Biochimica et Biophysica Acta, 1041 (1990) 129-132
12%
Elsevier BBAPRO33757
Radical Intermediates in veratryl alcohol oxidation by ligninase. NMR evidence Gianfranco Gilardi
1, Patricia J.
Harvey
2, Anthony
E.G. Cass 1 and John M. Palmer 2
1 Centrefor Biotechnology and 2 Department of Pure and Applied Biology, Imperial College of Science, Technologyand Medicine, South Kensington, London (U.K.)
(Received29May 1990)
Key words: NMR; Ligninase;Veratrylalcohol; Ligrtin;(P. chrysosporium) Proton nuclear magnetic resonance (NMR) spectra of veratryl alcohol (3,4-dimethoxybenzyl alcohol) were obtained during its oxidation by ligninase. It was observed that a substantial increase in the iinewidths of the resonances occurred only in the presence of both the enzyme and hydrogen peroxide. Quenching the reaction by the addition of alkali immediately restored the normal linewidths of the resonances. Furthermore, inversion-recovery experiments showed a decrease in the longitudinal relaxation time of the substrate when the enzyme was actively turning over. Changes in both these NMR parameters are consistent with the generation of radical intermediates during the Ugninase-catalysed oxidation of veratryl alcohol.
Introduction The white rot fungus Phanerochaete chrysosporium produces several extra-cellular enzymes during its growth on wood, some of which have been implicated in the oxidative degradation of lignin. One of these, known as ligninase, is a glycoprotein with a molecular mass of 40 kDa that has been shown to catalyse the hydrogen peroxide dependent oxidation of a wide variety of lignin model compounds and is thought to initiate the breakdown of lignin [1-2]. Stopped flow studies of this enzyme suggest that it reacts with hydrogen peroxide to give an intermediate, compound I [3-5]. Reduction of Compound I in a single electron step will yield Compound II and a second one electron reduction will return the enzyme to the resting state [5-7]. Electron paramagnetic resonance (EPR) spectroscopy has identified substrate radicals as intermediates during the oxidation of a wide range of aromatic compounds by ligninase and this is consistent with a catalytic cycle involving two one electron transfers from substrate to enzyme [8-10]. However, with the fungal secondary metabolite, veratryl alcohol, EPR spectroscopy failed to detect radical generation [11]. Tien et al. [11] considered that reduction of Compound II by veratryl alcohol was
Correspondence: A.E.G. Cass, Centre for Biotechnology,Imperial College of Science, Technologyand Medicine, South Kensington, LondonSW7 2AZ, U.K.
tOO slow tO play a significant role in catalysis and they proposed that this substrate was oxidised via a two-electron pathway without the formation of radicals. Moreover, Harvey et al. [5] found that at low pH values, the steady-state enzyme intermediate observed during the oxidation of veratryl alcohol exhibited features more similar to Compound I than to Compound II. Veratryl alcohol has, however, been shown to enhance the oxidation rate of monomethoxylated.aromatic compounds and these effects have been attributed to the known charge transfer properties of radical cations [12-13]. As veratryl alcohol is produced and secreted by the fungus during its growth on wood, Harvey et al. [12] proposed a mechanism whereby the veratryl alcohol radical cation is a freely diffusing mediator transferring oxidizing equivalents from the enzyme to the large, insoluble and hydrophobic lignin. They argued that veratryl alcohol must be oxidised either by a two-step one-electron and a one-step twoelectron pathway of enzyme reduction operating simultaneously or alternatively, a cycle involving a substrate modified Compound II intermediate capable of enhancing the rate of reduction of Compound II to the native state, but still involving the release of substrate radical cations from the enzyme active site [5]. An alternative method to detect free radicals relies on the paramagnetic broadening of nuclear magnetic resonance (NMR) signals by the unpaired electron on the radical.
0167-4838/90/$03.50 © 1990 ElsevierSciencePublishers B.V. BiomedicalDivision)
130 In this paper we present evidence that during the action of ligninase on veratryl alcohol the proton resonances of the latter undergo substantial broadening consistent with the formation of veratryl alcohol radicals. Materials and M e t h o d s
Ligninase was isolated from culture filtrates of P.
chrysosporium as described by Harvey et al. [5] and its activity was determined from the rate of oxidation of veratryl alcohol, followed spectrophotometrically at 310 nm [14]. All solutions were prepared in deuterated water and buffered with 20 mM deuterated phosphate; deuterium peroxide was prepared by a d d i n g sodium peroxide to deuterated water and deuterophosphoric acid likewise from phosphorus pentoxide. The p D of the solution was calculated b y subtracting 0.4 from the p H meter reading. All N M R experiments were performed in nitrogen flushed 5 r a m tubes~ with degassed solutions; the reaction vol. was_0.5 ml and enzyme turnover was initiated by addition of peroxide. Typical solutions were 2 m M veratryl alcohol or 1,4-dimethoxybenzene, 0.5 mM peroxide and 0.02 # M ligninase. ~H-NMR spectra were obtained on a Bruker AC200 spectrometer operating at 200 MHz. A pulse to pulse time of 10.8 s was employed to ensure complete recovery of the magnetization and the residual water was suppressed by a selective presaturation pulse. The effect of the enzyme reaction on the spin lattice relaxation time was determined with an inversion-recovery pulse sequence. All chemical shifts are reported relative to sodium 3-(trimethylsilyl)2,2,3,3-tetradeuteropropionate (TSP). Results and D i s c u s s i o n
The dependence of the activity of ligninase on p H in deuterated water was determined and the optimum is essentially unchanged at 2.75, and therefore all the N M R experiments were carried out at the optimum p H value. Oxidation of both veratryl alcohol and 1,4-dimethoxybenzene by purified ligninase and hydrogen peroxide results in substantial changes in the 1H-NMR spectra of the substrates. In both cases the effect on the spectrum is the same. Resonances broaden and then disappear. If the enzyme reaction is quenched, by the addition of alkali to increase the p H to 10, then the original spectrum is restored. Furthermore, the broaderring requires the presence of both hydrogen peroxide and the enzyme and the extent of broadening depends upon the concentration of the former. Fig. 1 shows the 1H-NMR spectrum of veratryl alcohol at increasing times after initiating the reaction. It
i
. . . .
i
. . . .
~0
i . . . . . . . . . . . . . . . . . . .
~ I0
5 I0
~ll0
3 I0
'
pprn
Fig. I. ~H-NMR spectra foUowin$ the oxidation of veratryl alcohol by
hgninase at pH 2.75. Reaction conditions: 0.09 #M ligninase in 20 mM deuterophosphatebuffer and 2 mM veratryl alcohol, to which 0.5 mM hydrogen peroxide was added to start the reaction. Peak assignments: methoxygroup (3.9 ppm), hydroxymethylgroup (4.6 ppm) and aromatic protons (6.9-7.0 ppm). The first spectrum was acquired before the addition of hydrogen peroxide and subsequent spectra were obtained at 3.7, 5.3, 7, 9.7, 12.3 and 15 min after initiating the reaction.
can be seen that all of the resonances of the substrate are broadened whilst that of the acetone (at 2.2 ppm) is unaffected. This specific broadening is consistent with the effect being due solely to the action of the enzyme on its substrate and does not arise from bulk susceptibility effects. In addition to the quenching of the enzyme, the veratryl alcohol resonances also reappear approx. 10 min after the start of the reaction, although this time is somewhat dependent on the peroxide concentration. When the concentration of hydrogen peroxide is decreased from that of the experiment in Fig. 1 to 0.25 mM, the broadening and loss of intensity of the veratryl alcohol resonances occurs more slowly and to a lesser extent. Conversely, increasing the concentration of hydrogen peroxide to 0.77 mM still results in a loss of intensity, but in this case resonances characteristic of veratraldehyde appear after about 10 min and the
131 9-
veratryl alcohol resonances are still broadened beyond detection. None of these changes is seen if enzyme that has been heated to 100 °C is used. We attribute this broadening and loss of intensity of the resonances of veratryl alcohol to the generation of the veratryl alcohol radical cation during enzyme turnover. The presence of an unpaired electron will result in the rapid relaxation of nearby protons leading to a decrease in the transverse relaxation time (T2) and hence to an increase in the linewidth. Previous EPR measurements performed under similar conditions failed to detect any radicals. In the NMR experiments reported here the broadening and complete loss of intensity of the resonances of veratryl alcohol (VA) can be explained if radicals ( V A ' ) are in redox exchange with a much larger population of neutral molecules:
8-
7-
I
15 4 -
2
o
1'o
do
do
Time (min)
Fig. 2. Width at half height of the methoxyl resonances. Reaction conditions same as in Fig. 1. • • , control conducted without addition of hydrogen peroxide; [] [], with hydrogen peroxide 0.5 m M ; • ll, stopped reaction after 2 min by addition of alkali.
V A + V A +' ~(-~-VA +' + V A
When the reaction was carried out at pH 4 no line broadening was observed, suggesting a much lower steady-state radical concentration and consistent with the reduced enzyme activity at higher pH values. Line broadening at pH 4 could, however, be observed if the hydrogen peroxide and the enzyme were both increased 10-fold. Providing that the exchange rate is rapid compared with the relaxation time of the neutral molecule then the observed relaxation time is given by the following equation [15]: fptpA2/4 1/T 2 =
1 + (fdt2pA2/4) + 2tpTl~ l
where fp and fd are the fractions of paramagnetic and diamagnetic species, respectively, tp is the lifetime of the paramagnetic state, Tie the electron longitudinal relaxation time and A the hyperfine coupling constant. This shows that the few radicals that are present can influence the NMR properties of the neutral form and can thus be detected through this indirect effect. The variation in the linewidth of the methoxy group during the experiment is shown in Fig. 2. As well as causing a decrease in T2, the presence of a free radical in fast exchange with the neutral molecule would also be expected to decrease the longitudinal relaxation time, T1. In order to observe this effect an inversion-recovery experiment was performed using the following pulse sequence: '
~r-~-vr/2-Collect
When ~"= 0.69 T1 then the NMR resonance is a null, at times shorter than this the signal is inverted and at longer times it is positive. When the inversion-recovery experiment is carried out at a • value which yields a null for the methoxy groups in the absence of the
'
'
'
I
'
'
I
7.0
'
'
'
4.0 pore
Fig. 3. I H - N M R spectra showing the decreasing longitudinal relaxa; tion time T1 of aromatic and methoxyl resonances of veratryl alcohol during oxidation by ligninase. A ~r-~-~r/2-Collect pulse sequence was used with ~ = 0.69 T 1 to give a null for the methoxy resonances in the absence of enzyme turnover. The first spectrum was acquired before the addition of hydrogen peroxide and subsequent spectra were obtained at 6 min intervals after initiating the reaction.
132 acid pH results in a much longer lived signal and the effect is still apparent even 40 min after starting the reaction. The NMR spectra of both veratryl alcohol and 1,4dimethoxybenzene obtained during turnover of ligninase show a broadening and loss of intensity of the resonances of these substrates, attributed to the generation of radical products in rapid exchange with the neutral molecule. This interpretation is consistent with previous EPR results which show that radical cations are formed during the ligninase catalysed oxidation of 1,4-dimethoxybenzene and strongly suggest that the same process is occurring with veratryl alcohol. The detection of veratryl alcohol radicals supports the hypothesis that they are acting as mediators in the oxidation of lignin.
Acknowledgement G. Gilardi acknowledges with gratitude Montedison S.p.A. (Italy) for the award of a fellowship. ppm
Fig. 4. 1H-NMR spectra following the oxidation of 1,4-dimethoxybenzene. Reaction conditions are the same as in Fig. 1. The first spectrum was acquired before the addition of hydrogen peroxide and subsequent spectra were obtained at 4, 7, 11 and 25 rain after initiating the reaction.
enzyme then a decrease in T1 during turnover is observed as the signal becoming positive. The results of this experiment are illustrated in Fig. 3 and they show that during enzyme action the methoxy group has a positive intensity, indicating that it is recovering more rapidly than when there is no reaction. That this increased longitudinal relaxation is associated with an enzymatic reaction is shown by the observation that when the reaction stops or is quenched then the methoxy group resonance is again a null. It proved impossible to perform a complete inversion-recovery experiment with a range of z values due to the limited time during which the enzyme reaction could be sustained. Another substrate for the lignin peroxidase is 1,4-dimethoxybenzene and when this compound is added to the enzyme and hydrogen peroxide a free radical signal can be detected by EPR [8]. Consistent both with this and with our interpretation of the veratryl alcohol oxidation, we observe that under conditions of enzyme turnover the characteristic broadening and loss of intensity is seen in the NMR spectrum of 1,4-dimethoxybenzene, Fig. 4. In this case, however, the greater stability of the radical derived from 1,4-dimethoxybenzene at
References 1 Tien, M. and Kirk, T.K. (1983) Science 221, 661-663. 2 Glenn, J.K., Morgan, M.A., Mayfield, M.B., Kuwahara, M. and Gold, M.H. (1983) BB Res. Commun. 114(3), 1077-1083. 3 Andrawis, A., Johnson, K.A. and Tien, M. (1988) J. Biol. Chem. 263, 1195-1198. 4 Marquez, L, Wariishi, H., Dunford, B.H. and Gold, M.H. (1988) J. Biol. Chem. 263, 10549-10552. 5 Harvey, P.J., Palmer, J.M., Schoemaker, H.E., Dekker, H.L. and Wever, K. (1989) Biochim. Biophys. Acta 99, 59-63. 6 Harvey, P.J., Schoemaker, H.E., Bowen, R.M. and Palmer, J.M. (1985) FEBS Lett. 18, 13-16. 7 Renganathan, V. and Gold, M.H. (1986) Biochemistry 25, 16261631. 8 Kersten, P.J., Tien, M., Kalyanaraman, B. and Kirk, T.K. (1985) J. Biol. Chem. 260, 2609-2612. 9 Kersten, P.J., Kalyanaraman, B., Hammel, K.E. and Kirk, T.K. (1987) Lignin Enzymic and Microbial Degradation, Vol. 40, pp. 75-79, INRA, Paris. 10 Hammei, K.E., Kalyanaraman, B. and Kirk, T.K. (1986) Proc. Natl. Acad. Sci. USA 83, 3708-3712. 11 Tien, M., Kirk, T.K., Bull, C. and Fee, J.A. (1986) J. Biol. Chem. 261, 1687-1693. 12 Harvey, P.J., Schoemaker, H.E. and Palmer, J.M. (1986) FEBS Lett. 195, 242-246. 13 Harvey, P.J., Schoemaker, H.E. and Palmer, J.M. (1987) in Lignin Enzymic and Microbial Degradation, Vol. 40, pp. 145-150, INRA, Paris. 14 Tien, M. and Kirk, T.K. (1984) Proc. Natl. Acad. Sci. USA 81, 2280-2284. 15 Kreilck, R.W. (1973) in N.M.R. of Paramagnetic Molecules (LaMar, G.N., Horrocks, W.J. and Holm, L.H., eds.), pp. 595-641, Academic Press, New York.
12%
Elsevier BBAPRO33757
Radical Intermediates in veratryl alcohol oxidation by ligninase. NMR evidence Gianfranco Gilardi
1, Patricia J.
Harvey
2, Anthony
E.G. Cass 1 and John M. Palmer 2
1 Centrefor Biotechnology and 2 Department of Pure and Applied Biology, Imperial College of Science, Technologyand Medicine, South Kensington, London (U.K.)
(Received29May 1990)
Key words: NMR; Ligninase;Veratrylalcohol; Ligrtin;(P. chrysosporium) Proton nuclear magnetic resonance (NMR) spectra of veratryl alcohol (3,4-dimethoxybenzyl alcohol) were obtained during its oxidation by ligninase. It was observed that a substantial increase in the iinewidths of the resonances occurred only in the presence of both the enzyme and hydrogen peroxide. Quenching the reaction by the addition of alkali immediately restored the normal linewidths of the resonances. Furthermore, inversion-recovery experiments showed a decrease in the longitudinal relaxation time of the substrate when the enzyme was actively turning over. Changes in both these NMR parameters are consistent with the generation of radical intermediates during the Ugninase-catalysed oxidation of veratryl alcohol.
Introduction The white rot fungus Phanerochaete chrysosporium produces several extra-cellular enzymes during its growth on wood, some of which have been implicated in the oxidative degradation of lignin. One of these, known as ligninase, is a glycoprotein with a molecular mass of 40 kDa that has been shown to catalyse the hydrogen peroxide dependent oxidation of a wide variety of lignin model compounds and is thought to initiate the breakdown of lignin [1-2]. Stopped flow studies of this enzyme suggest that it reacts with hydrogen peroxide to give an intermediate, compound I [3-5]. Reduction of Compound I in a single electron step will yield Compound II and a second one electron reduction will return the enzyme to the resting state [5-7]. Electron paramagnetic resonance (EPR) spectroscopy has identified substrate radicals as intermediates during the oxidation of a wide range of aromatic compounds by ligninase and this is consistent with a catalytic cycle involving two one electron transfers from substrate to enzyme [8-10]. However, with the fungal secondary metabolite, veratryl alcohol, EPR spectroscopy failed to detect radical generation [11]. Tien et al. [11] considered that reduction of Compound II by veratryl alcohol was
Correspondence: A.E.G. Cass, Centre for Biotechnology,Imperial College of Science, Technologyand Medicine, South Kensington, LondonSW7 2AZ, U.K.
tOO slow tO play a significant role in catalysis and they proposed that this substrate was oxidised via a two-electron pathway without the formation of radicals. Moreover, Harvey et al. [5] found that at low pH values, the steady-state enzyme intermediate observed during the oxidation of veratryl alcohol exhibited features more similar to Compound I than to Compound II. Veratryl alcohol has, however, been shown to enhance the oxidation rate of monomethoxylated.aromatic compounds and these effects have been attributed to the known charge transfer properties of radical cations [12-13]. As veratryl alcohol is produced and secreted by the fungus during its growth on wood, Harvey et al. [12] proposed a mechanism whereby the veratryl alcohol radical cation is a freely diffusing mediator transferring oxidizing equivalents from the enzyme to the large, insoluble and hydrophobic lignin. They argued that veratryl alcohol must be oxidised either by a two-step one-electron and a one-step twoelectron pathway of enzyme reduction operating simultaneously or alternatively, a cycle involving a substrate modified Compound II intermediate capable of enhancing the rate of reduction of Compound II to the native state, but still involving the release of substrate radical cations from the enzyme active site [5]. An alternative method to detect free radicals relies on the paramagnetic broadening of nuclear magnetic resonance (NMR) signals by the unpaired electron on the radical.
0167-4838/90/$03.50 © 1990 ElsevierSciencePublishers B.V. BiomedicalDivision)
130 In this paper we present evidence that during the action of ligninase on veratryl alcohol the proton resonances of the latter undergo substantial broadening consistent with the formation of veratryl alcohol radicals. Materials and M e t h o d s
Ligninase was isolated from culture filtrates of P.
chrysosporium as described by Harvey et al. [5] and its activity was determined from the rate of oxidation of veratryl alcohol, followed spectrophotometrically at 310 nm [14]. All solutions were prepared in deuterated water and buffered with 20 mM deuterated phosphate; deuterium peroxide was prepared by a d d i n g sodium peroxide to deuterated water and deuterophosphoric acid likewise from phosphorus pentoxide. The p D of the solution was calculated b y subtracting 0.4 from the p H meter reading. All N M R experiments were performed in nitrogen flushed 5 r a m tubes~ with degassed solutions; the reaction vol. was_0.5 ml and enzyme turnover was initiated by addition of peroxide. Typical solutions were 2 m M veratryl alcohol or 1,4-dimethoxybenzene, 0.5 mM peroxide and 0.02 # M ligninase. ~H-NMR spectra were obtained on a Bruker AC200 spectrometer operating at 200 MHz. A pulse to pulse time of 10.8 s was employed to ensure complete recovery of the magnetization and the residual water was suppressed by a selective presaturation pulse. The effect of the enzyme reaction on the spin lattice relaxation time was determined with an inversion-recovery pulse sequence. All chemical shifts are reported relative to sodium 3-(trimethylsilyl)2,2,3,3-tetradeuteropropionate (TSP). Results and D i s c u s s i o n
The dependence of the activity of ligninase on p H in deuterated water was determined and the optimum is essentially unchanged at 2.75, and therefore all the N M R experiments were carried out at the optimum p H value. Oxidation of both veratryl alcohol and 1,4-dimethoxybenzene by purified ligninase and hydrogen peroxide results in substantial changes in the 1H-NMR spectra of the substrates. In both cases the effect on the spectrum is the same. Resonances broaden and then disappear. If the enzyme reaction is quenched, by the addition of alkali to increase the p H to 10, then the original spectrum is restored. Furthermore, the broaderring requires the presence of both hydrogen peroxide and the enzyme and the extent of broadening depends upon the concentration of the former. Fig. 1 shows the 1H-NMR spectrum of veratryl alcohol at increasing times after initiating the reaction. It
i
. . . .
i
. . . .
~0
i . . . . . . . . . . . . . . . . . . .
~ I0
5 I0
~ll0
3 I0
'
pprn
Fig. I. ~H-NMR spectra foUowin$ the oxidation of veratryl alcohol by
hgninase at pH 2.75. Reaction conditions: 0.09 #M ligninase in 20 mM deuterophosphatebuffer and 2 mM veratryl alcohol, to which 0.5 mM hydrogen peroxide was added to start the reaction. Peak assignments: methoxygroup (3.9 ppm), hydroxymethylgroup (4.6 ppm) and aromatic protons (6.9-7.0 ppm). The first spectrum was acquired before the addition of hydrogen peroxide and subsequent spectra were obtained at 3.7, 5.3, 7, 9.7, 12.3 and 15 min after initiating the reaction.
can be seen that all of the resonances of the substrate are broadened whilst that of the acetone (at 2.2 ppm) is unaffected. This specific broadening is consistent with the effect being due solely to the action of the enzyme on its substrate and does not arise from bulk susceptibility effects. In addition to the quenching of the enzyme, the veratryl alcohol resonances also reappear approx. 10 min after the start of the reaction, although this time is somewhat dependent on the peroxide concentration. When the concentration of hydrogen peroxide is decreased from that of the experiment in Fig. 1 to 0.25 mM, the broadening and loss of intensity of the veratryl alcohol resonances occurs more slowly and to a lesser extent. Conversely, increasing the concentration of hydrogen peroxide to 0.77 mM still results in a loss of intensity, but in this case resonances characteristic of veratraldehyde appear after about 10 min and the
131 9-
veratryl alcohol resonances are still broadened beyond detection. None of these changes is seen if enzyme that has been heated to 100 °C is used. We attribute this broadening and loss of intensity of the resonances of veratryl alcohol to the generation of the veratryl alcohol radical cation during enzyme turnover. The presence of an unpaired electron will result in the rapid relaxation of nearby protons leading to a decrease in the transverse relaxation time (T2) and hence to an increase in the linewidth. Previous EPR measurements performed under similar conditions failed to detect any radicals. In the NMR experiments reported here the broadening and complete loss of intensity of the resonances of veratryl alcohol (VA) can be explained if radicals ( V A ' ) are in redox exchange with a much larger population of neutral molecules:
8-
7-
I
15 4 -
2
o
1'o
do
do
Time (min)
Fig. 2. Width at half height of the methoxyl resonances. Reaction conditions same as in Fig. 1. • • , control conducted without addition of hydrogen peroxide; [] [], with hydrogen peroxide 0.5 m M ; • ll, stopped reaction after 2 min by addition of alkali.
V A + V A +' ~(-~-VA +' + V A
When the reaction was carried out at pH 4 no line broadening was observed, suggesting a much lower steady-state radical concentration and consistent with the reduced enzyme activity at higher pH values. Line broadening at pH 4 could, however, be observed if the hydrogen peroxide and the enzyme were both increased 10-fold. Providing that the exchange rate is rapid compared with the relaxation time of the neutral molecule then the observed relaxation time is given by the following equation [15]: fptpA2/4 1/T 2 =
1 + (fdt2pA2/4) + 2tpTl~ l
where fp and fd are the fractions of paramagnetic and diamagnetic species, respectively, tp is the lifetime of the paramagnetic state, Tie the electron longitudinal relaxation time and A the hyperfine coupling constant. This shows that the few radicals that are present can influence the NMR properties of the neutral form and can thus be detected through this indirect effect. The variation in the linewidth of the methoxy group during the experiment is shown in Fig. 2. As well as causing a decrease in T2, the presence of a free radical in fast exchange with the neutral molecule would also be expected to decrease the longitudinal relaxation time, T1. In order to observe this effect an inversion-recovery experiment was performed using the following pulse sequence: '
~r-~-vr/2-Collect
When ~"= 0.69 T1 then the NMR resonance is a null, at times shorter than this the signal is inverted and at longer times it is positive. When the inversion-recovery experiment is carried out at a • value which yields a null for the methoxy groups in the absence of the
'
'
'
I
'
'
I
7.0
'
'
'
4.0 pore
Fig. 3. I H - N M R spectra showing the decreasing longitudinal relaxa; tion time T1 of aromatic and methoxyl resonances of veratryl alcohol during oxidation by ligninase. A ~r-~-~r/2-Collect pulse sequence was used with ~ = 0.69 T 1 to give a null for the methoxy resonances in the absence of enzyme turnover. The first spectrum was acquired before the addition of hydrogen peroxide and subsequent spectra were obtained at 6 min intervals after initiating the reaction.
132 acid pH results in a much longer lived signal and the effect is still apparent even 40 min after starting the reaction. The NMR spectra of both veratryl alcohol and 1,4dimethoxybenzene obtained during turnover of ligninase show a broadening and loss of intensity of the resonances of these substrates, attributed to the generation of radical products in rapid exchange with the neutral molecule. This interpretation is consistent with previous EPR results which show that radical cations are formed during the ligninase catalysed oxidation of 1,4-dimethoxybenzene and strongly suggest that the same process is occurring with veratryl alcohol. The detection of veratryl alcohol radicals supports the hypothesis that they are acting as mediators in the oxidation of lignin.
Acknowledgement G. Gilardi acknowledges with gratitude Montedison S.p.A. (Italy) for the award of a fellowship. ppm
Fig. 4. 1H-NMR spectra following the oxidation of 1,4-dimethoxybenzene. Reaction conditions are the same as in Fig. 1. The first spectrum was acquired before the addition of hydrogen peroxide and subsequent spectra were obtained at 4, 7, 11 and 25 rain after initiating the reaction.
enzyme then a decrease in T1 during turnover is observed as the signal becoming positive. The results of this experiment are illustrated in Fig. 3 and they show that during enzyme action the methoxy group has a positive intensity, indicating that it is recovering more rapidly than when there is no reaction. That this increased longitudinal relaxation is associated with an enzymatic reaction is shown by the observation that when the reaction stops or is quenched then the methoxy group resonance is again a null. It proved impossible to perform a complete inversion-recovery experiment with a range of z values due to the limited time during which the enzyme reaction could be sustained. Another substrate for the lignin peroxidase is 1,4-dimethoxybenzene and when this compound is added to the enzyme and hydrogen peroxide a free radical signal can be detected by EPR [8]. Consistent both with this and with our interpretation of the veratryl alcohol oxidation, we observe that under conditions of enzyme turnover the characteristic broadening and loss of intensity is seen in the NMR spectrum of 1,4-dimethoxybenzene, Fig. 4. In this case, however, the greater stability of the radical derived from 1,4-dimethoxybenzene at
References 1 Tien, M. and Kirk, T.K. (1983) Science 221, 661-663. 2 Glenn, J.K., Morgan, M.A., Mayfield, M.B., Kuwahara, M. and Gold, M.H. (1983) BB Res. Commun. 114(3), 1077-1083. 3 Andrawis, A., Johnson, K.A. and Tien, M. (1988) J. Biol. Chem. 263, 1195-1198. 4 Marquez, L, Wariishi, H., Dunford, B.H. and Gold, M.H. (1988) J. Biol. Chem. 263, 10549-10552. 5 Harvey, P.J., Palmer, J.M., Schoemaker, H.E., Dekker, H.L. and Wever, K. (1989) Biochim. Biophys. Acta 99, 59-63. 6 Harvey, P.J., Schoemaker, H.E., Bowen, R.M. and Palmer, J.M. (1985) FEBS Lett. 18, 13-16. 7 Renganathan, V. and Gold, M.H. (1986) Biochemistry 25, 16261631. 8 Kersten, P.J., Tien, M., Kalyanaraman, B. and Kirk, T.K. (1985) J. Biol. Chem. 260, 2609-2612. 9 Kersten, P.J., Kalyanaraman, B., Hammel, K.E. and Kirk, T.K. (1987) Lignin Enzymic and Microbial Degradation, Vol. 40, pp. 75-79, INRA, Paris. 10 Hammei, K.E., Kalyanaraman, B. and Kirk, T.K. (1986) Proc. Natl. Acad. Sci. USA 83, 3708-3712. 11 Tien, M., Kirk, T.K., Bull, C. and Fee, J.A. (1986) J. Biol. Chem. 261, 1687-1693. 12 Harvey, P.J., Schoemaker, H.E. and Palmer, J.M. (1986) FEBS Lett. 195, 242-246. 13 Harvey, P.J., Schoemaker, H.E. and Palmer, J.M. (1987) in Lignin Enzymic and Microbial Degradation, Vol. 40, pp. 145-150, INRA, Paris. 14 Tien, M. and Kirk, T.K. (1984) Proc. Natl. Acad. Sci. USA 81, 2280-2284. 15 Kreilck, R.W. (1973) in N.M.R. of Paramagnetic Molecules (LaMar, G.N., Horrocks, W.J. and Holm, L.H., eds.), pp. 595-641, Academic Press, New York.
