Reactivity of 1-deoxy-D-erythro-hexo-2,3-diulose; A key intermediate in the maillard chemistry of hexoses

Article abstract of DOI:10.1021/jf900459x

Degradation of 1-deoxyhexo-2,3-diulose, a key intermediate in Maillard chemistry, in the presence of L-alanine under moderate conditions (37 and 50 °C) was investigated. Different analytical strategies were accomplished to cover the broad range of product

Full text of DOI:10.1021/jf900459x

J. Agric. Food Chem. 2009, 57, 4765–4770 4765
DOI:10.1021/jf900459x
Reactivity of 1-Deoxy-
D-erythro-hexo-2,3-diulose: A Key
Intermediate in the Maillard Chemistry of Hexoses
MICHAEL
V
OIGT AND
MARCUS A. GLOMB*
Institute of Chemistry, Food Chemistry, Martin-Luther-University Halle-Wittenberg,
Kurt-Mothes-Strasse 2, 06120 Halle/Saale, Germany
Degradation of 1-deoxyhexo-2,3-diulose, a key intermediate in Maillard chemistry, in the presence of
L-alanine under moderate conditions (37 and 50 °C) was investigated. Different analytical strategies
were accomplished to cover the broad range of products formed and their differing chemical
properties. These involved GC-MS analysis of trimethylsilyl and O-benzyloxime trimethylsilyl
derivatives (after reaction with O-benzylhydroxylamine and N,O-bis(trimethylsilyl)acetamide), GC-
FID analysis of the decyl ester of acetic acid (after reaction with decyl chloroformate), and HPLC-UV
analysis of quinoxaline derivatives (after reaction with o-phenylenediamine). Among the compounds
identified were carboxylic acids (glyceric acid and acetic acid) that can be seen as stable Maillard
end-products. However, the formation of dicarbonyls (3,4-dihydroxy-2-oxobutanal, 1-hydroxybutane-
2,3-dione, and 4-hydroxy-2-oxobutanal) and of hydroxycarbonyls (acetol) was verified presenting
unstable, reactive Maillard intermediates. Results confirmed that β-dicarbonyl cleavage is a very
important pathway within the degradation of 1-deoxyhexo-2,3-diulose. Other reactions taking place
include enolization, water elimination, and oxidation.
KEYWORDS: Maillard reaction; carbohydrate degradation; 1-deoxyhexo-2,3-diulose; 1-deoxyglucosone;
R-dicarbonyl compounds; dicarbonyl cleavage; acetic acid
INTRODUCTION
4-dihydroxy-2,5-dimethylfuran-3(2H)-one 3 (acetylformoin) (9 )
or the furanones 4 and 5, already established earlier in Maillard
model systems and processed food (6, 10, 11). Despite the obvious
importance of 1-DG, there are, to the best of our knowledge, only
three studies so far investigating the reactions of 1-DG from the
native authentic compound (12-14), although an independent
synthesis was already developed several years ago (15). In these
studies, Davidek and co-workers proved that 1-DG is a precursor
of acetic acid under cooking conditions. The formation of acetic
acid was explained by a hydrolytic β-dicarbonyl cleavage of the
isomeric 1-deoxyhexo-2,4-diulose. It was also found that to a
minor extent 1-DG can undergo an oxidative R-dicarbonyl
cleavage leading to other carboxylic acids (13).
In our study, we succeeded to expand the knowledge on the
reactions of 1-DG and the products formed from it. Formation of
organic acids as well as the formation of dicarbonyls and
hydroxycarbonyls was established. Several different analytical
methods were accomplished to obtain quantitative results, which
allowed insights into the reaction mechanisms behind 1-DG
degradation.
The complex of reactions between reducing sugars and com-
pounds bearing an amino group, also referred to as the Maillard
reaction or nonenzymatic browning, results in its early stage in
the Amadori compound as a first analytically reachable inter-
mediate (1 ). Later, reactive intermediates are formed, which are
responsible for the formation of taste- and odor-active as well as
colored compounds in processed food (2 ). Among these inter-
mediates, dicarbonyls are of major importance. For example
the Amadori compound of glucose delivers 3-deoxyglucosone
(3-DG) via 1,2-enolization (3, 4). This well-investigated com-
pound reacts to, for example, 5-(hydroxymethyl)furan-2-carbal-
dehyde, an indicator for heat treatment of processed food. Glomb
and Tschirnich showed that 1-deoxyhexo-2,3-diulose 1 [1-deoxy-
glucosone, 1-DG, (4R,5R)-4,5,6-trihydroxyhexane-2,3-dione] is
of much higher reactivity in comparison to that of 3-DG (5 ).
1-DG is alternatively formed from the Amadori compound via
2,3-enolization. Numerous postulated reaction pathways in Mail-
lard chemistry use 1-DG to explain the formation of well-known
Maillard products (see Figure 1) such as 3,5-dihydroxy-6-methyl-
2,3-dihydro-4H-pyran-4-one 2 (γ-pyranone) (6-8). This com-
pound can be found in yields up to the g/kg range in heated food
indicating the occurrence of the Maillard reaction. Its formation
is explained by dehydration of the pyranoic hemiketal form of
1-DG. Analogous reactions of the furanoic form of 1 lead to 2,
MATERIALS AND METHODS
Materials. The following chemicals of analytical grade were commer-
cially available: acetol (Acros Organics, Beerse, Belgium), N,O-bis(tri-
methylsilyl)acetamide with 5% trimethylchlorosilane, acetic acid,
L-alanine,
D-erythronic acid γ-lactone,
D-(-)-erythrose, L(+)-erythrulose,
o-phenylenediamine (OPD), heptafluorobutyric acid, 2,3,5-triphenyltetra-
zolium chloride (TTC) (Fluka/Sigma-Aldrich, Seelze, Germany), dipotas-
sium hydrogen phosphate trihydrate, potassium dihydrogen phosphate
*To whom correspondence should be addressed. Fax: ++049-345-
5527341. E-mail: marcus.glomb@chemie.uni-halle.de.
© 2009 American Chemical Society
Published on Web 05/07/2009
pubs.acs.org/JAFC

4766 J. Agric. Food Chem., Vol. 57, No. 11, 2009
Voigt and Glomb
Figure 1. Formation of well-established Maillard products related to 1-deoxyhexo-2,3-diulose 1: 3,5-dihydroxy-6-methyl-2,3-dihydro-4H-pyran-4-one
(γ-pyranone) 2, 2,4-dihydroxy-2,5-dimethylfuran-3(2H )-one (acetylformoin) 3, 4-hydroxy-2-(hydroxymethyl)-5-methylfuran-3(2H )-one 4, and 4-hydroxy-
5-methylfuran-3(2H )-one 5.
(Merck, Darmstadt, Germany), calcium sulfate, diethyl ether, ethyl acetate
(EtOAC) (Roth, Karlsruhe, Germany), diethylenetriaminepentaacetic
Decylchloroformate Derivative of Acetic Acid. Adopting the
method described in ref (17), samples were spiked with chlorosuccinic acid
(50 μg) dissolved in water as internal standard, and pyridine (40 μL)
and decylchloroformate (50 μL) were added. The mixture was then
sonicated for 10 min. Decyl esters were extracted with hexane, and the
organic layer was analyzed by GC-FID. Quantitative results were
obtained by internal calibration using commercially available acetic acid.
(Data obtained with GC-FIDshowedstandard deviations <20 mmol/mol
1-DG, resulting in coefficients of variation <3.5%. Coefficient of deter-
mination was always >0.97.)
acid,
D(-)-threose (Sigma-Aldrich, Taufkirchen, Germany),
D-glyceric
acid calcium salt, 2-methylquinoxaline (Aldrich/Sigma-Aldrich, Stein-
heim, Germany), and Dowex 50W Â 8 (H+-form, 50-100 mesh)
(Serva/Boehringer, Ingelheim-Heidelberg, Germany). 1-Deoxy-4,5-O-iso-
propylidene-
3-diulose were synthesized as described in ref (15). Quinoxalines of 3,
4-dihydroxy-2-oxobutanal (threosone), 1-hydroxybutane-2,3-dione
D-erythro-hexo-2,3-diulose and 1-deoxy-D-erythro-hexo-2,
(1-deoxythreosone), and 4-hydroxy-2-oxobutanal (3-deoxythreosone)
were isolated from glucose-lysine-OPD reaction mixtures by multilayer
counter current chromatography (data not shown). Their identity and
purity were confirmed by NMR and high resolution mass spectrometry in
accordance with ref (16).
Quinoxaline Derivatives. Adopting the method described in
ref (5), samples were spiked with o-phenylenediamine (4.95 μmol) dis-
solved in water (30 μL) and kept 5 h at RT prior to injection into
the HPLC-UV-system (conversion of dicarbonyls to quinoxalines under
these reactions conditions was almost 100% according to ref (5)).
Quinoxalines were monitored at λΜ = 320 nm. Quantification was
carried out by comparison of peak areas obtained at 320 nm with those
of standard solutions containing known amounts of either commer-
cially available reference compound (2-methylquinoxaline) or pure
authentic quinoxalines isolated and elucidated from reaction mixtures
in our working group. (Data obtained with HPLC-UV showed
standard deviations <1.5 mmol/mol 1-DG, resulting in coefficients of
variation <5%.)
Degradation of 1-Deoxyhexo-2,3-diulose. A solution of 1-deoxy-
4,5-O-isopropylidene-D-erythro-hexo-2,3-diulose (0.2 mmol) in water
(2 mL) was stirred with Dowex 50W Â 8 (H+-form, 50-100 mesh,
2 mL) under argon atmosphere. Completeness of deprotection was
checked after 3 h with TLC (TTC detection). The resin was filtered off
and washed with MeOH. After evaporation of the combined solvents, the
residue was dissolved in phosphate buffer (0.1 M, pH 7.4, 2.4 mL).
Aliquots (150 μL) of this solution and a solution of
in phosphate buffer (2.4 mL) were dispatched in screw-cap vials (1.5 mL)
giving an incubation solution of 1-DG and -alanine (42 mM, respec-
L-alanine (0.2 mmol)
L
O-Benzyloxime Trimethylsilyl Derivatives. Hydroxycarbonyl
compounds were derivatized on the basis of ref (18) with some modifica-
tions. Samples were spiked with a solution of 6 μmol O-benzylhydrox-
ylamine hydrochloride in water (300 μL) and reacted 3 h at 37 °C. The
mixture was adjusted to pH 6 and extracted with 1 mL of diethyl ether. The
organic layer was dried with calcium sulfate, evaporated with argon, and
the residue silylated as described above. Quantitative results were obtained
by external calibration using commercially available acetol. (Data ob-
tained with GC-MS showed standard deviations <3 mmol/mol 1-DG,
resulting in coefficients of variation <7.5%. Coefficient of determination
was always >0.98. Recovery was between 95 and 105%.)
tively). Incubation solutions were heated (37 and 50 °C), and samples were
taken over time and analyzed by GC after derivatization. Values were
expressed as the means of at least three independent determinations.
Deaerated Incubations. Degradation of 1-deoxy-D-erythro-hexo-2,
3-diulose under deaerated conditions was carried out using phosphate
buffer with 1 mM diethylenetriaminepentaacetic acid. Buffer was degassed
with helium before samples were prepared; samples were deaerated with
argon before incubation.
Derivatization Reactions. Trimethylsilyl Derivatives. Adopting
the method described in ref (5), aliquots of the samples (50 μL) were dried
and residues dissolved in pyridine (50 μL), and N,O-bis(trimethylsilyl)
acetamide with 5% trimethylchlorosilane (50 μL) was added. Samples
were kept 3 h at RT prior to injection into the GC-MS-system. Quantifica-
tion was carried out by comparison of peak areas obtained in the TIC with
those of standard solutions containing known amounts of the pure
authentic reference compounds. Signals of target compounds were stan-
dardized using the signal of silylated phosphoric acid present in all
samples. Data for silylated compounds obtained by GC-MS showed
standard deviations <10 mmol/mol 1-DG, resulting in coefficients of
variation <5%.
GC-MS. Samples were analyzed on a Thermo Finnigan Trace GC
Ultra coupled to a Thermo Finnigan Trace DSQ (both from Thermo
Fisher Scientific GmbH, Bremen, Germany). The GC column was HP-5
(30 m  0.32 mm, film thickness 0.25 μm; Agilent Technologies, Palo Alto,
CA). Injector, 220 °C; split ratio, 1:30; transfer line, 250 °C. The oven
temperature program was as follows: 100 °C (0 min), 5 °C/min to 200 °C
(0 min), and 10 °C/min to 270 °C (10 min). Helium 5.0 was used as carrier
gas in constant flow mode (linear velocity 28 cm/s, flow 1 mL/min). Mass
spectra were obtained with EI at 70 eV (source: 210 °C) in Full scan mode
(mass range m/z 50-650).

Article
J. Agric. Food Chem., Vol. 57, No. 11, 2009 4767
GC-FID. Samples were analyzed on a HP 6890N chromatograph
(Agilent Technologies, Palo Alto, CA, USA) equipped with a flame
ionizationdetector. The column was HP-5(30 m  0.32 mm, film thickness
0.25 μm; Agilent Technologies, Palo Alto, CA). Injector, 250 °C; split
ratio, 1:10; detector, 270 °C. Helium 4.6 was used as carrier gas in constant
flow mode (linear velocity 28 cm/s, flow 1.6 mL/min).
HPLC-UV. A Jasco PU-2089 Plus quaternary gradient pump with
degasser was used combined with a Jasco AS-2055 Plus autosampler.
Elution of materials was monitored by a Jasco UV-2075 Plus UV detector
(all Jasco, Gross-Umstadt, Germany). Chromatographic separation was
carried out on a stainless steel column (Eurospher 100-5 C18, 250 Â
4.6 mm) by Knauer (Berlin, Germany) using a flow rate of 1 mL/min. The
mobile phase used was water (solvent A) and water/MeOH 3:7 (v/v,
solvent B). Heptafluorobutyric acid (0.6 mL/L) was added to both
solvents. Samples were injected at 20% B, gradient then changed linear
to 30% B in 35 min, to 40% B in 5 min, then to 70% B in 15 min (held
5 min), then changed to 100% B in 30 min (held 10 min), and then changed
to 20% B in 5 min (held 15 min).
Figure 2. Total ion chromatogram of a 1-DG/L-alanine incubation mixture
(42 mM in 0.1 M phosphate buffer, pH 7.4, after 24 h at 37 °C under
deaerated conditions). Signals are trimethylsilyl derivatives of lactic acid
RESULTS
Carboxylic Acids. Degradation of 1-deoxyhexo-2,3-diulose 1 in
the presence of amine lead to several peaks, which could be
analyzed by GC-MS after silylation. Figure 2 shows the total ion
chromatogram of a derivatized sample (after 12 h, at 37 °C under
deaerated conditions). Important signals that could be identified
were assigned to lactic acid 6, glyceric acid 9, 3,5-dihydroxy-6-
methyl-2,3-dihydro-4H-pyran-4-one 2, erythrulose 10, and 4-
hydroxy-2-(hydroxymethyl)-5-methylfuran-3(2H)-one 4. As ex-
6,
10, and 4-hydroxy-2-(hydroxymethyl)-5-methylfuran-3(2H )-one 4.
L
-alanine 7, phosphoric acid 8, glyceric acid 9, γ-pyranone 2, erythrulose
pected, silylated
signals.
L-alanine7 and phosphoric acid 8 gave the largest
Formation of glyceric acid is shown in Figure 3. Obviously, it
was formed rather fast within the first hour. Later on, concentra-
tions increased at a smaller rate, nearly linear over the whole
period of time monitored. Formation was significantly favored
under aerated conditions, for example, 171 mmol/mol 1-DG of
glyceric acid was formed after 12 h at 37 °C under aerated
conditions (vs 40 mmol/mol 1-DG under deaerated conditions).
Another carboxylic acid found was acetic acid. Figure 4 shows
a GC-FID-chromatogram of the decylester formed after deriva-
tization with decyl chloroformate (chlorosuccinic acid was used
as internal standard). Acetic acid was a major product formed
from 1 even under the less rigorous conditions used in the present
work. Formation was slightly preferred under deaerated condi-
tions (up to 650 mmol/mol 1-DG was obtained after 24 h at 37 °C
under deaerated conditions vs 550 mmol/mol 1-DG under
aerated conditions).
Figure 3. Formation of glyceric acid (1-DG/L-alanine, 42 mM in 0.1 M
phosphate buffer, pH 7.4).
Hydroxycarbonyl and Dicarbonyl Compounds. Besides car-
boxylic acids, we also detected several hydroxycarbonyl and
dicarbonyl compounds. Erythrulose was analyzed after silylation
via GC-MS. Formation of this C4-sugar was considerably
favored under deaerated conditions. An almost 4-fold higher
amount was obtained after 16 h at 37 °C (162 mmol/mol 1-DG vs
43 mmol/mol 1-DG). Levels reached a maximum at about 16 h
and decreased in later samples. This pattern of a peak concentra-
tion with subsequent degradation was also found inthe case ofthe
dicarbonyl compounds detected in the incubation mixture, i.e.,
threosone (3,4-dihydroxy-2-oxobutanal), 1-deoxythreosone
(1-hydroxybutane-2,3-dione), and 3-deoxythreosone (4-hydro-
xy-2-oxobutanal). Formation ofthese dicarbonyls reached a peak
level after about 1 h at 50 °C. The initial rate of formation was
slightly lower at 37 °C; maximum levels were reached after 2 h at
this incubation temperature. In addition, amounts of dicarbonyls
detected were generally much lower than those of carboxylic acids
or erythrulose. Formation of 1-deoxythreosone and 3-deoxy-
threosone was favored under deaerated conditions, while threo-
sone formation was preferred in presence of oxygen. Highest
Figure 4. Identification of acetic acid by GC-FID (acetic acid decyl ester 11
and chlorosuccinic acid decyl ester ISTD).
levels were obtained for 1-deoxythreosone (Figure 5), which was
formed up to 31 mmol/mol 1-DG under deaerated conditions.
About half of this amount (14 mmol/mol 1-DG) was measured
under aerated conditions. This 2:1 ratio changed to about 1.5:1 at
50 °C (28 vs 19 mmol/mol 1-DG). The other two dicarbonyls were

4768 J. Agric. Food Chem., Vol. 57, No. 11, 2009
Voigt and Glomb
Figure 5. Formation of 1-deoxythreosone (1-DG/L-alanine, 42 mM in
0.1 M phosphate buffer, pH 7.4).
Table 1. Degradation of 1-Deoxyhexo-2,3-diulose Leads to C4-Dicarbonyls
37 °Ca 50 °Cb
mmol/mol 1-DG
Figure 6. Mechanism of 1-deoxyhexo-2,3-diulose degradation via R- and
β-dicarbonyl cleavage leading to acetic acid.
aerated
deaerated
aerated
deaerated
1-deoxythreosone
3-deoxythreosone
threosonec
14
3
31
10
5
19
5
28
10
10
that nearly similar concentrations of acetic acid and propanoic
acid were formed in a 2,3-dicarbonyl model system containing
pentane-2,3-dione. However, yields were very low compared to
that in β-dicarbonyl fragmentation. Therefore, a minor addi-
tional oxidative R-dicarbonyl cleavage was suggested and re-
searched more closely in a later study (13). We also found large
yields of acetic acid even under the far more moderate conditions
used herein. The slightly higher amounts that were obtained
under deaerated conditions may be explained by the greater
stability of 1 under these conditions. Additional formation of
acetic acid via an oxidative R-dicarbonyl cleavage should result in
higher yields under aerated conditions, which could not be
attested from our data. Furthermore, we were not able to identify
erythronic acid, which would be the counterpart expected in that
mechanism. Thus, oxidative R-dicarbonyl cleavage might indeed
be only of minor relevance or even nonexistent under the reaction
conditions herein.
The enediol intermediate occurring within the β-dicarbonyl
cleavage is in our opinion a linchpin in the formation of the
second group of compounds (reactive intermediates) investigated
in this study. Tautomerization leads to erythrulose (see Figure 7),
which was obtained in significant amounts in the incubation
mixtures. It should be mentioned that the two possible isomeric
aldoses erythrose and threose could not be identified in our
samples. The 4-fold higher amount of erythrulose obtained under
deaeration can be explained by the instability of its direct and
intermediate precursors in the presence of oxygen, i.e., the C4-
enediol and 1-DG. The oxidation product of the enediol is
threosone, which was verified as a degradation product formed
from 1 in our samples especially under aeration. This compound
is ratherknown from nonenzymatic browningreactions involving
ascorbic acid (20). Usui and co-workers recently reported threo-
sone during the degradation of glucose and fructose (16). How-
ever, these authors explained its formation via glucosone and
did not consider 1-DG as a possible precursor in threosone
formation. Oxidation of an enediol is a well-established step,
for example, oxidation of the 1,2-enediol of glucose giving
14
14
a Maximum concentrations were measured after 2 h. b Maximum concentrations
were measured after 1 h. c Maximum concentrations were measured after 30 min.
measured in lower amounts (Table 1). In the same range were the
concentrations quantified for hydroxypropane-2-one (acetol).
This hydroxycarbonyl compound was determined from the
incubation by GC-MS analysis as its O-benzyloxime trimethylsi-
lyl derivative. Slightly lower amounts were obtained for deaerated
conditions, for example, 10 mmol/mol versus 18 mmol/mol 1-DG
(after 12 h at 37 °C) and 28 mmol/mol versus 38 mmol/mol 1-DG
(after 12 h at 50 °C).
DISCUSSION
Products formed during the degradation of 1 in the presence of
amine can be divided into two groups. On the one hand, there are
carboxylic acids representing stable degradation products that
accumulated over time. On the other hand, reactive intermediates
(dicarbonyl and hydroxycarbonyl compounds) were formed to a
certain amountand immediately degradedonce the precursor was
exhausted.
Carboxylic acids have been described as products of sugar
fragmentation very early in papers on Maillard reaction mechan-
isms. Hodge already postulated 1 being involved in acetic acid
formation (1 ). He assumed the formation of glycolaldehyde from
1-DG via retro-aldolization followed by a saccharinic acid
rearrangement to give acetic acid. Brands and van Boekel
proposed acetic acid to be formed from 1 via a hydrolytic
R-dicarbonyl cleavage (19). Davidek et al. recently showed that
large amounts (up to 85 mol %) of acetic acid were indeed formed
from 1 at 100 °C regardless of the presence of an amino acid or
amine. However, their mechanistic study (12) suggested a hydro-
lytic β-dicarbonyl cleavage of the 2,4-tautomer of 1 as the major
pathway for acetic acid formation (Figure 6). In addition, no
evidence was found for the hydrolytic R-dicarbonyl cleavage that
seemed well established in the literature. Instead, they reported
D-arabino-hexos-2-ulose (glucosone).

Article
J. Agric. Food Chem., Vol. 57, No. 11, 2009 4769
Interestingly, we measured only negligible amounts of 2-
methylquinoxaline (data not shown), although we expected
methylglyoxal to be an important compound formed from 1 via
retro-aldolization. Thus, retro-aldol fragmentation may not play
a major role in 1-DG chemistry under the conditions investigated
in our experiments. The different pathways of the degradation of
1 resulting in the products investigated in the present work are
summarized in Figure 7. The 2,4-tautomer of 1-deoxyhexo-2,
3-diulose plays a pivotal role in the formation of fragmentation
products with a C2, C3, and C4 carbon backbone.
LITERATURE CITED
(1) Hodge, J. E. Dehydrated foods: chemistry of browning reactions in
model systems. J. Agric. Food Chem. 1953, 1, 928–943.
(2) Ledl, F.; Schleicher, E. Die Maillard-Reaktion in Lebensmitteln und
::
im menschlichen Korper - neue Ergebnisse zu Chemie, Biochemie
und Medizin. Angew. Chem. 1990, 102, 597–626.
(3) Anet, E. F. L. J. 3-Deoxyhexosones. J. Am. Chem. Soc. 1960, 82, 1502.
(4) Anet, E. F. L. J. 3-Deoxyglycosuloses (3-deoxyglycosones) and the
degradation of carbohydrates.. Adv. Carbohydr. Chem 1964, 19,
181–218.
(5) Glomb, M. A.; Tschirnich, R. Detection of alpha-dicarbonyl com-
pounds in Maillard reaction systems and in vivo. J. Agric. Food
Chem. 2001, 49, 5543–5550.
(6) Mills, F. D.; Weisleder, D.; Hodge, J. E. 2,3-Dihydro-3,5-dihydroxy-
6-methyl-4H-pyran-4-one, a novel nonenzymatic browning product.
Tetrahedron Lett. 1970, 11, 1243–1246.
(7) Ledl, F.; Schnell, W.; Severin, T. Proof of 2,3-dihydro-3,5-dihy-
droxy-6-methyl-4H-pyran-4-on in foods. Studies on Maillard reac-
tion XI. Z. Lebensm. Unters.-Forsch 1976, 160, 367–370.
(8) Kim, M. O.; Baltes, W. On the role of 2,3-dihydro-3,5-dihydroxy-6-
methyl-4(H)-pyran-4-one in the Maillard reaction. J. Agric. Food
Chem. 1996, 44, 282–289.
(9) Hofmann, T. Acetylformoin: A chemical switch in the formation of
colored Maillard reaction products from hexoses and primary and
secondary amino acids. J. Agric. Food Chem. 1998, 46, 3918–3928.
(10) Severin, Th.; Seilmeier, W. Maillard reaction. II. Reaction of
pentoses in the presence of amine acetates. Z. Lebensm. Unters. -
Forsch. 1967, 134, 230–232.
Figure 7. Degradation of 1-deoxyhexo-2,3-diulose leads to C2, C3, and
C4 fragmentation.
The enediol may also be dehydrated at C-3 resulting in
3-deoxythreosone (4-hydroxy-2-oxobutanal), which we identified
as its stable quinoxaline in the incubation mixtures. The isomeric
2,3-enediol may readily be formed from the described 1,2-enediol
and should give rise to 1-deoxythreosone (1-hydroxybutane-2,3-
dione) in an equivalent reaction. This compound was also verified
as its quinoxaline in the degradation of 1. Both dicarbonyls were
originally reported by Nedvidek from glucose/amino acid model
systems (21). Davidek identified the quinoxaline of 1-deoxy-
threosone by GC-MS analysis in glucose/glycine incubations and
interpreted it as additional evidence of glucose degradation along
the 1-DG road (12).
The nucleophilic attack of a hydroxyl anion, which has to be
assumed within the β-dicarbonyl cleavage mechanism, can of
course also happen at the C-4 carbonyl group of the 2,4-tautomer
of 1. This results in the formation of glyceric acid and
1-hydroxypropan-2-one (acetol). Acetol formation from hexoses
was already investigated about 50 years ago, and Hayami
proposed β-dicarbonyl cleavage to explain such results (22, 23).
Weenen later reported acetol formation from glucose and pro-
posed 1 to be involved (24). Yaylayan carried out label incor-
(11) Hiebl, J.; Ledl, F.; Severin, T. Isolation of 4-hydroxy-2-(hydroxy-
methyl)-5-methyl-3(2H)-furanone from sugar amino-acid reaction
mixtures. J. Agric. Food Chem. 1987, 35, 990–993.
(12) Davidek, T.; Devaud, S.; Robert, F.; Blank, I. Sugar fragmentation
in the Maillard reaction cascade: Isotope labeling studies on the
formation of acetic acid by a hydrolytic beta-dicarbonyl cleavage
mechanism. J. Agric. Food Chem. 2006, 54, 6667–6676.
(13) Davidek, T.; Robert, F.; Devaud, S.; Vera, F. A.; Blank, I. Sugar
fragmentation in the Maillard reaction cascade: Formation of short-
chain carboxylic acids by a new oxidative alpha-dicarbonyl cleavage
pathway. J. Agric. Food Chem. 2006, 54, 6677–6684.
(14) Robert, F.; Vera, F. A.; Kervella, F.; Davidek, T.; Blank, I.
Elucidation of Chemical Pathways in the Maillard Reaction by O-
17-NMR Spectroscopy. In The Maillard Reaction: Chemistry at the
Interface of Nutrition, Aging, and Disease; Baynes, J. W., Monnier,
V. M., Ames, J. M., Thorpe, S. R., Eds.; Annals of the New York
Academy of Sciences; Wiley: New York, 2005; Vol. 1043, pp 63-72.
(15) Glomb, M. A.; Pfahler, C. Synthesis of 1-deoxy-D-erythro-hexo-2,3-
diulose, a major hexose Maillard intermediate. Carbohydr. Res.
2000, 329, 515–523.
(16) Usui, T.; Yanagisawa, S.; Ohguchi, M.; Yoshino, M.; Kawabata, R.;
Kishimoto, J.; Arai, Y.; Aida, K.; Watanabe, H.; Hayase, F.
Identification and determination of alpha-dicarbonyl compounds
formed in the degradation of sugars. Biosci. Biotechnol. Biochem.
2007, 71, 2465–2472.
(17) Husek, P. Amino-acid derivatization and analysis in 5 minutes.
FEBS Lett. 1991, 280, 354–356.
(18) Mason, B. S.; Slover, H. T. Gas Chromatographic method for
determination of sugars in foods. J. Agric. Food Chem. 1971, 19,
551–554.
poration analysis of
D-glucose/amino acid systems under
pyrolysis conditions (25, 26). He concluded that 70% of the
detected acetol was derived from 1-DG by retro-aldolization of
the isomeric 1-deoxyhexo-2,5-diulose. However, β-dicarbonyl
cleavage seems to be a more likely pathway or at least an
additional pathway. Glyceric acid formation from glucose was
also investigated by Davidek and co-workers. It was shown that
thermal treatment of [1,2-13C2]glucose and [3-13C]glucose isoto-
pomers resulted in unlabeled glyceric acid only (13). This is in
agreement with the β-dicarbonyl cleavage mechanism of the 2,4-
isomer of 1. The higher amounts of glyceric acid we obtained
under aerated conditions may be explained by additional
oxidative fragmentation pathways.

4770 J. Agric. Food Chem., Vol. 57, No. 11, 2009
Voigt and Glomb
(19) Brands, C. M. J.; van Boekel, M. A. J. S. Reactions of mono-
saccharides during heating of sugar-casein systems: Building of a
reaction network model. J. Agric. Food Chem. 2001, 49, 4667–4675.
(20) Schulz, A.; Trage, C.; Schwarz, H.; Kroh, L. W. Electrospray
ionization mass spectrometric investigations of alpha-dicarbonyl
compounds: Probing intermediates formed in the course of the
nonenzymatic browning reaction of L-ascorbic acid. Int. J. Mass.
Spectrom. 2007, 262, 169–173.
(23) Hayami, J. Studies on the chemical decomposition of simple sugars.
XII. Mechanism of the acetol formation. Bull. Chem. Soc. Jpn. 1961,
34, 927–932.
(24) Weenen, H. Reactive intermediates and carbohydrate fragmentation
in Maillard chemistry. Food Chem. 1998, 62, 393–401.
(25) Yaylayan, V. A.; Keyhani, A. Origin of carbohydrate degradation
products in L-alanine/D-[C-13]glucose model systems. J. Agric. Food
Chem. 2000, 48, 2415–2419.
(21) Nedvidek, W.; Ledl, F.; Fischer, P. Detection of 5-hydroxymethyl-2-
methyl-3(2H)-furanone and of alpha-dicarbonyl compounds in
reaction mixtures of hexoses and pentoses with different amines.
Z. Lebensm. Unters.-Forsch. 1992, 194, 222–228.
(26) Yaylayan, V. A.; Keyhani, A. Carbohydrate and amino acid
degradation pathways in L-methionine/D-[C-13] glucose model
systems. J. Agric. Food Chem. 2001, 49, 800–803.
(22) Goto, R.; Hayami, J.; Kudo, K.; Otani, S. Studies on the chemical
decomposition of simple sugars 0.10. Acetol formation from C-14-
labeled hexoses. Bull. Chem. Soc. Jpn. 1961, 34, 753–757.
Received February 10, 2009. Revised Manuscript Received April 15,
2009.