(Received for publication, May 21, 1996, and in revised form, October 10, 1996)
From Novo Nordisk Biotech, Davis, California 95616
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 O2 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 O2 to H2O. 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 mechanistic factors from laccase, phenolic substrate, and O2 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 O2 and type 2/type 3 (T2/T3)1 copper sites) on the rate of laccase-catalyzed O2 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.
Chemicals used as buffers and substrates were
purchased from Aldrich (except for methyl syringate, which was from
Acros) with the highest available grade. Recombinant Polyporus
pinsitus (or Trametes villosa) laccase
isoform-1 (PpL), Rhizoctonia solani laccase isoform-4 (RsL),
and Myceliophthora thermophila laccase (MtL) were purified
as reported previously (11, 14, 15). Fe(2,2-dipyridyl)2Cl3 and
Fe(2,2
-dipyridyl)2Cl2 were made by mixing
FeCl3 or FeCl2 with two molar equivalent
2,2
-dipyridyl, respectively. The Britton and Robinson (B&R) buffers
were made by mixing 0.1 M boric acid, 0.1 M
acetic acid, 0.1 M phosphoric acid with 0.5 M
NaOH to the desired pH.
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 (Eo) of the T1 copper in laccase was
measured as reported previously (11), except that B&R buffer was used.
Briefly, the Eo(PpL) was measured with 21 µM
PpL, 0.2 mM Fe(bipyridyl)2Cl2, and 0.1-0.4 mM Fe(bipyridyl)2Cl3
(Eo[Fe(2,2-dipyridyl)2Cl3/Fe(2,2
-dipyridyl)2Cl2] = 0.780 V); the Eo(RsL) was measured with 17 µM RsL, 0.2 mM
Fe(bipyridyl)2Cl2, and 0.05-0.2 mM
Fe(bipyridyl)2Cl3; and the Eo(MtL)
was measured with 0.14 mM MtL, 23 mM
K3Fe(CN)6, and 0-200 mM
K4Fe(CN)6
(Eo[K3Fe(CN)6/K4Fe(CN)6] = 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 pKa value of the phenolic substrate studied was
determined spectrally in B&R buffer. At alkaline pH, the phenolic substrates showed a deprotonation-induced spectral change. For syringaldehyde (4-hydroxy-3,5-dimethoxybenzaldehyde), the absorption maximum at 302 nm was replaced by two maxima at 250 and 365 nm (isosbestic points: 238, 266, and 327 nm); for acetosyringone (4-hydroxy-3
,5
-dimethoxyacetophenone), the absorption maximum at 284 nm was replaced by two maxima at 250 and 359 nm (isosbestic points:
237, 263, and 321 nm); and for methyl syringate (methyl 4-hydroxy-3,5-dimethoxybenzoate), the absorption maximum at 273 nm was
replaced by one shoulder around 236 nm and one maximum at 322 nm
(isosbestic points: 231, 251, and 293 nm). The pH profiles of the
absorption at 302 and 365 nm yielded a pKa of 7.0 ± 0.2 for syringaldehyde, a pKa of
7.8 ± 0.2 for acetosyringone, and a pKa of
8.7 ± 0.2 for methyl syringate. Under the same conditions, a
pKa values of 8.2 and 2.2 were previously found for
syringaldazine (4-hydroxy-3,5-dimethoxybenzaldehyde azine) and
2,2
-azinobis-(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS),
respectively (11).
The laccase-catalyzed oxidation of
syringaldehyde, acetosyringone, or methyl syringate was monitored by
O2 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 O2 electrode with 0.04-15 µM 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 = 36 mM
1
cm
1) or on a Molecular Devices Thermomax microplate
reader with a 96-well plate (Costar, tissue culture grade) (for
K4Fe(CN)6;
at 405 nm = 0.90 mM
1 cm
1). The initial
absorbance changes were used to calculate the initial reaction rate
v. The apparent kinetic parameter Km was determined by fitting initial reaction rate (v) and
substrate concentration to v = Vmax× [substrate]/(Km + [substrate]) with the Prizm program of GraphPad (San Diego, CA), and
the apparent kcat was determined from
kcat = Vmax/[laccase].
All the experiments were carried out in air-saturated solutions. Thus
the initial [O2] was kept constant (and was assumed to be
the same as in plain water (0.28 mM)), and the observed
Km and kcat were
"apparent" values ([O2] needs to be systematically
varied in order to measure the true Km and
kcat).
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 µM 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 I50, 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 Ki.
The pH-induced changes in the Eo 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 Eo decreased ~90 mV from pH
2.7 to 5 but increased ~20 mV from pH 5 to 11; for RsL, the
Eo decreased ~80 mV from pH 2.7 to 7 but increased ~30
mV from pH 7 to 11; and for PpL, the Eo did not change from
pH 2.7 to 7 but increased ~30 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
pKa 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, K4Fe(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 Km and
kcat for the nonphenolic substrates are shown in
Fig. 3 (A-D). With both PpL and MtL, the
profiles of the apparent kcat for ABTS and
K4Fe(CN)6 had a monotonic declining nature when
the pH changed from acidic range to alkaline range (Fig. 3,
B and D). With PpL,
K4Fe(CN)6 showed an apparent
Km that did not change much when the pH changed from
2.7 to 7, whereas ABTS showed an apparent Km 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,
K4Fe(CN)6 showed an apparent
Km 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 Km 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 K4Fe(CN)6 or ABTS was so
slow above pH 7 or pH 8 that no accurate Km and
kcat could be obtained.
The pH profiles of the apparent Km and
kcat for the phenolic substrates are presented
in Fig. 4 (A-D). Being oxidized by PpL, all
these substrates showed the highest apparent
kcat at pH 5 (Fig. 4B). With PpL, the
apparent Km 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 Km 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
Km for methyl syringate and acetosyringone showed minimal change at pH 2.7 to 9, but the apparent Km
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 kcat at pH 7, although an
increase in the apparent kcat 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 Km and
kcat could be obtained. Similar result was
previously observed with syringaldazine (11).
The addition of NaF resulted in an immediate laccase inhibition with
I50 0.1 mM at acidic pH. Because HF has a
pKa of 3.5, the laccase inhibition shown in Fig.
5 was most likely caused by F
. As the pH
increased, the I50 became larger, indicating a weaker F
inhibition at higher pH (Fig. 5). This increase of
I50 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).
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 O2 to H2O 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, O2, 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 Eo of a phenol decreases when
pH increases. At a rate of E/
pH = 0.059 V at 25 °C, a pH
change from 2.7 to 11 would result in an Eo(phenol)
decrease of 0.49 V. However, over the same pH range, the Eo
changes for the laccases studied were much smaller (
0.1 V), similar
to the case of the Rhus laccase (17). Such different pH
dependences of the Eo for phenolic substrate and laccase
would then result in a larger difference in redox potential
[
Eo = Eo(laccase, T1)
Eo(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) = 7.1 ×
Eo + 7.0 observed (at pH 5)
for a wide variety of substrates and laccases (16), this
Eo 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 kcat from
pH 4 to 2.7 for MtL-catalyzed phenol oxidation (Fig. 4D)
could also be related to the
Eo effect, becasue, as
shown in Fig. 1A, the Eo(MtL) decreased 80 mV
when pH changed from 2.7 to 4, thus reducing the oxidation potency of
MtL.
The loss of F inhibition at high pH did not result in
recovery or increase of laccase activity. Likely the observed pH
dependence of the F
inhibition was mainly due to an
OH
competition (with F
) for inhibiting MtL
and PpL, similar to the cases of other laccases in which
OH
and F
are shown to competitively bind to
the T2/T3 center and inhibit activity (1, 3, 19, 20, 21, 22). Such
OH
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
inhibition and
Eo contribute, respectively, at alkaline
and acidic pH mainly. The oxidation of ABTS (to the stable, preferred
cation radical) or K4Fe(CN)6 (to
K3Fe(CN)6) does not involve protons, and thus
possesses an Eo independent of pH (17). This would make any
pH effect of
Eo minimal for these two substrates. The
contribution of the OH
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 K4Fe(CN)6), suggesting the
contribution of a common mechanism involving the OH
inhibition of laccase.
The laccases studied showed different, spectrally detected acid-base
transitions (on their T1 center) with a pKa 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 pKa (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 Km 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 pKa or the apparent
Km of the laccases and substrates. As shown in Figs.
3, 4, and 6, the effects of Eo and OH
were
most significant on the apparent kcat or
kcat/Km, indicating that it
is the electron transfer kinetics rather than the substrate binding (as
reflected by Km) that plays a more important role in
determining the pH activity profile. Protic equilibria related to
O2 (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 O2 reduction potential on pH could also
impact the pH activity profile. At 20 °C, the
Eo(O2/H2O) 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 Eo(T1) in PpL and MtL, respectively. Because the
Eo(T2/T3) would generally be equal to or higher than the
Eo(T1) in laccase (23, 24), this pH-induced change of
Eo(O2/H2O) 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
inhibition.
The difference between the monotonic pH activity profile of
a nonphenolic substrate (such as ABTS,
K4Fe(CN)6, or Fe(EDTA)2)
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
Eo and
OH
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 Eo quite insensitive to pH change (
0.1 V over pH
3-10) (17) and is inhibited by OH
at alkaline pH (1, 3,
19, 20, 21, 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
Eo and OH
inhibition discussed above.
Under single-turnover conditions, the effect of the OH
inhibition (excised at the T2/T3 center) would become insignificant because both the reduction of O2 (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
Eo should result in a higher substrate oxidation rate at
more alkaline pH.
In summary, this study demonstrated that both the OH
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
K4Fe(CN)6) whose oxidation does not involve
protons and has a minimal Eo 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
inhibition on
the T2/T3 center. For a reducing substrate (such as syringaldehyde),
whose oxidation involves protons and has a significant Eo
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
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 Km and
kcat 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
[O2] is also systematically varied.
I thank Drs. Alan V. Klotz, Alexander M. Blinkovsky, Mark S. Madden, Stephen H. Brown, and Kim M. Brown of Novo Nordisk Biotech for critical reading and helpful suggestions.