Effects of redox potential and hydroxide inhibition on the pH activity profile of fungal laccases.

The electronic absorption spectrum, susceptibility to fluoride inhibition, redox potential, and substrate turnover of several fungal laccases have been explored as a function of pH. The laccases showed a single spectrally detectable acid-base transition at pH 6-9 and a fluoride inhibition that diminished by increased pH (indicating a competition with hydroxide inhibition). Relatively small changes in the redox potentials (≤0.1 V) of laccase were observed over the pH 2.7-11. Under the catalysis of laccase, the apparent oxidation rates (kcat and kcat/Km) of two nonphenolic substrates, potassium ferrocyanide and 2,2′-azinobis-(3-ethylbenzthiazoline-6-sulfonic acid),decreased monotonically as the pH increased. In contrast, the apparent oxidation rates (kcat and kcat/Km) of three 2,6-dimethoxyphenols (whose pKa values range from 7.0 to 8.7) exhibited bell-shaped pH profiles whose maxima were distinct for each laccase but independent of the substrate. By correlating these pH dependences, it is proposed that the balance of two opposing effects, one generated by the redox potential difference between a reducing substrate and the type 1 copper of laccase (which correlates to the electron transfer rate and is favored for a phenolic substrate by higher pH) and another generated by the binding of a hydroxide anion to the type 2/type 3 coppers of laccase (which inhibits the activity at higher pH), contributes to the pH activity profile of the fungal laccases.

Laccases (EC 1.10.3.2) are a family of multi-copper oxidases that catalyze the oxidation of a range of inorganic and aromatic substances (particularly phenols) with the concomitant reduction of O 2 to water (for recent reviews see Refs. [1][2][3][4][5][6][7][8]. In the past decades, significant progress has been made in elucidating the structure of the copper sites, the catalysis sequence, and the mechanism that governs the catalytic reduction of O 2 to H 2 O. Laccase is receiving increased attention as a model system for characterizing the structure-function relationship of copper-containing proteins because of its potential biotechnological application in the fields such as delignification, plant fiber derivatization, textile dye or stain bleaching, and contaminated water or soil detoxification (4).
One of the most important characteristics of laccase enzymology with phenolic substrates is the pH dependence. In general, the phenol oxidase activity of laccase has a bell-shaped (bi-phasic) pH profile whose optimal pH varies considerably among different laccases (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11). Various structural and mech-anistic factors from laccase, phenolic substrate, and O 2 may contribute to the pH activity profile. Detailed studies have been carried out with respect to the effect of protic equilibrium (related to the laccase-bound O 2 and type 2/type 3 (T2/T3) 1 copper sites) on the rate of laccase-catalyzed O 2 reduction (12,13). However, insight on other potential factors, such as those related to the type 1 (T1) copper and (reducing) substrate, is still limited, and a comprehensive understanding on the mechanism that governs the bi-phasic pH dependence has not been fully established. To address this problem, I investigated several fungal laccases for pH-induced changes in their electronic absorption spectrum, fluoride inhibition, redox potential, and oxidation rate of phenolic and nonphenolic substrates. The study showed that the redox potential difference between a phenolic substrate and the T1 copper of laccase could result in an increased substrate oxidation rate at higher pH, whereas the hydroxide anion binding to the T2/T3 coppers could lead to an inhibition of laccase activity at higher pH. The balance of these two opposing effects might play an important role in determining the pH activity profile of laccase.
Spectrophotometrical Redox Titration of Laccase-The spectrum of laccase was recorded in B&R buffer, pH 2.7-11, on a Shimadzu UV160U spectrophotometer with an 1-cm quartz cuvette. The redox potential (E o ) of the T1 copper in laccase was measured as reported previously (11) 6 ] ϭ 0.433 V). Under various potentials of the solution poised by various concentration ratio of the redox titrant couples, the absorbance changes of laccase in the range of 550 -800 nm were monitored, and the concentrations of the copper(II) and copper(I) states were calculated after the spectral change reached an equilibrium. Anaerobicity was achieved by repetitive evacuating and argon flushing of the reaction chamber at 4°C. * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Laccase Activity Assays-The laccase-catalyzed oxidation of syringaldehyde, acetosyringone, or methyl syringate was monitored by O 2 consumption as described in Ref. 16, and the oxidation of syringaldazine or ABTS was determined photometrically as described in Ref. 11. Briefly, the phenol oxidation was monitored by a Hansatech DW1/AD O 2 electrode with 0.04 -15 M laccases in 0.3-0.5 ml of B&R buffer at 20°C. After the voltage reading stabilized, laccase was added into the solution to initiate the reaction, and the initial output voltage changes were used to calculate the initial reaction rate (v). The oxidation of the nonphenolic substrates were photometrically monitored on either a Shimadzu UV160U spectrophotometer with a 1-cm quartz cuvette (for ABTS; ⌬⑀ at 418 nm ϭ 36 mM Ϫ1 cm Ϫ1 ) or on a Molecular Devices Thermomax microplate reader with a 96-well plate (Costar, tissue culture grade) (for K 4 Fe(CN) 6 ; ⌬⑀ at 405 nm ϭ 0.90 mM Ϫ1 cm Ϫ1 ). The initial absorbance changes were used to calculate the initial reaction rate v. The apparent kinetic parameter K m was determined by fitting initial reaction rate (v) and substrate concentration to v ϭ V max ϫ [substrate]/(K m ϩ [substrate]) with the Prizm program of GraphPad (San Diego, CA), and the apparent k cat was determined from k cat ϭ V max /[laccase]. All the experiments were carried out in air-saturated solutions. Thus the initial [O 2 ] was kept constant (and was assumed to be the same as in plain water (0.28 mM)), and the observed K m and k cat were "apparent" values ([O 2 ] needs to be systematically varied in order to measure the true K m and k cat ).
Fluoride Inhibition-The inhibition of laccase by NaF was assayed with laccase-catalyzed ABTS oxidation in B&R buffer. The assay solutions contained 2 mM ABTS and 0.6 M MtL or 40 nM PpL. Being a complex "linear mixed type," the inhibition showed convex type correlations when the slope and the y intercept of the Lineweaver-Burk plots (1/rate versus 1/[substrate]) were plotted against [NaF], similar to that observed with the Rhus laccase (12). The inhibition was quantitated by the parameter I 50 , the NaF concentration at which only 50% of the initial laccase activity remained, because the complexity of the plots complicated the extraction of the inhibition constant K i .

RESULTS
The pH-induced changes in the E o and the wavelength corresponding to the maximal absorption of the blue band of laccase are shown in Fig. 1 (A and B). At the pH range of 2.7-11, one spectrally detected acid-base transition occurred around pH 6.5 for MtL, pH 8 for RsL, and pH 9 for PpL (Fig.  1B). For MtL, the E o decreased ϳ90 mV from pH 2.7 to 5 but increased ϳ20 mV from pH 5 to 11; for RsL, the E o decreased ϳ80 mV from pH 2.7 to 7 but increased ϳ30 mV from pH 7 to 11; and for PpL, the E o did not change from pH 2.7 to 7 but increased ϳ30 mV from pH 7 to 11. Under the experimental conditions, no obvious correlation between the two types of pH profile in Fig. 1 (A and B) was observed. Fig. 2 (A-D) shows the pH dependence of the initial oxidation rate for four phenolic substrates and two nonphenolic substrates at a given concentration. With PpL, all four phenols, whose pK a range from 7.0 to 8.7, had bell-shaped pH profiles from pH 2.7 to 10 with the same pH optimum of 5. With MtL, all four phenols also had bell-shaped pH profiles from pH 4 to 10 with the same pH optimum of 7. In contrast, the two nonphenolic substrates, K 4 Fe(CN) 6 and ABTS, only showed monotonic pH profiles (with both PpL and MtL) in which the rate decreased as the pH increased.
The pH profiles of the apparent K m and k cat for the nonphenolic substrates are shown in Fig. 3 (A-D). With both PpL and MtL, the profiles of the apparent k cat for ABTS and K 4 Fe(CN) 6 had a monotonic declining nature when the pH changed from acidic range to alkaline range (Fig. 3, B and D). With PpL, K 4 Fe(CN) 6 showed an apparent K m that did not change much when the pH changed from 2.7 to 7, whereas ABTS showed an apparent K m that did not change much from pH 2.7 to pH 7 but increased approximately 4-fold from pH 7 to pH 8 (Fig. 3A). With MtL, K 4 Fe(CN) 6 showed an apparent K m that increased about 3-fold when the pH changed from 2.7 to 4 but did not change much from pH 4 to pH 7, whereas ABTS showed an apparent K m that increased 4-fold from pH 2.7 to pH 6 but 40-fold from pH 6 to pH 8 (Fig. 3C). The catalyzed oxidation of K 4 Fe(CN) 6 or ABTS was so slow above pH 7 or pH 8 that no accurate K m and k cat could be obtained.
The pH profiles of the apparent K m and k cat for the phenolic substrates are presented in Fig. 4 (A-D). Being oxidized by PpL, all these substrates showed the highest apparent k cat at pH 5 (Fig. 4B). With PpL, the apparent K m for syringaldehyde and acetosyringone did not change much from pH 4 to 7 but increased about 10-fold from pH 4 to 2.7, whereas the apparent K m for methyl syringate did not change much from pH 2.7 to 6 but decreased about 10-fold from pH 6 to 7 (Fig. 4A). With MtL, the apparent K m for methyl syringate and acetosyringone showed minimal change at pH 2.7 to 9, but the apparent K m for syringaldehyde increased about 3-fold from pH 6 to 2.7 (Fig.  4C). Being oxidized by MtL, all the phenols showed the highest apparent k cat at pH 7, although an increase in the apparent k cat was also observed when the pH changed from 4 to 2.7 (Fig. 4D). The catalyzed oxidation of all three phenols was so slow above pH 7 with PpL or pH 9 with MtL that no accurate K m and k cat could be obtained. Similar result was previously observed with syringaldazine (11).
The addition of NaF resulted in an immediate laccase inhibition with I 50 Յ 0.1 mM at acidic pH. Because HF has a pK a of 3.5, the laccase inhibition shown in Fig. 5 was most likely caused by F Ϫ . As the pH increased, the I 50 became larger, indicating a weaker F Ϫ inhibition at higher pH (Fig. 5). This increase of I 50 at higher pH did not correspond to an increase in laccase activity, however, because in the absence of NaF, the activity of laccase diminished as pH increased (Fig. 2, C and D). DISCUSSIONS Based on a wide range of physical and chemical characterizations, it is generally accepted that the catalysis of fungal laccase involves (a) the binding of a reducing substrate to the T1 pocket and subsequent reduction of the T1-Cu(II) to Cu(I), (b) the internal electron transfer from the T1 to the T2/T3 center, and (c) the binding and subsequent reduction of an O 2 to H 2 O at the T2/T3 center (1)(2)(3)(4)(5)(6)(7)(8). Potentially, any pH-induced structural or mechanistic changes in either the reducing substrate, O 2 , or laccase (particularly on its T1 and T2/T3 centers) could contribute to the observed pH activity profiles.
The oxidation of phenol by laccase depends on the redox potential difference between the phenol and the T1 copper (16). Due to the oxidative proton release, the E o of a phenol decreases when pH increases. At a rate of ⌬E/⌬pH ϭ 0.059 V at 25°C, a pH change from 2.7 to 11 would result in an E o (phenol) decrease of 0.49 V. However, over the same pH range, the E o changes for the laccases studied were much smaller (Յ0.1 V), similar to the case of the Rhus laccase (17). Such different pH dependences of the E o for phenolic substrate and laccase would then result in a larger difference in redox potential [⌬E o ϭ E o (laccase, T1) Ϫ E o (substrate, single electron)] or driving force (for the electron transfer from phenol to T1 copper) at higher pH (Fig. 6, A and C). Given the correlation of log(rate) ϭ 7.1 ϫ ⌬E o ϩ 7.0 observed (at pH 5) for a wide variety of substrates and laccases (16), this ⌬E o effect should lead to a pH dependence in which the activity increases as the pH increases, thus contributing to the ascending part of the bell-shaped pH activity profile for phenols shown in Figs. 2, 4, and 6. The increase in k cat from pH 4 to 2.7 for MtL-catalyzed phenol oxidation (Fig.  4D) could also be related to the ⌬E o effect, becasue, as shown in Fig. 1A, the E o (MtL) decreased 80 mV when pH changed from 2.7 to 4, thus reducing the oxidation potency of MtL.
The loss of F Ϫ inhibition at high pH did not result in recovery or increase of laccase activity. Likely the observed pH dependence of the F Ϫ inhibition was mainly due to an OH Ϫ competition (with F Ϫ ) for inhibiting MtL and PpL, similar to the cases of other laccases in which OH Ϫ and F Ϫ are shown to competitively bind to the T2/T3 center and inhibit activity (1, 3, 19 -22). Such OH Ϫ inhibition interrupts the internal electron transfer from the T1 to the T2/T3 centers in laccase and, together with other rate-diminishing deprotonations (to be discussed later), could contribute to the descending part of the pH activity profile of phenols shown in Figs. 2, 4, and 6.
For phenol substrates, a bell-shaped pH activity profile with an optimal pH dependent on laccase (not substrate) is consistent with the mechanism in which the opposing effects of the OH Ϫ inhibition and ⌬E o contribute, respectively, at alkaline and acidic pH mainly. The oxidation of ABTS (to the stable, preferred cation radical) or K 4 Fe(CN) 6 (to K 3 Fe(CN) 6 ) does not involve protons, and thus possesses an E o independent of pH (17). This would make any pH effect of ⌬E o minimal for these two substrates. The contribution of the OH Ϫ inhibition would then result in a monotonic pH activity profile consistent with the data presented in Figs. 2, C and D, and 3, C and D. As shown in Fig. 6, B and D, the descending part of the pH profile for the phenols (at neutral alkaline pH range) is similar to the profile for the nonphenolic substrates (ABTS and K 4 Fe(CN) 6 ), suggesting the contribution of a common mechanism involving the OH Ϫ inhibition of laccase.
The laccases studied showed different, spectrally detected acid-base transitions (on their T1 center) with a pK a ranging from 6.5 to 9. The pH-induced change in the blue absorption band around 600 nm was probably caused by an energy perturbation (on the T1 Cu(II)) related to protic equilibria of nearby amino acid residue(s). The three phenol substrates studied also have different acid-base transitions because of their different pK a (ranging from 7 to 8.7). These protic equilibria related to the reducing substrate (or its oxidized intermediate product) and the T1 pocket could affect substrate docking (e. g. making K m larger or smaller) or activation (e. g. transforming phenol to phenoxide or vice versa). However, such protic events seem to be small in this study because the optimal pH appeared to be independent of either the pK a or the apparent K m of the laccases and substrates. As shown in Figs. 3, 4, and 6, the effects of ⌬E o and OH Ϫ were most significant on the apparent k cat or k cat /K m , indicating that it is the electron transfer kinetics rather than the substrate binding (as reflected by K m ) that plays a more important role in determining the pH activity profile. Protic equilibria related to O 2 (or its reduced intermediate product) and the T2/T3 pocket could diminish the overall catalytic rate at high pH (8,12,13). However, such an effect also seems to be small in this study, because a dominant effect of this type would result in a pH profile with a maximum independent of the reducing substrate.
The dependence of O 2 reduction potential on pH could also impact the pH activity profile. At 20°C, the E o (O 2 /H 2 O) can drop from 1.23 V at pH 0 to 0.82 V at pH 7 or 0.58 V at pH 11 and thus become comparable with the E o (T1) in PpL and MtL, respectively. Because the E o (T2/T3) would generally be equal to or higher than the E o (T1) in laccase (23,24), this pH-induced change of E o (O 2 /H 2 O) would lead to a decrease in reaction rate at high pH (due to unfavorable thermodynamics). The descending part of the observed pH profiles might be the result of the combination of the effect and that of the OH Ϫ inhibition.
The difference between the monotonic pH activity profile of a nonphenolic substrate (such as ABTS, K 4 Fe(CN) 6 , or Fe(EDTA) 2Ϫ ) and the bi-phasic (or bell-shaped) pH activity profile of a phenolic substrate (such as syringaldehyde) reported for other laccases (11,(25)(26)(27)(28) could also be interpreted by the ⌬E o and OH Ϫ effects discussed above. Another observation that could be attributed to these effects involves the different pH dependence of activity observed under steady-state and single-turnover (or anaerobic) conditions. It was reported that under steady-state conditions, the rate of the Rhus laccase-catalyzed hydroquinone oxidation has a bell-shaped pH dependence (29), whereas under single-turnover conditions, the rate only increases when pH increases (12,19,20). Like PpL and MtL, the Rhus laccase has an E o quite insensitive to pH change (Յ0.1 V over pH 3-10) (17) and is inhibited by OH Ϫ at alkaline pH (1, 3, 19 -22). Thus the bell-shaped pH activity profile of the Rhus laccase under steady-state conditions could be attributed (at least partly) to the opposing effects of ⌬E o and OH Ϫ inhibition discussed above. Under single-turnover conditions, the effect of the OH Ϫ inhibition (excised at the T2/T3 center) would become insignificant because both the reduction of O 2 (at the T2/T3 center), and the internal electron transfer (from the T1 to the T2/T3 center) in laccase would contribute little to the oxidation of hydroquinone (at the T1 center). In consequence, the predominant effect of ⌬E o should result in a higher substrate oxidation rate at more alkaline pH.
In summary, this study demonstrated that both the OH Ϫ inhibition at the T2/T3 center and the redox potential difference between a reducing substrate and the T1 center could affect the pH activity profile of a laccase. For a reducing substrate (such as K 4 Fe(CN) 6 ) whose oxidation does not involve protons and has a minimal E o dependence on pH, the activity of a laccase could decline monotonically when the pH increases, as the result of the possible involvement of an OH Ϫ inhibition on the T2/T3 center. For a reducing substrate (such as syringaldehyde), whose oxidation involves protons and has a significant E o dependence on pH, the pH activity profile of a laccase could be bi-phasic, reflecting possible combinatory contribution from the opposing effects of the pH-induced redox potential change (on both the T1 center and substrate) and the OH Ϫ inhibition. It should be pointed out that laccase is a two-substrate enzyme and to obtain the true kinetic parameters, both substrates should be subjected to concentration variation. The K m and k cat reported above were observed with various reducing substrate concentrations in air-saturated solutions only and hence are apparent values. Thus the hypothesis proposed above should be further tested by a full kinetic analysis based on experiments in which [O 2 ] is also systematically varied.