Antioxidant chemistry: Oxidation of L-cysteine and its metabolites by chlorite and chlorine dioxide

Article abstract of DOI:10.1021/jp049748k

The oxidation of L-cysteine and its metabolites cystine and L-cysteinesulfinic acid by chlorite and chlorine dioxide has been studied in unbuffered neutral and slightly acidic media. The stoichiometry of the oxidation of L-cysteine was deduced to be 3ClO₂^- + 2H ₂NCH(COOH)CH₂SH → 3Cl^- + 2H ₂NCH(COOH)CH₂SO₃H with the final product as cysteic acid. The stoichiometry of the chlorite-cysteinesulfinic acid gave a ratio of 1:2, ClO₂^- + 2H₂NCH(COOH)CH ₂SO₂H → Cl^- + 2H₂NCH(COOH) CH₂SO₃H. There was no further oxidation past cysteic acid, and there was no evidence of sulfate formation which would have indicated the cleavage of the carbon-sulfur bond. The reaction is oligooscillatory in chlorine dioxide formation. In conditions of excess oxidant, the reaction is characterized by a short induction period followed by a rapid and autocatalytic formation of chlorine dioxide. Chlorine dioxide is formed by the reaction of intermediate HOCl with the excess chlorite: 2ClO₂^- + 2HOCl + H^- → 2ClO₂(aq) + Cl^- + H₂O. Oligooscillations observed in chlorine dioxide formation result from the competition between this pure oxyhalogen reaction and reactions that consume chlorine dioxide. The rate of the reaction of chlorine dioxide with cysteine and its metabolites is fast and is of comparable magnitude with the reactions that form chlorine dioxide. The reaction of chlorine dioxide with L-cysteine is first order in both oxidant and substrate, retarded by acid, and has a lower-limit bimolecular rate constant of 405 ± 50 M^-1 s^-1, while for the reaction with L-cysteinesulfinic acid the rate constant is 210 ± 15 M^-1 s^-1. It would appear that the existence of a zwitterion on the asymmetric carbon atom precludes the formation of N-chloramines as has been observed with taurine and aminomethanesulfonic acid. The mechanism for the reaction is satisfactorily described by a network of 28 elementary reactions which include autocatalysis by HOCl.

Full text of DOI:10.1021/jp049748k

5576
J. Phys. Chem. A 2004, 108, 5576-5587
Antioxidant Chemistry: Oxidation of L-Cysteine and Its Metabolites by Chlorite and
Chlorine Dioxide
James Darkwa*
Department of Chemistry, UniVersity of the Western Cape, BellVille 7535, South Africa
Rotimi Olojo, Edward Chikwana, and Reuben H. Simoyi*
Department of Chemistry, Portland State UniVersity, P.O. Box 751, Portland, Oregon 97207-0751
ReceiVed: January 17, 2004; In Final Form: April 27, 2004
The oxidation of L-cysteine and its metabolites cystine and L-cysteinesulfinic acid by chlorite and chlorine
dioxide has been studied in unbuffered neutral and slightly acidic media. The stoichiometry of the oxidation
of L-cysteine was deduced to be 3ClO2- + 2H2NCH(COOH)CH2SH f 3Cl- + 2H2NCH(COOH)CH2SO3H
with the final product as cysteic acid. The stoichiometry of the chlorite-cysteinesulfinic acid gave a ratio of
1:2, ClO2- + 2H2NCH(COOH)CH2SO2H f Cl- + 2H2NCH(COOH)CH2SO3H. There was no further oxidation
past cysteic acid, and there was no evidence of sulfate formation which would have indicated the cleavage
of the carbon-sulfur bond. The reaction is oligooscillatory in chlorine dioxide formation. In conditions of
excess oxidant, the reaction is characterized by a short induction period followed by a rapid and autocatalytic
formation of chlorine dioxide. Chlorine dioxide is formed by the reaction of intermediate HOCl with the
excess chlorite: 2ClO2- + 2HOCl + H+ f 2ClO2(aq) + Cl- + H2O. Oligooscillations observed in chlorine
dioxide formation result from the competition between this pure oxyhalogen reaction and reactions that consume
chlorine dioxide. The rate of the reaction of chlorine dioxide with cysteine and its metabolites is fast and is
of comparable magnitude with the reactions that form chlorine dioxide. The reaction of chlorine dioxide with
L-cysteine is first order in both oxidant and substrate, retarded by acid, and has a lower-limit bimolecular rate
constant of 405 ( 50 M-1 s-1, while for the reaction with L-cysteinesulfinic acid the rate constant is 210 (
15 M-1 s-1. It would appear that the existence of a zwitterion on the asymmetric carbon atom precludes the
formation of N-chloramines as has been observed with taurine and aminomethanesulfonic acid. The mechanism
for the reaction is satisfactorily described by a network of 28 elementary reactions which include autocatalysis
by HOCl.
Introduction
needed to prevent the formation and oppose the actions of
reactive oxygen species which are generated in vivo and cause
damage to DNA, lipids, and proteins.8,9 It would appear quite
straightforward to attempt to mechanistically characterize the
antioxidant effects of aminothiols, but so far nothing conclusive
has been obtained. There are several other areas in physiological
processes where thiols have been implicated. It is assumed that
one of their most important roles is the forming and breaking
of S-S bonds, especially in the endoplasmic reticulum (ER).10
Protein folding in the ER often involves the formation of
disulfide bonds. The oxidizing conditions required in the ER
are maintained through the release of small thiols, mainly
cysteine and glutathione.11
Living systems need sulfur for viability. In seawater (as well
as the soil), sulfur is available as sulfate, and this sulfate is the
source of sulfur for the synthesis of sulfur-containing amino
acids found in proteins, e.g., cysteine and methionine.1 Green
plants use sulfate as a source of sulfur for their own biosyn-
thesis.2 Animals must find sulfur in a reduced organic condition
in their food in order to synthesize proteins.2 Deficiency of
sulfur-containing amino acids can result in anaemia and necrosis
of the liver and kidneys.3,4 Animals cannot incorporate sulfate
in proteins because they are unable to reduce it. Nearly all sulfur-
type reactions in animals are oxidative, especially S-oxygen-
ation. The major exception is the sulfide-disulfide equilibrium.
All sulfur found in animal proteins comes from the S-H or
S-S of the sulfur-containing amino acids.5
Our research group established a series of studies aimed at
elucidating the kinetics and mechanisms of the interactions of
oxyhalogen ions with small organic sulfur molecules.12 Cysteine
is one of these organosulfur molecules we have studied.13
Cysteine is a nonessential amino acid which can be synthesized
in the human body by the metabolism of methionine. By
containing sulfur, cysteine can bond in a special way to maintain
protein structure in the body. Cysteine is, however, only incor-
porated into proteins at the rate of 2.8% relative to the other
amino acids, but the unique thiol side chain of this amino acid
is often heavily involved in the three-dimensional stability of
proteins and enzymes.14 The side chain is also often involved
in the chemistry occurring at the active sites of many enzymes.
Biological thiols are products of sulfur metabolism.6 The most
important thiols in biological chemistry are cysteine, homocys-
teine, and glutathione. Thiols without an amino group adjacent
the sulfhydryl group are easily oxidized to sulfate and a mixture
of alcohols, aldehydes, and carboxylic acids (depending upon
the amount and strength of oxidant). R-Aminothiols are very
reactive, but will not easily cleave the C-S bond, and can thus
be regenerated for further use after oxidation to the sulfenic
acid or the disulfide. These thiols are strongly implicated as
antioxidants in human health although the mechanistic basis for
such assertions is not yet firmly established.7 Antioxidants are
10.1021/jp049748k CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/29/2004

Oxidation of Cysteine and Its Metabolites
J. Phys. Chem. A, Vol. 108, No. 26, 2004 5577
SCHEME 1
Cysteine is also critical to the metabolism of a number of
essential biochemicals including coenzyme A, heparin, biotin,
lipoic acid, and glutathione.15
Chlorine dioxide was prepared by the standard method of
oxidizing sodium chlorate in a sulfuric acid/oxalic acid mix-
ture.24 The stream was passed through a sodium carbonate
solution before being collected in ice-cold water at 4 °C at a
pH of ∼3.5.
Scheme 1 shows all possible reactions of a generic thiol in
the human body. A number of studies have been done on process
B;16 but none to our knowledge, except one of our previous
studies,13 have been done on the S-oxygenation pathways A,
C, D, E, F, and G. There is some healthy debate as to whether
process B or D is the most important in the sulfide-disulfide
equilibrium.17 While, in general, sulfenic acids are rarely stable
enough to be isolated,18 sulfinic and sulfonic acids are stable
enough, especially if the thiol has an amino group at the R or
â carbon position.19 We intend to report, in this article, the
oxidation of L-cysteine, cystine, and L-cysteinesulfinic acid by
chlorite ions. This study should enable us to evaluate the
physiologically relevant reaction steps A + C + E + F; E + F
and G + J (or K). This relevancy arises from the fact that
physiologically, cysteine is catabolized to taurine and that in
the presence of the standard P450-type enzymes, one of the
major metabolites has been sulfate.20 The major oxidizing
species in chlorite oxidations is HOCl. It is known that
myeloperoxidase and eosinoperoxidase catalyze the oxidation
of chloride ions by H2O2 to produce HOCl21 which is then
deactivated by several antioxidants present in the physiological
medium such as cysteine, glutathione, and taurine. The mech-
anism of the oxidation of these aminothiols by oxyhalogens has
not yet been studied.
2NaClO3(s) + H2C2O4‚2H2O(s) + 2H2SO4(aq) f
2ClO2(g) + 2CO2(g) + 4H2O + 2NaHSO4
Standardization of ClO2 was also accomplished by iodometric
techniques through addition of excess acidified potassium iodide
and back-titration of the liberated iodine against standard sodium
thiosulfate using the following stoichiometry:
2ClO2 + 10I- + 8H+ f 5I2 + 2Cl- + 4H2O
The results obtained were confirmed spectrophotometrically by
using the absorptivity coefficient of ClO2 of 1265 M-1 cm-1 at
360 nm.
Tests for Adventitious Metal Ion Catalysis. Water used for
preparing reagent solutions was obtained from a Barnstead
Sybron Corporation water purification unit capable of producing
both distilled and deionized water (Nanopure). Not much
difference was observed in the general reaction kinetics observed
with distilled and deionized water. We utilized inductively
coupled plasma mass spectrometry (ICPMS) to quantitatively
evaluate the concentrations of a number of metal ions in the
water used for our reaction medium. ICPMS analysis showed
negligible concentrations of iron, copper, and silver and
approximately 1.5 ppb of cadmium and 0.43 ppb of lead. The
use of chelators to sequester metal ions gave kinetics and
reaction dynamics indistinguishable from those run in deionized
water and slightly slower kinetics than those from singly distilled
water. The addition, however, of 1.0 µM Cu2+ ions showed a
dramatic increase in rate of reaction, proving the acknowledged
copper catalysis in reactions involving organosulfur compounds
and also that our reaction media did not contain enough metal
ions to affect the overall reaction kinetics and mechanisms.
Methods. Stoichiometric determinations were carried out by
mixing various oxidant/reductant ratios in stoppered volumetric
flasks and scanning them spectrophotometrically for ClO2
activity over a period of 24 h. The products formed were
characterized by 1H NMR measurements using D2O as solvent
and internal standard (4.67 ppm). Reaction kinetics were
followed on a Hi-Tech Scientific DX2 double-mixing stopped-
flow spectrophotometer. Absorbance traces were obtained by
following either the appearance or consumption of ClO2 at 360
nm. All measurements were carried out at 25.0 ( 0.5 °C with
ionic strength maintained at 1.0 M using NaClO4. Qualitative
Experimental Section
Materials. The following analytical grade chemicals were
purchased from Fisher Scientific and used without further
purification: sodium chlorate, oxalic acid, sodium carbonate,
perchloric acid (70%), potassium iodide, hydrochloric acid,
sodium thiosulfate, starch, and sulfuric acid. L-Cysteine, cystine,
L-cysteinesulfinic acid, and L-cysteic acid were purchased in
analytical reagent grade quality from Sigma-Aldrich. Sodium
chlorite (Sigma-Aldrich) was recrystallized from its ca. 80%
technical grade purity to ∼99% assay. This was carried out from
a water-ethanol-acetone mixture, and the crystals that formed
were dried over a period of 1 week in a desiccator. The
recrystallized chlorite was standardized iodometrically by adding
acidified potassium iodide and titrating the liberated iodine
against sodium thiosulfate using freshly prepared starch as
indicator and employing the following stoichiometries:22,23
ClO2- + 4I- + 4H+ f 2I2 + Cl- + 2H2O
I2 + 2S2O3 f 2I- + S4O6
-2
2-

5578 J. Phys. Chem. A, Vol. 108, No. 26, 2004
Darkwa et al.
CHART 1: Structure of Cysteine Showing the Chiral
tests for the presence of sulfate were performed using BaCl2
for sulfate precipitation in the presence of excess reductant. The
stoichiometries of reactions involving chlorine dioxide were
determined by pure titration of the oxidant into excess substrate
with starch as indicator. Starch prepared with mercuric iodide
as a preservative gave a deep blue-black color with excess
chlorine dioxide.
Center and the Diastereotopic Methylene Protons
Results
spectroscopy was used to conclusively identify products of the
reaction. Figure 1, spectrum A is from pure L-cysteine in nearly
neutral pH conditions. It shows what appears to be a triplet
centered around 4.0 ppm from the asymmetric proton HA. Close
examination of this peak shows that it is a split doublet which
is expected from the effect of the diastereotopic HB and HC
protons (see Chart 1). Spectrum A also shows the complex
doublet of a doublet split integrating as one proton each arising
from the adjacent methylene protons HB and HC. In this
L-isomer, HB is the proton centered further upfield at 3.04 ppm
and HC is centered at 3.12 ppm. This splitting is expected due
to the dissimilar environments protons HB and HC will experi-
ence due to proton HA at the asymmetric center. Spectrum B is
from the reaction mixture with excess chlorite, which should
give stoichiometry R1. The spectrum shows the same types of
protons as in spectrum A; but all are shifted downfield due to
the formation of the more electron-withdrawing sulfonic acid
group in cysteic acid. The coupling constants of protons HB
and HC increase due to the steric hindrance brought about by
the bulky sulfonic acid moiety which renders protons HB (3.45
ppm) and HC (3.62 ppm) more dissimilar. This increase in
coupling constants now makes it clearer that the peak appearing
at 4.37 ppm is a split doublet and not a triplet. Spectrum C is
the product of excess chlorite with cysteinesulfinic acid. It shows
the same type of spectrum as that observed in spectrum B,
meaning that in both reactions the same product is obtained. A
spectrum of pure cysteic acid (Fisher Scientific) was indistin-
guishable from that of spectrum C, indicating that it was the
oxidation product in both cases. The reactions utilized to obtain
spectra B and C were not buffered, and thus the products in
spectra B and C were not at the same pH. The pH of the
medium, due to the formation of the zwitterion on the asym-
metric carbon, was extremely important in determining the
position of all the protons in the spectrum. Spectrum C was
taken at a lower pH than spectrum B due to the expected
increase in acid strength in going from cysteinesulfinic acid to
cysteic acid. This explains the small differences observed
between spectra B and C. The peak observed at 4.6 ppm in
spectrum C is due to an impurity that is generated during the
commercial production of L-cysteinesulfinic acid.
pH experiments were also performed to prove that an acidic
product was formed. Figure 2 shows a pH trace of an unbuffered
experimental reaction solution. This experiment shows that there
is an immediate and rapid formation of acid upon addition of
chlorite. The final pH of the solution could be rationalized from
the stoichiometry after allowing for the acid dissociation constant
of cysteic acid.
Reaction Dynamics. In stoichiometric amounts of oxidant
to substrate, the reaction shows a short induction period followed
by a rapid formation of chlorine dioxide which is next followed
by its consumption until it is all depleted. However, in excess
chlorite with ratios ([ClO2-]0/[cysteine]0 ) R) greater than 5
and in unbuffered pH conditions, the reaction displays immediate
and monotonic formation of chlorine dioxide (see trace e of
Figure 3). At lower ratios chlorine dioxide concentrations reach
a transient peak followed by a decrease which gives way to a
Stoichiometry. The stoichiometry of reaction between chlo-
rite and L-cysteine was determined as
3ClO2- + 2H2NCH(COOH)CH2SH f
3Cl- + 2H2NCH(COOH)CH2SO3H (R1)
This stoichiometry was deduced as the highest chlorite/cysteine
ratio that could be used without the production of chlorine
dioxide after an incubation period of more than 24 h. This ratio
was later reconfirmed by titrimetric techniques in which the
excess oxidizing power was evaluated in excess oxidant
conditions. Cysteic acid was the highest oxidation product
formed in all oxidations. The same technique was utilized for
the reaction of chlorite with cysteinesulfinic acid and the
stoichiometry was evaluated as
ClO2- + 2H2NCH(COOH)CH2SO2H f
Cl- + 2H2NCH(COOH)CH2SO3H (R2)
No sulfate production was observed in both reactions, indicating
that the C-S bond was not cleaved during the oxidation and
that the sulfur center never attains the oxidative saturation state
of +6. The chlorite-cystine reaction also gave cysteic acid as
the final oxidation product with the stoichiometry
5ClO2- + 2(H2NCH(COOH)CH2S)2 + 2H2O f
5Cl- + 4H2NCH(COOH)CH2SO3H (R3)
Chlorine dioxide oxidations did not give clean and sharp
stoichiometries because of the volatility of aqueous chlorine
dioxide. This was especially so in reactions that involved long
induction periods and in titrations performed in open containers.
Chlorine dioxide oxidation kinetics were so rapid that this
volatility was not significant for all our kinetics measurements.
Chlorine dioxide also oxidized cysteine to as far as cysteic acid
without the formation of sulfate:
6ClO2(aq) + 5H2NCH(COOH)CH2SH + 3H2O f
6Cl- + 5H2NCH(COOH)CH2SO3H + 6H+ (R4)
The reaction of chlorine dioxide with cysteinesulfinic acid gave
a stoichiometric ratio of 2:5 with the same cysteic acid product.
Product Identification. There were several possible products
that could arise from the oxidation of a thiol, and we had to
eliminate most of them to justify cysteic acid as the major
oxidation product. The stability of cysteinesulfinic acid also
suggested that it could be a possible product in the oxidation
of L-cysteine. The absence of sulfate as evidenced by the lack
of a precipitate with barium chloride clearly indicated that the
oxidation did not proceed to cleavage of the carbon-sulfur bond.
Previous experiments with other aminothiols had shown the
formation of chloramines (and bromamines) by the further
oxidation occurring at the amine center.25,26 Proton NMR

Oxidation of Cysteine and Its Metabolites
J. Phys. Chem. A, Vol. 108, No. 26, 2004 5579
Figure 1. (A) NMR spectrum of L-cysteinesulfinic acid in nearly neutral acidic conditions. It shows the expected split doublet from protons HB
and HC and the triplet from proton A. Each integrates as a single proton. (B) Spectrum from the products of cysteine and chlorite. Assignment: HA
(4.0 ppm); HB (3.04 ppm); HC 3.12 ppm). It shows basically the same spectrum as in A shifted slightly downfield due to the formation of the
sulfonic acid which accompanies the oxidation of cysteine. The relative assignments with respect to field for protons HA, HB, and HC in spectrum
A remain unchanged. The larger coupling constants can be attributed to the larger sulfonic acid group which renders protons HB and HC more
dissimilar. (C) NMR spectrum of the product of cysteinesulfinic acid with excess chlorite. The spurious peaks represent impurities produced in the
synthesis of the cysteinesulfinic acid (Fisher Scientific). However, both spectra show that there is no change in the carbon backbone. These spectra
were also indistinguishable from that of cysteic acid.
Figure 2. pH changes in the unbuffered chlorite-cysteine reaction
system. This pH profile agrees with the stoichiometry which gives
cysteic acid as the final product. The final pH obtained can be correlated
to the expected cysteic acid formed qualified by its acid dissociation
constant.
Figure 3. Variation of [ClO2-] in cysteine oxidation by chlorite
spanning chlorite/cysteine ratios of 4.0 to 7.0. High ratios show a
monotonic increase in chlorine dioxide without a transient peak.
final chlorine dioxide formation which terminates at the final
stoichiometric value (see traces a, b, and c in Figure 3). Figure
4 shows the effect of acid with all other initial conditions kept
constant. Higher acid concentrations appear to retard the initial
rapid formation of chlorine dioxide as well as its rate of
consumption after the transient peak. The final chlorine dioxide
concentrations in all the experiments shown in Figure 4 are the
same, but the rates of attaining this final value are affected by
acid. This effect of acid was a little surprising since most
oxyhalogen reactions are strongly catalyzed by acid.
[cysteine]0 ) 5 × 10-3 M, I ) 1.0 M. [ClO2-] ) (a) 0.02 M, (b) 0.025
M, (c) 0.0275 M, (d) 0.03 M, (e) 0.035 M.
Although acid did not seem to enhance the initial produc-
tion of chlorine dioxide, cysteine concentrations strongly
catalyze the initial rapid chlorine dioxide formation (Figure 5).
The data in Figure 5 show that at high cysteine concentra-
tions the initial formation of chlorine dioxide is so rapid
that it overshoots its stoichiometric value. The final stoichio-
metric concentration of chlorine dioxide in trace e is lower than

5580 J. Phys. Chem. A, Vol. 108, No. 26, 2004
Darkwa et al.
Figure 6. Variation of chlorite concentration in cysteine sulfinic acid
oxidation. [RCH2SO2H] ) 2.5 × 10-3 M, I ) 1.0 M. [ClO2-]0 ) (a)
0.01 M, (b) 0.0125 M, (c) 0.015 M, (d) 0.0175 M, (e) 0.02 M, (f)
0.0225 M, (g) 0.025 M.
Figure 4. Variation of [H+] in cysteine oxidation by chlorite. Low
acid concentrations catalyze transient formation of chlorine dioxide as
well as its rapid consumption. All traces shown here gave the same
final chlorine dioxide concentration. [cysteine]0 ) 5 × 10-3 M, [ClO2
]
0
-
) 0.025 M, I ) 1.0 M. [H+] ) (a) 0.015 M, (b) 0.045 M, (c) 0.075 M,
(d) 0.105 M, (e) 0.135 M.
Figure 7. Acid variation in the oxidation of cysteine sulfinic acid by
chlorite. Due to the high oxidant/reductant ratio, monotonic and nearly
pseudo-first-order kinetics are observed in the formation of chlorine
dioxide. [ClO2-] ) 0.025 M, [RCH2SO2H]0 ) 2.5 × 10-3 M, I ) 1.0
M. [H+]0 ) (a) no acid, (b) 3.0 × 10-3 M, (c) 6.0 × 10-3 M, (d) 9.0
× 10-3 M, (e) 1.2 × 10-2 M, (f) 1.5 × 10-2 M.
Figure 5. Variation of [cysteine] in cysteine oxidation by chlorite. At
constant pH conditions, higher cysteine concentrations encourage the
rapid formation of chlorine dioxide. [ClO2-]0 ) 2.5 × 10-2 M, I ) 1.0
M. [cysteine]0 ) (a) 1.0 × 10-3 M, (b) 2.0 × 10-3 M, (c) 3.0 × 10-3
M, (d) 4.0 × 10-3 M, (e) 5.0 × 10-3 M.
the formation of chlorine dioxide) and initial rates can be
evaluated and compared to initial conditions. Figure 7 shows a
series of such experimental data, and they show a steady increase
in rate of formation of chlorine dioxide with acid. The lower
reductant concentrations are insufficient to make a significant
impact on the rate of consumption of chlorine dioxide once the
reaction that forms ClO2 (referred to as reaction R6 later in
this article) commences, and hence a monotonic increase in
formation of chlorine dioxide will be observed.
Chlorine Dioxide Oxidation Reactions. Due to the oligo-
oscillatory nature of the chlorite-cysteine reaction, it would
appear that the direct oxidation of the reducing substrates in
solution by chlorine dioxide may hold the key to understanding
the reaction’s dynamics and mechanism. Figure 8A shows that
that expected for trace a, and yet the accumulation of chlorine
dioxide in the initial stages of the reaction in trace e (R ) 5)
far outstrips that from trace a with R ) 25.
Oxidation of L-Cysteinesulfinic Acid. Overall, this reaction
appeared to be much faster than the corresponding oxidation
of cysteine and it displayed more exotic dynamics. In the
absence of added acid, the reaction shows the typical sigmoidal
autocatalytic formation of chlorine dioxide at low values of R
(<6), (Figure 6). Surprisingly, this sigmoidal chlorine dioxide
formation disappears and is replaced by a monotonic production
at R > 8. By using an oxidant to reductant ratio above this
range, e.g., R ) 10, a series of acid dependence experiments
can be performed and the pseudo-first-order rate constants (for

Oxidation of Cysteine and Its Metabolites
J. Phys. Chem. A, Vol. 108, No. 26, 2004 5581
Figure 9. Variation of cysteine sulfinic acid in its oxidation by chlorine
dioxide. [ClO2]0 ) 8.7 × 10-4 M, I ) 1.0 M. [RCH2SO2H]0 ) (a)
1.25 × 10-4 M, (b) 2.50 × 10-4 M, (c) 5.0 × 10-4 M, (d) 1.0 × 10-3
M, (e) 1.5 × 10-3 M.
Figure 8. (A) Acid effects on cysteine oxidation by chlorine dioxide.
[cysteine]0 ) 5.0 × 10-4 M, [ClO2]0 ) 7.5 × 10-3 M, I ) 1.0 M.
[H+]0 ) (a) no acid, (b) 1.5 × 10-3 M, (c) 6.0 × 10-3 M, (d) 1.2 ×
10-2 M, (e) 1.8 × 10-2 M. (B) Variation of [cysteine] in cysteine
oxidation by chlorine dioxide. [ClO2]0 ) 7.5 × 10-4 M, I ) 1.0 M.
[cysteine]0 ) (a) 0.5 × 10-3 M, (b) 1.0 × 10-3 M, (c) 1.5 × 10-3 M,
(d) 2.0 × 10-3 M, (e) 2.5 × 10-3 M, (f) 3.0 × 10-3 M, (g) 3.5 × 10-3
M, (h) 4.0 × 10-3 M. Inset: Plot of initial rate of reaction vs cysteine
concentrations. Linearity is observed only at very low cysteine
concentration with a paid saturation.
Figure 10. Absorbance traces showing cystine oxidation by chlorite
in highly acidic medium; [H+]0 ) 0.05 M, [cystine]0 ) 0.01 M (trace
B). Trace A shows a control experiment without cystine. The chlorine
dioxide formed in trace A is from a pure oxyhalogen reaction without
the aid of a reducing substrate.
Cystine Oxidation. The most fascinating behavior was
observed in the oxidation of the cysteine dimer, cystine (Figure
10) by chlorite. Cystine only dissolves in highly acidic media.
As a result, the data shown in Figure 10 were obtained at lower
pH conditions than those used for the rest of the data shown in
this article. This particular oxidation shows a very rapid initial
formation of chlorine dioxide followed by its equally rapid
consumption and reformation (trace B). Trace A is a control
experiment in which conditions used to obtain trace A are
reproduced with the exception of cystine, which is omitted.
Highly acidic chlorite solutions produce chlorine dioxide
according to the stoichiometry27
acid retards the direct oxidation of cysteine by chlorine dioxide.
Figure 8B shows a positive effect on the initial rate of
consumption of chlorine dioxide with cysteine, but this effect
quickly saturates. Figure 9 shows the first-order dependence of
the initial rate with cysteinesulfinic acid for its direct oxidation
by chlorine dioxide. In this case the rates of oxidation of these
two substrates appear to be of the same or comparable order.
Low acid concentrations do not seem to exert any effect on the
reaction of cysteinesulfinic acid with chlorine dioxide. This can
be attributed to the fact that small acid concentrations will not
supersede the protons being generated by the sulfinic acid itself,
and so much higher acid concentrations may be needed for an
effect to be observed.
5ClO2- + 4H+ f 4ClO2(aq) + Cl- + 2H2O

5582 J. Phys. Chem. A, Vol. 108, No. 26, 2004
Darkwa et al.
Trace A represents the ClO2 formed solely from this reaction
without any contribution from the oxidation of cystine. Initial
conditions used for these data ensure that all the cystine is
consumed (R ) 5). The second ClO2 formation is the pure
oxyhalogen reaction above because from about 250 s onward,
the rates of formation of chlorine dioxide in both traces become
approximately equal.
Chlorite-L-Cysteine Reaction. The initial reaction step
should involve a two-electron transfer from chlorite to form a
sulfenic acid:
ClO2
- + RCH2SH + H+ f HOCl + RCH2SOH (R8)
R represents the asymmetric center of cysteine (HOOC)-
(H2N)C(H)-, and RCH2SOH is the first metabolite, an unstable
sulfenic acid. Sulfenic acids are well-known electron-
deficient molecules, and they should react rapidly with the
remaining thiol to give the dimeric form of cysteine, cystine
((RCH2S-)2:
Mechanism
There are three major reactions occurring in the reaction
mixture which are responsible for the observed oligooscillatory
production of chlorine dioxide. The first reaction is the oxidation
of the reductant to form products Cl- and cysteic acid as well
as the intermediate species in between: the sulfenic acid, sulfinic
acid, and hypochlorous acid. The second reaction involves the
rapid reaction of the hypochlorous acid with excess chlorite to
produce chlorine dioxide.28 The third reaction is the oxidation
of the reducing species in the reaction mixture by chlorine
dioxide. These can be summarized by the generic reaction
network shown below:
RCH2SH + RCH2SOH f (RCH2S-)2 + H2O (R9)
In the absence of further oxidant, cystine is quite stable, and it
is one of the metabolites used in this study. In the presence of
further oxidant, cystine should oxidize to produce two sulfenic
acid molecules:
(RCH2S-)2 + 2H2O f 2RCH2SOH + 2H+ + 2e- (R10)
ClO2- + Red f HOCl + Ox
(R5)
The oxidant could be any of the oxidizing species in solution:
ClO2-, HOCl, or ClO2(aq). Sulfenic acids, due to their instabil-
ity, can disproportionate into thiosulfinates,30 but these dispro-
portionations and dimerizations should be kinetically inconse-
quential in the presence of excess oxidant. Thus we would
expect further oxidation to the more stable sulfinic acid:
2ClO2- + HOCl + H+ f 2ClO2(aq) + Cl- + H2O (R6)
ClO2(aq) + Red f ClO2- + HOCl + Cl- + Ox (R7)
Red represents any two-electron reducing species, e.g., cysteine;
and Ox is the oxidized product after losing two electrons. Here
reactions R5 and R7 are deliberately written unbalanced. If we
assume a nonradical pathway for the reduction of ClO2-, then
a two-electron reduction of ClO2- should produce HOCl as the
reactive species which is involved in further oxidations of the
substrates.29 Our indicator reaction is the formation of chlorine
dioxide, R6. Reaction R6 is heavily dependent on reactions R5
and R7, and thus we can make correct predictions on the rate
of reaction R5 by observing R6. If reaction R6 was much faster
(by an order of magnitude or two) than reactions R5 and R7,
then oligooscillations will not be observed. Instead, a very sharp
induction period will be observed signifying the complete
consumption of the reducing substrate.13 The effective coupling
of reactions R5, R6, and R7, by virtue of being of comparable
magnitudes in rate, is responsible for the exotic dynamics
observed in these reaction systems. This reaction mechanism
is much simpler than normal sulfur-based oxidations because
the organic backbone of the thiol limits the numbers of possible
intermediates that can be generated during the oxidation. The
organosulfur compound, during the course of its oxidation, may
be expected to produce its dimer (cystine), sulfenic acid, and
its sulfinic acid. The oxyhalogen species, on the other hand,
can only produce HOCl, Cl-, ClO2(aq), and chlorate in high
pH environments.28 The oxidizing species thus are limited only
to ClO2-, HOCl, and ClO2(aq). In low acid concentrations
cystine is insoluble in water and will be expected to precipitate
out during the course of the reaction if its formation is
quantitative and long-lived. Since no precipitation was observed,
we can exclude cystine as a major intermediate during the
oxidation of cysteine in excess chlorite. Thus a full and
exhaustive mechanism for the oxidation of cysteine will involve
a maximum of nine reactions involving the oxidation of a sulfur
center coupled to the reduction of a chlorine center (combination
of three reductants and three oxidants) plus the standard
oxychlorine reactions and acid-base equilibria.
RCH2SOH + ClO2- + H+ f RCH2SO2H + HOCl (R11)
The production of the sulfinic acid gives us the other metabolite
whose oxidation kinetics were evaluated in this article. Our
1
stoichiometric as well as H NMR data suggest that the final
oxidation product, in these conditions, would be cysteic acid:
RCH2SO2H + ClO2- + H+ f RCH2SO3H + HOCl (R12)
Thus a stepwise oxidation of L-cysteine will cover the oxidation
of its metabolites as well (cystine and cysteinesulfinic acid). In
highly acidic conditions the oxidation of L-cysteinesulfinic acid
is faster than that of L-cysteine, suggesting that the rate-
determining reactions for the whole reaction scheme should lie
between reactions R8 and R11. The instability of the sulfenic
acid precludes reaction R11 from being rate-determining, leaving
reaction R8. Reaction R8 should be able to rationalize the
observed acid and species dependency of the reaction rates
observed in Figures 3-5. A full explanation of these data,
however, requires an understanding of the chlorine dioxide
oxidation reactions which will be handled later in this article.
Chlorite-L-Cysteinesulfinic Acid Reaction. Since we have
established that the oxidation of cysteine passes through
cysteinesulfinic acid, the proposed mechanism for the oxidation
of the sulfinic acid should then be consistent with its role in
the overall oxidation of cysteine to cysteic acid. No organosulfur
intermediates are possible in this simple two-electron oxidation
of the sulfinic acid to the sulfonic acid. All the nonlinearities
generated in this reaction system should have their genesis solely
from the oxyhalogen kinetics. Figure 6 shows autocatalysis in
the formation of ClO2(aq) especially at low chlorite concentra-
tions. Acid catalysis as observed in Figure 7 would be expected
if the formation of chlorine dioxide involves a pure oxyhalogen
reaction. The autocatalysis observed suggests that the formation
of chlorine dioxide is controlled by the reaction of an oxy-
chlorine intermediate and not ClO2- as in reaction R12. HOCl

Oxidation of Cysteine and Its Metabolites
J. Phys. Chem. A, Vol. 108, No. 26, 2004 5583
produced in R12 will react as shown in reaction R6 to form
chlorine dioxide.
Chlorine Dioxide Formation. HOCl used in reaction R6 has
to be autocatalytically produced for the observation of the
sigmoidal kinetics in Figure 6. Another pathway for the
production of chlorine dioxide would involve aqueous chlorine:
produced, they should be short-lived and can only be observed
with a specialized trap.32 A kinetically indistinguishable and
very plausible pathway to R18 + R19 would involve the
formation of a loose adduct of the thiol and chlorine dioxide.
This adduct would then react with another molecule of aqueous
chlorine dioxide to release the sulfenic acid and two chlorous
acid molecules:
HOCl + Cl- + H+ f Cl2(aq) + H2O
2ClO2- + Cl2(aq) f 2ClO2(aq) + 2Cl-
(R13)
(R14)
-
ClO2(aq) + RCH2SH h [RCH2S-ClO2 ] [H+] (R22)
-
[RCH2S-ClO2 ] + ClO2(aq) + H2O f
However, reaction sequence R13-R14 will only become viable
if reaction R6 is slow. Reaction R6 is a composite of sequence
of reaction steps that involve the often-proposed asymmetric
Cl2O2 intermediate in aqueous environments.29 The existence
of this intermediate was proved through isotopic labeling
experiments of Taube and Dodgen31
RCH2SOH + 2ClO2- + H+ (R23)
If one assumes that the protonated thiol is inert, then the rate
of reaction R20 will be given by
-d[ClO2] k0[RCH2SH]0[ClO2(aq)]
ClO2- + HOCl + H+ f Cl2O2 + H2O
Cl2O2 + ClO2- f 2HOCl + Cl-
(R15)
(R16)
)
(1)
1 + Kb[H+]
dt
where [RCH2SH]0 is the initial concentration of cysteine used.
Equation 1 can explain the observed acid retardation, and
This sequence shows that one HOCl molecule will produce two
HOCl molecules, and if HOCl is the major oxidant, then an
autocatalytic production of chlorine dioxide from reaction R6
can be observed. We can generalize reaction R16 to apply to
any two-electron reductant:
using this equation we deduced an upper-limit value of kR20
)
405 ( 50 M-1 s-1. Further reaction would then involve the
oxidation of cysteine and its metabolites by chlorite and
hypochlorous acid. Sulfinic acids are not as nucleophilic as
thiols, and this can explain why no appreciable effect of acid
on the oxidation of L-cysteinesulfinic acid by chlorine dioxide
was observed.
Cl2O2 + 2e- + 2H+ f 2HOCl
(R17)
Overall Mechanism. We can define a complete mechanism
that can adequately explain all the reactions reported in this
article. A single mechanism should be able to describe the
observed dynamics of the chlorite-cysteine, chlorite-cysteine-
sulfinic acid, chlorite-cystine, chlorine dioxide-cysteine, and
chlorine dioxide-cysteinesulfinic acid reactions. This mecha-
nism is shown in Table 1. It is comprised of three rapid
protolytic equilibria (reactions M1-M3), two pure oxyhalogen
reactions (M20 and M21), and two sulfur-sulfur dispropor-
tionation reactions (M27-M28). The rest of the reactions
involve the oxidation of a sulfur center coupled with the
reduction of a chlorine center. The complex acid depen-
dence observed in the reaction system (see Figure 4) can be
explained by the opposing effects of reactions M1, M3, M4,
and M5. High acid concentrations lower the nucleophilic nature
of the thiol group (reaction R21) but also protonate chlorite to
form chlorous acid which is a better electrophile. Since the
pKa of chlorous acid is approximately 2.0,33 reactions run
in pH conditions lower than 2.0 should see a dominance of
reaction M5 over M4 and hence acid retardation. The slower
tandem of M4 + M5 in high acid stunts the production of the
reactive and autocatalytic species HOCl, and hence lower
chlorine dioxide concentrations are formed as seen in Figure 4.
If the rates of reactions M4 and M5 are not equal, then the
whole reaction system will display a strong acid effect espe-
cially in conditions where the pH of the solution is greater than
or equal to the pKa of chlorous acid. The rate law that is
derived from this mechanism for the reaction of chlorite and
cysteine is
This autocatalysis will not only affect chlorine dioxide formation
but the oxidation of the substrate as well as its metabolites.
Consumption of Chlorine Dioxide. Parts A and B of Figure
8 show that there is a feasible and rapid reaction between
cysteine and chlorine dioxide. Figure 9 shows an equally rapid
reaction of the sulfinic acid with chlorine dioxide. Although
Figure 8A shows that the reaction of cysteine and chlorine
dioxide is retarded by acid, the reaction of the cysteinesulfinic
acid with chlorine dioxide was insensitive to acid for a wide
range of low acid concentrations. Chlorine dioxide is a radical
species and should react very rapidly with the nucleophilic center
of the thiol:
ClO2(aq) + RCH2SH h ClO2- + H+ + RCH2S‚ (R18)
ClO2(aq) + RCH2S‚ + H2O f ClO2- + RCH2SOH + H+
(R19)
Reactions R18 and R19 can be combined to the following
composite termolecular reaction:
2ClO2(aq) + RCH2SH + H2O f
2ClO2- + 2H+ + RCH2SOH (R20)
Our kinetics data suggests that the reaction is first order in both
cysteine and chlorine dioxide, which would implicate reaction
R18 as rate-determining. The inhibitory effect of acid renders
reaction R18 irreversible since acid would accelerate the reaction
by forming the weak acid HClO2 thus pushing the “equilibrium”
to the right and accelerating the reaction. The inhibition should
arise from the deactivation of the thiol group by protonation:
[RCH2SH]0[Cl(III)]T
(1 + Ka-1[H+])(1 + Kb[H+])
Rate )
{kM4 + kM5Ka-1[H+]}
RCH2SH + H+ h [RCH2SH2]+; Kb
Thiyl radicals were not detected in our reaction system, and if
(R21)
(2)
with a complex dependency on acid as experimentally observed.

5584 J. Phys. Chem. A, Vol. 108, No. 26, 2004
Darkwa et al.
TABLE 1: Full Mechanisma
M6) which can disproportionate to thiosulfinates in the absence
of excess oxidant.30 Further oxidation of the sulfenic acid will
give the sulfinic acid and finally cysteic acid. Our experimental
data gave a lower-limit rate constant for direct reaction of
L-cysteinesulfinic acid and chlorine dioxide of kM18 ) 210 (
no.
reaction
M1 ClO2- + H+ h HClO2; Ka-1
M2 OCl- + H+ h HOCl
M3 RCH2SH + H+ h [RCH2SH2]+; Kb
15 M-1 s-1
.
M4 RCH2SH + ClO2- f RCH2SOH + OCl-
This mechanism is effectively a combination of the four
oxidants in the reaction mixture, HClO2, ClO2-, HOCl, and
ClO2(aq) with the four reductants, RCH2SH, (RCH2S-)2, RCH2-
SOH, and RCH2SO2H. Apart from the protolytic equilibria and
adduct formations (e.g., reactions M12, M14, M16, and M18),
all reactions involving the oxidation of a sulfur center were
assumed irreversible. The sulfur-sulfur reactions, M27 and
M28, are important for stoichiometric consistency in the
presence of excess reductant. Under such conditions, the organo-
sulfur species would disproportionate such that the more stable
sulfinic and sulfonic acids are the dominant products. The
possibility also exists for the formation of various thiosulfinates
in excess reductant conditions.
M5 RCH2SH + HClO2 f RCH2SOH + HOCl
M6 (RCH2S-)2 + ClO2- + H2O f 2RCH2SOH + OCl-
M7 RCH2SOH + ClO2- f RCH2SO2H + OCl-
M8 RCH2SO2H + ClO2- f RCH2SO3H + OCl-
M9 RCH2SH + HOCl f RCH2SOH + H+ + Cl-
M10 RCH2SOH + HOCl f RCH2SO2H + H+ + Cl-
M11 RCH2SO2H + HOCl f RCH2SO3H + H+ + Cl-
M12 (RCH2S-)2 + ClO2(aq) + H2O h (RCH2S)2‚ClO2
M13 (RCH2S)2‚ClO2 + ClO2(aq) + H2O f 2 RCH2SOH + HClO2
M14 RCH2SH + ClO2(aq) h [RCH2SClO2]-‚H+
M15 RCH2SClO2- + ClO2(aq) + H2O f RCH2SOH + 2ClO2- + H+
M16 RCH2SOH + ClO2(aq) h [RCH2S(O)ClO2]-‚ H+
M17 RCH2S(O)ClO2- + ClO2(aq) + H2O f RCH2SO2H + 2ClO2- + H+
M18 RCH2SO2H + ClO2(aq) h [RCH2S(O2)ClO2]-‚ H+
M19 RCH2S(O2)ClO2- + ClO2(aq) + H2O f RCH2SO3H + 2ClO2- + H+
M20 ClO2- + HOCl + H+ h Cl2O2 + H2O
Computer Simulations. We utilized a unique approach to
modeling the dynamics of the oxidation of cysteine by chlorite.
We modeled the simplest system first, the reaction with the
smallest number of intermediates: the cysteinesulfinic acid-
chlorine dioxide system. The kinetics parameters derived from
this calculation were used to model the cysteine-chlorine
dioxide reaction. The full reaction scheme, chlorite-cysteine,
was then finally modeled using the data derived from the other
two calculations. This approach was possible because we had
established that the oxidation of cysteine passed through the
sulfinic acid before proceeding to cysteic acid.
M21 Cl2O2 + ClO2- h 2 ClO2(aq) + Cl-
M22 RCH2SH + Cl2O2 + H2O f RCH2SOH + 2HOCl
M23 RCH2SOH + Cl2O2 + H2O f RCH2SO2H + 2HOCl
M24 RCH2SO2H + Cl2O2 + H2O f RCH2SO3H + 2HOCl
M25 RCH2SOH + HClO2 f RCH2SO2H + HOCl
M26 RCH2SO2H + HClO2 f RCH2SO3H + HOCl
M27 2RCH2SOH h RCH2SO2H + RCH2SH
M28 RCH2SOH + RCH2SO3H h 2 RCH2SO2H
a Legend: RCH2SH, cysteine (“R” represents the asymmetric center
on cysteine); (RCH2S-)2, cystine; RCH2SOH, cysteinesulfenic acid;
RCH2SO2H, cysteinesulfinic acid; RCH2SO3H, cysteic acid.
Cysteinesulfinic Acid-Chlorine Dioxide. Table 2 shows the
very short and abbreviated mechanism used to simulate the
oxidation of the sulfinic acid by chlorine dioxide. Since our
experimental data have shown no strong acid dependence for
this specific reaction, we could eliminate reactions M1 and M3
from the mechanism. Autocatalysis was not observed in the
consumption of chlorine dioxide, and thus we did not have to
include the autocatalytic production of HOCl. Reaction P1 was
assumed to be bimolecular, as deduced from our kinetics data,
even though it is written as termolecular. The only kinetics
parameters that had to be guessed were those for P2 and P3.
Kinetics parameters for P1 were estimated from this study, and
those for P4 were derived from the literature.34 The model was
insensitive to values of kP2 and kP3 as long as they were not
rate-determining. Figure 11A shows the reasonably good fit
obtained using this very simple mechanism.
In eq 2, [Cl(III)]T represents the total chlorine(III) species and
is given by the relation
-
[Cl(III)]T ) [HClO2] + [ClO2 ]
(3)
In the limit of low acid concentrations, e.g., 8 > pH > 3.0, the
terms in the denominator as well as the last term drop out giving
Rate ) kM4[RCH2SH]0[Cl(III)]T
(4)
In highly basic pH conditions, higher than physiological pH,
the thiolate anion asserts itself, giving a different mechanism
and also possibly different oxychlorine products. This range was
not studied in this report.
Cysteine-Chlorine Dioxide. The modeling of this system
was merely an extension of that for the oxidation of the sulfinic
acid. We could now utilize the kinetics parameters deduced from
the model in Table 2. The mechanism used to simulate the
cysteine-chlorine dioxide reaction is shown in Table 3. The
observation of acid dependence in Figure 8A meant that we
had to include equilibria Q14 and Q15. Since acid inhibition
was observed below pH 2, we assumed that equilibrium Q15
was more important. This proved to be the case in our
simulations. In this model the protonated thiol was assumed to
The autocatalytic production of chlorine dioxide is fueled by
the composite reaction M20 + M21 coupled with reactions
M22-M24 that contain autocatalytic production of HOCl. The
mechanism in Table 1 acknowledges three main oxidizing
species in the reaction mixture: ClO2-, HOCl, and ClO2(aq).
The intermediate we propose, Cl2O2, is only introduced to justify
the observed autocatalysis. For the autocatalysis to prevail, the
reactions of this intermediate, M22-M24, should be slower than
the reactions involving HOCl, M9-M11. The initial oxidation
of cystine should give two sulfenic acid molecules (reaction
TABLE 2: Cysteinesulfinic Acid-Chlorine Dioxide Mechanism
no.
reaction
kf; kr
P1
P2
P3
P4
2ClO2(aq) + RCH2SO2H + H2O f 2HClO2 + RCH2SO3H
HClO2 + RCH2SO2H f RCH2SO3H + HOCl
RCH2SO2H + HOCl f RCH2SO3H + H+ + Cl-
2HClO2 + HOCl h 2ClO2(aq) + H+ + Cl- + H2O
210
350
5 × 103
1.01 × 106; 1 × 102

Oxidation of Cysteine and Its Metabolites
J. Phys. Chem. A, Vol. 108, No. 26, 2004 5585
Figure 11. (A) Computer simulations using the model shown in Table 2 for the reaction between chlorine dioxide and cysteinesulfinic acid.
Experimental data is represented by the solid line, while the simulations appeared as the dashed line. Conditions simulated: [ClO2]0 ) 8.7 × 10-4
M; [RCH2SO2H]0 ) 5.0 × 10-4 M. (B) Modeling the chlorite-cysteinesulfinic acid reaction at two separate acid concentrations. The data simulated
is shown in Figure 8A. (a) 0.0015 M H+ and (b) 0.018 M H+. The model shown in Table 3 correctly predicted the experimentally observed acid
dependence. Solid lines represent experimental data, and symbols denote data generated from the model. (C) Extension of the model shown in trace
a of part B showing the concentration trajectories for some of the products and reactive species which we could not experimentally measure. The
model shows that cysteic acid is formed quantitatively after a short induction period while the sulfenic and sulfinic acids attain transient peaks
before decaying to form products.
be completely inert (see eq 1). Depending upon pH, reactions
Q10-Q12 could be ignored or shut down without any loss of
accuracy in our model. The most important parameter in this
model was kQ1. Reactions Q8 and Q9 could not assert
themselves at the beginning of the reaction because the
concentrations of these reactive sulfur species would be vanish-
ingly small. This reaction did not display any strong auto-
catalysis, and so we could eliminate (or minimize) the sequence
of reactions that incorporate HOCl autocatalysis. Kinetics
parameters for reaction Q1 were obtained from this study (405
( 50 M-1 s-1) as well as parameters for reaction Q8. The rest,
apart from Q13-Q16, were estimated for best fit. At low pH
conditions, the model was still accurate from the use of only
reactions Q1-Q7 and Q15. Figure 11B shows the simulations
fit to the data. This simple model was able to predict the
observed acid dependence as well as the rate law in the form
of eq 1. Using this mechanism we could also model the
concentration profiles of some of the intermediates we could
not measure experimentally such as the sulfinic and sulfenic
acids (Figure 11C). Figure 11C shows that concentrations of
the sulfenic and sulfinic acid intermediates go through transient
maxima before decaying and making way for cysteic acid.
Cysteic acid shows a short induction period followed by its
monotonic formation. Hypochlorous acid gives the same
concentration profile as the sulfinic and sulfonic acids but only
at much lower concentrations.

5586 J. Phys. Chem. A, Vol. 108, No. 26, 2004
Darkwa et al.
TABLE 3: Cysteine-Chlorine Dioxide Mechanism
no.
reaction
kf; kr
Q1
Q2
Q3
Q4
Q5
Q6
Q7
Q8
RCH2SH + 2ClO2(aq) + H2O f RCH2SOH + 2HClO2
RCH2SOH + HClO2 f RCH2SO2H + HOCl
RCH2SO2H + HClO2 f RCH2SO3H + HOCl
RCH2SH + HClO2 f RCH2SOH + HOCl
405
5 × 103
2.5 × 103
8 × 103
RCH2SH + HOCl f RCH2SOH + H+ + Cl-
RCH2SOH + HOCl f RCH2SO2H + H+ + Cl-
RCH2SO2H + HOCl f RCH2SO3H + H+ + Cl-
RCH2SOH + 2ClO2(aq) + H2O f RCH2SO2H + 2HClO2
RCH2SO2H + 2ClO2(aq) + H2O f RCH2SO3H + 2HClO2
RCH2SH + ClO2- + H+ f RCH2SOH + HOCl
RCH2SOH + ClO2- + H+ f RCH2SO2H + HOCl
RCH2SO2H + ClO2- + H+ f RCH2SO3H + HOCl
2HClO2 + HOCl h 2ClO2(aq) + H+ + Cl- + H2O
ClO2- + H+ h HClO2; Ka-1
5 × 103
8 × 103
5 × 103
5 × 102
Q9
210
Q10
Q11
Q12
Q13
Q14
Q15
75
1.2 × 102
1 × 102
1.01 × 106; 1 × 102
1 × 109; 1.02 × 107
1 × 103; 5 × 109
RCH2SH + H+ h [RCH2SH2]+; Kb
TABLE 4: Mechanism Used for Modeling the Whole Reaction Scheme
no.
reaction
rate constants: kf; kr
S1
S2
S3
S4
S5
S6
S7
S8
ClO2- + H+ h HClO2; Ka-1
1 × 109; 1.02 × 107
1 × 103; 5 × 109
75
RCH2SH + H+ h [RCH2SH2]+; Kb
RCH2SH + ClO2- + H+ f RCH2SOH + HOCl
RCH2SH + HClO2 f RCH2SOH + HOCl
8 × 103
(RCH2S)2 + ClO2- + H+ + H2O f 2RCH2SOH + HOCl
RCH2SOH + ClO2- + H+ f RCH2SO2H + HOCl
RCH2SO2H + ClO2- + H+ f RCH2SO3H + HOCl
RCH2SH + HOCl f RCH2SOH + H+ + Cl-
RCH2SOH + HOCl f RCH2SO2H + H+ + Cl-
RCH2SO2H + HOCl f RCH2SO3H + H+ + Cl-
(RCH2S)2 + 2ClO2(aq) + 2H2O f 2RCH2SOH + 2HClO2
RCH2SH + 2ClO2(aq) + H2O f RCH2SOH + 2HClO2
RCH2SOH + 2ClO2(aq) + H2O f RCH2SO2H + 2HClO2
RCH2SO2H + 2ClO2(aq) + H2O f RCH2SO3H + 2HClO2
ClO2- + HOCl + H+ h Cl2O2 + H2O
1 × 102
1.2 × 102
1 × 102
5 × 103
S9
8 × 103
S10
S11
S12
S13
S14
S15
S16
S17
S18
S19
5 × 102
1 × 102
5 × 102
1 × 103
210
1.01 × 106; 0.1
1.5 × 103; 5.5 × 10-6
15
Cl2O2 + ClO2- h 2 ClO2(aq) + Cl-
RCH2SH + Cl2O2 + H2O f RCH2SOH + 2HOCl
RCH2SOH + Cl2O2 + H2O f RCH2SO2H + 2HOCl
RCH2SO2H + Cl2O2 + H2O f RCH2SO3H + 2HOCl
1 × 102
25
Modeling the Whole Reaction Scheme. A simplified mech-
anism that was distilled from the full mechanism given in Table
1 was used for modeling the overall reaction kinetics. A number
of assumptions were made in arriving at this simplified scheme
shown in Table 4. Reaction M2 was assumed, in this environ-
ment, to be very fast and was eliminated. Thus, it was combined
with reaction M4 to give a new reaction, S3, which though
termolecular, was treated as bimolecular. Reactions of chlorine
dioxide were also handled in the same manner, combining M12
+ M13, M14 + M15, M16 + M17, and M18 + M19 into
composite reactions which were still treated as bimolecular. The
kinetics constants for reactions S1 and S2 were estimated from
the pKa of chlorous acid33 and an estimated value for pKb of
the thiol group in cysteine. The simulations were insensitive to
the values of these constants as long as they were fast and not
rate-determining. Thus after estimating kS1, the value of k-S1
was then fixed by the Ka of chlorous acid. Kinetics constants
for reactions S11-S14 were estimated from this study; and those
for reactions S15 and S16 were derived from established
literature values. The rest of the constants were adjusted for
best fit. This mechanism was simulated using both the semi-
implicit fourth-order Runge Kutta techniques and the Chemical
Kinetics Simulator developed by IBM. The presence of rapid
reactions S1 and S2 made the overall mechanism extremely stiff
with calculations needing to proceed overnight on the fastest
Pentium IV processors. This model was able to satisfactorily
predict the data on chlorite variations shown in Figure 3 as well
as that in Figure 7. It was unable to predict, to the same degree
of precision, the second phase of chlorine dioxide formation
shown in Figure 4. One major reason for this failure is the
unknown activity of aqueous chlorine dioxide at various pHs
and temperatures. The reactions we studied are known to be
extremely exothermic.35 Although we tried to control the
temperature as efficiently as possible, after long periods we still
lost a lot of chlorine dioxide due to its volatility.
Conclusion
The oxidation of cysteine, in the absence of P450-type
enzymes and flavin-containing monooxygenases, proceeds only
as far as cysteic acid. This appears to be common to all thiols
with an amino group at the â or γ positions. The carboxylic
acid group on cysteine, however, precludes any formation of
chloramines as has been observed with taurine and hypotaurine.
Acknowledgment. We thank the University of the Western
Cape for awarding a leave of absence to one of us (J.D.). We
would like to thank Fungai Mukome for performing the ICPMS

Oxidation of Cysteine and Its Metabolites
J. Phys. Chem. A, Vol. 108, No. 26, 2004 5587
(17) Hogg, P. J. Redox Rep. 2002, 7, 71-77.
analysis on the Nanopure water used for reagent preparations.
This work was funded by Grant No. CHE 0137435 from the
National Science Foundation.
(18) Okazaki, R.; Goto, K. Heteroat. Chem. 2002, 13, 414-418.
(19) Makarov, S. V.; Mundoma, C.; Penn, J. H.; Svarovsky, S. A.;
Simoyi, R. H. J. Phys. Chem. A 1998, 102, 6786-6792.
(20) Hirschberger, L. L.; Hou, Y. C.; Stipanuk, M. H. FASEB J. 1994,
8, A463.
References and Notes
(21) Thomas, E. L.; Bozeman, P. M.; Jefferson, M. M.; King, C. C. J.
Biol. Chem. 1995, 270, 2906.
(22) Indelli, A. J. Phys. Chem. 1964, 68, 3027-3031.
(23) Kern, D.; Kim, C.-H. J. Am. Chem. Soc. 1965, 87, 5309-5313.
(24) Brauer, G. Handbook of PreparatiVe Organic Chemistry; Academic
Press: New York, 1963; p 301.
(1) Kelly, M.; Lappalainen, P.; Talbo, G.; Haltia, T.; van der Oost, J.;
Saraste, M. J. Biol. Chem. 1993, 268, 16781-16787.
(2) Hofgen, R.; Kreft, O.; Willmitzer, L.; Hesse, H. Amino Acids 2001,
20, 291-299.
(3) Young, V. R.; Wagner, D. A.; Burini, R.; Storch, K. J. Am. J. Clin.
Nutr. 1991, 54, 377-385.
(25) Chinake, C. R.; Simoyi, R. H. J. Phys. Chem. B 1997, 101, 1207-
1214.
(4) Grazioli, L.; Casero, D.; Restivo, A.; Cozzi, E.; Marcucci, F. J.
Biol. Chem. 1994, 269, 22304-22309.
(5) Parcell, S. Altern. Med. ReV. 2002, 7, 22-44.
(6) Stipanuk, M. H. Annu. ReV. Nutr. 1986, 6, 179-209.
(7) Halliwell, B.; Gutteridge, J. M. Arch. Biochem. Biophys. 1990, 280,
1-8.
(8) Halliwell, B.; Gutteridge, J. M. Lancet 1984, 2, 1095.
(9) Halliwell, B.; Gutteridge, J. M. Lancet 1984, 1, 1396-1397.
(10) Mirazimi, A.; Svensson, L. J. Virol. 1998, 72, 3887-3892.
(11) Carelli, S.; Ceriotti, A.; Cabibbo, A.; Fassina, G.; Ruvo, M.; Sitia,
R. Science 1997, 277, 1681-1684.
(12) Chinake, C. R.; Simoyi, R. H. Journal of Physical Chemistry 1993,
97, 11569-11570.
(13) Darkwa, J.; Mundoma, C.; Simoyi, R. H. J. Chem. Soc., Faraday
Trans. 1998, 94, 1971-1978.
(14) Jaenicke, R. Naturwissenschaften 1996, 83, 544-554.
(15) Han, D.; Handelman, G.; Marcocci, L.; Sen, C. K.; Roy, S.;
Kobuchi, H.; Tritschler, H. J.; Flohe, L.; Packer, L. Biofactors 1997, 6,
321-338.
(16) Saez, G.; Thornalley, P. J.; Hill, H. A.; Hems, R.; Bannister, J. V.
Biochim. Biophys. Acta 1982, 719, 24-31.
(26) Chinake, C. R.; Simoyi, R. H. J. Phys. Chem. B 1998, 102, 10490-
10497.
(27) Gordon, G. J. Am. Water Works Assoc. 2001, 93, 163.
(28) Jia, Z.; Margerum, D. W.; Francisco, J. S. Inorg. Chem. 2000, 39,
2614-2620.
(29) Chinake, C. R.; Mundoma, C.; Olojo, R.; Chigwada, T.; Simoyi,
R. H. Phys. Chem. Chem. Phys. 2001, 3, 4957-4964.
(30) Huang, K. P.; Huang, F. L. Biochem. Pharmacol. 2002, 64, 1049-
1056.
(31) Taube, H.; Dodgen, H. J. Am. Chem. Soc. 1949, 71, 3330-3336.
(32) Karoui, H.; Hogg, N.; Frejaville, C.; Tordo, P.; Kalyanaraman, B.
J. Biol. Chem. 1996, 271, 6000-6009.
(33) Cotton, F. A.; Wilkinson, G. AdVanced Inorganic Chemistry, 5th
ed.; John Wiley; Wiley-Interscience: New York, 1988; p 567.
(34) Peintler, G; Nagypal, I; Epstein, I. R. J. Phys. Chem. 1990, 94,
2954-2960.
(35) Chinake, C. R.; Simoyi, R. H. S. Afr. J. Chem. 1997, 50, 220-
226.