Reactions of the class II peroxidases, lignin peroxidase and Arthromyces ramosus peroxidase, with hydrogen peroxide. Catalase-like activity, compound III formation, and enzyme inactivation.

The reactions of the fungal enzymes Arthromyces ramosus peroxidase (ARP) and Phanerochaete chrysosporium lignin peroxidase (LiP) with hydrogen peroxide (H(2)O(2)) have been studied. Both enzymes exhibited catalase activity with hyperbolic H(2)O(2) concentration dependence (K(m) approximately 8-10 mm, k(cat) approximately 1-3 s(-1)). The catalase and peroxidase activities of LiP were inhibited within 10 min and those of ARP in 1 h. The inactivation constants were calculated using two independent methods; LiP, k(i) approximately 19 x 10(-3) s(-1); ARP, k(i) approximately 1.6 x 10(-3) s(-1). Compound III (oxyperoxidase) was detected as the majority species after the addition of H(2)O(2) to LiP or ARP, and its formation was accompanied by loss of enzyme activity. A reaction scheme is presented which rationalizes the turnover and inactivation of LiP and ARP with H(2)O(2). A similar model is applicable to horseradish peroxidase. The scheme links catalase and compound III forming catalytic pathways and inactivation at the level of the [compound I.H(2)O(2)] complex. Inactivation does not occur from compound III. All peroxidases studied to date are sensitive to inactivation by H(2)O(2), and it is suggested that the model will be generally applicable to peroxidases of the plant, fungal, and prokaryotic superfamily.

Compound III (oxyperoxidase) was detected as the majority species after addition of H 2 O 2 to LiP or ARP, its formation was accompanied by loss of enzyme activity.A reaction scheme is presented that rationalizes the turnover and inactivation of LiP and ARP with H 2 O 2 .A similar model is applicable to horseradish peroxidase.The scheme links catalase and compound IIIforming catalytic pathways, and inactivation at the level of the compound I-H 2 O 2 complex.
Inactivation does not occur from compound III.All peroxidases studied to date are sensitive to inactivation by H 2 O 2 and it is suggested that the model will be generally applicable to peroxidases of the plant, fungal and prokaryotic superfamily.
Peroxidases (donor: hydrogen peroxide oxidoreductases) are ubiquitous enzymes that catalyze the oxidation of substrate at the expense of hydrogen peroxide (H 2 O 2 ) 1 .The heme peroxidases have been classified into two distinct groups, termed the animal (found only in animals) and plant (found in plants, fungi and prokaryotes) superfamilies (1).The plant peroxidases, which share similar overall protein folds and specific features, such as catalytically essential histidine and arginine residues in their active sites, have been subdivided into three classes on the basis of sequence comparison (2,3).Class I are intracellular enzymes including yeast cytochrome c peroxidase (CcP), ascorbate peroxidase (APX) from plants, and bacterial gene-duplicated catalase-peroxidases (4).Class III contains the secretory plant peroxidases such as those from horseradish (HRP), barley or soybean.These peroxidases seem to be biosynthetic enzymes involved in processes such as plant cell wall formation and lignification.Class II consists of the secretory fungal peroxidases such as lignin peroxidase (LiP) from Phanerochaete chrysosporium, manganese peroxidase (MnP) from the same source, and Coprinus cinereus peroxidase (CiP) or Arthromyces ramosus peroxidase (ARP), which have been shown to be essentially identical in both sequence and properties (5).The main role of class II peroxidases appears to be the degradation of lignin in wood.
All peroxidases so far studied share much the same catalytic cycle that proceeds in three distinct and essentially irreversible steps (6), and is often referred to as the 'peroxidase ping-pong'.The resting ferric enzyme reacts with H 2 O 2 in a two-electron process to generate the intermediate known as compound I. Compound I is discharged in two sequential singleelectron reactions with reducing substrate yielding radical products, which are often highly reactive, and water.The first reduction step results in the formation of another enzyme intermediate, compound II.In the final step compound II is reduced back to ferric peroxidase.
The 'peroxidase ping-pong' provides an adequate description of the peroxidase reaction; however, continuing work has revealed some limitations of the basic model.The compound I reduction steps have been shown to consist of reversible substrate binding followed by substrate oxidation (7).The formation and nature of compound I has been intensively studied.Both a neutral peroxidase-peroxide complex and a charged complex (known as compound 0) have been observed (8), and variations have been identified in the electronic structures of the compound Is of different peroxidases (9)(10)(11)(12)(13)(14).Furthermore, certain peroxidases have been found to utilize H 2 O 2 to reduce compound I when no other substrate is available.Foremost amongst the enzymes capable of this are the catalase-peroxidases, which act as highly efficient catalases (k cat /K M ≈ 10 6 M -1 s -1 ) (15,16).Unexpectedly, APX appears incapable of turnover in the presence of only H 2 O 2 (17) despite also being classified in class I. HRP, from class III, has been shown to possess catalase activity, albeit with much lower efficiency (k cat /K M ≈ 10 2 -10 3 M -1 s -1 ) than the catalase-peroxidases (18), but to our knowledge, no data are available on catalase activity in class II peroxidases.HRP compound I can also carry out a single-electron reduction of H 2 O 2 to generate compound II and superoxide radical (O 2

• -
). Reaction of compound II with more H 2 O 2 yields compound III (a complex between ferric peroxidase and O 2 •also known as oxyperoxidase).
Catalase activity and compound III formation result in enzyme turnover in the absence of normal reducing substrates, however, a further reaction with H 2 O 2 has been identified that leads to progressive irreversible enzyme inactivation.In the case of HRP, the complex between compound I and peroxide ([compound I•H 2 O 2 ]) has been unambiguously identified as the pivotal point connecting the three simultaneous pathways (19,20).Thus, the

Materials
Enzymes -Non-glycosylated recombinant lignin peroxidase isoenzyme H8 (LiP) from Phanerochaete chrysosporium was expressed in Escherichia coli, the polypeptide was recovered from inclusion bodies and refolded in vitro in the presence of heme, calcium ions and glutathione (40).After purification the preparation had an RZ (A 409 nm / A 280 nm ) = 1.8 and was homogeneous by SDS/PAGE.Recombinant LiP has so far has exhibited identical behavior to the natural fungal enzyme (40,41).Arthromyces ramosus peroxidase (ARP, lot 48H0555, RZ (A 405 nm / A 275 nm ) = 2.5) and horseradish peroxidase (type IX, RZ (A 403 nm / A 275 nm ) = 3.2) were obtained from Sigma as lyophilized powders.The HRP preparation has been characterized previously (33) by isoelectric focussing as a single band with pI 8.5 and confirmed to be isoenzyme C. ARP was similarly characterized as a single peroxidase band with pI 3.5.Enzyme concentrations were determined spectrophotometrically using ε 403 nm = 100 mM -1 cm -1 (for HRP-C), ε 409 nm = 168 mM -1 cm -1 (for LiP) and ε 405 nm = 109 mM -1 cm -1 (for ARP).Bovine erythrocyte superoxide dismutase (SOD) (product code S-2515) was purchased from Sigma as a lyophilized powder containing 4200 U mg -1 .
Chemicals -Hydrogen peroxide (30 % by vol.) and buffer substances (analytical reagent grade) were obtained from Merck.The concentration of H 2 O 2 was determined using form of the crystallized diammonium salt was supplied by Boehringer-Mannheim, its concentration was measured spectrophotometrically using ε 340 nm = 36 mM -1 cm -1 .
Tetranitromethane (TNM) and manganese (II) chloride were from Aldrich.All solutions were prepared using de-ionized water drawn from a Milli-Q system (Millipore).

Methods
Oxygen production -Oxygen production was measured using a Clark-type electrode coupled to a Hansatech (Kings Lynn, Cambs., UK) CB1D oxygraph unit.The equipment was calibrated using the tyrosinase / 4-tert-butylcatechol method (42).The temperature of the reaction chamber was controlled at 25 ± 0.1ºC using a Haake circulating water bath.Nitrogen was bubbled through the reaction medium to remove dissolved oxygen.A baseline rise in oxygen concentration of less than 0.5 µM min -1 without enzyme was obtained.The reaction medium (2 ml total volume) contained H 2 O 2 at the appropriate concentrations (see results) in the following buffers: 50 mM Na acetate (pH 3.0), 50 mM Na citrate, (pH: 4.5, 5.5) and 50 mM Na phosphate (pH: 6.5, 7.0, 7.5).The reactions were started by addition of peroxidase.
Additional reagents (Mn 2+ , TNM or SOD) were added as required.Two types of experiments were performed using the system described.
Determination of the initial rate of H 2 O 2 -decomposing activity (V 0 ) -Values of V 0 were determined at short reaction times in the initial linear phase of oxygen production.The initial linear phase lasted approximately 10 to 30 seconds, departure from linearity at later times indicated that enzyme inactivation was being observed, and such data were therefore not used in calculations.Values of V max and K M were obtained by non-linear regression of a hyperbolic function to a plot of V 0 against [H 2 O 2 ] using the program SigmaPlot for Windows (version 2, Jandel Scientific Software, San Rafael, CA, USA).These data could be fitted using the Michaelis-Menten equation, which in this case takes the following form (18): where ), the steps to which k cm -1 ) in an assay system comprising 10 mM ABTS and 5 mM H 2 O 2 in 50 mM Na phosphate buffer (pH 6.5).The residual enzyme activity (A R ) (expressed as %) was taken as the activity remaining at the end of the reaction (A t ) compared to the initial activity (A 0 ).Assays were recorded on a Perkin-Elmer Lambda-2S UV-vis spectrophotometer.The temperature was controlled at 25 ± 0.1ºC using a Haake circulating water bath.The partition ratio (r, the number of turnovers with H 2 O 2 before the enzyme was inactivated; ( ) Kinetics of inactivation -The inactivation of ARP was followed against time at pH  The multivariate data sets from the stopped-flow were fitted globally using the program Pro/K (Applied Photophysics).3) × 10 -3 s -1 ) was 10-fold higher than for ARP (k i = (1.85 ± 0.18) × 10 -3 s -1 ).Thus the k i of ARP was approximately equivalent to the value for HRP-C (18) but LiP was inactivated much more rapidly (Table I).

Catalase activities of
Both ARP and LiP were more effective catalases at neutral than at acidic pH (data not shown).The apparent pK A ≈ 6 observed for ARP was quite similar to the value with HRP-C (18), suggesting the involvement of the conserved distal histidine residue, and that catalase activity is an enzymatic reaction.
It has recently been shown that O 2 •scavengers (SOD, TNM and Mn 2+ ) have only limited effects on HRP-C catalase activity (38,43).Fig. 2 shows the effects of O 2 concentrations and pH values were used to further probe the inactivation of LiP and ARP.In Fig. 3 it can be seen that LiP (panel A) was more sensitive to H 2 O 2 than was ARP (panel B), as expected from the O 2 measurements above.The insets show that both enzymes were more sensitive at acidic than at neutral pH.The values of r obtained here, calculated using Eq. 2, were in reasonable agreement with those from oxygen measurements at the same pH (Table I).This suggests that, as for HRP-C ( 18), the level of turnover through the catalase reaction is strongly correlated with protection of the peroxidase against inactivation by H 2 O 2 .
The kinetics of inactivation were directly measured by following the time-dependent fall in peroxidase activity, determined using ABTS, as a function of H 2 O 2 concentration.In indicated the generation of compound I followed by the reduction of compound I to a species with a compound II-like spectrum in a biphasic process with k fast = 2.3 s -1 (41).The heme in this species then bleached in an H 2 O 2 concentration-independent process with a rate k = 2.1 × 10 -3 s -1 at pH 4.0, rising to 9 × 10 -3 s -1 at pH 6.0.Compound III was not observed.
The kinetic constants determined in the present study for LiP and ARP are compared in Table I to those previously obtained for HRP-C.
order of magnitude higher than those of ARP or HRP-C.Thus, the partition ratio between turnover and inactivation ( -see Scheme I) was much lower for LiP.The pH dependencies of inactivation (more sensitive at acid pH) and the catalase activity (lower at acid pH), of LiP and ARP and the effects of O 2 •scavengers on the catalase reaction were essentially similar to HRP (38) suggesting the involvement of common active site residues and a similar overall mechanism for all three enzymes.An additional point to note is that LiP, during reaction with H 2 O 2 , undergoes an autocatalytic covalent modification of Trp 171 to a β-hydroxy adduct ( 13).This modification process may be connected to the spontaneous reduction of LiP compound I to a compound II-like species (41) followed by heme bleaching.
It is therefore possible that at very low H 2 O 2 concentrations and higher pH values, when heme bleaching from the compound II-type material is more rapid (in contrast to the general sensitivity to inactivation, suggesting that the mechanism is not the same), a proportion of LiP activity loss occurs in this way.There may be parallels between this reaction of LiP and the inactivation of APX (17) since it is unlikely that the catalase cycle of LiP will be active with so little H 2 O 2 and in neither example was compound III detected.
Upon addition of large stoichiometric excesses of H 2 O 2 to LiP or ARP the majority of the peroxidase was converted to compound III.A similar effect has been seen with HRP (19,35).The formation of compound III under conditions where inactivation occurs has been used in some studies with LiP and HRP as evidence that inactivation results from modification of this intermediate (24,43).However, computer simulations of Scheme I (done using the rate constants available for HRP) showed that compound III could be present at complex was not seen in the spectra obtained with LiP or ARP in this study, however, the absorption peak of this species is around 940-965 nm making its observation difficult.
Additionally, the amount of complex that accumulates towards the start of the reaction is related to the affinity of compound I for peroxide.Nor were the inactive forms of the fungal enzymes, compound P670, detected.This was probably due to the inherently unstable nature of this material that easily loses iron to leave colorless products.The P670 species of APX (17) and HRP-A2 (35) proved equally hard to detect and that of HRP-C was most stable at 10ºC (20).
The role of compound III in the activity of LiP has in the past been the subject of some discussion (22)(23)(24)(25)(26)(27).A particularly contentious issue has been the presence or absence Summing up, a number of conclusions can be drawn from the data presented here.
LiP has generally been considered to be a fragile enzyme in the presence of H 2 O 2 , however, it will, in fact, turn over H 2 O 2 quite successfully when compared to APX.LiP was actually a better catalase than either ARP or HRP, but, the higher rate constant of LiP inactivation led to the greater sensitivity of this enzyme.Many studies (1 and references therein) with LiP have tended to focus on its behavior at acid pH with veratryl alcohol (a secondary metabolite of P. chrysosporium which, when oxidized by LiP, is capable of attacking lignin ( 1)).The pH dependence of inactivation observed in this study suggests that LiP may be especially vulnerable under such conditions implying that the fungus must maintain strict control over the oxidant / reductant ratio in order to avoid inactivation of the peroxidase.The inactivation profile of ARP, as well as its catalase activity, were quite similar to HRP-C suggesting that the fungal enzyme could indeed be suitable for many applications normally performed by the plant peroxidase, at least from the stand-point of its stability under oxidizing conditions.The firm understanding of the effects of H 2 O 2 on ARP, presented here, should allow continued development of peroxide resistant mutants.
Finally, the general mechanistic model shown in Scheme I is applicable to peroxidases from class II.Since class III and class II peroxidases essentially catalyze the same reaction, but in opposite directions, this may justify the apparently closer mechanistic similarities observed here compared to the class I enzymes.Differences in detail may exist in the behavior of specific enzymes but the central motif of the scheme, the partition between turnover and inactivation from [compound I•H 2 O 2 ], appears in the reactions of peroxidases from each of the three classes of the plant superfamily.Therefore, just as the 'peroxidase ping-pong' represents the turnover of peroxidases with both an oxidizing and a reducing reactivity of peroxidases with peroxides that we fully expect to be applicable to enzymes that have not yet been studied in detail or that remain to be identified.proceeds in the transition phase until all the enzyme has been inactivated or the substrate exhausted.The distribution of enzyme between the catalytic and inactivating pathways is described by two partition ratios:  against pH.The heme concentration was 1 µM with the appropriate number of equivalents of H 2 O 2 in 50 mM Na acetate (½), Na citrate (i) or Na phosphate ({).Incubations were allowed to proceed to the end of reaction.Residual activity (A R ) was measured using the ABTS assay system.Values of r were calculated using Eq. 2 in methods.
by guest on August 17, 2017 http://www.jbc.org/Downloaded from 3 and K 2 refer being shown in Scheme I. Determination of the total oxygen produced in the reaction ([O 2 ] ∞ ) -The reaction endpoint was reached when no further O 2 production was observed.In all experiments, the final concentration of oxygen produced ([O 2 ] ∞ ) was much less than 0.24 mM (i.e. the oxygen concentration in air-saturated medium at 25ºC).Inactivation and the determination of residual activity -Peroxidase was inactivated at 25ºC in 100 µl incubations in similar buffers to those used in the oxygraph.Each incubation contained enzyme (1 µM) and H 2 O 2 at different concentrations (giving the desired [H 2 O 2 ]/[peroxidase] ratios).When the reaction was complete, the peroxidase activity was measured spectrophotometrically by the increase in the absorbance at 414 nm (ε = 31.1 mM -1 (5 µM) was measured at pH 4.0 and pH 6.0 in stopped-flow experiments done on an Applied Photophysics (Leatherhead, Surrey, UK) SX18.MV stopped-flow instrument fitted with a PDA.1 photodiode array detector.The temperature was controlled at 25 ± 0.1ºC using a Neslab circulating water bath.
LiP and ARP -Both LiP and ARP showed clear catalase activity.The H 2 O 2 concentration dependence of the initial rates (V 0 ) of O 2 production by the two fungal enzymes, shown in Fig.1, demonstrated that both exhibited saturation kinetics.From the curves, values of K 2 (K M ) and k 3 (k cat ) (see Scheme I) could be obtained using Eq. 1. Panel A shows the saturation curve for LiP with K 2 = 8.6 ± 0.4 mM and k 3 = 2.87 ± 0.21 s - 1 , in panel B the corresponding values for ARP were K 2 = 10.2 ± 2.3 mM and k 3 = 1.15 ± 0.09 s -1 .Thus both LiP and ARP had kinetic parameters for O 2 generation of a similar order of magnitude to those previously determined for HRP-C (18) and HRP-A2(35).However, during experiments it was noticed that the values of V 0 for LiP were maintained for only a very short time and that O 2 production essentially ceased in approximately 10 mins, whereas for ARP it took about 1 hour for activity to be lost.This suggested that LiP was being inactivated more rapidly than ARP.Determinations of the total O 2 gas produced by a given amount of enzyme before its inactivation (i.e. the infinite product of the catalase reaction) confirmed that LiP lost activity much more rapidly than ARP.The values of the partition ratios (r), obtained from the slopes of the straight line fits to plots of [O 2 ] ∞ against [peroxidase] (not shown)(18), indicated that LiP (r = 155 ± 25) was more sensitive to inactivation by H 2 O 2 than ARP (r = 620 ± 60).see Scheme I), it was calculated that the inactivation constant of LiP (k i = (18.5 ± 2.

Fig. 4
Fig. 4 panel A the inactivation curves for ARP are shown, and in panel B the H 2 O 2 concentration-dependent saturation curve from which values of k i = (1.37 ± 0.27) × 10 -3 s -1 and K I = 3.83 ± 0.35 mM were calculated (see Eq. 3).The equivalent experiments performed for LiP (data not shown) yielded values of k i = (20.0± 3.0) × 10 -3 s -1 and K I = 6.1 ± 0.45 high concentrations even though inactivation proceeded from the [compound I•H 2 O 2 ] complex(38).This was because of the overwhelming proportion of catalytic turnover via the catalase cycle (independent of compound III) and the low rate of reversion of compound III to ferric peroxidase by loss of O 2 • -.The involvement of O 2 •in the degeneration of LiP compound III to the ferric form has been demonstrated (26).The [compound I•H 2 O 2 ] of a second compound III species, termed LiP III*.Two distinct compound IIIs have also been postulated to exist in HRP(46) with the superoxide ligand either loosely or tightly bound to the iron.In HRP-A2 (35) the more stable form (observed after 5 days in the presence of excess H 2 O 2 ) has been suggested to account for the presence of a highly H 2 O 2 resistant fraction (≈ 20 %) of the peroxidase seen in plots of residual activity (% A R ) against [H 2 O 2 ]/[HRP-A2], similar to those shown in Fig.3.The data in this paper do not demonstrate the existence of a particularly long lasting form of LiP compound III or of a resistant enzyme fraction.The presence of LiP III* would not necessarily affect the inactivation profile unless it was either stable and unreactive towards H 2 O 2 or directly involved in the inactivation pathway.The first of these possibilities was not supported by the spectroscopic data and the second has been suggested previously(24) but can be discarded for the reasons discussed previously.No spectral or kinetic evidence of a second form of ARP compound III was found.
obtained in the present study suggest that this relationship also applies to LiP and ARP.The k i for LiP was 10-fold higher than that of ARP (or HRP-C) resulting in faster inactivation and a lower value of r for this enzyme despite a somewhat higher value of k 3 .An additional factor that may contribute to inactivation at low [H 2 O 2 ] in the case of LiP is the spontaneous conversion of LiP compound I to a compound II-like species followed by [H 2 O 2 ]-independent heme bleaching.

FIG. 1 .FIG. 2 .FIG. 3 .
FIG. 1. H 2 O 2 concentration dependence of catalase-like oxygen production by fungal peroxidases.A, LiP (0.25 µM) and B, ARP (0.25 µM) on H 2 O 2 concentration at pH 7.0 (50 (17)as calculated for each time point and H 2 O 2 concentration.The data were plotted and fits to a first-order exponential decay to obtain values of k obs (observed rate constants of inactivation) were made using SigmaPlot.The rate constant of inactivation (k i ) and dissocation constant (K i ) were calculated from a hyperbolic secondary plot of k obs against [H 2 O 2 ] using the Michaelis-Menten equation in the form(17):Spectral changes on addition of H 2 O 2 -The UV-vis spectral changes that occurred over time to peroxidase (LiP or ARP) upon addition of H 2 O 2 (100 or 500 equivalents respectively) were observed, at pH 3.0 and pH 7.0, using a Perkin-Elmer Lambda-2S spectrophotometer.The reaction of LiP (0.8 µM) with H 2 O 2 times after addition of the enzyme, aliquots were removed and the activity with ABTS was determined.The conditions used were such that the [H 2 O 2 ] added with the enzyme aliquot, which varied with time as inactivation proceeded, did not affect the assayed activity.The A R by guest on August 17, 2017 http://www.jbc.org/Downloaded from (

TABLE I
Rate constants for the catalase-like activity and inactivation of LiP and ARP comparedto those of HRP-C.

Mechanistic model of the reaction of peroxidase with H 2 O 2 in the absence of other substrates
. E is native ferric peroxidase; S is H 2 O 2 ; E′, E′′ and E′′′ are the enzyme intermediates, compound I, II, III, respectively; E′S and E′′S are complexes between the respective enzyme intermediates and H 2 O 2 , [compound I•H 2 O 2 ] and [compound II•H 2 O 2 ]; E i is inactive peroxidase.In the scheme, peroxidase exhibits two catalytic cycles (catalase and compound III pathways) with distinct rate constants and stoichiometries and the inactivation route to E i , with the [compound I•H 2 O 2 ] complex playing a central role.The reaction of peroxidase with H 2 O 2 Inactivation kinetics of Arthromyces ramosus peroxidase.A, Time courses determined with ARP (1 µM) and H 2 O 2 (1 mM, »; 2 mM, ¹; 5 mM, E; 10 mM, {; 20 mM, i; 50 mM, ½) in 50 mM Na phosphate buffer, pH 7.5, and used to obtain values of k obs .B, Plot of k obs against H 2 O 2 concentration from which k i and K I were calculated using Eq. 3 in methods.FIG. 5. Spectral changes upon addition of H 2 O 2 to the fungal peroxidases.A, 100 equivalents of H 2 O 2 to LiP (2.4 µM) or B, 500 equivalents to ARP (4.1 µM) in 50 mM Na phosphate buffer, pH 7.0.Repeat scans were obtained for 60 minutes.The expanded sections show the visible spectra of ferric peroxidase and of compound III (C-III) detected soon after H 2 O 2 addition.The concentration of compound III fell with time due to the inactivation process.