 35). One of them (Trp35) is part of the conserved active site sequence, Trp-Cys-Gly-(Ala/Pro)-Pro-Cys-(Lys/Arg). Mutation of this residue affects the environment of the two C ys residues [7] an d the protein biochemical activity [8]. Two t yrosine (positions 53 and 85) an d two methionine residues (positions 41 and 79) are also present. Their role in enzymatic a ctivity is not known. It is particularly interesting to evaluate the sensitivity of Trp35 towards free radicals; therefore in addition to the wild-type thioredoxin h from the green alga Chlamydomonas reinhardtii, we oxidized the mutant form W35A. The aspartic acid residue at position 30, which is highly conserved in Trx from different species [9], is also crucial for general acid–base catalysis in the reductive opening of the disulfide oxidized thioredoxin [10,11], yet NMR studies have shown that the D30A mutant has the same global fold as the wild-type (WT) protein. In order to e valuate t he importance of this residue in oxidative processes we also d etermined the reaction of the m utant D30A with the N 3 _ radical. Materials and methods Proteins Recombinant Trx h from the green alga C. reinhardtii was purified from E. coli as described p reviously [12]. D30A and W35A mutants were prepared, also as described previously [13]. Samples for i rradiation were dialyzed several t imes against phosphate buffer (final buffer: 20 m M phosphate, 100 m M NaN 3 , p H 7). Concentrations of the three forms of Trx were adjusted to 7 7 l M (unless otherwise stated) u sing absorbance and e 278 ¼ 14 500 M )1 Æcm )1 for the WT a nd D30A mutant, and e 278 ¼ 8900 M )1 Æcm )1 for W35A. Tryptophan Tryptophan s olutions (500 l M ) were prepared in 2 0 m M phosphate, pH 7, 500 m M NaN 3 buffer. Tryptophan solu- tions containing tert-butanol contained t he same buffer components, with the addition of 500 m M tert-butanol. Gamma and pulse radiolysis experiments Gamma radiolysis experiments were performed using a panoramic 60 Co source (IL60PL, Cis-Biointernational, Saclay, France). A Fricke dosimeter [4] w as used to determine the dose rate. Pulse radiolysis was performed using the linear electron accelerator of the Curie Institute in Orsay [14]. The doses per pulse (200 ns duration, 5–15 Gy) were calibrated from the absorption o f the thio cyanate radical (SCNÞ  À 2 obtained by radiolysis of thiocyanate ion solution in N 2 O-saturated phosphate buffer { 10 m M KSCN, 10 m M phosphate, pH 7, G[(SCNÞ  À 2 ] ¼ 0.55 lmolÆJ )1 , e 472 ¼ 7580 M )1 Æcm )1 }[15]. All protein samples were prepared in 20 m M phosphate buffer, pH 7, containing 100 m M NaN 3 and saturated with N 2 O by flushing N 2 O gas for 1 h over the samples, avoiding bubbling in the solution. Absorption and fluorescence Absorption spectra were recor ded at room temperature with a PerkinElmer (k9) spectrophotometer. Fluorescence spectra were recorded on a FL111 Spex fluorimeter. Electrophoresis SDS/PAGE was performed using a 12% (w/v) acrylamide/ bisacrylamide gel with a Tris/Tricine buffer [16]. Reductive conditions were obtained by a dding 2-mercaptoethanol to the protein. Proteins bands were stained with Coomassie blue R-250. HPLC analysis HPLC was performed on a Beckman Gold 168 (Beckman Coulter, Aulnay, France) with diode array detection. The analytical column was a C4 r everse-phase column (150 · 4.6 mm, 5 lm). The mobile phase eluants used were: (A) 0.1% (v/v) trifluoroacetic acid in water; (B) 0.1% (v/v) trifluoroacetic acid and 7 0% (v/v) CH 3 CN. G radient elution used was 40–60% of B in 20 m in, 1 mLÆmi n )1 flow at room temperature. Mass spectrometry All spectra were acquired in positive-ion m ode on a Voyager DE-STR MALDI-TOF mass spectrometer (Applied Bio- systems, Courtaboeuf, France) equipp ed with a 337 nm nitrogen laser. Determination of the molecular masses of irradiated proteins was performed in linear mode (acceler- ating voltage 25 kV, grid voltage 93%, guide wire 0.3%, delay 600 ns) with external calibration. Freeze-dried fractions obtained from HPLC purification of irradiated Trx w ere d iluted with 15 lL o f 30% (v/v) CH 3 CN, 0.3% ( v/v) trifluoroacetic acid. One m illilitre of the solution was mixed with 4 lL of a saturated solution of sinapinic acid in 30% (v/v) CH 3 CN, 0.3% (v/v) trifluoro- acetic acid. F inally, 1.5 lL of t his premix w as deposited onto the sample p late and a llowed to d ry at room temperature. Activity measurements The activity of Trx was measured using the reduction of insulin [17]. One millilitre of t he following solution was prepared: 1 00 m M phosphate, p H 7.1, 1 30 l M human insulin (zinc form), 2 m M EDTA; to which 30 lLofa Trx solution (77 l M Trx, 20 m M phosphate buffer, pH 7, 100 m M NaN 3 ) was added t o obtain a final c oncentration of 2.5 l M of the protein. The experiment was started imme- diately after the addition of 500 l M dithiothreitol and the activity was monitored using the change in absorbance at 650 n m. A blank was made u sing the same conditions without adding Trx. All these experiments were carried out at 27 °C. 3482 C. Sicard-Roselli et al.(Eur. J. Biochem. 271) Ó FEBS 2004 Results Transients Aqueous solutions of Trx ( 77 l M WT, W35A or D30A , in 20 m M phosphate buffer, 100 m M NaN 3 pH 7) were irradiated by pulse radiolysis in the presence of NaN 3 under an atmosphere of N 2 O. Under t hese conditions the N 3 _ radical was the only o xidant species generated by radiolysis. The absorption s pectra of protein free radicals are shown in Fig. 1. Comparison of Fig. 1A and B (WT and D30A) shows that both p roteins yield a similar behavior. The spe c- trum obtained 50 ls after the pulse exhibited a broad absorption band at 510 n m, characteristic of the tryptophanyl radical (Trp • ) and a narrow peak at 410 nm reminiscent of a tyrosinyl radical (TyrO • ) [18]. The second- order r ate c onstants o f formation o f the Trp • radical, determined under pseudo fi rst-order conditions, were o f the same order of magnitude for both proteins (Table 1). The intensity of the 410 nm peak subsequently increased at the expense of the 510 nm band and reached a pseudo-plateau after approximately 300 ls. The rate constants for th is reaction measured at 510 and 410 nm were very close to each other, suggesting quantitative oxidation of TyrOH by Trp • , as for other p roteins. However, the intramolecular r ate constants of charge transfer d iffered b y a factor of t wo between WT Trx and the D30A mutant (Table 1). The Trp • and TyrO • yields were estimated from the magnitudes of the absorbance changes at 510 and 410 nm, assuming that the extinction coefficients of protein bound radicals are the same as those for free amino acids or peptides, namely, for Trp • 1800 and 300 mol )1 ÆLÆcm )1 at 510 and 410 nm, respectively; for TyrO • , 70 and 2600 mol )1 ÆLÆcm )1 at 510 and 410 nm respectiv ely [18,19]. Thus, the stoichiometry N  À 3 /Trp • was estimated to be equal to 1 : 1 and the percentage of transfer to around 60% in both compounds. This transfer is not total, as some tryptophan radical persists at the end of the reaction (approximately 300 ls after the pulse). Figure 1C shows the absorption spectrum of the W35A mutant form of Trx obtained 50 and 300 ls after the pulse. Among the transients t hat were formed by o xidation of W35A, the Trp • 510 n m broad band was not detected anymore; instead, a peak at 420 nm and weak bands at 390 and 480 nm appeared. The peak at 420 nm indicates that tyrosine could be oxidized directly by N  3 radicals. The rate constant of formation was determined under p seudo first- order c onditions (Table 1) and was substantially lower than in other proteins. The bands at 390 and 480 nm might belong to a methionyl residue. In general, N  3 radicals are unable to perform oxidization of methionine becau se the o ne-electron reduction potential of methionine is higher than that of N  3 . However it w as shown that interaction with other r esidues and particularly c arbonyl groups, may lower t he methionine redox potential considerably [20]. Here t he methionine radical would app ear as a n S-O complex with an a bsorption spectrum peaking around 390 nm [21] and/or interacting with another sulfur atom, giving an absorption spectrum with a maximum at 480 nm. Both types of radicals can e xist simultaneously in the same molecule [20]. Alternatively the 480 nm band might be assigned to a Trp • radical according to Joshi and Mukherjee [22,23]. These authors oxidized tryptophan by CCl 3 O 2 radicals and observed a blue-shift from 510 nm to 480 nm, which t hey a scribed t o a change in polarity of the environment. However, this blue-shift was formerly interpreted by Packer et al. as an a dduct o f Cl 3 COO • on the C2 or C3 of the indole ring of tryptophan [24]. I n o rder to properly assign the 480 nm b and, we investigated the effect o f the solvent polarity o n the absorption of a Trp radical. N  3 -induced pulse radiolys is oxidation of tryptophan was performed in phosphate buffer solution in the presence of tert -butanol (500 m M NaN 3 , 500 m M tert-butanol). The presence of tert-butanol induced a b roadening and a red-shift of the absorption f rom 510 nm to 540 nm (not shown). Therefore, assigning the 480 nm band to Trp • appears to be unlikely. Analysis of final compounds Three analytical methods were used to gain insight into the nature o f the oxidized forms o f T rx generated by c-ray radiolysis up to 900 Gy. Fig. 1. Differential absorption spectra of WT, D30A and W35 Trx protein free radicals obtained by pulse r adiolytically generated N 3 • after 50 and 30 0 ls. Proteins were 77 l M throughout. The optical path was 2 cm. (A) WT Trx; (B) D30A; (C) W35A. The dose was 6.5 Gy. Ó FEBS 2004 One-electron oxidation resistance of thioredoxin Ch1 (Eur. J. Biochem. 271) 3483 Firstly, reducing and nonreducing SDS/PAGE analysis was carried out on the three forms of oxidized Trx. WT Trx (Fig. 2) and W35A exhibit a single band corresponding to the mass (12 kDa) expected from intact Trx. For D30A, two new higher molecular mass bands are generated in a dose-dependent manner, suggesting t he formation of aggre- gates. Secondly, UV absorption spectra of the three forms of Trx were recorded before and after one-electron oxidation, as tyrosine dimers can be evidenced with an absorption at 315 nm coupled to a fluorescence band at 410 nm. A new 315 nm a bsorption band i ncreased with dose for t he W35A mutant only. In addition to absorption, fluorescence analysis of the three forms of Trx after radiolytic oxidation was performed. Excitation at 315 nm induced a broad fluorescence band at 410 n m for the W35A mutant (Fig. 3) while no new signal could be seen for WT and D30A Trx. Thirdly, liquid chromatography was performed to isolate the degradation products of WT Trx, D30A and W35A after oxidation with a dose of 100 Gy. For each form, the chromatograms were very similar and showed the forma- tion of a s ingle major product. The y ield of formation of t his product (W35A, 47 nmolÆJ )1 ; WT, 35 nmolÆJ )1 ; D 30A, 40 nmolÆJ )1 ) w as calculated using the area of each peak assuming that the sum of both peak areas represents 100%. The product was isolated and analyzed using mass spectr- ometry. For D30A and WT Trx, no difference between the mass of the intact protein and the oxidized one could be detected (Table 2). I n the case of W35A, a small increase o f themass(<40Da)wasevidenced. Enzymatic activity Enzymatic activity of the three Trx was measured before and after exposure to 100 Gy (Fig. 4). Firstly, unirradiated D30A had the highest activity, compared to the WT and W35A forms. Secondly, as expected, the activity of W35A compared to that of WT Trx was reduced by a factor of approximately two [25]. T hirdly, a ctivity of D 30A was weakly modified only after irradiation w ith 100 Gy, while that of WT and of W35A increased by a factor of 1.5 and 1.2, respectively (Table 3). Discussion As already observed for several proteins or p eptides [6,26– 29], in WT and D30A the N  3 radical first oxidizes a tryptophan residue that is subsequently reduced in the course of intramolecular charge transfer to a tyrosine residue (Reactions 3 and 4). The rate constants of Reaction 3 are in th e same range as for tryptophan residues in other proteins (Table 1). The reaction between N  3 radicals and the Trp residue is stoichiometric, as reported for other proteins [6]. Mutating Asp30 by replacement with an alanine did not affect this reaction. Because the 510 nm band is missing in W35A, we suggest that in the case of WT and D30A, Trp35 is the residue oxidized by N  3 . Evolution of the transient absorption spectra indicates that Trp3 5 i s p artly ÔrepairedÕ by a t yrosine residue acting as a one-electron donor. Trx contains two tyrosines. From the known structure of WT and D30A Trx [30], the distances Table 1. Rate c onstants of the reaction of N 3 • with WT Trx , D30A an d W35A. These values a re compared to those proposed for other peptides o r proteins. Data are an average of values g iven in [29,30,43]. The distance is taken from Figs 4 and 5 of [43]. k (Reaction 3) (mol )1 ÆLÆs )1 ) Intramolecular rate constants (s )1 ) k reaction Trp-Tyr distance of the couple involved (A ˚ ) WT Trx (1.1 ± 0.2) · 10 9 (9.9 ± 0.4) · 10 3 18.4 (Trp35-Tyr85) D30A (1.2 ± 0.3) · 10 9 (4.4 ± 1.7) · 10 3 18.7 (Trp35-Tyr85) W35A (0.46 ± 0.01) · 10 9 Hen egg white lysozyme (7.9 ± 0.8) · 10 8 120 ± 10 [6] 14 (Trp62/63-Tyr53) Trp-(Pro) 3 -Tyr % 5 · 10 9 [31,43] a (1.5–2.3) · 10 3 [29,31,43] % 8 [43] Trp-(Pro) 4 -Tyr % 5 · 10 9 [31,43] a 5.13 · 10 2 [43] % 11 [43] Trp-(Pro) 5 -Tyr % 5 · 10 9 [31,43] a 3.05 · 10 2 [43] % 14 [43] a Authors gave only the order of magnitude of this rate constant. Fig. 2. SDS/PAGE analysis in the presence of 2-mercaptoethanol (12% acrylamide gel) of Trx after c-ray ir radiation (up to 900 Gy). Left : WT Trx; middle: W 35A; right: D30A. For ea ch lane, 2 lg of protein was used. The molecular mass standards used for W35A and D30A are shown on the gels. For D30A, the arrows point out two new bands increasing with the dose. 3484 C. Sicard-Roselli et al.(Eur. J. Biochem. 271) Ó FEBS 2004 between Trp35 and Tyr (53 and 85) were calculated at 29.8 and 18.4 A ˚ , respectively (Table 1). We therefore propose that th e tyrosine residue involved in the i ntramolecular Trp • fi Tyr transfer i s Tyr85. These distances are much larger than in the other proteins or peptides investigated to date (Table 1), y et the rate constant of this step is also much higher [5,6,31]. It is currently agr eed that the rate constant of long-range intramolecular electron transfer decays expo- nentially with the donor–acceptor distance [32,33] and that this dependence would be the same for all proteins and peptides [34]. O ur results c learly demonstrate t hat this correlation does not apply for Trx. For D 30A, the intramolecular rate constant is reduced by a factor of two (Table 1) compared with that for the WT protein although the distances between Trp35 and the tyrosine residues are the same. Sakata et al. [35] suggested that a change in orientation of the donor/acceptor residues could induce a slower rate constant. Whether this effect would be due to electrostatic changes s uch as d ipole moment d ifferent orientation or to modifications of the structure of the solvation layer is not known. Here, no significant difference between the Tyr and Trp residues orientation could be demonstrated by superimposing their respective crystallo- graphic structures. Asp30 has an important role i n driving the hydrogen bond network linking its carboxylic group to the active s ite. Changes in t he kinetics of intramolecular electron t ransfer b y mutation of Asp30 could thus b e a consequence of hydrogen bond rearrangement at the active site. Weak bands at 390 and 480 nm, but no 510 nm band that could be assigned to Trp, was observed in the case of W35A (Fig. 1 C). Several explanations were proposed for the band at 480 nm. Such a blue-shift was observed and assigned to Trp • radical i n c asein a nd bovine serum albumin as these proteins w ere transferred to a s olvent of low polarity inducing changes in the protein environment and conformation [22,23]. Under our conditions, a decrease in solvent po larity by addition of tert-butanol did not produce any shift of the T rp • absorption ban d. W e t hus propose that the bands at 390 and 480 nm could be related to oxidation of methionine residues. Indeed, Trx possesses two methio- nine residues at positions 41 and 79. Met79 is close to the carbonyl function of Phe31 (less than 4 A ˚ ). Hence its reduction potential could be lower than that of Met41 which is in a polar environment with solvent access [ 20] allowing Met79 o xidation by N  3 . T he end product would be a MetS + radical, which, in interaction with the oxygen atom of the carbonyl function, would lead to Met S–O radical absorbing at 390 nm [36]. In addition Met79 is at 5.2 A ˚ from the sulfur atom of Cys39 and c ould form a Met S–S + radical absorbing at 480 nm. Final products are different for t he three p roteins. Aggregation was observed only for D30A, for doses above 100 Gy. This aggregation was als o seen with electrophoresis under reducing conditions, w hich excludes t he formation o f a disulfide bond . Surprisingly, aggregation did not correlate with the appearance of 315 nm absorption/420 nm fluores- cence bands, a s could b e e xpected for c ovalent t yrosine dimerization [37,38]. Therefore, dityrosine is unlikely to be formed. We propose that the polypeptide chain can also take par t in the one-electron processes. For example, in lysozymes oxidation of tryptophan residues leads to poly- peptide bond cleavage [6]. Also, in hen egg white lysozyme, one-electron reduction of the 6-127 disulfide bond leads to peptide bond cleavage [39]. W e therefore propose hydrogen Table 2. Mass analysis of intact and oxidized thioredoxin after separ- ation using liquid chromatography. Intact protein Oxidized protein WT Trx 11711 ± 10 11713 ± 10 D30A 11677 ± 10 11676 ± 10 W35A 11638 ± 10 11606 ± 10 Table 3. Activity of the different forms of Trx before and after oxida- tion. Thioredoxin Non irradiated DA 650 Æmg )1 Æmin )1 Irradiated 100 Gy DA 650 Æmg )1 Æmin )1 WT 0.79 0.97 W35A 0.44 0.67 D30A 1.60 1.58 Fig. 3. Fluorescence s pe ctrum of W35A and D30A (inset) (77 l M ,20m M phosphate buffer pH 7) recorded at room temperature with excitation a t 31 5 nm. Doses for W35A: 0 Gy; 403 Gy and 646 Gy. Doses for D30A (inset): 0 Gy and 100 Gy. Ó FEBS 2004 One-electron oxidation resistance of thioredoxin Ch1 (Eur. J. Biochem. 271) 3485 loss from a Ca atom followed by aggregation through the carbon-carbon bond. This process occurs easily in polymers [4] and glycine r esidues are good candidates [40]. In thioredoxin, oxidized forms other than aggregates are formed. Oxidation of tyrosine residues in aqueous solution and in t he presence of oxygen by OH radicals leads to formation of 3 ,4-dihydroxyphenylalanine. However, th is requires oxygen (which is absent in our system) and therefore, we can exclude the formation of 3,4-dihydroxy- phenylalanine in the three forms of Trx after radiolytic oxidation. No aggregate was formed from the W35A mutant but a new fluorescent product appeared after oxidation. As this 420 nm fluorescence band is not due to dimerization, it could arise from a d egradation product of the Trp13 residue. Mass spectrometry indicates an increase of mass lower than 4 0 Da. This and fl uore scence experiments suggest t hat one-electron o xidation of W35A could produce N-fo rmylkynurenin at position 1 3. It would mean that Trp13 is involved in t he final step of the one-electron oxidation of W35A by azide. Although many examples of enzyme i nactivation have already been reported (reviewed in [41,42]), no inactivation of Trx was shown to result from oxidation. This means t hat no amino acid involved i n insulin reduction activity is irreversibly affected by oxidation and that the tertiary structure is not altered to a large extent. Moreover, an opposite effect is observed for W35A, i.e. this Trx is found to be more active after a 100 Gy irradiation. Conclusion The use of azide r adicals g enerated by pulse radiolysis allowed us to determine the reactivity of WT and two mutant forms, D30A and W35A, of T rx toward one-electron o xidants. We were particularly interested in the reactivity of the Trp35 residue, as this is h ighly conserved in the active site and may be a part of t he defence o f livin g organisms toward reactive oxygen species. We show h ere that oxidation of Trx (WT and D30A) with N  3 occurs first at the Trp35 residue. The Trp • radical s ubsequently undergoes intramolecular reduction by a tyrosine residue. The tyrosine residue involved in this transfer is probably Tyr85. Such intramolecular electron transfer thus protects Trp35, and hence the enzyme’s biological activity, i n cases of oxidative stress. When Trp35 is absent, the other Trp residue (Trp13) may be oxidized, however, indirectly through tyrosine and/or methionine oxidation followed by intramolecular electron transfer. This suggests that protection against oxidation is not due to the accessibility of s ensitive residues to free radicals, but rather to some kind of ÔrepairÕ through long range intramolecular electron transfer. The main point is that no degradation of Trx was observed. It means that in the case of oxidation stress, if thioredoxin reductase is active, oxidized thioredoxin can be r ecycled and the defenses of the cells are not affected. References 1. Holmgren, A. (1985) Thioredoxin. Annu. Rev. Biochem. 54, 237– 271. 2. Watson, W.H. & Jones, D.P. (2003) Oxidation of nuclear thior- edoxin during o xid ative stre ss. FEB S Let t. 543, 144–147. 3.Davies,M.J.,Fu,S.,Wang,H.&Dean,R.T.(1999)Stable markers of oxidant damage to proteins and their application in the study of human disease. Free Rad. Biol. Med. 27, 1151– 1163. 4. Spinks, J.W.T. & Woods, R.J. (1990) Introduction to Radiation Chemistry, 3rd edn. Wiley Interscience, New York. 5. Weinstein, M., Alfassi, Z.B., De Felippis, M.R. & Faraggi, M. (1991) Long range electron transfer between tyrosine an d trypt o- phan in hen egg-white lysozyme. Biochim. Biophys. Acta 1076, 173–178. 6. Stuart-Audette, M., Blouquit, Y., Faraggi, M., Sicard-Roselli, C., Houe ´ e-Levin, C. & Jolle ` s, P. (2003) Re-evaluation of intramolecular long range electron transfer between tyrosine and tryptophan in lysozymes. Evidence for the participation of other residues. Eur. J. Biochem. 270, 1–7. 7. M en chi se, V., Corbier, C., Didierjea n, C., Jacquot, J P., Bened- etti, E., Saviano, M. & Aubry, A. (2001) Crystal structure of the W35A mutant thioredoxin h from Chlamydomonas reinhardtii:the substitution of the conserved active site Trp leads to modifications in the environment of th e two catalytic cysteines. Biopolymers 56, 1–7. 8. Krimm, I., Lemaire, S., Ruelland, E., Miginiac-Maslow, M., Jacquot, J P., Hirasawa, M., Knaff, D.B. & Lancelin, J M. (1998) The single mutation Trp35 fi Ala in t he 35–40 redox site of Chlamydomonas reinhardtii thioredoxin h affects its biochemical activity and the pH dependence of C36–C39 1 H- 13 CNMR.Eur. J. Biochem. 255, 185–195. 9. Eklu nd, H., Gleason, F.K. & Holmgren, A. (1991) Structural and functional relations among thioredoxins of different species. Proteins 11, 13–28. 10. Chivers, P.T. & Raines, R.T. (1997) General acid/base catalysis in theactivesiteofEscherichia coli thioredoxin. Biochemi stry 36, 15810–15816. 11. Dyson, H.J., Jeng, M.F., Tennant, L.L., Slaby, I., Lindell, M., Cui, D.S., Kuprin, S. & Holmgren, A. (1997) Effects of buried Fig. 4. Activity determination of the three forms of Trx using insulin reduction. Each sample was assembled in 1 mL of 100 m M phosphate, pH 7.1, with 130 l M human insulin (zin c form), 2 m M EDTA and 30 lL of a Trx solution (77 l M Trx, 20 m M phosphate buffer, pH 7, 100 m M NaN 3 )toobtainafinalconcentrationof2.5l M Trx. Non- irradiated W35A mutant, d; W35A after 100 Gy, s; non-irradiated WT Trx, j; WT Trx after 100 Gy, h; non-irradiated D30A mutant, m; D 30A a fter 1 00 Gy, n. 3486 C. Sicard-Roselli et al.(Eur. J. Biochem. 271) Ó FEBS 2004 charged groups on cysteine thiol ionization and reactivity in Escherichia coli thioredoxin: structural and functional character- ization of mutants of Asp 26 and Lys 57. Biochemistry 36, 2622– 2636. 12. Stein, M., Jacquot, J P., Jeannette, E., Decottignies, P., Hodges, M., Lancelin, J.M., Mittard, V., Schmitter, J.M. & Miginiac- Maslow, M. ( 1995) Chlamydomonas reinhardtii thioredoxins: structure of the genes coding for the chloroplastic m a nd cytosolic h isoforms; expression in Esc herichia coli of the r ecombinant proteins, purification and biochemical properties. Plant Mol. Biol. 28, 487–503. 13. El Hanine Lmoume ` ne,C.,Conte,D.,Jacquot,J P.&Houe ´ e- Levin, C. (2000) Redox properties of protein disulfide bond in oxidized thioredoxin and lysozyme: a pulse radiolysis study. Bio- chemistry 39, 9295–9301. 14. Favaudon, V., Tourbez, H., Houe ´ e-Levin, C. & Lhoste, J M. (1990) CO  2 - radical induce d cleavage o f disulfide b onds in p ro- teins. A gamma-ray and p ulse rad i olysis mechanist ic investigation. Biochemistry 29, 10978–10989. 15. Buxton, G.V. & Stuart, C.R. (1995) Reevaluation of the thio- cyanate dosimeter for pulse radiolysis. J. Chem. Soc. Faraday Trans. 91, 279–281. 16. Scha ¨ gger, H. & von Jagow, G. (1987) Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range of 1–100 kDa. Anal. Biochem. 166, 368–379. 17. Holmgren, A. (1979) Thioredoxin c atalyzes the reduction of insulin disulfides by dithiothreitol and dihydrolipoamide. J. Biol. Chem. 254, 9627–9632. 18. Bensasson, R.V., Land, E.J. & Truscott, T.G. (1983) Flash Pho- tolysis and Pulse Radiolysis. Pergamon Press, Oxford. 19. Dudley, F.D., Santus, R. & Grossweiner, L.I. (1975) Laser flash photolysis o f aqueous trypt ophan. J. Phy s. Chem. 79 , 2711–2716. 20. Pogocki, D. & Schoneich, C. ( 2002) Redox p roperties of M et (35) in ne urotoxic beta-amyloid pe ptide . A molecular modeling study. Chem.Res.Toxicol.15, 408–418. 21. Schon eich, C., Pogocki, D.,Wisniowski,P.,Hug,G.L.& Bobrowski, K. (2000) Intramolecular sulfure-oxygen bond for- mation in radical cations of N-acetylmethionine amide. J. Am. Chem. Soc. 122, 10224–10225. 22. Joshi, R. & Mukherjee, T. (2003) Effect of solven t viscosity, polarity and pH on the charge transfer between tryptophan radical and tyrosine in bovine serum albumin: a pulse radiolysis study. Biophys. Chem. 103, 89–98. 23. Joshi, R. & Mukherjee, T. (2002) Charge transfer between tryp- tophan and tyrosine in casein: a pulse radiolysis study. Biophys. Chem. 96, 15–19. 24. Packer, J.E., Mahood, J.S., Willson, R.L. & Wolfenden, B.S. (1981) Reactions of the trichloromethylperoxy free radical (Cl 3 COO • ) with tryptophan, tryptophanyl-tyrosine and lysozyme. Int. J. Radiat. Biol. 39, 135–141. 25. Krause, G. & Holmgren, A. (1991) Substitution of the conserved tryptophan 31 in Escherichia c oli thioredoxin b y site-directed mutagenesis and structure-function analysis. J. Biol. Chem. 266, 4056–4066. 26. Pru ¨ tz, W.A. & Land, E.J. (1979) Charge transfer i n peptides. Pulse radiolysis investigation of one-electron reactions in dipeptides of tryptophan and t yrosine. Int. J. Radiat. B iol. 36 , 513–520. 27. Pru ¨ tz, W.A., Siebert, F., Butler, J., Land, E.J., Menez, A. & Montenay-Garestier, T. ( 1982) Charge transfer in peptides. Intramolecular radical t ransformations involving methionine, tryptophan and t yrosine. Bioc him. Biophys. Acta 70 5, 139–149. 28. Bobrowski, K., P oznanski, J., Holcman, J. & Wierzchowski, K.L. (1998) Long-range electro n transfer between p roline-bridged aromatic amino acids. Adv. Chem. Series, Photochem. Radiat. Chem. 254, 131–143. 29. DeFelippis, M.R., Faraggi, M. & Klapper, M.H. (199 0) Evidence for thro ugh-bond long range electron transfer i n peptides. J. Am. Chem. Soc. 112, 5640–5642. 30. Menchise, V., Corbier, C., Didierjean, C., Saviano, M., Benedetti, E.,Jacquot,J P.&Aubry,A.(2001)Crystalstructureofthewild- type and D30A mutant thioredoxin h of Chlamydomonas reinhardtii and i mplications for the catalytic mechanism. Biochem. J. 359, 65–75. 31. Faraggi, M., DeFelippis, M.R. & Klapper, M.H. (1989) Long- range electron transfer between tyrosineandtryptophaninpep- tides. J. Am. Chem. Soc. 111 , 5141–5145. 32. Marcus, R.A. & Sutin, N. (1985) Electron transfers in chemistry and biology. Biochim. Biophys. Acta 811, 265–322. 33. Hush, N.S. (1985) Dist ance de pendence of electron transfer rates. Coord. Chem. Rev. 64, 135–157. 34. McLendon, G. ( 1988) Long-distance electron t ransfer in proteins and model systems. Acc. Ch em. R es. 21 , 1 60–167. 35. Sakata, Y ., Nakashim a, S., G oto, Y., Tatemitsu, H., Misumi, S., Asahi,T.,Hagihara,M.,Nishikawa,S.,Okada,T.&Mataga.N. (1989) Synthesis of completely fixed porphyrin-quinone com- pounds and the mutual orientation effect on electron transfer. J. Am. Chem. Soc. 111, 8979–8981. 36. Scho ¨ neich, C., Pogoski, D., Hug, G.L. & Bobrowski, K. (2003) Free radical reactions of methionine in peptides: Mechanisms relevant to b-amyloid oxidation and Alzheimer’s disease. J. Am. Chem. Soc. 125, 13700–137013. 37. Malencik, Q .A., Spro use, J.F., S wanson, C.A. & An derson, S .R. (1996) Dityrosine: Preparation, isolation and analysis. Anal. Bio- chem. 242, 202–213. 38. Audette , M., Blouquit, Y. & Houe ´ e-Levin, C. (2000) Oxidative dimerisation of proteins: roleoftyrosineaccessibility.Arch. Bio- chem. Biophys. 76, 673–681. 39. Berge ` s,J.,Kassab,E.,Conte,D.,Adjadj,E.&Houe ´ e-Levin, C. (1997) Ab i nit io calculations on arginine-disulfide complexes modelling the one-electron reduction o f lysozyme. Co mparison to an experimental reinvestigation. J. Phys. Chem. 101, 7809–7881. 40. Rauk, A., Yu, D., Taylor, J., Shustov, G.V., Block, D.A. & Armstrong, D.A. (1999) Effects of structure on alpha C-H bond enthalpies of amino acid residues: relevance to H transfers in enzyme mec hanisms a nd in protein oxidation. Biochemistry 38, 9089–9096. 41. Houe ´ e-Levin, C. & Sicard-Roselli, C. (2001) In Radiation Chem- istry: Present Status and Future Prospects (Jonah,C.&Rao,B.M., eds), pp. 553–584. Elsevier, Amsterdam. 42. Von Sonntag, C . (1987) The Chemical Basis of Radiation Biology. Taylor & Francis, London. 43. Bobrowski, K., Holcman, J., Poznanski, J., Ciurak, M. & Wierzchowski, K.L. (1992) Pulse radiolysis studies of intra- molecular electron transfer in model peptides and proteins. 5. Trp Æ to Tyr Æ radical transformation in H-Trp-(Pro) n -Tyr-OH series of peptides. J. Phys. Chem. 96 , 1 0036–10043. Ó FEBS 2004 One-electron oxidation resistance of thioredoxin Ch1 (Eur. J. Biochem. 271) 3487 . Thioredoxin Ch1 of Chlamydomonas reinhardtii displays an unusual resistance toward one-electron oxidation Ce ´ cile Sicard-Roselli 1 , Ste ´ phane. Gy. Ó FEBS 2004 One-electron oxidation resistance of thioredoxin Ch1 (Eur. J. Biochem. 271) 3483 Firstly, reducing and nonreducing SDS/PAGE analysis was carried