Myeloperoxidase-catalyzed taurine chlorination: initial versus equilibrium rate

www.elsevier.com/locate/yabbi

ABBArchives of Biochemistry and Biophysics 466 (2007) 221–233

Myeloperoxidase-catalyzed taurine chlorination:Initial versus equilibrium rate

Daniel R. Ramos a, M. Victoria Garcıa a, Moises Canle L. a,J. Arturo Santaballa a, Paul G. Furtmuller b, Christian Obinger b,*

a Chemical Reactivity & Photoreactivity Group, Department of Physical Chemistry & Chemical Engineering I, University of A Coruna,

Alejandro de la Sota 1, E-15008 A Coruna, Spainb Department of Chemistry, Division of Biochemistry, BOKU-University of Natural Resources and Applied Life Sciences,

Muthgasse 18, A-1190 Vienna, Austria

Received 14 May 2007, and in revised form 15 July 2007Available online 17 August 2007

Abstract

Myeloperoxidase (MPO) catalyzes the two-electron oxidation of chloride, thereby producing hypochlorous acid (HOCl). Taurine(2-aminoethane-sulfonic acid, Tau) is thought to act as a trap of HOCl forming the long-lived oxidant monochlorotaurine [(N-Cl)–Tau], which participates in pathogen defense. Here, we amend and extend previous studies by following initial and equilibrium rateof formation of (N-Cl)–Tau mediated by MPO at pH 4.0–7.0, varying H2O2 concentration. Initial rate studies show no saturation ofthe active site under assay conditions (i.e. [H2O2] P 2000 [MPO]). Deceleration of Tau chlorination under equilibrium is quantitativelydescribed by the redox equilibrium established by H2O2-mediated reduction of compound I to compound II. At equilibrium regime themaximum chlorination rate is obtained at [H2O2] and pH values around 0.4 mM and pH 5. The proposed mechanism includes knownacid–base and binding equilibria taking place at the working conditions. Kinetic data ruled out the currently accepted mechanism inwhich a proton participates in the molecular step (MPO-I + Cl�) leading to the formation of the chlorinating agent. Results supportthe formation of a chlorinating compound I–Cl� complex (MPO-I–Cl) and/or of ClO�, through the former or even independently ofit. ClO� diffuses away and rapidly protonates to HOCl outside the heme pocket. Smaller substrates will be chlorinated inside the enzymeby MPO-I–Cl and outside by HOCl, whereas bulkier ones can only react with the latter.� 2007 Elsevier Inc. All rights reserved.

Keywords: Myeloperoxidase; Chlorination; Taurine; Monochlorotaurine; Hypochlorous acid; Compound I; Compound II

Myeloperoxidase (MPO)1 is a heme-containingdominant granule enzyme present in circulating polymor-phonuclear neutrophils, which represent the cornerstoneof cell-mediated antimicrobial activity in the human innate

0003-9861/$ - see front matter � 2007 Elsevier Inc. All rights reserved.

doi:10.1016/j.abb.2007.07.024

* Corresponding author. Fax: +43 1 36006 6059.E-mail address: [email protected] (C. Obinger).

1 Abbreviations used: MPO, myeloperoxidase; MPO–H, protonatedspecies of MPO; MPO–H–Cl, protonated myeloperoxidase–chloridecomplex; MPO-I, compound I; MPO-I–H, protonated species of com-pound I; MPO-I–Cl, compound I–chloride complex; MPO-I–Cl–H,protonated species of compound I–chloride complex; MPO-II, compoundII; MPO-II–HO2, compound II–hydroperoxide complex; Tau, taurine;(N-Cl)–Tau, monochlorotaurine.

immune system. Neutrophils endocytose pathogens anddeposit their granule contents intracellularly into targetphagosomes [1]. The principal reaction catalyzed byMPO under physiological conditions is thought to be theoxidation of Cl� by H2O2 to yield the highly-reactive oxi-dizing and chlorinating agent HOCl [1]. In this respectMPO is unique among the members of the animal peroxi-dase superfamily [2] because the high reduction potential ofits redox intermediate compound I (MPO-I) allows thetwo-electron oxidation of Cl� [3]. HOCl can participatein subsequent nonenzymatic reactions such as oxidationand chlorination of target cell components, and can alsoreact with substances in the immediate environment that

mailto:[email protected]

222 D.R. Ramos et al. / Archives of Biochemistry and Biophysics 466 (2007) 221–233

modulate its biological effects. For example HOCl chlori-nates ammonia and amines to yield chloramines, whichalso have oxidizing and chlorinating activity [4]. The mostimportant role of HOCl and chloramines in leukocytefunction is probably to attack essential microbial cellcomponents.

Taurine (Tau, 2-aminoethane-sulfonic acid) is one of themost abundant free amino acids in mammalian tissues, andaccounts for a large part of the amines that are availablefor reaction with chlorinating reagents. Intracellular con-centrations between 22 and 26 mM were reported forhuman leukocytes [5,6]. Although Tau is found in the cyto-plasm of neutrophils and not released into phagosomes, ithas been proposed that it could play a role in scavenging ofHOCl released from the phagosomes [4]. Thereby, thelong-lived oxidant monochlorotaurine [(N-Cl)–Tau] isformed, which is a hydrophilic anionic membrane-imper-meable oxidant with relatively low toxicity [7]. Monochlo-rotaurine was reported to be uptaken by red blood cellsand, finally, reduced by intracellular components, such asglutathione or hemoglobin, to Tau [4], which accumulatesinside these cells [8]. These results suggest that red bloodcells contribute to elimination of (N-Cl)–Tau in vivo, pre-venting its accumulation in the extracellular phase and pro-tecting other blood cells, plasma components, and tissuesfrom oxidative attack [8].

At pH 7.0 the formation of (N-Cl)–Tau from Tau andHOCl (Reaction 3 in Table 1) takes place with a measuredrate constant of 4.8 · 105 M�1 s�1 [9], whereas at pH 4.7the rate constant is 3.5 · 103 M�1 s�1 [10]. Further chlori-nation to Tau–dichloramine, which is more toxic than(N-Cl)–Tau and has significant cytolytic activity [8], is rel-atively slow. Generally, dichlorination is much slower thanmonochlorination. This is best demonstrated by the fact

Table 1Mechanisms of MPO-catalyzed Tau chlorination proposed to fit exper-imental kinetic data

Mechanism Reactions No.

I MPOþH2O2 ¢K1

MPO-IþH2O 1

MPO-Iþ Cl� þHþ!k2MPOþHOCl 2

HOClþ Tau!k3 ðN -ClÞ–TauþH2O 3

MPOþHþ ¢K4

MPO–H 4

MPO-IþHþ ¢K5

MPO-I–H 5

MPO–Hþ Cl� ¢K6

MPO–H–Cl 6

MPO-IþH2O2 ¢K7

MPO-IIþHO�2 þHþ 7

MPO-IIþHO�2 þHþ!k8MPOþH2OþO2 8

II MPOþH2O2 ¢K1

MPO-IþH2O 1

MPO-Iþ Cl�!k2bMPO-I–Cl 2b

MPO-I–Clþ TauþHþ!k3bðN -ClÞ–Tauþ MPOþH2O 3b

MPOþHþ ¢K4

MPO–H 4

MPO-IþHþ ¢K5

MPO-I–H 5

MPO–Hþ Cl� ¢K6

MPO–H–Cl 6

MPO-IþH2O2 ¢K7

MPO-IIþHO�2 þHþ 7

MPO-IIþHO�2 þHþ!k8MPOþH2OþO2 8

that formation of (N-Cl)–methylamine (2 · 108 M�1 s�1)is about 5 orders of magnitude faster than the consecutivereaction between HOCl and (N-Cl)–methylamine [11], orby the published rate of ammonia chlorination by HOCl(6.1 · 106 M�1 s�1), which is about 4 orders of magnitudefaster than further chlorination of monochloroamine [12].Since Tau is not an electron donor for either compound I(MPO-I) or compound II (MPO-II), it is an ideal moleculeto investigate the mechanism of its chlorination by MPO[10,13].

The chlorination of Tau by MPO has been studied byMarquez and Dunford [10,13]. However, in order to avoidconsideration of the complex scenario of reactions outsidethe chlorination cycle of MPO (Table 1), the authors con-ducted most of their measurements at low pH (4.7) andfocused only on initial rate data of Tau chlorination toeliminate the effect of formation of Tau–dichloramine[10,13]. From their kinetic data they could conclude thatthe chlorination reaction mediated by MPO in vivo mayinvolve an enzyme intermediate species rather than freeHOCl. Nevertheless, the authors did not include MPO-typ-ical reactions, which play an important role in catalysisin vitro and in vivo, and thus significantly affect chlorinationreactions, especially at high pH. These reactions need to beconsidered when investigating chlorination of Tau over thewhole period of the reaction. This is possible since, as men-tioned above, formation of Tau–dichloramine is negligible.Additionally, the influence of pH should be studied inorder to determine whether protons intervene in the reac-tion mechanism.

The reactions involved in the so-called chlorination cycleof MPO are the formation of MPO-I, which has beendescribed to be slightly reversible (Reaction 1 in Table 1)[14,15] and its direct reaction back to the ferric enzyme(Reaction 2 in Table 1). Moreover, both Fe(III)–MPOand MPO-I could participate in protonation reactions(Reactions 4 and 5 in Table 1). In addition to a redox reac-tion with H2O2, protonated native MPO can also bind Cl�

forming a high-spin complex (MPO–H–Cl, Reaction 6 inTable 1); this binding constant has been accurately mea-sured as a function of pH [16–18]. Very important is the roleof H2O2 as one-electron donor reducing MPO-I to MPO-IIand forming HO�2 (hydroperoxyl) radical (Reaction 7 inTable 1). Compound II is outside the halogenation cycle.The rate constant for the reduction of MPO-I to MPO-IIby H2O2 was reported to be 8.2 · 104 M�1 s�1 [14,15] atpH 7.0, so this reaction is competitive with MPO-I reduc-tion by Cl� at neutral pH regions [19]. The only alternativeof a compound I to compound II transition in the systemMPO/H2O2/Cl�/Tau (i.e. in the absence of an exogenousone-electron donor) is intramolecular electron transfer thatquenches the porphyryl radical of compound I and oxidizesan amino acid of the protein matrix. This is known to occurwith MPO but is extremely slow and thus of no significancein the presence of H2O2 [14,15].

The one-electron oxidation of H2O2 produces superoxide[14,15] which has been reported to mediate MPO-II

0 200 400 600 800 1000

-0.10

-0.05

0.00

0.05

0.10

0.15

A

t / s 0.0 0.5 1.0 1.5 2.0

-0.010

-0.005

0.000

0.005

0.010

0.015

A

t / s

0.0 0.5 1.0 1.5 2.00.00

0.01

0.02

0.03

0.04

t / s

df(A

)dt

df(A

)dt

100 120 140 160 180 200

-0.04

-0.03

-0.02

A

t / s

100 120 140 160 180 200

2.2x10-4

2.3x10-4

2.3x10-4

2.4x10-4

t / s

Fig. 1. (a) Kinetic trace obtained in the stopped-flow spectrophotometer under the following conditions: k = 252 nm, T = 25.0 �C, [MPO0] = 100 nM,[Tau] = 10 mM, [Cl�] = 100 mM, [H2O2] = 0.8 mM, pH (100 mM buffer) = 5.5. (b) Initial rate fit and (c) plot of the differentiated function. (d and e) Idemas (b and c) but under ‘‘equilibrium’’ conditions. Each figure is used for illustrative purposes; they do not necessarily belong to the same kinetic run.

D.R. Ramos et al. / Archives of Biochemistry and Biophysics 466 (2007) 221–233 223

reduction to the ferric enzyme [20] (Reaction 8 in Table 1). Asa consequence of these dynamic equilibria the kinetics of (N-Cl)–Tau formation exhibits a very complex behavior. A typ-ical kinetic trace measured at [H2O2] P 0.5 mM and pH P 5is shown in Fig. 1a. The detailed kinetic analysis shouldincorporate all feasible concurrent reactions, allowing amore accurate description of in vivo halogenating activity.

Here, we revisit and extend previous studies includingequilibrium conditions to incorporate concurrent side reac-tions. Two mechanisms (Table 1) for MPO-catalyzed chlori-nation using Tau as a trap were tested in the pH range 4.0–7.0that includes actual conditions during phagocytosis andMPO activity in vivo. The kinetic data suggest that there is

no proton intervening in the direct reaction between MPO-I and Cl�. Chlorination might be mediated mainly by theMPO-I–Cl complex at the active site or by ClO� that under-goes protonation to HOCl outside the enzyme.

Experimental procedures

Materials

Stock solutions of 70 mM Tau (Fluka Microselect, P99.5%), 2.5 MNaCl (Sigma Ultra, min 99.5%) and 18 mM H2O2 (Sigma–Aldrich) wereprepared daily and kept in the dark at 4 �C. The H2O2 concentration wasdetermined spectrophotometrically using e240 = 39.4 M�1 cm�1 [21]. Citricacid (Sigma, anhydrous)/Na2HPO4 (Sigma, ACS reagent) buffer, 0.5 M,


was used to control pH in the range 4.0–6.0. Phosphate buffer was used formeasurements at pH 7.0.

Highly purified MPO of a purity index (A430/A280) of at least 0.85 wasobtained from Planta Natural Products (http://www.planta.at). Its actualconcentration was determined spectrophotometrically by usinge430 = 91,000 M�1 cm�1 heme�1 [22]. Cu/Zn superoxide dismutase waspurchased from Sigma.

Methods

Based on the published in vivo concentrations [1,23], the typical chlori-nation assays contained 100 mM Cl�, 10 mM Tau and 100 nM MPO.Temperature and pH were set at 25.0 �C and 4.0–7.0, respectively. Hydrogenperoxide was varied from 0.2 to 2.0 mM. Samples were prepared immedi-ately before use by mixing solution 1, containing Tau and MPO, withsolution 2, containing NaCl and H2O2 in 200 mM phosphate/citrate buffer(pH 4.0–6.0) or phosphate buffer (pH 7.0), by using the conventional mixingstopped-flow technique (Applied Photophysics� SX-18 MV stopped-flowspectrophotometer equipped with a diode array detector). Single-mixingunit syringes were filled up with both sample solutions and, after thermo-stating at 25.0 ± 0.1 �C, a total of 100 lL were shot for each run. Reactionswere studied both at the whole wavelength range as well as at 252 nm, whichcorresponds to the absorbance maximum of (N-Cl)–Tau. The molarextinction coefficient of the mono-chlorinated and dichlorinated product at252 nm is 429 and 190 M�1 cm�1, respectively, but—based on the hugedifferences in the reaction rate with HOCl—formation of the dichlorinatedproduct was negligible, especially in case of high excess of Tau [11,12].Exploratory experiments showed no impact of [Tau] on the obtained kineticdata under the working conditions. Kinetic traces at 252 nm were followeduntil full consumption of H2O2, and led to re-formation of native MPO. Inorder to determine the stoichiometry of H2O2 consumption and (N-Cl)–Tauformation the molar extinction coefficient of the absorbing species at theworking wavelength had to be taken into account. As the concentration of(N-Cl)–Tau at any time t, [(N-Cl)–Tau]t, can be estimated, then [H2O2]tequals [H2O2]0 � [(N-Cl)–Tau]t.

In the initial phase of the reaction (Fig. 1b), the kinetics of absorbanceincrease at 252 nm showed a curvature that could be fitted to a doubleexponential (Eq. (1)). This behavior may arise from the fact that the rate isstrongly dependent on both H2O2 and the actual MPO-II concentrations.By using Eq. (1) any portion of the initial time traces can be fittedproperly. Thus, the derivative of the resulting function just after reagentsmixing allows the calculation of the initial rate (Fig. 1c). Approximately 10replicated runs were fitted and the initial rates obtained from these fitswere averaged

y ¼ y0 þ A1 � ð1� e�x=t1 Þ þ A2 � ð1� e�x=t2 Þ ð1Þ

This procedure seems more precise than classical methods where the initialrate is obtained from the slope of the more or less shorter portion at thebeginning of the kinetic trace; it implies uncertainties in the initial rate val-ues, especially relevant for noisy traces. The approach used here reducesthe scatter of kinetic data, and estimates the initial rate after mixing, i.e.

just at ‘‘zero time’’.Finally, a portion of the residual time trace, after dynamic equilibrium

has been achieved, was fitted using the same mathematical expression (Eq.(1)) yielding the equilibrium rate at a particular time (Figs. 1d and e). Thetotal increase in absorbance (DA) observed at this point allowed calcula-tion of the amount of (N-Cl)–Tau formed and, therefore, the remainingconcentration of H2O2, since there is no noticeable loss of H2O2 throughperoxidative processes. This is consistent with the fact that the presence ofCu/Zn superoxide dismutase had no impact on the equilibrium chlorina-tion rate of Tau (not shown).

Results

MPO-catalyzed chlorination of Tau follows a rathercomplex mechanism. So far only the initial stages of this

reaction at pH 4.7 have received mechanistic consideration[10,13]. Here, we followed the formation of (N-Cl)–Tauover the whole period of H2O2 consumption between pH4.0 and pH 7.0, varying [H2O2]. Experimental data were fit-ted to the two mechanisms summarized in Table 1. Thebasis of these mechanisms has been taken from the litera-ture [10,13–20]. The novelty in the current study is thatwe have considered all the known acid/base, binding andredox equilibria taking place. This enabled the estimationof intrinsic kinetic constants and consequently clarificationof the molecular steps involved in the reaction mechanism.

Mechanism I takes into account eight reactions thatinfluence the kinetics of (N-Cl)–Tau formation. Theseinclude oxidation of Fe(III)–MPO by H2O2 to MPO-I(Reaction 1) and the known MPO-typical one-electronreduction of MPO-I to MPO-II by H2O2, with subsequentrelease of HO�2 [14,15] (Reaction 7). Reduction of MPO-Iby Cl� yields Fe(III)–MPO (Reaction 2) and HOCl, whichis capable of oxidizing or chlorinating several substrates.These include Tau, which reacts to form (N-Cl)–Tau(Reaction 3). Mechanism I also includes protonation equi-libria of both Fe(III)–MPO (Reaction 4) and MPO-I(Reaction 5) at the active site, as well as the formation ofa high-spin complex by Cl� binding to protonatedFe(III)–MPO (Reaction 6). Superoxide mediated reductionof MPO-II to Fe(III)–MPO has been reported [20], thusReaction 8 is also included in the reaction mechanism.

In Mechanism II, according to the proposal by Marquezand Dunford [10,13], Reactions 2 and 3 are exchanged withReactions 2b and 3b, i.e. Cl� binds to MPO-I producing anew enzymatic complex, MPO-I–Cl, (Reaction 2b), whichacts as the actual halogenating agent that chlorinates Tauaccording to Reaction 3b. Despite the noticeable disparityon initial rate values [10,13] with respect to our data, thereare not further reasons to reject this pathway, so bothmechanisms are considered as departing point.

Initial rate (v0) chlorination measurements

In the beginning we investigated the dependence of theinitial Tau chlorination rate (v0) on both the [H2O2] andpH. Initial rate of Tau chlorination increased upon raisingpH from 4.0 to 7.0. At neutral pH the reaction was too fasteven for the stopped-flow technique. Data obtained for fivepH values are depicted in Fig. 2a. They arrange in straightlines. These linear regressions were obtained by using errorvalues as weight, although the corresponding error bars arenot shown for the sake of clarity. Fig. 2a unequivocallydemonstrates that saturation of the active site is neverreached under the assay conditions (i.e.: [H2O2] P 2000[MPO]). In this initial stage of (N-Cl)–Tau formationMPO-II does not participate in the enzyme turnover. Thisis valid based on the published kinetic data of MPO-Ireduction. It is well known that at pH 7 MPO-I reductionby Cl� to Fe(III)–MPO and by H2O2 to MPO-II occur atsimilar rates and that, in contrast to the H2O2-mediatedreaction, halide oxidation becomes significantly faster with

http://www.planta.at

0.0000 0.0005 0.0010 0.0015 0.00200.0000

0.0002

0.0004

0.0006

v 0 / M

s-1

[H2O2] / M

400 5000.00

0.01

0.02

λ / nm

A

430 nm

0.0 3.0x10-5 6.0x10-5 9.0x10-50.0

0.1

0.2

0.3

0.4

k obs /

s-1

[H+] / M

Fig. 2. (a) Initial rate versus [H2O2]. k = 252 nm, T = 25.0 �C,[MPO0] = 100 nM, [Tau] = 10 mM, [Cl�] = 100 mM, [H2O2] = 0.2–2.0 mM, pH (100 mM buffer) = 4.0 (j), 4.5 ( ), 5.0 ( ), 5.5 ( ), and6.0 ( ). Inset: UV–vis spectrum obtained at this phase of the reactionshowing the Soret band. (b) Observed kinetic constant (kobs) versus [H+].Reaction conditions are the same as in (a). Solid line corresponds to Eq.(2) fit, dotted line is the best fit with Eq. (3).


decreasing pH [14,15,19]. Moreover, in the different assays[Cl�] was at least 50-fold that of H2O2. As shown in theinset of Fig. 2a the Soret band appears at 430 nm, whichis typical for MPO in its Fe(III) and MPO-I forms. Thus,in Mechanisms I and II only Reactions 1–6 have to be con-sidered and Eqs. (12A) (Mechanism I) and (19A) (Mecha-nism II) from the Appendix can be applied (Table 2).

Table 2Initial rate and equilibrium regime rate equations obtained for Mecha-nisms I and II

Mechanism Initial rate equation Acidity dependence

I v0 ¼ k2 �K1 �½MPO0 ��½H2O2 ��½Cl��½Hþ�1þK4 �½Hþ��ð1þK6 �½Cl��Þ kobs ¼ v0

½H2O2 � ¼½Hþ�

aþb�½Hþ�

II v0 ¼ k2b �K1 �½MPO0 ��½H2O2��½Cl��1þK4 �½Hþ��ð1þK6 �½Cl��Þ kobs ¼ v0

½H2O2 � ¼a

bþ½Hþ�

Mechanism Equilibrium regime rate equation

I veq ¼ k2 �½MPO0 ��½Cl��½Hþ�1þK4 �½Hþ��ð1þK6 �½Cl��Þ

K1 �½H2O2 �þ1þK5 �½Hþ�þ

K07�½H2 O2 �

½Hþ�

II veq ¼ k2b �½MPO0 ��½Cl��1þK4 �½Hþ��ð1þK6 �½Cl��Þ

K1 �½H2O2 �þ1þK5 �½Hþ�þ

K07�½H2 O2 �

½Hþ�

Uncertainties in initial reaction rates are around 10% atlow pH and/or low [H2O2], except for the four deviatedvalues at pH 6, where uncertainty goes to 40%. At highpH and [H2O2] values kinetics are much faster, whichimplies less precise measurements. The deviation of thefour points at pH 6 is only apparent; the straight line is thatof the error-weighted least squares fit, which properlycrosses all error bars (not represented). Furthermore, thesefour deviated data pairs do not clearly point to a curvedtrace but to a lower slope, since a non-weighted leastsquares straight line could also be properly drawn throughthose data.

The dependence of the initial rate of Tau chlorinationon pH can be better illustrated by representing the slopesobtained in Fig. 2a (kobs) versus pH (Fig. 2b). This pseudo-constant is independent of [H2O2], and its meaningbecomes apparent by considering Eqs. (12A) and (19A)(Table 2). Thus, the plotted data have been fitted withtwo different functions (Eqs. (2) and (3)):

kobs ¼a

bþ ½Hþ� ð2Þ

kobs ¼½Hþ�

aþ b � ½Hþ� ð3Þ

As shown in Fig. 2b, Eq. (2) derived from the initial rateexpression of Mechanism II (Eq. (19A)), fits properly theexperimental data, which is an indication of the adequacyof this pathway, while Eq. (3) cannot fit the experimentalvalues, suggesting that Mechanism I does not satisfactorilydescribe the system. Mechanism II includes the formationof a MPO-I–Cl complex, which, finally, chlorinates Tauin a proton dependent reaction (Reaction 3b).

‘‘Equilibrium’’ rate (veq) chlorination measurements

Although contradictory at first sight, we decided to usethe term ‘‘equilibrium’’ instead of ‘‘steady-state’’ rate inorder to describe the study carried out beyond initial rate.Steady-state is the stage in which the concentration of anintermediate remains constant for a period of time. Thisconcept is used in enzyme kinetics when the concentrationof the enzyme–substrate complex (E–S) remains invariableand, therefore, the reaction rate being constant, with thecorresponding kinetic trace depicting a straight line. Onthe other hand, the transient state—or pre-steady state—can be defined as the initial phase in which both [E–S]and the reaction rate increase accordingly, until they stabi-lize and keep constant (steady-state). However, this is notthe case in the present study where kinetic traces exhibita more complex profile. Due to accumulation of MPO-IIthe initial chlorination rate decelerates rapidly and is fol-lowed by a phase, where chlorination proceeds very slowly.Finally, it smoothly increases again due to H2O2 consump-tion and decay of [MPO-II], before it stops due to completeperoxide depletion. The described profile does not suit typ-ical steady-state kinetics. While at the typical pre-steady


state the rate should increase to reach a steady-state situa-tion that corresponds to the maximum concentration ofactive E–S complex, our initial phase exhibits a decreasein reaction rate according to the formation of MPO-II.The monitored subsequent pseudo-straight phase is notdue to the constant maximum concentration of activeE–S complex, but to the equilibrium of formation, and,therefore, nearly constant concentration of inactiveMPO-II. Thus, in order to avoid confusion and highlightthe different concept, this stage of the reaction is describedas ‘‘at equilibrium’’.

Rates of Tau chlorination measured at equilibrium (veq)show a fully different profile when plotted against [H2O2](Fig. 3a). After an initial rise the curves bend down withincreasing [H2O2], this effect being more significant athigher pH’s. Tau chlorination rates for pH P 5.5 and[H2O2] P 0.25 mM were very slow, indicating that a signif-icant fraction of MPO is outside the chlorination cycle, thelatter including only Fe(III)–MPO and MPO-I. This wasthe reason why we finally focused on the pH range 4–6.Looking at the dominating enzyme redox intermediate inthis stage of H2O2 consumption revealed that the enzymegathered as MPO-II, which has its Soret maximum at456 nm (inset to Fig. 3a) and a prominent band at

0.0000 0.0005 0.0010 0.0015 0.00200.0

2.0x10-6

4.0x10-6

6.0x10-6

8.0x10-6

1.0x10-5

v eq /

M s-1

[H2O2] / M

400 5000.00

0.01

0.02

A

λ / nm

456 nm

0.0 3.0x10-5 6.0x10-5 9.0x10-50

20000

40000

60000

80000

100000

b

[H+] / M

Fig. 3. (a) Rate at equilibrium versus [H2O2]. k = 252 nm, T = 25.0 �C, [MPO(100 mM buffer) = 4.0 (j), 4.5 ( ), 5.0 ( ), 5.5 ( ), and 6.0 ( ). Inset: UV–vi(b) a versus [H+]. (c) b versus [H+]. (d) log(c) versus pH (slope = 1.05 ± 0.14).

625 nm (not shown). Thus, at equilibrium, i.e. after passingthrough many chlorination cycles in the initial phase,Reactions 7 and 8 became relevant and need to be consid-ered in Mechanisms I and II. The null effect of the additionof Cu/Zn superoxide dismutase on the equilibrium chlori-nation rate of Tau, suggests that superoxide is not releasedfrom the heme cavity into the bulk medium. Thus, it is rea-sonable to assume that it does not leave immediately theactive site but rapidly reacts in its vicinity where it rebindsto MPO-II (MPO-II–HO2) before it goes back to MPO-Ior donates its electron re-forming Fe(III)–MPO.

Accordingly, MPO-II–HO2 would be considered asintermediate species. Rebinding to the active site has beenalso reported to occur for O2 and NO at the distal hemecavity of myoglobin [24]. Eqs. (17A) (Mechanism I) and(20A) (Mechanism II) from the Appendix were applied,with Eq. (4) being valid for the two mechanisms (Table2). These sets of data were fitted iteratively (Figs. 3a–d),so that a- and b-values fit to a straight line with respectto [H+] (Eq. (5)) (Figs. 3b and c), whereas c-values versus

[H+] fit to Eq. (6) (Fig. 3d). The effect of [H+] on a, b

and c parameters has been worked out from the expres-sions of the rate at equilibrium for valid Mechanism II(Eq. (20A)) (Table 2).

0.0 3.0x10-5 6.0x10-5 9.0x10-50

20

40

60

80

100

120

140

a

[H+] / M

4.0 4.5 5.0 5.5 6.0

7.5

8.0

8.5

9.0

9.5

10.0

log

(c)

pH

0] = 100 nM, [Tau] = 10 mM, [Cl�] = 100 mM, [H2O2] = 0.2–2.0 mM, pHs spectrum obtained at this phase of the reaction showing the Soret band.


veq ¼1

a½H2O2�

þ bþ c � ½H2O2�ð4Þ

a ðand bÞ ¼ a0 þ b0 � ½Hþ� ð5Þ

c ¼ a00

½Hþ� ð6Þ

0 1000-0.1

0.0

0.1

t / s

A

350 400 450 500 5500.000

0.005

0.010

0.015

0.020

0.025456 nm

430 nm

A

λ / nm

350 400 450 500 5500.000

0.005

0.010

0.015

0.020

0.025

456 nm

A

λ / nm

0 1000-0.1

0.0

0.1

t / s

A

350 400 450 500 5500.000

0.005

0.010

0.015

0.020

0.025

430 nm

A

λ / nm

0 1000-0.1

0.0

0.1

t / s

A

Fig. 4. Representative UV–vis spectra obtained from MPO in the initialphase (a), at equilibrium (b), and during H2O2 depletion (c), under thefollowing conditions: T = 25.0 �C, [MPO0] = 100 nM, [Tau] = 10 mM,[Cl�] = 100 mM, [H2O2] = 0.8 mM, pH (100 mM buffer) = 5.5. Insets:Kinetic trace showing Tau chlorination at k = 252 nm. The correspondingphases from which the representative spectra were taken are highlighted inbold black.

Curvature of the kinetic traces was due to decelerationof the chlorination activity because of accumulation ofMPO-II (Fig. 4a). After reaching the dynamic equilibrium,the chlorination process proceeded very slowly since theconcentration of active compounds (MPO and MPO-I)was reduced to a small percentage of the total enzyme.Along the central portion of the kinetic trace the enzymestill cycled consuming H2O2 and chlorinating Tau, while[MPO-II] decreased according to H2O2 consumption(Fig. 4b). Formation of compound III [14] was notobserved at the concentration of H2O2 that was used.Upon depletion of H2O2, the absorbance at 430 nmincreased indicating accumulation of Fe(III)–MPO(Fig. 4c).

These findings show that the decrease in the chlorinationrate under equilibrium conditions with respect to initialrates can be ascribed to MPO-II accumulation. From theexperimental data (Figs. 3c and d) K 07ð¼ k7=ðk�7 þ k8ÞÞcould be calculated to be 0.77 ± 0.13 M. Data from Figs.2b and 3c yielded a value for K1 [(8.3 ± 3.7) · 103 M�1]very similar to the value calculated in a different mannerfrom Figs. 2b, 3b and c [(9.3 ± 4.2) · 103 M�1], whichmight reflect the high reduction potential of the redox cou-ple MPO-I/MPO [3]. Besides, the fit shown in Fig. 3callowed the calculation of k2b to be (1.58 ± 0.15) ·104 M�1 s�1, which is very similar to MPO-I reduction rateat physiological [Cl�] [19,25].

Additionally, the equilibrium constant of Reaction 5,K5, was calculated to be (1.46 ± 0.16) · 105 M�1, whichcorresponds to a pKa of MPO-I–H of 5.16. It has beendemonstrated that the protonated form of MPO-I effi-ciently oxidizes halides [19]. At acidic pH’s halides bindnear a protonated site, which has to be very close to the fer-ryl oxygen since there is a halide-dependent change in theEPR spectrum at acidic pH [26]. The crystal structures ofthe MPO–bromide and MPO–cyanide–bromide complexessuggest [27] that the distal histidine upon protonationcould interact with the ferryl oxygen and the bound halide.Published pKa values are in the range 4.5–5.2 [16,17,19,26]thus underlining the adequacy of Mechanism II.

Applying the published value (170 lM) for the dissocia-tion constant of the MPO–H–Cl complex [14]) togetherwith the fitting constants of Fig. 2b, K4 was calculated tobe (3.4 ± 1.1) · 103 M�1; this corresponds to pKa (MPO–H) = 3.53, which is close to the published value of 3.95 [18].

Discussion

The chlorination of Tau by MPO has been investigatedby Marquez and Dunford at very specific conditions[10,13], namely at pH 4.7 and by linearly fitting the initialportion of the time traces at 252 nm. We have extendedthese studies including equilibrium conditions and thepH range 4.0–7.0. Our data provide some discrepancieswith previous studies [10,13], and new mechanisticinsights regarding formation and nature of the chlorinat-ing agent.


Absence of active site saturation

We could not find that the initial rate of chlorination ofTau exhibits saturation behavior [10,13], probably due totwo main reasons. The first one is related with the rapidkinetics system. In former work [10,13] data collectionfrom kinetic runs was triggered 4 ms after mixing; nowa-days stopped-flow equipment allows effective mixing below1 ms, and such difference is important when dealing withthe estimation of initial rates for fast kinetics. When con-sidering the rate of formation of MPO-II measured bythe same authors [k = (8.2 ± 0.2) · 104 M�1 s�1] [15], itcan be noticed that, under conditions used in the study,within 4 ms already some formation of MPO-II occurred,as depicted in Fig. 1b of the same paper [15]. In case ofthe discussed paper [10] the time used for the fit seems tobe significantly higher (up to 100 ms) to avoid interferenceof noise in the linear fit.

The present work (Fig. 5) clearly demonstrates thatdecrease of the chlorination rate starts very fast after reac-tants mixing, and adequate initial rate values should not bemeasured after 4 ms. At low concentration of H2O2 theeffect of MPO-II formation is almost negligible (Fig. 5a)and the slope obtained after 5 s is still approximately50% of that at 1 ms. At higher [H2O2] MPO-II formationbecomes important. Furthermore, this fast (pseudo)inhibi-

0 3

-0.010

-0.005

0.000

0.005

0.010

A

t / s

0.00 0.02 0.04 0.06 0.08 0.10

-0.008

-0.006

-0.004

-0.002

0.000

0.002

0.004

A

t / s

1 2 4 5

Fig. 5. Initial kinetic traces obtained in the stopped-flow spectrophotom-eter under the following conditions: k = 252 nm, T = 25.0 �C,[MPO0] = 100 nM, [Tau] = 10 mM, [Cl�] = 100 mM, (a) [H2O2] =0.2 mM, pH (100 mM buffer) = 5.0; (b) [H2O2] = 1.0 mM, pH (100 mMbuffer) = 6.0.

tion becomes more evident at higher pH values. Fig. 5bdepicts that the mixing time lasts only about 1 ms, andthe reaction of chlorination and, more important, thebuild-up of MPO-II already takes place just after this shorttime. Within 100 ms the chlorination rate was reduced to26% of the initial rate. Consequently, we used a differentmethodology: the kinetic trace at very early stages of thereaction was fitted to an adequate equation and then theinitial rate was calculated from its derivative, as explainedunder Experimental procedures.

Therefore, most likely the previously observed satura-tion of the active site, which becomes more evident athigher pH’s, is an artifact coming from the delay in datacollection with respect to the physically available zero time.The reported initial rate values, especially for faster kinet-ics, certainly are a composite between initial rates and equi-librium regime rates. Thus, apparent saturation is found asthe reaction rate decreases upon approaching the equilib-rium regime.

On the other hand, the previously proposed mechanismdoes not take into account acid–base and binding equilib-ria relevant for derivation of the rate equations. Here, thecomplementary kinetic information obtained by initial rateand equilibrium rate studies of the same kinetic trace,enables us to draw conclusions up to now not present inthe literature.

Chloride oxidation does not depend on proton concentration

The first important outcome of the present work isrelated to the acidity dependence of the formation of thechlorinating agent; it has been found that under the work-ing conditions there is no proton intervening in the molec-ular step between MPO-I and Cl�. This result rules outmolecular steps like Reaction 2, or a modified version ofReaction 2b, i.e.:

MPO-Iþ Cl� þHþ !k2b0

MPO-I–Cl–H

with Tau being chlorinated by this protonated complex(MPO-I–Cl–H) in a subsequent process. Reaction 2b 0

could be proposed on the basis of the noticeable accelera-tion of the rate of Cl� oxidation by MPO at acidic pH val-ues [16,18,19,28]. However, in this case the correspondinginitial and equilibrium rate equations would be analogousto those obtained for Mechanism I (Table 2), which donot support the observed kinetic behavior. Equivalent rateequations are also deduced by considering the reversibilityof these steps:

MPO-Iþ Cl� þHþ ¢k2

k�2

MPOþHOCl

MPO-Iþ Cl� þHþ ¢k2b0

k�2b0MPO-I–Cl–H

In fact, it has been demonstrated that HOCl can oxidizeFe(III)–MPO to MPO-I (Reaction �2) similar to theH2O2-mediated reaction [29]. Any other mechanistic


alternatives involving a proton in the molecular reaction ofMPO-I and Cl� render rate equations similar to Mecha-nism I, which is not supported by kinetic data. The aboveindicated acceleration at low pH should be explained byother processes, like the pH-dependent equilibrium ofMPO-II formation (Reaction 7). In fact, it has been ob-served that deceleration of chlorination from this speciesincreases with both [H2O2] and pH.

The quest for the chlorinating species

Taking into account that Mechanism II properlydescribes experimental kinetic rates, opens the door tokinetically indistinguishable mechanisms [30] compatiblewith the proposals of the literature, the chlorinating agentbeing either HOCl or MPO-I–Cl. As the reaction rate, bothinitial and equilibrium, is independent of [Tau], the molec-ular steps where Tau participates must be faster than theformation of the chlorinating agent. Therefore, MechanismII could be extended beyond the rate-limiting step by thesteps depicted in Fig. 6 with no effect on the deduced rateequations. Once the intermediate complex MPO-I–Cl isformed (Reaction 2b), it could chlorinate substrates ableto diffuse into the heme cavity (Reaction 3b); the equivalentcompound I–Cl intermediate has been observed in chlorop-eroxidase (CPO) [31]. Furthermore, it could be involved inan acid–base equilibrium to yield MPO-I–Cl–H (Reaction10), which has also been found for CPO [31]; this speciescould merely react with Tau (Reaction 12) and/or likelydissociate to yield Fe(III)–MPO and HOCl (Reaction 13),which diffuses away to oxidize a variety of substrates(Reaction 3).

Another feasible pathway is the breaking of MPO-I–Clto give Fe(III)–MPO and ClO� (Reaction 9). This species,which is much less oxidizing than HOCl, is able to diffusefrom the active site. Away from the enzyme, it is readilyprotonated to HOCl (Reaction 11), which is the activechlorinating agent (Reaction 3). It could be argued thatunder the working conditions for reliable kinetic measure-ments (pH 4–6) ClO� should have a very short lifetime[pKa (HOCl) = 7.54 [32]] in favor of HOCl. However, thispKa value applies for bulk water but not for the inner pro-tein core, where it is produced [33]. Furthermore, the inner

MPO-I-Cl

MPO + ClO-

10 H+

MPO-I-Cl-H HOCl + MPO

H+

HOCl

MPO-I + Cl- 2bTau (N-Cl)-Tau

3 Tau2'

Tau

3b

3 Tau

9

11

12

13

Fig. 6. Mechanistic pathways for the formation of MPO-mediatedchlorinating agents.

environment of the enzyme seems to stabilize negativecharged species, Cl� for instance, and the same must holdfor ClO�. If free HOCl was formed inside the enzyme, itcould rapidly react with every oxidizable group in theneighborhood of the heme inhibiting the activity ofMPO. High diffusion rate and reactivity of ClO� andHOCl, respectively, impede their direct observation.

Additionally, it could also be possible to propose anindependent step to form Fe(III)–MPO and ClO� (Reac-tion 2 0 in Fig. 6), instead of a pathway via MPO-I–Cl; evenin this case the deduced rate equations would be analogousto those shown in Table 2. The same applies if the previousstep or the formation of MPO-I–Cl are considered asreversible:

MPO-Iþ Cl� ¢k20

k�20MPOþ ClO�

MPO-Iþ Cl� ¢k2b

k�2b

MPO-I–Cl

Summing up all the possibilities exposed above: the de-duced rate equation for any reaction mechanism compris-ing Mechanism II (Table 1) and involving some or allsteps after MPO-I–Cl formation (Reaction 2b) is the math-ematical expression shown in Table 2. On the other hand, ifboth Reactions 2b and 2 0 are taken into account, the math-ematical expression of the rate equations holds but not themeaning of the parameters, i.e. predicted dependence onreactants and acidity is equal, but k2b should be replacedby (k2b + k2 0). Finally, the solely consideration of Reaction2 0 as feasible pathway also leads to the same rate equation,with the kinetic constant k2 0 taking the place of k2b. Fur-thermore, according to this reaction scheme (Fig. 6), smallsubstrates, like Tau, could be chlorinated both inside theenzyme by the complex MPO-I–Cl, and outside by releasedClO�/HOCl, whereas the chlorination of bulkier substrateswould only take place outside the heme cavity by reactionwith a free released chlorinating agent (ClO�/HOCl).

Role and turnover of compound II

The second important outcome obtained from initialrate and equilibrium rate studies of the same kinetic trace,relates to the effect of formation of MPO-II (or MPO-II–HO2) on the chlorination cycle. Reactions 7 and 8 (Table1) are responsible for deceleration of the chlorination reac-tion rate. This can easily be deduced from the different ratevalues at initial and equilibrium regimes (Figs. 2a and 3a),and from their appearance in the corresponding mecha-nism rate equation (Table 2 and Eq. (20A)) associated tothe term K 07. This constant ½K 07 ¼ k7=ðk�7 þ k8Þ� involvesthe rate constant of MPO-II formation, k7, and of thereverse reaction, k�7 (i.e. re-formation of MPO-I andH2O2), and the rate constant for superoxide mediatedMPO-II reduction to Fe(III)–MPO, k8. It is important tonote that following Reaction 8 would mean a net loss ofhydrogen peroxide outside the chlorination cycle.

https://www.researchgate.net/publication/232804794_The_Investigation_of_Organic_Reactions_and_Their_Mechanisms?el=1_x_8&enrichId=rgreq-0d8c175d7d335ff371cfec41747fb2b5-XXX&enrichSource=Y292ZXJQYWdlOzU5NzEzNzI7QVM6MTAzNTAxMjA4MjI3ODUzQDE0MDE2ODgwMjE2ODI=

0.0 2.0x10-6 4.0x10-6

0.000

0.005

0.010

0.015

[MPO

-II-H

O2]·[

H+ ] /

([MPO

0]-[M

PO-II

-HO

2]) / M

[H2O2]2 / M2

0.0 3.0x10-5 6.0x10-5 9.0x10-5 1.2x10-4

0

1000

2000

3000

4000

α·K

7

[H+] / M

Fig. 7. (a) ½MPO-II–HO2 ��½Hþ�½MPO0 ��½MPO-II–HO2 � plotted versus [H2O2]2. Values of [MPO-II–

HO2] were calculated from the data in Figs. 2 and 3. Conditions are thesame as in these figures. (b) Slopes from (a) plotted against [H+].


Depending on the relative values of k�7 and k8 two lim-its can be considered. First limit implies k�7� k8, thus tak-ing into account the values of K 07 (0.77 M) and k7

(3.3 · 104 M�1 s�1) [15], the maximum value for k8 can becalculated as 4.29 · 104 M�1 s�1, significantly lower than106 M�1 s�1 proposed in the literature [34]. The latter valuewould imply a measurable loss of hydrogen peroxide out-side the chlorination cycle, which is contrary to our find-ings. Therefore, the previously reported value should beused with some caution.

The other limit, k�7� k8 and K 07 � K7, means that theequilibrium in Reaction 7 (Table 1) is thermodynamicallyslightly unfavorable, which is in agreement with therequired excess of H2O2 to drive out the formation ofMPO-II [15]. Thus, assuming that the re-formation ofMPO-I is faster than the formation of Fe(III)–MPO, i.e.

k�7� k8, and considering that MPO-II accumulation isthe only responsible for the slow rate under equilibriumconditions, we can calculate the fraction of chlorinating-active enzyme (here defined as the concentration of allenzymatic species but MPO-II) and, therefore, [MPO-II]at particular pH and [H2O2] conditions:

½MPO0� � ½MPO-II�½MPO0�

¼ veq

v0;calc

ð7Þ

where v0,calc is the estimated value of the initial rate, calcu-lated from fittings shown in Fig. 2a at the correspondingpH and [H2O2].

The one-electron oxidation of Reaction 7 produces HO�2radicals [14,15] which have been described to act as elec-tron donors of MPO-II [35]. According to this equilibriumwe obtain:

K7 ¼½MPO-II� � ½HO�2� � ½Hþ�½MPO-I� � ½H2O2�

ð8Þ

Kinetic runs in presence of Cu/Zn superoxide dismutasedid not reveal any effect on reaction rate, suggesting thatsuperoxide is not released but rapidly reacts in heme vicin-ity upon rebinding to MPO-II (MPO-II–HO2), thus Eq. (8)rearranges to:

K7 ¼½MPO-II–HO2� � ½Hþ�½MPO-I� � ½H2O2�

;

½MPO-II–HO2� ¼K7 � ½MPO-I� � ½H2O2�

½Hþ� ð9Þ

Since the initial amount of enzyme is regained as Fe(III)–MPO following H2O2 consumption (Fig. 4), the formationof this complex must be reversible and it does not lead toMPO permanent inhibition.

Analyzing in detail all reactions involved in valid Mech-anism II, we can extract the following dependences. Forevery pH value, [MPO-I] must be proportional to the frac-tion of chlorinating-active enzyme ([MPO0] � [MPO-II–HO2]) and, according to Reaction 1, the extent of this pro-portionality depends on the actual [H2O2]. Thus, weobtain:

½MPO-I� ¼ a � ½H2O2� � ð½MPO0� � ½MPO-II–HO2�Þ ð10Þ

where a is the proportionality constant. Eq. (9) changes to:

½MPO-II–HO2� � ½Hþ�½MPO0� � ½MPO-II–HO2�

¼ a � K7 � ½H2O2�2 ð11Þ

The left term of Eq. (11) versus [H2O2]2 is plotted in Fig. 7a.Straight lines were obtained for each pH, with the sloperepresenting a Æ K7, which indicates the correctness of thisapproach. The effect of protonation equilibria of Fe(III)–MPO and MPO-I (Reactions 4 and 5) on a can also bestudied. Thus, for every pH value, [MPO-I] is proportionalto the fraction of unprotonated active enzyme (all MPOspecies minus MPO-II and excluding all other protonatedcomplexes, i.e. [MPO0] � [MPO-II–HO2] � [MPO–H] �[MPO-I–H] � [MPO–H–Cl]. The potential protonatedMPO-I–Cl–H complex is not considered, as its concentra-tion should be negligible). Besides, [MPO-I] is inverselyproportional to [H+] according to Reaction 5:

½MPO-I�

¼ b � ½H2O2� � ð½MPO0� � ½MPO-II–HO2� � ½MPO–H� � ½MPO-I–H� � ½MPO–H–Cl�Þ½Hþ�

ð12Þ

Applying a similar approach to all protonated species, i.e.

[MPO–H], [MPO-I–H] and [MPO–H–Cl] must also be


proportional to ([MPO0] � [MPO-II–HO2]) and—accord-ing to Reactions 4 and 5—to [H+], Eq. (12) can be written:

½MPO-I� ¼ b � ½H2O2� � ð½MPO0� � ½MPO-II–HO2�Þ � ð1� c � ½Hþ�Þ½Hþ�

ð13Þ

with b and c representing the proportionality constants of[MPO-I] and concentration of protonated species, respec-tively. Comparing this expression with Eq. (10), the follow-ing relationship is obtained:

a ¼ b � ð1� c � ½Hþ�Þ½Hþ� ð14Þ

The slopes from Fig. 7a (a Æ K7) are plotted against [H+] inFig. 7b, and the data fitted to Eq. (15):

y ¼ a� b � ½Hþ�½Hþ� ð15Þ

Considering the different described approximations, Eq.(15) reasonably fits the data. This supports the proposedmechanism and the assumptions made to derivate expres-sions for the rate at equilibrium.

Thus the current data show that formation of MPO-II isthe only process that affects the chlorination rate at equilib-rium, decelerating it significantly compared to the initialrate. Here, for the first time, quantitative kinetic data areprovided that explain this phenomenon. Under workingconditions for reliable kinetic measurements (pH 4–6),the hydroperoxide radical produced by MPO-I-mediatedH2O2 oxidation (Reaction 7 in Table 1) does not leavethe heme cavity, as pointed out by the null effect on Tauchlorination rate when superoxide dismutase is added.Since there was no noticeable loss of H2O2 through perox-idative processes like MPO-II reduction by HO�2 to Fe(III)–MPO (Reaction 8 in Table 1), it is reasonable to assumethat the equilibrium described by Eq. (9) mostly determinesthe ratio [MPO-I]/[MPO-II], and thus the actual Tau chlo-rination rate at equilibrium.

4.04.5

5.05.5

6.0 0.0

0.40.8

1.21.6

pH[H 2

O 2] / mM

v eq

Fig. 8. Three-dimensional scheme showing the dependence of Tauchlorination rate by MPO on pH and [H2O2] at equilibrium (in vivo

conditions). Conditions are the same as in Fig. 3a.

Summary

In conclusion, we have demonstrated that kinetic studieson complex enzymatic mechanisms can be performed atdifferent stages of the process, using the same kinetic traces.In the particular case of human myeloperoxidase, the ratedependence on pH and [H2O2] shows a different behaviorfor initial rates and for rates at equilibrium regime. Despitethese data differ from previously published results [10,13], achlorination pathway via the complex MPO-I–Cl is alsosupported by the present findings. It is important to noticethat the present data also support the possibility of a freechlorinating agent that is released unprotonated (ClO�)from the active center and undergoes protonation to HOCloutside the heme cavity (Reactions 2 0 and/or 9 and 11 inFig. 6). In this case, HOCl is the actual chlorinating agent.On the other hand, actual results reject mechanisms similar

to Mechanism I, which involve formation of the strong oxi-dizing agent HOCl inside the heme cavity by reaction ofCl� and MPO-I with net consumption of one acid equiva-lent (Reaction 2 in Table 1). The effect of pH on kineticrates show that the produced halogenating species—atleast, prior to and at the rate-limiting step—must beunprotonated, either MPO-I–Cl or ClO�, which are kinet-ically indistinguishable. Therefore, determination of fac-tual chlorinating agent should be carried out by chemicalor biological means other than bare kinetic studies.

Evidence for the reversibility of the reaction of MPO-Iand H2O2 to yield MPO-II has been also provided bothmathematically and by the null consumption of H2O2 viapathways other than Tau chlorination. The slow chlorina-tion rate under equilibrium conditions could be quantita-tively ascribed to the chemical equilibrium betweenMPO-I and MPO-II. Still, these data do not permit us tocompletely reject the possibility of MPO turnover via Reac-tion 8, but they evidence that this pathway must only rep-resent a minor contribution on the overall process.Furthermore, the absence of effect of addition of Cu/Znsuperoxide dismutase on the reaction rate indicates thatHO�2 radicals are not released from the active site into thebulk medium, which supports the proposal of an enzymaticcomplex MPO-II–HO2.

Fig. 8 provides a visual comprehension of the behaviorat equilibrium. Both at high and low [H2O2], as well as inthe ends of the employed pH range, chlorination keeps ata low rate, whereas the maximum rate is obtained at[H2O2] and pH values around 0.4 mM and pH 5, respec-tively. These results are in agreement with the work of Pan-asenko and coworkers who found chlorination activity ofMPO to be optimal at acidic pH values, while its peroxida-tive effect takes place around neutral pH [36,37]. Simula-tion of the MPO activity in the neutrophil phagosome at


pH 7.4 supports the relevance of both processes near neu-tral pH [34].

Acknowledgments

Authors thank Dr. Christine C. Winterbourn for usefulcomments and fruitful discussion on the reactivity of MPO.Thanks are also given to the reviewers for their helpfulcomments. This work was supported by Austrian Sciences

Foundation FWF No. P15660. D.R.R. thanks the SpanishMinisterio de Educacion, Cultura y Deporte for a Ph.D.F.P.U. grant and for supporting visits to the BOKU-Uni-versity of Natural Resources and Applied Life Sciences.D.R.R., M.C.L., M.V.G., and J.A.S. thank research fundsprovided by the Ministerio de Educacion y Cultura of theSpanish Government (DGICYT project BQU2003-04168-C03), and by the Xunta de Galicia (Project PGIDIT05PXIC10306PN).

Appendix

Derivation of the rate equations for Mechanisms I andII is shown.

Mechanism I

A. Initial rate

Reaction rate expression from Reaction 3:

v0 ¼d½ðN -ClÞ–Tau�

dt¼ k3 � ½HOCl� � ½Tau� ð1AÞ

Mass balance equation:

½MPO0� ¼ ½MPO� þ ½MPO-I� þ ½MPO–H�þ ½MPO-I–H� þ ½MPO–H–Cl� ð2AÞ

Equilibrium equation for protonation of MPO:

K4 ¼½MPO–H�½MPO� � ½Hþ� ;

½MPO–H� ¼ K4 � ½MPO� � ½Hþ� ð3AÞ

Equilibrium equation for proton binding onto MPO-I:

K5 ¼½MPO-I–H�½MPO-I� � ½Hþ� ;

½MPO-I–H� ¼ K5 � ½MPO-I� � ½Hþ� ð4AÞ

Equilibrium equation for binding of chloride:

K6 ¼½MPO–H–Cl�½MPO–H� � ½Cl�� ;

½MPO–H–Cl� ¼ K6 � ½MPO–H� � ½Cl�� ð5AÞ

Substituting from Eqs. (3A)–(5A) into (2A):

½MPO0� ¼ ½MPO� � ð1þ K4 � ½Hþ� � ð1þ K6 � ½Cl��ÞÞþ ½MPO-I� � ð1þ K5 � ½Hþ�Þ ð6AÞ

Equilibrium equation for MPO-I formation:

K1 ¼½MPO-I�

½MPO� � ½H2O2�; ½MPO� ¼ ½MPO-I�

K1 � ½H2O2�ð7AÞ

Substituting from Eq. (7A) into (6A):

½MPO0� ¼1þ K4 � ½Hþ� � ð1þ K6 � ½Cl��Þ

K1 � ½H2O2�þ 1þ K5 � ½Hþ�

� �

� ½MPO-I�ð8AÞ

Steady-state approximation for HOCl:

d½HOCl�dt

¼ 0; k2 � ½MPO-I� � ½Cl�� ½Hþ� ¼ k3 � ½HOCl� � ½Tau�

ð9AÞ

From Eqs. (8A) and (9A), Eq. (1A) is converted to:

v0 ¼ k2 � ½MPO-I� � ½Cl�� ½Hþ�

¼ k2 � ½MPO0� � ½Cl�� ½Hþ�1þK4�½Hþ��ð1þK6�½Cl��Þ

K1�½H2O2�þ 1þ K5 � ½Hþ�

ð10AÞ

Considering:

1þ K4 � ½Hþ� � ð1þ K6 � ½Cl��ÞK1 � ½H2O2�

� 1þ K5 � ½Hþ� ð11AÞ

which is consistent with the experimental initial rate depen-dence on the H2O2 concentration (Fig. 2a).

Initial rate expression becomes:

v0 ¼K1 � k2 � ½MPO0� � ½H2O2� � ½Cl�� ½Hþ�

1þ K4 � ½Hþ� � ð1þ K6 � ½Cl��Þ ð12AÞ

B. Rate at equilibrium

Under equilibrium and according to the hypothesisabout HO�2 described in the paper, conservation equationchanges to:

½MPO0� ¼ ½MPO� þ ½MPO-I� þ ½MPO–H� þ ½MPO-I–H�þ ½MPO–H–Cl� þ ½MPO-II–HO2� ð13AÞ

Variation of [MPO-II–HO2] at equilibrium occurs veryslowly, thus steady-state approximation can be applied:

d½MPO-II–HO2�dt

¼ 0;

k7 � ½MPO-I� � ½H2O2� ¼ ðk�7 þ k8Þ � ½MPO-II–HO2� � ½Hþ�ð14AÞ

Eq. (14A) converts to:

½MPO-II–HO2� ¼k7 � ½MPO-I� � ½H2O2�ðk�7 þ k8Þ � ½Hþ�

¼ K 07 � ½MPO-I� � ½H2O2�½Hþ� ð15AÞ

Substituting from Eqs. (3A)–(5A) and (15A) into Eq.(13A):


½MPO0� ¼ ½MPO� � ð1þ K4 � ½Hþ� � ð1þ K6 � ½Cl��ÞÞ

þ ½MPO-I� � 1þ K5 � ½Hþ� þK 07 � ½H2O2�½Hþ�

� �

ð16AÞ

From Eqs. (7A), (9A) and (16A), rate expression 1A is con-verted to:

veq ¼ k2 � ½MPO-I� � ½Cl�� ½Hþ�

¼ k2 � ½MPO0� � ½Cl�� ½Hþ�1þK4�½Hþ��ð1þK6�½Cl��Þ

K1�½H2O2� þ 1þ K5 � ½Hþ� þK 0

7�½H2O2�½Hþ�

ð17AÞ

Mechanism II

A. Initial rate

Mass balance equation:

½MPO0� ¼ ½MPO� þ ½MPO-I� þ ½MPO–H� þ ½MPO-I–H�þ ½MPO–H–Cl� þ ½MPO-I–Cl� ð18AÞ

Applying steady-state approximation for MPO-I–Cl, con-sidering that [MPO-I–Cl] must be negligible (similarly asHOCl, this compound should be a strong oxidant that re-acts very rapidly with Tau) and rearranging, the initial rateequation becomes:

v0 ¼K1 � k2b � ½MPO0� � ½H2O2� � ½Cl��1þ K4 � ½Hþ� � ð1þ K6 � ½Cl��Þ ð19AÞ

B. Rate at equilibrium

The preceding rate expression is converted to:

veq ¼k2b � ½MPO0� � ½Cl��

1þK4�½Hþ��ð1þK6�½Cl��ÞK1�½H2O2�

þ 1þ K5 � ½Hþ� þK 0

7�½H2O2�½Hþ�

ð20AÞ

References

[1] S.J. Klebanoff, J. Leukoc. Biol. 77 (2005) 598–625.[2] A. Taurog, Biochimie 81 (1999) 557–562.[3] J. Arnhold, P.G. Furtmuller, C. Obinger, Redox Rep. 8 (2003) 179–

186.[4] E.L. Thomas, D.B. Learn, in: J. Everse, K.E. Everse, M.B. Grisham

(Eds.), Peroxidases in Chemistry and Biology, CRC Press, Boston,1991, pp. 83–104.

[5] M.B. Grisham, M.M. Jefferson, D.F. Melton, E.L. Thomas, J. Biol.Chem. 259 (1984) 10404–10413.

[6] K. Fukuda, Y. Hirai, H. Yoshida, T. Nakajima, T. Usui, Clin. Chem.28 (1982) 1758–1761.

[7] E.L. Thomas, Infect. Immun. 25 (1979) 110–116.

[8] E.L. Thomas, M.B. Grisham, D.F. Melton, M.M. Jefferson, J. Biol.Chem. 260 (1985) 3321–3329.

[9] L.K. Folkes, L.P. Candeias, P. Wardman, Arch. Biochem. Biophys.323 (1995) 120–126.

[10] L.A. Marquez, H.B. Dunford, J. Biol. Chem. 269 (1994) 7950–7956.

[11] D.W. Margerum, E.T. Gray, R.P. Huffman, in: F.E. Brinckman,J.M. Bellama (Eds.), Organometals and Organometalloids: Occur-rence and Fate in the Environment, American Chemical Society,Washington, DC, 1978, pp. 278–291.

[12] J.C. Morris, in: S.D. Faust, J.V. Hunter (Eds.), Principles andApplications of Water Chemistry, John Wiley & Sons Inc., NewYork, 1967, pp. 23–53.

[13] H.B. Dunford, L.A. Marquez-Curtis, Methods Enzymol. 354 (2002)338–350.

[14] B.G.J.M. Bolscher, R. Wever, Biochim. Biophys. Acta 788 (1984) 1–10.

[15] L.A. Marquez, J.T. Huang, H.B. Dunford, Biochemistry 33 (1994)1447–1454.

[16] A.R.J. Bakkenist, J.E.G. De Boer, H. Plat, R. Wever, Biochim.Biophys. Acta 613 (1980) 337–348.

[17] P.C. Andrews, N.I. Krinsky, J. Biol. Chem. 257 (1982) 13240–13245.

[18] M. Ikeda-Saito, J. Biol. Chem. 260 (1985) 11688–11696.[19] P.G. Furtmuller, U. Burner, C. Obinger, Biochemistry 37 (1998)

17923–17930.[20] A.J. Kettle, C.C. Winterbourn, Biochemistry 40 (2001) 10204–

10212.[21] D.P. Nelson, L.A. Kiesow, Anal. Biochem. 49 (1972) 474–478.[22] T. Odajima, I. Yamazaki, Biochim. Biophys. Acta 206 (1970) 71–77.[23] D.N. Baron, S.A. Ahmed, Clin. Sci. 37 (1969) 205–219.[24] M. Ikeda-Saito, Y. Dou, T. Yonetani, J.S. Olson, T. Li, R. Regan,

Q.H. Gibson, J. Biol. Chem. 268 (1993) 6855–6857.[25] P.G. Furtmuller, C. Obinger, Y. Hsuanyu, H.B. Dunford, Eur. J.

Biochem. 267 (2000) 5858–5864.[26] H.C. Lee, K.S. Booth, W.S. Caughey, M. Ikeda-Saito, Biochim.

Biophys. Acta 1076 (1991) 317–320.[27] M. Blair-Johnson, T. Fiedler, R. Fenna, Biochemistry 40 (2001)

13990–13997.[28] M. Ikeda-Saito, H.C. Lee, K. Adachi, H.S. Eck, R.C. Prince, K.S.

Booth, W.S. Caughey, S. Kimura, J. Biol. Chem. 264 (1989) 4559–4563.

[29] P.G. Furtmuller, U. Burner, W. Jantschko, G. Regelsberger, C.Obinger, Redox Rep. 5 (2000) 173–178.

[30] J.A. Santaballa, H. Maskill, M. Canle L., in: H. Maskill (Ed.),Investigation of Organic Reactions and their Mechanisms, BlackwellPublishing, Singapore, 2006, pp. 79–103.

[31] H.A. Wagenknecht, W.D. Woggon, Chem. Biol. 4 (1997) 367–372.[32] J.C. Morris, J. Phys. Chem. 70 (1966) 3798–3805.[33] T.J. Fiedler, C.A. Davey, R.E. Fenna, J. Biol. Chem. 275 (2000)

11964–11971.[34] C.C. Winterbourn, M.B. Hampton, J.H. Livesey, A.J. Kettle, J. Biol.

Chem. 281 (2006) 39860–39869.[35] A.J. Kettle, C.C. Winterbourn, Biochem. J. 252 (1988) 529–536.[36] I.I. Vlasova, J. Arnhold, A.N. Osipov, O.M. Panasenko, Biochem-

istry-Moscow 71 (2006) 667–677.[37] H. Spalteholtz, O.M. Panasenko, J. Arnhold, Arch. Biochem.

Biophys. 445 (2006) 225–234.




Myeloperoxidase-catalyzed taurine chlorination: initial versus equilibrium rate

Documents

Transcript of Myeloperoxidase-catalyzed taurine chlorination: initial versus equilibrium rate