Glycation damage targets glutamate dehydrogenase in the
rat liver mitochondrial matrix during aging
Maud Hamelin, Jean Mary, Michal Vostry, Bertrand Friguet* and Hilaire Bakala*
Laboratoire de Biologie et Biochimie Cellulaire du Vieillissement, Universite
´
Paris 7-Denis Diderot, France
Aging is characterized by gradual deterioration of cel-
lular functions [1] associated with cumulative damage
to intracellular macromolecules, particularly proteins
[2]. Several lines of evidence suggest that mitochondria
play a key role in aging, both by producing intracellu-
lar reactive oxygen species (ROS) and by being the
most adversely affected organelles during aging [3].
The electron transport chain located in the inner mito-
chondrial membrane is implicated in production of
ATP via oxidative phosphorylation, and is also known
to be the major intracellular site of free radical genera-
tion, such as that of superoxide anions and, sub-
sequently, other potentially deleterious ROS [4].
Numerous data have indicated an age-related increase
in the rate of mitochondrial free radical generation
and in the extent of oxidative damage to mitochondrial
macromolecules [5–9], especially enzymes involved in
the respiratory chain, leading to impairment of respira-
tory activity [7,9–12].
Although direct oxidation of proteins and other
macromolecules is believed to be the main type of
endogenous damage during aging [3,13,14], ROS can
Keywords
aging; glycation; liver mitochondria;
proteomics; urea cycle enzymes

Correspondence
H. Bakala, Laboratoire de Biologie et
Biochimie Cellulaire du Vieillissement,
EA3106 ⁄ IFR117, Universite
´
Paris 7-Denis
Diderot, 2 place Jussieu, 75251 Paris,
Cedex 05, France
Fax: +33 1 44 27 39 25
Tel: +33 1 44 27 82 27
E-mail:
*These authors contributed equally to this
work
(Received 22 May 2007, revised 21 Septem-
ber 2007, accepted 25 September 2007)
doi:10.1111/j.1742-4658.2007.06118.x
Aging is accompanied by gradual cellular dysfunction associated with an
accumulation of damaged proteins, particularly via oxidative processes.
This cellular dysfunction has been attributed, at least in part, to impair-
ment of mitochondrial function as this organelle is both a major source of
oxidants and a target for their damaging effects, which can result in a
reduction of energy production, thereby compromising cell function. In the
present study, we observed a significant decrease in the respiratory activity
of rat liver mitochondria with aging, and an increase in the advanced gly-
cation endproduct-modified protein level in the mitochondrial matrix. Wes-
tern blot analysis of the glycated protein pattern after 2D electrophoresis
revealed that only a restricted set of proteins was modified. Within this set,
we identified, by mass spectrometry, proteins connected with the urea cycle,
and especially glutamate dehydrogenase, which is markedly modified in
older animals. Moreover, mitochondrial matrix extracts exhibited a signifi-

cant decrease in glutamate dehydrogenase activity and altered allosteric
regulation with age. Therefore, the effect of the glycating agent methylgly-
oxal on glutamate dehydrogenase activity and its allosteric regulation was
analyzed. The treated enzyme showed inactivation with time by altering
both catalytic properties and allosteric regulation. Altogether, these results
showed that advanced glycation endproduct modifications selectively affect
mitochondrial matrix proteins, particularly glutamate dehydrogenase, a
crucial enzyme at the interface between tricarboxylic acid and urea cycles.
Thus, it is proposed that glycated glutamate dehydrogenase could be used
as a biomarker of cellular aging. Furthermore, these results suggest a role
for such intracellular glycation in age-related dysfunction of mitochondria.
Abbreviations
AGE, advanced glycation endproduct; CEL, N-(carboxyethyl)lysine; CML, carboxymethyl-lysine; GDH, glutamate dehydrogenase; GO, glyoxal;
MGO, methylglyoxal; OCT, ornithine carbamoyl transferase; RCR, respiratory control ratio; ROS, reactive oxygen species; TCA, tricarboxylic
acid.
FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS 5949
also affect protein function through either lipoperoxi-
dative production of reactive aldehydes [15] or glycoxi-
dation pathways [16]. Indeed, there is a causal
relationship between hyperglycemia-induced ROS
generation and intracellular advanced glycation end-
product (AGE) formation [17]. This rise in AGE was
shown to be primarily if not exclusively due to a rapid
increase in AGE-forming methylglyoxal concentration
[18].
Alpha-dicarbonyl compounds such as glyoxal (GO)
and methylglyoxal (MGO) are physiological, highly
reactive intermediates involved in the Maillard reaction
[19]. Interestingly, MGO may originate from various
biochemical pathways, including dephosphorylation of

glycolytic intermediates, metabolites of the polyol
pathway and metabolism of aminoacetones [20]. At
physiological concentrations, MGO primarily targets
the arginine residues of proteins, leading primarily to
AGE-imidazolone [21,22] and -lysine residues to form
AGE adducts: N-(carboxyethyl)lysine (CEL) and meth-
ylglyoxal lysine dimer [23].
In a previous study, using an immunochemical
method, we showed that glycated proteins accumulate
in the rat liver mitochondrial matrix with aging [24].
In the present study, we show that glycation targets a
limited set of proteins, the most severely affected being
glutamate dehydrogenase, a crucial enzyme at the
interface between the tricarboxylic acid (TCA) and
urea cycles, indicating the interference of matrix
enzymes in mitochondrial impairment function with
aging.
Results
Age-related changes in the respiratory function
of isolated mitochondria
Mitochondria were isolated from the livers of young
(3-month-old) and old (27-month-old) Wistar rats and
their respiratory parameters were measured. As shown
in Table 1, there was a significant decrease in the rates
of mitochondrial oxygen consumption between
3-month-old and 27-month-old rats with gluta-
mate ⁄ malate or succinate as substrates. The ADP ⁄ O
ratios (P ⁄ O) obtained whatever the substrate used indi-
cated no change in coupling efficiency. Although the
respiratory control ratio (RCR) slightly decreased with

glutamate ⁄ malate substrate during aging, no change
was observed using succinate. In the presence of
2,4-dinitrophenol, oxygen consumption rates were the
same in the presence of ADP (state 3) at each age.
Together, these results indicated age-related impair-
ment of mitochondrial function.
Determination of 6D12-detectable AGE content
in mitochondrial matrix proteins
We next investigated the levels of AGE-modified pro-
teins to determine whether dysfunction of the organelle
parallels the accumulation of modified mitochondrial
matrix proteins. For this purpose, we used monoclonal
anti-AGE IgG (6D12) in a competitive ELISA
using carboxymethyl-lysine (CML)-modified BSA as
standard to evaluate AGE-modified protein content
(Fig. 1). AGE content increased significantly, by 48%
from 12.01 ± 0.71 AUÆlg protein
)1
(n ¼ 8) in
3-month-old rats to 17.81 ± 1.83 AUÆlg protein
)1
(n ¼ 9) in 27-month-old rats (P<0.01). These data
indicated age-associated accumulation of AGE adducts
in mitochondrial matrix proteins.
Identification of glycated mitochondrial matrix
proteins by LC-MS
⁄
MS after 2D gel
electrophoresis and western blotting
2D gel electrophoresis and western blotting

Standard 2D gel electrophoresis of liver mitochondrial
matrix proteins from young and old rats was run in
parallel (Fig. 2). Electrophoregrams showed a typical
pattern of total protein, and more than a thousand
Table 1. Biochemical respiratory parameters in rat liver mitochondria from 3-month-old and 27-month-old rats. Oxygen consumption rates
were measured polarographically in the presence of either 10 m
M succinate or 5 mM glutamate ⁄ 5mM malate. State 3 respiration was deter-
mined after adding 310 nmol of ADP. State 4 respiration is the rate of O
2
consumption after depletion of ADP. Data represent the mean ±
SEM. n, number of animals. *P < 0.05; **P < 0.01 versus 3-month-old rats.
Succinate (n ¼ 7) Malate + glutamate (n ¼ 5)
3 months 27 months 3 months 27 months
State 4 (ng atom O min
)1
Æmg
)1
) 74 ± 7 47 ± 8* 37.1 ± 2.9 35.8 ± 4.6
State 3 (ng atom O min
)1
Æmg
)1
) 281 ± 24 167 ± 14** 153.3 ± 6.6 130.6 ± 4.5*
RCR 3.79 ± 0.36 3.55 ± 0.28 4.42 ± 0.2 3.81 ± 0.45
P ⁄ O 1.87 ± 0.09 1.91 ± 0.06 2.83 ± 0.06 2.84 ± 0.09
Glycation of glutamate dehydrogenase with aging M. Hamelin et al.
5950 FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS
spots were detected using 2D Elite Master Software.
Western blot analysis using anti-AGE IgG (Fig. 3) in
samples from both young and old rats revealed that

only a small number of proteins among the thousand
protein spots observed on gel stained with colloidal
Coomassie blue were targeted for glycation. Moreover,
AGE-modified proteins were already present, although
to a lesser extent, in samples from young rats. We
noted individual variations both in the nature of modi-
fied proteins and in their extent of labelling within
Fig. 1. Determination of 6D12-detectable AGE protein content in
liver mitochondrial matrix with aging. The AGE adduct content in
mitochondrial matrix proteins from 3-month-old and 27-month-old
rats was assayed by competitive ELISA using monoclonal anti-AGE
IgG (clone 6D12). The results are expressed as AUÆlg protein
)1
and
represent the mean ± SEM. **P < 0.01 versus 3-month-old rats.
Number of animals is given in parentheses above the bar graphs.
A
250
-
150
-
100
-
75
-
50
-
37
-
25

-
pI
310
B
-
-
-
-
-
-
-
Molecular Mass (kDa)
250
150
100
75
50
37
25
Molecular Mass (kDa)
pI
310
Fig. 2. 2D gel electrophoresis profile of liver
mitochondrial matrix proteins. Liver mito-
chondrial matrix proteins (150 lg) were sub-
jected to isoelectrofocusing and subsequent
SDS ⁄ PAGE electrophoresis under reducing
conditions. Gels containing samples from
(A) 3-month-old and (B) 27-month-old rats
were stained with colloidal Coomassie blue.

A
B
Fig. 3. 2D gel electrophoresis and western blotting analysis of liver
mitochondrial matrix proteins; identification of AGE-modified pro-
teins with aging. Samples (150 lg) were subjected to isoelectrofo-
cusing and subsequent SDS ⁄ PAGE under reducing conditions. Gels
were subjected to western blotting using monoclonal anti-AGE IgG
(clone 6D12) to detect AGE-modified proteins in samples from (A)
3-month-old and (B) 27-month-old rats. Matched proteins were
identified by tandem LC-MS ⁄ MS mass spectrometry as: (1) gluta-
mate dehydrogenase; (2) catalase; and (3) ornithine carbamoyl
transferase.
M. Hamelin et al. Glycation of glutamate dehydrogenase with aging
FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS 5951
each age group. Furthermore, we observed that anio-
nic isoforms appeared to be modified with age for
trails 1 and 3 (Fig. 3B). Only spots from trails of iso-
forms numbered 1–3 that exhibited a significantly
increased yield of modifications with aging in a repro-
ducible way were retained for further analysis
(Fig. 3A,B) and were matched with colloidal Coomas-
sie blue staining spots (Fig. 2B), which were subjected
to LC-MS ⁄ MS protein identification.
Protein identification
Proteins from selected spots were identified automati-
cally by a computer program. Three series of MS ⁄ MS
spectra were obtained (data not shown). At least two
spots for each trail of isoforms were identified with
sufficient peptide coverage (13–32%), and analysis led
to the characterization of three proteins in a narrow pI

range (7.15–9.12), with a molecular mass in the range
Table 2. Identification of proteins located in spots 1–3. Tryptic peptides from in-gel digestion were subjected to LC ⁄ MS ⁄ MS analysis as
described in the Experimental procedures. Peptide identification was evaluated using Xcorr scores that measure similarities between mass-
to-charge (m ⁄ z) ratios for fragment ions predicted from amino acid sequences and fragment ions observed in the MS ⁄ MS spectrum. Xcorr,
cross correlation scores; z, charge of the precursor ion; M*, oxidized methionine, C@, alkylated cysteine.
Spot No.
Protein UniProt
accession No.
Mass
(kDa) pI Peptides matches Covearge (%)
1 Glutamate
dehydrogenase 1,
mitochondrial precursor,
P10860
61.4 8.05 12 22
Sequence Position MH
+
DM z Xcorr
K.M*VEGFFDR.G 69–76 1016.45 0.44 2 2.02
R.RDDGSWEVIEGYR.A 124–136 1581.73 0.31 3 2.83
R.DDGSWEVIEGYR.A 125–137 1425.63 0.17 2 3.01
R.YSTDVSVDEVK.A 152–162 1241.59 0.27 2 2.55
K.ALASLM*TYK.C 163–171 1013.53 0.34 2 2.64
172–183 1219.61 0.29 2 2.65
K.KGFIGPGIDVPAPDM*STGER.E 212–231 2060.01 0.64 3 2.72
K.GFIGPGIDVPAPDM*STGER.E 213–231 1931.9 0.43 2 3.04
K.HGGTIPVVPTAEFQDR.I 481–496 1723.88 0.62 2 2.10
K.DIVHSGLAYTM*ER.S 504–516 1507.72 0.46 3 3.46
R.TAM*KYNLGLDLR.T 524–535 1410.74 0.53 3 3.29
K.YNLGLDLR.T 527–535 963.53 0.57 2 2.50

R.TAAYVNAIEK.V 536–545 1079.57 0.35 2 2.80
2 Catalase, P04762 59.6 7.15 6 13
Sequence Position MH
+
DM z Xcorr
K.LNIM*TAGPR.G 38–46 988.52 0.33 2 2.03
221–232 1337.65 0.37 2 2.52
R.LAQEDPDYGLR.D 252–262 1276.62 0.38 2 2.65
365–379 1803.92 0.53 2 2.07
R.FNSANEDNVTQVR.T 431–443 1493.70 0.44 2 4.38
K.DAQLFIQR.K 468–475 990.54 0.22 2 2.44
3 Ornithine
carbamoyltransferase,
mitochondrial, P00481
39.9 9.12 10 32
Sequence Position MH
+
DM z Xcorr
K.GRDLLTLK.N 39–46 915.56 0.51 2 2.17
K.GEYLPLLQGK.S 71–80 1117.63 0.36 2 2.42
R.VLSSM*TDAVLAR.V 130–141 1278.67 0.57 2 3.41
R.VYKQSDLDILAK.E 142–153 1392.77 0.53 2 2.01
K.FGM*HLQAATPK.G 211–221 1216.61 0.43 2 2.20
K.GYEPDPNIVK.L 222–231 1131.57 0.26 2 2.07
K.LSM*TNDPLEAAR.G 244–255 1333.64 0.20 2 3.23
R.LQAFQGYQVTM*K.T 278–289 1429.71 0.54 2 3.98
R.KPEEVDDEVFYSPR.S 307–320 1709.80 0.71 2 2.91
R.SLVFPEAENR.L 321–330 1161.60 0.62 2 2.87
Glycation of glutamate dehydrogenase with aging M. Hamelin et al.
5952 FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS

39.9–61.4 kDa (Table 2). The number of peptides lead-
ing to the identification of each protein varied between
six and 12. The proteins identified corresponded to
glutamate dehydrogenase (GDH) for trail 1, catalase
for trail 2 and ornithine carbamoyl transferase (OCT)
for trail 3. By contrast to catalase, a potent enzyme in
antioxidant defense, the other two enzymes (GDH and
OCT) involved in the mitochondrial urea cycle exhib-
ited a wide extent of modification at 27 months
compared to 3 months. Further investigations were
performed with GDH because this enzyme is relevant
due to its essential role in between the urea and TCA
cycles.
Age-associated increased glycation and
decreased activity of GDH
Using anti-GDH serum, we first checked by western
blotting that GDH content within the mitochondrial
matrix was unchanged with age (data not shown).
GDH was then immunoprecipitated from matrix sam-
ples and western blotting was carried out to investigate
the increased GDH glycation rate with aging (Fig. 4).
Spots of GDH protein exhibited identical intensity in
3-month-old (49.86 ± 3.10 AU) and 23-month-old
rats (54.98 ± 2.39 AU), whereas AGE-labelling inten-
sity significantly increased by 45%, from 32.71 ±
2.65 AU (n ¼ 4) in 3-month-old rats to 47.29 ±
3.67 AU (n ¼ 3) in 23-month-old rats (P<0.01)
(Fig. 4B). The rate of glycation expressed as the ratio
of AGE ⁄ GDH intensity was significantly increased by
30%, from 0.66 ± 0.05 (n ¼ 4) in 3-month-old rats to

0.86 ± 0.03 (n ¼ 3) in 23-month-old rats (P<0.01)
(Fig. 4B). These data indicate that GDH underwent
increased glycation with aging.
We next examined whether the glycation modifica-
tion demonstrated above was associated with impair-
ment of GDH function. For this purpose, we
measured GDH activity in mitochondrial matrix
extracts (50 lg of total protein) in the absence and
presence of allosteric effectors (ADP and GTP)
(Table 3). The results obtained revealed a significant
decrease (23%) in GDH activity with aging, from
2.47 ± 0.19 UÆmg
)1
in 3-month-old rats to 1.89 ±
0.07 UÆmg
)1
in 23-month-old rats ( n ¼ 4) (P ¼ 0.02).
In the presence of the ADP activator, this activity was
enhanced by 1.20-fold in young samples, although
slightly less (1.14-fold) in old samples. However, this
GDH activity markedly collapsed with the GTP inhibi-
tor, with residual activity barely representing 8% in
young samples compared to 12% in old samples
(P<0.01). These data suggest an inhibitory effect of
glycation on GDH activity, in addition to a slightly
altered response to allosteric regulation with aging.
Effect of MGO on purified GDH activity and its
allosteric regulation
To ascertain whether the glycation impact upon GDH
activity was due to modifications in lysine ⁄ arginine

A
B
Fig. 4. Immunochemical identification of glycated GDH from mitochondrial matrix. GDH from matrix samples (500 lg) was immunoprecipitat-
ed and samples resolved by SDS ⁄ PAGE electrophoresis. Two identical gels from each sample run in parallel were subjected to western
blotting using either monoclonal anti-AGE IgG or GDH polyclonal antibody to detect AGE modifications or GDH antigen content, respectively,
in samples from 3-month-old and 23-month-old rats (A). Blots were semiquantified by densitometry scanning and density expressed in
arbitrary units (AU). Results are presented as the ratio of the AGE ⁄ GDH rate (B). Values are the means ± SEM. **P<0.01 versus 3-month-
old rats.
M. Hamelin et al. Glycation of glutamate dehydrogenase with aging
FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS 5953
residues, we used MGO as a potent glycating agent
generated intracellularly to investigate the effect on
purified GDH.
In vitro MGO modification of GDH
A purified GDH sample (10 lg) was incubated with
varying concentrations of MGO for up to 24 h and
subjected to western blotting (Fig. 5) with monoclonal
anti-AGE IgG or polyclonal anti-GDH serum. After
24 h of incubation, the GDH protein was MGO-
derived AGE-modified and the extent of modification
increased with the concentration of MGO (Fig. 5A),
whereas the GDH immunolabelling intensity concur-
rently decreased (Fig. 5B). These results clearly demon-
strated that glycated enzyme partially lost its
antigenicity, indicating that some antigenic sites of
GDH could be masked by glycation adducts.
Effect of MGO on GDH activity and kinetic
parameters
In vitro MGO treatment of purified GDH resulted in
decreased activity with time of incubation and an

increasing MGO concentration (Table 4). With 1 mm
MGO, GDH activity significantly decreased at 30 min
of incubation, and markedly dropped by 37% within
5 h compared to control. This MGO effect on GDH
activity was compared with that induced by GO, an a-
dicarbonyl metabolite presumed to possess noxious
specificity like MGO. Treatment with GO led to a sig-
nificant inhibition effect with as little as 50 lm (38%),
whereas a high concentration (1 mm) led to strong
inhibition (67%), providing evidence that GDH activ-
ity is altered by glycation modification. Using amino
acid analysis, we noted that both arginine (24 : 30)
and lysine (22 : 32) residues were damaged in MGO-
treated GDH, whereas 22 and eight residues,
respectively, were damaged with GO, indicating that
MGO was equally noxious to lysine and arginine
residues, leading to the loss of part of its activity. In
addition, analysis of enzymatic parameters (V
max
and
K
m
) of GDH modified by 1 mm MGO for 24 h
using a-ketoglutarate as substrate (Fig. 6) showed
that MGO treatment altered only the maximum
velocity (V
max
), whereas the apparent K
m
value did not

change.
Effect of MGO on GDH allosteric regulation
We next analyzed the allosteric regulator effects on
purified GDH activity before and after treatment with
MGO. At indicated times, the activity of the enzyme
incubated with 1 mm MGO was measured in the pres-
ence of constant concentrations of allosteric effectors
(activators: 250 lm ADP or 10 mm leucine; inhibitor:
30 lm GTP). GDH activity at t
0
in the absence of ef-
fectors (28.6 UÆmg enzyme
)1
) was set at 100% activity.
As shown in Fig. 7, GDH activity spontaneously
decreased with time of incubation. This decline was
accentuated when GDH was incubated with 1 mm
MGO (Fig. 7A) and the activity ratio shifted from
0.85-fold to 0.71-fold at t
240
versus t
0
. In the presence
of an ADP effector (Fig. 7B), native GDH activity
exhibited a significant 1.66-fold increase (Fig. 7B, left
panel), whereas this increase fell to 1.44-fold when
GDH was preincubated with 1 mm MGO (Fig. 7B,
right panel). With leucine as an effector, we observed a
similar phenomenon (Fig. 7C); the activity increased
by 1.55-fold at 240 min with native enzyme (Fig. 7B,

left panel) and by 1.24-fold when the enzyme was
preincubated with MGO (Fig. 7B, right panel). These
Table 3. Age-related changes in GDH activity and allosteric effector
sensitivity. Liver mitochondrial matrix extracts (50 lg total protein)
from young (3-month-old) and old (23-month-old) rats were sub-
jected to the GDH activity assay in the direction of a-ketoglutarate
amination, in the absence (control) or presence of constant concen-
trations of allosteric effectors (activator: 250 l
M ADP; inhibitor:
10 l
M GTP). Values are expressed as specific activity (UÆmg
)1
)
(control) and activity determined in the presence of effectors is
given as a percentage of control in the absence of effectors, at
each age. Data represented the mean ± SEM (n ¼ 4). *P ¼ 0.02;
**P < 0.01 versus 3-month-old rats.
GDH activity
Age
3 months
(n ¼ 4)
23 months
(n ¼ 4)
GDH activity (no effector) 2.47 ± 0.19 1.89 ± 0.07*
% of GDH activity (+ ADP) 120.88 ± 4.75 114.71 ± 16.71
% of GDH activity (+ GTP) 8.23 ± 0.75 11.57 ± 0.59**
4321
EGA
A
4321

HDG
B
Fig. 5. Western blot detection of purified GDH modified by MGO.
Purified GDH (Sigma) samples (10 lg) were incubated with or with-
out varying MGO concentrations for 5 h and aliquots (1 lg) were
subjected to SDS ⁄ PAGE under reducing conditions. Gels were sub-
mitted to western blotting using either monoclonal anti-AGE IgG
(clone 6D12) to detect MGO-derived AGE in GDH samples (A) or
GDH polyclonal antibody to determine whether the GDH load was
preserved (B). Lane 1, control, 0 l
M MGO; lane 2, 50 lM MGO;
lane 3, 200 l
M MGO; lane 4, 1 mM MGO.
Glycation of glutamate dehydrogenase with aging M. Hamelin et al.
5954 FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS
results indicate that GDH stimulation induced by allo-
steric effectors was partly abolished when the enzyme
was previously treated with MGO. On the other hand,
the GTP inhibitor (Fig. 7D) exhibited an efficient
effect on GDH activity; the activity, which was barely
0.44-fold at 240 min (Fig. 7D, left panel), rose to 0.71-
fold when GDH was pretreated with MGO (Fig. 7D,
right panel), indicating the capacity of MGO to abro-
gate the inhibitory effect of GTP on GDH activity.
Together, these results clearly demonstrate the inhibi-
tory effect of MGO on GDH activity, which rein-
forced the decline of its activity with time. In addition,
this MGO modification deeply altered the responsive-
ness of GDH to its respective allosteric effectors, in
agreement with data observed in vivo.

Discussion
In the present study, we have demonstrated a decline in
mitochondrial respiratory chain activity upon aging
concomitant with an accumulation of AGE-modified
proteins in the mitochondrial matrix. These modifica-
tions affect several proteins that are targets for glyca-
tion damage and, although some of these proteins are
already modified at a young age (3-month-old rats), the
level of damage significantly increased with increasing
age (27-month-old rats). Among the glycated proteins,
trails of isoforms appeared, in agreement with modifi-
cations targeting the basic amino acids arginine and
lysine, subsequently turning off their ionic charge. In
support of this assertion, recent studies have shown
that glycation was associated with both loss of basic
groups and shifts in pK of the acidic groups, consistent
with a reduction in effective anionic charge [23,25]. The
identification of these proteins, which are increasingly
glycated with aging, revealed three enzymes, GDH,
OCT and catalase. Catalase, a crucial antioxidant
defense enzyme highly expressed in peroxisomes, and
also constitutively present in the heart and liver mito-
chondria [26–28], is maintained with GDH and OCT,
which belong to the urea cycle. GDH, in particular,
which markedly emerged as being modified at
27 months, plays a key role in connecting TCA to the
urea cycles, and exhibited a loss of activity and altera-
tions in its allosteric properties with aging. Interest-
ingly, all these modified proteins are different from
those preferentially found to be oxidized during aging

in the mitochondrial matrix, as previously identified
(i.e. aconitase [29] and adenine nucleotide translocase
[30]). To determine whether the age-related inhibition
of GDH activity demonstrated here was related to
glycation modification, purified GDH was treated with
the a-dicarbonyl metabolite MGO. Incubation of GDH
with this compound resulted in time-dependent inacti-
vation of the enzyme, consistent with the damaging
Table 4. Effect of MGO and GO concentrations on GDH activity. Purified GDH samples (10 lg) were incubated in 100 mM TEA-HCl buffer
pH 7.3 with varying concentrations (0.05, 0.200 and 1 m
M) of either MGO or GO as glycating agents for variable times. At indicated times,
enzyme activity was determined spectrophotometrically on aliquots (1 lg). Data are expressed as a percentage of GDH activity control at t
0
for
each concentration and represent the mean ± SEM (n, number of animals). *P < 0.05; **P < 0.01; ***P < 0.001 traited versus control at t
0
.
MGO [m
M]
GDH activity (n ¼ 4)
Incubation time
GO [m
M]
Incubation time
30 min 2 h 30 5 h 5 h
0 98.0 ± 2.2 91.6 ± 2.7 86.3 ± 6.0 0 82.4 ± 3.9
0.05 87.9 ± 5.2 94.2 ± 2.7 83.4 ± 6.1 0.05 51.2 ± 2.3**
0.2 94.0 ± 4.0 90.8 ± 3.4 73.7 ± 6.0 0.2 45.3 ± 1.9**
1 80.4 ± 5.7* 72.2 ± 5.4** 54.8 ± 7.2** 1 27.5 ± 1.5***
Fig. 6. Effect of MGO modification on GDH parameters V

max
and
K
m
. Kinetic analyses were performed to determine the effect of
MGO (1 m
M) on GDH enzymatic parameters V
max
and K
m
. Purified
GDH (Sigma) samples (10 lg) were incubated with or without
MGO for 24 h and aliquots (1 lg) were taken for enzymatic assay.
Lineweaver–Burk plots were used with a-ketoglutarate as substrate
(S) in native (1) or MGO-treated (2) GDH.
M. Hamelin et al. Glycation of glutamate dehydrogenase with aging
FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS 5955
effect of glycation. Interestingly, kinetic analysis of
modified GDH showed that treatment with MGO
reduced only the maximum velocity without affecting
K
m
, indicating that MGO-modified GDH is inacti-
vated. In addition, amino acid analysis performed on
MGO- and GO-treated GDH revealed that lysine resi-
dues were more sensitive to MGO than to GO modifi-
cations, whereas arginine was equally sensitive to both
dicarbonyl compounds, suggesting that GDH inactiva-
tion was at least partly due to MGO-lysine ⁄ arginine
modifications, in accordance with the observed trails of

isoforms in the 2D electrophoregram. Indeed, numer-
ous data have claimed that MGO modifications of criti-
cal arginine ⁄ lysine residues cause structural distortion,
leading to enzyme inactivation [23,31]. In addition,
recent data on GDH studies indicate that among
33 lysine residues constitutive of its primary sequence,
the prominent lysine 126 directly interacts with the
a-carbon constituent on the substrate [32], suggesting
that MGO modification affects enzymatic activity. In
support of this assertion, a lysine residue involved in
inactivation of brain GDH isoproteins by O-phthal-
aldehyde has been identified [33]. Moreover, loss of
GDH activity was reported under multiple system atro-
phy conditions in which GDH activity was decreased
to a greater extent than other mitochondrial enzymes
[34] indicating that GDH is more sensitive to insults.
Interestingly, a recent in vitro study showed that incu-
bation of mitochondria with MGO led to rapid inhibi-
tion of mitochondrial respiratory rates through
particular protein target modifications [35]. Both the
TCA cycle and the electron respiratory chain were
inhibited, indicating a link between mitochondrial
MGO modified enzymes and altered function. Mamma-
lian GDH is allosterically regulated by a number of
small molecules [36] and its regulation is of particular
biological importance, as exemplified by the observa-
tion that some regulatory mutations of the gene for
GDH are associated with severe clinical manifestation
in children [32]. In addition, as shown in a recent study
Control

0
20
40
60
80
100
120
140
160
180
01
MGO (m
M)
Activity (%)
0
15
120
240
ADP 250 µM
0
20
40
60
80
100
120
140
160
180
0

MGO (m
M)
Activity (%)
Leu 10 mM
0
20
40
60
80
100
120
140
160
180
0
MGO (m
M)
Activity (%)
GTP 30 µM
0
20
40
60
80
100
120
140
160
180
01

MGO (m
M)
Activity (%)
B
C
D
A
1
1
1
**
**
****
***
# # #
# # #
**
**
**
**
# # #
# # # #
****
****
****
****
# # # #
# # # #
# # # #
# # # #

# # #
# # #
Fig. 7. MGO effect on GDH activity and allosteric regulation. Puri-
fied GDH samples (10 lg; Sigma) was incubated with 1 m
M MGO
for up to 240 min and aliquots (1 lg) were subjected to the GDH
activity assay in the absence (A) or presence of constant concentra-
tions of allosteric activators ADP (B) and leucine (C), or inhibitor
GTP (D). Values expressed as specific activity (UÆmg
)1
) are given as
percentage of control (activity determined in the absence or pres-
ence of MGO at t
0
; left and right panels, respectively) in each
range. Data represented the mean ± SEM (n ¼ 6). *t-test: effector
versus control at each time; #t-test: MGO 1 m
M-treated versus
nontreated MGO at each time. °Equivalent to * or #; °P < 0.05;
°°P < 0.01; °°°P < 0.005; °°°°P < 0.001.
Glycation of glutamate dehydrogenase with aging M. Hamelin et al.
5956 FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS
[37], the arginine side chain at position 463 of GDH is
thought to be involved in ADP allosteric activation
because the R463A mutant form of this enzyme is
insensitive to ADP stimulation. In the present study,
MGO modifications altered the allosteric regulation
properties of the enzyme, suggesting that these effects
are not only due to a change in charge profile, but also
in the conformation of the molecule resulting from gly-

cation of the charged arginine ⁄ lysine side-chain resi-
dues. The fact that arginine ⁄ lysine residues are targets
of glycation [38,39] suggests that the modifications
observed in the present study involve some of these res-
idues, leading to an impairment of allosteric regulation
and the catalytic properties of the enzyme.
GDH is important in converting free ammonia and
a-ketoglutatrate to glutamate; it utilizes nicotinamide
nucleotide cofactor NAD
+
for nitrogen liberation and
NADP
+
for nitrogen incorporation; however, it
should be recognized that the reverse reaction is a key
anapleurotic process linking amino acid metabolism
with TCA cycle activity. In a reverse reaction, GDH
provides an oxidizable carbon source for the produc-
tion of energy, as well as the reduced electron carrier
NADH. Thus, GDH is considered to be significant not
only because it catalyzes a reaction directly connected
to the TCA cycle, but also because of the pivotal posi-
tion in metabolism occupied by both glutamate and a-
ketoglutarate as a result of their ability to enter into
many metabolic pathways [40]. Accordingly, the build-
up of an inactive form of GDH demonstrated in the
present study could contribute to a decreased produc-
tion of a-ketoglutarate and a diminished flux through
the TCA cycle, which might be at least partly be
responsible for impairment of mitochondrial function

with advanced age, as demonstrated by the decrease in
respiration driven by the glutamate–malate substrate.
In summary, the results obtained in the present study
demonstrate that age-related impairment of mitochon-
drial respiration runs parallel to an accumulation of
AGE-modified matrix proteins. Identification of selec-
tively glycated proteins revealed that two of these are
key urea cycle enzymes, among which GDH was the
main target protein and showed a loss of both activity
and sensitivity to allosteric effectors with aging. In vitro
alterations in both allosteric regulation and catalytic
properties of this enzyme by the glycating agent MGO
during short-term incubation support the notion of the
dysfunctional power of intracellular glycation. In line
with the central role played by this enzyme in cellular
metabolism and energy homeostasis, we hypothesize
that AGE modifications of GDH may contribute, at
least in part, to a defect in mitochondria with aging and
could be used as a biomarker of cellular aging.
Experimental procedures
Animals
Experiments were performed on male Wistar rats (WAG ⁄ Rij)
born and raised in the animal care facilities of the Commissar-
iat a
`
l’Energie Atomique (CEA, Gif-sur-Yvette, France). This
strain remains lean even when fed ad libitum and does not
suffer from age-associated nephropathy, hypertension or
diabetes [41]. Cohorts were constituted of young adult
(3-month-old) and senescent (27-month-old) animals. All stu-

dies were conducted in accordance with the animal care policy
of national and European regulations.
Chemicals
GDH (bovine liver; EC 1.4.1.3) a nd horseradish peroxidase-
conjugated anti-(mouse IgG) or anti-(rabbit IgG) sera were
purchased from Sigma Chemicals (Saint Quentin Fallavier,
France). GDH was dissolved in 100 mm of triethanolamine,
pH 7.3. Monoclonal antibody to AGE (clone no. 6D12) from
Trans Genic Inc. (Kumamoto, Japan) shows cross-reaction
both to CEL and CML [ 42]. Polyclonal antibody against GDH
was obtained from Interchim (Montluc¸ on, France) and pro-
tein G-agarose bed (ImmunoPure immobilized proteinG Plus)
from Pierce (Perbio Science Company, Brebie
`
res, Franc e).
Isolation of mitochondria
A 10% (w ⁄ v) tissue homogenate was prepared using a Pot-
ter apparatus in an ice-cold medium containing 220 mm
mannitol, 70 mm sucrose, 0.1 mm EDTA and 2 mm Hepes,
pH 7.4, supplemented with 0.5% BSA (w ⁄ v). Nuclei and
cellular debris were pelleted by centrifugation for 10 min at
800 g and 4 °C. Supernatant was centrifuged at 8000 g for
10 min at 4 °C. The mitochondrial pellet was then washed
three times with the homogenization medium and used for
polarographic measurements.
To prepare mitochondrial matrix extract, mitochondria
were suspended in 50 mm Tris ⁄ HCl, pH 7.9, then dis-
rupted by sonication (four times for 10 s). The resulting
suspension was centrifuged at 15 000 g for 10 min and
then at 100 000 g for 45 min at 4 °C. The supernatant

(containing matrix proteins) was stored at )80 °C for fur-
ther analysis of AGE-modified proteins. The protein con-
centration was assessed using a Bradford protein assay
(Biorad, Mu
¨
nchen, Germany).
To estimate contamination of mitochondrial preparation
with lysosomes, we used acid phosphatase activity as a
marker.
Measurements of mitochondrial respiration
Oxygen consumption was measured polarographically with
a Clark electrode in the sample, as described by Aprille
M. Hamelin et al. Glycation of glutamate dehydrogenase with aging
FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS 5957
and Asimakis [43], in a thermostatically controlled closed
2 mL chamber (30 °C). The rate of oxygen consumption
was measured in the presence of 310 nmol of ADP and
10 mm of succinate or 5 mm glutamate ⁄ 5mm malate
(state 3) and after all ADP had been consumed (state 4 or
resting state). Oxygen consumption rates are expressed as
ng atoms of oxygen consumedÆmin
)1
Æmg protein
)1
. The
rate of oxygen consumption in state 3 and in state 4,
RCR (the ratio of state 3 to state 4 respiration), an index
of electron transport chain activity, and the ADP ⁄ O
2
ratio

were calculated. Oxygen consumption in the presence of
40 lm of dinitrophenol (uncoupled state) was also
checked.
Determination of 6D12-detectable AGE content in
mitochondrial matrix proteins by competitive
ELISA
The ELISA assay was conducted as previously described
[24]. Briefly a 96-well microtiter Nunc-immuno plate
(Nunc, Roskilde Denmark) was coated with 100 lLof
CML-BSA (6.4 nmol CMLÆmL
)1
) by incubation overnight
at 4 °C. Wells were washed with NaCl ⁄ P
i
)0.05%
Tween 20 (v ⁄ v) (buffer A) and free binding sites were
blocked by incubation for 1 h at room temperature with
100 lL of NaCl ⁄ P
i
)6% skimmed milk or NaCl ⁄ P
i
)1%
BSA (w ⁄ v). After washing with buffer A, 50 lL of com-
peting antigen (test samples at 0.1 lgÆlL
)1
or serial dilu-
tions of standard CML-BSA from 0.64 mm to 128 mm)
was added, followed by 50 lL of monoclonal anti-AGE
IgG (clone 6D12) (dilution 1 : 3000). The plate was incu-
bated for 2 h at room temperature, washed and then

incubated with 50 lL horseradish peroxidase-conjugated
anti-(mouse IgG) (dilution 1 : 10 000) for 2 h at room
temperature. The wells were washed, then 100 lL of sub-
strate solution (40 mm ABTS and 200 lL of 30% hydro-
gen peroxide in 20 mL sodium acetate-phosphate buffer,
pH 7.2) were added per well and incubated. Absorbance
(A) was measured at 405 nm on a micro-ELISA plate
reader (Spectra Rainbow, SLT. Labinstruments, Salzburg,
Austria). Results are expressed as the ratio B ⁄ Bo (bound ⁄
total), calculated as: experimental A ) background A (no
antibody) ⁄ total A (no competitor) ) background A, versus
CML added, as pmol CMLÆlg protein
)1
. Finally, data
were expressed as arbitrary unitsÆlg protein
)1
(AUÆlg pro-
tein
)1
) because anti-AGE IgG recognizes both CML and
CEL.
The use of 6% skimmed milk (w ⁄ v) as an alternative to
1% BSA (w ⁄ v) as a blocking agent in the immunochemical
assay introduced an increment of less than 10% at the
CML level, with GO-modified BSA used as standard (data
not shown). We took advantage of these results and used
skimmed milk in all further immunochemical assays
(ELISA and western blotting).
1D and 2D gel electrophoresis of mitochondrial
matrix proteins and western blotting

Mitochondrial matrix protein samples (150 lg) from 3-
month-old and 27-month-old rats were mixed with 200 l L
of 2D sample buffer (7 m urea, 2 m thiourea, 4% Chaps,
1% dithithreitol, 2% Pharmalytes, Amersham Biosciences,
Saclay, France; pH 3.0–10.0). The strips were allowed to
rehydrate overnight. 1D isoelectric focusing was performed
on Immobiline Drystrips (Amersham Biosciences; pH 3.0–
10.0, 13 cm) in a Multiphor II device (Amersham Bio-
sciences) for 49 325 Vh. After electrofocusing, immobilines
were prepared for SDS ⁄ PAGE and 2D SDS ⁄ PAGE was
run vertically on a 12% polyacrylamide gel using the cool-
ing Protean II system (Bio-Rad, Marne La-Coquette,
France). The gels were either fixed and stained with colloi-
dal Coomassie blue for total protein pattern and LC-
MS ⁄ MS analysis, or western blotted onto a nitrocellulose
membrane (Bio-Rad) overnight at 30 V. The membrane
was saturated with NaCl ⁄ P
i
, pH 7.4, 0.1% Tween 20 (v ⁄ v),
5% skimmed milk (w ⁄ v) overnight at 4 °C, followed by
four washes (10 min each) with NaCl ⁄ P
i
, pH 7.4, 0.2%
Tween 20 (washing buffer). The membrane was then incu-
bated for 2 h at room temperature with monoclonal anti-
AGE IgG clone 6D12 (dilution 1 : 3000) in NaCl ⁄ P
i
, 0.1%
Tween 20 (v ⁄ v), washed four times, incubated for 1 h with
anti-(mouse IgG) coupled to horseradish peroxidase (dilu-

tion 1 : 3000) and given a final wash. The proteins were
revealed with a SuperSignal West Pico chemiluminescent
reagent (Perbio Science Company, Brebie
`
res, France).
Protein identification by LC-MS
⁄
MS
Colloidal Coomassie blue-stained spots matching with
bands immunolabelled by monoclonal anti-AGE IgG were
excised from gel, cut into 1 mm pieces and then treated for
LC-MS ⁄ MS analysis. Gel pieces were washed twice in
100 mm ammonium bicarbonate buffer pH 8.8 and then
dehydrated with acetonitrile. The gel pieces were rehydrated
in 10 mm dithithreitol ⁄ ammonium bicarbonate solution and
proteins alkylated with 50 mm iodoacetamide. After dehy-
dratation with acetonitrile, gel pieces were rehydrated on
ice for 10 min in 20 lLof20mm ammonium bicarbonate
containing 50 ngÆlL
)1
of sequence-grade modified porcine
trypsin (Promega, Madison, WI, USA); then supernatants
were replaced by 20 lLof20mm ammonium bicarbonate,
and in-gel digestion was performed for 15 h at 37 °C. The
resulting peptides were extracted twice with 20 l Lof
20 mm ammonium bicarbonate and then three times in
20 lL of 0.5% trifluoroacetic acid in 50% acetonitrile. The
peptide extracts were concentrated to 20 lL using an RC
10.22 evaporator concentrator (Jouan, Saint Herblain,
France). Samples were then subjected to mass spectrometry

analysis.
Glycation of glutamate dehydrogenase with aging M. Hamelin et al.
5958 FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS
LC-MS ⁄ MS analyses were performed with an electro-
spray ion-trap mass spectrometer (LCQ Advantage; Ther-
moElectron, Courtaboeuf, France) coupled on-line with
a Surveyor HPLC system (ThermoElectron, Courtaboeuf,
France). A 150 mm · 0.18 mm ThermoElectron HyPurity
C18 column (5 lm particle diameter; 190 A
˚
pore size) with
a mobile phase of solvent A [0.1% (v ⁄ v) formic acid in
water] and solvent B [0.1% (v ⁄ v) formic acid in 80% (v ⁄ v)
acetonitrile] was used with a linear gradient of 2–60% of
mobile phase B over 60 min at a flow rate of 250 lLÆmin
)1
.
The flow was split and 2 lLÆmin
)1
was directed to the col-
umn. The electrospray needle was operated with a voltage
of 2.5 kV, and the heated desolvation capillary was held at
180 °C. Nitrogen was used as the sheath gas. All scans were
acquired in positive ion mode. The mass spectrometer oper-
ated in a data-dependent MS ⁄ MS mode. The top three
most intense ions were selected from the full MS scan (m ⁄ z
range 300–2000) for MS
2
analysis. The isolation width of
the parent ions was set to 2 m ⁄ z units with 35% normal-

ized collision energy. An m ⁄ z ratio for an ion that had been
selected for fragmentation was placed on a list and dynami-
cally excluded from further fragmentation for 1 min.
Proteins were identified automatically by the computer
program bioworks, release 3.3 (Thermo Electron, San Jose,
CA, USA) which correlated experiment MS ⁄ MS spectra
against theoretical tandem mass spectra from the protein
database Uniprot (02 ⁄ 06 ⁄ 06) using the sequest algorithm.
Alkylated cysteines and oxidized methionines were exam-
ined as differential modifications. The mass tolerance for
both peptides and fragment ions was 1.0 amu. Peptide
identification was further evaluated using Xcorr (cross cor-
relation) scores that measure similarities between mass-
to-charge ratios (m ⁄ z) for fragment ions predicted from
amino acid sequences obtained from the database and frag-
ment ions observed in the MS ⁄ MS spectrum. As a general
rule, Xcorr values of greater than 1.5 for a single charged
ion, 2.0 for a doubly charged ion and 2.5 for a triply
charged ion were accepted as positive identification.
Immunochemical identification of glycated GDH
from mitochondrial matrix
Liver mitochondrial matrix from young and old rats was
immunoprecipitated to isolate GDH and subjected to wes-
tern blotting for subsequent detection of the glycation pro-
file. Samples (500 lg) in 400 lL of immunoprecipitation
buffer (50 mm Tris-acetate, pH 7.5, 120 mm NaCl, 1 mm
phenylmethanesulfonyl fluoride, 1 mm EDTA, 1%
Tween 20) were incubated with anti-GDH serum (5 lg) for
1 h at room temperature on a rotative wheel. Then, anti-
body and bound proteins were precipitated with 10 lLof

protein G-agarose beads (Pierce) for 1 h with gentle rota-
tion. Protein complexes were pelleted by centrifugation
(1000 g for 1 min) then washed three times with washing
buffer (50 mm Tris-acetate, pH 7.5, 120 mm NaCl, 0.1%
Tween 20, 0.1% SDS) and pellets were boiled in
SDS ⁄ PAGE sample buffer (50 lL) to recover antigen and
antibody. Following centrifugation, the resulting superna-
tants of all samples were separated into two equal aliquots
prior to SDS ⁄ PAGE electrophoresis and western blotting.
Each couple of membranes was incubated with either GDH
polyclonal antibody or monoclonal anti-AGE IgG as
described below. Blots were semiquantified by densitometry
scanning using image master 1d software (Amersham
Biosciences) (in arbitrary units: AU) and finally expressed
as the ratio of the AGE ⁄ GDH rate.
Treatment of purified glutamate dehydrogenase
with MGO or GO
Purified GDH (Sigma) (10 lg) was incubated with varying
concentrations of MGO or GO (0.05 mm, 0.2 mm,1mm)
in 100 mm TEA-HCl pH 7.3 for up to 24 h at 25 °C. These
two a-dicarbonyl compounds endogenously formed in phys-
iological systems are important glycating agents [20]. Trea-
ted samples were used either for enzyme activity
determination and amino acid analysis to identify the target
of modification or submitted to western blotting after
SDS ⁄ PAGE electrophoresis as described above using anti-
GDH serum (dilution 1 : 20 000) or monoclonal anti-AGE
IgG (dilution 1 : 3000) to detect AGE modifications.
Amino acid analysis of treated GDH was performed
after hydrolysis in 6 m HCl 0.2% phenol (Laboratoire de

Microsequenc¸ age des Prote
´
ines, Institut Pasteur, Paris,
France). There were 30 arginine and 32 lysine residues in
native GDH; MGO-modified GDH contained only six argi-
nine and ten lysine residues, whereas GO modified GDH
remained at eight and 24 residues, respectively, after 5 h of
incubation.
Measurement of GDH activity: effect of glycation
Liver mitochondrial matrix extracts (50 lg total protein)
from young (3-month-old) and old (23-month-old) rats
were used for GDH activity assay. To investigate the
impact of glycation upon GDH activity, aliquots (1 lg) of
MGO- or GO-treated purified GDH (see above) were taken
at indicated times and immediately used for the activity
assay. The assay was conducted in the direction of a-keto-
glutarate amination. The reaction mixture of 3 mL con-
tained 90 mm TEA-HCl, pH 7.4, 53 mm ammonium
acetate, 60 lm NADH, 250 lm EDTA, at 25 °C according
to the Sigma-Aldrich procedure. The enzyme reaction was
initiated by adding a-ketoglutarate to a final concentration
of 13 mm in the absence or presence of allosteric effectors
(activators: ADP, leucine; inhibitor: GTP). The activity of
the reaction mixture devoid of ammonium ions was taken
as blank. The decrease in absorbance was recorded during
the first 30 s (steady state) at 340 nm with a Uvikon
922 spectrophotometer (Kontron Instrument, Neufahrn,
M. Hamelin et al. Glycation of glutamate dehydrogenase with aging
FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS 5959
Germany) and the results were expressed as the ratio

(DA
340 nm
Æmin
)1
experimental ) DA
340 nm
Æmin
)1
Blank)(3) ⁄
(6.22)(0.1) (unitsÆmL enzyme
)1
) and then reported as spe-
cific activity (unitsÆmg protein
)1
). One unit will reduce
1.0 lmol of a-ketoglutarate to l-glutamate per min at
pH 7.4 and 25 °C in the presence of ammonium ions. All
assays were performed in triplicate.
Kinetic analyses were performed to determine the effect
of MGO on purified GDH kinetic parameters V
max
and
K
m
. GDH samples (10 lg) were incubated in the absence or
presence of 1 mm MGO for 24 h and aliquots (1 lg) dis-
solved in reaction mixture for enzymatic assay, as described
above. Several sets of experiments were performed using
varying concentrations of a-ketoglutarate as substrate
(16.25–2600 lm) and the results were represented according

to Lineweaver–Burk plots.
Statistical analysis
The results are presented as the mean ± SEM and differ-
ences between groups were assessed using Student’s unpaired
t-test. P < 0.05 was considered statistically significant.
Acknowledgements
The financial support of the French Ministry of
Research and of the EC ⁄ FP6 Proteomage grant
(518230) is gratefully acknowledged.
References
1 Suh JH, Heath SH & Hagen TM (2003) Two subpopu-
lations of mitochondria in the aging rat heart display
heterogenous levels of oxidative stress. Free Radic Biol
Med 35, 1064–1072.
2 Berlett BS & Stadtman ER (1997) Protein oxidation in
aging, disease, and oxidative stress. J Biol Chem 272,
20313–20316.
3 Hagen TM, Yowe DL, Bartholomew JC, Wehr CM,
Do KL, Park JY & Ames BN (1997) Mitochondrial
decay in hepatocytes from old rats: membrane potential
declines, heterogeneity and oxidants increase. Proc Natl
Acad Sci USA 94, 3064–3069.
4 Chance B, Sies H & Boveris A (1979) Hydroperoxide
metabolism in mammalian organs. Physiol Rev 59, 527–
605.
5 Shigenaga MK, Hagen TM & Ames BN (1994) Oxida-
tive damage and mitochondrial decay in aging. Proc
Natl Acad Sci USA 91, 10771–10778.
6 Sohal RS, Ku HH, Agarwal S, Forster MJ & Lal H
(1994) Oxidative damage, mitochondrial oxidant genera-

tion and antioxidant defenses during aging and in
response to food restriction in the mouse. Mech Aging
Dev 74, 121–133.
7 Martinez M, Hernandez AI, Martinez N & Ferrandiz
ML (1996) Age-related increase in oxidized proteins
in mouse synaptic mitochondria. Brain Res 731, 246–
248.
8 Sastre J, Pallardo FV, Pla R, Pellin A, Juan G, O’Con-
nor JE, Estrela JM, Miquel J & Vina J (1996) Aging of
the liver: age-associated mitochondrial damage in intact
hepatocytes. Hepatology 24, 1199–1205.
9 Smith RA, Kelso GF, Blaikie FH, Porteous CM, Ledg-
erwood EC, Hughes G, James AM, Ross MF, Asin-
Cayuela J, Cocheme HM et al. (2003) Using mitochon-
dria-targeted molecules to study mitochondrial radical
production and its consequences. Biochem Soc Trans 31,
1295–1299.
10 Miquel J, Ferrandiz ML, De Juan E, Sevila I & Marti-
nez M (1995) N-acetylcysteine protects against age-
related decline of oxidative phosphorylation in liver
mitochondria. Eur J Pharmacol 292, 333–335.
11 Kwong LK & Sohal RS (2000) Age-related changes in
activities of mitochondrial electron transport complexes
in various tissues of the mouse. Arch Biochem Biophys
373, 16–22.
12 Lenaz G, Bovina C, D’Aurelio M, Fato R, Formiggini
G, Genova ML, Giuliano G, Merlo Pich M, Paolucci
U, Parenti Castelli G et al. (2002) Role of mitochondria
in oxidative stress and aging. Ann NY Acad Sci 959,
199–213.

13 Ames BN & Shigenaga MK (1992) Oxidants are a
major contributor to aging. Ann NY Acad Sci 663,
85–96.
14 Stadtman SR (1992) Protein oxidation and aging.
Science 257, 1220–1224.
15 Pamplona R, Requena JR, Portero-Otin M, Prat J,
Thorpe SR & Bellmunt MJ (1998) Carboxymethylated
phosphatidylethanolamine in mitochondrial membranes
of mammals-evidence for intracellular lipid glycoxida-
tion. Eur J Biochem 255, 685–689.
16 Kang Y, Edwards LG & Thornalley PJ (1996) Effect of
methylglyoxal on human leukaemia 60 cell growth:
modification of DNA G1 growth arrest and induction
of apoptosis. Leuk Res 20, 397–405.
17 Giardino I, Edelstein D & Brownlee M (1996) BCL-2
expression or antioxidants prevent hyperglycemia-
induced formation of intracellular advanced glycation
endproducts in bovine endothelial cells. J Clin Invest 97,
1422–1428.
18 Nishikawa T, Du Edelstein DXL, Yamagishi S, Mat-
sumura T, Kaneda Y, Yorek MA, Beebe D, Oates PJ,
Hammes HP, Giardino I et al. (2000) Normalizing mito-
chondrial superoxide production blocks three pathways
of hyperglycaemic damage. Nature 404, 787–790.
19 Wells-Knecht KJ, Brinkmann E, Wells-Knecht MC,
Litchfield JE, Ahmed MU, Reddy S, Zyzak DV, Thorpe
SR & Baynes JW (1996) New biomarkers of Maillard
Glycation of glutamate dehydrogenase with aging M. Hamelin et al.
5960 FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS
reaction damage to proteins. Nephrol Dial Transplant

11, 41–47.
20 Thornalley PJ, McLellan AC, Lo TW, Benn J & Sonk-
sen PH (1996) Negative association between erythrocyte
reduced glutathione concentration and diabetic compli-
cations. Clin Sci (London) 91, 575–582.
21 Lo TW, Westwood ME, McLellan AC, Selwood T &
Thornalley PJ (1994) Binding and modification of pro-
teins by methylglyoxal under physiological conditions.
A kinetic and mechanistic study with N alpha-acetylarg-
inine, N alpha-acetylcysteine, and N alpha-acetyllysine,
and bovine serum albumin. J Biol Chem 269, 32299–
32305.
22 Speer O, Morkunaite-Haimi S, Liobikas J, Franck M,
Hensbo L, Linder MD, Kinnunen PK, Wallimann T &
Eriksson O (2003) Rapid suppression of mitochondrial
permeability transition by methylglyoxal. Role of revers-
ible arginine modification. J Biol Chem 278 , 34757–
34763.
23 Ahmed N, Dobler D, Dean M & Thornalley PJ (2005)
Peptide mapping identifies hotspot site of modification
in human serum albumin by methylglyoxal involved in
ligand binding and esterase activity. J Biol Chem 280,
5724–5732.
24 Bakala H, Delaval E, Hamelin M, Bismuth J, Borot-
Laloi C, Corman B & Friguet B (2003) Changes in rat
liver mitochondria with aging. Lon protease-like reactiv-
ity and N(epsilon)-carboxymethyllysine accumulation in
the matrix. Eur J Biochem 270, 2295–2302.
25 Winlove CP, Parker KH, Avery NC & Bailey AJ (1996)
Interactions of elastin and aorta with sugars in vitro

and their effects on biochemical and physical properties.
Diabetologia 39, 1131–1139.
26 Radi R, Turrens JF, Chang LY, Bush KM, Crapo JD
& Freeman BA (1991) Detection of catalase in rat heart
mitochondria. J Biol Chem 266, 22028–22034.
27 Fountoulakis M, Berndt P, Langen H & Suter L (2002)
The rat liver mitochondria proteins. Electrophoresis 23,
311–328.
28 Salvi M, Battaglia V, Brunati AM, La Rocca N, Tibaldi
E, Pietrangeli P, Marcocci L, Mondovi B, Rossi CA &
Toninello A (2007) Catalase takes part in rat liver mito-
chondria oxidative stress defence. J Biol Chem 282,
24407–24415.
29 Yan LJ, Levine RL & Sohal RS (1997) Oxidative
damage during aging targets mitochondrial aconitase.
Biochemistry 94, 11168–11172.
30 Yan LJ & Sohal RS (1998) Mitochondrial adenine
nucleotide translocase is modified oxidatively
during aging. Proc Natl Acad Sci USA 95, 12896–12901.
31 Goldberg AL (2003) Protein degradation and protection
against misfolded or damaged proteins. Nature 426,
895–899.
32 Smith TJ, Peterson PE, Schmidt T, Fang J & Stanley
CA (2001) Structures of bovine glutamate dehydroge-
nase complexes elucidate the mechanism of purine regu-
lation. J Mol Biol 307, 707–720.
33 Ahn JY, Choi S & Cho SW (1999) Identification of
lysine residue involved in inactivation of brain gluta-
mate dehydrogenase isoproteins by O-phthalaldehyde.
Biochimie 81, 1123–1129.

34 Sorbi S, Piacentini S, Fani C, Tonini S, Marini P &
Amaducci L (1989) Abnormalities of mitochondrial
enzymes in hereditary ataxias. Acta Neurol Scand 80,
103–110.
35 Rosca MG, Monnier VM, Szweda LI & Weiss MF
(2002) Alterations in renal mitochondrial respiration in
response to the reactive oxoaldehyde methylglyoxal. Am
J Physiol Renal Physiol 283, F52–F59.
36 Banerjee S, Schmidt T, Fang J, Stanley CA & Smith TJ
(2003) Structural studies on ADP activation of mamma-
lian glutamate dehydrogenase and the evolution of regu-
lation. Biochemistry
42, 3446–3456.
37 Fang J, Hsu BY, MacMullen CM, Poncz M, Smith
TJ & Stanley CA (2002) Expression, purification and
characterization of human glutamate dehydrogenase
(GDH) allosteric regulatory mutations. Biochem J 363,
81–87.
38 Ikeda K, Higashi T, Sano H, Jinnouchi Y, Yoshida M,
Araki T, Ueda S & Horiuchi S (1996) N (epsilon)-(carb-
oxymethyl) lysine protein adduct is a major immuno-
logical epitope in proteins modified with advanced
glycation end products of the Maillard reaction.
Biochemistry 35, 8075–8083.
39 Ahmed MU, Brinkmann Frye E, Degenhardt TP,
Thorpe SR & Baynes JW (1997) N-epsilon-(carboxyeth-
yl)lysine, a product of the chemical modification of pro-
teins by methylglyoxal, increases with age in human lens
proteins. Biochem J 324, 565–570.
40 Broca C, Brennan L, Petit P, Newsholme P & Maech-

ler P (2003) Mitochondria-derived glutamate at the
interplay between branched-chain amino acid and glu-
cose-induced insulin secretion. FEBS Lett 545, 167–
172.
41 Corman B & Michel JB (1987) Glomerular filtration,
renal blood flow, and solute excretion in conscious
aging rats. Am J Physiol 253, R555–R560.
42 Koito W, Araki T, Horiuchi S & Nagai R (2004) Con-
ventional antibody against Nepsilon-(carboxymethyl)
lysine (CML) shows cross-reaction to Nepsilon-(carb-
oxyethyl) lysine (CEL): immunochemichal quantification
of CML with a specific antibody. J Biochem (Tokyo)
136, 831–837.
43 Aprille JR & Asimakis GK (1980) Postnatal develop-
ment of rat liver mitochondria: state 3 respiration,
adenine nucleotide translocase activity, and the net
accumulation of adenine nucleotides. Arch Biochem
Biophys 201, 564–575.
M. Hamelin et al. Glycation of glutamate dehydrogenase with aging
FEBS Journal 274 (2007) 5949–5961 ª 2007 The Authors Journal compilation ª 2007 FEBS 5961