Spectroscopic and Kinetic Properties of the Oxidized Intermediates of Lignin Peroxidase from Phanerochaete chrysosporium"

Stopped-flow rapid scan techniques were used to obtain a spectrum of nearly homogeneous lignin peroxidase compound I (LiPI) under pseudo-first order conditions at the unusually low pH optimum (3.0) for the enzyme. The Lip1 spectrum had a Soret band at 407 nm with -60% reduced intensity and a visible maximum at 650 nm. Under steady-state conditions a Soret spectrum for lignin peroxidase compound I1 (LiPII) was also obtained. The Soret maximum of LiPII at 420 nm was only -15% reduced in intensity compared to native Lip. Transient state kinetic results confirmed the pH independence of Lip1 formation over the pH range 3.06-7.39. The rate constant was (6.5 2 0.2) X 10' M" s-I. Addition of excess veratryl alcohol to Lip1 resulted in its reduction to LiPII with subsequent reduction of LiPII to the native enzyme. Reactions of Lip1 and LiPII with veratryl alcohol exhibited marked pH dependencies. For the Lip1 reaction the rate constants ranged from 2.5 X 1 0 ' ~ " 8" at pH 3.06 to 4.1 X lo3 M" s-' at pH 7.39; for the LiPII reaction, 1.6 X lo6 M" s-' (pH 3.06) to 2.3 X los M" s-' (pH 5.16).

Under secondary metabolic conditions the wood-rotting fungus Phanerochaete chrysosporium secretes two extracellular heme peroxidases which are involved in the degradation of lignin. These enzymes, manganese peroxidase and lignin *This work was supported by Operating Grant A1248 from the Natural Sciences and Engineering Research Council of Canada (to H. B. D.) and by Grant DMB 8607279 from the National Science Foundation and Grant DE-FG06-86ER 13550 from the United States Department of Energy (to M. H. G.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
§ To whom correspondence should be addressed Oregon Graduate Center, Dept. of Chemical & Biological Sciences, 19600 N. W. Von Neumann Dr., Beaverton, OR 97006-1999. peroxidase (Lip),' have been purified and characterized (1)(2)(3)(4)(5). Lip, a glycoprotein with a molecular weight of 41,000 and a single iron protoporphyrin IX prosthetic group ( 3 4 , catalyzes the HzOz-dependent oxidation of lignin model compounds via the initial formation of a substrate aryl cation radical with subsequent nonenzymatic reactions to yield the final products (6-9). Our initial characterization of the oxidized intermediates LiPI, LiPII, and LiPIII has been reported (10). Herein we have utilized rapid-scan spectrophotometry to determine the spectra of the oxidized intermediates of Lip generated both at the pH optimum (3.0) and at pH 6.0 under pseudo-first order conditions with Hz02 in excess. We have also investigated the pH dependence of Lip1 formation and the rate of reduction of Lip1 and LiPII using VA as the reducing substrate.

EXPERIMENTAL PROCEDURES
The major isozyme of Lip was purified from cultures of P. chrysosporium as previously described (3,5). The purified protein was homogeneous and had an RZ value of -5.0. Enzyme concentrations were determined at 407.6 nm using a molar absorptivity of 133 mM" cm-' (3,5). The enzyme was dialyzed exhaustively against triply distilled water before use (10). The concentration of Hz02 (BDH Chemicals) was determined as reported (11).
Rapid scan spectra were recorded with a Union Giken RA601 Rapid Reaction Analyzer equipped with a 1-cm cell. The absorption spectra were measured by means of a multichannel photodiode array and memorized in a digital computer system (Sord M200 Mark 111). The analogue replica was plotted on an X-Y recorder. Spectral regions of 96 nm were scanned from 360 to 700 nm.
Kinetic measurements were conducted using the Union Giken RA601 in the stopped-flow mode. One of the drive syringes contained native Lip or Lip1 in deionized water while the other syringe contained the substrate (HZ02 or VA) and buffer. All experiments were performed at (25.0 & 0.5) "C in 0.1 M sodium citrate buffer <pH 4.5 and in 0.1 M sodium phosphate >pH 4.5.
Lip1 formation was followed at 407.6 nm using a final enzyme concentration of 1.5 p M and various concentrations of excess Hz02  to maintain pseudo-first order conditions. The reaction was observed from pH 3.06 to 7.39. LiPII formation and decay were followed at 426 nm. Lip1 (1.5 p~) was freshly prepared for each experiment. Various concentrations of VA or p-cresol were buffered at the pH indicated.

Spectroscopic Characterization of Oxidized Intermediates
Formation of Compound I-The pH optimum for Lip is -3.0 (5, 12, 13); thus, where possible, spectral measurements were made at this pH. Measurements were also made at pH 6.0 where Lip1 and LiPII are more stable (10). The spectrum of native Lip has maxima at 407, 496, and 632 nm (3, 5 , 10) ( Fig. 1). Within 0.2 s after addition of excess HzOz (30 eq) to the native enzyme at pH 3.06, the Soret peak height was reduced by -60%, and a new peak appeared at 650 nm, suggesting the formation of Lip1 (10). Isosbestic points between native Lip and Lip1 occur at 426 and 540 nm.
Formation of Compounds ZI and 111-Upon addition of 30 eq of HzOZ to the native enzyme at pH 3.06, a sequence of spectral changes was recorded over a 10-s time span. During the first 0.2 s, Lip1 formation was observed ( quently the Soret maximum red-shifted to 419 nm and new peaks appeared in the visible region at 545 and 579 nm indicating the formation of compound I11 (Fig. 2 A ) (10). A spectrum for LiPII could not be observed under these conditions. In contrast, within 2 s after the addition of 30 eq of H202 to the native enzyme at pH 6.21, the transient formation of LiPII (visible maxima at 525 and 555 nm) (10) was observed (Fig. 2B). Within 10 s, LiPII was converted to LiPIII (maxima at 419, 543, and 579 nm) (10).
Compound 11 Formation in the Steady State-In our previous work (10) LiPII was prepared by adding 2 eq of H202 to the native enzyme at pH 6.0. In order to prepare LiPII under pseudo-first order conditions with excess Hz02, steady-state conditions were applied. If the rate constant for the reduction of Lip1 ( k z ) is larger than that for the reduction of LiPII (k3) Wavelength ( and if sufficient Hz02 is present so that Lip1 formation is not rate-limiting, then the accumulation of LiPII can be expected in the steady state. With horseradish peroxidase, k~ is -10 times k3 (14).
Rapid scan spectra of Lip were recorded in the steady state in the presence of either p-cresol or VA at pH 6.21. When the concentrations of H202, p-cresol, and Lip were 60,15, and 1.5 p~, respectively, a spectrum for the Soret region of LiPII was obtained (Fig. 3). The Soret of LiPII had -85% of the molar absorptivity of the native enzyme. A steady-state spectrum for LiPII could also be obtained at pH 3.0 with VA as the reducing substrate (data not shown).

Kinetic Experiments
Compound I Formation-All kinetic traces were of a single exponential character. The observed rate constants were linearly proportional to the H202 concentration in the range used. The formation of Lip1 exhibited no pH dependence from pH 3.06 to 7.39 (data not shown). The mean rate constant was (6.5 * 0.2) X IO6 M" s -' .
Compound 11 Formation and Conversion to the Native Enzyme-The formation of LiPII and its subsequent conversion to the native enzyme in the presence of VA was followed at 426 nm, the isosbestic point between native Lip and LiPI. At this wavelength the absorptivity of LiPII is >2 times that of native Lip and Lip1 (10). Fig. 4 shows a typical time course trace of the changes in absorbance. The biphasic curve shows an initial increase in absorbance (formation of LiPII) which is completed in less than 20 ms (shown on the expanded scale) followed by a decline (reduction of LiPII). The traces were exponential in character when followed for sufficiently short intervals. The rate of formation of LiPII was followed at different pH values ranging from 3.06 to 7.39. Fig. 5A shows the pH dependence of the reaction of Lip1 with VA, The rate decreased dramatically with increasing pH, from 2.5 X lo6 M" s" at pH 3.06 to 4.1 X lo3 M" s-l at pH 7.39. This result is consistent with published findings (12, 13) that the activity of Lip increases to a maximum near pH 2.5 and approaches zero at pH values above 6.
The second portion of the biphasic curve (Fig. 4) corresponds to the slower one-electron reduction of LiPII to the native enzyme. Fig. 5B shows the pH dependence of this reduction. Again, a striking decrease in rate constant is ob-  Fig. 4. At each pH the VA concentration was varied from 20 to 200 PM. Sampling period ranged from 20 ms to 3 s. B, pH dependence of native Lip formation from LiPII. Experimental conditions as in Fig. 4. Sampling period ranged from 300 ms to 30 s. served as the pH is increased from 3.06 (1.6 X lo5 M" s-') to 5.16 (2.3 X lo3 M" S -' ) .

DISCUSSION
Recently, we were able to prepare the oxidized intermediates of the enzyme, LiPI, LiPII, and LiPIII, at pH 6.0 in the presence of stoichiometric amounts of Hz02 (10). Each of these intermediates has spectral characteristics similar to those of the corresponding intermediates of horseradish peroxidase (14). These results suggest that the catalytic cycle of Lip is similar to that of other peroxidases as shown in Reactions 1-3.

(2)
LiPII + AH + Native Lip + A + H20 However, the pH optimum of the enzyme is unusually low for a peroxidase (pH 3.0) (5, 12, 13), and the K,,, of the enzyme for HzOZ is -30 pM (5,12). Therefore, in order to study the oxidized intermediates of Lip at the pH optimum and under pseudo-first order conditions with Hz02 in excess, stoppedflow rapid scan techniques were utilized.
The spectrum of Lip1 obtained at pH 3.06 with 30 eq of H202 (Fig. 1) is very similar to that obtained previously at pH 6.0 (10) and to those obtained for horseradish peroxidase (14) and indicates that this Lip1 preparation is >95% pure. This is the first spectrum of a reasonably pure Lip1 at the pH optimum. A recently published spectrum of LiPI, limited to the Soret region, clearly shows contamination by native Lip (Soret intensity -82% of native Lip) rather than, as the authors claim, an unusually high Soret intensity (15). The reduced absorption of the Soret region of Lip1 compared to native Lip and the characteristic absorption band at 650 nm suggests that, like horseradish peroxidase I (14, 16, 17), Lip1 contains 2 oxidizing eq over the native enzyme (10). The first oxidizing equivalent apparently is contained in the ferry1 state of the iron (14) as recently confirmed by our resonance Raman experiments (18); the second equivalent is contained as a porphyrin x-cation radical (16, 17).
Our kinetic results (data not shown) for Lip1 formation demonstrate no pH dependence over a pH range of 3.06-7.39, confirming previous results (15). Formation of compound I for other peroxidases is pH-dependent, and the distal ionizable group has a low pK value: for horseradish peroxidase and chloroperoxidase, pK I 3.0 (19, 20) and for yeast cytochrome c peroxidase, pK 5 4.5 (21). Presumably this ionizable group plays a role in the heterolytic cleavage of H202. The unique lack of pH dependence for Lip1 formation suggests that such an ionization may not occur in this system. This may account for the decreased rate of Lip1 formation compared to that found for other peroxidases (19)(20)(21).
Formation of LiPII and LiPIII-At pH 3.0 in the presence of excess H202 and the absence of a reducing substrate, Lip is rapidly oxidized to LiPIII. Within 2 s after addition of 30 eq of H202 to native Lip at pH 3.0, a spectrum with characteristic peaks at 545 and 579 nm is observed (Fig. 2). Apparently, at this pH in the presence of 30 eq of H202, the halflife of LiPII is too short to obtain its spectrum on this time scale. Even at pH 6.0, the addition of only 25 eq of H202 to native Lip results in its conversion to LiPIII (10) whereas 250 eq of Hz02 are required to convert native horseradish peroxidase to horseradish peroxidase 111 at that pH. Our observation of the rapid formation of LiPIII at pH 3.0 is in contrast to a recent report (15) which claims the detection of LiPII 30 s after addition of 20 eq of H202 to native Lip at pH 3.5. However, in that report only the Soret region was scanned and in that region LiPII and LiPIII cannot be readily differentiated.
Evidence for the transient formation of LiPII in the presence of excess H202 at pH 6.0 is shown in Fig. 2B. Possibly, at low pH in the absence of another reducing substrate, HzOz is readily oxidized by LiPII to form HO; and native Lip. HO; may then complex with the FeIII in native Lip to form the Fe3+O; structure of LiPIII (10, 14, 22). The formation of horseradish peroxidase I11 from horseradish peroxidase I1 and H2O2 has been reported (14,22).
Compound 1 1 Formation in the Steady State-In order to prepare a relatively pure preparation of LiPII under pseudofirst order conditions with H202 in excess, steady-state conditions were applied. Under these steady-state conditions a Soret spectrum for LiPII was obtained (Fig. 3). The molar absorptivity of the Soret of LiPII was 20% higher than the value we obtained previously (10). These results suggest that, as for horseradish peroxidase (19, 20), Lip1 has a higher reactivity than LiPII for the reducing substrate.
Reduction of LiPI and LiPII-Our earlier work suggested that the oxidation of VA by Lip1 proceeds via two singleelectron steps (10) with the intermediate formation of a VA cation radical (6-8) rather than by a single two-electron step (12). The results in Fig. 4 show this directly. The initial conversion of Lip1 to LiPII takes place in 20 ms. This is followed by a slower conversion of LiPII to the native enzyme over 500 ms. Thus, at pH 3.0 in the presence of excess reducing substrate (VA), the normal catalytic cycle is completed via two single-electron steps as shown in Reactions 1-3. This result also demonstrates that LiPII is capable of efficiently oxidizing VA.
The pH dependence of the conversion of Lip1 to LiPII (kt) and LiPII to native Lip ( k 3 ) is shown in Fig. 5. These results demonstrate directly for the first time that the rate of these conversions is strongly dependent on pH with optimum activity at a low pH. Apparently, it is the pH dependence of the reduction of Lip1 and LiPII rather than the formation of LiPI which dictates the unusually low pH optimum for this enzyme (12, 13). These results suggest that an ionizable group(s) with a pK < 5.0 controls either the binding and/or oxidation of the reducing substrate by these oxidized enzyme intermediates. Preliminary results (data not shown) indicate that the binding of VA to the native enzyme has a similar dependence. The low pH optimum of the enzyme may facilitate the formation and stabilization of the aryl cation radical (6-8). Experiments designed to determine what ionizable group(s) on the enzyme dictate this pH dependence are planned.