The Mechanism of Indole-S-acetic Acid Oxidation by Horseradish Peroxidases*

The oxygen-consuming oxidation of indole-d-acetic acid (IAA) occurred much faster in the presence of horseradish peroxidase C (neutral isoenzyme) than in the presence of horseradish peroxidase A (acidic iso- enzyme). An intermediate oxidation product of IAA was found to be a hydroperoxide species that reacted with the ferric enzymes to form Compound I at second order rate constants of 6.8 X lo3 M-’ s-l for peroxidase A and 2.0 x lo6 Mm1 s-l for peroxidase C at pH 4.4. The hydroperoxide concentration reached about one-half of the initial IAA concentration at the end of the oxygen-consuming reaction and then decreased slowly. The main intermediate of the enzymes observed during the oxygen-consuming reaction was Compound II, which oxidized IAA to its free radical at rate constants of 1.5 X lo3 Mm1 s-l for peroxidase A and 1.2 X lo4 Me1 se1 for peroxidase C at pH 4.4. The results supported the mech- anism that the oxygen consumption occurs mainly ‘ through the reaction of with the free rad- ical formed from the peroxidatic oxidation of The ferric enzymes were not reduced

The oxygen-consuming oxidation of indole-d-acetic acid (IAA) occurred much faster in the presence of horseradish peroxidase C (neutral isoenzyme) than in the presence of horseradish peroxidase A (acidic isoenzyme).
An intermediate oxidation product of IAA was found to be a hydroperoxide species that reacted with the ferric enzymes to form Compound I at second order rate constants of 6.8 X lo3 M-' s-l for peroxidase A and 2.0 x lo6 Mm1 s-l for peroxidase C at pH 4.4. The hydroperoxide concentration reached about one-half of the initial IAA concentration at the end of the oxygenconsuming reaction and then decreased slowly. The main intermediate of the enzymes observed during the oxygen-consuming reaction was Compound II, which oxidized IAA to its free radical at rate constants of 1.5 X lo3 Mm1 s-l for peroxidase A and 1.2 X lo4 Me1 se1 for peroxidase C at pH 4.4. The results supported the mechanism that the oxygen consumption occurs mainly ' through the reaction of oxygen with the IAA free radical formed from the peroxidatic oxidation of IAA. The ferric enzymes were not reduced by IAA under strict anaerobic conditions in the presence of carbon monoxide but were reduced upon addition of a small amount of oxygen or hydrogen peroxide to the systems. The results suggested that the ferric enzyme is reduced by the IAA free radical but not by IAA itself. From a comparison of reactivities of oxyperoxidase and Compound II we concluded that the catalytic cycle of ferrous and oxyperoxidases is not involved in the IAA oxidase reaction.
Since Kenten (1) reported in 1955 that horseradish peroxidase catalyzes aerobic oxidation of the plant hormone indole-3-acetic acid (IAA)' in the absence of added hydrogen peroxide, considerable progress has been made toward understanding the mechanism of the reaction. The earlier &dies have been reviewed by Ray (2) and Galston and Purves (3). Various investigators (4-10) have suggested that the free radical derived from IAA' is an intermediate in the IAA degradation. Since the reaction is not inhibited by a catalytic amount of catalase (1, 7, ll), hydrogen peroxide may not be involved in the catalytic function of peroxidase. Therefore, it has been suggested (10-14) that the IAA oxidase activity of peroxidase is attributed to its oxygenase-like function where the oxygen binding to the enzyme is an essential process. On the other hand, we have suggested that a major function of peroxidase in the peroxidase-oxidase reaction is to produce free radicals of substrates through the peroxidase cycle consisting of the * 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. I The' abbreviation used is: IAA, indole-3-acetic acid.
In this paper, we present evidence to indicate that the peroxidase cycle functions during the IAA oxidase reaction and that direct oxygen activation by peroxidase is not involved in the reaction.

EXPERIMENTAL PROCEDURES
Peroxidases A and C were purified from wild horseradish roots by the method of Shannon et al. (16)  period in the reaction of peroxidase A was shortened to about one-tenth when the enzyme concentration was increased from 1 pM to 10 pM. It has been reported that the lag period is elongated by addition of a small amount of ascorbate (9), ferulic acid (14), or a metabolite of the insecticide carbofuran (20,21).
The second addition of IAA to the reaction solution containing peroxidase A caused rapid consumption of oxygen and the lag disappeared ( Fig. 2A). If the addition of IAA was preceded by that of ascorbate there appeared a lag again, the duration of which was very sensitive to the amount of ascorbate. It should be noted that the rate at steady state and the total amount of consumed oxygen were independent of the amount of ascorbate. The lag period was plotted against the amount of ascorbate in Fig. 2B, which shows that the lag period increased markedly above a critical concentration of ascorbate.
The critical concentration became clear when a similar experiment was conducted with peroxidase C. Fig. 3 shows that a lag began to appear sharply by addition of ascorbate above 44 PM. The inset of Fig. 3 shows that such increase in the lag period was observed at a much lower concentration of ascorbate when the reaction was started in a fresh solution. The results suggested that a hydroperoxide species was formed during the IAA oxidase reaction and served as a substrate in the peroxidase reaction. Many workers (6-10) have suggested without direct evidence that a hydroperoxide derived from IAA is an intermediate product in the IAA degradation.
It can be seen from Fig. 3 that the hydroperoxide was not decomposed by a catalytic amount of catalase.
The stoichiometric relationship in the reaction between the hydroperoxide and ascorbate is shown in Fig. 4. Inset A of Fig. 4 shows the formation of Compound I from the reaction of peroxidase C with the hydroperoxide.
A kinetic trace measured with a stopped flow apparatus in the presence of a limited amount of the hydroperoxide is shown in inset B of Its concentration was assumed to be 44 pM (see Fig. 3) and was scaled in the abscissa. Inset A shows rapid scan spectra on the Union Giken instrument after two solutions of 4.4 pM peroxidase C with 44 pM hydroperoxide were mixed. The final concentrations were reduced by one-half. Spectrum was scanned from 355 nm at indicated times in 5 ms. Inset B shows a kinetic trace in the stopped flow titration experiment. Two solutions of 20 pM peroxidase C and 2 pM hydroperoxide + 200 pM IAA, pH 7.5, were mixed. At this pH, Compound I was stable in a measured time even in the presence of IAA or ascorbate.
The titration experiments were carried out in the absence (0) and presence (0)  Cl hydroperoxide concentration at any time during the oxidase reaction, except for the very early stage of the reaction.
As reported already by others (6, 22-24), spectral changes occurred at least in two steps ( Japanese radish peroxidases used for the reaction. Fig. 50 the lag time uersus ascorbate concentration plots as described in Fig. shows that the products varied with the concentration of 3. To measure the hydroperoxide concentration in the reaction by peroxidase C but not with the rate of the reaction (Compare peroxidase A, 1 pM peroxidase C was added with ascorbate so as to respectively. The result led to a conclusion that the intermediate spectrum is ascribable to the hydroperoxide that reacts with the ferric enzyme to form Compound I. Ricard and Job (10) argued that the direct reduction of peroxidase by IAA is necessary for the oxidase reaction to proceed and also that most of Compound II is formed from a two-equivalent reduction of oxyperoxidase. Formation of Compound II and oxyperoxidase in the reaction of peroxidase C and IAA was investigated with a Union Giken rapid reaction analyzer. Experiments shown in Fig. 7 were performed in the presence of bovine serum albumin which inhibited degradation of peroxidase heme but had little effect on the rate of oxygen consumption.
By a rapid scanning method, the formation of Compound II was clearly demonstrated in the oxidase reaction (Fig. 7A). No significant amount of oxyperoxidase was formed in the early stage of the reaction. The concentrations of Compound II and oxyperoxidase can be measured from absorbance changes in the Soret band if the enzyme exists in a mixture of ferric, Compound II, and oxyforms. Wavelengths of 461.7 and 452.7 were isosbestic between ferric and oxyenzymes and between Compound II and ferric enzymes and were used for assays of Compound II and oxyperoxidase, respectively.
At these wavelengths, absorbance was not changed by the formation of a complex of peroxidase C and IAA. The formation of such a complex was observed spectrophotometrically by Ricard and Job (10). Spectrophotometric analyses of absorbance changes at those wavelengths indicated that Compound II was formed rapidly after a short lag and oxyperoxidase accumulated slowly and later than Compound II (Fig. 7B). The oxyperoxidase formation became evident as the pH decreased and the IAA concentration increased.
An important problem in the mechanism of IAA oxidase reaction might be to identify a chemical species that reduces the ferric enzyme and to clarify the role of its reduction in the overall oxidase reaction. Ricard and Nari (25,26) reported that peroxidase is reduced directly by IAA. In our experiments, however, neither peroxidase A nor C was reduced by IAA under strict anaerobic conditions in the presence of OL --I 550 600 Time ( carbon monoxide. Fig. 8A shows that the reduction of the enzymes occurred upon addition of a small amount of oxygen. Like the oxidase reaction shown in Fig. 1, the reduction was fast in peroxidase C while it occurred slowly after a distinct lag in peroxidase A. Fig. 8B shows the reduction of ferriperoxidase A caused by an addition of hydrogen peroxide. It should be noted that ascorbate inhibited the oxygen-induced reduction but not the hydrogen peroxide-induced reduction. The strict anaerobiosis in these experiments was obtained with using an enzyme system of ascorbate oxidase or glucose oxidase plus catalase. In an anaerobic solution obtained with commercial carbon monoxide without using these enzyme systems, an appreciable amount of ferriperoxidases was reduced simply by an addition of IAA. It was also found that hydroquinone, NADH, &hydroxyindole-3-acetate, and chlorogenic acid inhibited both the oxygen-induced formation of carbon monoxide complex and the IAA oxidase reaction, just as ascorbate did. The results do not conform with the mechanism of Ricard and Job (lo), who concluded that the inhibition by ascorbate is ascribable to its specific binding to peroxidase. Marklund et al. (27) reported that there is marked difference between peroxidases A and C in the rate of reactions of the enzymes with organic hydroperoxides or peracids (   10. Dependence of the ratio of the rate of oxygen consumption to that of IAA free radical formation upon the enzyme concentration. The ratio was calculated from the results in the inset. The rate of IAA free radical formation was assumed to equal 2 k&Compound II] [IAA]. The concentration of Compound II was measured as described in Fig. 7B. Diverse features of the IAA oxidase reaction are probably ascribable to the fact that free radical mechanism is involved in the reaction. We have proposed a general mechanism for the peroxidase-oxidase reaction (15,31) and a peculiar pattern of the NADH oxidase reaction catalyzed by peroxidase C has been simulated with a computer from nine rate equations (32). Contrary to the NADH oxidase reaction, the IAA oxidase reaction is insensitive to superoxide dismutase (9,12). Hydrogen peroxide accumulates in the NADH oxidase reaction (32) 9. Effect of pH on the kl and ks values. 50 ITIM potassium acetate (pH 4 to 5.3) and 50 InM potassium phosphate (pH 6 to 7). A, The second order rate constant (kl) for the reaction of peroxidase A with the hydroperoxide derived from IAA. 0.1 pM enzyme, 7.8 pM hydroperoxide. The rate was measured from similar experiments shown in Inset B of Fig. 4. B, the ha value was measured from the overall kinetics by the method described elsewhere (5). The rate of IAA free radical formation was assumed to equal that of the iron reduction. 0.15 pM peroxidase A (0) or 0.13 to 0.26 pM peroxidase C (a), 10 PM hydrogen peroxide, 50 pM IAA, 100 pM FeC13, 300 PM ophenanthroline under anaerobic conditions. Peroxidase + R'OOH k,(2 X 10G M-' SC') + Compound I + R'OH fast Compound I + RH -Compound II + R.
Compound II + RH where RH and R. denote IAA and its free radical. The rate of free radical formation was assumed to correspond to that of reduction of o-phenanthroline . ferric ion complex used as an electron acceptor for the free radical (5). The rate constants were increased with the decrease in pH as was observed in reduction of Compound II by anionic electron donors (30). At pH 4.4 the rate constant was measured at 1.5 X lo3 Mm' s-l for peroxidase A and 1.2 x lo* M-' s-l for peroxidase C.
It has been suggested (7-10) that the oxygen consumption in the IAA oxidase reaction is attributed to the incorporation of oxygen into a free radical of IAA; R. + O2 + ROO . . In order to check this point, the ratio of the rate of oxygen consumption to the rate of free radical formation through the peroxidase cycle was calculated from data shown in the inset of Fig. 10. Fig. 10 shows that the ratio depended on the concentration of peroxidase C. The ratio might be closely correlated with propagation and termination of chain reactions. At low concentrations of the enzyme the ratio became higher than 2, indicating an involvement of a chain propagating reaction: ROO . + RH --+ ROOH + R..
R'OOH ca. 0.1 min-' verv slow p~ x _I__f $methyleneoxindole (6) where RH and R'H denote IAA and skatole, respectively, and rate constants in parentheses are for the reactions of peroxidase C at pH 4.4. This mechanism implies that the catalytic function of peroxidase is to form the free radical R. and the enzyme is not directly involved in the activation of molecular oxygen. Pilet and Galston (33) found that the IAA oxidase activity is correlated with a peroxidase-generating capacity. Many investigators have suggested formation of a hydroperoxide from IAA, but nobody has shown that the hydroperoxide is substituted for hydrogen peroxide in the peroxidase reaction shown in Reactions 1 to 3. Catalase reacts with organic hydroperoxides but cannot decompose them catalytically (34). Therefore, it is conceivable that the IAA oxidase reaction is Indole-3-acetic Acid Oxidation by Peroxidases a77 inhibited by a large amount of catalase (1,11,35). Schonbaum (36) has shown that peroxidase C forms spectroscopically distinct complexes with aromatic hydroxamic acids, hydrazides, amides, and cY-hydroxyketones and also that peracids structurally related to these compounds react with peroxidase C at high rates to form Compound I (37). Peroxidase A has a smaller affinity for the hydroxamic acids (38) and reacts with the peracids or organic hydroperoxides at much slower rates (27) in comparison with peroxidase C. This tendency becomes more pronounced in the reaction with the hydroperoxide derived from IAA (Table I). Marked differences in the IAA oxidase activity of the two isoenzymes can be accounted for in terms of the hi and hi values. The rate of the IAA oxidase reaction at steady state is mainly controlled by the h3 value. The reaction reaches the steady state after the hydroperoxide accumulates to a certain concentration.
Since Reaction Fig. 1 can be explained by the fact that the hl value for peroxidase A differs widely from hydrogen peroxide to the hydroperoxide derived from IAA. Compound II appears exponentially after IAA is added to an aerobic solution of peroxidase C (Fig. 7B). This seems to reflect the fact that the hydroperoxide (R'OOH) is formed according to an autocatalytic mechanism, as can be seen in Reactions 1 to 5. The detailed analysis of similar reactions has been performed in the NADH oxidase reaction catalyzed by peroxidase (32,39). It can be said from the result in Fig. 10 that the chain propagation (Reaction 5) may work efficiently in the early stage of the reaction where the formation of hydroperoxide free radical (R'OO.) is extremely slow. We assume that a trace amount of the hydroperoxide is necessary to start the reaction (9).
It has been reported that oxyperoxidase is formed during the peroxidase-oxidase reactions (15, 31) and reacts with IAA at a relatively high rate in comparison with other electron donors (40,41). Specific interaction between peroxidase and indole derivatives has been reported (10, 42). Therefore, the mechanism that peroxidase behaves like an oxygenase to form an active ternary complex of the enzyme, IAA, and oxygen becomes very attractive. Ricard and Job (10) reported that the formation of Compound II is preceded by that of oxyperoxidase in the IAA oxidase reaction. Our results shown in Fig.  7, A and B seem to contradict their conclusion. The discrepancy may arise from the difference in experimental conditions and also from complexity in spectral analysis of the enzyme intermediates (10). Since Compound II reacts with IAA about 10 times as fast as oxyperoxidase does (41), the concentration of oxyperoxidase would become 10 times that of Compound II in the steady state if Compound II is formed via oxyperoxidase. The slow accumulation of oxyperoxidase shown in Fig.  7A implies that its formation is a side reaction and has no significant role in the catalytic oxidation of IAA. It seems likely that the ferric enzyme is reduced by the IAA free radical (R .) formed in Reactions 2 and 3 (5) and the reduced enzyme then reacts with oxygen to form oxyperoxidase (43,44). The mechanism of ascorbate oxidation by the hydroperoxide derived from IAA is not clearly understood. The stoichiometric relationship shown in Fig. 4 may suggest that the peroxidases catalyze the oxidation of ascorbate by the hydroperoxide. This simple explanation, however, is not true because 1) the ascorbate inhibition is much stronger than expected from competition of IAA and ascorbate for Compound II and 2) the ascorbate inhibition occurs in the presence of oxygen ( Fig. 8A) but not in the presence of hydrogen peroxide under anaerobic conditions (Fig. 8B). It seems that ascorbate reacts with the hydroperoxide free radical. The addition of ascorbate does not change the amount of oxygen consumption but decreases the amount of the hydroperoxide accumulation. Products of the IAA oxidation are not thoroughly analyzed in this experiment.
The fact that the mole ratio of oxygen consumed to IAA added is about 0.75 may suggest the formation of indole-3-methanol @'OH) as a product of Reaction 1. The major end product 3-methyleneoxindole is formed through Reaction 6, as suggested by Hinman and Lang (8). Minor products may be formed from disproportionation of the hydroperoxide free radical when the rate of its formation is high.