First evidence of catalytic mediation by phenolic compounds in the laccase-induced oxidation of lignin models Francesca d’Acunzo and Carlo Galli Dipartimento di Chimica, Universita ` ‘La Sapienza’, Roma, Italy; IMC-CNR Sezione Meccanismi di Reazione The sulfonephthalein indicator, phenol red, exhibits an unusually slow rate of oxidation by laccase from Polipo- rus pinsitus, in spite of the fact that it is a phenol and therefore a natural substrate for this phenoloxidase enzyme. Nevertheless, after prolonged exposure to laccase (24 h) phenol red is oxidized by more than 90%. We found that phenol red, which can be oxidatively converted into a reso- nance-stabilized phenoxy radical, performs as a mediator in the laccase-catalyzed oxidation of a nonphenolic substrate (4-methoxybenzyl alcohol) and also of a hindered phenol (2,4,6-tri-tert-butylphenol). In particular, phenol red was found to be at least 10 times more efficient than 3-hydr- oxyanthranilate (a reported natural phenolic mediator of laccase) in the oxidation of 4-methoxybenzyl alcohol. Other phenols, which do not bear structural analogies to phenol red, underwent rapid degradation and did not perform as laccase mediators. On the other hand, several variously substituted sulfonephthaleins, of different pK 2 values, mediated the laccase catalysis, the most efficient being dichlorophenol red, which has the lowest pK 2 of the series. The mediating efficiency of phenol red and dichlorophenol red was found to be pH dependent, as was their oxidation E p value (determined by cyclic voltammetry). We argue that the relative abundance of the phenoxy anion, which is easier to oxidize than the protonated phenol, may be one of the factors determining the efficiency of a phenolic mediator, together with its ability to form relatively stable oxidized intermediates that react with the desired substrate before being depleted in undesired routes. Keywords: laccase; phenolic mediators; lignin models; radi- cals; acidity. Laccases (EC 1.10.3.2) are multicopper oxidases, produced by micro-organisms and plants, which participate in nature in both the biosynthesis and degradation of lignin [1]. In the latter case, laccase operates in conjunction with other ligninolitic enzymes, such as lignin peroxidase and man- ganese peroxidase. The latter two enzymes, which are stronger oxidants than laccase, are able to oxidize most aromatic constituents of lignin, whereas laccase oxidizes directly only the phenolic subunits, which are easier to oxidize but relatively less abundant (15%). It has been speculated [1], however, that laccase may react indirectly with nonphenolic lignin components through mediation by phenolic species present in its natural environment. These Ônatural mediatorsÕ could be metabolites [2], or even lignin fragments [3] generated by the other ligninolitic enzymes, and would open up alternative, possibly radical, routes to the oxidation of nonphenolic components [1]. It is quite reasonable to argue that natural mediators may participate in the laccase-catalyzed oxidation of nonphenolic lignin subunits in those micro-organisms that only rely on laccase for their ligninolitic action [4,5], and a number of sub- stances, such as phenolic acids [6,7] and 3-hydroxyanthrani- late (HAA) [8], have indeed been proposed as natural laccase mediators. However, in order to prime laccase towards the oxidation of nonphenolic substrates, a large excess of these mediators is often needed [7,9] or, at least, a stoichiometric amount is required in the most favourable cases [10]. This might be because phenolic mediators are, themselves, ÔgoodÕ substrates for the enzyme. Laccase would oxidize them to short-lived reactive species that undergo further reactions (noncatalytic undesired routes in Fig. 1: radical coupling, fragmentation, etc.), preventing their reduction to the original state by the nonphenolic substrate they are supposed to oxidize (Fig. 1, catalytic cycle). As an example, dimerization of the phenolic compound, HAA, once oxidized by laccase, is so fast and quantitative that we exploited the formation rate of this dimer in a new spectrophotometric assay of laccase activity in mixed solvents [11]. In the context of our studies [12–15] on laccase-mediator systems, and of the speculations on the natural role of phenolic mediators [1], we were intrigued by the results of Li et al. [9] on the violuric acid-mediated oxidation of phenol red by laccase. It was reported, in fact, that some fungal laccases oxidize phenol red (Fig. 2; an advocated phenolic lignin model) at a negligible rate, but that this rate increases substantially in the presence of violuric acid when it is added in up to a 100· molar excess with respect to phenol red. Why is phenol red oxidized so slowly, by laccase, that a mediator (a large excess of violuric acid, in this case) is needed in order to speed up its metabolization? By studying Correspondence to C. Galli, Dipartimento di Chimica, Universita ` ÔLa SapienzaÕ, 00185 Roma, Italy. Fax: + 39 06 490 421, Tel.: + 39 06 4991 3386, E-mail: carlo.galli@uniroma1.it Abbreviations: ABTS, 2,2¢-azinobis(3-ethylbenzo-6-thiazolinesulfonic acid); HAA, 3-hydroxyanthranilic acid; HBT, 1-hydroxybenzo- triazole. Enzyme: laccases (EC 1.10.3.2). (Received 19 May 2003, revised 10 July 2003, accepted 14 July 2003) Eur. J. Biochem. 270, 3634–3640 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03752.x the peculiar behaviour of phenol red and other phenol- sulfonephthaleins we aimed to gain a better insight into the interaction between laccase and phenols in general. More specifically, we wished to identify the characteristics that a phenol should have in order to be a good laccase mediator, i.e. one that follows the catalytic cycle shown in Fig. 1 without being rapidly metabolized by the enzyme (undesired routes). In this study, we first investigated the kinetics of oxidation of phenol red by Poliporus pinsitus laccase in the absence of any mediator, and then evaluated the ability of phenol red to act as a laccase mediator in the oxidation of a hindered phenol and of a nonphenolic compound. We extended this study to other sulfonephthalein indicators, characterized by different pK 2 values, and investigated the effect of solution pH on the mediating ability of phenol red and dichlorophenol red. Materials and methods Laccase Laccase from P. pinsitus was kindly donated by Novo Nordisk Biotech; it was purified by ion-exchange chroma- tography on Q-Sepharose by elution with phosphate buffer; laccase fractions with an absorption (A) 280 /A 610 ratio of 20–30 were considered sufficiently pure [16]. The collected fractions were concentrated by dialysis in cellulose mem- brane tubing (Sigma) against poly(ethylene glycol) to a final activity of 9000 UÆmL )1 , as determined spectrophotometri- cally by the standard reaction with 2,2¢-azinobis(3-ethyl- benzo-6-thiazolinesulfonic acid) (ABTS) [17]. Reagents 4-Methoxybenzylalcohol, HAA, 1-hydroxybenzotriazole (HBT) and all phenols employed (Aldrich) were used as received from the manufacturer. Buffers were prepared using ultrapure water obtained from a MilliQ apparatus. Kinetics of laccase-catalyzed oxidation of phenol red and 2,4,6-trichlorophenol In a 10-mm quarz UV-Vis cell, the reagent (phenol red or 2,4,6-trichlorophenol; 40 l M in 0.1- M citrate buffer, pH 5.0) was allowed to react with 0.3 UÆmL )1 of laccase at 25 °C. Phenol red consumption was monitored at 420 nm, while dichloroquinone formation was monitored at 273 nm, using an HP 8453 diode array UV-Vis spectrophotometer run with HP UV - VIS CHEMSTATION software. The rate of phenol red oxidation is expressed as the initial rate of depletion of the absorption at the wavelength followed (e.g. DA 420 Æ min )1 ), in keeping with the literature [9]. The complexity of the kinetic trend, i.e. a biphasic profile (Fig. 3), and the possibility that the absorption at 420 nm may be also attributed to products other than phenol red, prevented us from converting data from absorbance units to concentration units. Laccase-catalyzed oxidation of phenol red Phenol red (20 l M in 0.1- M sodium citrate buffer, pH 5.0), was incubated with 4 UÆmL )1 of purified laccase for 24 h at 25 °C. The amount of residual phenol red, after metaboli- zation by the enzyme, was determined by HPLC. The internal standard (3,4-dimethoxybenzaldehyde) was added to the reaction mix, which was then diluted in the mobile Fig. 2. Structure of phenol red and related molecules. Fig. 3. Degree of conversion of phenol red (thick line) and 2,4,6-tri- chlorophenol (thin line) as a function of time. The kinetics of laccase- catalyzed oxidation of phenol red and 2,4,6-dichlorophenol were monitored spectrophotometrically through the decrease in absorption at 420 nm and the increase in absorption at 273 nm, respectively. Reaction conditions: phenol red or 2,4,6-trichlorophenol, 40 l M in 0.1 M citrate buffer, pH 5.0; laccase, 0.3 UÆmL )1 . Calculation of the degree of conversion: C ¼ 100 AðtÞÀAðt 0 Þ Aðt f ÞÀAðt 0 Þ where t f ¼ 20 h for phenol red (>90% conversion by HPLC), but t f ¼ 4 h for trichlorophenol [the time at which A(t) has already reached its plateau value]. The inset shows the spectrum of phenol red at t ¼ 0andt ¼ 20 h; residual absorption at 420 nm after 20 h cannot be entirely attributed to phenol red (according to HPLC data), but rather to some unidentified product(s). Fig. 1. The catalytic cycle of a laccase-mediator system, where the oxidized mediator partitions between a significant undesired side-reaction and a catalytic cycle of oxidation of a nonphenolic substrate. Ó FEBS 2003 A phenolic mediator for the enzyme laccase (Eur. J. Biochem. 270) 3635 phase and filtered through 0.2-lm Teflon syringe filters (Superchrom Varisep) prior to analysis. We used an Agilent Technologies HPLC system (pump, degasser, UV-Vis detector and solvent delivery system) equipped with a Zorbax Agilent Eclipse XDB-C 8 15 cm · 4.6 mm column, and run with Agilent CHEMSTATION FOR LC software. The elution was carried out with an MeOH/H 2 O (40 : 60) mobile phase containing 0.3& trifluoroacetic acid, at a flow rate of 0.8 mLÆmin )1 , and products were detected at 265 nm. Oxidation of lignin model compounds 4-Methoxybenzyl alcohol or 2,4,6-tri-t-butylphenol (Ald- rich) (20 l M in 0.1 M citrate buffer, pH 5.0), was incubated for 24 h at 25 °Cwith4UÆmL )1 of purified laccase and a mediator (HAA, HBT, or one of the phenolic derivatives listed in Table 1; 6.7 l M ). Reaction products were identified by GC-MS analysis, run on a HP 5892 GC equipped with a 12-m · 0.2 mm methyl silicone gum capillary column, coupled to an HP 5962 MSD instrument operating at 70 eV. The yields of oxidation (yield of 4-methoxybenzal- dehyde or consumption of 2,4,6-tri-t-butylphenol) were determined by GC analysis with respect to p-methoxy- acetophenone as the internal standard, using a Varian 3400 Star instrument fitted with a 20 m · 0.2 mm methyl silicone gum capillary column coupled to an FID detector. Cyclic voltammetry The electrochemical equipment consisted of a computer- controlled home-made potentiostat with a Vernier Soft- ware MULTI PURPOSE LABORATORY INTERFACE ( MPLI )pro- gram for WINDOWS . The three-electrode system consisted of a glassy-carbon disc (of 1.5 or 3 mm diameter) working electrode, an aqueous Hg/HgCl 2 /saturated KCl reference electrode (E° vs. NHE ¼ E° vs. SCE + 0.242 V), and a Pt reference electrode (1 cm 2 ). Prior to recording each scan, the working electrode was polished using a Cypress Systems polishing kit, sonicated for 1 min, and rinsed with distilled water. All scans were obtained at room temperature. NaH 2 PO 4 /Na 2 HPO 4 buffers (0.5 M )atpH4.0and pH 7.4, respectively, were prepared by carefully weighing appropriate amounts of monobasic and dibasic phosphates, which were dissolved in water previously filtered through a MilliQÒ apparatus (Millipore, France). The cyclic voltam- metry scans of 0.5-m M phenol red or dichlorophenol red in solutions of different pH were run at a rate of 0.5 VÆs )1 .The cyclic voltammetry scans of 0.5 m M phenol red in the presence of 2.5–30 m M 4-methoxybenzyl alcohol were run at a rate of 5 mVÆs )1 . Results and discussion Kinetics of the laccase-catalyzed oxidation of phenol red In a previous study, on the oxidation of oligomeric phenols, we unambiguously pinpointed poor substrate solubility and steric hindrance as two factors that make the use of a mediator necessary for the laccase-catalyzed oxidation of otherwise recalcitrant phenolic substrates [18]. Phenol red, however, does not seem to present any steric problem, and its solubility is not limited in aqueous solutions. Therefore, its reported [9] lack of reactivity with laccase is puzzling. By following the reaction spectrophotometrically at 420 nm, we found that the initial oxidation rate of phenol red (3 · 10 )3 DAÆmin )1 for a 1-h reaction time) in the absence of a mediator was significantly slower than that of 2,4,6-trichlorophenol (Fig. 3), a structurally simple phenol that laccase easily oxidizes to dichlorobenzoquinones (most probably a mixture of o-andp-benzoquinone isomers, the latter being the most abundant: see Fig. 4 for product identification by MS [19]). However, the decrease in absorption at 420 nm proceeded slowly throughout the 20-h time span during which the reaction was monitored (the rate at 10–20 h was 3 · 10 )5 DAÆmin )1 ). For a direct comparison of the kinetic data of phenol red and 2,4,6-trichlorophenol, the degree of conversion of the two phenols, as a function of time, was plotted (see the equation in the legend to Fig. 3). Figure 3 clearly shows that, while 2,4,6-trichlorophenol was quantitatively meta- bolized within 2 h, the conversion of phenol red was only % 80% after 5 h. HPLC analysis (where phenol red is chromatographically separated from other components of the reaction mixture that may absorb at 420 nm) indicated that <10% residual phenol red was present after 20 h of exposure to laccase. In Fig. 3 therefore the A 420 (at 20 h) was taken as an approximation of the residual absorption at ÔzeroÕ phenol red concentration. This residual absorption may be attributed to some product(s) of phenol red metabolization, which, in turn, could react slowly with laccase. As a result, the kinetic curve is biphasic and the data cannot be expressed simply in terms of phenol red concentration. Table 1. Phenolic structures investigated as laccase mediators in the oxidation of 4-methoxybenzyl alcohol at pH 5.0, unless stated otherwise. Reaction conditions: substrate, 20 m M in 0.1 M citrate buffer, pH 5.0; mediator concentration, 6.7 m M ; laccase concentration, 4 UÆmL )1 ; reaction time, 24 h. pK ArOH values obtained from refs [25], [26], [29] and [30]. Entry Mediators investigated Yield (%) of 4-methoxy- benzaldehyde pK ArOH 1 Phenol red 20 7.42 2 Phenol red, 24 h 35 3 Phenol red, pH 3.5 <5 4 Phenol red, pH 5.5 20 5 Phenol red, pH 6.0 15 6 Dichlorophenol red 55 5.74 7 Dichlorophenol red, pH 3.5 10 8 Dichlorophenol red, pH 5.5 30 9 Dichlorophenol red, pH 6.0 10 10 Cresol red 20 7.87 11 m-Cresol red 15 8.04 12 Phenolphthalein <5 9.4 13 Aurintricarboxylic acid <5 >7.2 14 Chrome azurol S <5 15 Resveratrol <5 16 Alizarin red <5 17 Quercetin <5 18 2,6-Dimethoxy-4-allylphenol <5 19 2,6-Dichloroindophenol <5 3636 F. d’Acunzo and C. Galli (Eur. J. Biochem. 270) Ó FEBS 2003 Oxidation of lignin model compounds by the laccase/phenol red system In the previous paragraph we presented evidence to suggest that the laccase/phenol red couple is a Ôlong-lastingÕ reacting system, compared, for example, with the laccase/2,4,6- trichlorophenol couple. Phenol red is probably oxidatively converted into a phenoxy radical [3], like any other laccase- oxidizable phenol. If its phenoxy radical, or secondary products thereof, were sufficiently long-lived, phenol red might behave as a laccase mediator in radical oxidation routes towards recalcitrant substrates [20]. We tested this hypothesis in the oxidation of the nonphenolic 4-meth- oxybenzyl alcohol, and also of the hindered 2,4,6-tri- t-butylphenol. The former was not oxidized by laccase in the absence of a mediator, whereas the latter yielded only a trace of di-t-butylbenzoquinone (predominantly the p-benzoqui- none isomer. See Fig. 4 for product identification by MS). Oxidation of 4-methoxybenzyl alcohol. In Table 2 we report the amount of substrate converted with phenol red acting as the mediator, compared to the laccase/HAA (a phenolic mediator) and laccase/HBT systems, HBT being a well-known mediator of laccase, having an >N-OH structure [15]. The mediator is always in defect with respect to the substrate, unlike many phenolic mediators reported in the literature [7,9]. The only product detected in the medi- ated oxidations of 4-methoxybenzyl alcohol is 4-methoxy- benzaldehyde, and recovery of the unreacted alcohol complements the amount of product quantitatively. Table 2 shows that, while the conversion of 4-meth- oxybenzyl alcohol is much higher with HBT, phenol red is at least 10 times more efficient than HAA, a reported naturally occurring phenolic laccase mediator [8]. The laccase/phenol red system maintains its oxidizing activity even during the second day of reaction, when phenol red is no longer present. Therefore, part of the mediation efficiency of phenol red, compared with other structurally simple phenols, is the result of by-products that are still reactive towards the substrate. Oxidation of 2,4,6-tri-t-butylphenol. Phenol red per- formed significantly better than HBT as a mediator of laccase in the oxidation of 2,4,6-tri-t-butylphenol to di- t-butylbenzoquinone (predominantly p-benzoquinone isomer: see Fig. 4 for product identification). The laccase- generated phenoxy radical of phenol red, in this case, would remove the H-atom from the OH group of the hindered phenol in an almost thermoneutral step; the resulting 2,4,6- tri-t-butylphenoxyl radical drives the reaction towards the loss of isobutene and results in the observed di-t-butyl- benzoquinones. Two moles of substrate are oxidized per mole of mediator, indicating that phenol red participates in a catalytic cycle such as the one described in Fig. 1. A mechanistic parallel between N-hydroxy and phenolic mediators The laccase mediator, HBT, is oxidatively converted by the enzyme into an >N–O • reactive intermediate [21,22]. Electrochemical data [23] show that the >N–O • radical of HBT is sufficiently long-lived to be able to abstract a benzylic hydrogen atom from a benzylic alcohol, converting it into the aldehyde through the intervention of dioxygen (Fig. 5). This route, via a radical, circumvents the low tendency of a benzyl alcohol to be involved in an electron- transfer route with laccase, and makes its oxidation possible [13,21]. This hydrogen abstraction route is thermodynami- cally feasible because the energy of the O–H bond that N–O • forms is comparable to that of the benzylic C–H bond that is cleaved from the substrate [12,21]. It can be suggested that, if the laccase-generated phenoxy radical from phenol red is resonance-stabilized [24], and therefore as long-lived as the >N–O • radical from HBT Fig. 4. Fragmentation patterns that are diagnostic for o- and p-benzo- quinone isomerism [19]. R ¼ Cl, m/z=52 (amenable to 2,6-dichloro- benzoquinone) is the base peak; m/z=110 is 52% of the base peak. R ¼ t-Bu, the abundance of m/z ¼ 52, 53, 67 (amenable to 2,6-di- t-butylbenzoquinone) altogether is five times that of m/z ¼ 164; the base peak is (M-43) ¼ 177. Table 2. Oxidation (%) of 4-methoxybenzyl alcohol and 2,4,6-tri-t- butylphenol by laccase/phenol red, compared with two other laccase- mediator systems. The reaction conditions were as follows: substrate, 20 m M in 0.1 M citrate buffer, pH 5.0; mediator concentration, 6.7 m M ; laccase concentration, 4 UÆmL )1 ; reaction time, 24 h. HAA, 3-hydroxyanthranilic acid; HBT, 1-hydroxybenzotriazole. Substrate Mediator Phenol red HAA HBT 4-Methoxybenzyl alcohol 20 a 2 a [15] 75 a [15] 35 a,b –– 2,4,6-tri-t-butylphenol 55 c –30 c a Yield of aldehyde. b Reaction time: 48 h. c Yields of di-t-butyl- benzoquinone were determined by gas chromatography. Ó FEBS 2003 A phenolic mediator for the enzyme laccase (Eur. J. Biochem. 270) 3637 [22], it has time to follow an analogous H-abstraction route of oxidation of the nonphenolic substrate. In fact, phenol red is more acidic (pK 2 ¼ 7.42) [25] than simple phenols (pK a ¼ 9–10) [26]. This is a result of the delocalization of the negative charge of the anion onto the adjacent quinoid ring (see Fig. 2). Analogously, delocalization of the un- paired electron of the corresponding phenoxy radical onto the quinoid ring would enhance the survival time of the phenoxy radical intermediate from phenol red, giving it a greater chance to take part in catalytic cycles (Fig. 1) before being consumed in undesired routes. Because O–H bond energies of phenols (approximately 82–88 kcalÆmol )1 )are similar to those of the C–H benzylic bonds of the alcohol [21], the radical route of the phenoxy radical is as thermodynamically feasible as that of the > N–O • species from a >N–OH mediator. This suggested route bears strong analogies to the reported mechanism of oxidation of primary alcohols to aldehydes by the enzyme galactose oxidase [27], where the phenoxy radical of a tyrosine residue abstracts an H-atom from the alcoholic substrate [28]. Our hypothesis, that part of the mediating ability of phenol red (i.e. besides the reactivity of its secondary oxidation products) may be attributed to its phenoxy radical, finds support in the following results obtained by cyclic voltammetry. Figure 6 shows the irreversible vol- tammogram corresponding to the oxidation of phenol red to its phenoxy radical. More precisely, at the pH of the experiment (7.4), phenol red (pK 2 ¼ 7.42) is % 50% deprotonated, so that the pH-dependent E p results from both the oxidation of the phenol to the corresponding radical cation (which rapidly releases a proton to yield the phenoxy radical) and to the oxidation of the phenolate ion to the phenoxy radical directly [29]. The scans run at pH 7.4, in the presence of an increasing excess of 4-methoxybenzyl alcohol, showed some increase in current intensity (% 10% increase with a fivefold excess of 4-methoxybenzyl alcohol; 50% increase with a 60-fold excess; we were unable to achieve a larger excess of alcohol because of its limited solubility in the buffer). This indicates that phenol red is, at least in part, regenerated from its phenoxy radical through the abstraction of a benzylic hydrogen from the substrate, so that it can be oxidized once again at the electrode, this resulting in an increase of charge transport and therefore of current intensity. Effect of the p K 2 of the mediator and of solution pH on the laccase-catalyzed oxidation of 4-methoxybenzylalcohol So far we have shown that the laccase/phenol red couple is a long-lasting reacting system that is able to perform as a laccase-mediator system. In our search for other types of phenolic mediators, we tested a number of other phenolic structures, mostly antioxidants, by means of the benchmark oxidation of 4-methoxybenzylalcohol. Most turned out to be unable to perform as mediators (Table 1, entries 15–19), because they undergo rapid degradation. We believe that the potential for the oxidative formation of highly conju- gated (and, hence, stabilized) oxyradicals in structures akin to phenol red may be responsible for their peculiar behavior. We have provided some partial support to this view through the electrochemical experiment depicted in Fig. 6 and described above, in the previous paragraph. These consid- erations, combined with the fact that sulfonephthaleins, phenolphthaleins and related structures (Fig. 2) cover a wide range of pK values for the dissociation of their phenolic moiety, and that the oxidation potential of phenols is modulated by solution pH (at pH < pK) [29], led us to expand on the subject of phenol red-like mediators. All sulfonephthaleins exhibit some mediating effect, while aurintricarboxylic acid and chrome azurol S, which are the only structures carrying an electron-withdrawing group on the phenolic ring (Fig. 2), are not even oxidized by laccase (as determined by UV-Vis spectrometry). The most suc- cessful mediator is dichlorophenol red, which is the most acidic (with a pK 2 of 5.74) [25] among the sulfonephthalein structures tested. In fact, in Table 1, a correlation is observed between 4-methoxybenzyl alcohol conversion and the pK 2 of the sulfonephthalein mediator. It can be argued that at the reaction pH of 5.0, the lower the pK 2 of the mediator, the easier it is for laccase to oxidize it, as the fraction of sulfonephthalein that is present in its phenolate form (which is expected to be a better one-electron reductant with respect to laccase, and consequently to be more easily converted into the phenoxy radical by electron transfer) is greater. In short, deprotonation of the phenolic moiety seems to be the factor that turns a phenol red-like structure from a somewhat slowly reacting laccase substrate Fig. 6. Cyclic voltammetry scans (m = 5mVÆs )1 ) for 0.5 m M phenol red at pH 7.4 in 0.5 M NaH 2 PO 4 /Na 2 HPO 4 buffer in the presence of 0 m M (m), 2.5 m M (d)or30m M (j) 4-methoxybenzyl alcohol. Fig. 5. The radical mechanism of oxidation of a nonphenolic substrate by a laccase->N–OH mediator system. 3638 F. d’Acunzo and C. Galli (Eur. J. Biochem. 270) Ó FEBS 2003 into an efficient phenolic laccase mediator. Once again, we resorted to electrochemical determinations in order to verify whether there was any pH dependence of the E p of sulfonephthaleins, and, if any, whether it proceeded in the same direction as the pH dependence of the sulfonephtha- lein mediators in Table 1. By increasing the pH (from 4.0 to 7.4) at which the cyclic voltammetry scans were run (see the Materials and methods), we found a decrease of E p from 0.84 to 0.68 V vs. SCE for phenol red (compared with a calculated increase in the phenolate/phenol ratio from 4 · 10 )4 to 1), and from 0.81 to 0.71 V for dichlorophenol red (phenolate/phenol ratio from 2 · 10 )2 to 45). These increments are in keeping with those reported by Li & Hoffman [29] for phenol and 2-chlorophenol, and support the idea that a slow-reacting phenolic mediator, which by virtue of its lower pK is present in solution in a more easily oxidizable form, should be a better laccase mediator. Further support of this idea is provided by the pH dependence of the laccase-mediating effect of phenol red and dichlorophenol red (Table 1). Both sulfonephthaleins are poor mediators at pH 3.5 compared with pH 5.0. On a further increase of the pH solution (to a pH of ‡6) the conversion of 4-methoxybenzyl alcohol again decreases, because P. pinsitus laccase starts to lose its activity. Hence, at % pH 5.0, a compromise is reached between activity of the enzyme and partial (or substantial) deprotonation of the mediator into a more oxidizable species. Conclusions Phenol red is a nonhindered, highly soluble phenol that gives rise to a long-lasting laccase-phenolic mediator system. It is an efficient mediator, compared with other phenolic molecules [7], in that it does not need to be present in large excess in order to obtain significant conversion of laccase-resistant substrates. This favourable feature of phenol red is shared by other sulfonephthalein phenols, whereas other phenols tested, which do not have this structural motif, were rapidly degraded by laccase and failed to mediate. The peculiar behaviour of phenol red (and related structures) may be the result of a combination of several factors, such as (a) its phenoxy radical may be longer lived than that of a simple phenol, because of resonance stabilization, (b) this longer-living phenoxy radical may give dimerization (a reaction generally respon- sible for the instability of phenoxy radicals) less extensively than other reactions, such as H-abstraction from a substrate [this is supported by the observed regeneration of phenol red, following its oxidation at the electrode, by the presence of 4-methoxybenzyl alcohol (cf. Figure 6)] and (c) this phenoxy radical, as suggested by spectrophoto- metric determinations and HPLC data, may, in turn, form secondary species (which we did not attempt to identify) that can also act as mediators; it is possible that the latter phenomenon occurs to an even greater extent with those mediators that need to be used in large excess with respect to their substrate [7,9]. In this study we also showed that other structures based on the phenol red template can mediate laccase catalysis towards recalcitrant substrates, and that a correlation exists between their efficiency and the acidity of their phenolic group. In particular, dichloro- phenol red, the most acidic of the series we tested, was the most efficient at mediating the oxidation of 4-methox- benzyl alcohol. This correlation can be rationalized, based on electrochemical evidence and pH dependence of mediation efficiency, in terms of the larger fraction of the more easily oxidizable phenolate vs. the slowly reacting phenol form of these sulfonephthaleins. Clearly, phenol red is only a modest model of the phenolic subunits of lignin. Nevertheless, the new biogenic hypotheses that ascribe the reported ligninolytic activity of some fungi to the laccase- catalyzed formation of phenoxy radical species [1] from suitable phenolic fragments, begin to receive support here. Acknowledgements Thanks are due to Novo Nordisk Biotech (Denmark) for their generous gift of laccase. We also thank the EU project OXYDELIGN (grant QLK5-CT-1999-01277) for financial support. References 1. ten Have, R. & Teunissen, P.J.M. (2001) Oxidative mechanisms involved in lignin degradation by white-rot fungi. Chem. Rev. 101, 3397–3413. 2. Bo ¨ hmer, S., Messner, K. & Srebotnik, E. (1998) Oxidation of phenanthrene by a fungal laccase in the presence of 1-hydroxy- benzotriazole and unsaturated lipids. Biochem. Biophys. Res. Commun. 244, 233–238. 3. Thurston, C.F. (1994) The structure and function of fungal lac- cases. Microbiology 140, 19–26. 4. Srebotnik, E., Jensen, K.A. Jr & Hammel, K.E. (1994) Fungal degradation of recalcitrant non-phenolic lignin structures without lignin peroxidase. Proc. Natl Acad. Sci. USA 91, 12794–12797. 5. Kapich, A.N., Hofrichter, M., Vares, T. & Hatakka, A. (1999) Coupling of manganese peroxidase-mediated lipid peroxidation with destruction of nonphenolic lignin model compounds and 14C-labeled lignins. Biochem. Biophys. Res. Commun. 259, 212– 219. 6. Itoh, K., Fujita, M., Kumano, K., Suyama, K. & Yamamoto, H. (2000) Phenolic acids affect transformations of chlorophenols by a Coriolus versicolor laccase. Soil Biol. Biochem. 32, 85–91. 7. Johannes, C. & Majcherczyk, A. (2000) Natural mediators in the oxidation of polycyclic aromatic hydrocarbons by laccase medi- ator systems. Appl. Environ. Microbiol. 66, 524–528. 8. Eggert,C.,Temp,U.,Dean,J.F.D.&Eriksson,K E.L.(1996)A fungal metabolite mediates degradation of non-phenolic lignin structures and synthetic lignin by laccase. FEBS Lett. 391, 144–148. 9. Li, K., Xu, F. & Eriksson, K E.L. (1999) Comparison of fungal laccases and redox mediators in oxidation of a nonphenolic lignin model compound. Appl. Environ. Microbiol. 65, 2654–2660. 10. Cantarella, G., Galli, C. & Gentili, P. (2003) Free radical vs. electron-transfer routes of oxidation of hydrocarbons by laccase/ mediator systems. Catalytic or stoichiometric procedures. J. Mol. Catal. B: Enzym. 22, 135–144. 11. Cantarella, G., d’Acunzo, F. & Galli, C. (2003) Determination of laccase activity in mixed solvents. A comparison between two chromogens in a spectrophotometric assay. Biotechnol. Bioeng. 82, 395–398. 12. d’Acunzo, F., Baiocco, P. & Galli, C. (2003) A study of the oxi- dation of ethers with the enzyme laccase under mediation by two N-OH-type compounds. New J. Chem. 27, 329–332. 13. Baiocco, P., Barreca, A.M., Fabbrini, M., Galli, C. & Gentili, P. (2003) Promoting laccase activity towards non-phenolic sub- strates: a mechanistic investigation with some enzyme/mediator systems. Org. Biomol. Chem. 1, 191–197. Ó FEBS 2003 A phenolic mediator for the enzyme laccase (Eur. J. Biochem. 270) 3639 14. Fabbrini, M., Galli, C., Gentili, P. & Macchitella, D. (2001) An oxidation of alcohols by oxygen with the enzyme laccase and mediation by TEMPO. Tetrahedron Lett. 42, 7551–7553. 15. Fabbrini, M., Galli, C. & Gentili, P. (2002) Comparing the cata- lytic efficiency of some mediators of laccase. J. Mol. Cat. B: Enzym. 16, 231–240. 16. Xu, F. (1996) Oxidation of phenols, anilines, and benzenethiols by fungal laccases: correlation between activity and redox potentials as well as halide inhibition. Biochemistry 35, 7608–7614. 17. Wolfenden, B.S. & Willson, R.L. (1982) Radical cations as reference chromogens in kinetic studies of one-electron transfer reactions: pulse radiolysis studies of ABTS. J. Chem. Soc. Perkin Trans. 2, 805–812. 18. d’Acunzo, F., Galli, C. & Masci., B. (2002) Oxidation of phenols by laccase and laccase-mediator systems. Solubility and steric issues. Eur. J. Biochem. 269, 5330–5335. 19. Zhang, M Y., Carpenter, B.K. & McLafferty, F.W. (1991) Gas- phase formation of four isomeric C 4 H 4 + ions. Ionic isomer quantitation with neutralization-reionization mass spectrometry. J. Am. Chem. Soc. 113, 9499–9503. 20. Bumpus, J.A., Tien, M., Wright, D. & Aust., S.D. (1985) Oxida- tion of persistent environmental pollutants by a white rot fungus. Science 228, 1434–1436. 21. d’Acunzo, F., Baiocco, P., Fabbrini, M., Galli, C. & Gentili, P. (2002) The radical rate-determining step in the oxidation of benzyl alcohols by two N-OH-type mediators of laccase: the polar N-oxyl radical intermediate. New J. Chem. 26, 1791–1794. 22. Castro, A.I.R.P., Evtuguin, D.V. & Xavier, A.M.B. (2003) Degradation of biphenyl lignin model compounds by laccase of Trametes versicolor in the presence of HBT and heteropolyanion [SiW 11 VO 40 ] 5– . J. Mol. Catal. B: Enzym. 22, 13–20. 23. Bourbonnais, R., Leech, D. & Paice, M.G. (1998) Electrochemical analysis of the interactions of laccase mediators with lignin model compounds. Biochim. Biophys. Acta 1379, 381–390. 24. Kirste,B.,Grimm,M.&Kurreck,H.(1989)EPR, 1 H, and 13 C ENDOR studies of a quintet-state 13 C-labeled galvinoxyl-type tetraradical. J. Am. Chem. Soc. 111, 108–114. 25. Casula, R., Crisponi, G., Cristiani, F., Nurchi, V., Casu, M. & Lai, A. (1993) Characterization of the ionization and spectral proper- ties of sulfonephthalein indicators. Correlation with substituent effects and structural features. Talanta 40, 1781–1788. 26. Perrin, D.D. (1965) Dissociation Constants of Organic Acids in Aqueous Solution. Butterworths, London. 27. Silverman, R.B. (2000) Reduction and oxidation. In The Organic Chemistry of Enzyme-Catalyzed Reactions, pp. 157–158. Academic Press, San Diego, CA. 28. Wachter, R.M. & Branchaud, B.P. (1996) Molecular modeling studies on oxidation of hexopyranoses by galactose oxidase. An active site topology apparently designed to catalyze radical reactions, either concerted or stepwise. J. Am. Chem. Soc. 118, 2782–2789. 29. Hoffman, M.Z. & Li, C. (1999) One-electron redox potentials of phenols in aqueous solution. J. Phys. Chem. B 103, 6653– 6656. 30. Muramatsu, M. (1958) Studies on the interaction of surface films with solute in solutions. Bull. Chem. Soc. Jpn 31, 864–871. 3640 F. d’Acunzo and C. Galli (Eur. J. Biochem. 270) Ó FEBS 2003 . First evidence of catalytic mediation by phenolic compounds in the laccase-induced oxidation of lignin models Francesca d’Acunzo. model of the phenolic subunits of lignin. Nevertheless, the new biogenic hypotheses that ascribe the reported ligninolytic activity of some fungi to the