Ann Ist Super Sanità 2002;38(4):393-399
Advanced glycosylation end product quantification:
differently produced polyclonal antisera do not share
the recognition of epitopes of different nature
Angela Maria BUONGIORNO (a), Elisabetta SAGRATELLA (a), Stefania MORELLI (a),
Antonio DI VIRGILIO (b) and Maurizio SENSI (c)
(a) Laboratorio di Biochimica Clinica; (b) Servizio di Sperimentazione Animale,
Istituto Superiore di Sanità, Roma
(c) Dipartimento di Scienze Cliniche, Università degli Studi “La Sapienza”, Roma
Summary. - Advanced glycosylation end products (AGE) which are probably involved in the
pathogenesis of diabetic complications, comprise a series of related chemical structures. Thus different
antisera might recognize particular AGE epitopes rather than the complete range of epitopes. To test this
hypothesis, two antisera were raised using different immunization techniques and different AGE-carrier
proteins as immunogens. The antisera reactivity towards different AGE-proteins under various experimental
conditions was compared. Both antisera recognized all AGE-proteins, although with different binding curves.
Following pre-incubation with carboxymethyllysine-BSA (CML-BSA) (an oxidation-derived AGE) one
antiserum partially retained its reactivity, suggesting recognition of non-oxidation-derived AGE. This result
was confirmed both in the cross-reactivity and the preincubation experiments and when the reactivity of the
antisera was tested against antigens incubated under oxidative and non-oxidative conditions. These results
confirmed the hypothesis that differently produced antisera may not share the recognition of epitopes of
different nature and suggest the necessity to adopt a standardized methodology for the production of antisera
for an accurate and reproducible determination of the in vivo AGE concentration.
Key words: advanced glycosylation end products (AGE), enzyme-linked immunosorbent assay (ELISA),
polyclonal antisera, AGE-epitopes.
Riassunto (Determinazione dei prodotti finali della glicosilazione non enzimatica: antisieri policlonali
prodotti in condizioni differenti non condividono il riconoscimento di epitopi di origine diversa). - I prodotti
avanzati della glicosilazione non enzimatica (AGE), coinvolti nella patogenesi delle complicanze diabetiche,
comprendono una serie di strutture chimiche tra loro correlate. Così, anticorpi di diversa origine potrebbero
riconoscere solo alcuni e non tutti gli epitopi AGE che si formano in natura. Per verificare questa ipotesi,
sono stati prodotti due diversi anticorpi utilizzando differenti tecniche di immunizzazione e differenti
proteine AGE come immunogeni. I due anticorpi prodotti sono stati messi a confronto in varie situazioni
sperimentali. Entrambi gli anticorpi riconoscono diverse proteine AGE, anche se con curve di legame
diverse. Dopo pre-incubazione con un AGE di origine ossidativa carbossimetillisina-BSA (CML-BSA), un
anticorpo conserva una reattività residua, suggerendo una sua capacità a riconoscere antigeni AGE non
ossidati. Questo risultato è stato confermato sia dagli esperimenti di reattività incrociata e di pre incubazione
che dalla diversa risposta verso antigeni ottenuti in condizioni ossidanti o antiossidanti. Questi risultati
confermano l’ipotesi che anticorpi prodotti in condizioni differenti possono non condividere il
riconoscimento di epitopi di differente origine. Da ciò si evince la necessità di adottare una metodologia
standardizzata per la produzione di antisieri per la determinazione, accurata e riproducibile, della
concentrazione di AGE in vivo.
Parole chiave: prodotti avanzati della glicosilazione non enzimatica (AGE), analisi ELISA, anticorpi
policlonali, epitopi-AGE.
Introduction
Glucose and proteins react non-enzymatically
under normal physiological conditions forming
reversible Amadori-type adducts [1]. In proteins with
long half-lives, such as collagen, the reaction proceeds
further and leads to the appearance of irreversible
advanced glycosylation end products (AGE) [2, 3].
Under physiological conditions of temperature and pH,
AGE levels depend principally on the body fluid
glucose concentration. Thus under normoglycemic
conditions, slow AGE accumulation in tissues
mediates protein senescence and this is suggested to be
one of the mechanisms associated with normal aging
Indirizzo per la corrispondenza (Address for correspondence): Angela Maria Buongiorno, Laboratorio di Biochimica Clinica, Istituto
Superiore di Sanità, Viale Regina Elena 299, 00161 Roma. E-mail: [email protected].
394
Angela Maria BUONGIORNO, Elisabetta SAGRATELLA, Stefania MORELLI et al.
[4, 5]. By contrast, under the hyperglycemic conditions
which characterize diabetes mellitus, the resulting
increased levels of AGE are considered to be involved
in the pathogenesis of late diabetic complications [69]. Thus, accepting that the accumulation of AGE is
physiologically and pathologically relevant, the exact
determination of the concentration of these compounds
in tissues and body fluids could be of considerable
clinical importance. A precise measurement may thus
indicate the rate at which eventual tissue damage
occurs and whether experimental pharmacological
interventions, such as those with aminoguanidine [1013] or D-lysine [14], produce effects.
The original spectrofluorimetric technique of AGE
determination is not believed to be sufficiently specific
and accurate [15, 16], especially if the concentration is
very low, like in plasma and other body fluids. In
addition, the specificity problem has been aggravated
by the demonstration of the existence of nonfluorescent AGE [17]. However, since the report by
Bassiouny that non-enzymatically glycosylated
collagen is antigenic [18], several antisera have been
produced which recognize AGE epitopes, thus
allowing their quantification in immunological assays
[19-22].
The most difficult problem to overcome, for their
immunological measurement is that AGE form a
heterogeneous group of chemical structures present on
carrier proteins, which themselves are much more
antigenic than the AGE hapten they carry. In addition,
the problem has been complicated by the fact that
various proteins have been used as carriers, different
experimental conditions of in vitro AGE production
have been employed and finally different
immunization techniques have been performed. Thus,
although valid polyclonal antisera have been produced
by several groups throughout the world [19, 21, 23,
24], there is not as yet an universally accepted
methodology which could be shared by the involved
laboratories.
In the present work, we have tried to confirm this
hypothesis by producing two completely different antiAGE antisera by means of two completely different
methodologies. In particular: i) the carrier molecules
were bovine pancreatic ribonuclease (Rnase) and
keyhole limpet hemocyanin (KLH), two proteins of
different structures, molecular weights, isoelectric
points and solubility; ii) the AGE-proteins were
produced by treating the native carrier proteins with
different glucose concentrations, temperatures and
incubation times; iii) the rabbits were immunized
using different immunization schedules; iv) the
immunological properties of the resulting antisera,
depending on the immunogen used, were tested under
various experimental conditions.
Experimental procedures
Chemicals. - Bovine serum albumin (BSA)
(fraction V powder), RNase, KLH, human serum
albumin (HSA) (fraction V powder), human
immunoglobulin G (IgG) (lyophilised), human high
density lipoproteins (HDL) and human low density
lipoproteins (LDL), sodium azide, p-nitrophenyl
phosphate and phosphate buffered saline (PBS) tablets,
complete and incomplete Freund’s adjuvants,
glyoxylic acid, sodium borohydride, diethylentriaminepentacetic acid (DTPA), phytic acid and
tris(hydroxymethyl) aminomethane (TRIS) were
purchased from Sigma Chemicals (St. Louis, Miss.,
USA). Alkaline-phosphatase conjugated anti-rabbit
IgG antibody was obtained from Boheringer
Mannheim Italy Spa (Milan, Italy) while D-glucose
and hydrochloridric acid (HCl) were obtained from
Merck (Darmstadt, Germany).
Preparation of AGE-proteins. - AGE-proteins were
prepared by incubating BSA (20.0 mg/ml), HSA (20.0
mg/ml), IgG (5.0 mg/ml), HDL (2.0 mg/ml), LDL (2.0
mg/ml), RNase (25 mg/ml) in 0.4 M sodium phosphate
buffer (pH 7.4) containing 0.5 M D-glucose and 0.06%
w/v sodium azide for 12 weeks at 37 °C in the dark.
AGE-KLH was prepared by incubating KLH (2.0
mg/ml) at 45 °C for 4 weeks in sterile 0.2 M sodium
phosphate buffer (pH 7.4) containing 1.0 M D-glucose.
At the end of the incubation period, all solutions were
extensively dialyzed against PBS (4 °C). Protein
concentrations were determined with the Bio-Rad
Protein Assay (Bio-Rad Laboratories, München,
Germany) using BSA as the standard. Aliquots were
then frozen at - 80 °C.
Preparation of CML-BSA. - CML-BSA (ratio 1:1)
was synthesized according to the method of Reddy
[23]. Briefly, CML-BSA was prepared using 0.2 M
phosphate buffer (pH 8.0) as solvent for 1.0 ml BSA
(fraction V, 175 mg/ml) which was simultaneously
mixed with 0.5 ml of glyoxylic acid (0.3 M) and 0.5 ml
freshly prepared sodium borohydride (0.9 M) and
allowed to mix for 10 min at 37 °C followed by
dialysis against PBS (2.0 M), containing 0.6% (w/v)
sodium azide, overnight at 5 °C. The purity of the
CML-BSA preparation was assessed by amino acid
composition analysis on a modified Beckman 121 MB
analyzer (Beckman, Fullerton, Calif., USA) after
hydrolysis in a closed vial with 6 M HCl (111 °C for
24 h). Lysine was the only amino acid residue
modified by CML formation. The CML-BSA
preparation contained 20 CML-modified lysine
residues per molecule albumin corresponding to 34%
of all of the available lysine residues. Agarose gel electrophoresis [25] in 0.1 M Tris and 0.038 M glycine
buffer (pH 8.7) of the CML-BSA preparation,
exhibited a faster anodic migration with a normal band
395
ADVANCED GLYCOSYLATED END PRODUCT QUANTIFICATION
width than native BSA, indicating the presence of a
homogeneous charge distribution among the AGEderivatized protein molecules.
Immunization techniques. - Twelve female New
Zealand white rabbits, (a 2.5 kg body weight) (Charles
River srl, Lecco, Italy) were utilized for immunization,
six for each antigen.
Protocol basic immunization. - AGE-KLH (0.50
mg/ml) was emulsified in the same volume of
complete Freund’s adjuvant and a total volume of 1.0
ml was inoculated intradermally into multiple sites in
the rabbit back. Inoculations were performed twice
with ten day intervals. Three animals showed a
significant primary response, hence they were boosted
one month later with 1.0 ml of the immunogen (0.25
mg/ml) emulsified in the same volume of incomplete
Freund’s adjuvant, followed by a second booster
inoculation ten days later.
Protocol of hyperimmunization technique. - AGERNAse (2.0 mg/ml) was emulsified in the same
volume of complete Freund’s adjuvant and 1.0 ml was
inoculated intradermally into multiple sites in six
animals back, four times at weekly intervals [21]. Two
animals showed a significant primary response, hence
they were boosted two weeks later with the same
amount of immunogen emulsified in an equal volume
of incomplete Freund’s adjuvant. Booster inoculations
were given four times at weekly intervals.
All five animals were bled seven days after the last
booster injection. The antisera were titered by enzymelinked immunosorbent assay (ELISA) (see below) and
the highest titered antisera obtained using AGE-KLH
(CF5) and AGE-RNase (CF199) as immunogen
selected for future experiments.
Antiserum titer determination. - The primary and
secondary response titer of all antisera produced were
determined by means of a non-competitive ELISA using
BSA and AGE-BSA as the adsorbed ligand antigens as
follows: ELISA 96-wells microtiter plates (Costar,
Cambridge, MA, USA) were coated with BSA or AGEBSA by adding 0.1 ml of each antigen, dissolved in
coating buffer (0.1 M sodium carbonate buffer, pH 9.6,
containing 0.06% w/v sodium azide) at a concentration
of 30 µg/ml and incubating at 4 °C overnight. Wells
were washed three times with 0.15 ml of washing buffer
(PBS containing 0.05% v/v Tween 20 and 0.06% w/v
sodium azide) and blocked by incubation for 1 h at room
temperature with 0.15 ml of blocking buffer
(“Superblock”, Pierce, USA). After three washes with
0.15 ml of washing buffer, 0.05 ml of each antiserum,
serially diluted (10-5 to 10-2) in dilution buffer (PBS
containing 0.02% v/v Tween 20 and 0.06% w/v sodium
azide) added to the wells. Plates were then incubated for
2 h at room temperature in a microtiter plate shaker (100
rpm/min) (Titramax100 Heidolfh Elektro GmbH & Co
KG, Germany). Wells were washed three times with
0.15 ml of washing buffer, 0.1 ml of alkaline
phosphatase conjugated anti-rabbit IgG antibody diluted
1:2000 in dilution buffer added and the plates were
incubated at 37 °C for 1 h. Following the final three
washes with washing buffer, 0.1 ml of p-nitrophenyl
phosphate at a concentration of 1.0 mg/ml in substrate
buffer (0.97% v/v diethanolamine, 0.01% w/v
magnesium chloride-6H2O and 0.06% w/v sodium
azide in double distilled water, pH adjusted to 9.8 with
5 N HCl) was added to the wells and the optical density
(OD) read at 405 nm after colour development. All
determinations were performed in triplicates and the
results expressed as OD values.
Immunological characteristics of the CF5
and CF199 antisera
AGE specificity. - The specificity of the two antisera
towards AGE was tested by means of a competitive
ELISA using AGE-BSA as the adsorbed ligand antigen
and native proteins (BSA, HSA, HDL, LDL and IgG) and
AGE-proteins (AGE-BSA, AGE-HSA, AGE-HDL,
AGE-LDL and AGE-IgG) as competitors. The
competitive ELISA technique differed from the noncompetitive one described above only for the following
step: after preincubation of the wells with blocking buffer,
0.05 ml of scalar concentrations (0.58, 2.3, 9.3, 37.5, 150
and 600 µg/ml) of each competitor and 0.05 ml of each
antiserum (at 1:1000 working dilution) were added to
the wells and the plates incubated for 2 h at room
temperature in microtiter plate shaker. Alkaline
phosphatase conjugated anti-rabbit IgG antibody addition
and colour development was followed as described in the
antisera titer determination section. All determinations
were performed in triplicates. The results were expressed
as %1-B/Bo, calculated as [1-[Experimental OD Background (no first antibody) OD]/ [Total (no competitor)
OD- Background OD] x 100.
Antiserum reactivity. - Direct epitope recognition
and cross-reactivity by both antisera were tested by
means of the competitive ELISA using either AGEBSA or CML-BSA as adsorbed ligand antigens and
either AGE-BSA or CML-BSA as competitors to a
scalar concentrations (0.58, 2.3, 9.3, 37.5, 150 and 600
µg/ml). The results were expressed as %1-B/Bo.
Antiserum reactivity after absorption. - Working
dilution (1:1000) of both antisera were incubated for 2 h
in wells coated with AGE-BSA or CML-BSA, in
microtiter plate shaker. The antisera were then
recovered and their residual reactivity was tested by
means of the competitive ELISA using AGE-BSA (30
µg/ml) as the adsorbed ligand antigen and AGE-BSA or
CML-BSA as the competitors to a scalar concentrations
(0.58, 2.3, 9.3, 37.5, 150 and 600 µg/ml). The results
were expressed as OD values.
Angela Maria BUONGIORNO, Elisabetta SAGRATELLA, Stefania MORELLI et al.
396
Antiserum reactivity against AGE generated in the
presence of antioxidants. - BSA and HSA (20.0 mg/ml
respectively) were incubated at 45 °C for 2 weeks in
sterile de-aerated and N2 saturated 0.4 M sodium
phosphate buffer (pH 7.4) containing 1.0 M D-glucose
with or without the presence of phitic acid (1.0 mM)
and DTPA (1.0 mM) as antioxidant compounds. At the
end of the incubation period, the solutions were
extensively dialyzed against PBS (4 °C) and the
protein concentrations determined as described above.
The reactivity of both antisera against these antigens
was tested by means of the competitive ELISA using
AGE-BSA as the adsorbed ligand antigen and, BSA
and HSA incubated in glucose under oxidative or nonoxidative conditions, as competitors. The AGE levels
were expressed as UAGE/ml.
DO
1.4
1.4
(A)
1.2
1.2
1
1
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
100
1 - B/Bo (%)
80
60
40
20
9.3 37.5 150 600
Twelve rabbits were immunized with one or the
other immunogen, but only five animals showed a
significant primary response and thus continued to
receive booster inoculations. The final titration curves
of the antisera obtained from the two animals with the
highest secondary response are shown in Fig. 1. The
CF5 antiserum titration curve (Fig. 1A) reaches at a
1:1000 dilution the highest difference (12%) between
the OD measured in the presence of native BSA as the
absorbed antigen or AGE-BSA. The CF199 antiserum
titration curve (Fig. 1B) does not reach a plateau but
the OD continues to rise as the titer of the antiserum
rises. Moreover, at a 1:1000 dilution, the OD in the
presence of native BSA amounts to 36% of that
measured in the presence of AGE-BSA.
Immunological characteristics of CF5
and CF199 antisera
0
10-5 10-4 10-3 10-2
10-5 10-4 10-3 10-2
Antiserum diluition
0.58 2.3
Immunization and antiserum titration
(B)
Fig. 1. - ELISA titration curves of the AGE-KLH
antiserum (CF5;panel A) and the AGE-RNAse antiserum
(CF199; panel B). Adsorbed antigens: AGE-BSA (J J);
native BSA (J J)
0
Results
0.58 2.3
9.3 37.5 150 600
Competitor concentration (µg/ml)
Fig. 2. - ELISA competition analysis of the specificity of
the AGE-KLH antiserum (CF5; panel A) and the AGERNAse antiserum (CF199; panel B) towards AGEproteins and native proteins. AGE-BSA (J J); AGE-HSA
(H H); AGE-LDL (B B); AGE-HDL (F F); AGE-IgG (* *);
BSA (J J); HSA (H H); LDL (B B); HDL (F F); IgG (* *).
AGEs specificity. - All AGE-proteins inhibited, in a
dose dependent manner, the binding of both antisera to
the adsorbed ligand antigen (AGE-BSA) (Fig. 2).
However, the inhibition curves of the AGE-proteins
were not similar, but showed different degrees of
binding, as indicated by the different values of %1B/Bo at increasing concentrations of the adsorbed
ligand competitors. With CF5, no inhibition was found
when the competitors were native proteins (Fig. 2A).
In the case of CF199, the antiserum binding to the
adsorbed ligand antigen was somewhat reduced at high
concentrations of native proteins (Fig. 2B).
Antiserum reactivity. - In the presence of AGEBSA (Fig. 3A) or the CML-BSA (Fig. 3B) as the
adsorbed ligand antigens, the binding of the CF5
antiserum to the adsorbed ligand antigens did not
change using alternatively each one of them as the
competitor. In the case of the CF199 antiserum, the
direct competition curves (i.e. adsorbed AGE-BSA vs
AGE-BSA as the competitor or adsorbed CML-BSA vs
CML-BSA as the competitor) are similar. By contrast,
cross-competition curves (i.e. adsorbed AGE-BSA vs
CML-BSA as the competitor or CML-BSA vs AGEBSA as the competitor) demonstrate a higher affinity of
the antiserum for AGE-BSA, since the competition by
CML-BSA as the competitor is greatly reduced even at
high concentrations (Fig. 3C), whereas competition by
AGE-BSA as the competitor is still high and increases
as its concentration increases (Fig. 3D).
Antiserum reactivity after absorption. - The
competition curves of both antisera preabsorbed either
with AGE-BSA or CML-BSA are shown in Fig.4.
Following pre-incubation with AGE-BSA, the affinity
of both antisera towards the AGE-protein antigens
397
ADVANCED GLYCOSYLATED END PRODUCT QUANTIFICATION
Table 1. - Differential recognition by AGE-KLH (CF5) and AGE-RNAse (CF199) antisera of specific glucose AGE
epitopes formed in vitro in the presence of DTPA and Phitic acid as antioxidants, obtained with ELISA competition.
Values expressed as UAGE/ml ± SD(*)
Antiserum
Antioxidants
Proteins
BSA
CF5 antiserum
w/o antioxidants
w antioxidants
% w vs w/o (**)
CF199 antiserum
w/o antioxidants
w antioxidants
% w vs w/o (**)
Level of significance (Welch’s approximate t test)
HSA
20.0 ± 8.2
0.75 ± 0.13
4.1 ± 1.5 (a)
79.2 ± 14.2
2.4 ± 0.19
3.0 ± 0.4 (c)
14.5 ± 0.50
6.7 ± 0.81
45.8 ± 6.3 (b)
58.0 ± 2.0
9.3 ± 0.55
16.1 ± 0.55 (d)
P < 0.01 (***) P < 0.001 (****)
(*): 1 UAGE = %1-B/Bo relative to an AGE-BSA standard concentration of 1 µg/ml. (**): UAGE [with antioxidants] as percentage
of UAGE [without antioxidants]. (***): (a) (n = 4) vs (b) (n = 3) for BSA and (****): (c) (n = 4) vs (d) (n = 3) for HSA.
1-B/Bo (%)
100
100
(A)
80
80
60
60
40
40
20
20
0
100
0.58 2.3
9.3 37.5 150 600
0
100
(C)
80
80
60
60
40
40
20
20
0
0.58 2.3
9.3 37.5 150 600
0
Discussion
(B)
0.58 2.3
9.3 37.5 150 600
(D)
0.58 2.3
9.3 37.5 150 600
Competitor concentration (µg/ml)
Fig. 3. - ELISA competition performed with AGE-KLH
antiserum (CF5; panels A and B) and AGE-RNAse
antiserum (CF199; panels C and D). Adsorbed ligand
antigens: AGE-BSA (panels A and C) and CML-BSA
(panels B and D). Competitors of the adsorbed ligand
antigens: AGE-BSA (J J) and CML-BSA (J J).
appear greatly reduced (Fig. 4, panels A and C). By
contrast, with CML-BSA preabsorbed antisera, CF199
and, to a lesser extent, also CF5, retain the capacity to
bind to the adsorbed ligand antigen AGE-BSA (Fig.4,
panels B and D).
Antiserum reactivity against AGE generated under
oxidative or non-oxidative conditions. - The results of
this experiment are shown in Table 1. Nearly all the
CF5 antiserum reactivity against antigens incubated
with glucose under either oxidative or non-oxidative
conditions was abolished. By contrast CF199
antiserum retained a statistically significant reactivity
(nearly 50% in the case of BSA) even against antigens
prepared under non-oxidative conditions.
Several reports [17, 19-21, 23, 24, 26-29] have
been published on the determination of AGE levels by
immunological assays using polyclonal antisera or
monoclonal antibodies produced with different
methodologies. Since AGE comprise a heterogeneous
group of chemical moieties, it is possible that their
immunogenicity may vary. Thus different antisera
might recognize and measure the levels of different
AGE epitopes which, in addition, might possess
different pathogenic potentials.
One of the main conclusions of this study is to
highlight the difficulty to produce highly specific antiAGE antisera. Thus only five out of twelve animals
showed a significant primary response and, of these,
only two produced antisera which could be used for the
study. The reason for this is probably related to the
difficulty by part of the immune system of efficiently
recognizing the AGE epitopes and not to different
animal behavior, since all twelve animals produced
highly specific antisera against the native carrier
proteins (date not shown).
Further analysis of the results demonstrates an
important difference between the two antisera. Thus, in
addition to their titration curves having different shapes,
they behave differently with respect of the recognition
of native BSA. This antigen is only recognized by CF5
at very low diluition (1:100) whereas a relevant
reactivity is present with CF199 and it is probably due
to the presence of AGE-like in native BSA.
Both antisera recognized AGE epitopes produced
over several different carrier proteins, but with
different bindings. Moreover, CF5 did not recognize
native carrier proteins. By contrast at higher
concentrations of native carrier proteins, CF199
Angela Maria BUONGIORNO, Elisabetta SAGRATELLA, Stefania MORELLI et al.
398
1.2
1.2
(A)
1
1
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0.58 2.3
9.3 37.5 150 600
0
0.58 2.3
9.3 37.5 150 600
O
0
(B)
1.2
1.2
(C)
1
1
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
0.58 2.3
9.3 37.5 150 600
0
(D)
0.58 2.3
9.3 37.5 150 600
Competitor concentration (µg/ml)
Fig. 4. - Reactivity, measured by ELISA competition, of
the AGE-KLH (CF5) and the AGE-RNAse (CF199)
antisera preabsorbed with AGE-BSA (panels A and
C, respectively) and with CML-BSA (panels B and D,
respectively). Adsorbed ligand antigen: AGE-BSA.
Competitors of the adsorbed ligand antigen AGEBSA (J J) and CML-BSA (J J).
antiserum seems to recognize some AGE-like epitopes,
similarly to what was found in the case of native BSA,
present naturally on native proteins.
Given the natural presence of oxygen in plasma
and body tissues, it has been suggested that AGE can
be formed from Amadori products via a process of
oxidation and free radical mediation [30-32]. A glycooxidation AGE epitope which can be artificially
produced is carboxy-methyl-lysine tagged to BSA
(CML-BSA) [33, 34]. Some authors have suggested
that CML may be the major AGE epitope in vivo [23,
24], although it is not a cross-linker and it does not
emit fluorescence when properly excited [17]. When
each antiserum reactivity was tested against AGE-BSA
or CML-BSA antigens, some further differences were
found. In the case of CF5 antiserum, the recognition of
both AGE antigens was similar, suggesting that this
antiserum specifically recognize oxidation-derived
AGE. By contrast, with CF199 antiserum, competition
by CML-BSA failed to completely abolish antiserum
binding to adsorbed AGE-BSA, suggesting saturation of
the antibody fraction that recognizes CML epitopes, even
at relative low concentration of CML-BSA (Fig. 3). This
finding was confirmed when the antisera were
preabsorbed with the antigens before testing their
residual reactivity. Thus preabsorption with AGE-BSA
resulted in the nearly complete removal of the
antibodies recognizing oxidative and/or other epitopes
from both antisera. On the other side, absorption with
CML-BSA did not completely abolish antiserum
reactivity, especially in the case of the CF199, since a
significant residual recognition of AGE-BSA was still
present. This indicates that CF199 contains antibodies,
which recognize epitopes other than CML, even if the
latter is the principal epitope. This corresponds to what
was found by Soulis et al. [35]
The results obtained both in the cross-reactivity
end preincubation experiments were confirmed when
the reactivity of the antisera was tested against BSA
and HSA antigens incubated in high glucose under
non-oxidative conditions. As far as the reactivity of
CF5 is concerned, the data of this experiment confirm
that this antiserum probably recognizes principally
oxidation-derived AGE, since it loses nearly all its
capacity to recognize non-oxidation antigens. By
contrast, CF199 antiserum retains a considerable
amount of its activity when tested against the same
non-oxidation antigens.
It is feasible that human sera contain variable
amounts of AGE, the characteristics of which may also
be different, since, in addition to those locally formed
on circulating proteins, they might derive from the
catabolism of different structural proteins [31]. The
pre-absorption, the cross-section experiment results
and the antiserum reactivity under oxidative and nonoxidative condition reveal a substantial difference in
the activity of the two antisera. From an interlaboratory point of view, the implications of this are
very important. Thus if, in intervention studies on late
diabetic complications, AGE were used as bio-marker
and their levels measured by the participating
laboratories using different antisera and ligand
antigens, the results of the trials may be judged
differently. Therefore a need exists to compare and
eventually calibrate all the different methodologies
used to create anti-AGE antisera and to measure AGE.
Recently, one step towards this aim was made by the
description of a method to standardize AGE
measurement using a reference human serum [27].
In conclusion, this study has demonstrated that
production of anti-AGE antisera with different
methodologies can still result in the formation of
specific antibodies which can recognize AGE epitopes
albeit with different specificity and bindings.
However, the study has also shown that some of the
AGE epitopes recognized by one antiserum may not be
recognized by the other. Thus normal and pathological
tissue and body fluid samples may be found to express
different levels of AGE, depending on which
antiserum is used.
The standardization of anti-AGE antiserum
production and immunological method may be the
only way to set exact guidelines in the difficult task of
determining the level of these physiologically and
pathologically important chemical compounds.
Received on 15 July 2002.
Accepted on 24 October 2002.
ADVANCED GLYCOSYLATED END PRODUCT QUANTIFICATION
REFERENCES
1. Kennedy L, Lyons TJ. Non-enzymatic glycosylation. Brit Med
Bull 1989;45:174-90.
2. Vlassara H, Brownlee M, Cerami A . Accumulation of diabetic rat
peripheral nerve myelin by macrophages increases with the
presence of advanced glycosylation endproducts. J Exp Med
1984;160:197-207.
3. Bucala R, Vlassara H, Cerami A . Advanced glycosylation endproducts. In: Harding JJ, Crabbe MJC (Ed.). Post-translational
modification of proteins. vol. 2. Boca Raton, Florida: CRC Press
Inc; 1992. p. 53-79.
4. Cerami A, Vlassara H, Brownlee M. Glucose and aging. Sci Am
1987;256:90-6.
5. Monnier VM. Toward a Maillard reaction theory of aging. In:
Baynes JW, Monnier VM (Ed.). The Maillard reaction in aging,
diabetes and nutrition. New York: Alan R. Liss; 1989. p. 1-22.
6. Brownlee M, Cerami A, Vlassara H. Advanced glycosylation endproducts in tissue and the biochemical basis of diabetic
complications. N Engl J Med 1988;318:1315-21.
7. Vlassara H, Bucala R, Striker L. Pathogenic effects of advanced
glycosylation: biochemical, biologic and clinical implications for
diabetes and aging. Lab Invest 1994;70:138-51.
8. Sensi M, Pricci F, Pugliese G, De Rossi MG, Petrucci AFG,
Cristina A, Morano S, Pozzessere G, Valle E, Andreani D, Di
Mario U. Role of advanced glycation end-products (AGE) in late
diabetic complications. Diab Res Clin Pract 1995;28:9-17.
9. Vlassara H, Bucala R Advanced glycation and diabetes, an
update. In: Marshall SM, Home P, Rizza RA. (Ed.). The Diabetes
Annual. Amsterdam: Elsevier; 1995. p. 227-44.
10. Brownlee M, Vlassara H, Kooney T, Ulrich P, Cerami A.
Aminoguanidine prevents diabetes-induced arterial wall protein
crosslinking. Science 1986;232:1629-32.
11. Soulis-Liparota T, Cooper M, Papazoglou D, Clarke B, Jerums G.
Retardation by aminoguanidine of development of albuminuria,
mesangial expansion and tissue fluorescence in streptozotocin
induced diabetic rats. Diabetes 1991;40:1328-35.
399
20. Horiuchi S, Araki N, Morino Y. Immunochemical approach to
characterize advanced glycation end products of the Maillard
reaction. J Biol Chem 1991;266:7329-32.
21. Makita Z, Vlassara H, Cerami A, Bucala R. Immunochemical
detection of advanced glycosylation end products in vivo. J Biol
Chem 1992;267:5133-8.
22. Mitsuhashi T, Nakayama H, Itoh T, Kuwajima S, Aoki S, Atsumi
T, Koike T. Immunochemical detection of advanced glycation end
products in renal cortex from STZ-induced diabetic rats. Diabetes
1993;42:826-32.
23. Reddy S, Bichler J, Wells-Knecht KJ, Thorpe SR, Baynes JW.
Nε-(Carboxymethyl)lysine is a dominant advanced glycation end
product (AGE) antigen in tissue proteins. Biochemistry
1995;34:10872-8.
24. Ikeda K, Nagai R, Sakamoto T, Sano H, Araki T, Sakata N,
Nakayama H, Yoshida M, Ueda S, Horiuchi S. Immunochemical
approches to AGE-structure: characterization of anti-AGE
antibodies. J Immunol Methods 1998;215:95-104.
25. Bjerrum OJ, lngild A, Lowenstein H, Weeke B. Quantitation of
human IgG by rocket immunoelectrophoresis at pH 5 by use of
carbamylated antibodies: a routine laborated method. Clin Chim
Acta: 1973;46:337-43.
26. Bucala R, Vlassara H. Advanced glycosylation end products in
diabetic renal disease: clinical measurement, pathophysiological
significance, and prospects for pharmacological inhibition. Blood
Purif 1995;13:160-70.
27. Mitsuhashi T, Vlassara H, Founds HW, Yong Ming Li.
Standardizing the immunological measurement of advanced
glycation endproducts using normal human serum. J Immunol
Methods 1997;207:79-88.
28. Ikeda K, Higashi T, Sano H, Jnnouchi Y, Yoshida M, Araki T,
Ueda S, Horiuchi S. Ne-(Carboxymethyl)lysine protein adduct is
a major immunological epitope in proteins modified with
advanced glycation end producys of the Maillard reaction.
Biochemistry 1996;35:8075-83.
12. Edelstein D, Brownlee M. Mechanistic studies of advanced
glycosylation end product inhibition by aminoguanidine.
Diabetes 1992;41:26-9.
29. Munch G, Keis R, Wessels A, Riederer P, Bahner U, Heidland A,
Niwa T, Lemke H.D, Schinzel R. Determination of advanced
glycation end products in serum by fluorescence spectroscopy
and competitive ELISA. Eur J Clin Chem Clin Biochem
1997;35:669-77.
13. Hammes HP, Strodter D, Weiss A, Bretzel RG, Federlin K,
Brownlee M. Secondary intervention with aminoguanidine
retards the progression of diabetic retinopathy in the rat model.
Diabetologia 1995;38:656-60.
30. Hunt JV, Smith CCT, Wolff SP. Autooxidative glycosylation and
possible involvement of peroxides and free radicals in LDL
modification by glucose. Diabetes 1990;39:1420-4.
14. Sensi M, De Rossi MG, Celi FS, Cristina A, Rosati C, Perrett D,
Andreani D, Di Mario U. D-Lysine reduces the non-enzymatic
glycation of proteins in experimental diabetes mellitus in rats.
Diabetologia 1993;36:797-801.
15. Brennan M. Changes in solubility, non-enzymatic glycation and
fluorescence of collagen in tail tendons from diabetic rats. J Biol
Chem 1989;264:20947-52.
16. Baynes JW. Role of oxidative stress in development of
complications in diabetes. Diabetes 1991;40:405-12.
17. Dyer DG, Dunn JA, Thorpe SR, Bailie KE, Lyons TJ, McChance
DR, Baynes JW. Accumulation of Millard reaction products in skin
collagene in diabetes and aging. J Clin Invest 1993;91:2463-9.
18. Bassiouny AR, Rosenberg H, McDonald TL. Glucosylated
collagen is antigenic. Diabetes 1981;32:1182-4.
19. Nakaiama H, Taneda S, Kuwajima S, Aoki S, Kuroda Y, Misawa
K, Nakagawa S. Production and characterization of antibodies to
advanced glycosylation products on proteins. Biochem Biophys
Res Commun 1989;162:740-5.
31. Hunt JV, Bottoms MA, Mitchinson MJ. Oxidative alteration in
the experimental glycation model of diabetes mellitus are due to
protein-glucose adduct oxidation. Biochem J 1993;291:529-35.
32. Fu MX, Wells-Knecht KJ, Blackledge JA, Lyons TJ, Thorpe SR
and Baynes JW. Glycation, glycoxidation and cross-linking of
collagen by glucose. Diabetes 1994;41:676-83.
33. Ahmed MU, Thorpe SR, Baynes JW. Identification of N epsiloncarboxymethillysine as a degradation product of fructolysine in
glycated protein. J Biol Chem 1986;261:4889-94.
34. Dunn JA, McChance DR, Thorpe SR, Lyons TJ, Baynes JW. Agedependent accumulation of Nε-(carboxymethyl)lysine and Nε(carboxymethyl)hydroxy-lysine in human skin collagen.
Biochemistry 1991;30:1205-10.
35. Soulis T, Cooper ME, Sastra S, Thallas V, Panagiotopoulus S,
Bjerrum OJ, Jerums G. Relative contributions of advanced
glycation and nitric oxide synthase inhibition to aminoguanidinemediated renoprotection in diabetic rats. Diabetologia
1997;40:1141-51.