Journal of Neuroscience Research 57:280–289 (1999)
Neurotoxicity of Methylglyoxal
and 3-Deoxyglucosone on Cultured Cortical
Neurons: Synergism Between Glycation
and Oxidative Stress, Possibly Involved
in Neurodegenerative Diseases
Seiji Kikuchi,1* Kazuyoshi Shinpo,1 Fumio Moriwaka,1 Zenji Makita,2 Toshio Miyata,3
and Kunio Tashiro1
1Department of Neurology, Hokkaido University School of Medicine, Sapporo, Hokkaido, Japan
2Department of Internal Medicine, Hokkaido University School of Medicine, Sapporo, Hokkaido, Japan
3Institute of Medical Science and Department of Medicine, Tokai University School of Medicine,
Isehara, Kanagawa, Japan
In this study, we investigate the neurotoxicity of
glycation, particularly early-stage glycation, and its
mechanisms, which are possibly synergized with oxida-
tive stress. Methylglyoxal (MG) and 3-deoxyglucosone
(3DG), intermediate products of glycation, are known
to further accelerate glycation and advanced glycation
endproducts (AGEs) formation. Both compounds
showed neurotoxicity on cultured cortical neurons
and these effects were associated with reactive oxygen
species production followed by neuronal apoptosis.
Pretreatment with N-acetylcysteine induced neuropro-
tection against MG and 3DG. Cotreatment, but not
pretreatment, with aminoguanidine protected neu-
rons against the neurotoxicities of both compounds.
The present study provides the first evidence that MG
and 3DG are neurotoxic to cortical neurons in culture.
Interference with the process by which glycation and
AGEs formation occur may provide new therapeutic
opportunities to reduce the pathophysiological changes
associated with neurodegeneration, if, as indicated
here, the participation of glycoxidation in the patho-
genesis of neurodegenerative diseases is essential. J.
Neurosci. Res. 57:280–289, 1999. r 1999 Wiley-Liss, Inc.
Key words: glycation; glycoxidation; neurotoxicity
INTRODUCTION
Glycation reactions are initiated by addition of
sugar aldehyde or ketone groups to free amino groups
mainly on lysine and arginine residues of proteins, or
N-terminal amino groups of proteins by a nonenzymatic
reaction known as the Maillard reaction (Monnier and
Cerami, 1981; Sell and Monnier, 1989; Smith et al.,
1995). A synthesis of intermediates up to Amadori
r 1999 Wiley-Liss, Inc.
compound is caused in the early stage of glycation, and
this reversible reaction depends on the concentration of
sugars and incubation time. In the late stage of glycation,
advance glycation endproducts (AGEs) are formed after a
complex cascade of repeated dehydration, condensation,
fragmentation, oxidation, and cyclization reactions,
through intermediates such as 3-deoxyglucosone (3DG).
These reactions are irreversible. The AGEs whose struc-
tures have been identified are as follows: N2-(carboxy-
methyl)lysine (CML) (Fu et al., 1996), pentosidine (Dyer
et al., 1991; Grandhee and Monnier, 1991), pyrralin
(Miyata and Monnier, 1992), imidazolon (Niwa et al.,
1997), crosslin, and X1, while other new compounds
remain to be identified. The change in the biological
activity and abnormal degradation process of proteins
such as superoxide dismutase-1 (SOD-1) (Arai et al.,
1987) are brought about by glycation or AGEs formation.
Moreover, there is a possibility that AGEs act as a protein
crosslinker by forming aggregates that are detergent-
insoluble and protease-resistant, and thereby inducing the
change in intracellular localization (Smith et al., 1995).
Such aggregates may interfere with the axonal transport.
Both the formation of adducts and the evolution in
AGEs modification of proteins are accelerated by oxygen
in a process called glycoxidation (Baynes, 1991; Smith et
al., 1995). On the other hand, the condensation of
reducing sugars with protein amino groups and subse-
Contract grant sponsor: Research Committee for CNS Degenerative
Diseases, the Ministry of Health and Welfare of Japan.
*Correspondence to: Dr. Seiji Kikuchi, Department of Neurology,
Hokkaido University School of Medicine, Kita 14 Nishi 5 Kitaku,
Sapporo, Hokkaido, Japan. E-mail: skikuti@med.hokudai.ac.jp
Received 24 September 1998; Revised 11 March 1999; Accepted 5
April 1999
 Neurotoxicity of MG and 3DG 281

Fig. 1. Methylglyoxal (MG) induces neuronal

cell death in a concentration- and time-
dependent manner in cultured cortical neu-
rons. A: Cells were exposed to 0 (a), 100 (b),
and 500 (c) µM MG for 24 hr. Cells were then
fixed and stained with anti-MAP2 antibody.
Bar ⫽ 10 µm. B: Cells were exposed for 24,
48, or 72 hr at the indicated concentrations.
Cell survivals were evaluated by counting the
number of MAP2-positive neurons.
282 Kikuchi et al.
quent Amadori rearrangement lead to redox-cycling ac-
tion, resulting in site-specific damage of proteins (Hunt et
al., 1988; Sakurai and Tsuchiya, 1988). These observa-
tions have led to the hypothesis that glycation-induced
pathology results from a cycle of oxidative stress, in-
creased chemical modification of proteins via the Mail-
lard reaction, and further AGEs-dependent oxidative
stress (Kaneto et al., 1996; Mullarkey et al., 1990;
Sakurai and Tsuchiya, 1988; Yan et al., 1994).
The contribution to disease made by glycation and
AGEs has been studied thus far in relation to diabetes and
diabetes-related complications (Myint et al., 1995; Ryle
et al., 1997). However, it has become clear that glycation
and AGEs also have an impact on physiological aging
(senile cataract, arteriosclerosis, etc.; Vlassara et al.,
1994), neurodegenerative diseases, such as Alzheimer’s
and Parkinson’s disease (Castellani et al., 1996), and
amyotrophic lateral sclerosis (ALS) (Arai et al., 1987;
Chou et al., 1998; Ookawara et al., 1992; Shibata et al.,
1997).
In Alzheimer’s disease (Horie et al., 1997; Takeda
et al., 1996), senile plaques and neurofibrillary tangles are
sometimes stained positively by anti-AGEs antibodies
(Smith et al., 1994).
In the present study, we investigate the neurotoxic-
ity of glycation, particularly early-stage glycation, and its
mechanisms, which are possibly synergized with oxida-
tive stress. For this purpose, we used a glycation-
accelerated system, since cultured cells cannot be main-
tained over 2 weeks. If glucose is used as a reducing
sugar, a much longer incubation may be needed to detect
AGEs, even in test tubes (Grandhee and Monnier, 1991).
3DG and methylglyoxal (MG), intermediate products of
glycation, are known to further accelerate glycation and
AGEs formation (Che et al., 1997; Lo et al., 1994; Okado
et al., 1996; Shinoda, 1994; Suzuki et al., 1998; Yamada
et al., 1994).
The pathomechanism of neurodegenerative dis-
eases, in which many factors are thought to be involved,
was examined from the viewpoint of the synergism
between glycation and oxidation. The present study
provides the first evidence that the glycation intermedi-
ates, MG and 3DG, are neurotoxic on cortical neurons in
culture. The possibility of a treatment is suggested
(Edelstein and Brownlee, 1992) if the participation of
glycoxidation in the pathogenesis of neurodegenerative
diseases is essential and glycoxidation is the modulating
factor of the neurotoxicity in these diseases.
MATERIALS AND METHODS
Materials
3-DG was kindly donated by Dr. F. Hayase (Depart-
ment of Agricultural Chemistry, Meiji University, Ka-
nagawa, Japan) and T. Miyata (coauthor). MG was
purchased from Sigma (St. Louis, MO). Neurobasal
medium and B27 supplements were purchased from
GibcoBRL Life Technologies (Gaithersburg, MD). Amino-
guanidine (AMG) was from Wako Chemicals (Osaka,
Japan). N-acetyl-L-cysteine (NAC) was from Katayama
Chemical (Osaka, Japan). 2,7-Dichlorofluorescin diac-
etate (DCF-DA) was from Molecular Probes (Eugene,
OR). Polyclonal anti-AGEs antibody was prepared by
one of the authors (Z. Makita).
Cell Culture and Experimental Treatments
Dissociated cell cultures were established from the
cortices of 14-day Sprague-Dawley rat fetuses under deep
anesthesia with diethylether. Cells were dissociated by
incubation for 20 min in phosphate-buffered saline (PBS)
containing 2 mg/ml trypsin and 0.2% DNase, followed by
trituration. Cells were plated onto poly-L-lysine-coated
8-well chamber slides at a density of approximately
600/mm2 in Eagle’s minimum essential medium supple-
mented with 10% heat-inactivated fetal bovine serum.
Twenty-four hr after plating, the culture medium was
replaced with neurobasal medium containing B27 supple-
ments. Experiments were performed on 10-day-old cul-
tures. Under these culture conditions, more than 90% of
the cells were found to be neurons. Cell cultures were
exposed to MG or 3DG of various concentrations for 24
hr with or without AMG or NAC.
Analyses of Neuronal Survival
Neuronal survival was assessed by counting the num-
ber of morphologically undamaged neurons on immuno-
cytochemical staining using monoclonal anti-microtubule-
associated protein 2 (anti-MAP2) antibody. After experi-
mental treatments, cultures were fixed in 4% paraformal-
dehyde for 30 min, and membranes were permeabilized
with 0.2% Triton X-100. Cells were incubated with
blocking solution (10% normal rabbit serum in PBS) for
30 min and then with monoclonal anti-MAP2 antibody in
blocking solution at a dilution of 1:500 overnight at 4°C.
Biotinylated secondary antibody, ABC solution, and
diaminobenzidine were used to visualize stained cells.
Analysis of Neuronal Apoptosis and Detection of
Reactive Oxygen Species (ROS)
As a measure of apoptosis, cells were fixed in 4%
paraformaldehyde, membranes were permealized with
0.2% Triton X-100, and cells were stained with the
fluorescent DNA-binding dye Hoechst 33258 (1 mg/ml).
Hoechst-stained cells were visualized and photographed
under epifluorescence illumination (excitation, 340 mn;
barrier filter, 510 nm).
Fig. 2. 3-Deoxyglucosone (3DG) induces
neuronal cell death in a concentration- and
time-dependent manner in cultured cortical
neurons. A: Cells were exposed to 0 (a), 100
(b), and 500 (c) µM 3DG for 24 hr. Cells
were then fixed and stained with anti-MAP2
antibody. Bar ⫽ 10 µm. B: Cells were
exposed for 24, 48, or 72 hr at the indicated
concentrations. Cell survivals were evalu-
ated by counting the number of MAP2-
positive neurons.
284 Kikuchi et al.
After incubation with the reagents described, ineter-
nucleosomal DNA cleavage was detected by ladder
formation as reported previously by Kaneto et al. (1996).
ROS were detected with DCF-DA, which produces a
green fluorescence when oxidized. Cultures were incu-
bated with 50 µM DCF-DA for 30 min at 37°C, rinsed
three times with PBS, and visualized by epifluorescence
microscopy (excitation, 488 nm; emission, 510 nm).
Immunocytochemical Study for AGEs
For the attempt to produce AGE on our cell culture
system, cells were treated with 50 µM 3DG for 3 days and
then fixed in 4% paraformaldehyde for 30 min, and
membranes were permeabilized witn 0.2% Triton X-100.
Cells were incubated with blocking solution (10% normal
goat serum in PBS) for 30 min and then with polyclonal
anti-AGE antibody in blocking solution at various dilu-
tions (1:100, 1:500, 1:1,000, and 1:10,000) overnight at
4°C. Biotinylated secondary antibody, ABC solution, and
diaminobenzidine were used to visualize stained cells.
RESULTS
MG and 3DG Have Neurotoxic Effects on Cultured
Cortical Neurons
The effects of MG on neuronal survival were first
examined morphologically. Incubation of cortical neu-
rons with 100 µM MG for 24 hr resulted in decreases in
cell numbers as well as in number and length of neurites
(Fig. 1A). Treatment with MG resulted in a concentra-
tion- and time-dependent decrease in the number of
viable neurons when cell survival was assessed 24 hr after
treatment (Fig. 1B). LC50 for MG neurotoxicity was 130 µM.
Incubation of cortical neurons with 100 µM 3DG
for 24 hr also resulted in decreases in cell numbers as well
as in number and length of neurites (Fig. 2A). Treatment
with 3DG resulted in a concentration- and time-
dependent decrease in the number of viable neurons when
assessed 24 hr after treatments (Fig. 2B). LC50 for 3DG
neurotoxicity was 209 µM.
MG and 3DG Induce Neuronal Apoptosis
Cells were treated with 100 µM of MG or 100 µM
of 3DG, and apoptosis was examined after 30 min by
assessment of chromatin condensation in cells stained
with the Hoechst 33258 for 30 min. Treatment of neurons
with both agents resulted in chromatin condensations (for
MG; Fig. 3, for 3DG; data not shown). Nuclear fragmen-
tation was demonstrared to start 30 min after the exposure
to MG, and was increasing even after 3 hr. DNA ladder
formation was induced after incubation with 100 µM of
MG for 3 and 24 hr (data not shown).
MG and 3DG Induce Production of Peroxide
Cells treated with 50 µM of MG for 30 min emitted
higher levels of peroxide (Fig. 4b) than those treated with
vehicle (Fig. 4a) as detected using the fluorescence
marker DCF-DA.
Neurotoxicity of MG and 3DG Is Attenuated
by NAC and AMG
Cultures were pretreated for 24 hr with 1 mM NAC
and then exposed to MG. Neuronal degenerations in-
duced by MG were significantly reduced in cultures
pretreated with NAC (Fig. 5).
Cultures were pretreated for 24 hr with 1 mM of
AMG and then exposed to 3DG. Pretreatment with AMG
showed no significant protective effect against neurotox-
icity induced by 3DG (data not shown). Cotreatment with
AMG, however, resulted in significant protection from
the neurotoxicity of 3DG (Fig. 6).
Exposure to 3DG Fails to Produce Positive AGEs
Immunostaining
For the attempt to produce AGEs in cultured
cortical neurons, cultured cells were exposed to 50 µM of
3DG for 3 days, which concentration caused an approxi-
mately 40% reduction in neuronal survival. There was no
significant staining for anti-AGEs antibody in cortical
neurons using this method of exposure (data not shown).
DISCUSSION
In the present study, we investigated the neurotoxic-
ity of glycation, particularly early-stage glycation, and its
mechanisms, which are possibly synergized with oxida-
tive stress. For this purpose, we used a glycation-
accelerated system. MG and 3DG are known to accelerate
glycation reactions and crosslinking of proteins. In cases
of hyperglycemia, it has been shown that the concentra-
tions of both compounds are elevated and that both
compounds accelerate glycation reactions.
MG (Che et al., 1997; Lo et al., 1994; Okado et al.,
1996; Suzuki et al., 1998) is one of a series of dicarbonyl
intermediates, which includes such intermediates as glu-
cosone, deoxyglucosone, dehydroascorbate, and glyoxal,
that have been identified in the Maillard reaction. MG is
formed nonenzymatically by amine-catalysed sugar frag-
mentations, and by spontaneous decomposition of triose
phosphate intermediates in glycolysis. It is also a product
of the metabolism of acetol, an intermediate in the
catabolism of both threonine and the ketone body ac-
etone. MG reacts rapidly with amino, guanidino, and thiol
functional groups in proteins leading to denaturation and
crosslinking of proteins. The physiological significance
 Neurotoxicity of MG and 3DG 285

Fig. 3. MG induces apoptotic nuclear alterations in cultured cortical neurons. Cells were
exposed to vehicle (a), or 100 µM of MG for 30 min (b) and 3 hr (c), then were stained with
Hoechst dye. MG induced nuclear condensation and fragmentation in neurons. Neurons were
photographed using a fluorescence microscope with a 20⫻ objective.

Fig. 4. MG induces peroxide accumulation in cultured cortical neurons. Cells were exposed to 0
(a) or 50 µM (b) of MG for 30 min. Cells were then washed and loaded with 50 µM of
2,7-Dichlorofluorescin diacetate (DCF-DA) for 30 min.
286 Kikuchi et al.

of the modification of proteins by MG has been difficult

to judge, since until recently, the physiological concentra-
tion of MG had not been determined. Lo et al. (1994)
have shown that at physiological concentration, MG
binds and irreversibly modifies plasma proteins.
3DG (Okado et al., 1996; Shinoda, 1994; Suzuki et
al., 1998; Yamada et al., 1994), another highly reactive
carbonyl compound, is formed after the multiple dehydra-
tion and rearrangements of Amadori products. 3DG
reacts again with free amino groups, leading to crosslink-
ing via the formation of AGEs in the late stage of the
Maillard reaction. Incubation of 3DG with proteins leads
to the formation of pyrraline and pentosidine (Dyer et al.,
1991), whose structures have identified them as AGEs
occurring in human tissue proteins.
Plasma concentrations of MG and 3DG reach 5 µM
and 1 µM, respectively, under diabetic conditions (Phill-
ips et al., 1993). Niwa et al. (1993) reported plasma
concentration of 3DG in DM patients with nephropathy
as 1,235 ng/ml (7.6 µM), versus 314 ng/ml (1.9 ⬍µM) in
healthy subjects. These compounds are capable of induc-
ing apoptotic cell death in the macrophage-derived cell
line, U937, with 10–300 µM of MG and 10–1,000 µM of Fig. 5. Effects of N-acetyl-L-cysteine (NAC) on MG neurotox-
3DG (Okado et al., 1996). In addition, PC12 cells are icity. Cells were preincubated for 24 hr in the presence or
susceptible to MG of 300 µM and over, and to 3DG of 10 absence of 10 mM of NAC, then exposed for 24 hr to MG.
mM and over (Suzuki et al., 1998). Taken together, these
results suggest that the concentrations of MG and 3DG
used here were reasonable and were appropriate for
investigating the acute phase of their neurotoxicity.
Until now, the validity of the incubation time for
glycation and AGEs formation has been investigated in
test tubes using reducing sugars and proteins such as
albumin, RNase, and collagen. Such investigations have
shown that a period of several days to a month is needed,
depending on the sugars used. However, the incubation
time in our experiments (24 hr) is sufficient to examine
the early phase of neurotoxicity of glycoxidation with
MG and 3DG, although it may be insufficient to detect
AGEs formation.
Recently, Niwa et al. (1998) reported that 3DG and
glyoxal accelerated the formation of CML, a chemically
defined AGE, in the explant-cultured neurons of dorsal
root ganglia. To our knowledge, there is no direct
pathway through which CML is produced from 3DG.
CML is, however, reported to be produced by lipid
peroxidation under the condition of oxidant stress. Even
though glyoxal is a known precursor of CML, glyoxal
itself is also formed during lipid peroxidation. (Moreover,
the significance of glyoxal has not been elucidated even
in diabetic complications.) It is the oxidant stress induced
by 3DG and MG (a derivative of glyoxal) that was demon- Fig. 6. Effects of AMG on 3DG neurotoxicity. Cells were
strated in our study. Taken together, it is much more possible exposed for 24 hr to 3DG in the presence or absence of 1 mM of
that, in their experiment (Niwa et al., 1998), CML was AMG.

synthesized by lipid peroxidation under the oxidant stress
induced by 3DG and glyoxal, and that they did not fully
elucidate the involvement of glycation in their system.
Contrary to our study, neurotoxicity was not demon-
strated in their study (Niwa et al., 1998), although they
used 3DG up to 1 mM which was enough to bring about
neurotoxicity in our study. Does it mean that CML is not
toxic enough to cause neuronal death? There are several
possibilities to explain this discrepancy. The most impor-
tant point is the difference of culture system, i.e., our
dissociated cultured neurons can be easily exposed to
3DG directly. On the other hand, their explant cultured
neurons (Niwa et al., 1998) were surrounded by many
non-neuronal cells which may have had protective effects
on neurons.
In the present study, we elucidated the neurotoxicity
of early intermediates of glycation through oxidant stress.
For the study of the neurotoxicity of AGE, which requires
several cascade reactions after early intermediates, analy-
ses with non-CML AGE are necessary because CML,
itself, can not generate reactive oxygen species or act as a
crosslinker. Long-term incubation (for example, with
slice culture system) is also necessary to make non-CML
AGE generate.
LC50 of MG and 3DG were different, i.e., 130 µM
and 209 µM, respectively. As mentioned above, in the
macrophage-derived cell line, U937, the concentrations
of MG and 3DG producing intracellular peroxide were
essentially the same, while nuclear fragmentation (Hoechst
stain) was detected by 200 µM MG and 1,000 µM 3DG
(Shinoda, 1994). In PC12 cells, cell viability was more
greatly diminished by MG treatment than by treatment
with 3DG (Suzuki et al., 1998). These differences are
possibly due to: (1) the cell permeability of MG (Che et
al., 1997) and 3DG; (2) their reactivity with proteins; (3)
the amount of reactive oxygen species produced during
glycation; (4) the toxicity of glycation products as a
protein crosslinker; or (5) the mechanisms of detoxifica-
tion (Phillips et al., 1993; Suzuki et al., 1998).
Previously, none of glycation products had been
investigated from a toxicological point of view. Future
studies will be needed to examine the difference between
MG and 3DG toxicity using a DCF experiment to
quantify the amount of ROS. With regard to the detoxifi-
cation mechanism, MG is mainly inactivated enzymati-
cally by the GSH-dependent glyoxalase pathway and
NADPH-dependent aldose reductase pathway (Phillips et
al., 1993). 3DG is mainly detoxified by NADPH-
dependent aldehyde reductase (Suzuki et al., 1998). In the
aldehyde reductase (ALR) gene-transfected PC12 cells,
the cytotoxicity of both MG and 3DG and apoptotic cell
death were decreased (Suzuki et al., 1998). This suggests
that intracellular ALR protects neural cells from cytotox-
Neurotoxicity of MG and 3DG 287
icity of MG and 3DG, and that neural cells, which normally
express a low level of ALR, might be susceptible to
glycation. Future studies will be needed to investigate the
detoxification mechanism in each cell type.
The intracellular peroxide level was assessed using
oxidant-sensitive fluorescent DCF-DA. Fifty µM of MG
is sufficient to induce ROS in cultured cortical neurons in
30 min. As for U937 cells, peroxide production has been
analyzed by fluorescent activated cell sorter (FACS), with
the result that MG over 10 µM was shown to produce
peroxide (Okado et al., 1996). NAC demonstrated neuro-
protective action by treatment prior to the exposure to
MG. NAC can raise intracellular GSH levels and thereby
provide cells with the cosubstrate required to eliminate
hydroperoxides, resulting in protection from ROS. In
addition, NAC also reacts with MG directly, rapidly, and
reversely to form the hemithioacetal adduct. To exclude
the latter possibility, NAC was added simultaneously
with the exposure to MG or added after it. We were
unable to demonstrate a protective effect of NAC using
this method of administration. The level of GSH induc-
tion and its time course are important subjects for future
research. The concentration of NAC showing neuroprotec-
tive action was 1 mM, which is the same as that used
previously (Kaneto et al., 1996). Although examination
with other antioxidants will be necessary, the participa-
tion of oxidative stress was confirmed by the DCF
experiment and by the protective action of NAC against
the neurotoxicity of 3DG and MG.
Aminoguanidine has been used already as an inhibi-
tor of AGEs crosslinking in experimental diabetes. The
primary mechanism of aminoguanidine is direct reaction
with Amadori derivative fragments, such as 3DG, which
prevents subsequent AGEs formation in susceptible pro-
teins. It has been reported, however, that aminoguanidine
has antioxidant properties in addition to the effects on
glycation (Scaccini et al., 1994). Aminoguanidine also
inhibits inducible nitric oxide synthase iNOS (Soulis et
al., 1997). In our study, cotreatment of aminoguanidine
protected cultured cortical neurons against 3DG neurotox-
icity. This suggests that the protective action of aminogua-
nidine in our experiment was due to a direct reaction
which did not require pretreatment to induce other
proteins, such as iNOS. The uptake of aminoguanidine
into the cell has not yet been confirmed.
It is difficult to conclude the pathological signifi-
cance of glycation in neurodegenerative diseases. In the
present study, we showed that glycation, even in the early
stage, can be harmful to neurons. And if it is a modulating
factor of the pathomechanism, interference with the
process by which AGEs formation occurs may provide
new therapeutic opportunities to reduce the pathophysi-
ological changes associated with neurodegeneration. In
288 Kikuchi et al.
addition to detailed research of the effects of aminoguani-
dine and various antioxidants, an examination of the
influence of various agents concerned with the trophic
factors, the signal transduction, and the cytokines on
glycation, will be necessary.
ACKNOWLEDGMENTS
We thank Dr. T. Hayase for kindly providing the
3DG. This work was supported, in part, by a research
grant from the Research Committee for CNS Degenera-
tive Diseases, the Ministry of Health and Welfare of
Japan.
REFFRENCES
Arai K, Maguchi S, Fujii S, Ishibashi H, Oikawa K, Taniguchi N. 1987.
Glycation and inactivation of human Cu-Zn-superoxide dismu-
tase. Identification of the in vitro glycated sites. J Biol Chem
262:16969–16972.
Baynes JW. 1991. Role of oxidative stress in development of complica-
tions in diabetes. Diabetes 40:405–412.
Castellani R, Smith MA, Richey PL, Perry G. 1996. Glycoxidation and
oxidative stress in Parkinson disease and diffuse Lewy body
disease. Brain Res 737:195–200.
Che W, Asahi M, Takahashi M, Kaneto H, Okado A, Higashiyama S,
Taniguchi N. 1997. Selective induction of heparin-binding
epidermal growth factor-like growth factor by methylglyoxal
and 3-deoxyglucosone in rat aortic smooth muscle cells. The
involvement of reactive oxygen species formation and a pos-
sible implication for atherogenesis in diabetes. J Biol Chem
272:18453–18459.
Chou SM, Wang HS, Taniguchi A, Bucala R. 1998. Advanced glycation
end products in neurofilament conglomeration of motoneurons
in familial and sporadic amyotrophic lateral sclerosis. Mol Med
4:324–332.
Dyer DG, Blackledge JA, Thorpe SR, Baynes JW. 1991. Formation of
pentosidine during nonenzymatic browning of proteins by
glucose. Identification of glucose and other carbohydrates as
possible precursors of pentosidine in vivo. J Biol Chem 266:
11654–11660.
Edelstein D, Brownlee M. 1992. Mechanistic studies of advanced
glycosylation end product inhibition by aminoguanidine. Diabe-
tes 41:26–29.
Fu MX, Requena JR, Jenkins AJ, Lyons TJ, Baynes JW, Thorpe SR.
1996. The advanced glycation end product, Nepsilon-(carboxy-
methyl)lysine, is a product of both lipid peroxidation and
glycoxidation reactions. J Biol Chem 271:9982–9986.
Grandhee SK, Monnier VM. 1991. Mechanism of formation of the
Maillard protein cross-link pentosidine. Glucose, fructose, and
ascorbate as pentosidine precursors. J Biol Chem 266:11649–
11653.
Horie K, Miyata T, Yasuda T, Takeda A, Yasuda Y, Maeda K, Sobue G,
Kurokawa K. 1997. Immunohistochemical localization of ad-
vanced glycation end products, pentosidine, and carboxymethyl-
lysine in lipofuscin pigments of Alzheimer’s disease and aged
neurons. Biochem Biophys Res Commun 236:327–332.
Hunt JV, Dean RT, Wolff SP. 1988. Hydroxyl radical production and
autoxidative glycosylation. Glucose autoxidation as the cause of
protein damage in the experimental glycation model of diabetes
mellitus and ageing. Biochem J 256:205–212.
Kaneto H, Fujii J, Myint T, Miyazawa N, Islam KN, Kawasaki Y,
Suzuki K, Nakamura M, Tatsumi H, Yamasaki Y, Taniguchi N.
1996. Reducing sugars trigger oxidative modification and
apoptosis in pancreatic beta-cells by provoking oxidative stress
through the glycation reaction. Biochem J 320:855–863.
Lo TW, Westwood ME, McLellan AC, Selwood T, Thornalley PJ.
1994. Binding and modification of proteins by methylglyoxal
under physiological conditions. A kinetic and mechanistic study
with N alpha-acetylarginine, N alpha-acetylcysteine, and N
alpha-acetyllysine, and bovine serum albumin. J Biol Chem
269:32299–32305.
Miyata S, Monnier V. 1992. Immunohistochemical detection of
advanced glycosylation end products in diabetic tissues using
monoclonal antibody to pyrraline. J Clin Invest 89:1102–1112.
Monnier VM, Cerami A. 1981. Nonenzymatic browning in vivo:
possible process for aging of long-lived proteins. Science
211:491–493.
Mullarkey CJ, Edelstein D, Brownlee M. 1990. Free radical generation
by early glycation products: a mechanism for accelerated
atherogenesis in diabetes. Biochem Biophys Res Commun
173:932–939.
Myint T, Hoshi S, Ookawara T, Miyazawa N, Suzuki K, Taniguchi N.
1995. Immunological detection of glycated proteins in normal
and streptozotocin-induced diabetic rats using anti hexitol-
lysine IgG. Biochim Biophys Acta 1272:73–79.
Niwa H, Takeda A, Wakai M, Miyata T, Yasuda Y, Mitsuma T,
Kurokawa K, Sobue G. 1998. Accelerated formation of Ne-
(carboxymethyl)lysine, an advanced glycation end product, by
glyoxal and 3-deoxyglucosone in cultured rat sensory neurons.
Biochem Biophys Res Common 248: 93–97.
Niwa T, Takeda N, Yoshizumi H, Tatematsu A, Ohara M, Tomiyama S,
Niimura K. 1993. Presence of 3-deoxyglucosone, a potent
protein crosslinking intermediate of Maillard reaction, in dia-
betic serum. Biochem Biophys Res Commun 196:837–843.
Niwa T, Katsuzaki T, Miyazaki S, Miyazaki T, Ishizaki Y , Hayase F,
Tatemichi N, Takei Y. 1997. Immunohistochemical detection of
imidazolone, a novel advanced glycation end product, in kidneys and
aortas of diabetic patients. J Clin Invest 99:1272–1280.
Okado A, Kawasaki Y, Hasuike Y, Takahashi M, Teshima T, Fujii J,
Taniguchi N. 1996. Induction of apoptotic cell death by
methylglyoxal and 3-deoxyglucosone in macrophage-derived
cell lines. Biochem Biophys Res Commun 225:219–224.
Ookawara T, Kawamura N, Kitagawa Y, Taniguchi N. 1992. Site-
specific and random fragmentation of Cu,Zn-superoxide dismu-
tase by glycation reaction. Implication of reactive oxygen
species. J Biol Chem 267:18505–18510.
Phillips SA, Mirrlees D, Thornalley PJ. 1993. Modification of the
glyoxalase system in streptozotocin-induced diabetic rats. Ef-
fect of the aldose reductase inhibitor Statil. Biochem Pharmacol
46:805–811.
Ryle C, Leow CK, Donaghy M. 1997. Nonenzymatic glycation of
peripheral and central nervous system proteins in experimental
diabetes mellitus. Muscle Nerve 20:577–584.
Sakurai T, Tsuchiya S. 1988. Superoxide production from nonenzymati-
cally glycated protein. FEBS Lett 236:406–410.
Scaccini C, Chiesa G, Jialal I. 1994. A critical assessment of the effects
of aminoguanidine and ascorbate on the oxidative modification
of LDL: evidence for interference with some assays of lipopro-
tein oxidation by aminoguanidine. J Lipid Res 35:1085–1092.
Sell DR, Monnier VM. 1989. Structure elucidation of a senescence
cross-link from human extracellular matrix. Implication of
pentoses in the aging process. J Biol Chem 264:21597–21602.

Shibata NHA, Kobayashi M, Dal Canto MC, Gurney ME, Ikeda K,
Horiuchi S. 1997. Advance glycosylation end products (AGE)
deposition in intraneuronal hyaline inclusions (IHIs) of spinal
cords from familial amyotrophic lateral sclerosis (ALS) patients
with superoxide dismutase-1 (SOD1) mutation and from trans-
genic mice expressing mutant human SOD1. Brain Pathol
7:1073.
Shinoda T. 1994. Suppression of cell-cycle progression during the S
phase of rat fibroblasts by 3-deoxyglucosone, a Maillard
reaction intermediate. Biosci Biotech Biochem 58:2215–2219.
Smith MA, Richey PL, Taneda S, Kutty RK, Sayre LM, Monnier VM,
Perry G. 1994. Advanced Maillard reaction end products, free
radicals, and protein oxidation in Alzheimer’s disease. Ann N Y
Acad Sci 738:447–454.
Smith MA, Sayre LM, Monnier VM, Perry G. 1995. Radical AGEing
in Alzheimer’s disease. Trends Neurosci 18:172–176.
Soulis T, Cooper ME, Sastra S, Thallas V, Panagiotopoulos S, Bjerrum
OJ, Jerums G. 1997. Relative contributions of advanced glyca-
tion and nitric oxide synthase inhibition to aminoguanidine-
mediated renoprotection in diabetic rats. Diabetologia 40:1141–
1151.
Neurotoxicity of MG and 3DG 289
Suzuki K, Koh YH, Mizuno H, Hamaoka R, Taniguchi N. 1998.
Overexpression of aldehyde reductase protects PC12 cells from
the cytotoxicity of methylglyoxal or 3-deoxyglucosone. J Bio-
chem (Tokyo) 123:353–357.
Takeda A, Yasuda T, Miyata T, Mizuno K, Li M, Yoneyama S, Horie K,
Maeda K, Sobue . 1996. Immunohistochemical study of ad-
vanced glycation end products in aging and Alzheimer’s disease
brain. Neurosci Lett 221:17–20.
Vlassara H, Bucala R, Striker L. 1994. Pathogenic effects of advanced
glycosylation: biochemical, biologic, and clinical implications
for diabetes and aging. Lab Invest 70:138–151.
Yamada H, Miyata S, Igaki N, Yatabe H, Miyauchi Y, Ohara T, Sakai
M, Shoda H, Oimomi M, Kasuga M. 1994. Increase in
3-deoxyglucosone levels in diabetic rat plasma. Specific in vivo
determination of intermediate in advanced Maillard reaction. J
Biol Chem 269:20275–20280.
Yan SD, Chen X, Schmidt AM, Brett J, Godman G, Zou YS, Scott CW,
Caputo C, Frappier T, Smith MA, Perry G, Yen SH, Stern D.
1994. Glycated tau protein in Alzheimer disease: a mechanism
for induction of oxidant stress. Proc Natl Acad Sci USA
91:7787–7791.