Oxyhalogen-sulfur chemistry - Kinetics and mechanism of the oxidation of cysteamine by acidic iodate and iodine

Article abstract of DOI:10.1139/v05-263

The oxidation of cysteamine by iodate and aqueous iodine has been studied in neutral to mildly acidic conditions. The reaction is relatively slow and is heavily dependent on acid concentration. The reaction dynamics are complex and display clock behavior, transient iodine production, and even oligooscillatory production of iodine, depending upon initial conditions. The oxidation product was the cysteamine dimer (cystamine), with no further oxidation observed past this product. The stoichiometry of the reaction was deduced to be IO ₃^- + 6H₂NCH₂CH₂SH → I^- + 3H₂NCH₂CH₂S-SCH ₂CH₂NH₂ + 3H₂O in excess cysteamine conditions, whereas in excess iodate the stoichiometry of the reaction is 2IO₃^- + 10H₂NCH₂CH₂SH → I₂ + 5H₂NCH₂CH₂S-SCH ₂CH₂NH₂ + 6H₂O. The stoichiometry of the oxidation of cysteamine by aqueous iodine was deduced to be I₂ + 2H₂NCH₂CH₂SH → 2I^- + H ₂NCH₂CH₂S-SCH₂CH₂NH ₂ + 2H⁺. The bimolecular rate constant for the oxidation of cysteamine by iodine was experimentally evaluated as 2.7 (mol L ^-1)^-1 s^-1. The whole reaction scheme was satisfactorily modeled by a network of 14 elementary reactions.

Full text of DOI:10.1139/v05-263

49
Oxyhalogen-sulfur chemistry — Kinetics and
mechanism of the oxidation of cysteamine by
acidic iodate and iodine
Alice Chanakira, Edward Chikwana, David H. Peyton, and Reuben H. Simoyi
Abstract: The oxidation of cysteamine by iodate and aqueous iodine has been studied in neutral to mildly acidic con-
ditions. The reaction is relatively slow and is heavily dependent on acid concentration. The reaction dynamics are com-
plex and display clock behavior, transient iodine production, and even oligooscillatory production of iodine, depending
upon initial conditions. The oxidation product was the cysteamine dimer (cystamine), with no further oxidation ob-
served past this product. The stoichiometry of the reaction was deduced to be IO3 + 6H2NCH2CH2SH → I– +
–
3H2NCH2CH2S-SCH2CH2NH2 + 3H2O in excess cysteamine conditions, whereas in excess iodate the stoichiometry of
–
the reaction is 2IO3 + 10H2NCH2CH2SH → I2 + 5H2NCH2CH2S-SCH2CH2NH2 + 6H2O. The stoichiometry of the oxi-
dation of cysteamine by aqueous iodine was deduced to be I2 + 2H2NCH2CH2SH → 2I– + H2NCH2CH2S-
SCH2CH2NH2 + 2H+. The bimolecular rate constant for the oxidation of cysteamine by iodine was experimentally eval-
uated as 2.7 (mol L–1)–1 s–1. The whole reaction scheme was satisfactorily modeled by a network of 14 elementary re-
actions.
Key words: cysteamine, cystamine, Dushman reaction, oligooscillations.
Résumé : L’oxydation de la cystéamine par l’iodate et l’iode aqueux a été étudiée dans des conditions allant de neutre
à faiblement acides. La réaction est relativement lente et dépend fortement sur la concentration en acide. La dyna-
mique de la réaction est complexe et présente un comportement d’horloge avec production transitoire d’iode allant jus-
qu’à une production oligooscillatoire d’iode suivant les conditions initiales. Le produit de l’oxydation est le dimère de
la cystéamine, la cystamine, sans oxydation apparente ultérieure de ce produit. On a déduit que, dans les conditions
où la cystéamine est en excès la stoechiométrie de la réaction est IO3 + 6H2NCH2CH2SH → I– + 3H2NCH2CH2S-
–
–
SCH2CH2NH2 + 3H2O alors que lorsque l’iodate est en excès, la stoechiométrie de la réaction est de la forme 2IO3
10H2NCH2CH2SH → I2 + 5H2NCH2CH2S-SCH2CH2NH2 + 6H2O. On a établi que la stoechiométrie de la réaction
d’oxydation de la cystéamine par l’iode aqueux est de la forme I2 + 2H2NCH2CH2SH → 2I– + H2NCH2CH2S-
+
SCH2CH2NH2 + 2H+. La valeur établie expérimentalement pour la constante de vitesse bimoléculaire d’oxydation de la
cystéamine par l’iode est de 2,7 (mol L–1)–1 s–1. On peut faire une modélisation complète du schéma réactionnel sur la
base de quatorze réactions élémentaires.
Mots clés : cystéamine, cystamine, réaction de Dushman, oligooscillations.
[Traduit par la Rédaction] Chanakira et al. 57
Introduction
tion mechanisms of organosulfur compounds have been
hampered by the propensity of sulfur chemistry to produce
complex and exotic dynamics (2), which are fueled by free-
radical mechanisms (3), autoxidations, and production of
varied products and intermediates (4, 5). With the sulfur
center having an oxidation state range of –2 to +6, it is easy
for these intermediates to assert themselves.
Cysteamine is a very important biological molecule.
Medically known as Cystagon, it can be used as oral therapy
for prevention of hypothyroidism and enhances growth in
patients with nephropathic cystinosis (6, 7). This is a
lysosomal storage disorder that was long considered primar-
ily a renal disease but is now recognized as a systemic disor-
der that eventually affects many organ systems in children
(8). This disorder is characterized by cystine accumulation
within cellular lysosomes, and cysteamine converts this
disulfide into cysteine and a mixed disulfide, cysteamine–
cysteine, which can easily be eliminated from the cystinotic
lysosomes, thus effecting depletion of cellular cystine (9).
Organosulfur compounds dominate much of synthetic, an-
alytical, and medicinal chemistry. Oxidations of organic sul-
fur compounds appear to be involved in many cellular
functions, including regulation of protein synthesis and pro-
tection of the cell from oxidative damage (1). Owing to the
nucleophilic nature of the thiol group, a better understanding
of the biological activities of these compounds can be de-
rived from their oxidation mechanisms. Studies of the oxida-
Received 4 April 2005. Published on the NRC Research Press
Web site at http://canjchem.nrc.ca on 25 February 2006.
A. Chanakira,1 E. Chikwana, D.H. Peyton, and
R.H. Simoyi.2 Department of Chemistry, Portland State
University, Portland, OR 97207-0751, USA.
1Present address: College of St. Catherine, St. Paul,
MN 55105, USA.
2Corresponding author (e-mail: rsimoyi@pdx.edu).
Can. J. Chem. 84: 49–57 (2006)
doi:10.1139/V05-263
© 2006 NRC Canada

50
Can. J. Chem. Vol. 84, 2006
Cysteamine and its disulfide, cystamine, can also be used in
topical eye drops to dissolve corneal cystine crystals (10).
Cysteamine is an excellent scavenger of OH and HOCl; it
did not contain enough metal ions to affect the overall reac-
tion kinetics and mechanism (21). The highest metal ion
concentration was cadmium at 1.5 ppb, followed by lead at
0.43 ppb.
·
also reacts with H2O2 and other oxygen-based toxic metabo-
lites (11). In addition to protection against radical damage in
DNA, cysteamine can also act as a repair agent for DNA
through the formation of the protective RSSR·–, which reacts
with the DNA·+ radical ion to regenerate DNA and form
cystamine (12). It has been shown that cysteamine and its
metabolite hypotaurine are far more likely to act as antioxi-
dants in vivo than is taurine, provided that they are present
in sufficient concentration at sites of oxidant generation (13).
Cysteamine is oxidized to the sulfinic derivative (hypotaurine)
only in the presence of cofactorlike compounds such as sul-
fide, methylene blue, and hydroxyalamine (14). Most meta-
bolic pathways give hypotaurine as a precursor to taurine
(15, 16).
In some of our recent work, we examined the reactivity of
the cysteamine metabolites: hypotaurine (cysteamine sulfinic
acid) and taurine (cysteamine sulfonic acid) (17). Oxidation
of hypotaurine by chlorite and chlorine dioxide occurred si-
multaneously at the sulfur center (giving taurine) and at the
nitrogen center (to give the chloramines) (18). On the other
hand, taurine is relatively inert to oxidation by chlorine di-
oxide, a reactive radical species, and acidified bromate (19).
When subjected to the strong oxidizing agent HOCl, the C—
S bond is not cleaved, and reaction occurs exclusively at the
nitrogen center, giving chlorotaurine (17).
The key to understanding the physiological role of
cysteamine is through its oxidation pathway: its reactive in-
termediates and oxidation products. In this manuscript we
report on the oxidation of cysteamine by the mild oxidizing
agents acidic iodate and aqueous iodine. In a nonenzymatic
pathway, could the oxidation of cysteamine yield hypo-
taurine or go all the way to taurine? Specifically, we are
aware that the action of most goitrogenic mechanisms in-
volves the consumption of the iodine atom, which is needed
for thyroid activity (20). What then, is the rate and mecha-
nism by which cysteamine reacts with molecular iodine?
Methods
All experiments were carried out at 25.0 0.5 °C and at a
constant ionic strength of 1.0 mol L–1 (NaClO4). CA, so-
dium perchlorate, and perchloric acid solutions were mixed
in one reactant vessel and iodate (or iodine) solutions in an-
other. All kinetics measurements were performed on a Hi-
Tech Scientific SF61–DX2 double mixing stopped-flow
spectrophotometer. Spectrophotometric determinations were
performed on a PerkinElmer Lambda 25 UV–vis spectro-
photometer.
Stoichiometric determinations
–
The stoichiometry for the CA–IO3 was determined both
in excess iodate and in excess CA. In excess iodate the total
oxidizing power was determined by titration. Excess acidi-
fied iodide was added to the reaction solution, and the re-
leased iodine was titrated against standard thiosulfate.
Spectrophotometry was also used to determine the amount
of iodine formed in excess iodate by its absorbance at
460 nm. In the I2–CA reaction the stoichiometry was deter-
mined by titrating standardized iodine solution from a bu-
rette into a solution of CA of known strength. The end point,
which was enhanced by starch indicator, was detected as the
point where the blue-black color lingered.
Results
Stoichiometry
The stoichiometry of the reaction was strictly 1:6, with
6 mol of cysteamine reacting with 1 mol iodate. This sug-
gested a one-electron oxidation of the sulfur center, and the
only product possible after one-electron oxidation of an
organosulfur center was the disulfide (H2NCH2CH2S-
SCH2CH2NH2, RSSR). Both spectrophotometric and
titrimetric techniques were used to deduce the following
stoichiometry:
Experimental
–
Materials
[R1] IO3 + 6H2NCH2CH2SH → I–
Cysteamine hydrochloride (CA, 2-aminoethanethiol hy-
drochloride) (98%), iodine, potassium iodide (Aldrich), so-
dium perchlorate (98%) (Acros), cystamine dihydrochloride
(MP Biomedical), potassium iodate, perchloric acid (72%),
soluble starch, sodium thiosulfate, and hydrochloric acid
(Fisher) were used without further purification. The concen-
tration of iodine was determined by standardization against
thiosulfate with starch as indicator. This standardization
allowed us to evaluate the absorptivity coefficient of iodine
at its isosbestic point of 460 nm with triiodide as
770 (mol L–1)–1 cm–1. This standardization was carried out
before each series of kinetic experiments were performed
because of the volatile nature of iodine. CA solutions were
prepared just before use and not kept for more than 24 h. All
solutions were prepared using distilled–deionized water
from a Barnstead Sybron Corporation water purification
unit. Inductively coupled plasma mass spectrometry
(ICPMS) was used to show that our aqueous reaction media
+ 3H2NCH2CH2S-SCH2CH2NH2 + 3H2O
Stoichiometry [R1] was obtained only at the stoichio-
metric ratio of 1:6. In lower concentrations of iodate, the re-
action mixture produced a mixture of both cysteamine and
cystamine. These products were confirmed by the NMR
spectrum of the mixture. In stoichiometric excess of iodate,
the reaction products included iodine, which was then spec-
trophotometrically examined to deduce the overall
stoichiometry of [R2]:
–
[R2] 2IO3 + 10H2NCH2CH2SH + 2H+ → I2
+ 5H2NCH2CH2S-SCH2CH2NH2 + 6H2O
If the iodate-to-substrate ratio was greater than 1:5, then
the amount of iodine produced was determined by the
amount of cysteamine as the limiting reagent. There was an
increase in the amount of iodine produced as the ratio was
© 2006 NRC Canada

Chanakira et al.
51
Fig. 1. CA variation in excess oxidant shows higher rates of io-
increased from 1:6. At 1:5, however, the increase saturated,
and no further increase in the production of iodine was ob-
served. The direct reaction of aqueous iodine and cyste-
amine was determined by a rapid and careful titration aided
by starch indicator. Although the end point of the titration
shifted upon prolonged standing, the correct stoichiometry
was attained at the first end point of 1:2, as shown below.
dine production, as well as higher final concentrations of iodine,
at the end of the reaction as the CA concentration is increased.
Trace h shows that at the stoichiometric ratio there is just a tran-
sient formation of iodine. [IO3 ]0 = 0.005 mol L–1, [H+]0 =
–
0.02 mol L–1, [CA]0 = (a) 0.0005, (b) 0.001, (c) 0.002,
(d) 0.003, (e) 0.004, (f) 0.005, (g) 0.015, (h) 0.03 mol L–1.
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
g
[R3] I2 + 2H2NCH2CH2SH → 2I–
+ H2NCH2CH2S-SCH2CH2NH2 + 2H+
The complications observed after allowing the titrated so-
lution to incubate arose from two equilibria: the sulfide–
disulfide equilibrium (22), which sought to reestablish itself
after the depletion of the iodine because of titration with
thiosulfate, and the iodine–thiol equilibrium (23, 24), which
is well-known and has been extensively discussed in this
manuscript under “charge-transfer” complex formation (see
below). Stoichiometry [R3] could also be obtained spectro-
photometrically in an excess of iodine and under high acid
conditions, where both of these complicating equilibria were
suppressed.
f
e
d
c
b
a
h
0
5
10
15
20
25
30
Time (s)
Confirmation of products
Reactions run in excess iodate did not afford sulfate as a
product, as shown by the lack of precipitation of barium sul-
fate with barium chloride. Before barium chloride was
added, excess iodate was removed by the use of excess acid-
ified iodide, followed by evaporation of the iodine formed to
avoid precipitation of barium iodate. Standard proton NMR
spectra were also used to confirm the formation of the
disulfide. Reagent grade cysteamine gave the expected spec-
trum of two triplets owing to the methylene proton sets on
the carbon skeleton. This reagent-grade quality cysteamine
also showed some cystamine impurity. Addition of DCl de-
stroyed the resolution of the cysteamine triplets and also
shifted them downfield because of the protonation of the
amino center. The NMR spectrum of the product confirmed
that the product was the dimeric cystamine. These product
identities were confirmed by an HPLC technique, which uti-
lized an isocratic 98%:2% water–methanol medium. The
aqueous medium was 0.05 mol L–1 phosphate at pH 4.0.
Both cysteamine and cystamine were modified by addition
of acid, and so a control spectrum was initially obtained
with cystamine mixed with acid before product UV spectral
analyses were performed.
tics: an initial quiescent period followed by formation of io-
dine. Trace g, on the other hand, shows two peaks in iodine
concentrations (oligooscillations), with one peak after ap-
proximately 1 s and the other peak (which represented the
maximum concentrations obtained) at 60 s (truncated from
Fig. 1). The only difference in the traces shown in Fig. 1 is
–
the oxidant-to-reductant ratio, R = [IO3 ]0 /[CA]0. The in-
crease in final iodine concentrations formed in going from
trace a to g is justified from the premise that stoichiometry
[R2] is a stepwise combination of rxn. [R1] and the
Dushman reaction, [R4] (5[R1] + [R4]) (27).
–
[R4] IO3 + 5I– + 6H+ → 3I2 + 3H2O
The collapse of observed iodine concentration in increas-
ing [CA]0 to trace h arises from the decrease in R (which is
–
[IO3 ]0/[CA]0) to below that expected for stoichiometry
[R2]. Figure 2a shows a series of experimental traces ob-
tained by varying initial iodate concentrations while keeping
all other species concentrations constant. Traces c–g satisfy
stoichiometry [R2], and thus, at t = ∞ all these traces give
the same iodine absorbance. Expansion of the first 4 s of the
reaction shows a distinct induction period, which could be
related to the initial concentration of iodate. We define an
induction period as the time taken before a rapid and mono-
tonic formation of iodine commences. A linear relationship
(see Fig. 2b) exists between the inverse of this induction pe-
riod and initial iodate concentrations. The intercept on the
iodate axis, the concentration of iodate for an induction pe-
riod of t = ∞ (1/t = 0), confirmed stoichiometry [R1]. The in-
tercept is approximately 0.0005 mol L–1 iodate for an initial
concentration of 0.003 mol L–1 CA (a 1:6 ratio). An induc-
tion time of infinity indicates that no iodine is formed. Any
higher iodate concentration over 0.0005 mol L–1 will yield
iodine as a final product and stoichiometry will shift from
[R1] towards [R2]. In Fig. 2c, increasing the initial acid con-
centration relative to Fig. 2a and amplifying the first 5 s
Reaction dynamics
The reaction dynamics were complex, with oligo-
oscillatory production of iodine (25), transient formation of
iodine, and clock reaction behavior (26). Figure 1 shows all
three types of complex dynamics. This is much more evident
if the region between t = 0 and 5 s is expanded and magni-
fied. Trace h, for example, is at the 1:6 stoichiometric ratio
of rxn. [R1] and should produce no aqueous iodine at the
end of the reaction. Trace h shows, however, an initial rapid
formation of iodine, followed by an equally rapid consump-
tion, all within 3 s. The rest of the traces, a–g, have stoichio-
metric excess of iodate according to [R1], while only traces
a and b have the stoichiometric ratio that satisfies stoichio-
metry [R2]. Traces a–e display clock reaction characteris-
© 2006 NRC Canada

52
Can. J. Chem. Vol. 84, 2006
Fig. 2. (a) Effect of [IO3 ] variation at constant [H+] and [CA]
–
shows, instead of an induction period, a transient peak in io-
dine concentrations, followed by a final and rapid formation
of iodine. In all cases shown in Fig. 2c, the reactions that
form iodine initially dominate those that consume iodine
(even in stoichiometric excess of CA), and hence the ob-
served initial iodine peak. Iodine formation is also almost in-
stantaneous. The onset of prolonged and monotonic iodine
formation is determined by the initial iodate concentrations.
For traces e and f, for example, the high ratio ensures rapid
consumption of the substrate, leaving only the Dushman re-
action, [R4], to form iodine without any other reactions in
the reaction medium that consume iodine.
showing a finite and measurable induction period before forma-
–
tion of iodine. The initial rate of reaction increased as [IO3 ]0
increased; [CA]0 = 0.003 mol L–1, [H+]0 = 0.001 mol L–1, [IO3 ]0 =
–
(a) 0.004, (b) 0.005, (c) 0.006, (d) 0.007, (e) 0.008, (f) 0.009,
(g) 0.01 mol L–1; (b) Linear plot of reciprocal induction time vs.
–
[IO3 ]0 for data shown in Fig. 2a; (c) Effect of progressively in-
creasing [IO3 ] at constant acid and [CA] close to stoichiometry
–
show oligooscillations. Traces below stoichiometry and at
stoichiometry a–c show just the transient formation of iodine.
[CA]0 = 0.015 mol L–1, [H+] = 0.02 mol L–1, [IO3 ]0 = (a) 0.0015,
–
(b) 0.002, (c) 0.0025, (d) 0.003, (e) 0.004, (f) 0.005 mol L–1.
0.16
Lower acid concentrations did not produce such oligo-
oscillatory behavior: only a period of no iodine formation is
observed, followed by monotonic iodine formation. The
strong catalytic effect of acid on the Dushman reaction is re-
sponsible for the olligooscillations. The effect of acid on the
reaction appears much more complex than that of iodate (see
Figs. 3a and 3b). There are several reactions that are viable
in this reaction mixture, and it would appear all of them are
affected by acid; hence, a clean assessment of the effect of
acid is not easily obtained. However, what is obvious from
Fig. 3a is that acid concentrations catalyze the formation of
iodine after the induction period. In all traces shown in
Fig. 3a, the final iodine concentrations were the same (6 ×
10–4 mol L–1, absorbance - 0.46). While trace g went to
completion in less than 250 s, it took more than 2000 s for
trace a to go to completion. A very interesting set of data are
shown in Fig. 3b. In this set of data the reaction conditions
are set to the stoichiometric ratio of [R1], and at t = ∞ no io-
dine is (nor should be) observed. Acid variations at these
conditions show that acid catalyzes the formation of iodine
and retards the consumption of iodine. Hence, high acid con-
centrations will show a wider “excursion” away from the
equilibrium conditions before settling down to stoichiometry
[R1].
Figure 4 shows the expected catalytic effect of iodide
ions. In this case, very low iodide ion concentrations can
drastically reduce the induction period (upon expanding the
first 2 s of the reaction). They also increase the rate and
amount of iodine obtained. This is expected in conditions
where the oxidant to reductant ratio (R) is greater than 0.2,
where the iodide ions initially added to the reaction mixture
combine with those formed in stoichiometry [R1] to fuel the
Dushman reaction, [R4]. This is also evident from Fig. 4,
which shows a long enough time evolution of the reaction.
Other experimental data (not shown) demonstrated a satura-
tion of the reduction of the induction period when the added
iodide ions equaled or exceeded the concentrations of the
substrate. Iodide acts solely as a catalyst: in stoichiometric
excess of the substrate (stoichiometry [R1]), it is used and
released intact at the end of the reaction. Iodide ultimately
does not participate in the overall stoichiometry except in
high excess of iodate conditions, where it is ultimately con-
verted to molecular iodine. Strictly speaking, the Dushman
reaction (27) is an extraneous reaction that only asserts itself
after stoichiometry [R1] of our fiduciary reaction has been
satiated. No simple relationship could be deduced between
induction period and (or) rate of formation of iodine with io-
dide (see Fig. 2b for the iodate dependency). This is due to
the fact that the effect of iodide on the whole reaction is
(a)
g
0.14
0.12
0.10
0.08
0.06
0.04
0.02
0.00
f
e
d
c
b
a
0
10
20
30
40
50
Time (s)
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
(b)
0.000
0.002
0.004
0.006
0.008
0.010
0.012
[IO3–] (mol L–1
)
0.5
(c)
0.4
0.3
0.2
0.1
0.0
f
e
d
c
b
a
0
1
2
3
4
5
Time (s)
© 2006 NRC Canada

Chanakira et al.
53
Fig. 3. (a) Effect of varying acid concentration on the oxidation
Fig. 4. [I–] effect on the iodate oxidation of CA. The effect of
of CA by iodate. There is an increase in rate of iodine formation as
iodide can be seen on the reduced induction period and in-
acid concentration is increased. [CA]0 = 0.003 mol L–1, [IO3 ]0 =
creased final absorbance of iodine as the iodide concentration is
–
0.003 mol L–1, [H+]0 = (a) 0.001, (b) 0.0015, (c) 0.002, (d) 0.0025,
(e) 0.003, (f) 0.00375, (g) 0.005 mol L–1; (b) Effect of acid
variation at stoichiometric point (trace h from Fig. 1). The effect
of acid can be seen in the form of a higher maximum iodine
increased. [CA]0 = 0.005 mol L–1, [H+]0 = 0.02 mol L–1, [IO3 ]0 =
–
0.005 mol L–1, [I–]0 = (a) no iodide, (b) 0.00025, (c) 0.0005,
(d) 0.00075, (e) 0.001 mol L–1.
1.2
concentration. [CA]0 = 0.03 mol L–1, [IO3 ]0 = 0.005 mol L–1,
–
[H+]0 = (a) 0.005, (b) 0.0075, (c) 0.01, (d) 0.0125, (e) 0.015,
(f) 0.02 mol L–1.
e
1.0
d
c
b
0.30
0.8
g
(a)
a
0.25
f
0.6
0.20
e
0.4
0.2
0.0
d
0.15
c
0.10
0.05
0.00
b
a
0
5
10
15
20
25
30
Time (s)
0
20
40
60
80
100
the optical observation of the iodine–CA reaction. Hence,
data interpretation of the iodide effect was not possible, al-
though a cursory glance showed that the iodide effect was
mildly inhibitory.
Time (s)
0.16
0.14
0.12
0.10
0.08
0.06
0.04
0.02
0.00
(b)
f
Mechanism
e
d
The mechanistic basis of the oxidation of CA by iodate–
iodine had fewer possible intermediates in going from CA
(reactant) to cystamine (product) when compared with other
organosulfur oxidations in which the oxidation pathway in-
volves formation of sulfinic and sulfonic acids, as well as
cleavage of the C—S bond to yield sulfate (28, 29). There
are only two possible pathways towards formation of the
dimer: through the cysteamine-thiyl radical (30) and (or) the
cysteamine sulfenic acid. In metal–ion mediated oxidation of
cysteamine, the radical pathway is highly favored (31).
c
b
a
0
1
2
3
4
5
Time (s)
[R5] H2NCH2CH2SH + Mn+ ꢀ H2NCH2CH2S·
+ M(n–1)+ + H+
complex and most likely involves a multiterm rate law. Al-
ternatively (as proved later) iodide ions most likely affect
both the formation and consumption of iodine reactions.
The simple and direct oxidation of CA by aqueous iodine
was also studied spectrophotometrically at 460 nm. Fig-
ure 5a shows that this reaction is relatively slow and is
mildly inhibited by acid. Under low acid concentrations,
Fig. 5b shows that there is an inverse relationship between
the acid concentrations and rate of consumption of iodine.
The positive intercept is important: it suggests two path-
ways, one inhibited by acid and one that is not. The gentle
slope of the plot suggests very mild inhibition by acid. This
linearity is lost at high acid concentrations (Fig. 5c). The ef-
fect of iodide was much more complex (Fig. 6). After mix-
ing iodine and CA, a rapid equilibrium sets up, which forms
some adduct that also absorbs at 460 nm. We could not lo-
cate a wavelength where this adduct did not adversely affect
Our reaction environment has been stripped of metal ions,
making the radical pathway unlikely. Oxyiodine species are
known to predominantly oxidize via two-electron steps,
making the pathway through the unstable sulfenic acid the
most likely pathway (32).
[R6] H2NCH2CH2SH + H2O → H2NCH2CH2SOH
+ 2H+ + 2e–
The sulfenic acid should rapidly combine with an
unoxidized CA molecule (a condensation reaction) to form
cystamine.
[R7] H2NCH2CH2SOH + H2NCH2CH2SH
→ H2NCH2CH2S-SCH2CH2NH2 + H2O
© 2006 NRC Canada

54
Can. J. Chem. Vol. 84, 2006
Fig. 5. (a) Effect of varying acid concentration on the reaction
of CA and iodine. The initial rate of reaction decreases as the
acid concentration is increased. [CA]0 = 0.001 mol L–1, [I2]0 =
0.00044 mol L–1, [H+]0 = (a) 0, (b) 0.001, (c) 0.002, (d) 0.003,
(e) 0.004 mol L–1; (b) Plot of inverse of the initial rate vs. acid
concentrations of the I2–CA reaction for low acid variations for
the type of data shown in Fig. 5a. [CA]0 = 0.001 mol L–1, [I2]0 =
0.00044 mol L–1; (c) Types of initial rates vs. inverse acid de-
pendence plots for the I2–CA reaction for a wider range of acid
concentrations than those selected for Fig. 5b. [CA]0 = 0.001 mol L–1,
[I2]0 = 0.00044 mol L–1.
Fig. 6. Effect of iodide variation on the I2–CA reaction. There is
an increase in the initial absorbance with addition of iodide, but
no parallel nature of the absorbance traces shows little effect of
iodide on the reaction rate, only a change in the equilibrium po-
sition of the I2–CA reaction mixture. [CA]0 = 0.0002 mol L–1,
[I2]0 = 0.00044 mol L–1, [I–]0 = (a) 4.0 × 10–5, (b) 6.0 × 10–5,
(c) 8.0 × 10–5, (d) 1.0 × 10–4 mol L–1.
0.29
0.28
0.27
0.26
(a)
0.25
0.24
0.23
0.26
d
c
b
0.25
a
e
d
c
0.24
0.22
b
0
10
20
30
40
50
60
a
Time (s)
0.21
20
40
60
80
100
120
Time (s)
This can bring about some stoichiometric uncertainties, but
if rxn., [R7] is very rapid, there will be very little deviation
in the expected 1:6 stoichiometry.
6
1.4×10
(b)
Iodate oxidations are strongly acid-catalyzed and will not
proceed except in low pH environments (34). All iodate oxi-
dations reported in this manuscript were at pH conditions
lower than 3.0. We expect, at these pH conditions, that the
amino group of the cysteamine will be protonated and that
the prevailing form of cysteamine will be [H3NCH2CH2SH]+.
The accepted pKa value of [H3NCH2CH2SH]+ is 10.8, with a
Kb of the amino group of cysteamine as 6.3 × 1010 (mol L–1)–1,
which calculates to a full protonation of the amino group in
pH conditions lower than 4.0. Thus any further protonation
to cysteamine will be in addition to the one already resident
on the amino group. Since this protonation is distal to the
active thiol group, it does not strongly influence its reactiv-
ity. We shall represent cysteamine, in this manuscript, how-
ever, as the neutral form.
Sulfenic acids are stronger acids than thiols and, if they
are long-lived, would exist predominantly as zwitterions
(produced in rxn. [R6] and others), +N(H)3CH2CH2SO–.
However, the strongly acidic environment would protonate
the oxygen atom, and it would exist predominantly as
+N(H)3CH2CH2SOH.
6
1.3×10
6
1.2×10
1.1×10
1.0×10
9.0×10
8.0×10
6
6
5
5
0.000
0.001
0.002
0.003
0.004
0.005
0.006
[H+] (mol L–1
)
–
6
6
7
7
1.05×10
–
1.00×10
9.50×10
9.00×10
8.50×10
8.00×10
7.50×10
7.00×10
–
–
–
7
Oxyiodine oxidizing species
–7
The chemistry of oxyiodine species has been extensively
studied, and it is well-established that the major oxidizing
species are HIO2, HOI, and molecular iodine (I2). These spe-
cies are derived from iodate through a series of well-known
reactions (35, 36).
–
7
7
–
0
200
400
600
800
1000
1200
–1
[H+]–1 ((mol L–1
) )
–
[R8] IO3 + 2H+ + I– ꢀ HIO2 + HOI
[R9] HOI + H+ + I– ꢀ I2 + H2O
Any uncondensed sulfenic acid left after all the CA has
been consumed will rapidly disproportionate through stan-
dard pathways to form thiosulfinates (33) as well as CA.
In the absence of iodide, the initiation reaction involves
the direct oxidation of CA by iodic acid.
© 2006 NRC Canada

Chanakira et al.
55
–
[R10] IO3 + H+ ꢀ HIO3
In this reaction scheme we are ignoring the iodine–
triiodide equilibrium (38, 39) by assuming that triiodide re-
acts just as rapidly as iodine.
[R11] HIO3 + H2NCH2CH2SH → H2NCH2CH2SOH
+ HIO2
–
[R21] I2 + I– ꢀ I3 ; KI = 770 (mol L–1)–1
[R12] HIO2 + H2NCH2CH2SH → H2NCH2CH2SOH
+ HOI
This assertion was based on our experiments that showed
that the effect of iodide on the reaction rate was inconclu-
sive. Addition of iodide to the reaction medium changed the
original absorbance, but did not alter the subsequent rate of
consumption of iodine (see Fig. 6).
[R13] HOI + H2NCH2CH2SH → H2NCH2CH2SOH
+ H+ + I–
Using [R18]–[R20], we can derive the following overall
rate law for the consumption of iodine.
Equations [R10]–[R13] are relevant at the beginning of
the reaction before iodide ions accumulate. After iodide con-
centrations have built up to effective levels, the reaction will
proceed through rxn. [R8], as it is the overall rate-
determining step in the consumption of CA. In the presence
of added iodide ions, as in Fig. 4, eqs. [R10]–[R13] are com-
pletely shunted out and are insignificant. The sum of rxns.
[R8] + [R10] + [R11] + [R12] + 2[R13] shows that in each
initial reaction cycle, 1 mol of iodide ultimately produces
2 mol of iodide in a quadratic autocatalytic sequence. This is
obvious because each iodate ion, whose reactivity is trig-
gered off by iodide, will ultimately end up as iodide, effec-
tively doubling the number of iodide ions initially present.
Reaction [R13] is the major oxidation route. This leaves two
important oxyiodine reactions, which involve the dispropor-
tionation of HOI (37). This disproportionation ensures that
there is no accumulation of HOI in the reaction medium, es-
pecially after all the reducing substrate has been consumed.
The stable species allowable at the end of the reaction are
iodate, iodine, and iodide.
d[I2]
[CA]0[I2]
(1 + Kb[H+])
[1]
Rate = −
=
(k1 + k2Kb[H+]
dt
In eq. [1], [CA]0 represents the amount of cysteamine
added to the reaction mixture. This is then partitioned be-
tween the protonated and unprotonated forms of the thiol,
based on the acid concentrations and Kb. Equation [1] will
give different acid dependence behavior depending upon the
relative values of k1 and k2. If k1 >> k2, the general form of
the equation for the initial rate of reaction vs. the inverse of
the acid concentration will give a nearly straight line at acid
concentrations where Kb[H+] << 1, with the second term in
eq. [1] assumed to be negligible. The plot shown in Fig. 5b
was restricted to low acid concentrations. The high acid con-
centration kinetics data were discarded. The low acid con-
centrations plot allows us to estimate the values of k1 and Kb
by assuming that the observed retardation is due to the re-
moval of active cysteamine by the protonation equilibrium.
After deleting the second term in eq. [1], the equation can
then be rearranged and written as:
[R14] HIO2 + I– + H+ ꢀ 2HOI
–
[R15] IO3 + HOI + H+ ꢀ 2HIO2
⎛
⎜
⎜
⎝
⎞
⎟
⎟
⎠
⎛
⎜
⎜
⎝
⎞
⎟
⎟
⎠
1
Kb
1
+
Iodine oxidation
[2]
=
[H ] +
mod(rate)
k1[CA]0[I2]0
k1[CA]0[I2]0
Acid dependence experiments (Figs. 5a and 5b) clearly
indicate an electrophilic mechanism with an initial attack by
iodine on the nucleophilic sulfur atom of CA.
Equation [2] is then used to plot the inverse of the modu-
lus of the initial rate vs. the acid concentrations (see
Fig. 5b), from where we can evaluate values of k1 and Kb.
From the intercept, a value of k1 = 2.7 0.3 (mol L–1)–1 s–1
was deduced. The slope gave Kb = 117 15 (mol L–1)–1.
This value of Kb has no thermodynamic significance apart
from the fact that it fits kinetics data through eq. [2]. This is
a value that is only relevant for the conditions used for
Figs. 5a and 5b. Literature quotes the pKa of the thiol group
of cysteamine as 8.6, but the protonation of the amino group
precludes an accurate measurement of pKa for
[H2NCH2CH2SH2]+, should this exist. Another possible
mechanism for the observed acid retardation of iodine oxida-
tion is the loss of nucleophilicity of the water molecules,
owing to the formation of the hydronium ions. However, this
retardation would be kinetically indistinguishable to the one
suggested through rxn. [R20]. A plot of inverse acid depend-
ence, which includes both high and low acid concentrations,
is shown in Fig. 5c. This is not a linear plot, as expected
from the form of eq. [1]. We could then work with the curve
generated in Fig. 5c and curve-fit it to variables k1, k2, and
Kb. This fitting procedure gave k2 = 0.37 (mol L–1)–1 s–1.
[R16] H2NCH2CH2SH + I-I → H2NCH2CH2S(H)I+ + I–
[R17] H2NCH2CH2S(H)I+ + H2O
→ H2NCH2CH2SOH + 2H+ + I–
The product is then formed by rxn. [R7]. This step
([R16] + [R17]) was qualitatively proved by the experimen-
tal observations of an immediate drop in pH upon mixing io-
dine and CA. Thus, acid dependence experiments can be
justified from the basis of an initial protonation of the thiol
group, which subsequently reacts much more slowly with
the electrophilic iodine atom.
[R18] H2NCH2CH2SH + H+ ꢀ [H2NCH2CH2SH2]+; Kb
[R19] H2NCH2CH2SH + I2 + H2O → H2NCH2CH2SOH
+ 2I– + 2H+; k1
[R20] [H2NCH2CH2SH2]+ + I2 + H2O
→ H2NCH2CH2SOH + 2I– + 3H+; k2
© 2006 NRC Canada

56
Can. J. Chem. Vol. 84, 2006
Table 1. Cysteamine–iodate–iodine reaction.
Reaction No.
Reaction
kf and kr
–
M1
M2
M3
M4
M5
M6
M7
M8
H+ + IO3 ꢀ HIO3
1×107, 1×109
HIO3 + RSH → HIO2 + RSOH
HIO2 + RSH → HOI + RSOH
HOI + RSH → I– + H+ RSOH
RSOH + RSH → RSSR + H2O
2.5×102
5×102
1×103
5×108
IO3 + I– + 2H+ ꢀ HIO2 + HOI
2.8, 1.44×103
2.1×108, 90
3.1×1012, 2.2
8.6×102, 2.00
6.2×109, 8.5×106
2.7
–
HIO2 + I– + H+ ꢀ 2HOI
HOI + I– + H+ ꢀ I2 + H2O
–
M9
IO3 + HOI + H+ ꢀ 2HIO2
–
M10
M11
M12
M13
M14
I2 + I– ꢀ I3
I2 + RSH + H2O → RSOH + 2H+ + 2I–
I3 + RSH + H2O → RSOH + 2H+ + 3I–
1.5
–
RSH + H+ ꢀ RSH2
1×108, 8.5×105
0.37
+
RSH2 + I2 + H2O → RSOH + 3H+ + 2I–
+
Note: RSH = cysteamine; RSOH = unstable cysteamine sulfenic acid intermediate; RSSR = cystamine.
Computer modeling
Fig. 7. Comparison of experimental and calculated absorbance
traces for the I2–CA reaction data shown in Fig. 5a (traces a and
d). The symbols represent calculations.
The reaction scheme was modeled using the Kintecus
software generated by James Ianni. The whole reaction
scheme is shown in Table 1. The 14-reaction scheme was
able to model satisfactorily all of the reaction dynamics ob-
served in Figs. 1–4. The data for the direct reaction of iodine
and CA could also be simulated by the same model after
nullifying rxns. [M1]–[M7] (i.e., setting all their rate con-
stants to zero).
0.26
0.25
0.24
Charge-transfer complexes
0.23
d
While the calculations were able to simulate the general
dynamics of the data shown in Figs. 1–4, they gave much
lower absorbance readings than those shown in these figures.
The absorbance readings only matched the calculations at
the beginning and at the end of the simulations. The experi-
mentally observed intermediate values were much higher,
owing to the expected charge-transfer complexes that
cysteamine is known to form with molecular iodine (23, 40).
a
0.22
0.21
20
40
60
80
100
120
Time (s)
[M2]. After setting up an initial guess value for kM2, the ki-
netics parameters for the rest of the reactions in that chain,
[M3] and [M4], became instantly irrelevant for as long as
they were faster than the set value for kM2 and with the rela-
tive magnitudes in the order of kM3 < kM4. Any reversal in
this order generated large amounts of HIO2 (in excess iodate
conditions), which never decayed as the reaction reached its
end point with respect to iodine production. The data shown
in Fig. 4 could be modeled without reactions [M1] through
[M4]. We settled on a value of kM2 of 250 (mol L–1)–1 s–1 as
the one that gave us the best fit to the induction period. The
two protolytic reactions, [M1] and [M13], were set to be
faster than all reactions in the mechanism. The role of [M1]
was to ensure bimolecular kinetics for rxn. [M2]. The role of
[M13] was to simulate acid retardation observed in the di-
rect oxidation of CA by iodine. The values of kM13 and k–M13
were linked together by the value of Kb derived in this study.
The fact that reaction [M13] was rapid in both directions
made it insignificant with respect to the sequence [M1]–
[M4], since kM2, the rate-determining step for that sequence,
was only 250 (mol L–1)–1 s–1. The effect of sequence
[M10] + [M12] was not utilized in the derivation of rate for
[R22] H2NCH2CH2SH + I2 ꢀ [H2NCH2CH2SH-I+][I–]
Absorbance readings from experiments and simulations
coincided under conditions where iodate was in stoichio-
metric excess over cysteamine such that at the end of the
reaction cysteamine concentrations had vanished, thus pre-
cluding the possibility of rxn. [R22]. While reactions of the
[R22]-type have been reported (40), no specific measure-
ments have been performed for the cysteamine–iodine mix-
tures. Reaction [R22] is suppressed in high acid conditions,
such that it was easier to fit simulations to the data in
Fig. 5a. Figure 7 shows a typical fit to the experimental data
in Fig. 5a at separate acid concentrations.
Adoption of rate constants
The model presented had very few rate constants that
needed to be guessed for the best simulations fit. For exam-
ple, kinetics constants for reactions [M6]–[M10] were ob-
tained from well-established literature values (27, 35, 37, 41,
42). In the absence of added iodide ions (as in data in
Fig. 4), the most important reaction in the whole model was
© 2006 NRC Canada

Chanakira et al.
57
15. J.H. Fellman, T.R. Green, and A.L. Eicher. Adv. Exp. Med.
Biol. 217, 39 (1987).
16. G. Ricci, R. Chiaraluce, G. Pitari, S. Dupre, O. Fujii, Y.
Hosokawa, and K. Yamaguchi. J. Nutr. Sci. Vitaminol. 35, 143
(1989).
eq. [1] and was not very effective in the mechanism for the
overall iodate–CA reaction. A very small degree of retarda-
tion was observed in the iodine–CA reaction upon asserting
this sequence (see Fig. 6). The possible direct reaction of the
protonated thiol with triiodide was also not included in the
mechanism.
17. C.R. Chinake and R.H. Simoyi. J. Phys. Chem. B, 101, 1207
(1997).
18. B.S. Martincigh, C. Mundoma, and R.H. Simoji. J. Phys.
Conclusion
Chem. A, 102, 9838 (1998).
19. R.H. Simoyi, K. Streete, C. Mundoma, and R. Olojo. S. Afr. J.
Chem. 55, 136 (2002).
20. C.C. Capen. Toxicol. Lett. 64/65, 381 (1992).
21. C.J. Doona and D.M. Stanbury. J. Phys. Chem. 98, 12630
(1994).
22. P.J. Hogg. Redox Rep. 7, 71 (2002).
23. J.P. Danehy. Q. Rep. Sulfur Chem. 2, 325 (1967).
24. J. Kurzawa and K. Janowicz. Acta Chim. Hung. 125, 147
(1988).
Our study shows that cysteamine, the well-known precur-
sor to taurine (14), is not easily oxidized to taurine by the
mild oxidizing agent iodine and without the standard P450-
type enzymes and the flavin-containing monooxygenases
(43). In its goitrogenic role (9), cysteamine most likely ab-
stracts the iodine atom in a reversible process to form the
disulfide, which can be converted back to the thiol by a very
mild redox potential.
25. G. Rabai and M.T. Beck. J. Chem. Soc. Dalton Trans. 1669
(1985).
26. M. Melichercik, A. Olexova, and L. Treindl. J. Mol. Catal. A:
Acknowledgement
We would like to thank Itai Chipinda for performing the
HPLC analysis for us. This work was supported by research
grant Nos. CHE 0137435 and CHE 0341769 from the Na-
tional Science Foundation and research grant No. R01
AI051509-01A2 from the National Institutes of Health.
Chem. 127, 43 (1997).
27. H.A. Liebhafsky and G.M. Roe. Int. J. Chem. Kinet. 11, 693
(1971).
28. T.R. Chigwada, E. Chikwana, and R.H. Simoyi. J. Phys.
Chem. A, 109, 1081 (2005).
29. T.R. Chigwada and R.H. Simoyi. J. Phys. Chem. A, 109, 1094
(2005).
30. T. Nauser and C. Schoneich. Chem. Res. Toxicol. 16, 1056
(2003).
31. L. Pecci, G. Montefoschi, G. Musci, and D. Cavallini. Amino
References
1. R.B. Freedman. FEBS Lett. 97, 201 (1979).
2. C.R. Chinake and R.H. Simoyi. J. Phys. Chem. 98, 4012 (1994).
3. B. Kalyanaraman. Biochem. Soc. Symp. 61, 55 (1995).
4. I.R. Wilson and G.M. Harris. J. Am. Chem. Soc. 82, 4515
(1960).
Acids, 13, 355 (1997).
32. R. Okazaki and K. Goto. Heteroat. Chem. 13, 414 (2002).
33. G. Capozzi and G. Modena. Oxidation of thiols. In The chem-
istry of the thiol group. Edited by S. Patai. Wiley and Sons,
New York. 1974. p. 791.
5. I.R. Wilson and G.M. Harris. J. Am. Chem. Soc. 83, 286
(1961).
6. J.M. Geelen, L.A. Monnens, and E.N. Levtchenko. Nephrol.
34. Y. Xie, M.R. McDonald, and D.W. Margerum. Inorg. Chem.
Dial. Transplant. 17, 1766 (2002).
38, 3938 (1999).
7. W.A. Gahl. Eur. J. Pediatr. 162 (Suppl. 1), S38 (2003).
8. R.L. Pisoni, G.Y. Park, V.Q. Velilla, and J.G. Thoene. J. Biol.
Chem. 270, 1179 (1995).
9. V.E. Kimonis, J. Troendle, S.R. Rose, M.L. Yang, T.C.
Markello, and W.A. Gahl. J. Clin. Endocrinol. Metab. 80, 3257
(1995).
35. G. Schmitz. Phys. Chem. Chem. Phys. 1, 1909 (1999).
36. K. Kustin and M. Eigen. J. Am. Chem. Soc. 84, 1355 (1962).
37. L. Kolaranic and G. Schmitz. J. Chem. Soc. Faraday Trans. 88,
2343 (1992).
38. M.-F. Ruasse, J. Aubard, B. Galland, and A. Adenir. J. Phys.
Chem. 90, 4382 (1986).
10. F. Iwata, E.M. Kuehl, G.F. Reed, L.M. Mccain, W.A. Gahl,
and M.I. Kaiser-Kupfer. Mol. Genet. Metab. 64, 237 (1998).
11. O.I. Aruoma, B. Halliwell, B.M. Hoey, and J. Butler. Biochem.
J. 256, 251 (1988).
39. D.H. Turner, G.W. Flynn, N. Sutin, and J.V. Beitz. J. Am.
Chem. Soc. 94, 1554 (1972).
40. W.H. Miller, R.O. Roblin, and E.B. Astwood. J. Am. Chem.
Soc. 67, 2201 (1945).
12. D. Becker, S. Summerfield, S. Gillich, and M.D. Sevilla. Int. J.
41. G. Schmitz. Phys. Chem. Chem. Phys. 2, 4041 (2000).
42. E.T. Urbansky, B.T. Cooper, and D.W. Margerum. Inorg.
Chem. 36, 1338 (1997).
Radiat. Biol. 65, 537 (1994).
13. M.M. Oosthuizen and D. Greyling. Redox Rep. 4, 277 (1999).
14. D. Cavallini, C. De Marco, R. Scandurra, S. Dupre, and M.
Graziani. J. Biol. Chem. 241, 3189 (1966).
43. W.X. Guo, L.L. Poulsen, and D.M. Ziegler. Biochem.
Pharmacol. 44, 2029 (1992).
© 2006 NRC Canada