Journal of Neuroscience Research 79:700 –706 (2005)
-Glutamylcysteine Ethyl Ester-Induced
Up-Regulation of Glutathione Protects
Neurons Against A (1– 42)-Mediated
Oxidative Stress and Neurotoxicity:
Implications for Alzheimer’s Disease
Debra Boyd-Kimball,1 Rukhsana Sultana,1 Haﬁz Mohmmad Abdul,1 and
D. Allan Butterﬁeld1,2*
Department of Chemistry, Center for Membrane Sciences, University of Kentucky, Lexington, Kentucky
Sanders-Brown Center on Aging, University of Kentucky, Lexington, Kentucky
Glutathione (GSH) is an important endogenous antioxi- ﬁeld et al., 2001, 2002a). Consequently, A (1– 42) has
dant found in millimolar concentrations in the brain. GSH been implicated as a causative agent in AD (Varadarajan et
levels have been shown to decrease with aging. Alzhei- al., 2000; Butteﬁeld, 2002, 2003). The lipid-soluble anti-
mer’s disease (AD) is a neurodegenerative disorder as- oxidant vitamin E has been shown to inhibit the oxidative
sociated with aging and oxidative stress. A (1– 42) has damage induced by A (1– 42), suggesting that reactive
been shown to induce oxidative stress and has been oxygen species (ROS) play a role (Yatin et al., 2000).
proposed to play a central role in the oxidative damage Glutathione (GSH) is a tripeptide ( -gluta-
detected in AD brain. It has been shown that administra- mylcysteinylglycine) found in intracellular concentrations
tion of -glutamylcysteine ethyl ester (GCEE) increases of 1–3 mM in the brain (Cooper, 1997). GSH is located in
cellular levels of GSH, circumventing the regulation of both the cytosol and the mitochondria within cells and acts
GSH biosynthesis by providing the limiting substrate. In as a vital endogenous antioxidant to combat oxidative
this study, we evaluated the protective role of up- stress. Cysteine is the limiting amino acid in GSH biosyn-
regulation of GSH by GCEE against the oxidative and thesis, and, as a result, -glutamylcysteine is the limiting
neurotoxic effects of A (1– 42) in primary neuronal cul- substrate for GSH synthesis. -Glutamylcysteine syn-
ture. Addition of GCEE to neurons led to an elevated thetase ( -GCS), the enzyme that catalyzes the formation
mean cellular GSH level compared with untreated con- of the dipeptide -glutamylcysteine, is the rate-limiting
trol. Inhibition of -glutamylcysteine synthetase by buthi- enzyme in GSH synthesis, and this enzyme is feedback
onine sulfoximine (BSO) led to a 98% decrease in total inhibited by GSH itself (Anderson and Luo, 1998). It has
cellular GSH compared with control, which was returned been proposed that administration of the rate-limiting
to control levels by addition of GCEE. Taken together, substrate, -glutamylcysteine, will circumvent GSH-
these results suggest that GCEE up-regulates cellular mediated feedback inhibition in GSH biosynthesis (Drake
GSH levels which, in turn, protects neurons against pro- et al., 2002), because this dipeptide is a substrate for GSH
tein oxidation, loss of mitochondrial function, and DNA synthase (Cooper, 1997). Moreover, modiﬁcation of the
fragmentation induced by A (1– 42). These results are substrate by esteriﬁcation [ -glutamylcysteine ethyl ester
consistent with the notion that up-regulation of GSH by (GCEE)] has been shown to facilitate transport of the
GCEE may play a viable protective role in the oxidative compound across the plasma membrane, where it is dees-
and neurotoxicity induced by A (1– 42) in AD brain. teriﬁed and can be acted upon by GSH synthetase to
© 2005 Wiley-Liss, Inc.
catalyze the formation of GSH (Anderson et al., 1985;
Anderson and Meister, 1989). Such up-regulation of GSH
Key words: Alzheimer’s disease; glutathione; amyloid
*Correspondence to: Prof. D. Allan Butterﬁeld, Department of Chemistry,
Center for Membrane Sciences, and Sanders-Brown Center on Aging, 121
A (1– 42) has been shown to induce oxidative stress Chemistry-Physics Building, University of Kentucky, Lexington, KY
both in vitro and in vivo (Yatin et al., 1999a; Butterﬁeld 40506-0055. E-mail: email@example.com
and Lauderback, 2002; Drake et al., 2003a). Oxidative Received 18 July 2004; Revised 3 November 2004; Accepted 8 November
stress is extensive in AD, a neurodegenerative disease 2004
associated with cognitive decline and aging (Subbarao et Published online 27 January 2005 in Wiley InterScience (www.
al., 1990; Hensley et al., 1995; Markesbery, 1997; Butter- interscience.wiley.com). DOI: 10.1002/jnr.20394
© 2005 Wiley-Liss, Inc.
GCEE Protects Against A (1– 42) Oxidative Stress 701
has been shown to protect both synaptosomes and mito- was aliquoted per well to establish a standard curve ranging from
chondria against peroxynitrite-mediated oxidative stress 0 to 16.0 M GSH. Fifty microliters of sample was added per
(Drake et al., 2002, 2003b). well. One hundred ﬁfty microliters of Assay Cocktail [consisting
A (1– 42) has been shown to deplete GSH levels in of 2-(N-morpholino)ethanesulfonic acid (MES) buffer, cofactor
astrocytes (Abramov et al., 2003). Such glial cells play a mixture (NADP and glucose-6-phosphate), enzyme mixture
major role in supplying the metabolic substrates and pre- (GSH reductase, and glucose-6-phosphate dehydrogenase] and
cursors of GSH to neurons. Consequently, A (1– 42) may 5,5 -dithiobis-2-nitrobenzoic acid (DTNB) were added to each
lead to depletion of GSH as an available antioxidant in well, and the absorbance was followed at 405 nm for 30 min. All
neurons. Additionally, GSH levels decrease with age, leav- measurements were made in triplicate. The average absorbance
ing neurons vulnerable to oxidative damage initiated by at 25 min was calculated for each standard and sample. A plot of
A (1– 42) (Liu and Choi, 2000). Up-regulation of GSH the corrected absorbance vs. the concentration of the GSH
could play a therapeutic role in AD (Butterﬁeld et al., standards (in M) was utilized to calculate the average concen-
2002b). tration of GSH present in the samples.
In this study, we evaluated the role of GCEE up-
regulation of GSH in neurons as a protective therapeutic Protein Carbonyls
mechanism against A (1– 42)-mediated oxidative stress Samples (5 l) were incubated for 20 min at room tem-
and neurotoxicity. We show that GCEE up-regulation of perature with 5 l of 12% sodium dodecyl sulfate (SDS) and
GSH in neurons protects against A (1– 42)-induced pro- 10 l of 2,4-dinitrophenylhydrazine (DNPH) that was diluted
tein oxidation, loss of mitochondria function, and DNA 10 times with water from a 200 mM stock. The samples were
fragmentation. neutralized with 7.5 l of neutralization solution (2 M Tris in
30% glycerol). The resulting sample (250 ng) was loaded per
MATERIALS AND METHODS well in the slot-blot apparatus. Samples were loaded onto a
Chemicals nitrocellulose membrane under vacuum pressure. The mem-
brane was blocked with 3% bovine serum albumin (BSA) in
All chemicals were of the highest purity and were ob- phosphate-buffered saline (PBS) containing 0.01% (w/v) so-
tained from Sigma (St. Louis, MO) unless otherwise noted. dium azide and 0.2% (v/v) Tween 20 (wash blot) for 1 hr and
A (1– 42) was purchased from Anaspec (San Jose, CA), with incubated with a 1:100 dilution of anti-DNP polyclonal anti-
HPLC and MS veriﬁcation of purity. The peptide was stored in body in wash blot for 1 hr. After completion of the primary
the dry state at –20°C until use. The OxyBlot protein oxidation antibody incubation, the membranes were washed three times in
detection kit was purchased from Chemicon International (Te- wash blot for 5 min each. An anti-rabbit IgG alkaline phospha-
mecula, CA). Anti-4-hydroxynonenal was purchased from Al- tase secondary antibody was diluted 1:8,000 in wash blot and
pha Diagnostic International (San Antonio, TX). added to the membrane for 1 hr. The membrane was washed in
wash blot three times for 5 min each and developed using
Cell Culture Experiments
Sigmafast Tablets (BCIP/NBT substrate). Blots were dried,
Neuronal cultures were prepared from 18-day-old scanned with Adobe Photoshop (San Jose, CA), and quantitated
Sprauge-Dawley rat fetuses (Yatin et al., 1999b). A peptides with Scion Image.
were dissolved in sterile water that had been stirred over
Chelex-100 resin. The peptides were preincubated for 24 hr at Electron Microscopy
37°C prior to addition to cultures. The ﬁnal concentration of Electron microscopy was used to assess the ability of the
the peptides in the cell culture was 10 M, and the effects of A A peptides to form ﬁbrils upon incubation in solution for
on the neurons were measured after 24 hr of exposure. GCEE 24 hr. Aliquots of 5 l of the peptide solutions that were used
was added to cell cultures 1 hr prior to addition of A peptide. for the cell culture experiments were placed on a copper mesh
Mitochondrial function was evaluated by the 3-[4,5- formvar carbon-coated grid. After 1–1.5 min of incubation at
dimethylthiazol-2-yl)-2,5-diphenyl] tetrazolium bromide (MTT) room temperature, excess liquid was drawn off, and samples
reduction assay. Brieﬂy, MTT was added to each well at a ﬁnal were counterstained with 2% uranyl acetate. Air-dried samples
concentration of 1.0 mg/ml and then incubated for 1 hr. The were examined in a Philips Tecnai Biotwin 12 transmission
dark blue formazan crystals formed in intact cells were extracted electron microscope (FEI, Eindhoven, The Netherlands) at
with 250 l dimethyl sulfoxide (DMSO), and the absorbance 80 kV. Images were captured with a 2K 2K digital camera
was read at 595 nm with a microtiter plate reader (Bio-Tek (Advanced Microscopy Techniques).
Instruments, Winooski, VT).
Analysis of DNA Fragmentation
GSH Assay Cell death was measured by Hoescht 332584 (1 g/ml)
Endogenous GSH was measured with a Glutathione Assay followed by propidium iodide (PI; 5 g/ml) staining and de-
Kit (Cayman Chemical, Ann Arbor, MI). Brieﬂy, cells were tected by ﬂuorescence microscopy (Darzynkiewicz et al., 1994).
deproteinated by treatment with 10% (w/v) metaphosphoric Cortical neuronal cells were treated with A (10 M) alone for
acid (Aldrich, Milwakee, WI) and centrifuged at 2,000g for 24 hr or pretreated with GCEE (for 1 hr) followed by A
2 min. The supernatant was collected and treated with 4 M (10 M) and incubated for 24 hr. Cultures were rinsed three
triethanolamine. Standards of oxidized GSH (GSSG) were pre- times in PBS, ﬁxed with 4% paraformaldehyde for 10 min at
pared varying from 0 to 8.0 M. Fifty microliters of standard 37°C, rinsed, and stained with Hoechst 332584 or PI for 10 min
702 Boyd-Kimball et al.
Fig. 2. Neurotoxicity A (1– 42) and protection by GCEE as measured
by MTT reduction. The results are shown as mean SEM of three
independent measurements. Relative to untreated neuons, treatment of
Fig. 1. Protein oxidation as indexed by protein carbonyls. Relative to neurons with 10 M A (1– 42) showed a signiﬁcant decrease in MTT
untreated control, increased protein carbonyls were observed for 24 hr reduction. Likewise, 1 hr of pretreatment of cell cultures with 500 M
of treatment of neuronal cell cultures with 10 M A (1– 42). Lower or 750 M GCEE resulted in a signiﬁcant decrease in A (1– 42)-
levels of protein carbonyls were observed with 10 M A (1– 42) and mediated MTT reduction. Such changes were not observed for 1 hr of
pretreatment with 500 M GCEE or 10 M A (1– 42) and pretreat- pretreatment with 1 mM GCEE, followed by 10 M A (1– 42). *P
ment with 750 M GCEE. Complete inhibition of A (1– 42)-induced 0.05, **P 0.0001; n 3. The statistical comparison was performed
protein oxidation was found with treatment of cultures with 1 mM between control and A treatment data sets.
GCEE 1 hr prior to addition of 10 M A (1– 42). Results are given as
the mean SEM of four independent treatments. *P 0.003, **P
0.001; n 4. The statistical comparison was performed between
untreated control and A treatment sets.
at room temperature. Images were obtained sequentially with
Hoechst 332584, then PI. Random areas (approximately seven)
in the cell culture dishes were selected, and the number of
apoptotic cells was counted in neurons from untreated control,
treated with A (1– 42) alone, or pretreated with GCEE fol-
lowed by A (10 M) groups. The average of the apoptotic cells
in each of the respective groups was determined as a percentage
of the untreated control.
ANOVA followed by Student’s t-test was used to deter-
mine statistical signiﬁcance. P 0.05 was considered signiﬁcant.
GCEE Inhibited A (1– 42)-Induced Protein
Oxidation in a Concentration-Dependent Manner
A 60% increase in protein oxidation in neurons was Fig. 3. GSH assay. Cell cultures were pretreated with 500 M buthi-
induced following 24 hr of treatment with A (1– 42) onine sulfoximine (BSO) for 6 hr to inhibit -glutamylcysteine syn-
(Fig. 1). Pretreatment of neurons with GCEE was shown thetase, and 1 mM GCEE was then added. Cells were collected after
24 hr of treatment with GCEE and assayed for total GSH. The results
toinhibitA (1– 42)-mediatedproteinoxidationinaconcen- are shown as mean SEM of four independent measurements. Treat-
tration-dependent manner. Protein oxidation was signiﬁ- ment of neurons with BSO led to a signiﬁcant loss of GSH. No
cantly decreased by 1 hr of pretreatment of neuronal signiﬁcant difference was noted in GSH levels between control,
cultures with 750 M GCEE; however, protein carbonyl GCEE-, and BSO- plus GCEE-treated neurons. *P 0.0005; n 4.
values reached control levels following pretreatment with The statistical comparison was performed between control and each
1 mM GCEE for 1 hr prior to the addition of A (1– 42). treatment data set.
GCEE Protects Against A (1– 42) Oxidative Stress 703
Fig. 4. Phase-contrast provides morphological insight. A: Control cells show extensive intact neu-
ronal network and cell bodies. B: Neuronal culture treated with 10 M A (1– 42) shows atrophied
neurons and loss of neuronal connections. C: Neuronal culture treated with 1 mM GCEE 1 hr prior
to treatment with 10 M A (1– 42) shows intact neuronal connections.
This 1 mM level is well within the level of GSH in the the deesteriﬁed dipeptide serve as a substrate for GSH
brain (Cooper, 1997). synthase, in that de novo synthesis of GSH via
-glutamylcysteine synthetase was not possible in the pres-
GCEE Protects Mitochondria From A (1– 42)- ence of BSO.
Mediated Oxidative Stress in a Concentration-
Dependent Manner GCEE Protects Against A (1– 42) Loss of
A (1– 42) was shown to decrease MTT reduction by Neuronal Network and Neuronal Death
45% compared with untreated control (Fig. 2). Pretreat- Phase-contrast microscopy was used to examine
ment of neurons with 500 M GCEE for 1 hr followed by the morphological changes in the neurons following
A (1– 42) did not show any change in MTT reduction treatment (Fig. 4). Neurons that were exposed to
compared with A (1– 42) alone. Pretreatment of neurons 10 M A (1– 42) for 24 hr (Fig. 4B) showed loss of
with 750 M GCEE and A (1– 42) did show some pro- neuronal network and vacuoles in the perikarya, indi-
tection of mitochondrial function compared with that cating dying cells. Conversely, neurons pretreated with
induced by A (1– 42) alone but still showed a signiﬁcant 1 mM GCEE 1 hr prior to addition of A (1– 42) (Fig.
decrease in MTT reduction compared with untreated 4C) showed intact networks and cell bodies similar to
control. However, addition of 1 mM GCEE 1 hr prior to those of control neurons (Fig. 4A). The apoptotic (later
the addition of A (1– 42) showed no signiﬁcant loss in stage) and necrotic cells are positive for Hoechst stain-
MTT reduction compared with control, which is consis- ing and PI staining, whereas early apoptotic stages are
tent with the concentration of GCEE that protected neu- negative for PI staining. We used Hoechst staining and
rons against A (1– 42)-mediated protein oxidation (Fig.
1). Because 1 mM GCEE protected neurons against pro- PI for DNA fragmentation to distinguish between late-
tein oxidation and prevented loss of mitochondrial func- stage apoptosis and necrosis vs. early-stage apoptosis
tion, this concentration was used for all subsequent exper- (Fig. 5). Neurons treated with A (1– 42) (Fig. 5B)
iments. showed extensive DNA fragmentation by both stains,
from which we conclude that late apoptotic and ne-
GCEE Increases Endogenous GSH Levels crotic cells are found under the conditions of this ex-
To determine whether GCEE protected neuons by periment. In contrast, pretreatment of neurons with
acting as a GSH mimetic or by serving as a substrate for 1 mM GCEE followed by addition of A (1– 42) (Fig.
GSH synthase for the up-regulation of GSH, an assay of 5C) resulted in a signiﬁcant reduction in DNA frag-
total GSH was conducted (Fig. 3). Treatment of cells with mentation and apoptotic cells similar to those in un-
1 mM GCEE led to a mean 20% increase in GSH com- treated control cells (Fig. 5A). The averages of late
pared with control, but this increase was not statistically apoptotic cells and necrotic cells were calculated and are
different from the GSH level of untreated controls. Inhi- reﬂected in the bar graph (Fig. 5D). The results suggest
bition of -glutamylcysteine synthetase by 500 M bu- that there is signiﬁcant (*P 0.05) DNA fragmentation
thionine sulfoximine (BSO) led to a signiﬁcant reduction as a measure of cells undergoing late-stage apoptosis and
in cellular GSH levels (2% of control); however, inhibition necrosis in the cells treated with A (10 M).
of -glutamylcysteine synthetase followed by addition of
1 mM GCEE raised total GSH levels to 88% of control GCEE Does Not Interfere With A (1– 42) Fibrils
and was not found to vary signiﬁcantly from control. That To determine whether GCEE interacted with
is, GCEE apparently led to elevated GSH levels by having A (1– 42) in such a way as to disturb the ﬁbril morphol-
704 Boyd-Kimball et al.
Fig. 5. Hoechst and propidium iodide staining for DNA fragmenta- text. D: The averages of late-stage apoptotic cells and necrotic cells
tion. Control neurons (A) and neurons treated for 1 hr with 1 mM were calculated and are reﬂected in this bar graph. The results presented
GCEE followed by 10 M A (1– 42) (C) show little evidence of are the mean SEM expressed as percentage of control values. Each
apoptotic neurons. Conversely, 10 M A (1– 42) treatment of neurons experiment was repeated three times with three independent samples.
shows evidence of late-stage apoptosis and necrosis (arrows; B). See *P 0.05 compared with the untreated control.
ogy of the peptide, EM studies were conducted (Fig. 6). DISCUSSION
No signiﬁcant difference was observed in the ﬁbril mor- Oxidative stress can lead to a variety of cellular
phology between A (1– 42) (Fig. 6A) and A (1– 42) in consequences, including altered protein conformation,
the presence of 1 mM GCEE (Fig. 6B). loss of protein function, increased protein aggregation,
GCEE Protects Against A (1– 42) Oxidative Stress 705
Fig. 6. EM studies of ﬁbril morphology. The extensive ﬁbril formation of A (1– 42) (A) is unaffected
in the presence of 1 mM GCEE (B).
decreased protein turnover, altered cellular redox poten- GCEE was unable to protect neurons against lipid
tial, altered Ca2 homeostasis, and ultimately cell death. peroxidation (data not shown). This ﬁnding is particularly
A (1– 42) has been shown to induce oxidative stress and important for two reasons. First, GSH is located in the
neurotoxicity in vitro in a manner inhibited by the chain- cytosol and is not lipid soluble. Consequently, GSH would
breaking antioxidant vitamin E, suggesting that ROS play not play a primary role in protection of the lipid bilayer
a pivotal role (Yatin et al., 1999a, 2000; Behl, 1999). from free radical attack. Second, and more importantly,
A (1– 42) has been shown to deplete GSH levels in as- A (1– 42) may exert its cytotoxic effect by inserting into
trocytes, leading to a signiﬁcant loss of neurons in vitro the lipid bilayer as small, soluble aggregates, where it
(Abramov et al., 2003). A (1– 42) has been implicated as initiates a lipid peroxidation cascade, followed by protein
a causative agent in AD (Selkoe, 2001a), which is associ- oxidation (Varadarajan et al., 2000; Butterﬁeld, 2002; But-
ated with oxidative stress and aging (Butterﬁeld and Laud- terﬁeld and Lauderback, 2002). GCEE-mediated up-
erback, 2002). GSH levels have been shown to decrease regulation of GSH does not interfere with A (1– 42)
with aging, leaving neurons vulnerable to ROS attack and aggregation (Fig. 6) or insertion into the lipid bilayer,
subsequent damage (Lui and Choi, 2000). Additionally, because lipid peroxidation was still present (data not
exogenous GSH has been shown to prevent A (1– 42)- shown). However, GCEE up-regulation of GSH was able
induced apoptosis in human cortical neurons (Medina et to protect the cell from protein oxidation. This is partic-
al., 2002). In this study, we tested the hypothesis that ularly important in that the index of lipid peroxidation
up-regulation of GSH by GCEE would protect primary used was 4-hydroxy-2-trans-nonenal (HNE). HNE is an
neuronal cell cultures against A (1– 42)-mediated oxida- , -unsaturated alkenal and a strong electrophilic product
tive stress and neurotoxicity. of lipid peroxidation that can readily react by Michael
Treatment of neurons with 1 mM GCEE was shown addition with cysteine, lysine, and histidine residues in
to elevate GSH levels 24 hr after administration, suggest- proteins to introduce a carbonyl functionality (Esterbauer
ing that GCEE was able to cross the plasma membrane of et al., 1991; Butterﬁeld and Stadtman, 1997; Mark et al.,
the neuron and provide a substrate for the synthesis of 1997; Uchida, 2003). As noted, in this study, the index of
GSH. Moreover, BSO was used to inhibit -gluta- protein oxidation was protein carbonyls. Protein carbony-
mylcysteine synthetase, which prevented de novo synthe- lation in neurons pretreated with GCEE followed by
sis of GSH and signiﬁcantly depleted the cellular GSH A (1– 42) was not found to be signiﬁcantly different from
level. Upon addition of GCEE to BSO-treated neurons, that in control. These results suggest that, although GCEE
GSH levels returned to control, which is consistent with up-regulation of GSH was unable to prevent lipid peroxi-
the notion that -glutamylcysteine was a substrate for dation, it was able to protect against damage mediated by
glutathione synthase, an enzyme that catalyzes the addition the lipid peroxidation product HNE. GSH has been
of C-terminal glycine to this dipeptide to form GSH. shown to react directly with electrophiles, such as HNE,
Additionally, pretreatment of neurons with GCEE pro- or indirectly in a reaction catalyzed by glutathione
tected these cells against A (1– 42)-mediated loss of mi- S-transferase (Esterbauer et al., 1991; Xie et al., 1998).
tochondrial function, protein oxidation, loss of neuronal Moreover, HNE can diffuse from its site of origin to react
network, and apoptotic cells. Taken together, these results with cytosolic proteins (Butterﬁeld and Stadtman, 1997).
suggest GCEE induces up-regulation of GSH, which, in In summary, we have shown that GCEE can cross
turn, protects neurons against protein oxidation and neu- the plasma membrane of neurons in culture and increase
rotoxicity. cellular GSH levels. Moveover, this increase in GSH levels
706 Boyd-Kimball et al.
can protect neurons from alterations in mitochondrial Darzynkiewicz Z, Li X, Gong J. 1994. Assays of cell viability: discrimina-
function, increased protein oxidation, loss of neuronal tion of cells dying by apoptosis. Methods Cell Biol 41:15–38.
network, and apoptosis induced by A (1– 42). This ﬁnd- Drake J, Kanski J, Varadarajan S, Tsoras M, Butterﬁeld DA. 2002. Elevation
ing is of importance in that A (1– 42) may play a central of brain glutathione by -glutamylcysteine ethyl ester protects against
role in the pathogenesis of AD, and GSH is an vital peroxynitrite-induced oxidative stress. J Neurosci Res 68:776 –784.
Drake J, Link CD, Butterﬁeld DA. 2003a. Oxidative Stress precedes ﬁbrillar
endogenous antioxidant found in millimolar concentra- deposition of Alzheimer’s disease amyloid -peptide (1– 42) in a trans-
tions in the brain (Cooper, 1997), although GSH levels genic Caenorhabditis elegans model. Neurobiol Aging 24:415-420.
decrease with age (Liu and Choi, 2000) . Thus, agents such Drake J, Sultana R, Aksenova M, Calabrese V, Butterﬁeld DA. 2003b.
as GCEE, which may react directly with ROS or by Elevation of mitochondrial glutathione by gamma-glutamylcysteine ethyl
increasing the availability of GSH in the brain, may pro- ester protects mitochondria against peroxynitrite-induced oxidative stress.
vide therapeutic beneﬁt in oxidative stress-associated neu- J Neurosci Res 74:917–927.
rodegenerative diseases such as AD (Butterﬁeld et al., Esterbauer H, Schaur RJ, Zollner H. 1991. Chemistry and biochemistry of
2002b). 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Rad Biol
ACKNOWLEDGMENTS Hensley K, Hall N, Subramaniam R, Cole P, Harris M, Aksenov M,
This work was supported in part by NIH grants Aksenova M, Gabbita P, Wu JF, Carney JM, Lovell M, Markesbery WR,
AG-05119 and AG-10836 to D.A.B. Butterﬁeld DA. 1995. Brain regional correspondence between Alzhei-
mer’s disease histopathology and biomarkers of protein oxidation. J Neu-
REFERENCES rochem 65:2146 –2156.
Abramov AY, Canevari L, Duchen MR. 2003. Changes in intracellular Liu R, Choi J. 2000. Age-associated decline in -glutamylcysteine syn-
calcium and glutathione in astrocytes as the primary mechanism of amy- thetase gene expression in rats. Free Rad Biol Med 28:566 –574.
loid neurotoxicity. J Neurosci 23:5088 –5095. Mark RJ, Lovell MA, Markesbery WR, Uchida K, Mattson MP. 1997. A
Anderson ME, Lou JL. 1998. Glutathione therapy: from prodrugs to genes. role for 4-hydroxynonenal, an aldehydic product of lipid peroxidation, in
Semin Liver Dis 18:415– 424. disruption of ion homeostasis and neuronal death induced by amyloid
Anderson ME, Meister A. 1989. Glutathione monoesters. Anal Biochem beta-peptide. J Neurochem 68:255–264.
183:16 –20. Markesbery MR. 1997. Oxidative stress hypothesis in Alzheimer’s disease.
Anderson ME, Powrie F, Puri RN, Meister A. 1985. Glutathione mono- Free Rad Biol Med 23:134 –147.
ethyl ester: preparation, uptake by tissues, and conversion to glutathione. Medina S, Martinez M, Hernanz A. 2002. Antioxidants inhibit the human
Arch Biochem Biophys 239:538 –548. cortical neuron apoptosis induced by hydrogen peroxide, tumor necrosis
Behl C. 1999. Vitamin E and other antioxidants in neuroprotection. Int J factor alpha, dopamine, and beta-amyloid peptide 1– 42. Free Rad Res
Vit Nutr Res 69:213–219. 36:1179 –1184.
Butterﬁeld DA. 2002. Amyloid beta-peptide (1– 42)-induced oxidative Selkoe DJ. 2001. Alzheimer’s disease results from the cerebral accumulation
stress and neurotoxicity: implications for neurodegeneration in Alzhei- and cytotoxicity of amyloid beta-protein. J Alzheimers Dis 3:75– 80.
mer’s disease brain. Free Rad Res 36:1307–1313. Subbarao KV, Richardson JS, Ang LC. 1990. Autopsy samples of Alzhei-
Butterﬁeld DA. 2003. Amyloid beta-peptide [1– 42]-associated free radical- mer’s cortex show increased peroxidation in vitro. J Neurochem 55:342–
induced oxidative stress and neurodegeneration in Alzheimer’s disease
brain: mechanisms and consequences. Curr Med Chem 10:2651–2659.
Uchida K. 2003. 4-Hydroxynonenal: a product and mediator of oxidative
Butterﬁeld DA, Lauderback CM. 2002. Lipid peroxidation and protein
stress. Prog Lipid Res 42:318 –343.
oxidation in Alzheimer’s disease brain: potential causes and consequences
Varadarajan S, Yatin S, Aksenova M, Butterﬁeld DA. 2000. Review:
involving amyloid beta-peptide-associated free radical oxidative stress.
Alzheimer’s amyloid beta-peptide-associated free radical oxidative stress
Free Rad Biol Med 32:1050 –1060.
Butterﬁeld DA, Stadtman ER. 1997. Protein oxidation processes in aging and neurotoxicity. J Struct Biol 130:184 –208.
brain. Adv Cell Aging Gerontol 2:161–191. Xie C, Lovell MA, Markesberry WR. 1998. Glutathione transferase pro-
Butterﬁeld DA, Drake J, Pocernich C, Castegna A. 2001. Evidence of tects neuronal cultures against four hydroxynonenal toxicity. Free Rad
oxidative damage in Alzheimer’s disease brain: central role for amyloid Biol Med 25:979 –988.
beta-peptide. Trends Mol Med 7:548 –554. Yatin SM, Varadarajan S, Link CD, Butterﬁeld DA. 1999a. In vitro and in
Butterﬁeld DA, Castegna A, Lauderback CM, Drake J. 2002a. Evidence vivo oxidative stress associated with Alzheimer’s amyloid -peptide (1–
that amyloid beta-peptide-induced lipid peroxidation and its sequelae in 42). Neurobiol Aging 20:325–330.
Alzheimer’s disease brain contribute to neuronal death. Neurobiol Aging Yatin SM, Yatin M, Aulick T, Ain KB, Butterﬁeld DA. 1999b. Alzheimer’s
23:655– 664. amyloid -peptide associated free radicals increase rat embryonic neuronal
Butterﬁeld DA, Pocernich CB, Drake J. 2002b. Elevated glutathione as a polyamine uptake and ornithine decarboxylase activity: protective effect
therapeutic strategy in Alzheimer’s disease. Drug Disc Res 56:328 – 437. of vitamin E. Neurosci Lett 263:17–20.
Cooper AJL. 1997. Glutathione in the brain: disorder of glutathione me- Yatin SM, Varadarajan S, Butterﬁeld DA. 2000. Vitamin E prevents Alz-
tabolism. In: The molecular and genetic basis of neurological disease. heimer’s amyloid beta-peptide (1– 42)-induced neuronal protein oxida-
Newton, MA: Butterworth-Heinemann. p 1195–1230. tion and reactive oxygen species production. J Alzheimers Dis 2:123–131.