Oxidative metabolites of 5-S-cysteinylnorepinephrine are irreversible inhibitors of mitochondrial complex I and the α-ketoglutarate dehydrogenase and pyruvate dehydrogenase complexes: Possible implications for neurodegenerative brain disorders

Article abstract of DOI:10.1021/tx990170t

The major initial product of the oxidation of norepinephrine (NE) in the presence of L-cysteine is 5-S-cysteinylnorepinephrine which is then further easily oxidized to the dihydrobenzothiazine (DHBT) 7-(1-hydroxy-2-aminoethyl)-3,4-dihydro-5-hydroxy-2H-1,4-benzothia zine-3-carboxylic acid (DHBT-NE-1). When incubated with intact rat brain mitochondria, DHBT-NE-1 evokes rapid inhibition of complex I respiration without affecting complex II respiration. DHBT-NE-1 also evokes time- and concentration-dependent irreversible inhibition of NADH-coenzyme Q₁ (CoQ₁) reductase, the pyruvate dehydrogenase complex (PDHC), and α-ketoglutarate dehydrogenase (α-KGDH) when incubated with frozen and thawed rat brain mitochondria (mitochondrial membranes). The time dependence of the inhibition of NADH-CoQ₁ reductase, PDHC, and α-KGDH by DHBT-NE-1 appears to be related to its oxidation, catalyzed by an unknown component of the inner mitochondrial membrane, to electrophilic intermediates which bind covalently to active site cysteinyl residues of these enzyme complexes. The latter conclusion is based on the ability of glutathione to block inhibition of NADH-CoQ₁ reductase, PDHC, and α-KGDH by scavenging electrophilic intermediates, generated by the mitochondrial membrane-catalyzed oxidation of DHBT-NE-1, forming glutathionyl conjugates, several of which have been isolated and spectroscopically identified. The possible implications of these results to the degeneration of neuromelanin-pigmented noradrenergic neurons in the locus ceruleus in Parkinson's disease are discussed.

Full text of DOI:10.1021/tx990170t

Chem. Res. Toxicol. 2000, 13, 749-760
749
Oxid a tive Meta bolites of 5-S-Cystein yln or ep in ep h r in e
Ar e Ir r ever sible In h ibitor s of Mitoch on d r ia l Com p lex I
a n d th e r-Ketoglu ta r a te Deh yd r ogen a se a n d P yr u va te
Deh yd r ogen a se Com p lexes: P ossible Im p lica tion s for
Neu r od egen er a tive Br a in Disor d er s
Wenkuan Xin, Xue-Ming Shen, Hong Li, and Glenn Dryhurst*
Department of Chemistry and Biochemistry, University of Oklahoma, Norman, Oklahoma 73019
Received October 5, 1999
The major initial product of the oxidation of norepinephrine (NE) in the presence of L-cysteine
is 5-S-cysteinylnorepinephrine which is then further easily oxidized to the dihydrobenzothiazine
(DHBT) 7-(1-hydroxy-2-aminoethyl)-3,4-dihydro-5-hydroxy-2H-1,4-benzothiazine-3-carboxylic
acid (DHBT-NE-1). When incubated with intact rat brain mitochondria, DHBT-NE-1 evokes
rapid inhibition of complex I respiration without affecting complex II respiration. DHBT-NE-1
also evokes time- and concentration-dependent irreversible inhibition of NADH-coenzyme Q1
(CoQ1) reductase, the pyruvate dehydrogenase complex (PDHC), and R-ketoglutarate dehy-
drogenase (R-KGDH) when incubated with frozen and thawed rat brain mitochondria
(mitochondrial membranes). The time dependence of the inhibition of NADH-CoQ1 reductase,
PDHC, and R-KGDH by DHBT-NE-1 appears to be related to its oxidation, catalyzed by an
unknown component of the inner mitochondrial membrane, to electrophilic intermediates which
bind covalently to active site cysteinyl residues of these enzyme complexes. The latter conclusion
is based on the ability of glutathione to block inhibition of NADH-CoQ1 reductase, PDHC, and
R-KGDH by scavenging electrophilic intermediates, generated by the mitochondrial membrane-
catalyzed oxidation of DHBT-NE-1, forming glutathionyl conjugates, several of which have
been isolated and spectroscopically identified. The possible implications of these results to the
degeneration of neuromelanin-pigmented noradrenergic neurons in the locus ceruleus in
Parkinson’s disease are discussed.
pars compacta (SNc), are believed to underlie the classic
motor symptoms of Parkinson’s disease (PD) (1). On the
basis of pathobiochemical changes in the parkinsonian
SNc and in the nigrostriatal system of animals exposed
to the dopaminergic neurotoxin 1-methyl-4-phenyl-1,2,3,6-
tetrahydropyridine (MPTP) (2), we recently proposed a
new mechanism for the neurotoxic process (3). The first
step in this mechanism was proposed to be a large but
transient impairment of DA neuron energy metabolism.
In the case of MPTP, this would be caused by its active
metabolite, 1-methyl-4-phenylpyridinium (MPP+) (4), a
reversible inhibitor of mitochondrial complex I (5, 6) and
R-ketoglutarate dehydrogenase (R-KGDH) (7). The nor-
mal decline of neuron energy metabolism with age (8),
periodic exposure to environmental toxins that interfere
with mitochondrial respiratory enzymes (6, 9-12), par-
ticularly in view of both age-based (13) and genetically
based impairments of xenobiotic metabolism (14-17),
together with systemic mitochondrial enzyme defects (18,
19) might represent a combination of factors that evoke
a profound depletion of neuronal energy in the parkin-
sonian SNc. This energy impairment with resultant
depolarization of both the neuronal and mitochondrial
membranes was proposed to mediate a massive release
of DA (20) together with both cytoplasmic and mitochon-
drial glutathione (GSH) (21-23). Furthermore, the pro-
found energy impairment and relief of the Mg2+ block of
N-methyl-D-aspartate (NMDA) receptors was proposed
In tr od u ction
Massive decrements of striatal dopamine (DA),1 the
result of the degeneration of neuromelanin (NM)-
pigmented dopaminergic neurons in the substantia nigra
* To whom correspondence should be addressed. Telephone:
(405) 325-4811. Fax: (405) 325-6111. E-mail: gdryhurst@
chemdept.chem.ou.edu.
1
Abbreviations: Asp, L-aspartate; BSA, bovine serum albumin; BT,
benzothiazine; BT-NE-1, 7-(1-hydroxy-2-aminoethyl)-5-hydroxy-1,4-
benzothiazine-3-carboxylic acid; BT-NE-2, 7-(1-hydroxy-2-aminoethyl)-
5-hydroxy-1,4-benzothiazine; BT-NE-2R, 7-(1-hydroxy-2-aminoethyl)-
3,4-dihydro-5-hydroxy-2H-1,4-benzothiazine; CD, circular dichroism;
CoQ1, coenzyme Q1; CysGly, cysteinylglycine; CySH, L-cysteine; DA,
dopamine; DHBT, dihydrobenzothiazine; DHBT-NE-1, 7-(1-hydroxy-
2-aminoethyl)-3,4-dihydro-5-hydroxy-2H-1,4-benzothiazine-3-carboxy-
lic acid; DP, dipeptidases; DTT, D,L-dithiothreitol; EDTA, ethylenedi-
aminetetraacetic acid; ESI-MS, electrospray ionization mass spec-
trometry; FAB-MS, fast atom bombardment mass spectrometry; Glu,
L-glutamate; GSH, glutathione; γ-GT, γ-glutamyl transpeptidase; HO•,
hydroxyl radical; KCN, potassium cyanide; R-KGDH, R-ketoglutarate
dehydrogenase; LC, locus ceruleus; MeCN, acetonitrile; MeOH, metha-
nol; MPP+, 1-methyl-4-phenylpyridinium; MPTP, 1-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine; NaBH4, sodium borohydride; NAD+, nico-
tinamide adenine dinucleotide; NADH, reduced nicotinamide adenine
dinucleotide; NE, norepinephrine; NM, neuromelanin; NMDA, N-
methyl-D-aspartate; O2-•, superoxide anion radical; PD, Parkinson’s
disease; PDHC, pyruvate dehydrogenase complex; 5-S-CyS-DA, 5-S-
cysteinyldopamine; 2-S-CyS-NE, 2-S-cysteinylnorepinephrine; 5-S-
CyS-NE, 5-S-cysteinylnorepinephrine; SNc, substantia nigra pars
compacta; SOD, superoxide dismutase; TFA, trifluoroacetic acid; TPP,
thiamine pyrophosphate; 5, 6-S-glutathionyl-DHBT-NE-1; 6, 8-S-
glutathionyl-DHBT-NE-1; 8, (2R)-2-S-glutathionyl-BT-NE-1; 9, (2S)-
2-S-glutathionyl-BT-NE-1; 10, (2R,3R)-2-S-glutathionyl-DHBT-NE-1;
11, (2S,3S)-2-S-glutathionyl-DHBT-NE-1.
10.1021/tx990170t CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/08/2000

750 Chem. Res. Toxicol., Vol. 13, No. 8, 2000
Xin et al.
Sch em e 1
to mediate their activation by basal extracellular levels
of L-glutamate (Glu) and L-aspartate (Asp) (24) with
resultant superoxide (O2-•) generation in excess of the
scavenging capacity of superoxide dismutase (SOD) (25).
-•
During the DA neuron energy impairment, O2 was
proposed to release low-molecular weight Fe2+ from iron-
containing proteins (26-28) that catalytically decomposes
H2O2 in the presence of ascorbate, forming hydroxyl
radical (HO•) which oxidizes extracellular GSH (basal and
that released from DA neurons). This was proposed to
trigger release of GSH from glia as a mechanism for
protecting neighboring neurons against damage by ex-
tracellular HO• (29).
When the neuron energy impairment begins to subside,
increasing ATP production would initiate reuptake of DA
-•
(30) that is oxidized by and hence would scavenge O2
(31), thus blocking Fe2+ mobilization, HO• formation, and
oxidation of extracellular GSH. However, replenishing
intraneuronal GSH, released from DA neurons during
the energy impairment, requires its continued release
from glia (29, 32) and degradation initially by γ-glutamyl
transpeptidase (γ-GT) to Glu and cysteinylglycine (Cys-
Gly) which is further hydrolyzed by dipeptidases (DP) to
L-glycine (Gly) and L-cysteine (CySH) (33). CysGly and
CySH are then translocated into DA neurons to provide
CySH, normally employed for intraneuronal GSH bio-
synthesis (32, 34, 35). Indeed, when perfusions of neu-
rotoxic concentrations of MPP+ into the rat striatum or
SNc are discontinued, a massive elevation of extracellular
GSH levels occurs, believed to reflect its release from glia
(3). The subsequent decline of the extracellular GSH level
from its peak concentrations is accompanied by elevation
of the extracellular Glu level (36) and a smaller increase
in the extracelular CySH level, the latter effect possibly
reflecting its rapid translocation into DA neurons (3).
During the period of recovering but still reduced DA
neuron ATP production, partial relief of the Mg2+ block
of NMDA receptors should permit their continued activa-
tion by elevated extracellular levels of Glu with resultant
oxidation of NE in the presence of translocated CySH as
a noradrenergic neuron energy impairment subsides
might lead to endogenous toxicants that contribute to LC
cell death. The in vitro oxidation of NE generates NE-
o-quinone that is scavenged by CySH to give, initially,
5-S-cysteinylnorepinephrine (5-S-CyS-NE) (Scheme 1)
(48). However, 5-S-CyS-NE is then further oxidized to
o-quinone 1 which rapidly cyclizes to o-quinone imine 2.
5-S-CyS-NE is then oxidized by 2 to give radical 3
forming radical 4. Disproportionation of 3 forms 5-S-CyS-
NE and 1. Disproportionation of radical 4 forms 2 and
7-(1-hydroxy-2-aminoethyl)-3,4-dihydro-5-hydroxy-2H-
1,4-benzothiazine-3-carboxylic acid (DHBT-NE-1). In view
of the possibility that endogenously generated oxidative
metabolites of 5-S-CyS-NE might contribute to the
pathogenesis of LC cell death in PD, this communication
describes the effects of DHBT-NE-1 on mitochondrial
respiration and the activities of NADH-coenzyme Q1
(CoQ1) reductase, R-KGDH, and the pyruvate dehydro-
genase complex (PDHC).
-•
O2 generation that oxidizes returning DA to DA-o-
quinone which, we propose, reacts with translocated
CySH forming, initially, 5-S-cysteinyldopamine (5-S-CyS-
DA) (31). Indeed, 5-S-CyS-DA, normally a minor metabo-
lite of DA, becomes a major metabolite in the parkinso-
nian SNc as indicated by a large increase in the 5-S-CyS-
DA/DA concentration ratio (37). However, 5-S-CyS-DA
is more easily oxidized than DA (31) to dihydrobenzothi-
azines (DHBTs) and benzothiazines (BTs) (38) that
include irreversible inhibitors of mitochondrial complex
I (39, 40) and R-KGDH (41). This raises the possibility
that DHBTs and BTs, oxidative metabolites of 5-S-CyS-
DA, might be endogenously generated toxicants that
contribute to the mitochondrial complex I (42) and
R-KGDH (43) deficiencies and dopaminergic SNc cell
death in PD.
NM-pigmented noradrenergic cells in the locus cer-
uleus (LC) also degenerate in PD (44). Resultant decre-
ments of norepinephrine (NE) levels in projection areas
of LC neurons have been implicated with dementia,
depression, and other nonmotor symptoms of PD (45-
47). However, little is known about pathobiochemical
changes that occur in the parkinsonian LC. Nevertheless,
the fact that LC and SNc cells both contain easily oxidized
catecholamine neurotransmitters and degenerate in PD
suggests that similar neuropathological mechanisms may
be involved. In other words, intraneuronal O2-•-mediated
Ma ter ia ls a n d Meth od s
Ch em ica ls. NE (hydrochloride salt and free base), CySH
(free base), GSH, SOD, catalase, 2-aminoethanediol, nicotina-
mide adenine dinucleotide (NAD+), reduced nicotinamide ad-
enine dinucleotide (NADH), adenosine 5′-diphosphate (ADP,
sodium salt), L-(-)-malic acid (disodium salt), R-ketoglutaric acid
(sodium salt), succinic acid (disodium salt), pyruvic acid (sodium
salt), glycylglycine, ethylenediaminetetraacetic acid (potassium
salt, K2EDTA), mannitol, Trizma hydrochloride, coenzyme A,
D,L-dithiothreitol (DTT), thiamine pyrophosphate (TPP), fatty
acid-free bovine serum albumin (BSA), Triton X-100, and
asolectin (phospholipid from soybean) were obtained from Sigma
(St. Louis, MO). MPP+ (iodide salt) was obtained from RBI
(Natick, MA) and rotenone from Aldrich (Milwaukee, WI). These
and all other commercially available chemicals were of the
highest purity available and were used without further purifica-
tion. Coenzyme Q1 (ubiquinone-1) was synthesized as described
previously (39).
Sp ectr oscop y. NMR spectra were recorded on a Varian (Palo
Alto, CA) VXR-500 spectrometer. Low- and high-resolution fast
atom bombardment mass spectrometry (FAB-MS) employed a
VG Instruments (Manchester, U.K.) ZAB-E spectrometer. Elec-
trospray ionization mass spectrometry (ESI-MS) employed a PE
Sciex (Thornhill, ON) API III biomolecular mass analyzer.
Circular dichroism (CD) spectra were obtained with an AVIV
(Lakewood, NJ ) 62DS spectropolarimeter. UV/visible spectra

Mitochondrial Toxicity of Cysteinylnorepinephrine
Chem. Res. Toxicol., Vol. 13, No. 8, 2000 751
were recorded on a Hewlett-Packard (Palo Alto, CA) 8452A diode
array spectrophotometer.
for isolation and purification of glutathionyl conjugates 10 and
11. The flow rate for methods I-IV was 7 mL/min.
Syn th esis of 5-S-CyS-NE a n d DHBT-NE-1. NE (24.7 mg,
free base) dissolved in aqueous 0.1 M HCl (100 mL) was oxidized
by controlled potential electrolysis (1.0 V vs SCE) at pyrolytic
graphite electrodes using equipment described previously (48).
The electrolysis was terminated after 30 min, and CySH (18
mg) was added to the bright yellow solution of NE-o-quinone,
causing the color to rapidly fade to pale yellow. The major
products that formed were 5-S-CyS-NE together with smaller
yields of 2-S-CyS-NE (48). These conjugates were separated
using preparative HPLC method I when 5-S-CyS-NE eluted
with a retention time (tR) of 19 min. The solution eluted under
this peak was collected and purified by preparative HPLC using
1:1 MeCN in deionized water as the mobile phase. The solution
eluted under the peak for 5-S-CyS-NE was collected and then
freeze-dried to give a chromatographically pure (analytical
An a lytica l HP LC. Analytical HPLC employed a Gilson
(Middleton, WI) gradient system equipped with dual model 302
pumps, a Rheodyne (Cotati, CA) 7125 loop injector (20 µL
sample loop), and a Waters (Milford, MA) 440 UV detector (254
nm). Two mobile phase solvents were employed. Solvent A was
prepared by adding 10 mL of concentrated ammonium hydroxide
(NH4OH) to 4 L of deionized water, the pH being adjusted to
2.2 by addition of concentrated trifluoroacetic acid (TFA).
Solvent B was prepared by adding 10 mL of NH4OH to 2 L of
HPLC grade methanol (MeOH) and 2 L of deionized water, the
pH being adjusted to 2.2 with TFA. A reversed phase column
(Columbus C18, 5 µm, 100 mm × 3.2 mm, Phenomenex, Tor-
rance, CA) was employed. The following mobile phase gradient
was employed: from 0 to 2 min, 100% solvent A; from 2 to 43
min, linear gradient to 21.5% solvent B; from 43 to 45 min,
linear gradient to 100% solvent B; and from 45 to 48 min, 100%
solvent B. The flow rate was 0.6 mL/min. A standard calibration
curve of HPLC peak height versus DHBT-NE-1 concentration
was employed for analysis studies. This analytical HPLC
method was used to monitor the oxidation of DHBT-NE-1 and
the effects of GSH on this reaction. In a typical experiment,
DHBT-NE-1 (1.0 mM) was incubated with rat brain mitochon-
drial membranes (2 µg of mitochondrial protein/µL) in 20 mM
potassium phosphate buffer (pH 8.0) at 30 °C for times of e2 h.
After the predetermined incubation time, an aliquot of the
reaction mixture (50 µL) was removed and immediately diluted
to 500 µL with ice-cold deionized water and analyzed by
analytical HPLC. In experiments designed to assess benzothi-
azine (BT) metabolites, the diluted incubtion mixture was
treated with NaBH4 immediately before HPLC analysis. NaBH4
reduces BT-NE-1 to DHBT-NE-1 and BT-NE-2 to BT-NE-2R.
Calibration curves prepared with authentic samples of DHBT-
NE-1 and BT-NE-2R were employed for quantitative HPLC
analysis of these compounds. Peak identification was based on
coelution and peak fitting with authentic samples.
1
HPLC) white solid. The H NMR and mass spectra of 5-S-CyS-
NE were identical to those previously reported (48).
To synthesize DHBT-NE-1, an aqueous solution of 5-S-CyS-
NE (ca. 5 mM) was adjusted to pH 10-11 with a solution of
concentrated KOH and stirred at room temperature for 30 min.
The resultant product solution was adjusted to pH 2.2 by
addition of HCl and then injected onto the reversed phase
column, and components were separated using preparative
HPLC method II. The solution eluted under the peak corre-
sponding to DHBT-NE-1 was collected and then desalted by
preparative reversed phase HPLC using solvent E as the mobile
phase. The 1H NMR and mass spectra of chromatographically
pure DHBT-NE-1 were identical to those reported previously
(48).
Syn t h esis of 7-(1-H yd r oxy-2-a m in oet h yl)-3,4-d ih yd r o-
5-h yd r oxy-2H-1,4-ben zoth ia zin e (BT-NE-2R). NE (1 mM)
and 2-aminoethanol (1 mM) dissolved in 30 mL of potassium
phosphate buffer (pH 7.4; µ ) 0.2) were oxidized by controlled
potential electrolysis at pyrolytic graphite electrodes (70 mV vs
SCE). After 60 min, the entire pale yellow product solution was
introduced onto the preparative reversed phase column, and
components were separated using HPLC method II. The tR for
BT-NE-2R was 44 min. The solution eluted under this peak was
collected at -80 °C (dry ice/acetone bath). Following several
repetitive experiments, the combined solutions containing BT-
NE-2R were purified by preparative HPLC method III and then
freeze-dried.
P r ep a r a tive HP LC. Preparative HPLC employed a Gilson
gradient system equipped with dual model 306 pumps, a
reversed phase column (Bakerbond C18, 10 µm, 250 mm × 21.2
mm, P. J . Cobert Associates, St. Louis, MO), and a UV detector
(254 nm). Six mobile phase solvents were employed. Solvent C
was prepared by adding 7.5 mL of NH4OH to 1 L of deionized
water and then adjusting the pH to 2.2 with concentrated HCl.
Solvent D was 1:1 (v/v) acetonitrile (MeCN) in deionized water
adjusted to pH 2.2 with HCl. Solvent E was 1:1 (v/v) MeCN in
deionized water. Solvent F was deionized water adjusted to pH
2.15 with TFA. Solvent G was 1:1 MeCN in deionized water
adjusted to pH 2.15 with TFA. Solvent H was 1:1 MeOH in
deionized water adjusted to pH 2.15 with TFA. Preparative
HPLC method I, used to separate and purify cysteinyl conju-
gates of NE, employed solvents C and D and the following
gradient: from 0 to 2 min, 100% solvent C; and from 2 to 30
min, linear gradient to 15% solvent D. Preparative HPLC
method II, used to separate and purify DHBT-NE-1 and for
initial isolation of BT-NE-2R, employed solvents D and E and
the following gradient: from 0 to 2 min, 100% solvent D; and
from 2 to 30 min, linear gradient to 30% solvent E. Preparative
HPLC method III, used to purify BT-NE-2R, employed solvents
F and G and the following gradient: from 0 to 20 min, 100%
solvent F; from 20 to 50 min, linear gradient to 3% solvent G;
from 50 to 53 min, linear gradient to 100% solvent G; and from
53 to 68 min, 100% solvent G. Preparative HPLC method IV
employed solvents F and H and the following gradient: from 0
to 20 min, 100% solvent F; from 20 to 175 min, linear gradient
to 20% solvent H; from 175 to 179 min, linear gradient to 100%
solvent H; and from 179 to 190 min, 100% solvent H. Method
IV was used to separate and purify the 6-(S)-5 and 8-(S)-6
glutathionyl conjugates of DHBT-NE-1, for isolation and puri-
fication of metabolites formed by the mitochondrial membrane-
catalyzed oxidation of DHBT-NE-1 in the presence of GSH, and
BT-NE-2R was isolated as a very pale yellow solid that in
the HPLC mobile phase (pH 2.15) exhibited UV bands (λmax) at
294, 288, 254 (sh), and 218 nm. FAB-MS (thioglycerol matrix):
1
m/z 227.0900 (MH+, 3%, C10H15N2O2S; calcd m/z 227.0854). H
NMR (Me2SO-d6): δ 9.60 [br s, 1H, C(5)-OH], 7.85 (br s, 3H,
NH3+), 6.48 [d, J ) 1.5 Hz, 1H, C(8)-H], 6.40 [d, J ) 1.5 Hz,
1H, C(6)-H], 4.51 [dd, J ) 9.5, 3.0 Hz, 1H, C(â)-H], 3.50-3.47
[m, 2H, C(3)-H2], 2.99-2.96 [m, 2H, C(2)-H2], 2.92-2.86 [m, 1H,
C(R)-H], 2.77-2.70 [m, 1H, C(R)-H].
Syn th esis of 6-S-Glu ta th ion yl-7-(1-h yd r oxy-2-a m in oet-
h yl)-3,4-d ih yd r o-5-h yd r oxy-2H -1,4-b en zot h ia zin e-3-ca r -
b oxylic Acid (5) a n d 8-S-Glu t a t h ion yl-7-(1-h yd r oxy-2-
a m in oeth yl)-3,4-d ih yd r o-5-h yd r oxy-2H-1,4-ben zoth ia zin e-
3-ca r boxylic Acid (6). DHBT-NE-1 (1 mM) and GSH (5 mM)
dissolved in 30 mL of potassium phosphate buffer (pH 7.4; µ )
0.2) were oxidized by controlled potential electrolysis at pyrolytic
graphite electrodes (70 mV vs SCE). After 60 min, the entire
pale yellow solution was introduced onto the preparative
reversed phase column, and components were separated by
HPLC method IV. Two major products eluted at a tR of 112 min
(6) and a tR of 141 min (5). The solutions eluted under these
peaks were collected and stored separately at -80 °C. Following
purification by HPLC method IV, the resulting solutions were
freeze-dried.
Compound 5 was isolated as a pale yellow solid that in the
HPLC mobile phase (pH 2.15) exhibited UV bands (λmax) at 318,
282 (sh), and 248 nm and an ESI-MS with m/z 576 (MH+, 100%).
FAB-MS (dithiothreitol/dithioerythritol matrix): m/z 576.1452

752 Chem. Res. Toxicol., Vol. 13, No. 8, 2000
Xin et al.
(MH+, 2%, C21H30N5O10S2; calcd m/z 576.1434). 1H NMR (D2O):
δ 6.71 [s, 1H, C(8)-H], 5.27 [m, 1H, C(â)-H], 4.45 [m, 1H, C(3)-
H], 4.27 [dd, J ) 8.5, 4.5 Hz, 1H, C(d)-H], 3.75 [d, J ) 18.0 Hz,
1H, C(f)-H], 3.70 [d, J ) 18.0 Hz, 1H, C(f)-H], 3.68 [t, J ) 6.5
Hz, 1H, C(a)-H], 3.18 [m, 1H, C(2)-H], 3.10-3.05 [m, 2H, C(2)-
H, C(R)-H], 3.02-2.92 [m, 3H, C(e)-H2, C(R)-H], 2.34-2.22 [m,
2H, C(c)-H2], 1.97-1.92 [m, 2H, C(b)-H2].
Compound 6 was a pale yellow solid that in the HPLC mobile
phase exhibited UV bands λmax at 318 and 246 nm and an ESI-
MS with m/z 576 (MH+, 100%). FAB-MS (thioglycerol matrix):
m/z 576.1452 (MH+, 1%, C21H30N5O10S2; calcd m/z 576.1434).
1H NMR (D2O): δ 6.70 [s, 1H, C(6)-H], 5.39 [dd, J ) 8.0, 4.0
Hz, 1H, C(â)-H], 4.40 [dd, J ) 4.5, 3.5 Hz, 1H, C(3)-H], 4.30
[dd, J ) 8.5, 4.5 Hz, 1H, C(d)-H], 3.73 [d, J ) 18.0 Hz, 1H, C(f)-
H], 3.69 [d, J ) 18.0 Hz, 1H, C(f)-H], 3.65 [t, J ) 6.5 Hz, 1H,
C(a)-H], 3.20 [dd, J ) 13.0, 4.5 Hz, 1H, C(2)-H], 3.11-3.01 [m,
3H, C(2)-H, C(e)-H, C(R)-H], 2.99-2.93 [m, 2H, C(e)-H, C(R)-
H], 2.32 [t, J ) 7.5 Hz, 2H, C(c)-H2], 2.01-1.90 [m, 2H, C(b)-
H2].
State 3 (complex I) respiration was then initiated by addition
of 2.0 µL of 0.27 M ADP (final concentration of 0.9 mM).
NADH-CoQ1 Red u cta se Assa ys. Rat brain mitochondria
were exposed to six freeze-thaw cycles immediately before use
to ensure maximal NADH-CoQ1 reductase activity. The activity
of NADH-CoQ1 reductase was measured by monitoring the
decrease in NADH concentration with time at 30 °C using a
Beckman DU640 spectrophotometer as described previously (39,
42). Rat brain mitochondrial membranes were g95% rotenone
sensitive in NADH-CoQ1 reductase assays.
Meth od I. This method was employed to follow the time
course of NADH-CoQ1 reductase inhibition by the compounds
of interest. Frozen and thawed rat brain mitochondria (200-
400 µg of protein) were incubated with (sample) or without
(control) DHBT-NE-1 in 250 µL of 20 mM potassium phosphate
buffer (pH 8.0) at 30 °C for times ranging from 0 to 120 min.
Then 25 µL of the resulting solution (containing 20-40 µg of
mitochondrial protein) was transferred to the assay solution that
consisted of 850 µL of 20 mM potassium phosphate buffer (pH
8.0), 50 µL of 30 mM KCN, 50 µL of asolectin (15 mg/mL), and
50 µL of 1 mM CoQ1 (in 10% ethanol/water, v/v) contained in a
0.5 cm optical path length quartz UV cell. With the exception
of CoQ1, all other reagents were dissolved in 20 mM potassium
phosphate buffer (pH 8.0). The solution was incubated at 30 °C
for 5 min, and then 25 µL of 10 mM NADH was added to initiate
the reaction. The activity of NADH-CoQ1 reductase (nanomoles
of NADH per minute per milligram of mitochondrial protein)
was based on the decrease in the concentration of NADH
measured at 340 nm using a molar absorptivity of 6.22 mM-1
Isola tion of Meta bolites F or m ed by th e Mitoch on d r ia l
Mem br a n e-Ca ta lyzed Oxid a tion of DHBT-NE-1 in th e
P r esen ce of GSH. DHBT-NE-1 (1 mM), GSH (5 mM), and rat
brain mitochondrial membranes (2 µg of mitochondrial protein/
µL) were incubated in 10 mL of 20 mM potassium phosphate
buffer (pH 8.0) at 30 °C for 4-12 h. Following centrifugation
(10000g for 10 min at 2 °C) and filtration of the resulting
supernatant (0.45 µm Millipore HA-type filter), the filtrate was
injected onto the reversed phase column and components were
separated by HPLC method IV. The solutions eluted under the
peaks corresponding to 6 (tR ) 112 min), 8 (tR ) 136 min), 5 (tR
) 141 min), and 9 (tR ) 150 min) were separately collected,
immediately frozen, and stored at -80 °C. Following several
repetitive experiments, collected solutions containing 5 or 6 were
purified (HPLC method IV) and then freeze-dried.
cm-1
.
Met h od II. This method was employed to investigate the
irreversible inhibition of NADH-CoQ1 reductase. Frozen and
thawed rat brain mitochondria (300-400 µg of protein) were
incubated at 30 °C for 60 min in a total volume of 200 µL of 20
mM potassium phosphate buffer (pH 8.0) in the absence (control)
or presence of DHBT-NE-1. In some experiments, GSH, cata-
lase, or SOD was included in this incubation mixture. After 60
min, 400 µL of 20 mM phosphate buffer (pH 8.0) was added
and then the mixture was centrifuged (150000g for 10 min at 5
°C). The supernatant was discarded and the pellet gently
resuspended in 100 µL of 20 mM phosphate buffer (pH 8.0). A
25 µL aliquot of this solution (containing 50-100 µg of mito-
chondrial protein) was removed and assayed for NADH-CoQ1
reductase activity as described in Method I.
Efforts to isolate 8 and 9 as pure solids were unsuccessful
because of their decomposition during the final freeze-drying
step. Accordingly, vigorously stirred, freshly chromatographed
solutions of 8 or 9 were treated with solid NaBH4 until the pH
reached 6-7. After 10 min, the resulting solution was adjusted
to pH 2.15 with TFA and then pumped directly onto the reversed
phase column and components were separated by preparative
HPLC method IV. The reduced (DHBT) forms of 8 and 9, i.e.,
10 and 11, respectively, eluted at 165 and 163 min, respectively.
The solutions containing 10 or 11 were purified (HPLC method
IV) and freeze-dried.
Ra t Br a in Mitoch on d r ia l P r ep a r a tion s. Brain mitochon-
dria were prepared from male albino Sprague-Dawley rats
(Harlan Sprague-Dawley, Madison, WI) using procedures de-
scribed in detail elsewhere (39). The final mitochondrial pellet
was suspended in medium A [300 mM mannitol, 15.1 mM
Trizma hydrochloride, 15 mM glycylglycine, and 0.1 mM K2-
EDTA (pH 7.4)]. For oxygen consumption (mitochondrial res-
piration) experiments, mitochondria were prepared fresh each
day and were stored on ice prior to use. For NADH-CoQ1
reductase, R-KGDH, and PDHC assays, mitochondria were
stored at -80 °C until they were needed. The amount of
mitochondrial protein was determined by the method of Lowry
et al. (49).
P DHC a n d R-KGDH Assa y. Procedures for PDHC and
R-KGDH assays were based on those described by Tabatabaie
et al. (50) and Lai and Cooper (51), respectively. Briefly, to 500
µL of a solution containing 6 mM NAD+, 2 mM MgCl2, and 0.4
mM TPP in 50 mM potassium phosphate buffer (pH 7.4) were
added 100 µL of CoA (6 mM), 50 µL of pyruvic acid (20 mM, for
the PDHC assay) or R-ketoglutarate (20 mM, for the R-KGDH
assay), 50 µL of 2% Triton X-100, and 60 µL of DTT (5 mM), all
dissolved in 50 mM phosphate buffer (pH 7.4), and 200 µL of
50 mM phosphate buffer (pH 7.4). The resultant solution, in an
optical quartz UV cuvette, was thoroughly mixed, and after
thermal equilibrium had been attained (5 min at 30 °C), 40 µL
of frozen and thawed rat brain mitochondria in medium A
(containing 80 µg of mitochondrial protein) was added to initiate
the reaction. The activity of PDHC or R-KGDH was based on
the rate of increase of NADH production (nanomoles of NADH
formed per minute per milligram of mitochondrial protein)
measured at 340 nm between 30 and 240 s after initiation of
the reaction.
Oxygen Electr od e Stu d ies. Oxygen consumption by intact
rat brain mitochondria was assessed with a YSI (Yellow Springs
Instrument Co., Yellow Springs, OH) model 5300 biological
oxygen monitor equipped with a model 5357 micro oxygen probe
assembly (600 µL volume) and thermostated at 30 °C. Mito-
chondria (200-400 µg of protein) were suspended in 600 µL of
air-saturated medium B [300 mM mannitol, 10 mM Trizma
hydrochloride, 10 mM KCl, 5 mM potassium phosphate, 0.1 mM
K2EDTA, and 1.0 mg/mL BSA (pH 7.2)] without (control) or with
The effect of DHBT-NE-1 on the activity of PDHC or R-KGDH
was investigated by incubating this putative metabolite in 50
mM potassium phosphate buffer (pH 7.4) with frozen and
thawed rat brain mitochondria (2 µg/µL of protein) for times of
e120 min. At predetermined times, a 40 µL aliquot of this
solution was transferred into the normal assay solution to
determine PDHC activity.
known concentrations of DHBT-NE-1 for
a predetermined
incubation time (2-10 min) prior to measurement of oxygen
consumption. Then, 2.0 µL of pyruvate with malate (each at
concentrations of 0.75 M to give a final concentration of 2.5 mM)
was added, and oxygen consumption was measured for 2 min.

Mitochondrial Toxicity of Cysteinylnorepinephrine
Chem. Res. Toxicol., Vol. 13, No. 8, 2000 753
Ta ble 1. Tim e-Dep en d en t In h ibition of NADH-CoQ1 Red u cta se by DHBT-NE-1 Wh en In cu ba ted w ith F r ozen a n d Th a w ed
Ra t Br a in Mitoch on d r ia a
NADH-CoQ1 reductase activity expressed as % of control activity at time zerob,c
incubation
time (min)
control
0.1 mM DHBT-NE-1 0.5 mM DHBT-NE-1 2.0 mM DHBT-NE-1 5.0 mM DHBT-NE-1 10 mM DHBT-NE-1
0
20
40
60
80
100 ( 1.9d
92.7 ( 1.4
89.5 ( 1.5
83.4 ( 1.6
73.9 ( 1.9
99.9 ( 4.1
88.0 ( 3.3
83.7 ( 1.8
74.5 ( 3.0
64.9 ( 2.3*
96.2 ( 5.9
91.8 ( 4.2
80.2 ( 4.0
67.5 ( 5.5*
54.5 ( 5.8**
100.0 ( 8.6
77.5 ( 3.9
65.0 ( 5.9*
50.8 ( 3.6***
47.9 ( 5.6***
93.8 ( 4.2
87.1 ( 3.3
79.5 ( 4.2
77.3 ( 4.9
66.5 ( 4.3*
53.1 ( 5.0***
45.9 ( 1.7***
65.0 ( 5.9*
50.8 ( 3.6***
37.2 ( 2.3***
a
Frozen and thawed rat brain mitochondria (ca. 1.2 µg/µL of mitochondrial protein) were incubated in 20 mM potassium phosphate
buffer (pH 8.0) in the absence (control) or presence of DHBT-NE-1 at the specified concentration. Aliquots were assayed by method I at
the indicated times for NADH-CoQ1 reductase activity. Data are means ( SEM (n g 4). c *p < 0.05. **p < 0.001. ***p < 0.0001 (statistical
b
analysis compared the effects of DHBT-NE-1 on the activity of NADH-CoQ1 reductase to control activity measured at the same time).
The control activity at time zero of NADH-CoQ1 reductase was 221.8 ( 4.3 nmol of NADH min-1 (mg of mitochondrial protein)-1
.
d
F igu r e 2. Irreversible inhibition of NADH-CoQ1 reductase by
DHBT-NE-1 when incubated for 60 min with frozen and thawed
rat brain mitochondria. Frozen and thawed mitochondria (ca.
1.8 µg of protein) were incubated in 200 µL of 20 mM potassium
phosphate buffer (pH 8.0) at 30 °C with the indicated concentra-
tion of DHBT-NE-1. After 60 min, 400 µL of 20 mM phosphate
buffer (pH 8.0) was added and the resulting solution centrifuged
(150000g for 10 min at 5 °C). The washed pellet was gently
resuspended in 100 µL of 20 mM phosphate buffer (pH 8.0) and
a 25 µL aliquot removed and assayed for NADH-CoQ1 reductase
activity. Data are expressed as means ( SEM (bars) (n g 4).
One asterisk indicates p < 0.05; two asterisks indicate p < 0.001
and three asterisks p < 0.0001.
F igu r e 1. Effect of (A) 1 mM and (B) 2 mM DHBT-NE-1 on
malate- and pyruvate-supported state 3 (complex I) oxygen
consumption (respiration) after preincubation for the times
indicated with intact rat brain mitochondria (ca. 1.2 µg of
protein/µL) in medium B (pH 7.2). Oxygen consumption was
assessed with a Clark-type oxygen electrode assembly. Data are
means ( SEM (n g 6). One asterisk indicates p < 0.05; two
asterisks indicate p < 0.001.
when incubated for up to 10 min with intact rat brain
mitochondria.
Effects of DHBT-NE-1 on NADH-CoQ1 Red u cta se
Activity. When incubated with frozen and thawed rat
brain mitochondria (mitochondrial membranes), DHBT-
NE-1 evoked a time- and concentration-dependent inhibi-
tion of NADH-CoQ1 reductase (Table 1). A solution of 0.1
mM DHBT-NE-1 required incubation for >60 min with
mitochondrial membranes before the inhibition of NADH-
CoQ1 reductase reached statistical significance. However,
even the highest concentration of DHBT-NE-1 (10 mM)
that was employed required approximately incubation for
40 min with mitochondrial membranes before the inhibi-
tion of NADH-CoQ1 reductase reached statistical signifi-
cance.
Ir r ever sible In h ibition of NADH-CoQ1 Red u cta se
by DHBT-NE-1. When mitochondrial membranes were
washed following incubation (60 min) with DHBT-NE-1
(0.1-10 mM), NADH-CoQ1 reductase activity remained
significantly reduced compared to control activities (Fig-
ure 2). These results indicate that DHBT-NE-1 evokes a
time-dependent and irreversible inhibition of NADH-
CoQ1 reductase.
Sta tistics. All results were obtained at least in triplicate and
are presented as means ( SEM. A Student’s t test was employed
to determine statistical significance with a p of <0.05 being
taken to indicate a significant difference.
Resu lts
Effects of DHBT-NE-1 on Mitoch on d r ia l Resp ir a -
tion . When preincubated with intact rat brain mitochon-
dria for 2 or 5 min prior to addition of malate and
pyruvate and then (2 min later) ADP (to stimulate state
3 or complex I), 1 mM DHBT-NE-1 had no significant
effect on state 3 respiration compared to controls (Figure
1A). However, preincubation of 1 mM DHBT-NE-1 for
10 min with mitochondria evoked a small but statistically
significant decrease of state 3 respiration (Figure 1A). A
2 mM solution of DHBT-NE-1 had no effect on state 3
respiration when preincubated with mitochondria for 2
min. However, after longer preincubation times (5 and
10 min), 2 mM DHBT-NE-1 evoked a progressive de-
crease of state 3 oxygen consumption (Figure 1B). DHBT-
NE-1 (5 mM) evoked significant inhibition of complex I
respiration (50 ( 14% of control oxygen consumption; p
< 0.01) when preincubated with intact mitochondria for
only 2 min. DHBT-NE-1 (e5 mM) had no significant
effect on succinate-supported (complex II) respiration
When SOD (1.2 units/µL) and/or catalase (4.7 or 13
units/µL) were included in the initial incubation of
mitochondrial membranes (2 µg of protein/µL) with
DHBT-NE-1 for 60 min, NADH-CoQ1 reductase remained
inhibited after the membranes were washed (data not
shown). However, when no lower than equimolar GSH
was included in the initial incubation mixture, the

754 Chem. Res. Toxicol., Vol. 13, No. 8, 2000
Xin et al.
Ta ble 2. Ra t Br a in Mitoch on d r ia l Mem br a n e-Ca ta lyzed Oxid a tion of DHBT-NE-1 (1 m M) a n d Ir r ever sible In h ibition of
NADH-CoQ1 Red u cta sea ,b
DHBT-NE-1 oxidizedc
NADH-CoQ1 reductase activity expressed
mitochondrial
as % of control activity at time zero
autoxidation membrane-catalyzed BT-NE-1d formed BT-NE-2e formed
time (min)
0
(nmol)
oxidation (nmol)
(nmol)
(nmol)
control
experimental
0
11 ( 2
NDf
NDf
100 ( 2
(216 ( 5)g
84 ( 3
93 ( 6
60
120
21 ( 3
43 ( 7
185 ( 1
200 ( 0
1 ( 1
0 ( 0
110 ( 6
41 ( 2
62 ( 5**
35 ( 4***
74 ( 4
a
DHBT-NE-1 (1 mM, 200 nmol) in 20 mM potassium phosphate buffer (pH 8.0) at 30 °C was incubated with frozen and thawed rat
brain mitochondria (2 µL of protein/µL). At the indicated times, aliquots were removed for analytical HPLC analysis for DHBT-NE-1,
BT-NE-1, and BT-NE-1 and assayed for irreversible inhibition of NADH-CoQ1 reductase activity as described in Materials and Methods.
b
Data are means ( SEM (n ) 6). c The amount of DHBT-NE-1 remaining in the incubation solution was determined by analytical HPLC
d
analysis of an aliquot of the incubation solution (50 µL) following 10-fold dilution. The amount of BT-NE-1 that formed was measured
in a 10-fold-diluted aliquot of the incubation solution following NaBH4 reduction by the increase of the magnitude of the DHBT-NE-1
peak in analytical HPLC. e The amount of BT-NE-2 was measured in a 10-fold-diluted aliquot of the incubation solution following NaBH4
reduction by the height of the resultant BT-NE-2R peak in analytical HPLC. f Not determined. g Control activity of NADH-CoQ1 reductase
at time zero (nanomoles of NADH per minute per milligram of mitochondrial protein). **p < 0.001. ***p < 0.0001.
F igu r e 3. Effects of GSH on the irreversible inhibition of
F igu r e 4. Effects of DHBT-NE-1 (1.0 mM) on the inhibition
of (A) PHDC and (B) R-KGDH when incubated with rat brain
mitochondrial membranes (ca. 2 µg of mitochondrial protein/
µL) in phosphate buffer (pH 7.4). Enzyme activities were
measured as described in Materials and Methods. Black col-
umns are control and stipled columns are experimental activities
presented as means ( SEM (n g 6). Statistical analysis-
compared activities of experimentals and controls at the same
time. One asterisk indicates p < 0.05; two asterisks indicate p
< 0.001 and three asterisks p < 0.0001.
NADH-CoQ1 reductase by 10 mM DHBT-NE-1. Frozen and
thawed rat brain mitochondria (ca. 2 µg of protein/µL) were
incubated for 60 min in 20 mM potassium phosphate buffer (pH
8.0) together with GSH at the indicated concentrations. After
the mitochondrial membranes had been washed, NADH-CoQ1
reductase activities were measured via method II (Materials and
Methods). Data are means ( SEM (n g 6). One asterisk
indicates p < 0.05, and two asterisks indicate p < 0.001.
irreversible inhibition of NADH-CoQ1 reductase normally
evoked by DHBT-NE-1 was completely blocked
(Figure 3).
E ffect s of DHBT-NE -1 on P DHC a n d R-KGDH
Activities. When incubated with rat brain mitochondrial
membranes in 50 mM phosphate buffer (pH 7.4), DHBT-
NE-1 evoked a time-dependent inhibition of both PDHC
(Figure 4A) and R-KGDH (Figure 4B). Neither SOD nor
catalase was able to block or attenuate the inhibition of
PDHC or R-KGDH by DHBT-NE-1. However, no lower
than equimolar GSH completely blocked the time-de-
pendent inhibition of PDHC and R-KGDH evoked by
DHBT-NE-1 (data not shown).
Mitoch on d r ia l Mem br a n e-Ca ta lyzed Oxid a tion of
DHBT-NE-1. In the presence of rat brain mitochondrial
membranes, the oxidation of DHBT-NE-1 was apprecia-
bly more rapid than its autoxidation (Table 2). The rate
of oxidation of DHBT-NE-1 increased with increasing
concentrations of mitochondrial protein (data not shown),
indicating that mitochondrial membranes catalyze the
oxidation reaction.
F igu r e 5. Analytical HPLC chromatograms of the product
solutions formed following incubation of DHBT-NE-1 (1.0 mM)
with frozen and thawed rat brain mitochondria (2 µg of protein/
µL) in 20 mM potassium phosphate buffer (pH 8.0) at 30 °C for
(A) 0, (B) 1, (C) 2, and (D) 1 h followed by NaBH4 reduction and
for (E) 2 h followed by NaBH4 reduction.
Analytical HPLC chromatograms of product solutions
formed at various times during incubation of DHBT-NE-1
(1 mM) with rat brain mitochondrial membranes at pH
8 are presented in Figure 5A-C. Thus, the HPLC peak
for DHBT-NE-1 completely disappeared in <120 min,
although under the chromatographic conditions that were
employed no significant product peaks appeared. The
expected products of oxidation of DHBTs such as DHBT-
NE-1 are BTs. The latter compounds are unstable in the
low-pH HPLC mobile phases that were employed (52).
However, BTs are stable for several hours in solution at
pH 7.4 and under such conditions can be quantitatively
reduced by NaBH4 to their respective and more stable
DHBTs that can be readily assessed by analytical HPLC

Mitochondrial Toxicity of Cysteinylnorepinephrine
Chem. Res. Toxicol., Vol. 13, No. 8, 2000 755
Sch em e 2
Ta ble 3. Effect of 5 m M GSH on th e Ra t Br a in
Mitoch on d r ia l Mem br a n e-Ca ta lyzed Oxid a tion of 1 m M
DHBT-NE-1 a t p H 8.0a
using low-pH mobile phases (52). An analytical HPLC
chromatogram of the product solution formed after
incubation of DHBT-NE-1 with mitochondrial mem-
branes for 1 h, following NaBH4 reduction, is presented
in Figure 5D. The large peak for BT-NE-2R indicates that
BT-NE-2 (see Scheme 2 for structures) is a major product
of the mitochondrial membrane-catalyzed oxidation of
DHBT-NE-2 at this stage of the reaction. However, after
2 h, the magnitude of the chromatographic peak for BT-
NE-2R significantly decreased (Figure 5E), indicating
that BT-NE-2 is further oxidized by mitochondrial mem-
branes. NaBH4 reduction of product solutions formed
after incubations of DHBT-NE-1 with mitochondrial
membranes for 1 h (Figure 5D) or 2 h (Figure 5E) did
not result in significant increases in the magnitude of
the chromatographic peak for DHBT-NE-1, indicating
that very little free BT-NE-1 is formed in the reaction.
HPLC analysis of product solutions formed when
DHBT-NE-1 was incubated with mitochondrial mem-
branes revealed not only that combined yields of BT-NE-1
and BT-NE-2 were smaller than the amount of DHBT-
NE-1 oxidized but also that the latter metabolites were
also oxidized (Table 3). Furthermore, the time depen-
time DHBT-NE-1 oxidized BT-NE-1 formed BT-NE-2 formed
(min)
(nmol)b
(nmol)b,c
(nmol)b,c
0
60
120
8 ( 2
172 ( 8
197 ( 1
NDd
18 ( 5
6 ( 2
NDd
7 ( 2
5 ( 1
a
DHBT-NE-1 (1.0 mM, 200 nmol) and GSH (5.0 mM, 1000
nmol) in 200 µL of 20 mM potassium phosphate buffer (pH 8.0)
were incubated at 30 °C with 400 µg of frozen and thawed rat
b
brain mitochondrial protein. Data are means ( SEM (n ) 6).
c BT-NE-1 and BT-NE-2 were determined by analytical HPLC
following NaBH4 reduction as described in Materials and Methods.
d
Not determined.
dence of the irreversible inhibition of NADH-CoQ1 re-
ductase appeared to be related to the mitochondrial
membrane-catalyzed oxidation of DHBT-NE-1 and then
BT-NE-1 and/or BT-NE-2. It has not yet been possible
to isolate pure samples of either BT-NE-1 or BT-NE-2.
However, to investigate the influence, particularly of BT-
NE-2, on NADH-CoQ1 reductase activity, the product
solutions formed following a 60 min incubation of DHBT-

756 Chem. Res. Toxicol., Vol. 13, No. 8, 2000
Xin et al.
NE-1 (1 mM) with mitochondrial membranes (Table 2)
were centrifuged (14000g for 4 °C at 10 min) and the
supernatant was removed. These supernatants contained
0.54 ( 0.03 mM BT-NE-2 but only traces of BT-NE-1
(<0.01 mM) and DHBT-NE-1 (<0.08 mM). The super-
natant (100 µL) was incubated with fresh mitochondrial
membranes (2.4 µg of protein/µL) in a total volume of 110
µL of 20 mM phosphate buffer (pH 8.0) at 30 °C for 60
min. HPLC analysis of the supernatant at this time
revealed that the concentration of BT-NE-2 had de-
creased to 0.18 ( 0.02 mM. Furthermore, an aliquot (25
mL) of the supernatant was assayed for NADH-CoQ1
reductase activity. This assay revealed that the activity
of NADH-CoQ1 reductase was 67.3 ( 4.4% (p < 0.05; n
) 5) of control activity, suggesting that the mitochondrial
membrane-catalyzed oxidation of BT-NE-2 was related
to the inhibition of this enzyme complex.
Although no lower than equimolar concentrations of
GSH blocked the inhibition of NADH-CoQ1 reductase,
PDHC, and R-KGDH, even a large excess of the tripeptide
had no significant effect on the mitochondrial membrane-
catalyzed oxidation of DHBT-NE-1 (Table 3). However,
GSH dramatically decreased the yield of BT-NE-2 formed
in this reaction and increased the yields of BT-NE-1,
although these products only accounted for a small
fraction of the oxidized DHBT-NE-1. An analytical HPLC
chromatogram of the product solution formed 2 h after
incubation of DHBT-NE-1 (1 mM) and GSH (5 mM) with
mitochondrial membranes is presented in Figure 6B and
shows not only a massive decrease in the level of DHBT-
NE-1 from its initial concentration (Figure 6A) but also
the appearance of a number of new metabolites. These
included the 6-(S)-5 and 8-(S)-6 glutathionyl conjugates
of DHBT-NE-1 and compounds 8 and 9. Following NaBH4
reduction of the latter product solution, analytical HPLC
showed small chromatographic peaks for DHBT-NE-1
and BT-NE-2R, indicative of the formation of BT-NE-1
and BT-NE-2, respectively (Figure 6C). In addition,
following NaBH4 reduction, the peaks for compounds 8
and 9 disappeared and were replaced by a single peak
corresponding to a mixture of 10 and 11, which coeluted
under the analytical HPLC conditions that were em-
ployed (Figure 6C).
Freshly collected solutions of 8 and 9, prepared using
preparative HPLC method IV, were bright yellow and
exhibited identical UV/vis spectra (λmax ) 354 and 252
nm at pH 2.15) and ESI-MS data (m/z 576, MH+, 100%),
indicating that they are isomers. Because of their de-
composition during the final freeze-drying step, it was
not possible to isolate 8 and 9. Nevertheless, the UV/vis
spectra of 8 and 9 were identical to those of epimers of
2-S-glutathionyl-7-(2-aminoethyl)-5-hydroxy-1,4-benzothi-
azine-3-carboxylic acid (ESI-MS, m/z 558) that differ only
in the absence of the side chain â-OH residue (41),
suggesting that 8 and 9 are epimers of the 2-S-glutathio-
nyl conjugate of BT-NE-1. Compounds 10 and 11, formed
by the NaBH4 reduction of 8 and 9, respectively, could
be separated by preparative HPLC method IV. Dissolved
in the HPLC mobile phase (pH 2.15), 10 and 11 also
exhibited identical UV spectra [λmax ) 302, 248 (sh), and
228 nm] and ESI-MS data (m/z 576, MH+, 100%) corre-
sponding to a glutathionyl conjugate of DHBT-NE-1. The
UV spectra of 10 and 11, however, were different from
those of 5 [λmax ) 318, 282 (sh), and 248 nm at pH 2.15]
and 6 (λmax ) 318 and 248 nm at pH 2.15) as were their
tR values in analytical (and preparative) HPLC (Figure
F igu r e 6. Analytical HPLC chromatograms of the product
solutions formed following incubation of DHBT-NE-1 (1.0 mM)
and GSH (5 mM) with frozen and thawed rat brain mitochondria
(2 µg of protein/µL) in 20 mM potassium phosphate buffer (pH
8.0) at 30 °C for (A) 0, (B) 2, and (C) 2 h followed by NaBH4
reduction.
6C). This indicated that 10 and 11 are 2-S-glutathionyl
conjugates of DHBT-NE-1 and, indeed, that the UV
spectra of these compounds are virtually identical to
those of the diastereomeric 2-S-glutathionyl conjugates
of 7-(2-aminoethyl)-5-hydroxy-2H-1,4-benzothiazine-3-
carboxylic acid (41). It was not possible to isolate suf-
ficient quantities of 10 and 11 to permit their complete
structure elucidation by spectroscopic methods (1H NMR
or FAB-MS). However, the CD spectra of 10 and 11

Mitochondrial Toxicity of Cysteinylnorepinephrine
Chem. Res. Toxicol., Vol. 13, No. 8, 2000 757
catalyze this reaction. The fact that the inhibition of
NADH-CoQ1 reductase, PDHC, and R-KGDH by DHBT-
NE-1 is unaffected by SOD and/or catalase implies that
this inhibition is not caused by oxygen radical-mediated
damage to these enzyme complexes.
When incubated with intact mitochondria, DHBT-NE-1
(1-2 mM) evokes significant inhibition of complex I
respiration within e10 min (Figure 1). However, the
same concentrations of DHBT-NE-1 require g40 min to
evoke significant inhibition of NADH-CoQ1 reductase
when incubated with frozen and thawed mitochondria
having approximately the same protein content (Tables
1 and 2 and Figure 2). These observations may indicate
that either intact mitochondria transport DHBT-NE-1
such that it attains high intramitochondrial concentra-
tions needed to evoke rapid inhibition of complex I or the
rate of its intramitochondrial oxidation is significantly
greater than with mitochondrial membranes.
The observation that no lower than equimolar concen-
trations of GSH block inhibition of NADH-CoQ1 reduc-
tase, PDHC, and R-KGDH when DHBT-NE-1 is incu-
bated with mitochondrial membranes with resultant
formation of glutathionyl conjugates 5, 6, 8, and 9
(Scheme 2) suggests that electrophilic intermediates
formed by the catalyzed oxidation of DHBT-NE-1 and BT-
NE-1 are responsible for inhibition of these enzyme
complexes. The initial step in this reaction is the oxida-
tion of DHBT-NE-1 to o-quinone imine 2 that, in the
presence of GSH, is scavenged to give glutathionyl
conjugates 5 and 6 (Scheme 2). However, a fraction of 2
rearranges and/or decarboxylates to form BT-NE-1 and
BT-NE-2 that are further oxidized in a reaction catalyzed
by mitochondrial membranes (Table 2). The proximate
oxidation product of BT-NE-1 is o-quinone imine 7 which
reacts with GSH to give epimers 8 and 9.
F igu r e 7. CD spectra of (A) 10 and 11 and (B) DHBT-NE-1.
Compounds were dissolved in water at a concentration of
approximately 0.4 mM.
The preceding results suggest that, in the absence of
GSH, electrophilic o-quinone imine 2, formed by the
mitochondrial membrane-catalyzed oxidation of DHBT-
NE-1, may evoke irreversible inhibition of NADH-CoQ1
reductase by covalent modification of cysteinyl-SH resi-
dues at or close to the active site of complex I (53, 54),
forming adducts conceptualized as 12 and 13 (Scheme
2). Similarly, o-quinone imine 7 formed by the mitochon-
drial membrane-catalyzed oxidation of BT-NE-1 may also
covalently modify such cysteinyl residues forming 14.
The major initial metabolite of the mitochondrial
membrane-catalyzed oxidation of DHBT-NE-1 is BT-
NE-2 which is further oxidized (Table 2) to an intermedi-
ate which appears to have high thiol reactivity (Table 3)
and inhibits NADH-CoQ1 reductase. These results sug-
gest that BT-NE-2 is oxidized to o-quinone imine 15
which reacts with active site cysteinyl residues of complex
I forming 16 (Scheme 2), evoking irreversible inhibition
of NADH-CoQ1 reductase. However, this conclusion is
somewhat speculative in view of the fact that glutathionyl
conjugates of BT-NE-2 have not yet been detected fol-
lowing incubations of DHBT-NE-1 (or BT-NE-2) and
mitochondrial membranes in the presence of GSH. Co-
valent modification of active site cysteinyl residues of
PDHC (55, 56), R-KGDH (57), and/or their dihydrolipoate
cofactor by 2, 7, and possibly 15 might account for
inhibition of these enzyme complexes by DHBT-NE-1.
A hypothesis for the pathogenic mechanism that might
underlie the degeneration of dopaminergic SNc cells in
PD and of nigrostriatal DA neurons evoked by MPTP/
MPP+ was summarized in the Introduction (3). Key steps
(Figure 7A) indicated that these compounds exhibited
opposite ellipticities. The CD spectrum of DHBT-NE-1,
which retains the R chirality of CySH at C(3) (Figure 7B),
exhibits the same ellipticity as 10 (Figure 7A), leading
to the conclusion that 10 is the 2R,3R diastereomer of
2-S-glutathionyl-DHBT-NE-1 and hence that 8 is the 2R
epimer of 2-S-glutathionyl-BT-NE-1. Accordingly, 11 is
the 2S,3S diastereomer of 2-S-glutathionyl-DHBT-NE-
1, and hence, 9, is the 2S epimer of 2-S-glutathionyl-BT-
NE-1.
Discu ssion
DHBT-NE-1 evokes a time- and concentration-depend-
ent inhibition of complex I, but not complex II, respiration
when incubated with intact rat brain mitochondria
(Figure 1), implying that this compound can reach the
inner mitochondrial membrane. DHBT-NE-1 also evokes
time- and concentration-dependent irreversible inhibition
of NADH-CoQ1 reductase (Table 1 and Figure 2) and
PDHC and R-KGDH (Figure 4), all enzyme complexes
associated with the inner mitochondrial membrane.
In vitro, mitochondrial membranes catalyze the oxida-
tion of DHBT-NE-1 to BT-NE-1 and BT-NE-2 that are
further oxidized (Table 2). The time dependence of
NADH-CoQ1 reductase inhibition by DHBT-NE-1 ap-
pears to be related to its oxidation by a constituent of
the inner mitochondrial membrane, known to contain
many iron and copper enzyme complexes that might

758 Chem. Res. Toxicol., Vol. 13, No. 8, 2000
Xin et al.
disorders. In Neurodegeneration and Neuroprotection in Parkin-
son’s Disease (Olanow, C. W., J enner, P., and Youdim, M., Eds.)
pp 23-45, Academic Press, New York.
in this mechanism are (1) a massive release of DA and
GSH from dopaminergic neurons during a transient
energy impairment, (2) reuptake of DA together with
CySH as the energy impairment subsides, and (3) intra-
neuronal oxidation of DA by O2-• in the presence of CySH,
forming 5-S-CyS-DA and thence DHBTs and BTs that
inhibit mitochondrial complex I and R-KGDH. The CySH
required for the latter reaction is proposed to be formed
by the γ-GT/DP-mediated degradation of GSH released
(2) Gerlach, M., and Riederer, P. (1996) Animal models of Parkinson’s
disease: an empirical comparison of the phenomenology of the
disease in man. J . Neural Transm. 103, 987-1041.
(3) Han, J ., Cheng, F.-C., Yang, Z., and Dryhurst, G. (1999) Inhibitors
of mitochondrial respiration, iron(II) and hydroxyl radical evoke
release and extracellular hydrolysis of glutathione in rat striatum
and substantia nigra: potential implications to Parkinson’s
disease. J . Neurochem. 73, 1683-1695.
(4) Ransom, B. R., Kunis, D. M., Irwin, I., and Langston, J . W. (1987)
Astrocytes convert the parkinsonian inducing neurotoxin, MPTP,
to its active metabolite, MPP+. Neurosci. Lett. 75, 323-328.
(5) Ramsay, R. R., Dadgar, J ., Trevor, A., and Singer, T. P. (1986)
Energy driven uptake of N-methyl-4-phenylpyridine by brain
mitochondria mediates the neurotoxicity of MPTP. Life Sci. 39,
581-588.
(6) Cooper, J . M., and Schapira, A. H. V. (1997) Mitochondrial
dysfunction in neurodegeneration. J . Bioenerg. Biomembr. 29,
175-183.
(7) Mizuno, Y., Saitoh, T., and Sone, N. (1987) Inhibition of mito-
chondrial R-ketoglutarate dehydrogenase by 1-methyl-4-phe-
nylpyridinium ion. Biochem. Biophys. Res. Commun. 143, 971-
976.
(8) Beal, M. F. (1995) Mitochondrial Dysfunction and Oxidative
Damage in Neurodegenerative Diseases, R. G. Landes Co., Austin,
TX.
(9) Fleming, L., Mann, J . B., Bean, J ., Briggle, T., and Sanchez-
Ramos, J . R. (1994) Parkinson’s disease and brain levels of
organochlorine pesticides. Ann. Neurol. 36, 100-103.
(10) Sanchez-Ramos, J ., Fucca, A., Basit, A., and Song, S. (1998)
Toxicity of dieldrin for dopaminergic neurons in mesencephalic
cultures. Exp. Neurol. 150, 263-271.
(11) Meco, G., Bonifati, M. G., Vanacore, N., and Fabrizio, E. (1994)
Parkinson’s after exposure to the fungicide maneb (manganese
ethylene-bis-dithiocarbamate). Scand. J . Environ. Health 20,
301-305.
(12) Bachurin, S. O., Shevtzova, E. P., Lermontova, N. N., Serkova,
T. P., and Ramsay, R. R. (1996) The effect of dithiocarbamates
on neurotoxic action of 1-methyl-4-phenyl-1,2,3,6-tetrahydropy-
ridine (MPTP) and on mitochondrial respiration chain. Neuro-
toxicology 17, 897-903.
-•
from glia, as is the Glu that mediates intraneuronal O2
generation by activation of NMDA receptors. This hy-
pothesis is consistent with the fall of the nigral GSH level
without corresponding increases in the GSSG level (58),
increased SOD activity (59), an increased 5-S-CyS-DA/
DA ratio (37), and decreased complex I (42) and R-KGDH
(43) activities in the parkinsonian SNc. However, because
little is known about pathobiochemical changes that
occur in the parkinsonian LC, it is at present difficult to
judge whether a similar mechanism involving intraneu-
ronal O2-•-mediated oxidation of NE in the presence of
CySH forming 5-S-CyS-NE and thence DHBT-NE-1 with
resultant inhibition of mitochondrial complex I, R-KGDH,
and PDHC and impaired ATP production might contrib-
ute to the degeneration of noradrenergic neurons.
It is perhaps relevant, however, that the profound
neuronal energy impairment evoked by cerebral ischemia
initiates processes that ultimately result in the delayed
loss of specific subpopulations of neurons during subse-
quent recirculation. These include dopaminergic termi-
nals in the striatum and noradrenergic terminals in a
number of brain regions (60). Factors implicated in the
mechanism underlying the degeneration of these neurons
include the ischemia-mediated release of DA (61, 62) and
NE (63, 64), which reverses upon reperfusion, loss of GSH
without corresponding increases in the levels of GSSG
(65-67), elevation of extracellular levels of Glu (61), part
of which in addition to CySH (67) is derived from the
degradation of released GSH (68), with resultant NMDA
(13) Le Couteur, D. G., and McClean, A. J . (1998) The ageing liver:
drug clearance and an oxygen diffusion barrier hypothesis. Clin.
Pharmacokinet. 34, 359-373.
(14) Steventon, G. B., Heafield, M. T. E., Waring, R. H., and Williams,
A. C. (1989) Xenobiotic metabolism in Parkinson’s disease.
Neurology 39, 883-887.
-•
receptor activation (69) that mediates intraneuronal O2
generation (70), and the development of changes in
mitochondrial function, including reduced state 3 respi-
ration and decreased PDHC activity (71, 72). Taken
together, these observations suggest that the intraneu-
ronal O2-•-mediated oxidation of DA or NE in the pres-
ence of CySH might lead to the formation of DHBTs that
inhibit mitochondrial enzyme complexes and contribute
to the degeneration of dopaminergic and noradrenergic
neurons evoked by ischemia reperfusion. However, it
must be emphasized that there is presently no direct
evidence for formation of cysteinyl conjugates of DA or
NE or resultant DHBT mitochondrial toxicants in the
brain following transient cerebral ischemia. Indeed, these
and earlier (39, 40) results suggesting that DHBTs can
be accumulated by brain mitochondria and catalytically
oxidized to electrophilic intermediates that covalently
bind to mitochondrial proteins imply that such putative
metabolites may be difficult to detect as free bases.
(15) Menegon, A., Board, P. G., Blackburn, A. C., Mellick, G. D., and
Le Couteur, D. G. (1998) Parkinson’s disease, pesticides, and
glutathione transferase polymorphisms. Lancet 352, 1344-1346.
(16) Payami, H., and Zareparsi, S. (1998) Genetic epidemiology of
Parkinson’s disease. J . Geriatr. Psychiatry Neurol. 11, 98-106.
(17) Bandmann, O., Vaughan, J ., Holmans, P., Marsden, C. D., and
Wood, N. W. (1997) Association of slow acetylator genotype for
N-acetyltransferase 2 with familial Parkinson’s disease. Lancet
350, 1136-1139.
(18) Swerdlow, R. H., Parks, J . K., Miller, S. W., Tuttle, J . B., Trimmer,
P. A., Sheehan, J . P., Bennett, J . P., Davis, R. E., and Parker, W.
D. (1996) Origin and functional consequences of the complex I
defect in Parkinson’s disease. Ann. Neurol. 40, 663-671.
(19) Kobayashi, T., Matsumme, H., Matuda, S., and Mizuno, Y. (1998)
Association between the gene encoding the E2 subunit of the
R-ketoglutarate dehydrogenase complex and Parkinson’s disease.
Ann. Neurol. 43, 120-123.
(20) Rollema, H., Damsma, G., Horn, A. S., De Vries, J . B., and
Westerink, B. H. C. (1986) Brain dialysis in conscious rats reveals
an instantaneous massive release of striatal dopamine in response
to MPP+. Eur. J . Pharmacol. 126, 345-346.
(21) Za¨ngerle, L., Cuenod, M., Winterhalter, K. H., and Do, K.-Q.
(1992) Screening of thiol compounds: depolarization-induced
release of glutathione and cysteine from rat brain slices. J .
Neurochem. 59, 181-189.
(22) Wu, E. Y., Smith, M. T., Bellomo, G., and Di Monte, D. (1990)
Relationships betwen the mitochondrial transmembrane poten-
tial, ATP concentration, and cytotoxicity in isolated rat hepato-
cytes. Arch. Biochem. Biophys. 282, 258-262.
Ack n ow led gm en t. This work was supported by
National Institutes of Health Grant GM32367. Additional
support was provided by the Vice President for Research
at the University of Oklahoma.
Refer en ces
(23) Rigobello, M. P., and Bindoli, A. (1993) Effect of pyruvate on rat
heart thiol status during ischemia and hypoxia followed by
reperfusion. Mol. Cell. Biochem. 122, 93-100.
(1) J enner, P., and Olanow, C. W. (1996) Pathological evidence for
oxidative stress in Parkinson’s disease and related degenerative

Mitochondrial Toxicity of Cysteinylnorepinephrine
Chem. Res. Toxicol., Vol. 13, No. 8, 2000 759
(24) Zeevalk, G. D., and Nicklas, W. J . (1991) Mechanisms underlying
initiation of excitotoxicity associated with metabolic inhibition.
J . Pharmacol. Exp. Ther. 257, 870-878.
(25) Lafon-Cazal, M., Pletri, S., Culcasi, M., and Bockaert, J . (1993)
NMDA-dependent superoxide production and neurotoxicity. Na-
ture 364, 535-537.
(26) Brieland, J . K., and Fantone, J . C. (1991) Ferrous iron release
from transferrin by human neutrophil-derived superoxide an-
ion: effect of pH and iron saturation. Arch. Biochem. Biophys.
284, 78-83.
(27) Flint, D. H., Tuminello, J . F., and Emptage, M. H. (1993) The
inactivation of Fe-S cluster containing hydrolyases by superox-
ide. J . Biol. Chem. 268, 22369-22376.
(28) Yoshida, T., Tanaka, M., Somomatsu, A., and Hirai, S. (1995)
Activated microglia cause superoxide-mediated release of iron
from ferritin. Neurosci. Lett. 190, 21-24.
(29) Cooper, A. J . L. (1998) Role of astrocytes in maintaining cerebral
glutathione homeostasis and protecting the brain against xeno-
biotics and oxidative stress. In Glutathione in the Nervous System
(Shaw, C. A., Ed.) pp 91-115, Taylor and Francis, Washington,
DC.
(46) Scatton, B., J avoy-Agid, F., Rouquier, L., Dubois, B., and Agid,
Y. (1983) Reduction of cortical dopamine, noradrenaline, serotonin
and their metabolites in Parkinson’s disease. Brain Res. 275,
321-328.
(47) Gerlach, M., J ellinger, K., and Riederer, P. (1994) The possible
role of noradrenergic deficits in selected signs of Parkinson’s
disease. In Noradrenergic Mechanisms in Parkinson’s Disease
(Briley, M., and Marien, M., Eds.) pp 59-71, CRC Press, Boca
Raton, FL.
(48) Shen, X.-M., and Dryhurst, G. (1996) Oxidation chemistry of (-)-
norepinephrine in the presence of L-cysteine. J . Med. Chem. 39,
2018-2029.
(49) Lowry, O. H., Rosebrough, N. J ., Farr, A. L., and Randall, R. J .
(1951) Protein measurement with the Folin phenol reagent. J .
Biol. Chem. 193, 265-275.
(50) Tabatabaie, T., Potts, J . D., and Floyd, R. A. (1996) Reactive
oxygen species-mediated inactivation of pyruvate dehydrogenase.
Arch. Biochem. Biophys. 336, 290-296.
(51) Lai, J . C. K., and Cooper, A. J . L. (1996) Brain R-ketoglutarate
dehydrogenase complex: kinetic properties, regional distribution,
and effects of inhibitors. J . Neurochem. 47, 1376-1386.
(52) Shen, X.-M., Zhang, F., and Dryhurst, G. (1997) Oxidation of
dopamine in the presence of cysteine: characterization of new
toxic products. Chem. Res. Toxicol. 10, 147-155.
(30) Cao, C. J ., Eldefrawi, A. T., and Eldefrawi, M. E. (1990) ATP-
regulated neuronal catecholamine uptake: a new mechanism. Life
Sci. 47, 655-667.
(53) Runswick, M. J ., Gennis, R. B., Fearnley, I. M., and Walker, J .
E. (1989) Mitochondrial NADH:ubiquinone reductase: comple-
mentary DNA sequence of the import precursors of the bovine
75-kDa subunit. Biochemistry 28, 9452-9459.
(54) Masui, R., Wakabayashi, S., Matsubara, H., and Hatefi, Y. (1991)
The amino acid sequences of two 13 kDa polypeptides and partial
amino acid sequence of 30 kDa polypeptide of complex I from
bovine heart mitochondria: possible location of iron-sulfur
clusters. J . Biochem. (Tokyo) 109, 534-543.
(31) Spencer, J . P. E., J enner, P., Daniel, S. E., Lees, A. J ., Marsden,
C. D., and Halliwell, B. (1998) Conjugates of catecholamines with
cysteine and GSH in Parkinson’s disease. Possible mechanisms
of formation involving reactive oxygen species. J . Neurochem. 71,
2112-2122.
(32) Sagara, J ., Miura, K., and Bannai, S. (1993) Maintenance of
neuronal glutathione by glial cells. J . Neurochem. 61, 1672-1676.
(33) Tate, S. S., and Meister, A. (1985) γ-Glutamyl transpeptidase from
kidney. Methods Enzymol. 113, 400-437.
(55) Graupe, K., Trommer, W. E., and Bisswanger, H. (1989) Interac-
tion between catalytic and regulatory sites of the pyruvate
dehydrogenase from Escherichia coli studied by the ESR tech-
nique. Biochim. Biophys. Acta 999, 176-182.
(56) Ali, M. S., Roche, T. E., and Patel, M. S. (1993) Identification of
the essential cysteine residue in the active site of bovine pyruvate
dehydrogenase. J . Biol. Chem. 268, 22353-22356.
(57) Bunik, V. I., Buneeva, O. A., and Gomazkova, V. S. (1990)
Regulation of R-ketoglutarate dehydrogenase cooperative proper-
ties in substrate binding by thiol-disulfide exchange. Biochem.
Int. 21, 873-881.
(34) Bannai, S. (1984) Transport of cystine and cysteine in mammalian
cells. Biochim. Biophys. Acta 779, 289-306.
(35) Dringen, R., Pfeiffer, B., and Hamprecht, B. (1999) Synthesis of
the antioxidant glutathione in neurons: supply by astrocytes of
CysGly as precursor for neuronal glutathione. J . Neurosci. 19,
562-569.
(36) Carboni, S., Melis, F., Pani, L., Hadjiconstantinou, M., and
Rossetti, Z. L. (1990) The non-competitive NMDA-receptor an-
tagonist MK-801 prevents the massive release of glutamate and
aspartate from rat striatum induced by 1-methyl-4-phenylpyri-
dinium (MPP+). Neurosci. Lett. 117, 129-133.
(58) Sian, J ., Dexter, D., Lees, A., Daniel, S., Agid, Y., J avoy-Agid, F.,
J enner, P., and Marsden, C. D. (1994) Alterations in glutathione
levels in Parkinson’s disease and other neurodegenerative disor-
ders affecting the basal ganglia. Ann. Neurol. 36, 348-355.
(59) Saggu, H., Cooksey, J ., Dexter, D., Wells, F. R., Lees, A. J ., J enner,
P., and Marsden, C. D. (1989) A selective increase in particulate
superoxide dismutase activity in parkinsonian substantia nigra.
J . Neurochem. 53, 692-697.
(37) Fornstedt, B., Brun, A., Rosengren, E., and Carlsson, A. (1989)
The apparent autoxidation rate of catechols in dopamine-rich
regions of human brains increases with the depigmentation of
substantia nigra. J . Neural Transm.: Parkinson’s Dis. Dementia
Sect. 1, 279-295.
(38) Shen, X.-M., and Dryhurst, G. (1996) Further insights into the
influence of L-cysteine on the oxidation chemistry of dopamine:
reaction pathways of potential relevance to Parkinson’s disease.
Chem. Res. Toxicol. 9, 751-763.
(60) Sims, N. R., and Zaidan, E. (1995) Biochemical changes associated
with selective neuronal death following short-term cerebral
ischaemia. Int. J . Biochem. Cell Biol. 27, 531-550.
(39) Li, H., and Dryhurst, G. (1997) Irreversible inhibition of mito-
chondrial complex I by 7-(2-aminoethyl)-3,4-dihydro-5-hydroxy-
(61) Globus, M. Y.-T., Busto, R., Dietrich, W. D., Martinez, E., Valde´s,
I., and Ginsberg, M. D. (1988) Intraischemic extracellular release
of dopamine and glutamate is associated with striatal vulner-
ability to ischemia. Neurosci. Lett. 91, 36-40.
2H-1,4-benzothiazine-3-carboxylic acid (DHBT-1):
nigral endotoxin of relevance to Parkinson’s disease. J . Neuro-
chem. 69, 1530-1541.
a putative
(40) Li, H., Shen, X.-M., and Dryhurst, G. (1998) Brain mitochondria
catalyze the oxidation of 7-(2-aminoethyl)-3,4-dihydro-5-hydroxy-
2H-1,4-benzothiazine-3-carboxylic acid (DHBT-1) to intermediates
that irreversibly inhibit complex I and scavenge glutathione:
potential relevance to the pathogenesis of Parkinson’s disease.
J . Neurochem. 71, 2049-2062.
(41) Shen, X.-M., Li, H., and Dryhurst, G. (2000) Oxidative metabolites
of 5-S-cysteinyldopamine inhibit R-ketoglutarate dehydrogenase
complex: potential relevance to the pathogenesis of Parkinson’s
disease. J . Neural Transm. (in press).
(42) Schapira, A. H. V., Mann, V. M., Cooper, J . M., Dexter, D., Daniel,
S. E., J enner, P., Clark, J . B., and Marsden, C. D. (1990) Anatomic
and disease specificity of NADH-CoQ1 reductase (complex I)
deficiency in Parkinson’s disease. J . Neurochem. 55, 2142-2145.
(43) Mizuno, Y., Matuda, S., Yoshino, H., Mori, H., Hattori, N., and
Ikebe, S. (1994) An immunohistochemical study on R-ketoglut-
arate dehydrogenase complex in Parkinson’s disease. Ann. Neurol.
35, 204-210.
(62) Globus, M. Y.-T., Ginsberg, M. D., Dietrich, W. D., Busto, R., and
Scheinberg, P. (1987) Substantia nigra lesion protects against
ischemic damage in the striatum. Neurosci. Lett. 80, 251-256.
(63) Globus, M. Y.-T., Busto, R., Dietrich, W. D., Martinez, E., Valde´s,
I., and Ginsberg, M. D. (1989) Direct evidence for acute and
massive norepinephrine release in the hippocampus during
transient ischemia. J . Cereb. Blood Flow Metab. 9, 892-896.
(64) Busto, R., Harik, S. I., Yoshida, S., Scheinberg, P., and Ginsberg,
M. D. (1985) Cerebral norepinephrine depletion enhances recovery
after brain ischemia. Ann. Neurol. 18, 329-336.
(65) Rehncrona, S., Folbergrova´, J ., Smith, D. S., and Siesjo¨, B. (1980)
Influence of complete and pronounced incomplete cerebral is-
chemia and subsequent recirculation on cortical concentrations
of oxidized and reduced glutathione in the rat. J . Neurochem. 34,
477-486.
(66) Cooper, A. J . L., Pulsinelli, W. D., and Duffy, T. E. (1980)
Glutathione and ascorbate during ischemia and postischemic
reperfusion in rat brain. J . Neurochem. 35, 1242-1245.
(67) Slivka, A., and Cohen, G. (1993) Brain ischemia markedly elevates
levels of the neurotoxic amino acid, cysteine. Brain Res. 608, 33-
37.
(44) J ellinger, K. (1989) Pathology of Parkinson’s syndrome. In
Handbook of Experimental Pharmacology (Calne, D. B., Ed.) Vol.
8, pp 47-112, Springer-Verlag, Berlin.
(45) Riederer, P., Birkmayer, W., Seemann, D., and Wuketich, S.
(1977) Brain noradrenaline and 3-methoxy-4-hydroxyphenylglycol
in Parkinson’s syndrome. J . Neural Transm. 41, 245-251.
(68) Yang, C. S., Lin, N. N., Liu, L., Tsai, P. J ., and Kuo, J . S. (1995)
Lowered brain glutathione by diethylmaleate decreased the

760 Chem. Res. Toxicol., Vol. 13, No. 8, 2000
Xin et al.
glutamate release by cerebral ischemia in anesthetized rats. Brain
Res. 698, 237-240.
(69) Lee, J .-M., Zipfel, G. J ., and Choi, D. W. (1999) The changing
landscape of ischaemic brain injury mechanisms. Nature 399
(Suppl.), A7-A14.
(70) Yang, G. Y., Chan, P. H., Chen, J ., Carlson, E., Chen, S. F.,
Weinstein, P., Epstein, C. J ., and Kamii, H. (1994) Human
copper-zinc superoxide dismutase transgenic mice are highly
resistant to reperfusion injury after focal cerebral ischemia. Stroke
25, 165-170.
(71) Sims, N. R., and Pulsinelli, W. A. (1987) Altered mitochondrial
respiration in selectively vulnerable brain subregions following
transient forebrain ischemia in the rat. J . Neurochem. 49, 1367-
1374.
(72) Zaidan, E., and Sims, N. R. (1997) Reduced activity of the
pyruvate dehydrogenase complex but not cytochrome c oxidase
is associated with neuronal loss in the striatum following short-
term forebrain ischemia. Brain Res. 772, 23-28.
TX990170T