Oxidation of 4-methoxymandelic acid by lignin peroxidase. Mediation by veratryl alcohol.

The mechanism of veratryl alcohol-mediated oxidation of 4-methoxymandelic acid by lignin peroxidase was studied by kinetic methods. For monomethoxylated substrates not directly oxidized by lignin peroxidase, veratryl alcohol has been proposed to act as a redox mediator. Our previous study showed that stimulation of anisyl alcohol oxidation by veratryl alcohol was not due to mediation but rather due to the requirement of veratryl alcohol to complete the catalytic cycle. Anisyl alcohol can react with compound I but not with compound II. In contrast, veratryl alcohol readily reduces compound II. We demonstrate in the present report that the oxidation of 4-methoxy mandelic acid is mediated by veratryl alcohol. Increasing veratryl alcohol concentration in the presence of 2 mM 4-methoxymandelic acid resulted in increased oxidation of 4-methoxymandelic acid yielding anisaldehyde. This is in contrast to results obtained with anisyl alcohol where increased concentrations of veratryl alcohol caused a decrease in product formation. ESR spectroscopy demonstrated that 4-methoxymandelic acid caused a decrease in the enzyme-bound veratryl alcohol cation radical signal, which is consistent with its reaction at the active site of the enzyme.

The mechanism of veratryl alcohol-mediated oxidation of 4-methoxymandelic acid by lignin peroxidase was studied by kinetic methods. For monomethoxylated substrates not directly oxidized by lignin peroxidase, veratryl alcohol has been proposed to act as a redox mediator. Our previous study showed that stimulation of anisyl alcohol oxidation by veratryl alcohol was not due to mediation but rather due to the requirement of veratryl alcohol to complete the catalytic cycle. Anisyl alcohol can react with compound I but not with compound II. In contrast, veratryl alcohol readily reduces compound II. We demonstrate in the present report that the oxidation of 4-methoxy mandelic acid is mediated by veratryl alcohol. Increasing veratryl alcohol concentration in the presence of 2 mM 4-methoxymandelic acid resulted in increased oxidation of 4-methoxymandelic acid yielding anisaldehyde. This is in contrast to results obtained with anisyl alcohol where increased concentrations of veratryl alcohol caused a decrease in product formation. ESR spectroscopy demonstrated that 4-methoxymandelic acid caused a decrease in the enzymebound veratryl alcohol cation radical signal, which is consistent with its reaction at the active site of the enzyme.
To degrade the aromatic polymer lignin, the white-rot fungus Phanerochaete chrysosporium secretes H 2 O 2 (1), and two families of H 2 O 2 -utilizing enzymes, the lignin peroxidase (LP) 1 and manganese peroxidases (MnP). The MnPs catalyze the oxidation of Mn 2ϩ to Mn 3ϩ (2) whereas the LPs catalyze the oxidation of phenolic and nonphenolic methoxylated aromatic substrates (3). The catalytic cycle of both peroxidases is similar to that of other peroxidases. The enzyme is first oxidized by H 2 O 2 to form a two-electron-oxidized intermediate, compound I (4). Compound I then returns to the resting ferric enzyme by two sequential one-electron reduction steps producing two, oneelectron-oxidized products. The one-electron oxidized species of the enzyme formed during turnover is referred to as compound II. Unique to these fungal peroxidases is the recalcitrant nature of their substrate (5). Both act on substrates that other peroxidases are not capable of oxidizing.
The role and mechanism by which the LP and MnP interact with lignin is yet to be elucidated. For both enzymes, the role of mediators has been proposed. With MnP, the mediation phenomenon is well characterized and widely accepted. Complexed Mn 2ϩ (6, 7) is oxidized by MnP and diffuses away from the active site of the enzyme to oxidize lignin (8). In the case of LP, the role of mediators and the mechanism by which lignin is depolymerized by LP is still unknown. A key component of LP catalysis is veratryl alcohol, a secondary metabolite also produced by ligninolytic cultures of P. chrysosporium (9). Harvey et al. (10) proposed that veratryl alcohol is the mediator for LP as Mn 2ϩ is the mediator for MnP. They proposed that the initial one-electron oxidation of veratryl alcohol forms the cation radical which then diffuses away from the active site and oxidizes other substrates such as anisyl alcohol and 4-methoxymandelic acid (4-MMA). The mediation hypothesis explains the ability of LP to oxidize substrates such as anisyl alcohol and 4-MMA only in the presence of veratryl alcohol (10). The mediation mechanism is also attractive as it accounts for how a large bulky enzyme can interact with large bulky insoluble substrates.
Harvey's hypothesis was supported by the work of Khindaria et al. (11). They were able to generate the radical chemically and also enzymatically. Both forms were detected by ESR spectroscopy; the half-life was much longer for the enzyme-bound form (12). Candeias and Harvey (13) also generated the veratryl cation radical by chemical means and characterized it by pulse radiolysis. They determined the lifetime of the radical and calculated that it was capable of diffusing up to 7 m. However, the work of Candeias and Harvey (13) also showed that the veratryl alcohol cation radical did not oxidize anisyl alcohol or 4-MMA. They postulated that the enzyme-bound radical could be more reactive or that it would have a longer half-life, thus allowing the radical to oxidize these monomethoxylated substrates.
Our previous work demonstrated that the veratryl alcohol "mediation" phenomenon with anisyl alcohol was not due to redox mediation (14). The enhancement was due to the requirement of veratryl alcohol to complete the catalytic cycle. Whereas most substrates can be oxidized by both compound I and II of LP, anisyl alcohol can only be oxidized by compound I (thus being stuck at compound II). Inclusion of veratryl alcohol results in its oxidation by compound II allowing the enzyme to complete the catalytic cycle. In the present study, we demonstrate, to our surprise, that the oxidation of the other monomethoxylated substrate that Harvey and co-workers (13) characterized, 4-MMA, is mediated by veratryl alcohol. Steady-state methods show that 4-MMA is oxidized by the enzyme-generated veratryl alcohol cation radical. We also provide ESR spectroscopy data consistent with this where the ESR signal of the veratryl alcohol cation radical is quenched by 4-MMA.

MATERIALS AND METHODS
Enzyme Purification-LP isozyme H1 (pI 4.7) was isolated from P. chrysosporium strain PSBL-1 as described previously (15). The H1 fraction from the Mono Q column was further purified by preparative isoelectric focusing. For stopped-flow experiments, the enzyme was dialyzed against distilled-deionized water. The concentration of LP was determined at 409 nm using an extinction coefficient of 169 mM Ϫ1 cm Ϫ1 (16).
Chemicals-Hydrogen peroxide solutions were prepared daily and the concentration was determined using ⑀ 240 nm ϭ 39.4 M Ϫ1 cm Ϫ1 (17). Veratryl alcohol, veratraldehyde, 4-MMA, and anisaldehyde were purchased from Aldrich. Veratryl alcohol was vacuum-distilled before use; all other chemicals were used without further purification.
Product Analysis by High Pressure Liquid Chromatography-Reaction mixtures were sampled (100 l) and added to 400 l of methanol. A 10-l aliquot of the resultant solution was injected onto a reverse phase C18 (VYDAC TM ) column eluted at 0.5 ml/min with a 5-100% linear methanol gradient in water. Products were monitored at 280 nm and identified by comparison to retention times of authentic standards.
Steady-state Kinetics-The rate of veratraldehyde and anisaldehyde formation was determined spectrophotometrically. The absorption spectra of the alcohols and their corresponding aldehydes are shown in Fig. 1. Veratraldehyde formation was monitored at 330 nm (⑀ ϭ 1.9 mM Ϫ1 cm Ϫ1 ) rather than at 310 nm as in previous studies since anisaldehyde is transparent at this wavelength. Anisaldehyde formation was determined by the absorbance increase at 300 nm (⑀ ϭ 4.5 mM Ϫ1 cm Ϫ1 ). The veratraldehyde contribution to the absorbance increase at 300 nm (⑀ ϭ 7.4 mM Ϫ1 cm Ϫ1 ) was determined by its absorbance at 330 nm. Presteady-state Kinetics-The KinTek three-syringe stopped-flow apparatus was used as described previously (7). The reactions of LP compound I were monitored at 426 nm, the isosbestic wavelength of native enzyme and compound I. Rate constants for compound II were determined from experiments where compound II was formed from the reaction of compound I with the reductant of choice. The reactions were at 28°C under pseudo-first order conditions. The double mixing system contained 3 M LP in water which was mixed initially with 3 M H 2 O 2 . The resultant compound I was then mixed with the contents of the third syringe which contained the reductant (4-MMA) in 150 mM sodium tartrate, pH 3.5.
ESR Quantitation of Cation Radical-The veratryl alcohol cation radical formed from LP was quantitated by acid quench with a threesyringe fast-flow system as described previously (11). The contents of the first two syringes, one containing LP and veratryl alcohol, the other containing H 2 O 2 , were mixed for 0.5 s. The contents were then quenched with 10% HNO 3 . Radical concentration was determined by double integration of the first derivative spectrum. Tempol was used as a standard. Tempol was quantitated at 240 nm using an extinction coefficient of 1,440 M Ϫ1 cm Ϫ1 (18).

RESULTS
Steady-state Kinetics-The effect of veratryl alcohol concentration on the rate of anisaldehyde formation from 4 mM 4-MMA was determined spectrophotometrically (Fig. 2). Formation of veratraldehyde and anisaldehyde in these steadystate experiments was confirmed by high pressure liquid chromatography analyses (data not shown). Little or no anisaldehyde is formed in the absence of veratryl alcohol. Increasing veratryl alcohol concentration resulted in an increase in the rate of anisaldehyde formation. The hyperbolic curve fits Michaelis-Menton kinetics with an apparent K m of 240 M for veratryl alcohol. This K m is within experimental error for the K m value of 172 M determined for veratryl alcohol in the absence of 4-MMA (14). Due to the high 4-MMA concentration used in these experiments and the little or no change in the apparent K m for veratryl alcohol, this would indicate that 4-MMA minimally competes with veratryl alcohol at the active site. The effect of pH on 4-MMA oxidation is similar to that obtained with just veratryl alcohol oxidation (data not shown). The V max increases as the pH decreases.
The effect of 4-MMA concentration on veratraldehyde and anisaldehyde formation at a constant veratryl alcohol concentration is shown in Fig. 3. With a constant veratryl alcohol concentration of 2 mM, increasing 4-MMA caused a rapid decrease in the rate of veratraldehyde formation. Complete inhibition was observed at a concentration of less than 1 mM. Concomitant with the decrease in veratraldehyde formation is the increase in anisaldehyde formation. The rate reached a peak at 150 M 4-MMA. This was followed by a slow and gradual decrease in rate. The decrease in rate can be accounted for by competition of 4-MMA with veratryl alcohol at the active site. Based on the shape of the curve, the K i is predicted to be in excess of 4 mM. This is consistent with the results shown in Fig. 2 where the K m for veratryl alcohol determined in the absence of 4-MMA approximates the value determined in the presence of 4 mM 4-MMA.
Kinetic traces monitoring veratraldehyde formation at 330 nm showed lag phase kinetics at very low 4-MMA concentrations (Fig. 4). This is reminiscent of the effect of guaiacol on veratryl alcohol oxidation (14). The length of the lag period was roughly proportional to the 4-MMA concentration, suggesting that veratryl alcohol oxidation did not proceed until most if not Presteady-state Kinetics-The reactivity of 4-MMA with compound I and II of LP was studied with a three-syringe stoppedflow technique under pseudo-first order reaction conditions. Compound I was generated in the stopped-flow by mixing one equivalent of LP with H 2 O 2 and aged for 4 s before mixing with 4-MMA. Similar to results obtained with anisyl alcohol (14), 4-MMA reacts with compound I but much more slowly than veratryl alcohol (Fig. 5). A rate constant of 4.3 ϫ 10 3 M Ϫ1 s Ϫ1 was calculated from the slope for the reaction of compound I with 4-MMA. Reactions of compound II with other reductants always exhibit a hyperbolic concentration dependence. We were not able to observe this with 4-MMA at higher concentrations (data not shown). This suggests that 4-MMA has little or no reactivity with compound II. When extremely high concentrations are required to observe a reaction, a low level contaminant could account for the reaction; this we have observed with anisyl alcohol (14).
ESR Spectroscopy-The ESR spectrum of the veratryl alcohol cation radical is shown in Fig. 6. The radical was generated by enzymatic incubations quenched with HNO 3 . Stoichiometric amounts of 25 M enzyme was incubated with H 2 O 2 in the presence of 2 mM veratryl alcohol and varying concentrations of 4-MMA. The ESR signal intensity was determined by double integration and calculated as described previously (12). Spectrum A is an incubation without 4-MMA; a value of 7.3 M was calculated for the cation radical. Addition of 0.5 mM 4-MMA decreased the signal intensity by 53%. Addition of 1 mM 4-MMA resulted in an 85% decrease in signal intensity. DISCUSSION Veratryl alcohol was first detected in ligninolytic cultures of P. chrysosporium by Lundquist and Kirk in 1978 (9), long before the discovery of LP (20,21). It was not until the discovery of LP that a role for veratryl alcohol in lignin degradation was first proposed. Veratryl alcohol not only protects the enzyme from H 2 O 2 -dependent inactivation (22) but also serves as a convenient assay substrate for the enzyme (23). The most debated role is that of a redox mediator between LP and polymeric lignin. This role was initially formulated from data showing the ability of veratryl alcohol to enhance the oxidation of recalcitrant compounds that are not directly oxidized by the enzyme. Lignin (24), anisyl alcohol (10), 4-MMA (10), chloropromazine (25), and guaiacol (26) are only oxidized by LP if veratryl alcohol is included in the reaction mixture. Although the mediation mechanism provided a suitable explanation for the phenomenon and provided a model for how a large bulky enzyme could degrade an insoluble large bulky polymer, there were chemical arguments against the model. Despite its proposed existence as a stable mediator able to participate in oxidation reactions at a distance, there was no direct proof of the radical for many years. Khindaria et al. (11) were the first to detect the veratryl alcohol cation radical by ESR spectroscopy. They first generated and characterized the species by oxidation with Ce(IV) in 10% HNO 3 . After establishing the ESR parameters, these workers were then able to detect the enzyme-bound radical during steady-state turnover of LP at pH 3.5. At this pH, the radical was not detected when generated by Ce(IV) oxidation indicating that the radical is not stable free in solution. Khindaria et al. (12) proposed that the enzyme extends the half-life of the radical.
Candeias and Harvey (13) also characterized the veratryl alcohol cation radical. They generated the species with thallium (II) and studied it by pulse radiolysis. The spectral and kinetic properties of the radical determined by Candeias and Harvey (13) did not totally agree with those of Khindaria et al. (12). Nevertheless, these workers determined that the half-life of the radical was approximately 60 ms in solution thus allowing it to diffuse a distance of 7 m. This would allow for it to act as a diffusible oxidant. Khindaria et al. (12) determined a half-life of 0.57 ms resulting in a much lower diffusion radius. Regardless, Candeias and Harvey (13) demonstrated that the radical was capable of diffusing far enough to react with the polymeric dye Poly R-478. Surprisingly, no reaction was observed with either anisyl alcohol or 4-MMA. Candeias and Harvey (13) rationalized that the enzyme could serve to in-crease the reactivity of the radical (with positively charged residues) thus allowing the veratryl alcohol cation radical to oxidize both anisyl alcohol and 4-MMA. Alternatively, they indicated that if the enzyme extended the half-life of the cation radical, the probability of reaction would be increased.
Our studies with anisyl alcohol clearly showed that its oxidation by LP is not mediated by veratryl alcohol (14). We found that anisyl alcohol is not oxidized by compound II. Compound I, in contrast, is able to readily oxidize anisyl alcohol. Thus with only anisyl alcohol, the enzyme would be stuck at the compound II intermediate. The addition of veratryl alcohol, which is readily oxidized by compound II would allow the enzyme to complete the catalytic cycle. We showed that this mechanism would predict stimulation of anisyl alcohol oxidation at low veratryl alcohol concentrations. This is because reactions with compound II to complete the catalytic cycle (enhancement) would offset the effect of veratryl alcohol reacting with compound I (inhibitory). At higher concentrations of veratryl alcohol, the rate of anisyl alcohol oxidation should decrease because it would more effectively compete with anisyl alcohol at compound I. Consistent with the proposed mechanism, our data show that an optimal rate of anisyl alcohol oxidation is observed at 0.1 mM veratryl alcohol (14).
Two recent studies have shown that veratryl alcohol enhanced oxidation of a second substrate is due to mediation.  (13). B, proposed mechanism for mediation by veratryl alcohol. The mechanism is adapted from work of Hammel et al. (28) and Schmidt et al. (29). The oxidation of veratryl alcohol under aerobic conditions results in formation of the benzylic carbon-centered radical which then adds dioxygen yielding the ␣-hydroxybenzylperoxyl radical. Decomposition of this radical yields superoxide and veratraldehyde. Alternatively, the cation radical can oxidize 4-MMA. The mechanism shows electron abstraction from the carboxylate, however, abstraction from the aromatic ring yielding the cation radical is not ruled out. Elimination of CO 2 results in formation of the benzylic carbon-centered radical which leads to anisaldehyde formation by a mechanism similar to that proposed for veratraldehyde. that the oxidation of guaiacol was mediated by veratryl alcohol. The proposed mechanism for mediation of these substrates is shown in Fig. 7. Kinetic simulation of mediation reveals two unique aspects which are born out by the data: (i) veratryl alcohol oxidation does not occur until the secondary substrate is depleted (lag phase kinetic traces) and (ii) the effect of veratryl alcohol on the oxidation of the secondary substrate exhibits saturation kinetics. Both of these properties are observed with chloropromazine (25) and guaiacol (26). These properties differ from those observed with anisyl alcohol where no lag phases are observed and an optimal rate is observed at low veratryl alcohol concentrations.
Results obtained here with 4-MMA oxidation by LP resemble those obtained with guaiacol and chloropromazine oxidation. The addition of 4-MMA caused lag phase kinetics on veratryl alcohol oxidation (Fig. 4) and saturation kinetics were observed when the veratryl alcohol concentration was varied (Fig. 2). These two results clearly point to mediation of 4-MMA oxidation by veratryl alcohol. Mediation is also consistent with the ESR results. The veratryl alcohol cation radical generated by compound I decreased in concentration in incubations containing 4-MMA. The decrease in signal intensity can be attributed to either competitive inhibition at the active site or to the reaction of 4-MMA with the cation radical. If the inhibition were due to competition, the decrease in signal intensity can be predicted based on the relative rate constants of compound I with veratryl alcohol and 4-MMA (1.5 ϫ 10 5 M Ϫ1 s Ϫ1 versus 4.3 ϫ 10 3 M Ϫ1 s Ϫ1 ). These predicted values are 0.7% inhibition at 0.5 mM 4-MMA and 1.4% inhibition at 1 mM 4-MMA. The data indicates 53 and 85% inhibition; this is consistent with 4-MMA reacting with the cation radical thus causing the decrease in the ESR signal intensity.
Our results with anisyl alcohol would predict that 4-MMA oxidation is not mediated by veratryl alcohol. Like anisyl alcohol, 4-MMA is a monomethoxylated substrate whose reduction potential would be predicted to be higher than that of veratryl alcohol. Examination of the 4-MMA structure reveals how it, despite being a monomethoxylated substrate, could be oxidized by the veratryl alcohol cation radical. The elimination of the ␤-carboxyl group as CO 2 would drive the reaction forward (Fig.  7B). Candeias and Harvey (13) suggested that the lack of reactivity between the veratryl alcohol cation radical and 4-MMA in solution can be explained by the short half-life of the cation radical. Our kinetic simulations, described below agree with this conclusion. We simulated the effect of 4-MMA on veratraldehyde formation (similar to data shown in Fig. 4) with a short-lived and a long-lived veratryl alcohol cation radical. The mechanism used for this simulation is shown in Fig. 8A. Khindaria et al. (12) measured the half-life of the cation radical in solution and enzyme-bound to be 0.57 and 370 ms, respectively. Candeias and Harvey (13) estimated the rate constant between the veratryl alcohol cation radical with 4-MMA to be 5 ϫ 10 4 M Ϫ1 s Ϫ1 (we used 9 ϫ 10 4 M Ϫ1 s Ϫ1 ). With a k cat of 17 s Ϫ1 for LP and a half-life of 370 ms for the enzyme bound cation radical, the simulations shown in Fig. 8B were obtained which resemble the real data shown in Fig. 4. When the half-life of the cation radical is decreased to 0.57 from 370 ms, no inhibition of veratraldehyde formation is observed (Fig. 8B), indicating that its half-life was too short for reactions with 4-MMA.
In conclusion, we have shown that 4-MMA oxidation is mediated by veratryl alcohol. This work supports the hypothesis put forth by Candeias and Harvey (13) suggesting that extending the half-life of the radical can increase the reaction yield. However, our results still do not support the ability of veratryl alcohol to act as a diffusible oxidant. The cation radical is too short-lived when free in solution to oxidize other substrates. The results of Khindaria et al. (12) and ours clearly show that it must be stabilized to participate in electron transfer reactions with other aromatic substrates. Nevertheless, the list of substrates for which veratryl alcohol serves as an electrontransfer agent continues to grow. It therefore, would not be surprising to find that it serves such a role with lignin. Furthermore, as suggested by Shoemaker et al. (27), other mechanisms may be operating in fungal cultures to stabilize the cation radical and allow it to act as a diffusible oxidant. FIG. 8. A, mechanism used in kinetic simulation. The mechanism shown is simplified to illustrate, among other points, the different effect of 4-MMA on veratraldehyde formation depending on the half-life of the cation radical. Rate constants are: k 1 ϭ 1.5 ϫ 10 5 M Ϫ1 s Ϫ1 ; k 2 ϭ 17 s Ϫ1 ; k 3 ϭ 1.9 s Ϫ1 for decay of the enzyme bound species or 1.2 ϫ 10 3 s Ϫ1 for the free cation radical; and k 4 ϭ 9 ϫ 10 4 M Ϫ1 s Ϫ1 . VA is veratryl alcohol; VA .
ϩ is the cation radical; and AAD is anisaldehyde. B, simulation of the effect of 4-MMA on veratraldehyde formation. Increasing concentrations of 4-MMA, as specified in the figure, comparable to results shown in Fig. 4, caused a proportionate increase in the lag period. All but one of the simulations used a decay rate for the cation radical was 1.9 s Ϫ1 . The simulation shown in the uppermost line labeled 200 M 4-MMA used a rate constant of 1.2 ϫ 10 3 s Ϫ1 for the decay rate of the cation radical.