Important role of the 3-mercaptopropionamide moiety in glutathione: Promoting effect on decomposition of the adduct of glutathione with the oxoammonium ion of TEMPO

Article abstract of DOI:10.1021/jo050783c

Cyclic voltammetry of TEMPO in aqueous 0.1 M NaOH in the presence of glutathione (GSH) or cysteine (Cys) indicated the following points: (i) Both of the thiols rapidly formed adducts 3 with oxoammonium ion 1 anodically generated from TEMPO, (ii) 3 generat

Full text of DOI:10.1021/jo050783c

Important Role of the 3-Mercaptopropionamide Moiety in
Glutathione: Promoting Effect on Decomposition of the Adduct of
Glutathione with the Oxoammonium Ion of TEMPO
Hatsuo Maeda,* Hong-Yan Wu, Yuji Yamauchi, and Hidenobu Ohmori
Graduate School of Pharmaceutical Sciences, Osaka University,
Yamada-oka 1-6, Suita, Osaka 565-0871, Japan
h-maeda@phs.osaka-u.ac.jp
Received April 18, 2005
Cyclic voltammetry of TEMPO in aqueous 0.1 M NaOH in the presence of glutathione (GSH) or
cysteine (Cys) indicated the following points: (i) Both of the thiols rapidly formed adducts 3 with
oxoammonium ion 1 anodically generated from TEMPO. (ii) 3 generated from GSH entered a
succeeding reaction that generated N-oxide anion 2- (the reduced TEMPO). (iii) 3 produced from
Cys remained intact over the time scale of voltammetry. A structural feature of GSH was considered
to contribute to the observed behavior of this tripeptide. Possible structural features were evaluated
by screening various thiols on the basis of whether they provided GSH-like voltammetric results.
The 3-mercaptopropionamide group with an amide hydrogen in GSH was determined to be
responsible for the observed difference between GSH and Cys. The likely function is to transform
3 from GSH into a 5-imino-1,2-oxathiolane intermediate, thereby releasing 2-. Product analysis
for reactions of model thiols representing GSH and Cys with 1 provided support for this argument
and suggested that the reaction of GSH or Cys with 1 would produce the corresponding disulfides,
regardless of whether a five-membered ring intermediate was formed. The proposed function of
the 3-mercaptopropionamide moiety of GSH may provide useful insight for the molecular design
of exogenous thiol compounds as novel drugs for the treatment of GSH-depletion-related disorders.
Introduction
reactive nitrogen species (RNS) but also as a redox buffer
to maintain thiol-disulfide redox balances in intracel-
lular proteins.3 Recently, these functions have attracted
much attention from the field of molecular biology. This
is primarily because reactions with ROS or RNS trans-
form GSH into its activated form, which is likely to
induce S-glutathiolation between GSH and a thiol-
containing protein (Scheme 1). S-glutathiolation is thought
to alter the functional conformations of proteins, thereby
leading to the gene expression for cellular resistance
The tripeptide L-γ-glutamyl-L-cysteinylglycine (glu-
tathione or GSH) exists in mammalian cells at the
millimolar level. Although GSH is also present in its
oxidized form, glutathione disulfide (GSSG), the intra-
cellular ratio of GSH to GSSG is more than 100 in normal
cells.1 Thus, GSH acts as the most prevalent intracellular
thiol, exhibiting both strong nucleophilic and strong
reducing capabilities.2 Therefore, GSH functions not only
as a scavenger of reactive oxygen species (ROS) and
* To whom correspondence should be addressed. Phone and fax: +81
6 6879 8206.
(1) Deneke, S. M. Curr. Top. Cell. Regul. 2000, 36, 151-180.
(2) (a) Meister, A.; Anderson, M. E. Annu. Rev. Biochem. 1983, 52,
711-760. (b) Moran, L. K.; Gutteridge J. M. C.; Quinlan, G. J. Curr.
Med. Chem. 2001, 8, 763-772.
(3) (a) Gilbert H. F. Methods Enzymol. 1984, 107, 330-351. (b)
Gilbert H. F. Methods Enzymol. 1995, 251, 8-28. (c) Cotgreave, I. A.;
Gerdes, R. G. Biochem. Biophys. Res. Commun. 1998, 242, 1-9. (d)
Sies, H. Free Radical Biol. Med. 1999, 27, 916-921. (e) Åslund, F.;
Beckwith, J. Cell 1999, 96, 751-753. (f) Klatt, P.; Lamas, S. Eur. J.
Biochem. 2000, 267, 4928-4944.
10.1021/jo050783c CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/22/2005
8338
J. Org. Chem. 2005, 70, 8338-8343

Role of the 3-Mercaptopropionamide Moiety in GSH
SCHEME 1. A Proposed Mechanism for
S-Glutathiolation Induced by ROS and RNS
FIGURE 1. Cyclic voltammograms of (a) 1.0 mM TEMPO,
and 1.0 mM TEMPO in the presence of (b) 1.0 mM serine, (c,
d) 0.3 and 1.0 mM Cys, and (e, f) 0.3 and 1.0 mM GSH,
respectively, in 0.1 M aqueous NaOH.
toward oxidative stress and the modulation of signal-
transduction pathways such as tyrosine phosphorylation,
regulation of transcription, proteolytic processes, ubiq-
uitination, and degradation of proteins.3 When the in-
tracellular concentration of GSH falls below a critical
level, the S-glutathiolation-dependent and other func-
tions of GSH are partially lost. Further, GSH deficiency
has been demonstrated to have a correlation with patho-
logical conditions such as cancer, neurodegenerative
disorders, HIV, and aging.4 These observations have
prompted research to elucidate a molecular link of GSH
with GSH-dependent biological and pathological pro-
cesses. However, no studies have been done to address
why the thiol functionality of GSH has been naturally
selected to perform crucial in vivo functions. GSH is
thought to contain structural features that preferentially
allow this thiol to maintain cellular homeostasis. The
only available information on the structural features of
GSH is that it contains an exceptional γ-peptide linkage
between the glutamyl and cysteinyl moieties that aids
GSH in resisting intracellular aminopeptidases.3d Un-
derstanding the structural chemistry of GSH in connec-
tion with its chemical reactivity will help provide a
criterion for the molecular design of novel exogenous thiol
compounds with GSH-like biological acitivity. This in-
formation can then be used for the treatment of GSH-
depletion-related disorders.
A previous investigation to develop an electrochemical
detection-HPLC system with the recourse of TEMPO as
an electrochemical mediator for anodically inactive com-
pounds (such as alcohols and sugars)5 initiated the
application of this system toward the analysis of biologi-
cal thiols. To date, alkanethiols have been converted into
their corresponding disulfides by anodic oxidation with
TEMPO as an electrochemical mediator.6 However, no
voltammetric behavior of TEMPO in the presence of
thiols, such as GSH and Cys, in aqueous media has been
examined. In an effort to determine whether these thiols
can be detected using an electrochemical response due
to the catalytic oxidation of the thiols by a redox coupling
CHART 1
of TEMPO and its oxoammonium ion 1 (eq 1), cyclic
voltammetry for aqueous solutions of TEMPO in either
the absence or the presence of GSH or Cys was per-
formed. When aqueous 0.1 M NaOH was used as the
medium, GSH and Cys were determined to exhibit
uniquely different effects with respect to the voltammet-
ric response of TEMPO. The results implied that the
adduct of GSH and 1 smoothly decomposed, thereby
generating N-oxide anion 2- (eq 1) (the reduced TEMPO).
In contrast, the reaction of Cys with 1 formed a similar
adduct, which remained intact during the time scale of
voltammetry. This interesting observation was investi-
gated more thoroughly because the difference in the
chemical structures between GSH and Cys was thought
to be responsible for the voltammetric results. In this
paper we describe an important function of the 3-mer-
captopropionamide group, located on the cysteinylglycine
moiety of GSH, in promoting the adduct of GSH and 1
to decompose almost certainly into 2- and GSSG. The
proposed mechanism provides a reasonable explanation
for the recent report that N-acetylcysteinamide scavenges
peroxides more effectively than N-acetylcysteine itself.7
Results and Discussion
Figure 1 compares cyclic voltammograms of TEMPO
(1.0 mM) in aqueous 0.1 M NaOH either in the absence
or in the presence of L-serine, Cys (Chart 1), or GSH.
TEMPO itself exhibited a redox wave with anodic and
cathodic peaks at 530 and 460 mV vs SCE, respectively
(Figure 1a). The addition of serine brought about an
increase of the anodic peak and the disappearance of the
cathodic peak on the redox wave of TEMPO (Figure 1b).
These voltammetric results clearly indicate that serine
undergoes catalytic oxidation by the redox cycle of
TEMPO. This finding is in good agreement with known
(4) (a) Dro¨g, W. Exp. Gerontol. 2002, 37, 1331-1343. (b) Townsend,
D. M.; Tew, K. D.; Tapiero H. Biomed. Pharmacother. 2003, 57, 145-
155. (c) Pastore, A.; Federici, G.; Bertini, E.; Piemonte, F. Clin. Chim.
Acta 2003, 333, 19-39.
(5) (a) Yamauchi, Y.; Maeda, H.; Ohmori, H. Chem. Pharm. Bull.
1996, 44, 1021-1025. (b) Yamauchi, Y.; Maeda, H.; Ohmori, H. Chem.
Pharm. Bull. 1997, 45, 2024-2028.
(6) Kashiwagi, Y.; Ohsawa, A.; Osa, T.; Ma, Z.; Bobbitt, J. M. Chem.
Lett. 1991, 581-584.
(7) Grinberg, L.; Fibach, E.; Amer, J.; Atlas, D. Free Radical Biol.
Med. 2005, 38, 136-145.
J. Org. Chem, Vol. 70, No. 21, 2005 8339

Maeda et al.
FIGURE 2. Changes in the peak current for the wave at 350
mV in the voltammograms of 1.0 mM TEMPO plotted against
the concentration of (closed circles) GSH or (open circles) Cys.
results for primary and secondary alcohols.8 The thiol
counterpart of serine, Cys, produced a different effect on
the voltammetric response of TEMPO. This conflicted
with the expectation that Cys could be oxidized catalyti-
cally by the redox cycle of TEMPO. Cys itself exhibited
no voltammetric waves between 0 and 0.7 V. When 1.0
mM TEMPO in the presence of 0.3 mM Cys was con-
ducted, a new anodic peak emerged at 350 mV, with a
contracted redox wave (Figure 1c). The new peak became
larger and the redox wave for TEMPO disappeared when
the concentration of Cys was increased to 1.0 mM (Figure
1d). These results suggested that Cys entered a stoichio-
metric reaction with TEMPO or its oxidized form 1. As
with Cys, GSH yielded no voltammetric waves in the
examined potential window and did not interfere with
observing the 350 mV anodic peak. However, the anodic
wave of TEMPO always remained fairly intact, regard-
less of the concentration of GSH added to the TEMPO
solution (Figure 1e,f). In contrast to Cys, GSH likely
entered a catalytic process with a redox cycle of TEMPO.
To distinguish the electrode processes of TEMPO in the
presence of these thiols, a relationship between the
concentration of added Cys or GSH and the peak current
of the new anodic wave at 350 mV in the voltammetry of
TEMPO was established (Figure 2). When the concentra-
tion of Cys was greater than or equal to TEMPO, the
peak height became almost constant. This observation
was in contrast to that for GSH. The anodic peak at 350
mV became larger as the concentration of GSH increased,
and continued to increase even when GSH was added at
higher concentrations than TEMPO.
The following two pathways could be envisioned for the
resulting peak at 350 mV, which was observed in the
presence of the thiols GSH and Cys (Scheme 2). (Route
1) A chemical reaction followed by an electrode reaction:
Both Cys and GSH are likely to exist as the correspond-
ing thiolate anion in aqueous 0.1 M NaOH. These thiolate
anions would be oxidized by TEMPO, yielding the cor-
responding thiyl radical and the reduced TEMPO 2- (eq
1), and anodic oxidation of either product might be
responsible for the peak at 350 mV. (Route 2) An
electrode reaction followed by a fast chemical reaction:
The oxoammonium 1 (eq 1), generated through anodic
oxidation of TEMPO, would be subjected to a markedly
fast reaction with the thiolate anion of Cys or GSH,
FIGURE 3. Absorption spectra observed for CH3CN-H2O
(5:95) solutions of (a) 4 mM TEMPO, and 4 mM 1 (b) before
and (c) immediately after addition of 8 mM GSH in aqueous
24 mM NaOH.
SCHEME 2. Qualitative Description for the
Observed Voltammetric Results of TEMPO in the
Presence of Cys or GSH
leading to a shift of the anodic peak of TEMPO to the
negative direction.9 In an effort to examine which route
causes the thiol-induced voltammetric peak at 350 mV,
ESR spectra of TEMPO and absorption spectra of 1 as
well as TEMPO were measured either in the absence or
in the presence of Cys or GSH. No changes in the ESR
signal observed for 10 µM TEMPO in aqueous 0.1 M
NaOH were induced upon addition of 40 µM Cys or GSH
(data not shown). TEMPO itself exhibited a well-defined
peak at 424 nm (Figure 3a). Essentially the same
spectrum was observed for a mixture of TEMPO and the
thiolate anion of Cys or GSH (data not shown). These
results negated the possibility of a direct chemical
reaction between TEMPO and the thiolate anions of Cys
and GSH, that is, route 1. In the spectra of 1 electro-
chemically generated from TEMPO, a poorly defined peak
around 464 nm was observed (Figure 3b). This peak due
to 1 disappeared totally and immediately after addition
of Cys or GSH in aqueous NaOH (Figure 3c). Similar
phenomena were recognized even upon reaction of 1 with
Cys or GSH in the absence of NaOH (data not shown). It
is worth commenting that no peak due to TEMPO was
observed after reaction of 1 with Cys or GSH. These
observations suggested that Cys or GSH rapidly enters
reaction with the oxoammonium 1, which accompanies
no regeneration of TEMPO. Thus, route 2 is thought to
be responsible for the 350 mV peak; the anodic oxidation
(9) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Funda-
mentals and Applications; John Wiley & Sons: New York, 1980; pp
429-487.
(8) Adam, W.; Saha-Mo¨ller, C. R.; Ganeshpure, P. A. Chem. Rev.
2001, 101, 3499-3548.
8340 J. Org. Chem., Vol. 70, No. 21, 2005

Role of the 3-Mercaptopropionamide Moiety in GSH
CHART 2
FIGURE 4. (A) Cyclic voltammograms of 1.0 mM TEMPO in
the presence of (a) 1.0 mM Glu-Cys or (b) 1.0 mM Cys-Gly in
0.1 M aqueous NaOH and (B) changes in the peak current of
the wave at 350 mV in the voltammograms of 1.0 mM TEMPO
plotted against the concentration of (open circles) Glu-Cys or
(closed circles) Cys-Gly.
SCHEME 3. Proposed Mechanism for the
Succeeding Reaction of Adduct 3
of TEMPO is observed as a prewave in the voltammo-
grams, due to the significantly higher reactivity of the
sulfhydryl groups of Cys and GSH toward 1 compared
with the hydroxyl group in serine. It is well-known that
the oxidation of alcohols with 1 is initiated by the attack
of a hydroxyl oxygen atom on the nitrogen atom of 1.8,10
A similar initiation step would be included in the present
reaction of thiols with 1. The common effects of Cys and
GSH on the voltammetry of TEMPO and spectrophoto-
metry of 1 are consistent with this adduct formation
between 1 and Cys or GSH. Therefore, the way in which
Cys and GSH react with 1 is likely linked to the
decomposition rates of their adducts 3 (Scheme 2). The
adduct of GSH and 1 will smoothly decompose into 2-
and an intermediate or product, whereas 3 from Cys and
1 will be stable and not susceptible to further transfor-
mations on the voltammetric time scale. Anion 2- will
be oxidized to 1 at a potential similar to that of TEMPO,
allowing the turnover of the catalytic cycle. This is the
reason the stoichiometry of the reaction of Cys with 1
was 1:1, whereas 1 reacted catalytically with GSH.
The susceptibility of 3 from GSH to a succeeding
reaction was expected to correlate with the structural
features of GSH. In an attempt to determine a structural
moiety in GSH, as a factor in promoting the catalytic
reaction with 1, cyclic voltammetry of TEMPO was
conducted in the presence of various thiols with partial
structures of GSH (Chart 2). The voltammetric evalua-
tion was based on whether they provided GSH-like
voltammetric results (GSH-like effect). The first screen-
ing of a key structural feature of GSH was performed
with L-γ-glutamyl-L-cysteine (Glu-Cys), L-cysteinylglycine
(Cys-Gly), and N-acetyl-L-cysteine (NAC). As shown in
Figure 4, Glu-Cys exerted a Cys-like effect on the
voltammetric response of TEMPO, whereas the addition
of Cys-Gly produced results essentially identical to those
of GSH. In addition, NAC induced the same changes as
Cys as far as the voltammetric behavior of TEMPO.
These results suggested that the key partial structure
for the observed reactivity inherent in GSH is included
in Cys-Gly. The deamino derivative of Cys-Gly, 4a11
(Chart 2), exhibited a GSH-like effect on the voltammo-
grams of TEMPO. Its decarboxy derivative 4b also
induced essentially the same voltammetric results as
GSH. However, Cys-like voltammetric results were ob-
served in the presence of 4c bearing no amide hydrogen.
These results suggested that a 3-mercaptopropionamide
group with an amide hydrogen is the key chemical
structure for the observed voltammetric behavior of
TEMPO in the presence of GSH.
The observed importance of the amide hydrogen in
GSH, 4a, and 4b suggested that deprotonation of the
amide hydrogen would be included in the decomposition
process for the adducts generated from 1 and these thiols.
On the basis of this speculation, a plausible two-pathway
succeeding reaction of 3 from GSH, 4a, or 4b was
proposed in Scheme 3. One pathway yields the 5-imino-
1,2-oxathiolane intermediate 5 via attack of the amide
oxygen atom on the sulfur atom (path A). The other
proceeds via bond formation between the sulfur and the
amide nitrogen atom, providing the 1,2-thiazolidin-3-one
intermediate 6 (path B). Both intermediates are expected
to react with another molecule’s thiol group, producing
the corresponding disulfide 7. Formation of five-mem-
(10) (a) Semmelhack, M. F.; Schmid, C. R.; Corte´s, D. A. Tetrahedron
Lett. 1986, 27, 1119-122. (b) de Nooy, A. E. J.; Besemer, A. C.
Tetrahedron 1995, 51, 8023-8032.
(11) Condon, M. E.; Petrillo, Jr. E. W.; Ryono, D. E.; Reid, J. A.;
Neubeck, R.; Puar, M.; Heikes, J. E.; Sabo, E. F.; Losee, K. A.;
Cushman, D. W.; Ondetti, M. A. J. Med. Chem. 1982, 25, 250-258.
J. Org. Chem, Vol. 70, No. 21, 2005 8341

Maeda et al.
SCHEME 4. Model Thiols for Elucidating Which
Is a Plausible Intermediate, 5 or 6
SCHEME 5. Product Analysis for Reaction of 1
with 4g or 4h as a Model Compound of GSH or
Cys, Respectively
SCHEME 6. Proposed Mechanism for the Reported
Difference in Scavenging Ability of Peroxides
between AD4 and NAC
bered ring intermediates is accompanied by the release
of 2- (the reduced TEMPO), and this anion is oxidized
to 1 via TEMPO, allowing the turnover of the TEMPO
redox cycle.
The sulfhydryl and amide groups at positions that
permit the closure of a five-membered ring such as 5 or
6 is important. Compounds 4d12 and 4e13 (Scheme 4) are
able to form four- or six-membered ring analogues such
as 8 and 9 if the corresponding adducts 3 from these
thiols behave similarly to GSH. However, 4d and 4e
exerted Cys-like effects. The formation of four- and six-
membered ring intermediates from adduct 3 appeared
to be much less favored than the production of 5 or 6, as
is generally observed in ring-closure reactions. The
voltammetric results on 4f (Scheme 4) were informative
to regard 5 rather than 6 as a plausible intermediate.
The corresponding adduct of 1 and 4f might undergo ring
formation by attack of the amide nitrogen atom on the
sulfur atom, yielding the five-membered ring intermedi-
ate 10. However, Cys-like results were obtained in cyclic
voltammetry of TEMPO in the presence of 4f, suggesting
that a five-membered intermediate such as 6 would not
be formed through the attack of an amide nitrogen atom
in 3. Thus, it was concluded that the voltammetric
behavior of TEMPO, in the presence of GSH, originates
from the formation of 5, as an activated form of GSH
toward attack of another molecule of GSH, via path A
accompanied by release of 2-.
The formation of a disulfide compound as the final
product in the reaction of 1 with a thiol was confirmed
by the following experiments. The oxoammonium 1 was
prepared as a 50% aqueous CH3CN solution by constant-
current electrolysis of TEMPO in a divided cell at room
temperature, in which 1 F/mol of electricity was passed.
The obtained solution was added to a solution of 4g or
4h14 (Scheme 5) in aqueous NaOH. The resulting mixture
was stirred at room temperature for 1 h, and product
analysis was performed on the reaction mixture. Com-
pounds 4g and 4h were used as model thiols because it
was easy to isolate the products. Cyclic voltammetry
experiments revealed that 4g and 4h represented GSH
and Cys, respectively. The reaction of 1 with both of these
thiols provided the corresponding disulfides 7g and 7h.
Thus, it was demonstrated that thiols react with 1 to
provide the corresponding disulfides, regardless of their
classification (GSH- or Cys-like) with respect to their
effects on the voltammetric response of TEMPO. Al-
though it was reported that various types of 1,2-thiazo-
lidin-3-one compounds 6 could be prepared and isolated,15
no formation of the corresponding 6 from 4g was noted.
Further, this supports path A via 5 as the route for
adduct 3 from GSH. It is important to note that the
isolated yield of 7g was higher than that of 7h. These
results implied that disulfide formation from a thiol by
a reaction with 1 proceeds more effectively when a thiol
has an amide group that permits formation of 5 from the
corresponding 3. Adduct 3 from a thiol with Cys-like
reactivity was thought to react directly with another thiol
molecule on the preparative time scale, producing the
corresponding disulfide.
The proposed mechanism for the reaction of GSH and
GSH-like thiols with 1 generated from TEMPO suggests
that 5 is the activated form of GSH in biological reactions
induced by ROS or RNS. Moreover, 3-mercaptopropiona-
mides with amide hydrogens are potent candidates as
exogenous thiol compounds which can be used for the
treatment of GSH-depletion-related disorders. Recently,
it was reported that N-acetylcysteinamide (AD4 in Scheme
6) reduced H2O2 and t-BuOOH much more effectively
than NAC and was stronger as a peroxide scavenger by
(12) Delamarche, E.; Hoole, A. C. F.; Michel, B.; Wilkes, S.; Despont,
M.; Welland, M. E.; Biebuyck, H. J. Phys. Chem. B 1997, 101, 9263-
9269.
(13) Paulini, R.; Frankamp, B. L.; Rotello, V. M. Langmuir 2002,
18, 2368-2373.
(14) Duan, L.; Garrett, S. J. Langmuir 2001, 17, 2986-2994.
(15) Takahaki, S.; Inaba, T.; Matsunaga, A.; Sakurasawa, M.;
Ichikawa, S. Japanese Patent Appl. JP 2002-173079, 2002; Chem.
Abstr. 2003, 139, 44173.
8342 J. Org. Chem., Vol. 70, No. 21, 2005

Role of the 3-Mercaptopropionamide Moiety in GSH
a factor of about 2 than GSH.7 These observations can
be explained by the effective formation of an oxathiolane
intermediate such as 5i from AD4. Provided that the
reaction of a thiol with a peroxide is reversible, the
reducing ability of the thiol will depend on the rate of
transformation of a sulfenic acid, such as 11 or 12, to
the corresponding disulfide. The amide group of AD4 is
thought to promote disulfide formation via 5i, while NAC
lacks a promoting functional group. NAC exerted the
same influence on the voltammograms of TEMPO as Cys;
therefore, AD4 is a better reducing agent of peroxides
than NAC. The observed increase in reactivity of AD4
with respect to GSH toward peroxides is related to the
presence of two amide hydrogen atoms, which increases
the likelihood of forming a cyclic intermediate (5).
Experimental Section
Materials. Milli-Q water and HPLC-grade CH3CN and
MeOH were used. All other chemicals were purchased from
commercial suppliers and used without further purification.
General Procedure for Cyclic Voltammetry. A three-
electrode configuration was employed: a glassy carbon (L )
3 mm) electrode as a working electrode, a saturated calomel
electrode (SCE) as a reference electrode, and a platinum wire
electrode as a counter electrode. The glassy carbon electrode
was polished with polishing paper (no. 1500) and alumina
powder (0.05 µM) on a polishing cloth, sonicated in water,
MeOH, and water, and dried with a stream of N2 gas prior to
use. Test solutions were deoxygenated by bubbling N2 gas for
several minutes and subjected to voltammetric measurements
at room temperature and 100 mV/s over a potential range
between 0.0 and 0.7 V vs SCE under a N2 atmosphere.
General Procedure for Spectrophotometry of 1. TEMPO
(156 mg, 1.0 mmol) was subjected to constant-current elec-
trolysis (1.5 mA, 1 F/mol) in 25 mL of CH3CN-H2O (1:1)
containing 0.1 M NaClO4 at room temperature under a N2
atmosphere in a divided cell equipped with a glassy carbon
anode (15 × 30 mm) and a platinum foil cathode (12 × 20 mm).
After the electrolysis, the analyte was used as a stock solution
(40 mM) of 1. Absorption spectra of 1 itself were obtained after
the stock solution (1 mL) was diluted by a factor of 10 with
H2O. Effects of Cys or GSH on the absorption spectra of 1 were
evaluated as follows. A thiol solution (40 mM, 2 mL) in H2O,
NaOH (1 M, 240 µL) in H2O, and the stock solution (1 mL) of
1 were added in this order to the appropriate volume of H2O.
The total volume of the mixture was adjusted to 10 mL by
addition of H2O, and the thus obtained solution was im-
mediately subjected to spectrophotometry.
Conclusions
The biological thiols GSH and Cys produce different
effects on the voltammetric behavior of TEMPO in
aqueous NaOH. The characteristic effect of GSH is that
it allows the adduct 3 of GSH and 1 (the oxidized
TEMPO) to form an initial intermediate which readily
decomposes, thereby generating 2- (the reduced TEMPO)
and allowing the turnover of the TEMPO redox cycles
via anodic oxidation of this anion. Examination of various
thiol compounds based on whether a GSH-like voltam-
metric result was provided, revealed that the 3-mercap-
topropionamide group with an amide hydrogen is essen-
tial for GSH to exhibit its inherent reactivity toward 1.
The proposed function of this structure is to form the
oxathiolane 5 from 3 of GSH and 1, resulting in an
activated form of GSH. This, in turn, reacts with another
molecule of GSH, producing GSSG. This mechanism
provides a reasonable explanation for the observation
that AD4 works as a stronger scavenger for peroxides
than NAC and GSH. Thus, the present investigation
provides acceptable criteria for the molecular design of
exogenous thiol compounds as novel drugs for the treat-
ment of GSH-depletion-related disorders.
Acknowledgment. This work was supported in part
by a Grant-in-Aid for the Encouragement of Young
Scientists (A) (12771378) from the Ministry of Educa-
tion, Culture, Sports, Science, and Technology of Japan.
Supporting Information Available: Synthesis and char-
acterization of 4a-h and experimental procedures for product
analysis on reaction of 1 with 4g or 4h. This material is
available free of charge via the Internet at http://pubs.acs.org.
JO050783C
J. Org. Chem, Vol. 70, No. 21, 2005 8343