Characterization of products from the reactions of dopamine quinone with N-acetylcysteine

Article abstract of DOI:10.1006/bioo.1996.0011

The reactions between electrochemically prepared dopamine (DA) quinone and N-cetylcysteine (NACγSH), a protein model nucleophile, have been investigated at pH 7 and pH 2. Major products were purified by semipreparative reversed-phase liquid chromatography and identified by mass spectrometry and nuclear magnetic resonance spectroscopy to be nucleophilic addition products of the quinone with NACγSH. The principal product is the monoaddition adduct of the quinone at C(5) of the ring, 5-5-(N-acetylcysteinyl)dopamine. The diaddition adduct of the quinone at C(2) and C(5) of the ring, 2,5-S,S'-di(N-acetylcysteinyl)dopamine, and the monoaddition product at C(2) of the ring, 2-S-(N-acetylcysteinyl)dopamine, are produced to lesser extents. The relative molar ratio of 2-S-(N-acetylcysteinyl)dopamine, 5-S-(N-acetylcysteinyl)dopamine, and 2,5-S,S'-di-(N-acetylcysteinyl)dopamine produced at pH 7 is approximately 8:59:33, whereas at pH 2 the ratio is 10:89:1. The uv/vis spectroscopic analysis shows that the two monoaddition products have maxima at 258 and 294 nm, whereas the diaddition product has a maximum at 276 nm and a shoulder at 302 nm. Cyclic voltammetry and chronoamperometry demonstrate that these adducts are oxidized more readily than DA, which causes the anodic current for the oxidation of DA in the presence of NACγSH to be kinetically controlled by the rates of the addition reactions. A reaction scheme is proposed for the formation of these products.

Full text of DOI:10.1006/bioo.1996.0011

BIOORGANIC CHEMISTRY 24, 110–126 (1996)
ARTICLE NO. 0011
Characterization of Products from the Reactions of Dopamine
Quinone with N-Acetylcysteine
RONGDA
X
U,* XIN
H
UANG,* KARL J. KRAMER,† AND M. DALE
H
AWLEY
*
*
Department of Chemistry, Kansas State University, Manhattan, Kansas 66506; and †U.S. Grain
Marketing Research Laboratory, Agricultural Research Services, United States Department of
1
Agriculture, Manhattan, Kansas 66502
Received August 3, 1995
The reactions between electrochemically prepared dopamine (DA) quinone and N-
acetylcysteine (NACySH), a protein model nucleophile, have been investigated at pH 7 and
pH 2. Major products were purified by semipreparative reversed-phase liquid chromatography
and identified by mass spectrometry and nuclear magnetic resonance spectroscopy to be
nucleophilic addition products of the quinone with NACySH. The principal product is the
monoaddition adduct of the quinone at C(5) of the ring, 5-S-(N-acetylcysteinyl)dopamine.
The diaddition adduct of the quinone at C(2) and C(5) of the ring, 2,5-S,SЈ-di-
(N-acetylcysteinyl)dopamine, and the monoaddition product at C(2) of the ring,
2
2
-S-(N-acetylcysteinyl)dopamine, are produced to lesser extents. The relative molar ratio of
-S-(N-acetylcysteinyl)dopamine, 5-S-(N-acetylcysteinyl)dopamine, and 2,5-S,SЈ-di-(N-
acetylcysteinyl)dopamine produced at pH 7 is approximately 8:59:33, whereas at pH 2 the
ratio is 10:89:1. The uv/vis spectroscopic analysis shows that the two monoaddition products
have maxima at 258 and 294 nm, whereas the diaddition product has a maximum at 276 nm
and a shoulder at 302 nm. Cyclic voltammetry and chronoamperometry demonstrate that
these adducts are oxidized more readily than DA, which causes the anodic current for the
oxidation of DA in the presence of NACySH to be kinetically controlled by the rates of the
addition reactions. A reaction scheme is proposed for the formation of these products.
 1996
Academic Press, Inc.
INTRODUCTION
2
Metabolism of dopamine (DA), which is a neurotransmitter in the central ner-
vous system of humans and other vertebrate animals, normally involves the forma-
tion of 3-methoxytyramine via O-methylation, 3,4-dihydroxyphenylacetic acid and
4-hydroxy-3-methoxyphenylacetic (homovanillic) acid via oxidative deamination,
and norepinephrine via ͱ-hydroxylation (1). In insects, DA is a neurotransmitter
and is N-acylated to N-acetyldopamine and N-ͱ-alanyldopamine, the main precur-
1
The Agricultural Research Service, USDA is an equal opportunity/affirmative action employer and
all agency services are available without discrimination.
2
Abbreviations used: DA, dopamine; NACySH, N-acetylcysteine; LC, liquid chromatography; FAB–
MS, fast atom bombardment mass spectrometry; NMR, nuclear magnetic resonance spectroscopy; CV,
cyclic voltammetry; CA, chronoamperometry; EC, electrochemical detector; DEPT, distortionless en-
hancement by polarization transfer; 5-NACyS-DA, 5-S-(N-acetylcysteinyl)dopamine; 2-NACyS-DA, 2-
S-(N-acetylcysteinyl)dopamine; 2,5-di-NACyS-DA, 2,5-S,SЈ-di-(N-acetylcysteinyl)dopamine.
1
10
0
045-2068/96 $18.00
Copyright  1996 by Academic Press, Inc.
All rights of reproduction in any form reserved.

DOPAMINE QUINONE AND N-ACETYLCYSTEINE
111
sors for quinonoid metabolites involved in sclerotization reactions that lead to
adduct and cross-link formation with proteins (2). In addition, oxidation of DA to
DA quinone is often observed (3). In insects at physiological pH, oxidation of
DA can involve an intramolecular cyclization reaction that ultimately leads to the
formation of a black, insoluble melanin polymer for pigmentation of the cuticle
(2). Dopamine oxidation and melanin formation might also be associated with
Parkinson’s disease, where DA-containing cells in the nervous system are suscepti-
ble to degeneration (4, 5). DA quinone can also react with nucleophiles such as
glutathione, cysteine (6, 7), and sulfhydryl groups in proteins (8, 9). Some of these
reactions occur physiologically; for example, the nucleophilic addition product, 5-
S-cysteinyldopamine, has been detected in striatal tissue from various species (10).
The reactions of DA quinone with
L-cysteine at physiological pH have been
studied previously by continuously electrolyzing a mixture of DA and excess
L
-
cysteine (11). Since the initial cysteinyl adducts are oxidized to their corresponding
quinones at the applied potential, and since the amine moiety of the cysteinyl
adduct also functions as a nucleophile, numerous products were formed, many of
which were not identified.
The goal of our work was to clarify the initial steps of the oxidation reaction
pathway of DA when a model protein nucleophile is present in excess. To determine
the initial site(s) of nucleophilic attack on DA quinone, DA was first oxidized in
the absence of a nucleophile to produce DA quinone. The electrolysis was termi-
nated before mixing DA quinone with a nucleophile. This procedure minimizes
the generation of additional oxidation products that would be formed if excess
chemical oxidant were used to oxidize DA or if exhaustive electrolysis of DA were
carried out in the presence of a nucleophile. Since the amino group in cysteine and
cysteinyl adducts can act as a nucleophile (11), the amino group was acetylated in
order to render it unreactive.
Products from the reactions of DA quinone with NACySH were analyzed using
reversed-phase liquid chromatography (LC). The major products were purified
using semipreparative LC and characterized by fast atom bombardment mass spec-
trometry (FAB–MS), nuclear magnetic resonance spectroscopy (NMR), uv/vis
spectroscopy, and cyclic voltammetry (CV). CV was used to characterize the electro-
chemical behavior of this system. Rate constants for the addition reactions were
estimated using chronoamperometry (CA).
MATERIALS AND METHODS
Chemicals
The following chemicals were obtained from commercial sources and used as
3
received: DA and NACySH (Sigma Chemical Co., St. Louis, MO); formic acid,
ammonium formate and disodium ethylenediaminetetraacetate (EDTA) (Fisher
Scientific Co., Pittsburgh, PA); methanol (uv cutoff: 204 nm) (Baxter Healthcare
3
Mention of a proprietary product does not constitute a recommendation by the USDA.

112
XU ET AL.
Corporation, Burdick & Jackson Division, Muskegon, MI); KCl (Mallinckrodt Spe-
cialty Chemicals, Chesterfield, MO); and HCl (J. T. Baker Chemical Co., Phil-
lipsburg, NJ).
Electrochemical Preparation of DA Quinone
DA was oxidized in 0.01
M
HCl and 0.09
M
KCl (pH 2.0) in a custom-made
coulometric cell. A platinum gauze working electrode and a Ag/AgCl (saturated
KCl) reference electrode were placed in the sample compartment. A platinum
auxiliary electrode was separated from this compartment by means of a 0.45-Ȑm
nylon 66 membrane so that electrolysis could be conducted efficiently and without
mixing of the anode and cathode reaction products. A custom-built three-electrode
potentiostat was used to apply a constant potential and to give instantaneous analog
current and digital coulomb readouts (12). The potential for coulometric oxidation
of DA was controlled at 700 mV vs Ag/AgCl (saturated KCl). Electrolysis was
terminated when the current-time curve reached a plateau value, indicating that
DA oxidation was complete. Electrolysis for 3 min was sufficient for complete
oxidation (Ͼ99%) of 0.5 ml of 1 m
M
DA. DA quinone was found to be relatively
stable at pH 2.0 in a separate uv spectrophotometric experiment with the decrease
in absorbance at 390 nm being less than 0.75% after 3 min (Huang et al., unpub-
lished data).
Analytical LC and Electrochemical Studies of the Reactions of DA
Quinone with NACySH
The compositions of reaction mixtures that resulted from mixing 0.4 ml 1 m
M
DA quinone with 0.2 ml 10 m NACySH at either pH 7.4 or pH 2.0 were analyzed
M
by LC. The LC system consisted of a Beckman (Berkeley, CA) Model 332 gradient
liquid chromatography system equipped with two Model 110A pumps and a Model
4
20 controller, a Hewlett–Packard (Palo Alto, CA) HP 8452A diode array spectro-
photometer equipped with a 1-cm quartz flow cell (Pyrocell Manufacturing Co.,
Inc., Westwood, NJ), and a Bioanalytical Systems (West Lafayette, IN) LC-4B dual
amperometric detector connected to a Hewlett–Packard HPLC ChemStation via
a Hewlett–Packard 35900 multichannel interface. The uv/vis spectra of the LC
effluent were recorded every 10 s for the duration of the experiment within the
2
20- to 800-nm wavelength range. The dual amperometric detector that was used
for electrochemical detection of the LC effluent consisted of two glassy carbon
working electrodes, a Ag/AgCl (3 KCl) reference electrode, and a stainless
M
steel auxiliary electrode. Electrode potentials were 800 and Ϫ100 mV. The former
potential is sufficient to oxidize DA and all N-acetylcysteinyl adducts of DA,
whereas the latter potential is sufficient to reduce DA quinone and all adduct
quinones. The two working electrodes were arranged in a parallel configuration so
that both electrochemically oxidizable species and reducible species in the LC
effluent could be detected simultaneously. Two chromatograms were recorded from
the electrochemical (EC) detector and are referred to as LC–EC (oxidation) and
LC–EC (reduction) chromatograms. Separation was achieved on a Microsorb-MV
C18 column (5 Ȑm, 4.6 ϫ 250 mm) (Rainin Instrument Co., Inc., Woburn, MA)
with a binary mobile phase system in which solvents A and B were used. Solvent

DOPAMINE QUINONE AND N-ACETYLCYSTEINE
113
A was 150 m
.0), and solvent B was 50% methanol, 180 m
formate, and 0.1 m
M
formic acid, 30 m
M
ammonium formate, and 0.1 m
M
EDTA (pH
ammonium
3
M
formic acid, 8 m
M
M
EDTA (pH 3.0). The mobile phase gradient was 0 min, 95%
solvent A and 5% solvent B; 0–30 min, linear gradient from 5% solvent B to 30%
solvent B; 30–40 min, linear gradient from 30% solvent B to 100% solvent B. The
flow rate was at 1 ml/min; a 180-Ȑl sample injector loop was used.
A Bioanalytical Systems BAS-100W electrochemical system was used to perform
2
CV and CA. A BAS glassy carbon electrode with an area of 7.1 mm was used as
the working electrode. The reference and auxiliary electrodes were a Ag/AgCl
(
saturated KCl) electrode and a Pt wire electrode, respectively. A small electrochem-
ical cell with a capacity of 1 ml was used. CV of 0.3 m DA in the presence of 0,
.3, or 1.5 m NACySH in 0.1 phosphate buffer (pH 7.0) and 0.3 m DA in the
presence of 0 or 3.0 m NACySH in 0.01 HCl and 0.09 KCl (pH 2.0) was
carried out. The scan rate was 200 mV/s. CA data were acquired for 0.3 m DA
in the absence and presence of 3.0 m NACySH. The potential was stepped from
Ϫ0.2 to 0.6 V when in 0.1 phosphate buffer at pH 7.0 and from 0 to 0.8 V when
in 0.01 HCl and 0.09 KCl at pH 2.0. These oxidation potentials were sufficiently
M
0
M
M
M
M
M
M
M
M
M
M
M
positive to oxidize DA and all N-acetylcysteinyl adducts under these experimen-
tal conditions.
Synthesis and Purification of Products from Reaction of DA
Quinone with NACySH
A DA quinone solution, prepared by the electrolysis of 10 ml 15 m
M
DA in 0.01
M
HCl and 0.09 KCl, was added dropwise to 5 ml 150 m NACySH in 0.5
M
M
M
phosphate buffer (pH 7.4). The final reaction mixture was colorless and its pH
remained at 7.4. The reaction mixture was lyophilized to reduce the volume, and
the concentrated solution was filtered and subjected to a semipreparative LC separa-
tion. The LC system described above was used, but with the electrochemical detector
disconnected. Separation was achieved on a Phenomenex (Torrance, CA) C18
semipreparative column (10 Ȑm, 250 ϫ 10 mm). The mobile phase gradient em-
ployed was 0 min, 95% solvent A and 5% solvent B; 0–40 min, linear gradient from
5% solvent B to 40% solvent B. The flow rate was 4 ml/min; a 5-ml sample loop
was used. Each product effluent was collected and subjected to a desalting step
using the semipreparative LC column. After a product solution was injected onto
the column, the salt components were eluted first using distilled water as the mobile
phase; the product was then eluted using 25% methanol. The effluent containing
the product was subsequently lyophilized to dryness.
Characterization of Products
FAB–MS was performed on a Hewlett–Packard HP 5989A MS Engine. Proton
NMR and distortionless enhancement by polarization transfer (DEPT) NMR were
performed with a Varian (Palo Alto, CA) UNITYplus 400 MHz spectrometer.
Products were dissolved in 0.8 ml D O. The uv/vis spectroscopic measurements of
2
the reaction products in 0.01
M
HCl were performed using the HP 8452A diode-
array spectrophotometer with a 1.0 ϫ 0.4-cm quartz cuvette. Cyclic voltammograms

114
XU ET AL.
of reaction products in 0.1
M
phosphate buffer (pH 7.0) were obtained using the
BAS-100W electrochemical system. The scan rate was 200 mV/s.
RESULTS
LC Analysis of Reaction Products of DA Quinone with NACySH
The products obtained from mixing 0.4 ml 1 m
M
DA quinone with 0.2 ml 10 m
M
NACySH at either pH 7.4 or pH 2.0 were separated by analytical LC. No discernible
peaks were observed in the LC–EC (reduction) chromatograms (data not shown).
Four major peaks and several minor peaks were obtained in the LC–EC (oxidation)
chromatogram for the reaction conducted at pH 7.4 (Fig. 1A). The peak at 7.26
min is due to DA (1, see Scheme 1), while the peaks at 11.23, 21.24, and 25.62 min
are due to three major reaction products, identified here as compounds 3, 4, and
7
1
, respectively. The same products are observed at pH 2.0, but two of the products,
and 7, are formed in much lower yields than at pH 7.4 (Fig. 1B). The minor peak
at 9.1 min is due to NACySH, whereas the minor peak at 8.2 min is due to an
impurity in the starting material. The other minor peaks are unidentified.
NMR and MS Identification of Reaction Products
Products 3, 4, and 7 were first characterized by FAB–MS. The results showed
ϩ
ϩ
that 3 and 4 both have MH at m/zϭ315 and that 7 has MH at m/zϭ476 (spectra
not shown). These results strongly suggest that 3 and 4 are both monoaddition
adducts (theoretical m/z: 315) of NACySH with DA quinone, whereas 7 is a diaddi-
tion adduct of two NACySH molecules with DA quinone (theoretical m/z: 476).
The structures of these adducts were elucidated by NMR spectroscopy (Table
1
1
). The H NMR spectrum of 4 shows that the molecule contains two aromatic
protons that give meta-coupling (J ϭ 2.0 Hz) (Fig. 2A). Because this result indicates
that the N-acetylcysteinyl moiety is bound to C(5) of the aromatic ring of DA, 4
1
is identified as 5-S-(N-acetylcysteinyl)dopamine (5-NACyS-DA). The H NMR
spectrum of 3 shows that the molecule contains two aromatic protons that display
ortho-coupling (J ϭ 8.2 Hz), indicating that the N-acetylcysteinyl moiety is bound
to C(2) of the aromatic ring of DA (Fig. 2B). Therefore, 3 is identified as 2-S-(N-
acetylcysteinyl)dopamine (2-NACyS-DA).
1
The H NMR spectrum of the diaddition adduct 7 shows only a singlet in the
aromatic region (Fig. 2C), which suggests that both of the N-acetylcysteinyl moieties
are attached to the aromatic ring of DA. Since the signals of the DA side chain
protons overlap with the signals of the protons in the N-acetylcysteinyl moiety in
1
this H NMR spectrum, a DEPT NMR experiment was conducted to confirm that
the N-acetylcysteinyl moieties are not attached to the DA side chain carbons (Fig.
3
). The DEPT NMR spectrum exhibits four 180Њ out-of-phase signals, demonstrating
that four –CH – groups are present in the molecule, two of which are in the two
2
N-acetylcysteinyl moieties and two of which are in the DA side chain. This confirms
that both N-acetylcysteinyl moieties are bound to the aromatic ring of DA. Based
on the fact that there is little or no 6-monoaddition product present in the reaction

DOPAMINE QUINONE AND N-ACETYLCYSTEINE
115
F
IG. 1. LC–EC (oxidation) chromatograms of the reaction products of DA quinone and NACySH
at (A) pH 7.4 and (B) pH 2.0.
mixture, C(6) of the DA quinone ring is presumed to be the site that is least reactive
with NACySH. Accordingly, because assignments where two N-acetylcysteinyl moi-
eties are bound to the 2,6-carbons or 5,6-carbons of the DA aromatic ring are
excluded, the two N-acetylcysteinyl moieties must be bound to C(2) and C(5) of
the DA aromatic ring in adduct 7. This species is identified as 2,5-S,SЈ-di-(N-
acetylcysteinyl)dopamine (2,5-di-NACyS-DA).
The uv/vis and Electrochemical Characterization of the Reaction Products
The adducts were also characterized using uv/vis spectrometry and CV. The
uv/vis spectra show that, whereas both monoaddition adducts (3 and 4) have the

116
XU ET AL.
S
CHEME 1. Proposed mechanism for formation of adducts from reactions of DA quinone with NACySH.

DOPAMINE QUINONE AND N-ACETYLCYSTEINE
117
F
IG. 2. 1H NMR spectra of addition products from reaction of DA quinone with NACySH: (A)
monoaddition product 5-NACyS-DA (4), (B) monoaddition product 2-NACyS-DA (3), and (C) diaddi-
tion product 2,5-di-NACyS-DA (7).
same ␭max values, 258 and 294 nm, they have different molar absorption coefficients
at each ␭max (Table 2). In contrast, the diaddition adduct (7) has only one ␭max at
276 nm and a shoulder at 302 nm. CV data show that all three adducts have slightly
less positive redox potentials than DA at pH 7.0, indicating that they would be
readily oxidized by any unreacted DA quinone in the reaction mixture (Table 2).
Other characteristic data of these adducts are also presented in Table 2.
To estimate the relative molar ratios of adducts 3, 4, and 7 formed at pH 7.4 and
pH 2.0, LC–UV chromatograms (absorbance at ␭max versus time) of the reaction
mixtures were obtained (not shown). Based on the peak areas and respective molar
absorption coefficients, the relative molar ratios of adducts 3, 4, and 7 are estimated
to be 8:59:33 at pH 7.4 and 10:89:1 at pH 2.0.
Electrochemical Studies of DA in the Presence of NACySH
The cyclic voltammogram of DA at pH 7.0 in the absence of NACySH consists
of a single anodic peak near 0.26 V for the two-electron oxidation of DA to its
corresponding quinone and a single cathodic peak near 0.13 V for the reduction

118
XU ET AL.
FIG. 2—Continued
of the quinone to starting material (Fig. 4A). The absence of any other discernible
redox process indicates that DA quinone is relatively stable under these conditions
(
pH 7.0, scan rate: 200 mV/s). In the presence of an equimolar amount of NACySH,
the anodic peak for the oxidation of DA nearly doubles in height (Fig. 4B). Concomi-
tantly, the magnitude of the cathodic peak for the redox of DA quinone is reduced
substantially. The cathodic peak for the reduction of DA quinone is eliminated
completely when the ratio of NACySH to DA is five (Fig. 4C).
The absence of a cathodic peak in the cyclic voltammogram of DA in the presence
of a fivefold excess of NACySH is probably due to either the quinone being con-
sumed by the nucleophile or the nucleophile transferring electrons to the quinone.
To determine which of these possibilities is occurring, the nucleophile was added
to a solution of electrochemically prepared diadduct quinone. If an electron-transfer
process were to occur between the quinone and the nucleophile, the diadduct
quinone would be reduced and the nucleophile would be oxidized. On the other
hand, if a nucleophilic reaction were to occur between the quinone and the nucleo-
phile, a new species would be formed, which should be detected by LC analysis.
After the diadduct, 2,5-di-NACyS-DA, was partially oxidized electrochemically,
the resulting quinone was then mixed with NACySH. The chromatogram of the
NACySH/quinone mixture had one additional major anodic peak at 21.2 min in

DOPAMINE QUINONE AND N-ACETYLCYSTEINE
119
FIG. 2—Continued
addition to the diadduct peak at 13.0 min (data not shown). The effluent that gives
rise to this additional peak exhibits a ␭max at 309 nm and a shoulder at 323 nm in
its uv/vis spectrum (data not shown), which is consistent with the formation of the
triadduct, 2,5,6-tri-NACyS-DA. Addition of the NACyS- group to the aromatic
ring would be expected to cause a red shift in the ␭max: DA, 202 and 280 nm; 2-
or 5-NACyS-DA, 258 and 294 nm; and 2,5-di-NACyS-DA, 276 and 302 nm. These
F
IG. 3. DEPT NMR spectrum of diaddition product 2,5-di-NACyS-DA (7) from reaction of DA
quinone with NACySH.

120
XU ET AL.
TABLE 1
1
a
H NMR Data and Structural Assignments of NACySH Adducts of DA
2-NACyS-DA (3)
5-NACyS-DA (4)
2,5-di-NACyS-DA (7)
ͳ
ͳ
ͳ
(ppm) Coupling (Hz) Assignment (ppm) Coupling (Hz) Assignment (ppm) Coupling (Hz)
Assignment
6
6
4
3
3
3
1
.96 d(8.24)
.86 d(8.24)
.27 dd(7.17, 3.97)
.40 dd(13.74, 3.97)
H5
H6
H3Ј
H2Јa
H7, 8
H2Јb
H6Ј
6.98 d(2.13)
6.84 d(1.84)
4.26 dd(8.24, 3.97)
3.40 dd(14.34, 3.97)
3.27 t(7.02)
3.14 dd(14.35, 8.24)
2.90 t(7.02)
H2
H6
H3Ј
H2Јa
H8
H2Јb
H7
7.05
4.34
s
m
H6
H3Ј, 3Љ
3.47 dd(14.34, 4.27) H2Јa (or 2Љa)
3.38 dd(13.88, 4.28) H2Љa (or 2Јa)
3.22
1.92
1.86
.21
m
m
s
s
H7, 8, 2Јb, 2Љb
H6Љ
H6Ј
.11 dd(14.04, 7.32)
.84
s
1.91
s
H6Ј
a
2
Data were measured in D O at 27ЊC, using HOD (ͳ 4.8 ppm) as a reference signal. See Scheme 1 for structures
of the adducts.
data suggest that NACySH reacts with 2,5-di-NACyS-DA quinone primarily by
nucleophilic addition to form a triadduct, rather than by reduction to regenerate
the diadduct starting material. To demonstrate that there is no additional anodic
peak at 21.2 min when the quinone is reduced, a separate experiment was conducted
in which an electron acceptor, ascorbic acid, was added to a solution of the diadduct
quinone. As expected, the LC chromatograms exhibited only one major anodic
peak at 13.0 min, which is due to the reduced diadduct (data not shown).
These results demonstrate that (1) DA quinone reacts rapidly with NACySH to
give a species that is oxidizable at the applied potential and (2) all electrogenerated
quinones are consumed under these conditions by follow-up chemical reactions.
TABLE 2
Characteristic Data of DA and Its NACySH Adducts
Compounda
DA (1)
M
b
RT (min)c Relative yieldd
UV: ␭max (log ␧)e
CV: Epa, Epc (mV)f
r
153
314
314
6.21
8.87
n/a
8%
202 (3.43), 280 (3.41)
258 (3.43), 294 (3.41)
258 (3.49), 294 (3.36)
276 (3.95), 302 (sh, 3.46)
255, 125
228, 134
179, 133
226, 105
2
5
2
-NACyS-DA (3)
-NACyS-DA (4)
13.74
15.14
59%
33%
,5-di-NACyS-DA (7) 475
a
See Scheme 1 for structures of the adducts.
The molecular weights were determined using FAB–MS.
Retention times were obtained from the LC–EC chromatogram of the reaction mixture at pH 7.4
b
c
(
Fig. 1A).
d
The relative yields of the three major products were estimated on the basis of their peak areas in
the LC–DAD chromatogram (not shown) of the reaction mixture at pH 7.4 and their molar absorp-
e
tion coefficients .
e
The uv/vis molar absorption coefficients of products were obtained in 0.01
M
HCl (Fig. 4).
Cyclic voltammetric data were obtained on a 7.1-mm glassy carbon electrode at a scan rate of 200
mV/s in 0.1 phosphate buffer at pH 7.0 (Fig. 5).
f
2
M

DOPAMINE QUINONE AND N-ACETYLCYSTEINE
121
F
IG. 4. Cyclic voltammograms of 0.3 m
M
DA in the presence of (A) 0 m
M
, (B) 0.3 m
M
, and (C) 1.5
mM
NACySH in 0.1 phosphate buffer (pH 7.0). The scan rate was 200 mV/s.
M
Although the absence of the cathodic current during the reversed scan is not fully
understood, it might be due to the fact that multiple adducts are formed and that
all of the quinones are consumed either by adduct formation or by acting as oxidants
of the adducts.
Attempts to determine the rate constant of the initial reaction of DA quinone
with NACySH at pH 7.0 by CA were unsuccessful. Although the apparent number
of electrons transferred, napp , for the oxidation of DA in the presence of NACySH
was kinetically controlled, napp exceeded 4.0 throughout the accessible time range
of 100 ms Յ t Յ 5 s when [NACySH]/[DA] ϭ 10 (Fig. 5). This result permits lower
Ϫ
1
4
Ϫ
1
Ϫ
1
limits of 100 s and 3.3 ϫ 10
M
s
to be set for the pseudo-first-order and
second-order rate constants, respectively, for the initial reactions of NACySH with
Ϫ
1
DA quinone (13), which are consistent with the estimation of 315 s and 6.3 ϫ
5
Ϫ
1
Ϫ
1
10
M
s
that Tse and co-workers have reported for the addition of cysteine to
dopamine at pH 7.4 (7).
Even though the experiment is subject to several sources of error, the variation
of napp from a value near 4.0 when t ϭ 100 ms to 5.8 when t ϭ 5 s suggests that

122
XU ET AL.
F
IG. 5. CA data and working curve for the addition of NACySH to NACyS-DA quinone in 0.1
phosphate buffer (pH 7.0). Concentrations of DA and NACySH were 0.3 and 3.0 m , respectively.
Potential step was from Ϫ0.2 to 0.6 V vs. Ag/AgCl (saturated KCl). Working curve for ECE process
K ϭ 10) for the reaction described by Eq. [5], —; experimental data, ●●●
M
M
(
.
the rate constant for the addition of NACySH to NACyS-DA quinone might be
amenable to estimation (Fig. 5). When napp vs t was fit to the theoretical working
curve for a homogeneous chemical reaction interposed between electron transfer
Ϫ
1
reactions, a reasonable fit was obtained when k ϭ 0.76 s was used for the pseudo-
first-order rate constant and kЈ ϭ k/[NACySH] ϭ 2.5 ϫ 102
M
Ϫ
1
s for the second-
Ϫ
1
order rate constant. The theoretical working curve was obtained by digital simula-
tion of the sequence of reactions that is described by Eqs. [1] to [5].
ϩ
Ϫ
DA s DA quinone ϩ 2H ϩ 2 e
[1]
[2]
[3]
[4]
fast
DA quinone ϩ NACySH Ǟ NACyS-DA
fast
NACyS-DA ϩ DA quinone Ǟ NACyS-DA quinone ϩ DA
kЈ
NACyS-DA quinone ϩ NACySH Ǟ (NACyS) -DA
2
Kϭ10
NACyS-DA quinone ϩ (NAcyS) –DA
NACyS-DA
ϩ (NACyS) -DA quinone [5]
2
2
The model assumes that reactions [1] through [3] occur rapidly and that reaction
4] is rate controlling. Since each of the electrochemical oxidations in reactions [1]
[
and [3] involves two electrons per molecule of DA and NACyS–DA, the minimum
value of napp at small values of kt, where k is the pseudo-first-order rate constant
and t is time, should be 4.0. In contrast, when kt is large, a limiting value of 6.0 is
predicted for this reaction sequence (13). The final two electrons per molecule of
DA arise from the homogeneous redox reaction that is described by Eq. [5]. The
value of the equilibrium constant for this reaction is based on the redox potentials

DOPAMINE QUINONE AND N-ACETYLCYSTEINE
123
F
IG. 6. Cyclic voltammogram of 0.3 m
M
DA in the presence of 3.0 m
M
NACySH in 0.01
M
HCl and
0
.09 M KCl (pH 2.0). The scan was initiated in the positive-going direction at a rate of 200 mV/s. The
numbers 1 and 2 indicate the first and second cycles, respectively.
of DA and the monoadducts (see Table 2). Departure from semiinfinite linear
diffusion at longer times limited our experimental time range to t Յ 5 s. Because
NACySH is also oxidized slowly at the applied potential of 0.6 V, currents for the
oxidation of DA in the presence of NACySH were corrected for this back-
ground process.
The rate of the initial reaction of NACySH with DA (Eq. [2]) is significantly
slower at pH 2.0 than at pH 7.0. As shown in Fig. 6, the cyclic voltammogram
consists of a single, kinetically controlled anodic peak for the oxidation of DA to
DA quinone near 0.59 V on the first, positive-going sweep. Upon reversal of the
potential scan at 0.80 V, the negative-going sweep consists of two cathodic peaks.
The more positive of these peaks is due to the reduction of predominately 5-
NACyS-DA quinone, whereas the more negative peak near 0.35 V is due to the
reduction of unreacted DA quinone. On the second, positive-going sweep, a new
anodic peak appears at 0.48 V, which is assigned to the two-electron oxidation of
5
-NACyS-DA to 5-NACyS-DA quinone. As expected when the initial electrode
product, DA quinone, is consumed by a follow-up chemical reaction to produce a
new electroactive species, the relative magnitudes of the cyclic voltammetric peaks
for the adduct redox couple, 5-NACyS-DA/5-NACyS-DA quinone, increase at the
expense of the peaks for the DA/DA quinone redox couple with increasing number
of cyclic voltammetric cycles. The observations that Ep,a for 5-NACyS-DA occurs
at a less positive potential than that for DA, whereas Ep,c for 5-NACyS-DA quinone
occurs at a more positive potential than that for DA quinone, are attributed to the
greater electrochemical reversibility of the 5-NACyS-DA/5-NACyS-DA quinone
redox couple than of the DA/DA quinone redox couple.
The rate constant for the reaction of DA quinone with NACySH at pH 2.0 was
determined by CA. In the absence of NACySH, the oxidation of DA to DA quinone
is a diffusion-controlled two-electron process. However, when [NACySH]/[DA] ϭ
10, napp is kinetically controlled, varying from its lower limit near 2.0 when t ϭ 100

124
XU ET AL.
F
IG. 7. CA data and working curve to estimate the rate constant for the addition of NACySH to DA
quinone in 0.01 HCl and 0.09 KCl (pH 2.0). Concentrations of DA and NACySH were 0.3 and
.0 m , respectively. Potential step was from 0 to 0.8 V vs Ag/AgCl (saturated KCl). Working curve
for ECE process (K ϭ 1) for the reaction described by Eq. [6], —; experimental data, ●●●
M
M
3
M
.
ms to 4.0 when t ϭ 5 s. These data were then fit to the theoretical working curve
for an ECE process in which a pseudo-first-order chemical reaction, the addition
of NACySH to DA quinone (Eq. [2]), is interposed between the two-electron
oxidations of DA (Eq. [1]) and 5-NACyS-DA (Eq. [3]) to DA quinone and 5-
Ϫ
1
2
Ϫ
1
NACyS-DA quinone, respectively (13, Fig. 7). Values of 0.53 s and 1.8 ϫ 10
s
M
Ϫ
1
were calculated for the pseudo-first-order and second-order rate constants,
respectively, for the reaction described by Eq. [2] at pH 2.0. The equilibrium constant
that was used for the homogenous redox reaction was 1.0, which is based on the
fact that (Ep,a ϩ E )/2 for both redox couples is 0.46 V at pH 2.0 (Eq. [6]).
p,c
Kϭ1
5-NACyS-DA quinone ϩ DA
5-NACyS-DA ϩ DA quinone
[6]
Again, since NACySH is also oxidized slowly at the applied potential of 0.8 V, the
kinetically controlled current for oxidation of DA in the presence of excess NACySH
in Fig. 7 was corrected for this background process.
DISCUSSION
Cyclic voltammetric studies have shown that the mono- and diaddition adducts
of NACySH with DA quinone are oxidized more readily than DA at pH 7.0. Since
our goal was to study the initial products that arise from the reaction of DA quinone
with NACySH, electrolysis of DA was effected in the absence of added nucleophiles
and was terminated when its oxidation was complete. DA quinone was determined
to be stable under the conditions of its preparation, but was found to react rapidly
when it was added to a solution of NACySH.
The most abundant product detected in our system is 5-NACyS-DA (4), the

DOPAMINE QUINONE AND N-ACETYLCYSTEINE
125
monoaddition adduct of the quinone at C(5) of the ring of DA. The second major
product produced at pH 7.4 is the diaddition adduct of the quinone bonded at
C(2) and C(5) of the ring (2,5-di-NACyS-DA, 7). In addition, we found that the
monoaddition product linked at C(2) of the ring (2-NACyS-DA, 3) is also produced
at pH7.4, but at a lower yield than that of either 5-NACyS-DA or 2,5-di-NACyS-
DA. The formation of 5-NACyS-DA is consistent with the detection of 5-S-cys-
teinyldopamine in a previous study where DA was electrolyzed in the presence of
L-cysteine (11). However, the monoaddition adduct of the quinone bonded at C(2)
of the ring of DA was not detected in that study, apparently because continuous
electrolysis caused electrochemical oxidation of the initial products and formation
of many other products. CV of DA in the presence of NACySH reveals a kinetically
controlled anodic current that is the result of reactions of DA quinone with
NACySH to yield adducts that are also electroactive at the applied potential.
Since electrolysis was terminated prior to the addition of NACySH, oxidation of
the monoaddition adducts will be effected only by any DA quinone that remains
unreacted. If the quinones of the monoaddition adducts were to react also with
NACySH, the product distribution would be predicted to contain equimolar
amounts of DA and the 2,5-di-NACyS-DA adduct. The experimental results are
consistent with this expectation at both pH 2.0 and pH 7.4.
A pathway is proposed for the reaction of electrogenerated DA quinone with
an excess of NACySH (Scheme 1). After DA quinone (2) is mixed with NACySH,
nucleophilic attack of NACySH occurs on the ring of DA quinone principally at
C(5) and to a lesser extent at C(2), to form 5-NACyS-DA (4) and 2-NACyS-DA
(3), respectively. Portions of these two monoaddition adducts are subsequently
oxidized by unreacted DA quinone to form the corresponding quinones of the
monoaddition adducts (5 and 6) and to regenerate DA. 2-NACyS-DA quinone (5)
and 5-NACyS-DA quinone (6) are then subject to nucleophilic attack by NACySH
at the 5- and 2-positions of the ring, respectively, forming 2,5-di-NACyS-DA (7).
In view of our observation that the rate constant for the addition of NACySH to
DA quinone is at least 100 times greater than that for the addition of NACySH to
the adduct quinone, it is somewhat surprising that any 2,5-di-NACyS-DA is formed
under our reaction conditions. If homogeneous electron transfer is rapid (Eq. [5]),
then the most rational explanation requires the equilibrium constant for the oxida-
tion of the monoaddition adduct by DA quinone (Eq. [6]) to be larger than the
estimate of 1 at pH 2.0 and to increase with increasing pH.
Although the 2-cysteinyldopamine adduct has not been reported previously to
our knowledge, a similar C(2) addition adduct, 2-cysteinyldopa, has been obtained
from 3,4-dihydroxyphenylalanine (dopa) quinone generated enzymatically using
mushroom tyrosinase (14). 2-Glutathionyldopa has also been obtained from the
tyrosinase-catalyzed conjugation of dopa with glutathione (15). It also has been
reported that tyrosinase-generated dopa quinone reacts with protein sulfhydryl
groups (16), but the structure(s) of the adduct(s) formed is unknown.
Other adducts that might be formed are the C(6) monoaddition and the C(2),
C(5), C(6) triaddition adducts. However, no C(6) addition adduct was detected and
no such C(6) addition product has been reported for the reaction of either DA
quinone with -cysteine (11) or in studies of the oxidation of catechol, 4-methylcate-
L

126
XU ET AL.
chol and N-acetyldopamine with
L-cysteine (17). On the other hand, 6-cysteinyldopa
and 6-glutathionyldopa have been detected in the tyrosinase-catalyzed oxidations
of dopa (14, 15).
While the formation of a triaddition adduct is possible, our conditions do not
favor its formation. CA of DA in the presence of NACySH reveals that the reaction
of the first molecule of nucleophile with DA quinone to form monoaddition adducts
is much faster than the addition of a second molecule of nucleophile. Accordingly,
the addition of a third molecule would be expected to be even slower. Also, there
is a deficiency of oxidant (unreacted DA quinone) for oxidation of the diaddition
adduct to its quinone, which is an advantage of this study in terms of preventing
subsequent reactions of the initial products. However, under the conditions of a
continuous electrochemical oxidation, a diaddition adduct of -cysteine and DA
L
cyclized at the C(6) position has been detected (11).
In summary, this study has delineated the initial reactions of DA quinone and
NACySH. Previously, the pathway for the initial reaction of DA quinone with L-
cysteine had been proposed to be the addition of -cysteine exclusively to C(5) of
L
DA quinone (11). However, we have shown that other nucleophilic additions also
occur, including addition to C(2) of DA quinone and formation of the C(2), C(5)
diaddition adduct.
ACKNOWLEDGMENTS
The authors thank B. Plashko for performing FAB–MS experiments and M. Collinson for providing
the BAS-100W electrochemical system. We are grateful to T. L. Hopkins and T. Morgan for assistance
with this study and for reviewing the manuscript. We also thank G. Dryhurst (University of Oklahoma)
and M. Sugumaran (University of Massachusetts Boston) for critical comments about the manuscript.
This research was supported in part by National Science Foundation Grants DCB-9019400, MCB-
9
418129, and CHE-9216101.
REFERENCES
1
. NAGATSU, T. (1986) in Quantitative Analysis of Catecholamines and Related Compounds (Krstu-
lovic, A. M., Ed.), pp. 209–210, Ellis Horwood Limited, Chichester.
RAMER, K. J. (1992) Annu. Rev. Entomol. 37, 273–302.
. GRAHAM, D. G. (1978) Mol. Pharmacol. 14, 633–643.
. HIRSCH, E., GRAYBIEL, A. M., AND GID, Y.A. (1988) Nature 334, 345–348.
2
3
4
5
. HOPKINS, T. L., AND K
A
. KASTNER, A., HIRSCH, E. C., LEJEUNE, O, JAVOY-AGID, F., RASCOL, O., AND
AGID, Y. (1992) J.
Neurochem. 59, 1080–1089.
6
7
8
9
. ZHANG, F., AND
. TSE, D. C. S., M
. HASTINGS, T. G., AND
D
RYHURST, G. (1993) Bioorg. Chem. 21, 392–410.
REERY, R. L., AND DAMS, R., N. (1976) J. Med. Chem. 19, 37–40.
IGMOND, M. J. (1994) J. Neurochem. 63, 1126–1132.
. HASTINGS, T. G. (1995) J. Neurochem. 64, 919–924.
ARLSSON, A. (1990) J. Neurochem. 54, 578–586.
RYHURST, G. (1994) J. Med. Chem. 37, 1084–1098.
WAAY, M. (1978) J. Chem. Ed. 55, A7–A12.
3. HAWLEY, M. D., AND ELDBERG, S. W. (1966) J. Phys. Chem. 70, 3459–3464.
4. ITO, S., AND ROTA, G. (1977) Experientia 33, 1118–1119.
5. ITO, S., PALUMBO, A., AND ROTA, G. (1985) Experientia 41, 960–961.
6. KATO, T., ITO, S., AND UJITA, K. (1986) Biochim. Biophys. Acta 881, 415–421.
7. SUGUMARAN, M., DALI, H., AND EMENSI, V. (1989) Arch. Insect Biochem. Physiol. 11, 127–137.
C
C
A
Z
1
1
1
1
1
1
1
1
0. FORNSTEDT, B., BERGH, I., ROSENGREN, E., AND
1. ZHANG, F., AND
2. VAN
C
D
S
F
P
P
F
S