One- and two-electron oxidations of tyrosine, monoiodotyrosine, and diiodotyrosine catalyzed by hog thyroid peroxidase.

Stopped flow experiments were carried out with purified hog thyroid peroxidase (A413 nm/A280 nm = 0.42). In the steady state of oxidations of L- and D-tyrosines, N-acetyltyrosinamide, and monoiodotyrosine, thyroid peroxidase existed in the form of Compound I, the primary catalytic intermediate of peroxidase in its reaction with H2O2. Kinetic results led us to conclude that thyroid peroxidase catalyzes two-electron oxidations of these molecules. In the steady state of oxidation of diiodotyrosine, on the other hand, the enzyme was found in the form of compound II at pH 7.4, but in the form of compound I at pH 5.5. The result implies that the mechanism of diiodotyrosine oxidation varied from a one-electron to a two-electron type as the pH decreased. The selection of mechanisms of oxidation appears to be peculiar to thyroid peroxidase; horseradish peroxidase and lactoperoxidase catalyzed only one-electron oxidations of these five donor molecules. Rate constants for rate-limiting steps in the reactions of these donor molecules with the three peroxidases were measured by overall kinetic and stopped flow kinetic methods.

Stopped flow experiments were carried out with purified hog thyroid peroxidase (A413 nm/A280 nm = 0.42). In the steady state of oxidations of L-and D-tyrosines, Nacetyltyrosinamide, and monoiodotyrosine, thyroid peroxidase existed in the form of Compound I, the primary catalytic intermediate of peroxidase in its reaction with H202. Kinetic results led us to conclude that thyroid peroxidase catalyzes two-electron oxidations of these molecules. In the steady state of oxidation of diiodotyrosine, on the other hand, the enzyme was found in the form of Compound II at pH 7.4, but in the form of Compound I at pH 5.5. The result implies that the mechanism of diiodotyrosine oxidation varied from a one-electron to a two-electron type as the pH decreased.
The selection of mechanisms of oxidation appears to be peculiar to thyroid peroxidase; horseradish peroxidase and lactoperoxidase catalyzed only one-electron oxidations of these five donor molecules. Rate constants for rate-limiting steps in the reactions of these donor molecules with the three peroxidases were measured by overall kinetic and stopped flow kinetic methods.
Thyroid peroxidase is involved in the biosynthesis of thyroid hormone (1-4). The primary reaction catalyzed by the enzyme is iodination of the tyrosyl residues in thyroglobulin. Thyroid peroxidase does catalyze the conversion of diiodotyrosine to thyroxine (5-7). Oxidative coupling reactions of diiodotyrosine and other tyrosine derivatives have also been studied with other peroxidases (5, [8][9][10][11][12][13]. Horseradish peroxidase and lactoperoxidase have been used to characterize the specificity of thyroid peroxidase in the coupling reaction. Although the catalytic properties vary slightly among peroxidases (9-11, 13), these differences do not seem essential for characterizing the nature of thyroid peroxidase.
From stopped flow experiments with detergent-solubilized (14) and purified (15) thyroid peroxidase preparations, we have shown that thyroid peroxidase, unlike lactoperoxidase, catalyzes the oxidation of L-tyrosine by way of two-electron transfer. This finding provides the first clear evidence for twoelectron oxidation of organic molecules by the peroxidase systems. Organic molecules such as phenol derivatives are believed to be oxidized to their free radical forms in the presence of peroxidase and H202 (16,17). Therefore, it seems very important to determine whether thyroid peroxidase exhibits a unique catalytic property in the oxidation of monoiodotyrosine and diiodotyrosine as well as L-tyrosine. We have carried out kinetic experiments with purified hog thyroid peroxidase and the results are reported in this paper.

MATERIALS AND METHODS
Hog thyroid peroxidase used in this experiment was prepared as described previously (15). The ratio of A4L3to A m nm of our enzyme preparation was 0.42, and its concentration was tentatively calculated by using a value of 114 for emM at 413 nm, which is used for lactoperoxidase (18). Horseradish peroxidase used was isoenzyme C (A4w d A~w -= 3.4), and lactoperoxidase ,,,,,/A2~ = 0.90) was donated by Dr. S. Nakamura, Hirosaki University (Hirosaki, Japan) and Dr.

RESULTS
In general, peroxidase reactions are formulated as follows (16,19,20). Peroxidase + Hz02 -Compound I (1) Compound I + AH? -Compound I1 + AH. k2 (2) Compound I1 + AH1peroxidase + AH. (3) 2AH. -AH-AH (or, A + AHd (4) In most cases, the catalytic intermediate observed in the steady state is Compound 11, since Reaction 1 is fast and k , > 10k3 (19). When the electron donor (AH2) is an organic molecule, it is believed to be oxidized by way of one-electron transfer (17). If peroxidase is present as Compound I1 in the steady state of the reaction, Reaction 3 is rate-limiting, and the oxidation of AH, occurs by way of one-electron transfer. This conclusion can be derived from the analysis of stopped flow traces at two wavelengths isosbestic between any pair of the three enzyme forms involved in the catalytic cycles. Such kinetic results on the oxidations of L-and D-tyrosines, Nacetyltyrosinamide, monoiodotyrosine, and diiodotyrosine, in the presence of horseradish peroxidase and lactoperoxidase (Fig. I), led us to conclude that these reactions occurred by way of one-electron transfer. When Reaction 2 is relatively slow, Compound I must be seen transiently before the formation of Compound 11. This was observed during the oxidation of diiodotyrosine by horseradish peroxidase and Hz02 (Fig. 2 A ) . Fig. 2 shows that the catalytic intermediate of horseradish peroxidase and lactoperoxidase in the steady state of diiodotyrosine oxidation was Compound 11. The results shown in Figs. 1 and 2 are in accord with the accepted mechanism (Reactions 1 to 3).  The reaction catalyzed by thyroid peroxidase was not simple. The kinetic traces shown in Fig. 3 were grouped into two patterns. There was no essential difference among the three peroxidases in the mechanism of diiodotyrosine oxidation at pH 7.4. The enzyme intermediate observed in the steady state of diiodotyrosine oxidation was apparently Compound 11, and the rate constant for Reaction 3 was calculated from the plot in the inset of Fig. 3 to be 2.0 X lo4 M" s-l according to Chance's equation (21). In contrast, the intermediate of thyroid peroxidase appearing in the steady state of oxidations of L-and D-tyrosines, N-acetyltyrosinamide, and monoiodotyrosine was not Compound I1 (Fig. 3). We have identified the intermediate in the reaction of thyroid peroxidase with Ltyrosine as Compound I (15). Kinetic traces for D-tyrosine resembled those for L-tyrosine, and were omitted in Fig. 3. From comparison of Figs. 3 and 4, we concluded that Compound I accumulated in the steady state of oxidations of Dtyrosine, N-acetyltyrosinamide, and monoiodotyrosine. The slow accumulation of a small amount of Compound I1 would be ascribable to the reduction of Compound I by endogenous donor present in the enzyme preparation. The reaction pattern depended on the pH of the reaction solutions when the electron donor was diiodotyrosine. From the shapes of the two difference spectra shown in Fig. 4B, the enzyme intermediate appearing during the oxidation of diiodotyrosine was found to be Compound I1 when the reaction was carried out at pH 7.4, but was Compound I at pH 5.5. The intermediate was Compound I in the oxidation of monoiodotyrosine at either pH 7.4 or 5.5 (Fig. 44).
The rate constant for Reaction 3 could be measured directly .......... FIG. 4. Difference spectra of thyroid peroxidase observed during the oxidations of monoiodotyrosine (A) and diiodotyrosine (B). The spectra were obtained at 1 s from the stopped flow traces at varying wavelengths at pH 5.5 (0) and 7.4 (0). The experimental conditions were as described in Fig. 3. The spectra decreased with time without changing their shape. from the reaction of Compound I1 with electron donors.

I"
Kinetic traces for monoiodotyrosine and diiodotyrosine are shown in Fig. 5. Similar experiments had been carried out with detergent-solubilized hog thyroid peroxidase (22). In this experiment, care was taken to start the reaction after the amount of HZ02 became negligibly small in Compound I1 solutions. Since Compound I1 was slowly reduced back to the ferric enzyme at the expense of endogenous donor, the reaction was started by the addition of an electron donor at a time when a small percentage of the peroxidase was present as the ferric enzyme (upper traces in Fig. 5). The observed first order rate constant was proportional to the concentration of added electron donor (Fig. 5, lower portion), and the second order rate constants thus calculated are listed in Table I, together with data on L-and D-tyrosines and N-acetyltyrosinamide.
Despite the difference in the mechanism of oxidations of Land D-tyrosines, N-acetyltyrosinamide, and monoiodotyrosine among thyroid peroxidase and the other peroxidases, no apparent difference could be detected in the difference spectra of their oxidation products. In Fig. 6, the spectral data on L-   ' The larger value was obtained by adding diiodotyrosine at a time when the Compound I1 concentration was decreased to about onehalf. The larger value seemed accurate, though its precise estimation became difficult.  tyrosine and monoiodotyrosine are shown. By following the absorbance changes, the rates of oxidations of the electron donors were measured. Under experimental conditions in which Reaction 1 was not rate-limiting, the rate must be proportional to the donor concentration. Fig. 7 shows an example in which diiodotyrosine was used as the electron donor. From these overall kinetic experiments, the rate constant for Reaction 2' or 3 could be measured, and the results are shown in Table 11. Here, Reaction 2' is formulated as follows. e Data on D-tyrosine are not shown in figures, but the reaction patterns were similar to those for L-tyrosine. 'Only this value is measured at pH 5.5.
A value of 75 was obtained i n our previous experiment (15).   (Fig. 3) might be explained by this effect, and the rate constant calculated from the trace would be a tentative value.
The pH dependence of thyroid peroxidase activity has been measured for oxidations of guaiacol (23,24) and iodide (24,25), and for iodination (9, 10,25) and coupling (10). The measurement of the rates of oxidations of tyrosine and their derivatives at varying pH value was not easy because of the pH dependence of absorbance (26, 27). Fig. 8A shows pH activity curves for thyroid peroxidase when the electron donor was L-tyrosine, monoiodotyrosine, and diiodotyrosine. The rates of oxidation were estimated by using data shown in Fig.  8B. There was a remarkable difference in the optimal pH between monoiodotyrosine and diiodotyrosine. The difference would be ascribable to the fact that the phenolic group is deprotonated more easily in diiodotyrosine than in monoiodotyrosine.

DISCUSSION
Since a typical feature in peroxidase reactions is the formation of radical species of bivalent redox molecules, it has been suggested that free radicals are formed as intermediates in the processes of thyroid hormone biosynthesis, particularly in iodination (2, 4, 5, 13, 28, 29) and in the coupling reaction (2, 4, 13). Bjorksten (30), however, found that iodide reduced Compound I of horseradish peroxidase directly to the ferric form. The mechanism was confirmed by stopped flow experiments (31). From the analysis of enzyme intermediates during the oxidation of iodide, thyroid peroxidase and lactoperoxidase were found to catalyze a two-electron oxidation of iodide (14). Therefore, the enzymatic iodinating species is not an iodine free radical, but probably an iodinium cation, as suggested by several workers (2, 12, 14,25, 32, 33). The proposed mechanism would also explain the inhibitory action of antithyroid agents and excess iodide on the peroxidase-catalyzed iodination (34, 35). Now, a similar question is directed to the coupling reaction. We suggest, from data shown in Figs. 3 and 4, that, at physiological pH, diiodotyrosine is oxidized to its free radical by the thyroid peroxidase system. The rate constant for the reduction of Compound I1 by diiodotyrosine at pH 7.4 is measured by three different methods, as shown in Figs. 3, 5, and 7. The values thus obtained lie between 1 X lo4 and 3.7 X lo4 M" s" (Tables I and 11). The rate constant might be concluded to be about 2 X lo4 M" s-l, though further experiments are needed for its accurate estimation.
The reactions of Compound I of thyroid peroxidase with Land D-tyrOSineS, N-acetyltyrosinamide, and monoiodotyrosine are rather exceptional. The accumulation of Compound I in the steady state of peroxidase reactions would be interpreted in terms of alternative mechanisms. Approximate rate constants (lo4, M" s") are shown in parentheses.
The rate constant for the reaction of Compound I with DIT should be greater than 2 X lo5 M" s " because the [Compound I]/[Compound 111 ratio in the steady state is less than 0.1 (Fig. 4B).
excluded in the reactions with D-tyrosine, N-acetyltyrosinamide, and monoiodotyrosine, because the rate constants for Reaction 3 shown in Table I are too small to explain the absence of Compound I1 in the steady state of Case 2 reactions.
Therefore, we conclude that the accumulation of Compound I during the oxidation of these donor molecules should be explained by the case 1 mechanism. When diiodotyrosine is oxidized at pH 5.5, the overall rate is close to k , (Tables I and   11). The rate, however, should be much slower if Compound I accumulates via the Case 2 mechanism. So it seems that, at pH 5.5, diiodotyrosine is oxidized mostly through the Case 1 mechanism. The detection of phenoxy radicals by ESR spectroscopy may afford conclusive evidence, since clear ESR signals have been observed during the oxidation of cresol by the horseradish peroxidase system (36, 37). At physiological pH, thyroid peroxidase catalyzes a one-electron oxidation of diiodotyrosine, and two-electron oxidations of L-and D-tyrosines, N-acetyltyrosinamide, and monoiodotyrosine, as shown in Fig. 9. The catalytic feature of the enzyme to differentiate the mechanism must be most interesting from the physiological point of view. Deme et al. (38) reported regulatory effects of diiodotyrosine on the biosynthesis of thyroid hormone, but at present it is uncertain whether their findings are related to our mechanism. Table I1 shows that the rate constant measured by the overall kinetics is much smaller than that measured by the stopped flow method, in the reactions of thyroid peroxidase, particularly with L-and D-tyrosines. This discrepancy might be explained by the fact that Compound I1 accumulates slowly through side reactions, and its reduction by Land D-tyrosines is very slow (Table I). Since very good agreement is observed for turnover numbers determined by the overall and stopped flow methods with all five substrates in the reactions of lactoperoxidase and horseradish peroxidase, the discrepancy observed with thyroid peroxidase should be ascribed to the nature of the enzyme. Further studies are needed to explain it quantitatively.
The one-electron and two-electron mechanisms in enzymatic oxidation-reduction reactions have been discussed in detail (17, 39). There are several examples in which the electron transfer mechanism is affected by experimental conditions. As regards the pH dependence, a two-electron mechanism tends to change into a one-electron type as the pH increases. Such reactions are reductions of O2 and p-benzoquinone by the xanthine oxidase system (40-42) and the oxidation of sulfite by the horseradish peroxidase system (43). The reason might be that the two-electron transfer is feasible only when the supply of proton is not limited at a crucial moment of the electron transfer. The fact that diiodotyrosine deprotonates with a pK, value of 6.37 (44) may be related to the pH-dependent change in the mechanism of electron transfer from diiodotyrosine to Compound I of thyroid peroxidase.
It has been reported (45,46) that the reduction of Compound I of horseradish peroxidase to the ferric enzyme is accompanied by an uptake of 2 protons. The apparent two-electron transfer from L-tyrosine and monoiodotyrosine to Compound I of thyroid peroxidase is notable as regards the coupling mechanism. The primary product of phenol compounds (ROH) is a phenoxy radical (RO.) in a one-electron mechanism and probably a phenoxy cation (RO') in a two-electron mechanism. The two-electron transfer tends to occur in a specifically bound complex between enzyme and substrate (17). An apparent two-electron oxidation has also been pro- Several investigators have questioned whether thyroid peroxidase functions specifically and differently in thyroid hormone biosynthesis, as compared with other peroxidases (1, 9, 10, 14, 15,28,49-51). Regarding iodination, lactoperoxidase is as active or more active than thyroid peroxidase (9,49). Nunez (4) has suggested specificity of thyroid peroxidase in the coupling reaction. Michot et al. (52) reported that iodide increases the rate of tyrosine coupling in the presence of thyroid peroxidase, but not of horseradish peroxidase and lactoperoxidase. As shown in Table 11, thyroid peroxidase catalyzes the oxidations of tyrosine and diiodotyrosine much faster than others do. Thyroid peroxidase and lactoperoxidase resemble each other in many respects, but a distinct difference between the two enzymes is now found in the mechanism of oxidation. Its physiological implication is most interesting and should be elucidated by further experiments. Since the COUpling reaction in vivo occurs in thyroglobulin molecules, the environment of tyrosyl groups would control both iodination and coupling reactions. We are now studying whether the mechanism derived from the kinetic data with free L-tyrosine and iodotyrosines is applicable to the reactions in thyroglobulin.