Kinetic evidence that cysteine reacts with dopaminoquinone via reversible adduct formation to yield 5-cysteinyl-dopamine: An important precursor of neuromelanin

Article abstract of DOI:10.1039/b316294j

The reaction of cysteine (cys) with dopaminoquinone (DQ) to form (mainly) 5-cysteinyl-dopamine (5-cys-DA) is of interest because it is known to play a role in the production of melanin in the mammalian brain. To gain insight into this important reaction, an in vitro detailed kinetic study was undertaken. It has been established that cys reacts with DQ via the initial reversible formation of an intermediate adduct or complex and that this adduct then decomposes to form 5-cys-DA. A little 2-cys-DA, is almost certainly formed at the same time but its presence could not be kinetically investigated. Clarification of the kinetic data was aided by following the reaction of DQ with a cys analogue, mercaptoacetic acid (maa). Maa was found to react in a similar fashion, but also forms, reversibly, a bis-complex. This bis-complex, 2,5-(maa)₂-dopaminoquinone, is in equilibrium with the di-protonated compound but neither of these species reacts further over the timescale employed in these kinetic studies. Equilibrium constants and first-order rate constants have been extracted from the data and the cys complex is found to be weaker than its maa analogue by an order of magnitude (K_cys = (1.09 ± 0.02) × 10^-3; K_1,maa = (7.45 ± 0.11) × 10^-3). (Note that the possibility that cys also forms a bis-complex at much higher cys concentrations cannot be excluded.) The rates of decomposition differ markedly - the cys complex has the value k_cys = 1830 ± 50 s^-1 whereas the rate constant for the decomposition of the maa complex is k_maa = 69.3 ± 0.02 s^-1 and we attribute this difference to the effect of the positive charge carried by the amino-group on cys. Finally, the constants obtained are used to compare the reactivity of thiol addition with ring cyclization (U. El-Ayaan, E. Herlinger, R. F. Jameson, and W. Linert, J. Chem. Soc., Dalton Trans., 1997, 2813-2818) and we show how this has important implications concerning the production of neuromelanin.

Full text of DOI:10.1039/b316294j

View Article Online / Journal Homepage / Table of Contents for this issue
Kinetic evidence that cysteine reacts with dopaminoquinone via
reversible adduct formation to yield 5-cysteinyl-dopamine:
an important precursor of neuromelanin
Guy N. L. Jameson,*a Jie Zhang,a,b Reginald F. Jamesonc and Wolfgang Linert*a
a Institute of Applied Synthetic Chemistry, Vienna University of Technology,
Getreidemarkt 9/163-AC, A-1060 Vienna, Austria. E-mail: gnlj@hotmail.com;
wlinert@mail.zserv.tuwien.ac.at; Fax: ϩϩ 43 1 58801 16299; Tel: ϩϩ 43 1 58801 15350
b Institute of Crystal Materials, Shandong University, Jian, Shandong, China.
E-mail: jzhang@sdu.edu.cn
c Department of Chemistry, The University, Dundee, Scotland, UK DD1 4HN.
E-mail: RJDundee@aol.com
Received 12th December 2003, Accepted 22nd January 2004
First published as an Advance Article on the web 11th February 2004
The reaction of cysteine (cys) with dopaminoquinone (DQ) to form (mainly) 5-cysteinyl-dopamine (5-cys-DA) is of
interest because it is known to play a role in the production of melanin in the mammalian brain. To gain insight into
this important reaction, an in vitro detailed kinetic study was undertaken. It has been established that cys reacts with
DQ via the initial reversible formation of an intermediate adduct or complex and that this adduct then decomposes
to form 5-cys-DA. A little 2-cys-DA, is almost certainly formed at the same time but its presence could not be
kinetically investigated. Clarification of the kinetic data was aided by following the reaction of DQ with a cys
analogue, mercaptoacetic acid (maa). Maa was found to react in a similar fashion, but also forms, reversibly, a
bis-complex. This bis-complex, 2,5-(maa)2-dopaminoquinone, is in equilibrium with the di-protonated compound
but neither of these species reacts further over the timescale employed in these kinetic studies. Equilibrium constants
and first-order rate constants have been extracted from the data and the cys complex is found to be weaker than its
maa analogue by an order of magnitude (Kcys = (1.09 0.02) × 10Ϫ3; K1,maa = (7.45 0.11) × 10Ϫ3). (Note that the
possibility that cys also forms a bis-complex at much higher cys concentrations cannot be excluded.) The rates of
decomposition differ markedly—the cys complex has the value kcys = 1830 50 sϪ1 whereas the rate constant for the
decomposition of the maa complex is kmaa = 69.3 0.02 sϪ1 and we attribute this difference to the effect of the
positive charge carried by the amino-group on cys. Finally, the constants obtained are used to compare the reactivity
of thiol addition with ring cyclization (U. El-Ayaan, E. Herlinger, R. F. Jameson, and W. Linert, J. Chem. Soc.,
Dalton Trans., 1997, 2813–2818) and we show how this has important implications concerning the production of
neuromelanin.
ity of γ-glutamyl transpeptidase has been reported9 in the sub-
stantia nigra of those patients that suffered from Parkinson’s
Introduction
5-cysteinyl-dopamine (5-cys-DA) is found in certain dopa-
minergic regions of the mammalian brain,1–3 including caudate
nucleus, putamen, globus pallidus and substantia nigra. Indeed
chemical degradation of neuromelanin in the substantia nigra
has suggested that 5-cys-DA is a precursor of melanin,4–6 via its
co-polymerisation during the oxidation of dopamine†. During
the progression of Parkinson’s disease, however, increased
ratios7 of 5-cys-DA over dopamine are observed in the substan-
tia nigra. It has therefore been postulated that, rather than
being involved in melanin production, the formation of
5-cys-DA diverts production. In fact, with increasing molar
excess of cysteine, the formation of melanin in vitro is decreased
and ultimately stopped.4 Moreover, due to its lower redox
potential, 5-cys-DA can further oxidize and cyclize intramo-
lecularly to form toxic dihydrobenzothiazines.8
Although 5-cys-DA can be formed by nucleophilic attack by
cysteine (cys) on dopaminoquinone (DQ), concentrations
of cysteine in the brain are generally low. Alternatively, or in
parallel, glutathione, present in much higher concentrations,
could attack to form 5-glutathionyl-dopamine (5-glu-DA),
which could then undergo enzymatic cleavage by γ-glutamyl
transpeptidase and peptidase to form 5-cys-DA. Evidence for
the presence of 5-glu-DA is not forthcoming but elevated activ-
disease.
Previous studies have focused on the interaction of cysteine
with quinones because only cysteinyl catechols have been
unequivocally detected in vivo8 and these studies have employed
continuous electrolytic production of the quinones8,10,11 in the
presence of excess -cysteine. Because of oxidation of the initial
substitution product (i.e. 5-cys-DA) and intramolecular cycliz-
ation by the amine group, a bewildering number of products
were produced and the initial stages of the reaction were
obscured. Xu et al.12,13 have, however, studied this initial step by
first oxidizing dopamine in the absence of a nucleophile and
then adding the nucleophile. They were able to determine the
initial products through a mixture of spectroscopic techniques;
mainly 5-cys-DA, some of the di-substituted product 2,5-di-
cys-DA and a small amount of 2-cys-DA were formed. A
mechanism was proposed10,12 but no kinetic data was supplied
to back their claims. The interpretation of their results was also
hampered by the fact that oxygen was not excluded.
In the present study the oxidation of dopamine to the quin-
one was achieved electrolytically and the reactions with cysteine
and mercaptoacetic acid‡ (maa) followed spectrophoto-
metrically in a stopped-flow apparatus. Mercaptoacetic acid
was included in this investigation in order to clarify the data
† Dopamine: 2-(3,4-dihydroxyphenyl)ethylamine.
‡ Mercaptoacetic acid: 2-mercaptoethanoic acid.
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 7 7 – 7 8 2
777

View Article Online
ϩ
Table 1 The variation of the observed first-order rate constant for the
and to understand the role played, if any, by the -NH3 group
obs
present in cysteine. The absorption of dopaminoquinone14 at
385 nm (ε = 1650 MϪ1 cmϪ1) was used to follow its consump-
tion and, in order to make use of pseudo first-order kinetics,
the respective thiols were always used in large excess. Further-
more, the exclusion of all oxygen has enabled us to study, for
the first time, the initial stages of the basic reactions and to
establish that they involve the initial reversible formation of
adducts (or ‘complexes’) between the quinones and the thiols,
which then react to form the final product, 5-cys-DA or 5-maa-
DA.
decay of DQ, k , with ionic strength, I ([cys] = 4 mM)
cys
obs
cys
pH
I/M
k
/sϪ1
1.08
1.09
1.15
1.09
4.65
4.65
4.65
4.65
0.200
0.300
0.400
0.500
0.300
0.370
0.460
0.500
0.918
0.895
0.814
1.16
345
350
353
333
Materials and methods
Chemicals
Dopamine was obtained from Merck; -cysteine and citric acid
from Sigma; mecaptoacetic acid and acetic acid from Aldrich.
All were used without further treatment.
Generation of dopaminoquinone
Solutions of 1 mM dopamine were made in 0.4 M KCl and
0.1 M HCl. Low pH’s were used to hinder the ring closure
reaction14 of the resultant quinone by protonating the amine
group; high ionic strength was used to facilitate fast oxidation,
thus also reducing ring closure and minimizing the chance of
further oxidation. An IsoTech IPS 1810H Laboratory DC
power supply unit produced a constant voltage across two plat-
inum electrodes in a custom two-compartment glass cell separ-
ated by a glass frit. A current of 40 mA at an emf of 5 V was
maintained for 3½ min to ensure fast conversion of dopamine
to quinone (10–40% conversion depending on the concen-
trations required; the UV-Vis spectrum was measured and
checked before each run of experiments). Anaerobic conditions
were maintained by bubbling argon through the solutions
before, during and after electrolysis.
Fig. 1 Time-dependent electronic spectra of the reaction of dopamino-
quinone with -cysteine, showing the disappearance of quinone at
385 nm and appearance of products absorbing at 255 and 300 nm.
Spectra were taken on a logarithmic timescale from 1 ms–1 s. Initial
concentrations: [DQ] = 0.32 mM, [cys] = 4 mM; pH = 3.15.
obs
cys
The dependence of k
on the concentration of cysteine
followed the expression (2).
(2)
An example of the fit obtained using (2) for a particular pH is
given in Fig. 2 and the results are summarised in Table 2. It was
also established that both A and B are inversely proportional to
the proton concentration (Fig. 3).
Stopped-flow
Solutions of cysteine (2–16 mM) and mercaptoacetic acid
(1–12 mM) were made in various buffers (acetate, citric acid, or
HCl) depending on the pH. The ionic strength was kept con-
stant at 0.5 M with KCl. All solutions (both thiol and quinone)
were transferred to the SX-17MV stopped-flow spectrometer
(fitted with a photo-multiplier detector) from Applied Photo-
physics via gas tight Hamilton syringes. Single wavelength
measurements were carried out at 385 nm and time-dependent
spectra by combining many single wavelength measurements.
Final pH’s were measured using a Radiometer PHM64
instrument.
obs
Fig. 2 Variation of the observed first-order rate constant k with
cys
Results
cysteine concentration for a representative pH of 5.04. The solid line
through the data points follows eqn. (2) using the constants A = 1500
and B = 3.
Cysteine
Experiments were performed over a large range of pH
values (pH = 1–5), concentrations of cysteine (2–16 mM)
and of quinone (0.1–0.4 mM) and over a range of ionic
strengths, I. Time-dependent spectra (see Fig. 1) showed the
disappearance of the quinone and increases in absorbances at
255 and 300 nm corresponding to the formation of both 5 and
2-cys-DA12 (they absorb in the same regions of the electronic
spectrum).
The rate of consumption of dopaminoquinone (DQ) obeyed
the rate law (1) and was found to be essentially independent of
ionic strength (Table 1).
Fig. 3 Plots of A (᭺) and B (᭹) vs. 1/[Hϩ] for the reaction of cys with
DQ. The solid lines represent the best fit lines through the origin. The
(1)
slopes are A: 2.00 0.03 and B: (1.09 0.02) × 10Ϫ3
.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 7 7 – 7 8 2
778

View Article Online
Table 2 Kinetic data obtained for a range of pHs and of cysteine
concentrations for the reaction of cysteine with dopaminoquinone
Table 4 Kinetic data obtained at a range of pHs and concentrations
of the reaction of DQ and maa
obs
cys
obs
/sϪ1
maa
pH
[cys]T/M
k
/sϪ1
A
B
pH
[maa]T/M
k
A
B
C
2.72
0.002
0.004
0.006
0.008
0.010
0.004
0.006
0.008
0.010
0.004
0.006
0.008
0.010
0.004
0.006
0.008
0.010
0.016
0.004
0.006
0.008
0.010
0.002
0.004
0.006
0.008
1.72
5.83
8.58
11.1
15.4
5.44
8.22
10.8
12.7
77.3
113
126
162
90.5
148
211
293
438
159
264
355
422
361
602
835
1060
1500
1400
3
5
2.74
0.002
0.004
0.006
0.008
0.010
0.012
0.002
0.004
0.006
0.008
0.010
0.012
0.001
0.002
0.004
0.006
0.008
0.010
0.012
0.002
0.004
0.006
0.008
0.010
0.012
0.001
0.004
0.006
0.008
0.010
0.006
0.008
0.010
0.012
0.002
0.004
0.006
0.008
0.010
0.012
1.60
2.09
2.15
1.92
1.53
1.35
2.73
5.40
4.92
4.68
4.20
3.78
8.06
9.83
950
3
48000
2.76
4.11
4.22
3.16
4.07
2650
5800
40
75
50000
55000
21000
33000
15
18
10.6
10.8
8.99
7.53
6.43
4.45
5.04
55000
30
4.22
4.45
14.2
8800
90
51000
53000
16.6
15.6
13.9
11.3
11.2
18.0
23.7
23.2
22.8
18.1
30.1
28.8
25.4
23.8
35.0
35.4
32.5
29.1
26.5
24.5
220000
120
16000
210
Table 3 The variation of ionic strength, I, with the observed first-
obs
maa
order rate constant, k , of the reaction of maa with DQ ([maa] =
4.84
5.04
36000
56000
520
820
80000
8 mM)
obs
pH
I/M
k
/sϪ1
maa
120000
4.66
4.65
4.64
4.65
0.300
0.370
0.460
0.500
24.7
23.8
26.3
25.6
Mercaptoacetic acid
The rate of consumption of dopaminoquinone once again
obs
maa
followed the pseudo first-order expression (1) and k
was
once again independent of ionic strength (Table 3). The
obs
maa
dependence of k
on mercaptoacetic acid concentration
was shown to follow eqn. (3). A representative example
of the fits obtained is presented in Fig. 4 and the results
are summarized in Table 4. Although A and B are inversely
proportional to the proton concentration, the plot of C vs.
1/[Hϩ]2 is linear with an intercept on the ordinate axis (Figs. 5
and 6).
Fig. 5 Plots of A (᭺) and B (᭹) vs. 1/[Hϩ] for the reaction of maa with
dopaminoquinone. The solid lines represent the best fit straight lines
through the origin. The slopes are A = 0.516 0.003 and B = (7.45
(3)
0.11) × 10Ϫ3
.
Discussion
Cysteine
The departure from first-order dependence of the rate on one
of the reactants that is illustrated in Fig. 2, has been known for
some time. It has been shown to be explicable in terms of
the prior, reversible, formation of a reactive intermediate.15,16
Indeed similar curves have been observed while investigating
the addition of sulfite to p-benzoquinone.17 Note that inde-
pendence of the observed rate constants with ionic strength
(Table 1) when using cysteine as nucleophile neither supports
nor disproves the existence of this intermediate because one of
obs
Fig. 4 Variation of observed first-order rate constant, k , with maa
maa
concentration at pH = 2.74. The solid line is a theoretical fit using
equation (3) with A = 950, B = 3, and C = 48000.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 7 7 – 7 8 2
779

View Article Online
Scheme 1 The proposed reaction mechanism of cysteine or mercaptoacetic acid with dopaminoquinone. Note that only the formation of 5-cys-DA
and of 5-maa-DA is depicted. Formation of the 2-substituted products could occur in a similar manner but is disfavoured, possibly for steric reasons.
would be equally valid to describe reversible formation of the
intermediate cys-DQ as an attack by the deprotonated cysteine
thiolate on the quinone. In other words the proton could be
lost before reaction rather than during (as described by
eqn. (4)). This ambiguity would of course be resolvable if the
reaction produced directly the product, without the reversible
formation of an intermediate. In the absence of a unique
expression, the following formulation was adopted because the
resulting equations are simpler and only differ in a numerical
factor, namely the protonation constant of the SϪ group of the
cysteine anion.§
The total amount of quinone is given by eqn. (6).
Fig. 6 Plot of C vs. 1/[Hϩ]2 for the reaction of maa with dopamino-
quinone. The solid line represents the best fit straight line through the
[DQ]T = [DQ] ϩ [cys-DQ] =
data points (the ordinate axis is depicted logarithmically to show clearly
[DQ] {1 ϩ (Kcys[cys]T/[Hϩ])} (6)
the goodness of fit). The slope is (5.85 0.11) × 10Ϫ6 and the intercept is
(5.02 0.08) × 104.
Therefore the rate of consumption of dopaminoquinone is
given, from eqns. (4) and (5), by eqn. (7):
the feasible reactants, namely the cysteine, bears an overall
charge of zero at most of the pH’s under investigation. This is
not the case with mercaptoacetic acid.§
(7)
To take into account the form of the rate eqn. (2) the follow-
ing reaction sequence is proposed, eqns. (4) and (5). (See also
Scheme 1.) Please note that proton ambiguity means that it
Making use of eqns. (6) and (7), and putting [cys] = [cys]T
because cysteine is in large excess yields eqn. (8).
(4)
(5)
(8)
§ The protonation constants24 for cysteine are: log K1 = 10.29, log K2 =
8.15, log K3 = 1.88; and maa log K1 = 10.11, log K2 = 3.48.
Thus, comparing eqn. (8) with eqns. (1) and (2), the slope of
A vs, 1/[Hϩ] = kcysKcys = 2.00 0.03, and that of B vs. 1/[Hϩ] =
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 7 7 – 7 8 2
780

View Article Online
Kcys = (1.09 0.02) × 10Ϫ3. The ratio of the two slopes gives kcys
The values accepted are Kcys = (1.09 0.02) × 10Ϫ3 and kcys
1830 sϪ1
.
=
Thus the plot of C vs. 1/[Hϩ]2 yields the following: the slope =
K1,maaK2,maa = 5.85 × 10Ϫ6 which makes K2,maa = (7.85 0.11) ×
10Ϫ4. The intercept = K1,maaK2,maaβ2H = (5.02 0.08) × 104 which
gives βH2 = 8.58 × 109 (or log βH2 = 9.94).
.
Mercaptoacetic acid
Conclusions
The method employed above can be used, mutatis mutandis, to
explain the rate eqn. (3) obtained for the reaction of maa with
DQ but with the reversible attachment of a second maa allowed
for. This second maa is presumably bound para to the first,
i.e. in the 2-position. (The structure of the bis-complex is
formulated in Scheme 1.) The following equilibria are involved
(eqns. (9) and (10)).
There are relatively few mechanistic studies of the addition
of sulfur nucleophiles to quinones and until now none has
provided clear evidence for the existence of a tetrahedral inter-
mediate (Scheme 1), similar to those found in other addition–
elimination reactions. The existence of such an intermediate is
kinetically important and has mechanistic ramifications.
Firstly, production of the product is a first-order process and
not, as is often assumed, a second-order step. This fundamental
difference may sometimes be over-looked when the range of
concentrations used is too small. This has been the case in pre-
vious studies of additions to quinones13,18,19 and is because at
(9)
(10)
obs
lower cys concentrations the curvature of plots of [cys] vs. k
cys
T
The total amount of quinone is thus given by eqn. (11):
(e.g. Fig. 2) is less pronounced and can be assumed to be linear
and thus obeying simple second-order kinetics. Indeed, we only
became aware of this problem when investigating the reaction
of maa with DQ when the deviation from second-order kinetics
is very pronounced (see Fig. 4).
(11)
Secondly, intermediate formation allows us to explain the
difference in behaviour between cysteine and mercaptoacetic
acid. The stability constant for the formation of maa-DQ is
almost an order of magnitude greater than that for cys-DQ
(7.45 × 10Ϫ3 in contrast to 1.09 × 10Ϫ3). This could be ascribed
to steric effects, but the great difference in the rate constant for
the decomposition step (ca. 70 sϪ1 for the maa intermediate and
1830 sϪ1 for the corresponding cys complex) strongly supports
the suggestion that this is due to the positive charge on the
amino group in cysteine, possibly by stabilizing build-up of
negative charge on the carbonyl oxygen of C4 (see Scheme 1).
Furthermore, the relative weakness of the first association
equilibrium for cysteine with dopaminoquinone coupled with
its high rate of reaction is sufficient to explain why a second
association step is not observed—at least with the cysteine
concentrations employed in this study.
Because both reactants are charged species, the independence
of the observed rate constants with ionic strength confirms the
existence of a reactive intermediate. Furthermore, the form of
the experimental rate equation, eqn. (3), taken in conjunction
with the speciation, shows that only the mono-adduct reacts,
i.e.:
(12)
Making use of eqns. (11) and (12) and putting [maa] = [maa]T
yields:
The detailed mechanistic information provided in this paper
is of great value when discussing the relative importance of
possible pathways in the formation of pheomelanin and neuro-
melanin, and especially when abnormal conditions prevail as
is the case in neurodegenerative diseases such as Parkinson’s
disease. Indeed, this provided the impulse to undertake this
study. Our previous work on catecholamine oxidations14,20–22
has shown that the initial product, a quinone, undergoes intra-
molecular Michael addition of the amine side-chain at C6 to
form leucodopaminochrome. Further oxidations and conden-
sations lead finally to melanin. Neuromelanin is found to con-
tain 5-cysteinyl conjugates and levels of 5-cys-DA are more
pronounced during the development of Parkinson’s disease.7
Addition of cysteine is therefore in competition with the ring
cyclized product. With this study we are now in a position to
compare the relative importance of these two reactions. The
dependence of the observed first-order rate of decay of
(13)
Comparison of eqns. (1), (3), and (13) enables kmaa and K1,maa
to be calculated from the plots of A and B vs. 1/[Hϩ]. The values
accepted are K1,maa = (7.45 0.11) × 10Ϫ3 and kmaa = 69.3 0.1
s
Ϫ1. Although in large excess, maa did not remove all quinone.
This implies that the formation of product (5-maa-DA) is in
fact reversible and therefore the rate constant obtained is a
combination of the forward and backward reactions (kϩmaa and
kϪmaa). The complicated nature of the overall reaction means that
it is not possible to measure an equilibrium constant for this
reaction and therefore not possible to separate kϩmaa and kmϪaa
.
The plot of C vs. 1/[Hϩ]2 is linear with a slope of (5.85 0.11)
× 10Ϫ6, and with an intercept of (5.02 0.08) × 104. This
dopaminoquinone to leuchodopaminochrome on pH is given
obs
in El-Ayaan et al.14 At physiological pH (≈ 7) k is calculated
implies that the bis-complex exists in two forms described by
eqn. (14):
cyc
to be 0.37 sϪ1. Should cysteine be present we can use eqn. (8) to
obs
cys
calculate k for the same pH and any required cysteine concen-
tration. It is believed that cysteine concentrations in the brain
23
are in the order of 31–95 nmol gϪ1 or approximately 50 µM.
(14)
obs
This produces a value for k of 644 sϪ1
.
cys
We suggest that the protonation is of the two –OϪ functions
on the ring (see Scheme 1). This means that the last term in the
denominator of the theoretical rate equation, eqn. (13),
becomes eqn. (15):
This is an interesting result in that it suggests that the pres-
ence of cysteine would completely inhibit ring cyclization. The
fact that neuromelanin contains lower amounts of 5-cysteinyl
conjugates suggests two possibilities. (i) The quinone concen-
tration is higher than that of cysteine but this in turn implies
that oxidation of dopamine is relatively high. (ii) Oxidation of
K
1,maaK2,maa[maa]T2 (1 ϩ β2H[Hϩ]2)/[Hϩ]2
(15)
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 7 7 – 7 8 2
781

View Article Online
4 R. Carstam, C. Brinck, A. Hindemith-Augustsson, H. Rorsman and
dopamine occurs in a region where cysteine is largely excluded
as suggested by Carstam et al.4
E. Rosengren, Biochim. Biophys. Acta, 1991, 1097, 152–160.
5 G. Odh, R. Carstam, J. Paulson, A. Wittbjer, E. Rosengren and
H. Rorsman, J. Neurochem., 1994, 62, 2030–2036.
6 L. Zecca, C. Mecacci, R. Seraglia and E. Parati, Biochim. Biophys.
Acta, 1992, 1138, 6–10.
7 B. Fornstedt, A. Brun, E. Rosengren and A. Carlsson, J. Neural
Transm., 1989, 1, 279–295.
8 F. Zhang and G. Dryhurst, J. Med. Chem., 1994, 37, 1084–1098.
9 J. Sian, D. T. Dexter, A. J. Lees, S. Daniel, P. Jenner and C. D.
Marsden, Ann. Neurol., 1994, 36, 356–361.
10 X. Huang, R. Xu, M. D. Hawley, T. L. Hopkins and K. J. Kramer,
Arch. Biochem. Biophys., 1998, 352, 19–30.
11 X.-M. Shen and G. Dryhurst, Chem. Res. Toxicol., 1996, 9, 751–763.
12 R. Xu, X. Huang, K. J. Kramer and M. D. Hawley, Bioorg. Chem.,
1996, 24, 110–126.
13 R. Xu, X. Huang, T. D. Morgan, O. Prakash, K. J. Kramer and
M. D. Hawley, Arch. Biochem. Biophys., 1996, 329, 56–64.
14 U. El-Ayaan, E. Herlinger, R. F. Jameson and W. Linert, J. Chem.
Soc., Dalton Trans., 1997, 2813–2818.
It is possible, however, that 5-cys-DA is formed by the addi-
tion of glutathione followed by cleavage of the initial product
(5-glutathionyl-dopamine) by γ-glutamyl transpeptidase and
peptidase. Credence is given to this mechanism by the report of
elevated activity9 of this enzyme during Parkinson’s disease.
Preliminary experiments in our laboratory suggest that,
although complicated, glutathione reacts with dopaminoquin-
one in an analogous fashion to cysteine but slower. Glutathione
is present in much higher concentrations than cysteine and
therefore kgluobs will again be several orders of magnitude higher
than the corresponding rate constant for ring cyclization. These
results suggest that (ii) above is the more likely explanation for
the 5-cys-DA concentrations observed in vivo but much more
work, outwith the scope of the present purely chemical study,
needs to be done.
15 W. G. Movius and R. G. Linck, J. Am. Chem. Soc., 1969, 91, 5394–
5396.
16 W. G. Movius and R. G. Linck, J. Am. Chem. Soc., 1970, 92, 2677–
2683.
Acknowledgements
17 M. P. Youngblood, J. Org. Chem., 1986, 51, 1981–1985.
18 A. Thompson, E. J. Land, M. R. Chedekel, K. V. Subbarao, and
T. G. Truscott, Biochim. Biophys. Acta, 1985, 843, 49–57.
19 M. R. Chedekel, E. J. Land, A. Thompson and T. G. Truscott,
J. Chem. Soc., Chem. Commun., 1984, 1170–1172.
20 W. Linert, R. F. Jameson, and E. Herlinger, Inorg. Chim. Acta, 1991,
187, 239–247.
Thanks for financial support are due to the “Fonds zur
Förderung der Wissenschaftlichen Forschung in Österreich”
(Project 15874-NO3) and to the “Jubiläumsfonds der Öster-
reichischen Nationalbank” (Project 10668).
21 W. Linert, E. Herlinger and R. F. Jameson, J. Chem. Soc.,
Perkin Trans. 2, 1993, 2435–2439.
22 U. El-Ayaan, R. F. Jameson, and W. Linert, J. Chem. Soc.,
Dalton Trans., 1998, 1315–1320.
References
1 E. Rosengren, E. Linder-Eliasson and A. Carlsson, J. Neural
Transm., 1985, 63, 247–253.
2 B. Fornstedt, E. Rosengren and A. Carlsson, Neuropharmacology,
1986, 25, 451–454.
3 B. Fornstedt, I. Bergh, E. Rosengren and A. Carlsson,
J. Neurochem., 1990, 54, 578–586.
23 A. Palumbo, M. d’Ischia, G. Misuraca, L. De Martino and G. Prota,
Biochim. Biophys. Acta, 1995, 1245, 255–261.
24 D. D. Perrin, ‘IUPAC Chemical Data Series, No. 22: Stability
Constants of Metal-Ion Complexes, 2nd Edn’, 1979.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 7 7 7 – 7 8 2
782