On the mechanism of chlorination by chloroperoxidase

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 252, No. 1, January, pp. 292302.198’7

On the Mechanism of Chlorination

by Chloroperoxidase’

H. BRIAN DUNFORD,*” ANNE-MARIE LAMBEIR,*33 MOHAMMED A. KASHEM*,4 AND MICHAEL PICKARDt *Department

of Chemistry and +Department of Microbiology, University Edmonton, Alberta, Canada T6G .%‘G2

Received July 15,1986, and in revised form September

of Alberta,

17,1986

Spectral-scan results obtained on the millisecond time scale are reported for reactions of chloroperoxidase with peracetic acid and chloride ion in both the presence and the absence of monochlorodimedone. A multimixing experiment is performed in which stoichiometric amounts of chloroperoxidase and peracetic acid are premixed for 0.7 s before the resultant compound I is reacted with chloride ion. The combined results show that the only detectable enzyme intermediate species is compound I (except in very late stages of the reaction), that the disappearance of compound I is accelerated by the presence of chloride ion, and that it is further accelerated if both chloride and monochlorodimedone are present. It is concluded that compound I is an obligate intermediate species in the reaction. Experiments are performed on the reaction of monochlorodimedone with hypochlorous acid in both the presence and the absence of added chloride ion, but in the absence of chloroperoxidase. The presence of chloride ion greatly accelerates the reaction rate apparently by setting off a chlorine chain reaction. This reaction would be important in the enzyme-catalyzed reaction if hypochlorous acid were liberated into the solution. A careful analysis of steady-state kinetic results shows that in the chlorination of monochlorodimedone at least, liberation of free hypochlorous acid is not important in the enzyme-catalyzed pathway. Rather the reaction proceeds from compound I to formation of iron(III)-OCI by chloride ion addition to the ferry1 oxygen atom. This obligate intermediate species then chlorinates the substrate. It is well described as enzyme-activated hypochlorous acid, in which replacement of the proton in HOC1 by the heme iron ion produces a Cl+ species of great potency. Thus the enzyme controls chlorination of monochlorodimedone rather than unleashing an uncontrolled chain reaction in which it would be rapidly destroyed. o 1381 Academic PWS, IIK. Several other recent papers have addressed the mechanism of chlorination of organic substrates by peroxide or peracid and Cl- 5 catalyzed by chloroperoxidase. A theme

Halide and ligand binding properties of chloroperoxidase have been studied extensively by a variety of techniques (l-5). ’ This work was supported in part by Natural Sciences and Engineering Research Council of Canada Operating Grants A1248 to H.B.D. and A6482 to M.A.P. * To whom correspondence should be addressed. a A.-M.L. has been a postdoctoral fellow supported by the Alberta Heritage Foundation for Medical Research. Present address: Institute for Cellular Pathology, Tropical Disease Unit, Avenue Hippocrates ‘74, B-1200 Brussels, Belgium. 4M.A.K. is holder of an Alberta Heritage Foundation for Medical Research studentship. 0003-9X361/87 $3.00 Copyright All righta

Q 1987 by Academic Press, Inc. of reproduction in any form reserved

’ Abbreviations used: E or Fe+(W), native chloroperoxidase (where + refers to the net charge on the ferric ion as affected by the pyrrole nitrogens of the heme); ROOH, peracetic acid; Fe+(V)=O, compound I (in which V is a formal oxidation state only); ICI or Fe(III)-OCl, enzyme chlorinating intermediate formed by addition of Cl- to compound I; HA or MCD, monochlorodimedone; CIA. dichlorodimedone; EC1 or Fe(III)-Cl, native enzyme-chloride ion complex; v, initial rate; [Eb, total enzyme concentration; +, ionic strength. 292

MECHANISM

OF CHLORINATION

common to all of the papers is a role for HOCl, whether free in solution or enzyme bound (6-10). Although Clz formation was observed in the absence of organic substrates, it was not thought to be an intermediate in the chlorination reaction. Rather, chlorination was believed to proceed through compound I formation and an iron(II1) hypochlorite intermediate (6). In a steady-state kinetic study, chlorination by free HOC1 and enzyme-activated hypochlorous acid (the same iron(III)-OCl species described above) was considered. Available evidence appeared to favor the latter (7). Two product analysis studies led to the conclusion that free HOC1 acid is the most likely chlorinating reagent (8) and that reaction does not occur at the enzyme active site (9). However, there does appear to be a difference in product distribution between the uncatalyzed and chloroperoxidase-catalyzed reactions (1,ll). Finally a mixed enzymatic-nonenzymatic chain reaction was proposed for the halogenation of MCD and it was speculated that an ironchloride ion-dioxygen complex preceded HOC1 formation (10). In this paper we report the results of a variety of experiments which are relevant to the mechanism of chlorination of MCD and assessthe current state of knowledge of this reaction. MATERIALS

AND

293

BY CHLOROPEROXIDASE

using a l-cm observation chamber. Study of the reaction of chloroperoxidase compound I with Cll was made possible by use of a Durrum Model D-110 stopped-flow apparatus equipped with a Dionex Model D-132 multimixing attachment (dead time 50 ms). Experimental details have been described elsewhere (A.-M. Lambeir, H. B. Dunford, and M. A. Pickard, submitted for publication). All experiments on the Cary spectrophotometer and Union Giken rapid reaction analyzer were performed at 25.0 f 0.2”C. The multimixing experiments were performed at 22 * 1°C. RESULTS

The Reaction of Monochlorodimedone Hypochlorous Acid

with

(i) In the absence of added chloride ion. Pseudo-first-order rate constants were determined at 278 nm in phosphate buffer, pH 2.8, both with excess HOC1 and with excess MCD. In the former case simple exponential traces were obtained; in the latter the traces consisted of an initial rapid exponential curve followed by a slow nonexponential portion. Experimental conditions and results of all exponential traces are summarized in Fig. 1 where the slope

400;

METHODS

Chloroperoxidase was prepared by reported methodology (12). MCD was obtained from Sigma, peracetic acid from FMC Corporation, and reagent grade KC1 from Anachem. A neutral stock solution of HOC1 was prepared by bubbling Clz through 0.5 mM NaOH. The Clz was prepared by dripping concentrated HCl (Anachem) on reagent grade MnOz (Fisher). Reagent grade NaOH was obtained from BDH. The concentration of the neutral HOC1 solution was determined by placing an aliquot in a solution 0.1 M with respect to NaOH and measuring the absorbance at 290 nm where cock- = 0.35 mM-' cm-i (13). All solutions were prepared with water obtained from a Mini-Q water purification system (Millipore). In general, solutions were prepared and handled as described in (7). Routine spectral measurements were made on a Cary 219 spectrophotometer. Stopped-flow and rapidscan experiments were performed on a Union Giken rapid reaction analyzer Model 601 by procedures previously described (14,15). This instrument has a dead time of 4 ms when, as in the present study, one is

WW

( l

) or

WOW

( 0 1 (PM)

FIG. 1. Plot of the observed pseudo-first-order rate constant versus concentration of reactant, either monochlorodimedone or hypochlorous acid, for the reaction between the two species. The reaction was followed at 278 nm on a stopped-flow apparatus. The slope yields the second-order rate constant for the reaction, /carp = 1.6 X lo6 M-' s-l. (0) [HOCl], 8.5 PM, excess MCD; (0) [MCD], 5 pM, excess HOCl. Phosphate buffer, pH 2.8, p 0.10 M with NazSO, added to maintain p where necessary, 25°C.

294

DUNFORD

is the same whichever reagent is in excess. From the slope a rate constant of 1.6 X lo6 M-l s-l is obtained for the reaction of MCD with HOC1 at pH 2.8. The pH dependence wasdetermined in experiments in which only the MCD solution contained buffer prior to mixing. The results are shown in Fig. 2. The rate constant increases with increasing pH. When HOC1 is also buffered prior to mixing there is no effect at pH 2.8 (phosphate) but there is at pH 4.2 and 4.6 (citrate). The simplest explanation appears to be that HOC1 reacts with citrate buffer. The same result was obtained when citrate was replaced with formate. (ii) In the presence of chloride iolz If both solutions contained buffer and 0.05 M Cl-, the reaction was too fast to observe in the stopped-flow apparatus ([MCD], 5.0 PM; initial [HOCl], 47 PM). If both solutions contained 0.05 M Cl- but only the MCD was buffered, the rate constant increased by factors of 2.2 at pH 3.0 and 1.1 at pH 4.6 (compared to no added Cl-). The latter result shows that MCD can compete with Clas a reactant with HOC1 if the HOC1 and 70

8 22

I

I

I

I

ET AL.

Cl- are not allowed to preequilibrate acid. Rapid Scan Spectra during the Reaction of Peracetic Acid, Chloride Ion, and Chlwoperoxidase

When chloroperoxidase plus added Clare reacted with peracetic acid, the native enzyme is converted into compound I (Fig. 3). In the latter stages of the reaction the appearance of a small amount of a third enzyme species is indicated by the lack of clearly defined isosbestic points. The third species exhibits the spectral properties of compound II. Rapid Scan Spectra during the SteadyState Chlorination of Monochlorodirnedme Catalyzed by Chlwoperoxidase

The steady-state kinetics of chlorination of MCD in the presence of peracetic acid and Cl- has been studied intensively over the pH range 2.6-4.0 where chlorination rate is optimal (7). Similar conditions are reported in Fig. 4 for rapid scan experiments. It can be seen that only two chloroperoxidase species are present, the native enzyme and compound I. As the concentration of Cl- is decreased, the percentage of compound I increases until nearly 100% of the enzyme is present as compound I. Disappearance of Compnd

6.5

in

I in the

Presence of Chloride

$

6.0

I

I

1

3

I

4

PH FIG. 2. Plot of the log of the second-order rate constant versus pH for the reaction of monochlorodimedone with hypochlorous acid. At pH 2.8 and 3.0, phosphate buffer was used; at higher pH values citrate was used. (0) Only MCD buffered; (0) both MCD and HOC1 buffered. The HOC1 appears to react with citrate buffer where the two reagents are mixed prior to the stopped-flow experiment. [MCD], 5 PM; [HOCl], 61 pM. Other conditions as in Fig. 1.

The results of multimixing experiments are shown in Fig. 5. Unbuffered chloroperoxidase (8 PM) was premixed with 8 PM peracetic acid weakly buffered with phosphate at pH 6.8. After a delay time of 700 ms to allow for compound I formation, the resultant solution was mixed with citrate buffer of pH 5.0. In one experiment no chloride was present; in another 4 mM chloride was present in the citrate buffer. The final pH was 5.0. It can be seen that the rate of disappearance of compound I is greatly enhanced by chloride in the time interval immediately after mixing. In view of the results displayed in Fig. 2, some of

MECHANISM

OF CHLORINATION

295

BY CHLOROPEROXIDASE

is a very slow reaction and within experimental error the rate of its disappearance is not accelerated by the presence of Cl-. Reaction between Peroxide w Peracid with Chloride Ion When either H202 (2 mM) or peracetic acid (2 mM) is mixed with Cl- (0.02 M), no reaction (15 min) is detected at 235 nm where both ROOH species absorb. DISCUSSION

The kinetics of halogen hydrolyses have been studied intensively (16). For chlorine the overall reaction in unbuffered solution

I 360

I 400

440

Wavelength (nm) FIG. 3. Spectral changes occurring in the Soret region where chloroperoxidase plus chloride ion in one reservoir are mixed with peracetic acid in the other. The top curve is the spectrum of the native enzyme. The remaining curves 1-5 were obtained over a 1-ms interval commencing 1, 22,44, 110, and 288 ms after mixing. The only enzyme species present appear to be native enzyme and compound I until the late stages of the reaction. Compound I is approaching its steadystate concentration as the reaction progresses. Curve 5 does not match the isosbestic points of the earlier curves. The small deviations can be explained by the formation of some compound II. [Chloroperoxidase], 2.4 pM; [Cl-], 5 mM; [ROOH], 2 mM, pH 3.5.

the citrate buffer may be chlorinated in the multi-mixing experiments.

I )

400

Wavelength

The Peroxidation of Monochlorodimedone and Its Possible Rote in Catalyzing the Chlorination Reaction When 7.0 X lop4 M MCD is mixed with 4.2 mM peracetic acid and 0.3 M KCl, the rate of disappearance of MCD can be followed in the Cary spectrophotometer by monitoring the disappearance of the MCD band centered at -280 nm. One obtains a rate constant of 7.0 X lo-* s-l. When the KC1 is replaced by 0.24 M KzS04, the rate constant is 6.6 X lo-* s-l (data not shown). Thus the uncatalyzed peroxidation of MCD

D

(nm)

FIG. 4. Spectra taken during the steady state chlorination of monochlorodimedone as a function of chloride ion concentration. In all cases [chloroperoxidase], 2.4 pM; [ROOH], 2 mM; [MCD], 0.1 mM; pH 2.8. All spectra were recorded in the 3- to 4-ms interval after mixing. The isosbestic points indicate clearly that the only detectable enzyme species are native enzyme and compound I. The top curve is the native enzyme spectrum. For the remaining curves from top to bottom, the values of [Cl-] and the yields of eompound I in the steady state are (1) 3 X 10-z M, 33%; (2) 1 X lo-‘, 50%;(3) 3 X lo-’ M, 70%;(4) 3 X lo-’ M, 97%. Thus the lower the [Cl-], the higher the [compound I] since the chlorination reaction is proceeding more slowly.

296

DUNFORD 0.4s

FIG. 5. Multimixing experimental results showing directly that chloride ion catalyzes the disappearance of compound I. Chloroperoxidase is premixed with a stoichiometric amount of peracetic acid for -0.7 s before the resultant compound I is reacted with Cl-, or is mixed solely with buffer. The reaction is followed at 400 nm. The dashed line indicates the spontaneous disappearance of compound I, the solid line its disappearance in the presence of Cl-. Initial concentrations: [chloroperoxidase] = [peracetic acid] = 8 PM; [Cl-], 4 mM. The peracetic acid was weakly buffered at pH 6.8; the NaCl was more strongly buffered at pH 5.0. Final pH 5.0 for both experiments.

Clz + Hz0 ; HOC1 + H+ + Cl-

I

PI

The reaction actually passes through an intermediate species C&OH- but only the overall process need be considered here. The value of k is 11.0 s-* and of kclz 1.80 X lo4 M-‘s-~ at 20°C (16). Other potentially relevant chlorine chemistry consists of Clz + Cl- * Cl, K= 0.18 M-’ at 25°C

Ref. (17)

HOC1 + H+ + ClOK= 3.2 X 10e8 M at 25°C

Ref. (18)

However for 0.05 M Clz and 0.1 M Cl-, a combination of concentrations which approximate an upper limit for a set of chlorination conditions in the presence of chloroperoxidase, [Cl;] would be 9 X 10e4 M at equilibrium and so can be neglected. In acid solution the presence of hypochlorite anion ClO- can also be neglected. Chloroperoxidase-Catalyzed For chlorination reactions or peracid and chloride ion, chloroperoxidase, it would three chlorinating reagents sidered: hypochlorous acid

Chlorination by peroxide eatalyzed by appear that need be con(HOCl), en-

ET AL.

zyme-activated hypochlorous acid (iron(III)-OCl in the enzyme active site), and molecular chlorine (Cl,). In order to choose among them we examine their possible mechanisms of formation and test these mechanisms with the available experimental evidence. Compound I can be detected directly during the steady-state chlorination of monochlorodimedone (Fig. 4). The inverse correlation of compound I concentration with chloride ion concentration is entirely compatible with the formation of compound I as the primary step in the overall reaction. This is followed by its reaction with chloride ion, direct evidence for which is presented in Fig. 5. A first obligatory step in the reaction is compound I formation Fe+(III)

+ ROOH+

Fe+(V)=0

+ ROH

[2]

The choice of peracetic acid is far superior to use of hydrogen peroxide. Compound I is so unstable itself that the further complication of its catalatic reaction is highly undesirable. The catalatic reaction is minimized by choice of peracetic acid. Indeed it was only through the combined use of peracetic acid and rapid scan experiments that the spectrum of pure compound I was finally measured (14). Chloride ion is known to bind weakly to native chloroperoxidase. However, this reaction does not appear to be the most reasonable choice as the initiator of the chlorination reaction (10). The chloride ion is competitive with cyanide binding, indicating that it is binding in the sixth coordination position of the ferric ion (7). In the case of myeloperoxidase, Cl- binding to the active site has been confirmed by resonance Raman spectroscopy (20). Thus when chloride binding occurs it is blocking the active site of the native enzyme to subsequent reaction with ROOH. The subsequent reaction of compound I with Cl- (Fig. 5) is of key importance. If a species such as iron(III)-OCl is formed by direct addition of Cl- to compound I it must have a very transient existence. No trace of it can be seen in the data in Fig. 4 nor in the crucial initial reaction period in Fig. 3. The combined data in Figs. 4 and 5 show for the first time that Cl- catalyzes the de-

MECHANISM

OF CHLORINATION

composition of compound I back to the native enzyme. It would appear reasonable that the reaction Fe+(V)=0 is followed

Fe(III)-OCl

+ H’ + Fe+(III)

+ HOC1 [4]

+ Cla + Hz0

[5]

+ HA + H+ + Fe+( III) + ClA + Hz0

[6]

can compete successfully with Reactions [4] and [5] as we shall show in the following discussion. In a steady-state study of the chlorination of monochlorodimedone by peracetic acid and Cl- catalyzed by chloroperoxidase (7) the two most plausible mechanisms were considered to be

Fe+(III)

I:

+ ROOH 2 Fe+(V)=0

+ ROH

PI Fe’(V)=O+Cll+H+2 Fe+(III)

+ HOC1

HOC1 + HA 2 ClA + Hz0 Mechanism

+ Cll2

[8]

PI

II:

Fe’(II1)

PI

[lo]

+ ClA + Hz0

[ll]

Appropriate corrections were made for the weak binding of Cl- to the native enzyme + Cl- 5 Fe(III)-Cl

WI

which slightly inhibits the formation of compound I. (In Eq. [12] and the subsequent discussion, K is defined as a dissociation constant.) Plots of v/[El, vs [ROOH] yielded rectangular hyperbolae of the form Bi[ROOH] B2 + [ROOH]

-= ,;I0 From mechanism B,=kL[Cl-1;

u31

I,

&=F(l,Y)

[14] 1

and from mechanism

II,

Ic,[HAl[Cl-] B1 = (k,/&J[HA]+ [Cl-]’ & WW[HAlWl (l+[Cl-]/K)=(k,/&)[HA]+[Cl-]

P51

According to mechanism I, plots of B1 vs [Cl-] and B,/(l + [Cl-]/K) vs [Cl-] should be linear. Clearly they are not (Figs. 3 and 4, Ref. (2)). (The correction term for chloride binding to native enzyme (1 + [Cl-]/ K) is not significant in Fig. 4 of Ref. (Z).) According to mechanism II plots of B1 and B2 vs [Cl-] should be rectangular hyperbolae, and this is clearly the case. However, the possibility of other mechanisms must be considered (21). The power of the kinetic method lies not in its ability to prove mechanisms but to disprove them. The steady state results clearly exclude mechanism I; they do not prove that mechanism II is important (7). Here we extend our analysis to include several other possibilities. The following steps may be added to the first two steps of mechanism I (Eqs. [7] and [8]) to produce mechanism III. III

Fe+(III)+ROOH~Fe+(V)=O+ROH

Fe(III)-OCl

+ HA + H’ 2

Fe+(III)

in the absence of organic substrate to be chlorinated. Either or both of these reactions must be fast since there is no clear evidence of Fe(III)-OCl formation in Fig. 3, but ample evidence of compound I formation. At the present time it is not possible to distinguish between Reactions [4] and [5]. It might appear that stoichiometric addition of chloride ion to compound I would eliminate Reaction [5]. However Reaction [5] can compete with Reaction [3] as soon as some halogenating intermediate is formed. Thus, in the absence of HA, one cannot distinguish between CIZ formed in Eq. [5] and the reverse reaction in Eq. [l]. In the presence of MCD (HA) the reaction

Mechanism

Fe( III)-OCl

+ Cll + 2H+ + Fe+(III)

Fe(III)-OCl

Fe+(V)=0

[3]

by either of

Fe(III)-OCl Fe(III)-OCl

+ Cl--*

297

BY CHLOROPEROXIDASE

HOCl+Cl-+H+%la+HZO Cl,+HA&lA+H++Cl-

WI r171

298

DUNFORD

This does not lead to any change in B1 and I (see Appendix for this and subsequent steady-state derivations). Therefore mechanism III can be excluded. Similarly mechanism IV, a chain reaction, can be excluded. Mechanism IV is Eqs. [7] and [8] followed by

B2 as defined by mechanism

IV

HOC1 + Cl- + H+ 2 Clz + HZ0

Initiation

WI

Clz + HA 2 A. +Cl+ +H++Cl-

[18]

Propagation Cl. +HA$A-

+H++Cl-

PI

ET AL.

hydrogen peroxide or peracetic acid with Cl- rule out any reaction by uncatalyzed formation of Cla or HOCI. The blank experiments involving MCD, peracetic acid, and either Cl- or SO:- would appear to eliminate a reaction initiated by autoxidized MCD. Furthermore, the rate of MCD autoxidation is not appreciably enhanced by 1 nM chloroperoxidase (7), the concentration used in the steady-state experiments (A.-M. Lambeir, unpublished results). This would appear to eliminate a reaction initiated by chloroperoxidasecatalyzed autoxidation of MCD.

(ii) Reaction with hypochlorous acid in presence and absence of chloride. The re-

sults of Fig. 1 establish the most accurately rate constant to date for the &lA+ClPO1 determined i Cl,+A. reaction of HOC1 with MCD, k = 1.6 X lo6 Termination Cl. + A + $ ClA WI Mpl s-l at pH 2.8 phosphate buffer, 25”C, CL = 0.10. If mechanism II is correct for the B1 and B2 have the same values as in mechchloroperoxidase-catalyzed chlorination of anisms I and III so that mechanism IV can MCD (discussed above), then the HOC1 rebe eliminated. Other possible termination action with MCD provides the correct blank reactions lead to the same conclusion. for a comparison of the rate of chlorination Similarly, mechanism II (starting with by HOC1 compared to Fe(III)-OCl. It was Eqs. [7] and [lo]) can be modified so that pointed out correctly that Cll should also the third step (Reaction [ll]) is replaced by be present if the above experiment is regarded as a blank for the chloroperoxidaseV Fe(III)-OCl+Cl-+2H++ catalyzed reaction where free HOC1 could Fe+(III) + Cla + Ha0 [22] be generated from excess Cl- and peracetic Cl,+HA-+ClA+H++Cl[171 acid (21). Our results indicate that in the presence of Cl- the reaction of HOC1 and to give Mechanism V. MCD is too fast to be measured on a Now stopped-flow apparatus. This is strong evidence that in the presence of HOC1 and Cl- but in the absence of chloroperoxidase, BI = kz+ k,[C1-l; a chlorination reaction of Cl, formed via reaction (21) is occurring. The crucial question is whether this reaction is relevant to the chloroperoxidase-catalyzed mechanism. We examine this possibility. and this mechanism is invalid. Finally mechanism VI (V modified to include a It turns out that it is simply mechanism III (non-chain) or mechanism IV (chain) chain reaction) can also be eliminated (data described in this paper. Neither fits the kinot shown). Therefore of all these possinetic data, so both can be eliminated. Thus bilities only mechanism II fits the experiwe have the interesting case of the nonmental data. A reasonable probability would appear to be that mechanism II is enzymatic reaction of HOCl, Cl-, and MCD which occurs at a very fast rate but is not correct. relevant to the chloroperoxidase-catalyzed reaction of peracetic acid (or hydrogen Noncatalyzed Halogen&m peroxide), Cl-, and MCD. In the particular (i) Reaction initiated by Cl- and peroxetic case of MCD as substrate with chloroperacid. The blank experiments using either oxidase as catalyst the HOC1 appears not

MECHANISM

OF CHLORINATION

to be released to the solution. Rather the species described as enzyme-activated HOC1 or an enzyme-bound chlorinating intermediate Fe(III)-OCl is utilized directly. This species is formed simply by addition of Cl- to the ferry1 oxygen atom of compound I and is utilized to chlorinate substrate before HOC1 is released to the solution. It is esthetically pleasing to conclude that an enzyme has evolved to chlorinate a substrate specifically. If the enzyme were to unleash an indiscriminate chain reaction it would destroy itself. However esthetic considerations did not influence our conclusion, which is based solely on a kinetic and spectral analysis of the problem.

Equilibrium

Cl- Binding E + Cl- ti EC1

K = MCl-1 [EC11 Mass Balance [El,, = [EC11 + [E] + [compound I] = 7[EIW1 + [E] + [compound I] + [compound I] = kgcompound I][Cll] kl[ROOH]

1 + [Cll] ( --IK

+ [compound I]

APPENDIX

= [compound I]

As far as possible the rate constants are numbered the same as in Ref. (7). Reactions are written only to indicate the correct order of reaction. Mechanism

299

BY CHLOROPEROXIDASE

III (Non-chain

1 + MCl-1 kJROOH] Initial

Reaction)

1 +K-1

( -I K

Rate ‘u = k,[Cl,][HA]

E+Cl-SEC1

= kc,JHOCll[C1-][H+]

E + ROOH 2 Compound I + ROH

= k’z[compound I][Cl-]

Compound I + Cl- -% E + HOC1 Vu=

HOC1 + Cll + H+ 2 Cl2 + Hz0

ki[ROOH] (

Cl,+HA&A+H++Cl-

d[compound dt

--IK

k’dCl-] k$X] ’ + kJROOH]

Steady-State -!I!&+ dt

~k[C1-I[Elo 1+ k;lC1-1 1+w-1

I]

k;ICl-][ROOH]

-=V

= k,[EXROOH] - k;[compound

IKCll] = 0

d[HOCl] = k;Ecompound IlCll] dt - k,$HOCl][Cl-][H+] 9

-zz [;&

&[ROOH] [ROOH] + B2

=0 B1 = k%Cl-]

= kcJHOC1][Cl-][H+] - kJCl&HA]

=0

300

DUNFORD

Therefore B1 and B2 are unchanged from mechanism I and mechanism III is not valid. Mechanism

ET AL.

k&l

* IHA]

[A.]=g[Cl.]

IV (Chain Reaction)

= kJC121A. ] [Cl.]=‘$l..;]*’

2

k&Al MC1*I”-= qc121 ki[CMHAl

E+Cl-SEC1 E + ROOH 2 Compound I + ROOH

kJA*pj$$=ki[Cl,IHA]

Compound I + Cl- 2 HOC1 + E [Cl-]=

&C,]

[A.]=

@HA]

HOC1 + Cl- + H+ 3 Clz + Hz0

Initiaticm

Equilibrium

Cl,+HAzA.

+Cl.

+H++Cl-

Cl- Binding

As in Mechanism

Propagation

III.

Mass Balance

Cl. +HA%A.

1 Cl,+A.

+H++Cl-

As in Mechanism

III.

&lA+Cl.

Initial Rate

Termination

v=k&l,][A.]+k&l.l[A.] Cl. +A.

&CIA

= k&l,][A

. ] + k[Cl,IHA]

Steady-State d[compound

= qw

I] and d[HOCl]

dt

dt

as in mechanism d[Cl . ] = k#&IHA]

dt

- k&l

III.

4A.l dt

= kJC1,1[HA] + k&l.

approximation

applies

XHA]

of the above two equations k[Cl,aHA]

&‘++l,][HA]

The steady-state to Cl&

-k&l,IA.]-k&l.l[A.]=O Addition to

v=(k,,

* IHA]

+k&1,1[A.]-k&Jl.IA.]=O -

[HA] + ki[Cl‘JHA]

leads

= kt[Cl . IA. ]

Substitution into the above equations d[Cl * ]/dt and d[A * ]/dt yields

k&l,IA -I= 0 cc1I = h.xJHOWCl-IH+l 2 ki[HAl+ kdlA*I - k[Cl,rHA]

for

2k[ClzIHA]

- k5[C1. HHA] + k#&JA

.] = 0

2k[ClaHA]

+ k,[Cl . l[HA] - kJCl,~A

.] = 0

Subtraction that

of the last two equations shows

-

= kc,JHOCIIC1-HH’]

Successive substitution of the steady-state values for [HOC11 and [Compound I], and use of the conservation relation for [Eb yields

MECHANISM

OF CHLORINATION

k;lCl-HE]0

[f&l =

Finally substitution of [Cl,] into expression for v gives

the Mass

k;lCl-l[ROOH]

Vu=

301

BY CHLOROPEROXIDASE

Balance

Plo

[ROOH]+F(I+y)

= [E] + [EC11+ [compound I] + [ICI]

1

LEl= kdcompound II EW The values of Bl and B2 are identical to those obtained from mechanisms I and III so that mechanism IV is invalid. Mechanism V (Formation of Cl2 via Fe(III)-OCl)

kl [Eb =

[ROOH]

k&C1-1

[compound I]

ki[ROOH]

+ [compound I] + [ICI] [Ej,, = [compound I] 1 +

E+Cl-:ECl

MCI-1 kJROOH]

E + ROOH 2 Compound I + ROOH Compound I + Cl- 2 ICI [compound I] = 2[ICl]

ICl+Cl-&$+E Cl,+HA%lA+H++Cl-

[EJ, = $[ICl]

Steady-State

- k.JCompound Il[Cl-] = 0 $9

= kJCompound I][Cl-] - k,[IclxcI-]

WLI = kdIClICl-] dt

Equilibrium

Binding

Unchanged.

=0

- k4[C12KHA]= 0

302

DUNFORD

ET AL.

Initial Rate v = kzJCl,IHA]

4.

= k&ClXCl-] V,=

5.

kK-IWO

6.

MCl-XROOH]

7. 8.

[ROOH]+y(l+y) 1

9.

B1 = &[Cl-] 10. 11. 12. 13.

14. 15.

This result is the same as mechanism I except for the factor (1 + k6/k.J. It therefore cannot be fit to the experimental data and is also invalid. The only valid mechanism is mechanism II.

16. 17. 18.

REFERENCES 19. 1. SONO, M., DAWSON, J. H., HALL, K., AND HAGER, L. P. (1986) Biochemistry 25,347-356. 2. CHAMPION, P. M., MIJNCK, E., DEBRUNNER, P. G., HOLLENBERG, P. F., AND HAGER, L. P. (1973) Biochemistry 12,426-435. 3. KREJCAREK, G. E., BRYANT, R. G., SMITH, R. J.,

20. 21.

AND HAGER, L. P. (1976) Biochemistry 15,25082511. REMBA, R. D., CHAMPION, P. M., FITCHEN, D. B., CHIANG, R., AND HAGER, L. P. (1979) Biochemistry 18,2280-2290. HOLLENBERG, P. F., HAGER, L. P., BLUMBERG, W. E., AND PEISACH, J. (1980) J. Biol. Chem 255, 4801-4807. LIBBY, R. D., THOMAS, J. A., KAISER, L. W., AND HAGER, L. P. (1982) J. Biol. Chem. 257, 50305037. LAMBEIR, A.-M., AND DUNFORD, H. B. (1983) J. Bid Chem. 258,13558-13563. GEIGERT, J., NEIDLEMAN, S. L., AND DALIETOS, D. J. (1983) J. Biol. Chem. 258,2273-2277. RAMAKRISHNAN, K., OPPENHUIZEN, M. E., SAUNDERS, S., AND FISHER, J. (1983) Biochemistry 22,3272-3277. GRIFFIN, B. W. (1983) B&hem. Biophys. Res. Commun. 116,873-879. BROWN, F. W., AND HAGER, L. P. (1967) J. Amer. Chem sot. 89,719-720. PICKARD, M. A. (1981) Canad J. h&&i01 27,12981305. ZGLICZYNSKI, J. M., STELMASZYNSKA,T., DOMANSKI, J., AND OSTROWSKI,W. (1971) Biochim Biophys. Acta 235,419-424. PALCIC, M. M., AND DUNFORD, H. B. (1980) J. Biol Chem. 255,6128-6132. HARRISON, J. E., ARAISO, T., PALCIC, M. M., AND DUNFORD, H. B. (1980) Biochem Biophys. Res. Commun. 94,34-40. EIGEN, M., AND KUSTIN, K. (1962) J. Amer. Chem, Sot. 84,1355-1361. ZIMMERMANN, G., AND STRONG, F. C. (1957) J. Amer. Chem Sot 79,2063-2066. LATIMER, W. (1952) Oxidation Potentials, 2nd ed., Prentice-Hall, New York. PALCIC, M. M., RUTTER, R., ARAISO, T., HAGER, L. P., AND DUNFORD, H. B. (1980) Biochem. Biophys. Res. Commun 94,1123-1127. IDEDA-SAITO, M., ARGADE, P. V., AND ROUSSEAU, D. L. (1985) FEBS Lett. 184,52-55. GRIFFIN, B. W., AND ASHLEY, P. L. (1984) Arch B&hem. Biophys. 233,188-196.