Comparison of 2-acetylfuran formation between ribose and glucose in the Maillard reaction

Article abstract of DOI:10.1021/jf802683a

Sugar type is a major factor regulating the reaction rates and pathways in Maillard reaction. Ribose and glucose were used to compare their reactivities and pathways of 2-acetylfuran formation. A stable isotope labeling method was used to study their reactivity. A 1:1 mixture of [¹³C ₆]glucose and unlabeled ribose (or other unlabeled sugar) was reacted with proline at 145 °C for 40 min. The reactivity of each sugar was revealed by the ratio of isotopomers. The reactivity of sugars in 2-acetylfuran formation decreased in the order ribose, fructose, glucose, rhamnose, and sucrose. This method simplified the reaction system and the calculation process and gave a direct comparison of reactivity as seen via mass spectrum. The difference between glucose and ribose in 2-acetylfuran formation was that glucose could form 2-acetylfuran directly from cyclization of its intact carbon skeleton, whereas ribose first underwent degradation into fragments before forming a six-carbon unit leading to 2-acetylfuran. In the presence of cysteine, ribose could not generate 2-acetylfuran at a detectable level. When ribose was reacted with glycine, formaldehyde generated from glycine combined with ribose to form 2-acetylfuran. In other amino acids, a symmetric structure of the ribose intermediate was formed, making fragmentation more complicated.

Full text of DOI:10.1021/jf802683a

J. Agric. Food Chem. 2008, 56, 11997–12001 11997
Comparison of 2-Acetylfuran Formation between
Ribose and Glucose in the Maillard Reaction
YU WANG AND CHI-TANG HO*
Department of Food Science, Rutgers University, New Brunswick, New Jersey 08901-8520
Sugar type is a major factor regulating the reaction rates and pathways in Maillard reaction. Ribose
and glucose were used to compare their reactivities and pathways of 2-acetylfuran formation. A stable
isotope labeling method was used to study their reactivity. A 1:1 mixture of [13C6]glucose and unlabeled
ribose (or other unlabeled sugar) was reacted with proline at 145 °C for 40 min. The reactivity of
each sugar was revealed by the ratio of isotopomers. The reactivity of sugars in 2-acetylfuran formation
decreased in the order ribose, fructose, glucose, rhamnose, and sucrose. This method simplified the
reaction system and the calculation process and gave a direct comparison of reactivity as seen via
mass spectrum. The difference between glucose and ribose in 2-acetylfuran formation was that glucose
could form 2-acetylfuran directly from cyclization of its intact carbon skeleton, whereas ribose first
underwent degradation into fragments before forming a six-carbon unit leading to 2-acetylfuran. In
the presence of cysteine, ribose could not generate 2-acetylfuran at a detectable level. When ribose
was reacted with glycine, formaldehyde generated from glycine combined with ribose to form
2-acetylfuran. In other amino acids, a symmetric structure of the ribose intermediate was formed,
making fragmentation more complicated.
KEYWORDS: 2-Acetylfuran; reactivity; labeling study; ribose; glucose
INTRODUCTION
for 2-acetylfuran could be either from glucose or from glucose
and glycine (5). In the presence of phenylalanine, cysteine, or
serine, glucose can undergo cyclization and form 2-acetylfuran
via 1,4-dideoxyosone. However, when glucose reacted with
lysine, alanine, proline, or arginine, it degraded into different
fragments, which then recombined to form 2-acetylfuran.
Glycine, which is involved in 2-acetylfuran formation, degraded
into formaldehyde that then connected with a [C-5] unit of
glucose (5).
The Maillard reaction has been extensively studied for the
development of flavor and color in processed foods. In recent
years some studies have also reported the potential toxicological
effects and health problems of Maillard reaction products (1, 2).
Formation of these Maillard compounds depends on the types
of sugars and amino acids as well as reaction conditions that
may alter reaction kinetics and pathways. For example, fructose
shows a higher reactivity in 5-hydroxymethyl-2-furaldehyde
(HMF) generation compared to glucose and sucrose (3, 4), and
the formation pathways of 2-acetylfuran vary with different
amino acids (5).
2-Acetylfuran is an important sweet balsamic-cinnamic note
in natural or processed foods. It has been detected in sweet corn
products, fruits, flowers, wine, beer, and cola (6-8). In some
model system studies, such as serine/threonine/glutamine with
ribose/glucose/fructose, 2-acetylfuran was also determined to
be an important flavor compound (9).
Formation pathways of 2-acetylfuran from glucose based on
selected amino acids have been studied recently (5), but the
influence of sugars on 2-acetylfuran generation, particularly the
reactivity and mechanistic pathways from pentose, has not been
evaluated. Therefore, this study first aimed at elucidating the
reactivity of sugars in 2-acetylfuran formation and their genera-
tion pathways from ribose. Another aim of this study was to
develop a fast and visual method to compare the reactivity of
sugars. Labeled and unlabeled sugars were used together in the
same system, and reactivity order of sugars was obtained by
calculating the ratio of different isotopomers using GC-MS.
The formation pathways of 2-acetylfuran have not been
systematically studied. Yaylayan et al. (10) proposed that
2-acetylfuran was generated from 1-deoxyglucosone-6-phos-
phate during the thermal degradation of glucose-6-phosphate.
In our previous study we showed that the formation pathways
MATERIALS AND METHODS
Materials. L-Phenylalanine, L-glycine, L-cysteine, L-proline, L-lysine,
formaldehyde (37 wt % in water), phosphate buffer (pH 7.4, 0.01 M),
sodium hydroxide, anhydrous sodium sulfate, 2-acetylfuran, 2,5-
dimethyl-4-hydroxy-3(2H)-furanone, [13C6]-D-glucose, D-glucose, D-
ribose, D-fructose, L-rhamnose, and sucrose were purchased from Sigma
Chemical Co. (St. Louis, MO). [2-13C]-L-Glycine, [13C5]ribose, and
* Address correspondence to this author at the Department of Food
Science, Rutgers University, 65 Dudley Rd., New Brunswick, NJ
08901-8520 [telephone (732) 932-9611, ext. 235; fax (732) 932-6776;
e-mail ho@aesop.rutgers.edu.
10.1021/jf802683a CCC: $40.75  2008 American Chemical Society
Published on Web 11/19/2008

11998 J. Agric. Food Chem., Vol. 56, No. 24, 2008
Wang and Ho
[5-13C]ribose were obtained from Cambridge Isotope Laboratories
(Andover, MA). HPLC grade dichloromethane was purchased from
Fisher Scientific (Springfield, NJ).
Preparation of Model Systems. Labeled or unlabeled sugars and
different amino acids were dissolved separately in the phosphate buffer
(0.01 M, pH 7.4). The pH was adjusted with 1 M sodium hydroxide.
The concentrations were 0.6 and 0.2 M for sugars and amino acids,
respectively. An aliquot (1 mL) of sugar (1:1 mixture of labeled and
unlabeled sugars or only labeled sugars) was mixed separately with 1
mL of different amino acids. All samples were prepared in sealed glass
tubes, heated at 145 °C for 40 min, and then cooled in an ice bath. The
reaction mixture was extracted three times with 10 mL of dichlo-
romethane. The organic phase was separated, dried over anhydrous
sodium sulfate, and concentrated under nitrogen gas for the GC-MS
analysis. All of the models were analyzed in duplicate.
GC-MS System. The analyses of volatiles were performed with a
HP6890 gas chromatograph. An Agilent gas chromatograph (6890
series) is equipped with an autosampler (7673 series injector) and an
Agilent 5973 mass spectrometric detector (EI, 70 eV). The column
was an HP-1701 [14% (cyanopropyl-phenyl)methylpolysiloxane capil-
lary (60 m × 0.25 mm i.d., film thickness ) 0.25 µm]. The injector
was in 1:1 split mode. The constant carrier gas (helium) flow rate was
set at 1.0 mL/min. The GC oven temperature was programmed as
follows: the initial oven temperature of 40 °C was set and increased to
280 °C at a rate of 5 °C/min and then held at 280 °C for 12 min. The
total run time was 60 min. The injector and detector temperatures were
both 250 °C.
Figure 1. Comparison of reactivity between glucose and other sugars in
2-acetylfuran formation in the presence of proline: black line at 1 represents
glucose reactivity. All models were analyzed in duplicate and heated at
145 °C for 40 min. The calculation was based on mass spectral data
and equations provided under Materials and Methods.
Calculation of Labeling Percentage. The percentage of labeling
distribution for 2-acetylfuran was calculated by subtracting the natural
abundance of 13C (1.1%). All percentages below 1% were taken as
0%.
Comparison of Reactivity. Comparison of reactivity between
glucose and other sugars was based on mass spectral data. [13C6]Glucose
and other unlabeled sugars were used in a ratio of 1:1, and values of
the reactivity were calculated using the following equations:
R ) PIO/PIG
PIO ) P[12Cn] + (1/n) × P[12C1] + (2/n) × P[12C2] + ...+
((n - 1)/n) × P[12Cn-1
]
PIG ) P[13C6] + (1/6) × P[13C1] + (2/6) × P[13C2] + ...+
Figure 2. Mass spectra showing the comparison of reactivity between
ribose and glucose in 2-acetylfuran formation.
(5/6) × P[13C5]
followed by fructose at 3.3-fold, and sucrose was only 0.3-fold
of glucose. The reactivity of sugars in 2-acetylfuran formation
declined in the order ribose, fructose, glucose, rhamnose, and
sucrose. It is generally accepted that the higher the concentration
of the open-chain form of the sugar, the higher the reactivity
(13). In other words, compared to glucose, ribose and fructose
are more able to retain the open-chain form, which can then
influence the reactivity of 2-acetylfuran formation.
The uniqueness of this developed method for measuring the
reactivity of sugars in Maillard compound formation gave a
direct visual comparison via the mass spectrum of the com-
pound. A higher activity of ribose compared to glucose in
2-acetylfuran generation is clearly shown in Figure 2. Another
example of this is 6-deoxysugars, such as rhamnose, which are
more effective in forming 2,5-dimethyl-4-hydroxy-3(2H)-fura-
none (DMHF) as compared to hexoses (14). This is in
accordance with the spectrum shown in Figure 3a, where a
1:1 mixture of labeled glucose and unlabeled rhamnose was
used. In the presence of proline, the intensity of m/z 128
([12C6]DMHF) generated from [12C6]rhamnose was significantly
higher than that of m/z 134 ([13C6]DMHF) formed from
cyclization of [13C6]glucose. In addition, this method can also
be used to study the formation mechanism of a Maillard
compound. It can differentiate precursors of compounds from
different sugars. For instance, acetylformoin is considered to
R ) reactivity comparison, PIO ) percentage of isotopomers from other
sugars, PIG ) percentage of isotopomers from glucose, and P )
percentage of n-number of carbons in a specific sugar.
RESULTS AND DISCUSSION
Measuring Reactivity of Sugars on 2-Acetylfuran Forma-
tion. Maillard reaction is a complicated reaction consisting of
many parallel and consecutive reactions. Many factors such as
types of reactants and reaction conditions (temperature ranges,
pH, and water activity) play key roles in controlling the reaction
rates and mechanisms. Reactants generally regulate the type of
Maillard products and formation pathways, and reaction condi-
tions influence the kinetics (11). However, for some specific
compounds, type of sugar or amino acid can also affect the
reaction rate. For example, fructose browned more quickly than
glucose and sucrose (12). Therefore, in the current study
different sugars were used to study their reactivity in 2-acetyl-
furan formation. A novel method was developed to compare
reactivity by reacting a 1:1 mixture of [13C6]glucose and another
unlabeled sugar and then comparing the ratio of labeled and
unlabeled isotopomers. Figure 1 shows a comparison of
reactivity between glucose and other sugars in 2-acetylfuran
formation in the presence of proline. At 145 °C and pH 7.4,
the reactivity of ribose was 6.5-fold higher than that of glucose,

2-Acetylfuran Formation in the Maillard Reaction
J. Agric. Food Chem., Vol. 56, No. 24, 2008 11999
Figure 3. GC-MS spectra of 2,5-dimethyl-4-hydroxy-3(2H)-furanone (DMHF) (a) and acetylformoin (b) from a 1:1 mixture of [13C6]glucose and [12C6]rhamnose
in the presence of proline.
Table 1. Percent Labeling Distribution of 2-Acetylfuran Generated in
Different Models
be a precursor of DMHF from the degradation of hexoses (15).
In the same model of a 1:1 mixture of labeled glucose and
unlabeled rhamnose, the [M + 6] isotopomer of acetylformoin
model
M M + 1 M + 2 M + 3 M + 4 M + 5 M + 6
from glucose was observed, whereas the [M] isotopomer from
rhamnose was not identified. Thus, DMHF is formed from
rhamnose through a different precursor (Figure 3b). The same
method was used in the determination of the DMHF formation
pathway from methylglyoxal and 2-methylfuranthiol formation
from norfuraneol (16, 17).
[13C6]Glc + Rib + Cys
[5-13C]Rib + Rib + Cys
[13C6]Glc + Rib + Gly
0
0
71
0
0
7
0
0
4
4
14
44
0
0
6
7
0
3
0
0
0
0
0
0
0
0
9
0
0
0
100
0
4
13
0
0
[13C6]Glc + Rib + [2-13C]Gly 45 31
[5-13C]Rib + Rib + Pro
[5-13C]Rib + Pro
41 45
47
6
Formation of 2-Acetylfuran via Ribose. It is a challenge
to study 2-acetylfuran formation pathways from ribose,
because pentose must degrade into different fragments and
form six-carbon 2-acetylfuran via a complicated recombina-
tion reaction. Glucose has been reported to have three
pathways that form 2-acetylfuran on the basis of the type of
amino acid. In the presence of glycine, the [C-5] unit of
glucose combines with formaldehyde from glycine, leading
to 2-acetylfuran. For other amino acids, either cyclization
of intact glucose or recombination of glucose fragments can
lead to 2-acetylfuran formation (5).
reacted with cysteine, the reactivity of ribose was higher than
that of glucose, and besides the [13C6] isotopomer, all other
isotopomers formed in this reaction system could be attributed
to ribose. However, it was very interesting to observe that the
only isotopomer to form is the [13C6] isotopomer (Table 1),
indicating cysteine acts as a fragmentation inhibitor or a
scavenger. To prove the inhibition or scavenging ability of
cysteine, unlabeled ribose was reacted with cysteine, and no
2-acetylfuran was detected. The difference between glucose and
ribose in 2-acetylfuran formation was that glucose could undergo
cyclization via its intact carbon skeleton to form 2-acetylfuran,
whereas ribose must degrade into fragments before it can form
2-acetylfuran with a six-carbon unit. Thus, 2-acetylfuran could
not be formed from ribose in the presence of a fragmentation
inhibitor or scavenger.
In the presence of cysteine, a 1:1 mixture of [13C6]-2-
acetylfuran and [12C6]-2-acetylfuran is observed from a 1:1
mixture of [13C6]glucose and [12C6]glucose, suggesting no
fragmentation of glucose during 2-acetylfuran formation (5).
When a 1:1 mixture of [13C6]glucose and unlabeled ribose was

12000 J. Agric. Food Chem., Vol. 56, No. 24, 2008
Wang and Ho
Figure 4. Proposed formation of 2-acetylfuran from ribose and formaldehyde (A) and formation of formaldehyde from a symmetric structure of ribose
(B).
When glucose reacts with glycine, 2-acetylfuran is formed
from the degradation of the [C-5] unit of glucose and the
formaldehyde of glycine together. Glycine can be degraded into
formaldehyde via Strecker degradation. Further fragmentation
of the 1-deoxyglucosone can lead to the five-carbon hydroxyl-
carbonyl intermediate, which can then react with formaldehyde
via enolization, aldol condensation, and cyclization leading to
2-acetylfuran (5). Two models containing a 1:1 mixture of
[13C6]glucose and unlabeled ribose with glycine or [2-13C]glycine
were set up to study the formation of 2-acetylfuran from ribose
and glycine to determine their similarity to glucose. The
percentage of [12C6] isotopomer was higher than that of [13C6]
isotopomer, suggesting that in the presence of glycine, ribose
still had a higher reactivity. When [2-13C]glycine was used
instead of unlabeled glycine, the percentage of [13C6] isotopomer
increased 9%, whereas the percentage of [13C5] decreased 9%
(Table 1). This result indicates the involvement of one carbon
atom from glycine in the 2-acetylfuran formation from glucose,
in accordance with our previous data (5). The same result was
observed in ribose. The percentage of [12C6] isotopomer
decreased 26%, whereas [12C5] isotopomer increased 24%. To
elucidate formaldehyde reacting with ribose to form 2-acetyl-
furan, a [13C5]ribose was reacted with formaldehyde. The [M
+ 5] isotopomer of 2-acetylfuran was obtained, suggesting one
possible formation pathway of 2-acetylfuran from ribose and
formaldehyde (data not shown). After dehydration, 1-deoxy-
pentosone could react with formaldehyde from glycine to form
2-acetylfuran via enolization and aldol condensation, which is

2-Acetylfuran Formation in the Maillard Reaction
J. Agric. Food Chem., Vol. 56, No. 24, 2008 12001
(4) Lee, H. S.; Nagy, S. Relative reactivities of sugars in the formation
of 5-hydroxymethyl furfural in sugar-catalyst model systems. J.
Food Process. PreserV. 1990, 14, 171–178.
(5) Wang, Y.; Ho, C. T. Amino acid-dependent formation pathways of
2-acetylfuran and 2,5-dimethyl-4-hydroxy-3(2H)-furanone in the
Maillard reaction. Food Chem. 2008(submitted for publication).
(6) Buttery, R. G.; Stern, D. J.; Ling, L. C. Studies on flavor volatiles
of some sweet corn products. J. Agric. Food Chem. 1994, 42,
791–795.
(7) Yamaguchi, K.; Shibamoto, T. Volatile constituents of Castanopsis
flower. J. Agric. Food Chem. 1979, 27, 847–850.
(8) Yuan, J. P.; Chen, F. Separation and identification of furanic
compounds in fruit juices and drinks by high-performance liquid
chromatography photodiode array detection. J. Agric. Food Chem.
1998, 46, 1286–1291.
(9) Chen, J. H.; Ho, C. T. Comparison of volatile generation in serine/
threonine/glutamine-ribose/glucose/fructose model systems. J.
Agric. Food Chem. 1999, 47, 643–647.
(10) Yaylayan, V. A.; Machiels, D.; Istasse, L. Thermal decomposition
of specifically phosphorylated D-glucose and their role in the
control of the Maillard reaction. J. Agric. Food Chem. 2003, 51,
3358–3366.
(11) Van Boekel, M. A. J. S. Formation of flavour compounds in the
Maillard reaction. Biotechnol. AdV. 2006, 24, 230–233.
(12) Reyes, F. G. R.; Poocharoen, B.; Wrolstad, R. Maillard browning
reaction of sugar-glycine model systems: changes in sugar
concentration, color and appearance. J. Food Sci. 1982, 47, 31376–
31377.
(13) Van Boekel, M. A. J. S. Kinetic aspects of the Maillard reaction:
a critical review. Nahrung/Food 2001, 45, 150–159.
(14) Hofmann, T.; Schieberle, P. Identification of potent aroma
compounds in thermally treated mixtures of glucose/cysteine and
rhamnose/cysteine using aroma extract dilution techniques. J.
Agric. Food Chem. 1997, 45, 898–906.
similar to the proposed pathway for the formation of 2-acetyl-
furan from glucose (Figure 4A).
The third type of 2-acetylfuran formation from ribose was
fragmentation and recombination. A 1:1 mixture of [5-13C]ribose
and unlabeled ribose was reacted with proline. A 1:1 mixture
of [M] and [M + 1] isotopomers was observed, and the
percentage of [M + 2] isotopomer was found to be only 14%
(Table 1). These data were not enough to speculate on the
formation pathways. First, a 1:1 mixture of [5-13C]ribose and
unlabeled ribose was reacted with either phenylalanine or lysine;
the percentages of isotopomers distribution were the same as
when reacted with proline, suggesting that, with the exception
of cysteine and glycine, other amino acids do not have a great
influence on the fragmentation of ribose (data not shown).
Therefore, proline was used as a representative. Second, only
[5-13C]ribose mixed with proline. A 1:1 mixture of [M + 1]
and [M + 2] isotopomers is shown (Table 1) and compared
with the model of a 1:1 mixture of [5-13C]ribose and unlabeled
ribose (Table 1), indicating that the possibility of breakages of
ribose from C1-C2 or C2-C3 was equal to that from C3-C4
or C4-C5, respectively. In other words, a symmetric structure
occurred from ribose during 2-acetylfuran formation. One
possible pathway is proposed in Figure 4B. Ribose was
rearranged into 3-pentulose through keto-enol tautomerism. By
retro-aldol condensation, the carbon at either position 1 or
position 5 can be cleaved into formaldehyde, which could react
with ribose to form 2-acetylfuran.
In conclusion, the results of this study showed the difference
between ribose and glucose in 2-acetylfuran formation from
reaction reactivity and mechanism pathways. The reactivity of
ribose in 2-acetylfuran formation was higher than that of
glucose. In the presence of cysteine, 2-acetylfuran could not be
generated from ribose. When ribose was reacted with glycine,
formaldehyde generated from glycine was involved in the
formation of 2-acetylfuran. For other amino acids, a symmetric
structure of ribose formed.
(15) Hofmann, T.; Schieberle, P. Acetylformoinsan important progeni-
tor of 4-hydroxy-2,5-dimethyl-3(2H)-furanone and 2-acetyltet-
rahydropyridine during thermal food processing. In FlaVour 2000,
Perception Release EValuation Formation Acceptance Nutrition/
Health, Proceedings of the 6th Wartburg Aroma Symposium;
Rothe, M., Ed.; Eisenach, Germany, 2001; pp 311-322.
(16) Wang, Y.; Ho, C. T. Formation of 2,5-dimethyl-4-hydroxy-3(2H)-
furanone through methylglyoxal: a Maillard reaction intermediate.
J. Agric. Food Chem. 2008, 54, 7405–7409.
LITERATURE CITED
(1) Mottram, D. S.; Wedzicha, B. I.; Dodson, A. T. Acrylamide is
formed in the Maillard reaction. Nature 2002, 419, 448–449.
(2) Lo, C. Y.; Li, S.; Wang, Y.; Tan, D.; Pan, M. H.; Sang, S.; Ho,
C. T. Reactive dicarbonyl compounds and 5-(hydroxymethyl)-2-
furfural in carbonated beverages containing high fructose corn
syrup. Food Chem. 2008, 107, 1099–1105.
(3) Antal, M. J.; Mok, W. S. L.; Richards, G. N. Mechanism of
formation of 5-hydroxymethy)-2-furaldehyde from D-fructose and
sucrose. Carbohydr. Res. 1990, 199, 91–109.
(17) Cerny, C.; Davidek, T. Formation of aroma compounds from
ribose and cysteine during the Maillard reaction. J. Agric. Food
Chem. 2003, 51, 2714–2721.
Received for review August 30, 2008. Revised manuscript received
November 8, 2008. Accepted November 9, 2008.
JF802683A