Tài liệu Báo cáo khóa học: Inactivation of copper-containing amine oxidases by turnover products doc

Inactivation of copper-containing amine oxidases by turnover products Paola Pietrangeli 1 , Stefania Nocera 1 , Rodolfo Federico 2 , Bruno Mondovı ` 1 and Laura Morpurgo 1 1 Department of Biochemical Sciences ‘A. Rossi Fanelli’ and C.N.R. Institute of Molecular Biology and Pathology, University of Rome ‘La Sapienza’, Rome, Italy; 2 Department of Biology, 3rd University of Rome, Rome, Italy For bovine serum amine oxidase, two different mechanisms of substrate-induced inactivation have been proposed. One consists of a slow oxidation by H 2 O 2 of a conserved residue in the reduced enzyme after the fast turnover phase [Pietr- angeli, P., Nocera, S., Fattibene, P., Wang, X.T., Mondovı ` , B. & Morpurgo, L. (2000) Biochem. Biophys. Res. Commun. 267, 174–178] and the other of the oxidation by H 2 O 2 of the dihydrobenzoxazole in equilibrium with the product Schiff base, during the catalytic cycle [Lee, Y., Shepard, E., Smith, J., Dooley, D.M. & Sayre, L.M. (2001) Biochemistry 40, 822–829]. To discriminate between the two mechanisms, the inactivation was studied using Lathyrus cicera (red vetch- ling) amine oxidase. This, in contrast to bovine serum amine oxidase, formed the Cu + -semiquinolamine radical with a characteristic UV-vis spectrum when oxygen was exhausted by an excess of any tested amine in a closed cuvette. The inactivation, lasting about 90 min, was simultaneous with the radical decay and with the formation of a broad band (shoulder) at 350 nm. No inactivation occurred when a thousand-fold excess of amine was rapidly oxidized in an L. cicera amine oxidase solution stirred in open air. Thus, the inactivation is a slow reaction of the reduced enzyme with H 2 O 2 , following the turnover phase. Catalase protected L. cicera amine oxidase from inactivation. This effect was substrate-dependent, varying from full protection (benzyl- amine) to no protection (putrescine). In the absence of H 2 O 2 , a specific inactivating reaction, without formation of the 350 nm band, was induced by some aldehydes, notably putrescine. Some mechanisms of inactivation are proposed. Keywords: copper amine oxidase; trihydroxyphenylalanine quinone; inactivation; hydrogen peroxide; aldehydes. Copper-containing amine oxidases [amine:oxygen oxido- reductase (deaminating) (copper containing); E.C.1.4.3.6] are ubiquitous enzymes that catalyze the oxidative deami- nation of primary amines, transferring two electrons to molecular oxygen in a ping-pong reaction producing H 2 O 2 , aldehydes, and ammonium ions [1,2]. E ox þ R-CH 2 -NH þ 3 ! E red -NH þ 3 þ R-CHO E red -NH þ 3 þ O 2 ! E ox þ NH þ 4 þ H 2 O 2 Together with copper, they contain an organic prosthetic group reactive with semicarbazide, phenylhydrazine and similar inhibitors of catalytic activity. This prosthetic group was identified [3] as trihydroxyphenylalanine quinone (TPQ), a post-translationally oxidized tyrosine residue [4]. It has been known for some time that copper amine oxidases are inactivated by the turnover product, H 2 O 2 ,as the presence of catalase protects the enzyme from inactiva- tion. This was described for the diamine oxidases from pea seedling [5] and pig kidney [6], and for the bovine serum amine oxidase (BSAO) [7]. In the latter case, it was not possible to identify the modification induced by H 2 O 2 , but the similarity of the behavior of several amine oxidases suggested that it consists of the oxidation of a conserved residue at the active site [7]. Tryptophan or metal-coordi- nated histidine residues, oxidized by H 2 O 2 in copper- and manganese-containing superoxide dismutases, were not affected in BSAO [7]. A more recent report [8] ascribed BSAO inactivation by benzylamines to a partitioning reaction, occurring during the catalytic cycle, between H 2 O mediated hydrolysis of the product Schiff base, and H 2 O 2 mediated oxidation of dihydrobenzoxazole in equi- librium with it, yielding aldehyde and benzoxazole, respect- ively. Inactivation by aldehydes is well documented for plant amine oxidases, such as lentil seedling amine oxidase treated with stoichiometric amounts of tryptamine under anaerobic conditions [9], or treated under turnover condi- tions with haloamines, 1,2-diaminoethane and 1,3-diamino- propane [10], and with the mechanism-based inhibitor, 2-butyne-1,4-diamine [11]. The focus of the present work was whether amine oxidase inactivation was due to H 2 O 2 reacting with the reduced protein after the turnover phase [7] or with Correspondence to P. Pietrangeli, Department of Biochemical Sciences ‘A. Rossi Fanelli’, University ‘La Sapienza’, P. le A. Moro 5, 00185 Rome, Italy. Fax: + 39 06 4440062, Tel.: + 39 06 49910639, E-mail: paola.pietrangeli@uniroma1.it Abbreviations: LCAO, Lathyrus cicera amine oxidase; BSAO, bovine serum amine oxidase; TPQ, 2,4,5-trihydroxyphenylalanine quinone; AGAO, Arthrobacter globiformis amine oxidase; HPAO, Hansenula polymorpha amine oxidase; ECAO, Escherichia coli amine oxidase. Enzyme: amine:oxygen oxidoreductase (deaminating) (EC 1.4.3.6). Dedication: This paper is dedicated to the memory of Eraldo Antonini, eminent biochemist, who died prematurely 20 years ago on March 19th, 1983. (Received 18 September 2003, revised 6 November 2003, accepted 10 November 2003) Eur. J. Biochem. 271, 146–152 (2004) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03913.x dihydrobenzoxazole produced by partitioning during the catalytic cycle [8]. The copper-containing amine oxidase purified from Lathyrus cicera (red vetchling) seedling (LCAO)wasusedasitformsCu 2+ -quinolamine in equilibrium with Cu + -semiquinolamine under reducing conditions as do other plant amine oxidases [2]. The peculiar spectroscopic properties of the latter radical allowed the inactivation process to be followed, while BSAO was spectroscopically silent under similar conditions. The first hypothesis [7] received supporting evidence and in addition LCAO could also be inactivated by the aldehyde produced by some substrates in the absence of H 2 O 2 . Materials and methods Protein purification LCAO was purified from L. cicera seedlings using a simple procedure that utilizes only two chromatographic steps (Table 1). Seeds, obtained from the local market, were soaked in aerated tap water for 12 h and grown in moistened vermiculite for 7 days in the dark at 23 °C. Seedling shoots (400 g) were homogenized in a Waring blender with 4 vols of 50 m M KH 2 PO 4 , pH 4.3. At this pH, most of the amine oxidase activity (> 90%) associated with cell walls and fibres was solubilized by increasing the ionic strength. The homo- genate was then strained through cheese cloth and the solid residue was washed four times with 3 vols of the same buffer. The enzyme was then eluted with 1 vol of 20% saturated ammonium sulphate in 50 m M KH 2 PO 4 ,pH4.3.The suspension was pressed through cheese cloth, and centri- fuged at 15 000 g for 30 min. The supernatant was brought to pH 7.0 with KOH, and to 70% saturation with solid ammonium sulphate, then it was stirred for 30 min and centrifuged at 15 000 g for 30 min. Although the purification was less than twofold, the advantage of this procedure was a substantial volume reduction. The precipitate was collected, resuspended in 0.2 vols of 15 m M potassium phosphate buffer, pH 7 and dialyzed overnight against the same buffer. The dialysate was loaded onto a DEAE–cellulose (What- man) column (20 · 4 cm i.d.) equilibrated with 15 m M potassium phosphate pH 7.0. During loading and washing with buffer, the eluate at A 280 > 0.1 was collected. As the enzyme did not bind, the solution, containing more than 80% of the total loaded amine oxidase activity, was adjusted to pH 5.5 with 1 M H 3 PO 4 and applied directly onto a SP Hi Trap (Pharmacia) column (5 · 0.8 cm i.d.) equilibrated with 50 m M potassium phosphate buffer, pH 5.5. The column was washed with the same buffer and also with the same buffer containing 0.1 M NaCl, then the amine oxidase was eluted using buffer containing 0.2 M NaCl. Fractions with high enzymatic activity were pooled and analyzed. Activity and protein assays The purified proteins moved as single bands on SDS/ PAGE. The concentration was measured by employing the molar extinction coefficients reported for the pea seedling enzyme (PSAO) [12], namely e 280nm ¼ 300 000 M )1 Æcm )1 and e 500nm ¼ 4900 M )1 Æcm )1 (Results). The copper content was assayed by atomic absorption spectrometry with a Perkin Elmer apparatus equipped with a HGA-400 graphite furnace and by the biquinoline spectrophotomet- ric method [13]. The amine oxidase activity was assayed spectrophotometrically at 25 °C with 1.0 m M putrescine in 0.1 M potassium phosphate buffer, pH 7.2, by measuring the formation of H 2 O 2 from the absorbance of the pink adduct (e 515nm ¼ 2.6 · 10 4 M )1 Æcm )1 ) produced by the horseradish peroxidase catalyzed oxidation of aminoanti- pyrine, followed by condensation with 3,5-dichloro-2- hydroxybenzensulfonic acid [14]. Samples with specific activity ¼ 70 IUÆmg )1 (micromoles of substrate oxid- izedÆper min) were employed. The vis-UV spectra were recorded with an AVIV (Lakewood, NJ, USA) spectro- photometer, Model 14 DS, equipped with a thermostatted cell holder. Chemicals Amines, aminoantipyrine, 3,5-dichloro-2-hydroxybenzen- sulfonic acid, catalase, horse radish peroxidase were purchased from Sigma Chemical Co. All other chemicals were commercial products of analytical purity grade. Steady-state kinetic measurements Kinetic data were obtained by measuring the velocity of H 2 O 2 formation as described above for the enzymatic activity determination. K m and k cat values were obtained by fitting the kinetic data to the Michaelis–Menten equation v ¼ V max [S]/(K m + [S]) by nonlinear regression analysis using Microcal ORIGIN 3.5 software. The data were the average of two/three experiments, carried out at 25 °C, using at least eight amine concentrations each. The standard error was ± 8%. The catalytic parameters were measured in 0.1 M potassium phosphate buffer, pH 7.2, at 120 m M ionic strength. Tris/HCl buffer at 0.1 M , pH 7.2 and 120 m M ionic strength was used for spermine. LCAO inactivation by amines The experiments were performed under the same conditions as described previously for BSAO [7], that is by incubating 0.4 l M LCAO with substrate, in 0.1 M potassium phosphate buffer, at three different pH values of 6.5, 7.2 and 8.0 in a Table 1. Purification of LCAO. Purification step Total Volume (mL) Total activity (IU) Total protein (mg) Specific activity (IUÆmg )1 ) Purification (fold) Yield (%) Crude extract 440 2540 190 13.4 1 100 (NH 4 ) 2 SO 4 70% precipitate, dialysis 65 1790 80 22.5 1.7 70 DEAE–cellulose chromatography 70 1420 45 32 2.5 57 SP Hi Trap chromatography 7 1260 18 70 5.2 50 Ó FEBS 2003 Inactivation of copper-containing amine oxidases (Eur. J. Biochem. 271) 147 1 mL test tube, open to air, using a 37 °C thermostatted water bath. At given time intervals, aliquots of the solutions were tested for activity with 1.0 m M putrescine at 25 °C, after dilution to approximately 2 n M LCAO. In another set of experiments the inactivation was carried out by incuba- ting LCAO (4.0–6.0 l M ) with 1.0 m M substrate, both at 25 °Cand37°C, in a cuvette provided with a Teflon stopper, thus limiting the amount of available oxygen and allowing the monitoring of the UV-vis spectrum of reacting species. Results LCAO purification and characterization The purification method described above was more rapid and allowed a larger recovery than the previously reported one [15]. Thus, it is highly suitable for a large scale preparation of LCAO. Molecular and enzymatic properties of LCAO were found to be very similar to those of other Cu-containing amine oxidases, particularly those from plant sources. As reported in Materials and methods, the LCAO concentra- tion was measured by using the PSAO molar extinction coefficients [12]. These values were chosen because they provided protein concentrations, identical at 280 nm and 500 nm, in good agreement with the copper content of 2.0 ± 0.1 ions per dimer and with the content of reactive TPQ groups. Lower coefficients were reported previously for LCAO [15] and for the homologous enzyme from Lathyrus sativus [16]. The content of TPQ was measured by titration with benzylhydrazine and with 2-hydrazinopyri- dine. The reaction with benzylhydrazine produced a stable adduct absorbing at 380 nm, with an extinction coefficient e 380nm ¼ 65 000 M )1 Æcm )1 , accounting for 1.9 ± 0.1 TPQ per dimer. These properties were quite similar to those of the corresponding adducts of LSAO [17] and PSAO [12]. The reaction with 2-hydrazinopyridine formed an adduct absorbing at 420 nm, with e 420nm ¼ 58 000 M )1 Æcm )1 , accounting for 1.8 ± 0.1 TPQ groups per dimer. LCAO steady-state kinetic parameters Table 2 reports the steady-state kinetic parameters of the LCAO catalyzed oxidation of some primary amines. All kinetic measurements were carried out on protein samples from the same batch. The substrates are arranged in the table in order of decreasing k cat , which shows  500-fold decrease in the list. The values of k cat /K m are less variable with the exception of putrescine and cadaverine. LCAO inactivation in air LCAO was irreversibly inactivated, as is BSAO, by incubation with excess amine. Table 3 shows the residual activity after 30 and 90 min incubation with some substrates in a test tube opened to air at 37 °C. The loss of activity was dependent upon the incubation time and the amine concentration and much less on the nature of the amine used. In general it approached a value of 60% after 30 min, and of > 90% after 90 min. Inactivation was reduced to about 30% after 20 and 60 min using gentle shacking in a water bath and abolished by very efficient stirring. The latter result was achieved by introducing 300 lL of a solution that contained 2.0 l M LCAO and 2.0 m M cadaverine, in a 3 mL spectrophotometer cuvette provided with a small magnetic stirrer and thermostatted at 25 °C. Aliquots of the solution were withdrawn at time intervals and tested for activity and H 2 O 2 content, after proper dilution. No loss of activity occurred in these conditions after 2, 5 and 20 min, although all of the amine was already oxidized after 2.0 min, as measured by the concentration of produced H 2 O 2 . Stirred LCAO was neither inactivated during the turnover phase, nor in the phase subsequent to amine exhaustion. The inactivation was also reduced by the presence of catalase, especially in the first 30 min. The effect of catalase was considerably dependent on the substrate, varying from full protection with benzylamine and agmatine to no protection at all with putrescine. Changes of pH in the range 6.5–8.0 had a relatively small effect on the inactiva- tion, while the protecting effect of catalase was usually larger at pH 6.5 than at pH 8.0. Figure 1 shows the results obtained with spermidine at three different pH values as an example. UV-vis effects of LCAO inactivation As described in Materials and methods, the process of inactivation was monitored spectrophotometrically by incubating LCAO (4.0–6.0 l M )at37°Cor25°C, with 1.0 m M substrate in a spectrophotometer cuvette closed by a Teflon stopper. The oxygen was consumed by most substrates within the mixing time as revealed by the immediate bleaching of the TPQ 500 nm band and by the appearance of absorption peaks at 465, 435 and 360 nm, identical with those of the Cu + -semiquinolamine radical, which is formed under anaerobic conditions by some copper amine oxidases [2]. With some substrates, such as 2-aminomethylpyridine, a few minutes were required to develop the radical signal, that is to exhaust the oxygen, but this did not change the general behavior. In any case, the spectrum of the radical showed a similar intensity at a given temperature and faded slowly away within about 90 min, together with the catalytic activity and with the formation of a broad band (shoulder) at  350 nm. At 25 °C, the initial intensity of the spectrum was lower than Table 2. Steady state kinetics: k cat and substrate specificity (k cat /K m ) for the oxidative deamination of primary amines catalyzed by LCAO. Substrate k cat (s )1 ) k cat /K m (s )1 Æ M )1 ) Putrescine 262 0.97 · 10 6 Cadaverine 159 1.6 · 10 6 Spermidine 100 4.8 · 10 4 Agmatine 45.9 0.94 · 10 5 Tyramine 32.9 1.1 · 10 4 Spermine 28.3 4.5 · 10 4 Histamine 10.3 1.3 · 10 4 Benzylamine 3.7 0.90 · 10 4 2-Aminomethylpyridine 0.47 1.5 · 10 4 148 P. Pietrangeli et al. (Eur. J. Biochem. 271) Ó FEBS 2003 at 37 °C, in agreement with the reported temperature dependence of the equilibrium between Cu + -semiquol- amine and Cu 2+ -quinolamine [18]. The inactivation was also slower at 25 °Cthanat37°C, but the overall behavior was similar. No recovery of activity or spectro- scopic properties occurred after extensive dialysis. The addition of 2-hydrazinopyridine up to a concentration of 0.1 m M , to solutions that were not completely inactivated, slowly formed a band at 420 nm, typical of the 2-hydraz- inopyridine adduct of TPQ. The final band intensity matched the residual activity of the solution. The reaction was slow, implying previous reoxidation of active mole- cules by oxygen and competition of 2-hydrazinopyridine with excess substrate. Also implied is that the TPQ of inactivated molecules was no longer able to bind inhibitors. The decay of the radical spectrum with time, which was very similar with all substrates, is shown in Fig. 2 for putrescine. This substrate was chosen because of its different behavior from other substrates in the presence of catalase (Table 2 and below). Isosbestic points are present in Fig. 2, at least in the early stages of the reaction, because of the broad band (shoulder) formed in the 350 nm region. By subtracting the spectrum of native LCAO from the spectrum of the radical, or from the spectra of the inactivated protein, a peak around 310–320 nm was observed with all substrates. The peaks produced by putrescine are shown in Fig. 3. The band at 350 nm is Table 3. Residual LCAO activity after incubation with substrate. Experimental conditions: 2.8 l M LCAO, 2 m M substrate, unless otherwise stated; 0.1 M potassium phosphate buffer pH 7.2, 37 °C. The activity was measured at 25 °C after proper sample dilution. Substrate [Amine] (m M ) Residual activity (% of starting activity) Residual activity in the presence of catalase (% of starting activity) Time (min) Time (min) 30 90 30 90 Putrescine 2.0 0 0 0 0 0.5 40 20 42 25 Cadaverine 2.0 20 17 79 77 10.0 3.0 3.0 75 48 Spermidine 2.0 19 6.0 89 43 Agmatine 2.0 41 37 100 100 Tyramine 2.0 10 10 70 70 Spermine 2.0 44 10 85 15 10.0 15 2.0 51 8.0 Histamine 2.0 35 30 90 90 Benzylamine 2.0 40 30 100 100 2-Aminomethylpyridine 2.0 45 15 55 35 Fig. 1. Time course of LCAO inactivation by spermidine. LCAO, 2.8 l M , was inactivated at 37 °Cby2.0m M substrate in 0.1 M potas- sium phosphate buffer at different pH values (open symbols) and in presence of catalase (full symbols): pH 6.5 (triangles); pH 7.2 (circles); pH 8.0 (squares). Fig. 2. Decrease with time of the UV-vis spectrum of the radical formed by putrescine-reacted LCAO. The spectra were recorded 1 (top spec- trum), 10, 30, 50, 70, 90, 120 min after the addition of 1.0 m M putrescine to 5.5 l M LCAO, in 0.1 M potassium phosphate buffer pH 7.2, at 37 °C. Ó FEBS 2003 Inactivation of copper-containing amine oxidases (Eur. J. Biochem. 271) 149 evident in the spectrum of the inactivated protein. A peak at 315 nm was found in the difference spectra of inactivated BSAO and was taken to be diagnostic of reduced TPQ [7,19]. The slight variability of the peak maximum wave- length may be due to the fact that this is not a real band but the result of the bleaching of an intense TPQ band at 270 nm [20]. In an inactivation experiment carried out at 25 °C, in order to slow down the reaction to obtain more accurate readings, the residual activity was measured in solution aliquots withdrawn after recording the intensity of the 465 nm peak. Five minutes after substrate addition, the radical was formed and the protein was fully active. Then the loss of activity and the loss of radical intensity took place at almost coincident rates (Fig. 4). At the end of the experiment, a concentration of 0.25 ± 0.01 m M H 2 O 2 was measured immediately after the cuvette was opened to air, in good agreement with the initial oxygen content of the solution, indicating that significant oxygen leaks had not occurred during the incubation and that a similar amount of oxygen was available in all experiments performed at the same temperature. This was confirmed by incubating 0.3 l M LCAO with 1 m M putrescine, at 25 °C, in a 2-mm path-length cuvette opened to air, which also contained the H 2 O 2 -detecting system. The absorbance at 515 nm reached a value of 1.27 (corresponding to 0.245 m M H 2 O 2 )within 5 min, remained constant for about 15 min, then started to decrease. The decrease was of 2% in the subsequent 40 min, while a dark red layer about 2 mm thick formed at the surface of the solution. This experiment demonstrated that in the absence of stirring, extra oxygen is not readily available in the bulk solution after exhaustion of the initial amount present and that diffusion is prevented by the reaction with excess amine in the top layer. In stirred solutions, all the amine was oxidized within 2 min, as shown by the equivalent amount of H 2 O 2 detected in solution at this stage. Thus, the turnover of a thousand-fold amine excess did not cause inactivation as the protein remained reduced for a too short time. Inthepresenceofcatalase,thedecayoftheradical spectrum either did not occur, as in the case of cadaverine, in agreement with the results on PSAO [5], or was greatly reduced, to  20% in the case of spermine (not shown). The only exception was putrescine (Fig. 5). The decay of the radical, slower than in absence of catalase, did not produce isosbestic points nor a shoulder in the 350 nm region, while the 500 nm band of the cofactor remained bleached. Discussion The process of BSAO inactivation required a long incubation with substrate, was inhibited by the presence of catalase, which eliminates H 2 O 2 , but was not produced by exogenous H 2 O 2 added to the resting enzyme [7]. These results were taken to imply that the inactivation is a Fig. 3. Difference spectra of putrescine-reacted LCAO. LCAO, 7.8 l M , was reacted with 1.0 m M putrescine. The spectra, recorded immedi- ately (solid line) and after 5 h incubation (dashed line) were subtracted of the native protein spectrum. 0.1 M potassium phosphate buffer pH 7.2 at 25 °C. Fig. 4. LCAO inactivation. Decay with time of the enzyme activity (s) and of the radical band at 465 nm (d) upon addition of 1.0 m M putrescine to 4.4 l M LCAO in 0.1 M potassium phosphate buffer pH 7.2 at 25 °C.Theactivitybeforeputrescineadditionwastakenas 100. Fig. 5. Decrease with time of the UV-vis spectrum of the radical formed by putrescine-reacted LCAO in the presence of catalase. The spectra were recorded 1 (top spectrum), 10, 40, 70, 110, 150 min after addition of 1.0 m M putrescine to 4.4 l M LCAO, in 0.1 M potassium phosphate buffer pH 7.2, at 37 °C, in the presence of 100 catalase units. 150 P. Pietrangeli et al. (Eur. J. Biochem. 271) Ó FEBS 2003 slow process, involving H 2 O 2 and a substrate-reduced form of the protein. These conclusions are confirmed by the similar results obtained with LCAO. The alternative mechanism proposed involving a partitioning reaction during turnover [8] was excluded as LCAO was fully active after either oxygen was consumed by an excess of amine in a closed cuvette, or the amine was consumed by oxygen in a solution stirred in air, with the rapid turnover of a thousand-fold amine excess. Furthermore, the loss of activity in a closed cuvette was a slow reaction, subse- quent to oxygen exhaustion and the turnover phase, simultaneous with the loss of intensity of the UV-vis spectrum of the Cu + -semiquinolamine radical (Fig. 4). All substrates displayed a similar inactivation time, independ- ent of their highly different catalytic parameters (Table 2) and formed the same band around 350 nm in a closed cuvette. This shows that the reaction was independent of the substrate or related aldehyde and that H 2 O 2 , radical and/or Cu 2+ -quinolamine were the only reacting species. Each species was always present at same concentration, as shown by the measured amounts of H 2 O 2 , by the initial intensity of the radical spectrum and by the bleached 500 nm band. Which of the two equilibrium species, the radical or the Cu 2+ -quinolamine, participated in the reaction is not certain. On one hand, the decay of the radical to the broad band at 350 nm formed isosbestic points along a large part of the reaction [Fig. 3]. On the other hand, BSAO was similarly inactivated [7], although it does not form the radical [2]. A possible explanation of these results is suggested by a recent report [21], in which the quinolamine was prepared by reducing Co- and Ni-substituted Arthrobacter globi- formis amine oxidase (AGAO) with substoichiometric amounts of substrate under anaerobic conditions. By addition of an excess of exogenous aldehyde, a band at 350 nm was slowly formed, which was assigned to a back- reaction generating the neutral form of the product Schiff base. This is more stable than the protonated form [22] preferred by Cu-AGAO, toward hydrolysis to aldehyde and quinolamine. The band disappeared on admission of oxygen into the solution. The band at 350 nm formed by LCAO, upon bleaching of the radical, suggests that the neutral form of the product Schiff base was stabilized by the modification responsible for the loss of catalytic activity, causing back-reaction of aldehydes with Cu 2+ -quinol- amine. This implies that the reaction with H 2 O 2 did not modify the quinolamine but oxidized another conserved residue at the active site. The similar inactivation time of all substrates suggests that the oxidation by H 2 O 2 was the rate- determining step, slower than the back-reaction with aldehyde. At difference from the 350 nm band formed by Co- and Ni-AGAO, the LCAO band was not affected by the admission of oxygen in solution. Thus, the inactivated protein was unable to hydrolyze the aldehyde and to react with oxygen. In previous work on BSAO [7] it was proposed that the H 2 O 2 target is a conserved residue affected by TPQ reduction. This residue was tentatively identified as Tyr371, corresponding to Tyr369 in E. coli amine oxidase [23,24] or Tyr305 in Hansenula polymorpha amine oxidase [25]. The short hydrogen bond (2.4 A ˚ )ofTyr369toTPQO4in ECAO has been taken to imply O4 deprotonation [23]. The TPQ O4 basic character increases considerably in the reduced cofactor [20], causing a partial deprotonation of the Tyr hydroxyl. The mutation of Tyr305 in HPAO [25] and Tyr369 in ECAO [24] decreased the enzyme catalytic activity, to a variable degree depending on the type of mutation and modified the active site hydrogen bond network and cofactor mobility. In the presence of catalase, some substrates produced partial inactivation when high aldehyde concentrations were reached upon prolonged incubation in air (Table 2). The effect does not appear to be related to the kinetic and structural properties of the substrate but rather to the specific reactivity of the corresponding aldehyde. This is evident in the benzylamine/agmatine or putrescine/cadaver- ine couples. The behavior of putrescine was unique as it was able to inactivate completely the protein in absence of H 2 O 2 . The process bleached the radical spectrum without forma- tion of the 350 nm band [Fig. 5]. The back-reaction of the aldehyde or pyrroline with quinolamine did not occur, as the neutral form of the product Schiff base was not stabilized in absence of H 2 O 2 . The aldehyde or pyrroline may react with a nucleophilic residue as often reported for plant amine oxidases [9–11]. BSAO has a very low reactivity with this substrate, k cat ¼ 0.017 s )1 [26]. In conclusion, the inactivation is a slow reaction of the reduced protein with H 2 O 2 , subsequent to turnover and occurring in a similar way for all amines examined. As it is common to all copper amine oxidases investigated so far, it might be relevant to some in vivo functions of amine oxidases. Acknowledgements This paper was supported by Murst and by a C.N.R. grant no. G002FD1 Agenzia 2000. References 1. McIntire, W.S. & Hartmann, C. (1992) Copper containing amine oxidases. In Principles and Applications of Quinoproteins (Davidson, V.L., ed.), pp. 97–171. Marcel Decker, New York. 2. Padiglia, A., Medda, R., Bellelli, A., Agostinelli, E., Morpurgo, L., Mondovı ` ,B.,FinazziAgro ` , A. & Floris, G. (2001) The reductive and oxidative half-reactions and the role of copper-ions in plant and mammalian copper-amine oxidases. Eur. J. Inorg. Chem. 35–42. 3. Janes, S.M., Mu, D., Wemmer, D., Smith, A.J., Kaur, S., Maltby, D., Burligame, A.L. & Klinman, J.P. (1990) A new redox cofactor in eucariotic enzymes: 6-hydroxy-dopa at the active site of bovine serum amine oxidase. Science 248, 981–987. 4. Cai, D. & Klinman, J.P. (1994) Evidence for a self-catalytic mechanism of 2,4,5-trihydroxyphenylalanine quinone biogenesis in yeast copper amine oxidase. J. Biol. Chem. 269, 32039–32042. 5. Mann, P.J.G. (1955) Purification and properties of the amine oxidase of pea seedlings. Biochem. J. 59, 609–620. 6. Mondovı ` , B., Rotilio, G., Finazzi-Agro ` , A. & Costa, M.T. (1967) Diamine oxidase inactivation by hydrogen peroxide. Biochim. Biophys. Acta 132, 521–523. 7. Pietrangeli, P., Nocera, S., Fattibene, P., Wang, X.T., Mondovı ` . B. & Morpurgo, L. (2000) Modulation of bovine serum amine oxidase activity by hydrogen peroxide. Biochem. Biophys. Res. Commun. 267, 174–178. 8. Lee, Y., Shepard, E., Smith, J., Dooley, D.M. & Sayre, L.M. (2001) Catalytic turnover of substrate benzylamines by the qui- none-dependent plasma amine oxidase leads to H 2 O 2 -dependent Ó FEBS 2003 Inactivation of copper-containing amine oxidases (Eur. J. Biochem. 271) 151 inactivation: evidence for generation of a cofactor-derived ben- zoxazole. Biochemistry 40, 822–829. 9. Medda, R., Padiglia, A., Finazzi-Agro ` , A., Pedersen, J.Z., Lorrai, A. & Floris, G. (1997) Tryptamine as substrate and inhibitor of lentil seedling copper amine oxidase. Eur. J. Biochem. 250, 377–382. 10. Medda, R., Padiglia, A., Pedersen, J.Z., Finazzi- Agro ` ,A., Rotilio, G. & Floris, G. (1997) Inhibition of copper amine oxidase by haloamines. A killer product mechanism. Biochemistry 36, 2595–2602. 11. Fre ´ bort, I., S ˇ ebela, M., Svendsen, I., Hirota, S., Masaaki, E., Yamauchi, O., Bellelli, A., Lemr, K. & Pec, P. (2000) Molecular mode of interaction of plant amine oxidase with the mechanism- based inhibitor 2-butyne-1,4-diamine. Eur. J. Biochem. 267, 1423–1433. 12. McGuirl, M.A., McCahon, C.D., McKeown, K.A. & Dooley, D.M. (1994) Purification and characterization of pea seedling amine oxidase for crystallization studies. Plant Physiol. 106, 1205–1211. 13. Brumby, P.E. & Massey, V. (1967) Determination of nonheme iron, total iron and copper. Methods Enzymol. 10, 463–474. 14. Angelini, R., Rea, G., Federico, R. & D’Ovidio, R. (1996) Spatial distribution and temporal accumulation of mRNA encoding diamine oxidase during lentil (Lens culinaris Medicus) seedling development. Plant Sci. 119, 103–113. 15. Cogoni, A., Farci, L., Medda, R., Rinaldi, A. & Floris, G. (1989) Amine oxidase from Lathyrus cicera and Phaseolus vulgaris:puri- fication and properties. Preport Biochem. 19, 95–112. 16. S ˇ ebela, M., Luhova, L., Fre ´ bort, I., Fullhammer, H., Hirota, S., Zajonkova, L., Stuzka, V. & Pec, P. (1998) Analysis of the active sites of the copper/topa quinone-containing amine oxidases from Lathyrus odoratus and L. sativus seedlings. Phytochem. Anal. 9, 211–222. 17. Padiglia, A., Medda, R. & Floris, G. (1992) Lentil seedling amine oxidase: interaction with carbonyl reagents. Biochem. Int. 28, 1097–1107. 18. Dooley, D.M., McGuirl, M.A., Brown, D.E., Turowsky, P.N., McIntire, W.S. & Knowles, P.F. (1991) A Cu(I)-semiquinone state in substrate-reduced amine oxidases. Nature 349, 262–264. 19. Su, Q. & Klinman, J.P. (1998) Probing the mechanism of proton coupled electron transfer to dioxygen: the oxidative half- reaction of bovine serum amine oxidase. Biochemistry 37, 12513– 12525. 20. Bossa, M., Morpurgo, G.O. & Morpurgo, L. (1994) Models and molecular orbital semiempirical calculations in the study of the spectroscopic properties of bovine serum amine oxidase quinone cofactor. Biochemistry 33, 4425–4431. 21. Kishishita, S., Okajima, T., Kim, M., Yamaguchi, H., Hirota, S., Suzuki,S.,Kuroda,S.,Tanizawa,K.&Mure,M.(2003)Roleof copper ion in bacterial copper amine oxidase: spectroscopic and crystallographic studies of metal-substituted enzymes. J. Am. Chem. Soc. 125, 1041–1055. 22. Cai, D., Dove, J., Nakamura, N., Sanders-Loehr, J. & Klinman, J.P. (1997) Mechanism-based inactivation of a yeast methylamine oxidase mutant: Implications for a functional role of the consensus sequence surrounding topaquinone. Biochemistry 36, 11472– 11478. 23. Wilmot, C.M., Murray, J.M., Alton, G., Parsons, M.R., Convery, M.A., Blakeley, V., Corner, A.S., Palcic, M.M., Knowles, P.F., McPherson, M.J. & Phillips, S.E.V. (1997) Catalytic mechanism of thequinoenzymeamineoxidasefromEscherichia coli:exploring the reductive half-reaction. Biochemistry 36, 1608–1620. 24. Murray, J.M., Kurtis, C.R., Tambyrajah, W., Saysell, C.G., Wilmot, C.M., Parsons, M.R., Phillips, S.E.V., Knowles, P.F. & McPherson, M.J. (2001) Conserved tyrosine-369 in the active site of Escherichia coli copper amine oxidase is not essential. Biochemistry 40, 12808–12818. 25. Hevel, J.M., Mills, S.A. & Klinman, J.P. (1999) Mutation of a strictly conserved, active-site residue alters substrate specificity and cofactor biogenesis in a copper amine oxidase. Biochemistry 38, 3683–3693. 26. De Matteis, G., Agostinelli, E., Mondovı ` ,B.&Morpurgo,L. (1999) The metal function in the reactions of bovine serum amine oxidase with substrates and hydrazine inhibitors. J. Biol. Inorg. Chem. 4, 348–353. 152 P. Pietrangeli et al. (Eur. J. Biochem. 271) Ó FEBS 2003 . Inactivation of copper-containing amine oxidases by turnover products Paola Pietrangeli 1 , Stefania Nocera 1 ,. Catalytic turnover of substrate benzylamines by the qui- none-dependent plasma amine oxidase leads to H 2 O 2 -dependent Ó FEBS 2003 Inactivation of copper-containing