Conversion of 5-hydroxymethylfurfural into 6-(hydroxymethyl)pyridin-3-ol: A pathway for the formation of pyridin-3-ols in honey and model systems

Article abstract of DOI:10.1021/acs.jafc.0c01679

The formation of 6-(hydroxymethyl)pyridin-3-ol by ring expansion of 5-(hydroxymethyl)furfural (HMF) in the presence of ammonia-producing compounds was studied to determine the routes of formation of pyridin-3-ols in foods. 6-(Hydroxymethyl)pyridin-3-ol was produced from HMF in model systems, mostly at neutral pH values, as a function of reaction times and temperature and with an activation energy (E_a) of 74 ± 3 kJ/mol, which was higher than that of HMF disappearance (43 ± 4 kJ/mol). A reaction pathway is proposed, which is general for the formation of pyridin-3-ols from 2-oxofurans. Thus, it explains the conversions of furfural into pyridin-3-ol and of 2-acetylfuran into 2-methylpyridin-3-ol, which were also studied. When honey and sugarcane honey were heated, they produced different pyridin-3-ols, although 6-(hydroxymethyl)pyridin-3-ol was the pyridine-3-ol produced to the highest extent. Obtained results suggest that formation of pyridin-3-ols in foods is unavoidable when 2-oxofurans are submitted to thermal heating and ammonia (or ammonia-producing compounds) is present.

Full text of DOI:10.1021/acs.jafc.0c01679

Subscriber access provided by BIU Pharmacie | Faculté de Pharmacie, Université Paris V
Food and Beverage Chemistry/Biochemistry
Conversion of 5-Hydroxymethylfurfural into 6-(Hydroxymethyl)pyridin-3-ol:
A Pathway for the Formation of Pyridin-3-ols in Honey and Model Systems
Francisco J. Hidalgo, Cristina M. Lavado-Tena, and Rosario Zamora
J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.0c01679 • Publication Date (Web): 22 Apr 2020
Downloaded from pubs.acs.org on April 23, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 34
Journal of Agricultural and Food Chemistry
Conversion of 5-Hydroxymethylfurfural into 6-
(Hydroxymethyl)pyridin-3-ol: A Pathway for the Formation of
Pyridin-3-ols in Honey and Model Systems
Francisco J. Hidalgo, Cristina M. Lavado-Tena, and Rosario Zamora*
Instituto de la Grasa, Consejo Superior de Investigaciones Científicas, Carretera de
Utrera km 1, Campus Universitario–Edificio 46, 41013-Seville, Spain
*Corresponding author:
Rosario Zamora
+34954611550
+34954616790
rzamora@ig.csic.es
Phone:
Fax:
e-mail:
1
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 2 of 34
1
ABSTRACT
2
3
The formation of 6-(hydroxymethyl)pyridin-3-ol by ring expansion of 5-
(hydroxymethy)furfural (HMF) in the presence of ammonia-producing compounds was
studied to determine routes of formation of pyridin-3-ols in foods. 6-
(Hydroxymethyl)pyridin-3-ol was produced from HMF in model systems, mostly at
neutral pH values, as a function of reaction times and temperature, and with an activation
energy (Ea) of 74  3 kJ/mol, which was higher than that of HMF disappearance (43  4
kJ/mol). A reaction pathway is proposed, which is general for the formation of pyridin-
3-ols from 2-oxofurans. Thus, it explains the conversions of furfural into pyridin-3-ol and
of 2-acetylfuran into 2-methylpyridin-3-ol, which were also studied. When honey and
sugarcane honey were heated, they produced different pyridin-3-ols, although 6-
(hydroxymethyl)pyridin-3-ol was the pyridine-3-ol produced to the highest extent.
Obtained results suggest that formation of pyridin-3-ols in foods is unavoidable when 2-
oxofurans are submitted to thermal heating and ammonia (or ammonia-producing
compounds) is present.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
KEYWORDS: 2-Acetylfuran; Carbonyl-amine reactions; Furfural; Honey; 5-
Hydroxymethylfurfural; Maillard reaction; Pyridin-3-ols; Reactive carbonyls
19
2
ACS Paragon Plus Environment

Page 3 of 34
Journal of Agricultural and Food Chemistry
20
INTRODUCTION
21
22
23
24
25
26
27
28
29
Maillard reaction is a major route for the formation of aromas as a consequence of
food processing.1–3 The compounds responsible for these aromas have a great diversity
of structures. Thus, the formation of aldehydes, alcohols, thiols, sulfides, acids, lactones,
and a large variety of heterocyclic compounds, has been described.4 Among these last
compounds, the stale flavors of furan-2-carbaldehydes,5 or the roasted flavors of
pyrazines,6 for example, have been widely studied. However, some of the compounds
responsible for the produced aromas are not always stable and aroma changes are usually
observed with further processing or storage. Many of these changes are still very poorly
understood, which difficulties the prediction of the evolution of food aromas.
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
Among the different flavor-related substances produced in foods as a consequence of
Maillard reaction, 5-hydroxymethylfurfural (HMF) is an important compound commonly
produced to significant extents. Its flavor has been described as caramel, waxy, fatty,
musty, or cardboard. HMF is considered as an endogenous contaminant because of its
potential adverse effects,7 and it is also a quality indicator as its levels in foods provide
evidence of overheating during processing or inadequate storage conditions.8 In addition,
it has also been associated with color development in some foods.9 This compound has
been shown to be thermally unstable. Thus, its conversion upon heating into 5,5’-oxy-
dimethylene-bis(2-furaldehyde), 5-methylfurfural, 2,5-furandicarboxaldehyde, or 2-
acetyl-5-methylfuran, among others, has been described.10,11 In addition, diverse
chemical syntheses suggest that conversion of HMF into 6-(hydroxymethyl)pyridin-3-ol
is possible. To this respect, Müller et al.12 showed that brief exposure to bromine in water-
methanol at 0 ºC smoothly and effectively converts hydroxymethylfurfurylamines,
originated from either fructose or inuline sources, into 6-substituted pyridin-3-ols.
Furthermore, formation of pyridin-3-ols by ring expansion 2-acylfurans is well-known in
3
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 4 of 34
45
46
47
48
49
organic synthesis.13,14 Nevertheless, to the best of our knowledge, a relationship between
pyridin-3-ols, like 6-(hydroxymethyl)pyridin-3-ol, and 2-oxofurans, like HMF, has not
been established so far in foods. However, 6-(hydroxymethyl)pyridin-3-ol, like other
pyridin-3-ols, has been found in the hydrolysis of melanoidins15 or as a product of
Maillard reaction.16
50
51
52
53
54
55
56
57
58
59
Pyridines are typical food flavors, which are produced as a consequence of
processing.17 The formation pathways of alkylpyridines seem to be related to the
cyclizations and oligomerizations suffered by lipid-derived aldehydes in the presence of
ammonia-producing compounds.18 In addition, formation of alkylpyridiniums has been
described in roasted coffee by trigonelline decomposition.19 On the other hand, the
formation pathways of pyridin-3-ols are lesser known in spite of being pyridin-3-ol a
common flavor in coffee,20 tea,21 or caramel colors,22 for example. In an attempt to
explore the role of 2-oxofurans in the appearance of pyridin-3-ols in foods, this study
investigates the formation of 6-(hydroxymethyl)pyridin-3-ol, pyridin-3-ol, and 2-
methylpyridin-3-ol in honeys and model systems.
60
MATERIALS AND METHODS
61
62
63
64
65
66
67
68
69
Materials and chemicals. As model food products, two kinds of honeys were studied.
One of them was a high quality multi-floral honey. The other one was a traditional plant
syrup, namely ‘miel de caña’ (sugarcane honey or sugarcane syrup). This syrup is
traditionally manufactured in Andalusia (Spain) from crushed sugarcane (Saccharum
officinarum) and, although produced only in limited quantities, it is highly appreciated by
consumers. For its preparation, the fresh cane juice is heated rapidly for decantation, then
filtered, and, finally, slowly heated until it looks like dark honey. A characterization of
this syrup was described previously.23 Both products were purchased at local
supermarkets.
4
ACS Paragon Plus Environment

Page 5 of 34
Journal of Agricultural and Food Chemistry
70
The tested pyridin-3-ols were pyridin-3-ol, 2-methylpyridin-3-ol, and 6-
(hydroxymethyl)pyridin-3-ol. As 2-oxofurans, HMF, furfural, and 2-acetylfuran were
employed. These and other chemicals employed in this study had the highest available
grade and were purchased from reliable commercial sources, including Sigma-Aldrich
(St. Louis, MO), Fluka (Buchs, Switzerland), and Alfa Aesar (Thermo Fischer GmbH,
Karlsruhe, Germany).
71
72
73
74
75
76
77
78
Conversion of 2-Oxofurans into Pyridin-3-ols in Model Systems. Mixtures of the
2-oxofuran (10 mol dissolved in 20 L of methanol) and an ammonia-producing
compound (30 mol in 50 L of water) were singly homogenized with 200 mg of 0.063–
79
80
81
82
83
0.200 mm silica gel (Macherey-Nagel, Düren, Germany), and 30 L of 0.3 M buffer, pH
3–10. Samples were heated in closed test tubes for the indicated times (0–22 h) and
temperatures (100-180 ºC). The employed buffers were sodium citrate, pH 3–6, sodium
phosphate, pH 6–8, and sodium borate, pH 8–10. Times and temperatures employed in
each experience are indicated in the respective figure legends.
84
85
86
87
88
89
At the end of the heating period, samples were cooled and 1 mL of methanol and 20
L of the internal standard solution (13 mg of 1-octadecanol in 25 mL of methanol) were
added. Suspensions were stirred for 1 min and centrifuged for 5 min at 2000 g. The
supernatant was collected, evaporated to dryness, and the residue was successively treated
with 200 L of N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA), heated for 30 min at
60 ºC, and studied by gas chromatography coupled to mass spectrometry (GC-MS).
90
91
The 2-oxofurans tested were HMF, furfural, and 2-acetylfuran. The assayed ammonia-
producing compounds were ammonia and ammonium chloride.
92
93
Formation of Pyridin-3-ols in Honeys. Honeys (1 g) were heated in closed test tubes
for 0–10 h at either 60 or 100 ºC. After cooling, 100 L of water was added and the
5
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 6 of 34
94
95
96
97
98
99
mixture was stirred. Then, 4 mL of acetonitrile and 20 L of the internal standard solution
(13 mg of 1-octadecanol in 25 mL of methanol) were added. Mixtures were successively
stirred for 1 min, sonicated for 10 min, stirred for 1 min, and the acetonitrile layer was
decanted and taken to dryness. The residue was treated with 300 L of N,O-
bis(trimethylsilyl)trifluoroacetamide (BSTFA), heated for 30 min at 60 ºC, and studied
by gas chromatography coupled to mass spectrometry (GC-MS).
100
101
102
103
104
105
106
107
108
GC-MS Analyses. GC-MS analyses were conducted with an Agilent 7820A gas
chromatograph coupled with an Agilent 5977 mass selective detector (quadrupole type)
using a fused-silica HP-5MS UI capillary column (30 m length, 0.25 mm inner diameter,
0.25 m coating thickness) from Agilent. One microliter of sample was injected in the
pulsed splitless mode. Working conditions were as follows: carrier gas, helium (1 mL/min
at constant flow); injector, 250 ºC; transfer line to mass selective detector, 280 ºC;
electron ionization (EI), 70 eV; ion source temperature, 230 ºC; and mass range, 28-550
amu. Oven temperature conditions were from 80 ºC (1 min) to 140 ºC at 20 ºC/min, then
to 300 ºC at 50 ºC/min, and finally held at 300 ºC for 4 min.
109
110
111
112
113
114
115
116
Identification of 6-(Hydroxymethyl)pyridin-3-ol, Pyridin-3-ol, 2-Methylpyridin-
3-ol and HMF. Identification of the three pyridin-3-ols and HMF was carried out by
comparison of retention indexes and mass spectra, and by co-elution with authentic
standards, which were derivatized analogously. Mass spectra of the assayed compounds
as well as that of the internal standard are collected in the Supporting Information. The
following ions (M+ – methyl) were employed for quantitation purposes: m/z 254 for 6-
(hydroxymethyl)pyridin-3-ol, m/z 152 for pyridin-3-ol, m/z 166 for 2-methylpyridin-3-ol,
m/z 183 for HMF, and m/z 327 for the IS.
117
118
Quantitation of 6-(Hydroxymethyl)pyridin-3-ol, Pyridin-3-ol, 2-Methylpyridin-
3-ol and HMF. Quantitation of the three pyridin-3-ols and HMF in model systems and
6
ACS Paragon Plus Environment

Page 7 of 34
Journal of Agricultural and Food Chemistry
119
honeys was carried out by preparing standard curves of those compounds in the
corresponding assayed systems and following the whole procedures described above
(without heating). For each curve, six different concentration levels of the three pyridin-
3-ols and HMF were used. Pyridin-3-ols and HMF contents were directly proportional to
the corresponding compound/internal standard area ratios (r > 0.99, p < 0.001). The
coefficients of variation at the different assayed concentrations were < 10%.
120
121
122
123
124
125
126
127
128
129
130
Statistical Analysis. All data given are mean  standard deviation (SD) of at least
three independent experiments. Analysis of variance was employed to compare the
different groups. When significant F values were obtained, group differences were
evaluated by the Tukey test.24 These studies were conducted using Origin version 7.0
(OriginLab Corp., Northampton, MA, USA). The significance level is p < 0.05 unless
otherwise indicated.
131
RESULTS AND DISCUSSION
132
133
134
135
136
Effect of Reaction Conditions on the Conversion of HMF into 6-
(Hydroxymethyl)pyridin-3-ol. When HMF was heated in the presence of either
ammonia or ammonium chloride, the formation of 6-(hydroxymethyl)pyridin-3-ol was
observed. The conversion yield depended on the reaction conditions, including reaction
pH, reactant concentrations, time and temperature.
137
138
139
140
141
142
Figure 1 shows the effect of pH on both the formation of 6-(hydroxymethyl)pyridin-
3-ol and the disappearance of HMF. As observed, HMF disappearance and 6-
(hydroxymethyl)pyridin-3-ol formation increased when the pH increased from 3 to 7. The
maximum amount of 6-(hydroxymethyl)pyridin-3-ol was produced at pH 6–8 with
sodium phosphate buffer. At higher pH values and in the presence of borate buffer, the
amount of the produced 6-(hydroxymethyl)pyridin-3-ol was much lower. For some
7
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 8 of 34
143
144
145
146
147
148
149
150
unknown reason, HMF disappearance also occurred at pH 8–10 in the presence of sodium
borate buffer, but this disappearance did not produce the pyridine to a high extent.
Between pH 3 and pH 8 the formation of 6-(hydroxymethyl)pyridin-3-ol and the
disappearance of HMF were correlated (r = 0.96, p = 0.002). The conversion yield of the
furan into the pyridine depended on the pH and ranged from 0.4% at pH 3 to 1.3% at pH
7–8. Although there was not a significant difference among the amounts of 6-
(hydroxymethyl)pyridin-3-ol produced at either pH 7 or pH 8, pH 8 was selected for the
rest of this study because the minimum amount of HMF was recovered at this pH.
151
152
153
154
6-(Hydroxymethyl)pyridin-3-ol formation also depended on the HMF/ammonia ratio.
Figure 2 shows the effect of increasing amounts of HMF on the formation of 6-
(hydroxymethyl)pyridin-3-ol in the presence of ammonium chloride. The maximum
amount of the pyridine was achieved when 10 mol of HMF were heated in the presence
155
156
157
158
159
of 30 mol of ammonium chloride. Nevertheless, the conversion yield was higher when
smaller amounts of HMF were employed. Thus, conversion yield was 3% with 2.5 mol
of HMF, and this yield decreased when higher amounts of HMF were employed. Most of
this study was carried out with 10 mol of HMF because the highest amount of 6-
(hydroxymethyl)pyridin-3-ol was obtained by using these conditions.
160
161
162
163
Conversion yield also depended on the concentration of the ammonia-producing
compound. Figure 3 shows the effect of increasing amounts of ammonia and ammonium
chloride on the formation of 6-(hydroxymethyl)pyridin-3-ol. The amount of this pyridine
increased linearly (r > 0.9993, p < 0.0007) between 0 and 20 mol of the amino compound
164
165
166
and it remained constant when more than 30 mol of the amino compound was added.
The behavior was inverse for the HMF recovered. It decreased linearly (r = 0.996, p <
0.004) between 0 and 20 mol of the amino compound and it remained constant when
8
ACS Paragon Plus Environment

Page 9 of 34
Journal of Agricultural and Food Chemistry
167
more than 30 mol of the amino compound was added. In fact, formation of 6-
(hydroxymethyl)pyridin-3-ol and disappearance of HMF were correlated (r = 0.98, p <
0.0001). Moreover, the behavior of the two amino compounds assayed (ammonia and
ammonium chloride) was very similar. Both of them produced analogous disappearance
of HMF, although the conversion yield for 6-(hydroxymethyl)pyridin-3-ol seemed to be
slightly higher when ammonium chloride was employed. For that reason, ammonium
chloride was employed as ammonia-producing compound in most of the experiments
carried out in this study. In addition, 30 mol of ammonium chloride was selected because
this was the minimum amount of the ammonia-producing compound that produced the
pyridine to a higher extent when starting from 10 mol of HMF.
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
Finally, time and temperature also played a major role on the conversion of HMF into
6-(hydroxymethyl)pyridin-3-ol. Figure 4 shows the time-courses for the formation of 6-
(hydroxymethyl)pyridin-3-ol at 100–160 ºC. As observed, the amount of the pyridine
increased linearly (r = 0.91, p < 0.04) as a function of heating time. Furthermore,
formation rates, obtained from the slopes of the lines of the best fit, increased with
temperature. In addition, there was a small lag period, which was higher when
temperature was lower. Reaction rates for this formation were employed in an Arrhenius
plot to obtain the activation energy (Ea) for the formation of 6-(hydroxymethyl)pyridin-
3-ol from HMF. The obtained plot is shown in Figure 5. The activation energy was
obtained from the slope of the line of best fit and resulted to be 74  3 kJ/mol.
187
188
189
190
Similarly, time-courses for the disappearance of HMF at 100–160 ºC are shown in
Figure 6. Recovered HMF decreased linearly (r > 0.99, p < 0.04) as a function of heating
times and disappearance rates increased with temperature. Disappearance rates were
employed in an Arrhenius plot for determining the Ea for HMF disappearance. It was 43
9
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 10 of 34
191
192
193
 4 kJ/mol. This value was lower than that found above for 6-(hydroxymethyl)pyridin-3-
ol formation. This explains that HMF is easily decomposed, as observed previously,10 and
it is only partially converted into 6-(hydroxymethyl)pyridin-3-ol.
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
Conversion of 2-Oxofurans into Pyridin-3-ols. The conversion of HMF into the
corresponding pyridin-3-ol is not exclusive for this furan. To confirm that other 2-
oxofurans can suffer analogous transformations, the conversion of other two products of
the Maillard reaction (furfural25 and 2-acetylfuran26) into the corresponding pyridin-3-ols
(pyridin-3-ol and 2-methylpyridin-3-ol, respectively) was studied. Figure 7 shows the
formation of the corresponding pyridin-3-ols after 3 and 22 h at 100 ºC. The figure also
shows the formation of 6-(hydroxymethyl)pyridin-3-ol by heating of HMF under the
same reaction conditions for comparison purposes. As expected, the corresponding
pyridin-3-ols were always produced, although the conversion yield depended on the
employed 2-oxofuran. Thus, the highest yield was observed for 6-
(hydroxymethyl)pyridin-3-ol. This pyridine was produced with a yield of 0.5% after 3 h
at 100 ºC, and the yield increased to 1.3% after 22 h at 100 ºC. The yields of the other
two pyridines were similar among them and were lower than those of 6-
(hydroxymethyl)pyridin-3-ol. The pyridine that was produced to a higher extent was
pyridin-3-ol (0.03% and 0.11% after 3 and 22 h, respectively). This pyridine has been
found in different foods that had suffered Maillard reaction.16–18 The yield for 2-
methylpyridin-3-ol was 0.01% and 0.06% after 3 and 22 h, respectively.
211
212
213
214
215
Proposed Pathway for the Conversion of 2-Oxofurans into Pyridin-3-ols. Above
described results suggest that, in the presence of ammonia and moderate temperatures, 2-
oxofurans are decomposed and the formation of pyridin-3-ols is produced. A pathway
that explains this transformation is shown in Figure 8. This reaction pathway is based on
that previously published by Gruber.14 Thus, in the presence of ammonia, the
10
ACS Paragon Plus Environment

Page 11 of 34
Journal of Agricultural and Food Chemistry
216
217
218
219
220
221
corresponding imine should be produced in a first step. Then, the addition of a second
molecule of ammonia would produce the opening of the ring with the formation of a
compound derived from the dicarbonyl precursor of the furan. This attack is consequence
of the polarization of the C–O bond in the furan ring because of the difference of
electronegativities between both atoms.27 Finally, cyclization of this intermediate with
the exit of a molecule of ammonia would produce the corresponding pyridine derivative.
222
223
224
225
226
227
228
229
230
231
232
This reaction only requires the presence of an oxo group at position 2 of the furan ring.
This means that it should be produced with any other 2-oxofuran that might be present.
This explains, for example, the presence of both 6-methylpyridin-3-ol and 5-
methylfurfural in coffee aroma.28 In addition, it could also be expected to be produced in
analogous 2-oxoheterocycles. For example, 2-acetylthiophene is a well-known Maillard
reaction product.29 According to the proposed reaction pathway, it should be converted
to a certain extent into 2-methylpyridin-3-thiol. To the best of our knowledge, this flavor
compound has not been yet identified in foods. However, other Maillard reaction product,
2-thiophenecarboxaldehyde, which is present in coffee30 or pork broth,31 for example,
should produce pyridin-3-thiol. This last pyridine has been identified in soy sauce, which
is also a source of 2-acetylthiophene.32
233
234
235
236
237
238
239
240
Although the presence of the oxo group is a requisite for the reaction, reaction yield
also depended on the existence of substituents in the carbon at position 5 of the furan ring.
The hydroxyl group of the hydroxymethyl substituent at position 5 of HMF has a negative
inductive effect. This effect should contribute to increase the electron deficiency of
carbon at position 5 of the furan ring, therefore converting this atom into a better
electrophile. For this reason, HMF is converted more easily into 6-
(hydroxymethyl)pyridin-3-ol than furfural into pyridin-3-ol. This suggests that, although
pyridin-3-ol is more frequently described than 6-(hydroxymethyl)pyridin-3-ol in most
11
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 12 of 34
241
242
243
244
foods, this last pyridine is likely to be present to a higher extent, but it has mostly gone
unnoticed until now. To confirm this last hypothesis, and to test the formation of pyridin-
3-ols in food products, the effect of thermal heating on commercial honeys was studied.
Obtained results will be discussed in the next subsection.
245
246
247
248
249
Formation of Pyridin-3-ols in Honeys. Two different types of honeys were studied.
One of them was a high quality honey, which had not suffered any previous thermal
overheating. The other one was a sugarcane honey. As described in the Materials and
Methods section, the production of this last honey requires thermal treatment that is likely
to promote Maillard reaction to a high extent.
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
Figure 9 shows the effect of heating on the formation of both HMF and 6-
(hydroxymethyl)pyridin-3-ol in the studied honey. As expected, because of its high
quality, very small amounts of HMF and 6-(hydroxymethyl)pyridin-3-ol were detected in
this honey at the initial step. However, when the honey was heated, both compounds were
produced to a significant extent. On the other hand, formation of pyridin-3-ol was not
observed, and only trace amounts of 2-methylpyridin-3-ol were detected when the honey
was heated at 100 ºC for long heating periods. This indicated that, similarly to that
observed in model systems, 6-(hydroxymethyl)pyridin-3-ol is produced to a higher extent
than other pyridines. In addition, HMF and 6-(hydroxymethyl)pyridin-3-ol formation
time-courses were different. Thus, HMF increased exponentially as a function of heating
time, as observed previously.33 On the other hand, 6-(hydroxymethyl)pyridin-3-ol
concentration increased linearly (r = 0.94, p < 0.0002). This difference of time-courses
does not necessarily mean that 6-(hydroxymethyl)pyridin-3-ol is not produced by ring
expansion of HMF. As observed in the studied model systems (Figure 2), increased
amounts of HMF do not always produce increasing amounts of 6-
(hydroxymethyl)pyridin-3-ol. This conversion requires the presence of ammonia (or other
12
ACS Paragon Plus Environment

Page 13 of 34
Journal of Agricultural and Food Chemistry
266
267
268
ammonia-producing compound) that can be the limiting factor in this conversion. In
addition, and analogously to that observed in model systems, HMF disappearance should
produce other compounds in addition to 6-(hydroxymethyl)pyridin-3-ol.
269
270
271
272
273
274
275
276
277
Results obtained with the sugarcane honey studied were someway different. Because
this food product suffers a heating process during its production, both HMF and 6-
(hydroxymethyl)pyridin-3-ol were already present in the initial sample. In addition, both
of them increased linearly (r > 0.93, p < 0.0008) when the sugarcane honey was heated.
In fact, there was a correlation between them at both 60 ºC (r > 0.97, p = 0.005) and 100
ºC (r > 0.97, p < 0.0001). Surprisingly, 6-(hydroxymethyl)pyridin-3-ol/HMF ratio was
higher at 60 ºC (0.27) than at 100 ºC (0.034). This suggests that disappearance of HMF
at low temperature mainly produces the pyridine. However, higher temperatures favor the
transformation of HMF into other compounds.
278
279
280
281
282
283
Differently to the studied honey, the studied sugarcane honey also contained the other
analyzed pyridines (pyridin-3-ol and 2-methylpyridin-3-ol) at the initial step. In addition,
concentrations of both pyridines increased linearly (r > 0.90, p < 0.002) as a function of
heating. However, and analogously to the observed in the studied model systems and the
studied honey, the concentrations of both pyridines were always lower than that of 6-
(hydroxymethyl)pyridin-3-ol.
284
285
286
287
288
289
Obtained results suggest that, analogously to that observed in honey, sugarcane honey,
and model systems, conversion of 2-oxofurans into pyridin-3-ols is unavoidable when 2-
oxofurans are submitted to thermal heating and ammonia (or an ammonia-producing
compound) is present. This suggests that, although not frequently found in foods, pyridin-
3-ols should be common components of foods in which Maillard reaction has occurred,
even in those foods submitted to soft heating or stored for a limited time period.
13
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 14 of 34
290
291
292
293
294
295
296
297
298
299
Obtained results also show that 6-(hydroxymethyl)pyridin-3-ol is an additional,
previously unknown in foods, transformation product of HMF. Some authors have shown
doubts about the usefulness of using HMF as a quality indicator of inadequately processed
foods because of its instability,9 although in many food systems, after a short lag time, a
constant increase in HMF concentration is observed, which indicates a faster formation
than degradation. Nevertheless, because aromaticity of pyridines is usually higher than
that of furans,35 6-(hydroxymethyl)pyridin-3-ol is expected to be more stable than HMF.
Therefore, although reaction yield of 6-(hydroxymethyl)pyridin-3-ol is much lower than
that of HMF, the potential use of 6-(hydroxymethyl)pyridin-3-ol as an alternative marker
of thermal stress in some foods might be suggested.
300
301
302
303
304
305
Finally, a last consequence of the described reaction is related to the observed changes
in the antioxidant activity of honeys submitted to thermal heating.34,35 As described
above, 6-(hydroxymethyl)pyridin-3-ol increases upon heating, and pyridin-3-ols have
been shown to be efficient chain-breaking antioxidants.36 The potential contribution of 6-
(hydroxymethyl)pyridin-3-ol to the observed increase of antioxidant activity in honey
might be hypothesized.
306
307
308
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge at
309
310
Mass spectra of trimetylsilyl derivatives of pyridin-3-ol, 2-methylpyridin-3-ol, 6-
(hydroxymethyl)pyridin-3-ol, 5-hydroxymethylfurfural (HMF), and 1-octadecanol (PDF)
311
312
AUTHOR INFORMATION
Corresponding author
14
ACS Paragon Plus Environment

Page 15 of 34
Journal of Agricultural and Food Chemistry
313
314
*Telephone: +34 954 611 550. Fax: +34 954 616 790. E-mail: rzamora@ig.csic.es.
Funding
315
316
317
318
This study was supported by the Ministerio de Ciencia, Innovación y Universidades
(MCIU) from Spain, the Agencia Estatal de Investigación (AEI) from Spain, and the
Fondo Europeo de Desarrollo Regional (FEDER) from the European Union (Project
RTI2018-096632-B-100).
319
320
321
Notes
The authors declare no competing financial interest.
15
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 16 of 34
322
REFERENCES
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
(1) McGorrin R. J. Key aroma compounds in oats and oat cereals. J. Agric. Food Chem.
2019, 67, 13778–13789.
(2) Poisson, L.; Schaerer, A.; Spreng, S.; Mestdagh, F.; Blank, I.; Davidek, T.
Generation of -diketones and 4-hydroxy-2,5-dimethyl-3(2H)-furanone upon
coffee roasting–Impact of roast degree on reaction pathways. J. Agric. Food Chem.
2019, 67, 13829–13839.
(3) Brehm, L.; Frank, O.; Junger, M.; Wimmer, M.; Ranner, J., Hofmann, T. Novel
taste-enhancing 4-amino-2-methyl-5-heteroalkylpyridines formed from thiamine
by Maillard-type reactions. J. Agric. Food Chem. 2019, 67, 13986–13997.
(4) Starowicz, M.; Zielinski, H. How Maillard reaction influences sensorial properties
(color, flavor and texture) of food products? Food Rev. Int. 2019, 35, 707–725.
(5) Vrzal, T.; Sterba, K.; Jurková, M.; Olsovska, J. The usage of a reflectometric
method for 5-(hydroxymethyl)furan-2-carbaldehyde determination as a stale flavor
sensor for beer. Food Packag. Shelf Life 2019, 19, 1–6.
(6) Liu, C. J.; Yang, Q.; Linforth, R.; Fisk, I. D.; Yang, N. Modifying Robusta coffee
aroma by green bean chemical pre-treatment. Food Chem. 2019, 272, 251–257.
(7) Zeng, R.; Zhang, G.; Zheng, J.; Zhou, H.; Wang, Y.; Huang, C.; Hu, W.; Ou, S.
Formation and identification of two hydroxymethylfurfural-glycine adducts and
their cytotoxicity and absorption in Caco-2 cells. J. Agric. Food Chem. 2020, 68,
384–389.
(8) Lee, C.-H.; Chen, K.-T.; Lin, J.-A.; Chen, Y.-T-; Chen, Y.-A.; Wu, J.-T.; Hsieh,
C.-W. Recent advances in processing technology to reduce 5-
hydroxymethylfurfural in foods. Trends Food Sci. Technol. 2019, 93, 271–280.
16
ACS Paragon Plus Environment

Page 17 of 34
Journal of Agricultural and Food Chemistry
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
(9) Francisquini, J. D.; Rocha, J.; Martins, E.; Stephani, R.; da Silva, P. H. F.; Renhe,
I. R. T.; Perrone, I. T.; de Carvalho, A. F. 5-Hydroxymethylfurfural formation and
color change in lactose-hydrolyzed Dulce de leche. J. Dairy Res. 2019, 477–482.
(10) Chambel, P., Oliveira, M. B., Andrade, P. B., Fernandes, J. O., Seabra, R. M.,
Ferreira, M. A. Identification of 5,5’-oxy-dimethylene-bis(2-furaldehyde) by
thermal decomposition of 5-hydroxymethyl-2-furfuraldehyde. Food Chem. 1998,
63, 473–477.
(11) Nikolov, P. Y.; Yayalayan, V. A. Thermal decomposition of 5-(hydroxymethyl)-2-
furaldehyde (HMF) and its further transformations in the presence of glycine. J.
Agric. Food Chem. 2011, 59, 10104–10113.
(12) Müller, C.; Diehl, V.; Lichtenthaler, F. W. Building blocks from sugars. Part 23.
Hydrophilic-3-pyridinols from fructose and isomaltulose. Tetrahedron 1998, 54,
10703–10712.
(13) Leditschke, H. die Synthese von 3-Oxyl-2-aryl-pyridinen aus Phenyl--furyl-
ketonen. Chem. Ber 1952, 85, 202–204.
(14) Gruber, W. Synthesis of 3-hydroxy-2-alkylpyridines. Can. J. Chem. 1953, 31, 564–
568.
(15) Tsuchida, H.; Komoto, M.; Kato, H.; Fujimaki, M. Isolation and identification of
two -hydroxy-pyridine derivatives from the nondialyzable melanoidin-
hydrolyzate. Agric. Biol. Chem. 1973, 37, 403–409.
(16) Olsson, K.; Pernemalm, P.-A.; Theander, O. Formation of aromatic compounds
from carbohydrates. VII. Reaction of D-glucose and glycine in slightly acidic,
aqueous solution. Acta Chem. Scand. B 1978, 32, 249–256.
(17) Maga, J. A. Pyridines in foods. J. Agric. Food Chem. 1981, 29, 895–898.
17
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 18 of 34
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
(18) Zamora, R.; Lavado-Tena, C. M.; Hidalgo, F. J. Oligomerization of reactive
carbonyls in the presence of ammonia-producing compounds: A route for the
production of pyridines in foods. Food Chem. 2020, 304, 125284.
(19) Stadler, R. H.; Varga, N.; Milo, C.; Schilter, B.; Vera, F. A.; Welti, D. H.
Alkylpyridiniums. 2. Isolation and quantification in roasted and ground coffees. J.
Agric. Food Chem. 2002, 50, 1200–1206.
(20) Moon, J.-K.; Shibamoto, T. Role of roasting conditions in the profile of volatile
flavor chemicals formed from coffee beans. J. Agric. Food Chem. 2009, 57, 5823–
5831.
(21) Kuo, P. C.; Lay, Y. Y.; Chen, Y. J.; Yang, W. H.; Tzen, J. T. Changes in volatile
compounds upon aging and drying in oolong tea production. J. Sci. Food Agric.
2011, 91, 293–301.
(22) Myers, D. V.; Howell, J. C. Characterization and specifications of caramel colours:
an overview. Food Chem. Toxicol. 1992, 30, 359–363.
(23) Ruiz-Matute, A. I.; Soria, A. C.; Sanz, M. L.; Martínez-Castro, I. Characterization
of traditional Spanish edible plant syrups based on carbohydrate GC-MS analysis.
J. Food Comp. Anal. 2010, 23, 260–263.
(24) Snedecor, G. W.; Cochran, W. G. Statistical Methods, 7th ed.; Iowa State
University Press: Ames, IA, 1980.
(25) Lu, S. Y.; Cui, H. P.; Zhan, H.; Hayat, K.; Jia, C. S. Hussain, S.; Tahir, M. U.;
Zhang, X. M.; Ho, C. T. Timely addition of glutathione for its interaction with
deoxypentosone to inhibit the aqueous Maillard reaction and browning of
glycylglycine-arabinose system. J. Agric. Food Chem. 2019, 67, 6586–6593.
18
ACS Paragon Plus Environment

Page 19 of 34
Journal of Agricultural and Food Chemistry
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
(26) Kanzler, C.; Schestkowa, H.; Haase, P. T.; Kroh, L. W. Formation of reactive
intermediates, color, and antioxidant activity in the Maillard reaction of maltose in
comparison to D-glucose. J. Agric. Food Chem. 2017, 65, 8957–8965.
(27) Adam, W.; Grimison, A. Sigma-polarization in 5-membered heterocyclic ring
systems. Theoret. Chim. Acta 1967, 7, 342–351.
(28) Dong, W.; Hu, R.; Long, Y.; Li, H.; Zhang, Y.; Zhu, K.; Chu, Z. Comparative
evaluation of the volatile profiles and taste properties of roasted coffee beans as
affected by drying method and detected by electronic nose, electronic tongue, and
HS-SPME-GC-MS. Food Chem. 2019, 272, 723–731.
(29) Begum, N.; Raza, A.; Song, H. L.; Zhang, Y.; Zhang, L.; Liu, P.. Effect of thermal
treatment on aroma generation from bovine bone marrow extract during enzymatic
hydrolysis. J. Food Process. Preserv. 2019, 43, e14105.
(30) Pua, A.; Lau, H.; Liu, S. Q.; Tan, L. P.; Goh, R. M. V.; Lassabliere, B.; Leong, K.-
C.; Sun, J.; Cornuz, M.; Yu, B.. Improved detection of key odourants in Arabica
coffee using gas chromatography-olfactometry in combination with low energy
electron ionisation gas chromatography-quadrupole time-of-flight mass
spectrometry. Food Chem. 2020, 302, 125370.
(31) Wang, Y.; Song, H.; Zhang, Y.; Tang, J.; Yu, D. Determination of aroma
compounds in pork broth produced by different processing methods. Flavour Frag.
J. 2016, 31, 319–328.
(32) Kim, H.-J.; Lee, E.-J.; Shin, O.-S.; Ji, W.-D; Choi, M.-R.; Kim, J.-K. Volatile
components in the soy sauce manufactured by bacillus species and fused yeast. J.
Microbiol. Biotechnol. 1996, 6, 194–201.
(33) Chua, L. S. The extent of hydroxymethylfurfural formation in honey by heating
temperature and duration. Lett. Org. Chem. 2018, 15, 233–240.
19
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 20 of 34
418
419
420
421
422
423
424
425
426
427
(34) Braghini, F.; Biluca, F. C.; Gonzaga, L. V.; Kracik, A. S.; Vieira, C. R. W.; Vitali,
L.; Micke, G. A.; Costa, A. C. O.; Fett, R. Impact of short-term thermal treatment
on stingless bee honey (Meliponinae): Quality, phenolic compounds and
antioxidant capacity. J. Food Process. Preserv. 2019, 43, 13954.
(35) Molaveisi, M. ; Beigbabaei, A. ; Akbari, E.; Noghabi, M. S. ; Mohamadi, M.
Kinetics of temperature effect on antioxidant activity, phenolic compounds and
color of Iranian jujube honey. Heliyon 2019, 5, 01129.
(36) Kumar, S.; Johansson, H.; Kanda, T.; Engman, L.; Mueller, T.; Jonsson, M.;
Pedulli, G. F.; Petrucci, S.; Valgimigli, L. Catalytic chain-breaking pyridinol
antioxidants. Org. Lett. 2008, 10, 4895–4898.
428
429
20
ACS Paragon Plus Environment

Page 21 of 34
Journal of Agricultural and Food Chemistry
FIGURE CAPTIONS
Figure 1. Effect of pH on the formation of 6-(hydroxymethyl)pyridin-3-ol (circles) and
on the disappearance of 5-hydroxymethylfurfural (HMF, up pointing triangles). HMF (10
mol) was heated in the presence of ammonium chloride (30 mol) and 30 L 0.3 M
buffer for 22 h at 100 ºC. Three buffers were assayed: sodium citrate buffer (open
symbols), sodium phosphate buffer (half-closed symbols), and sodium borate buffer
(closed symbols). Data are given in nmol of the produced 6-(hydroxymethyl)pyridin-3-ol
per mol of the amount of HMF added to the reaction mixture.
Figure 2. Effect of 5-hydroxymethylfurfural (HMF, ) concentration on the formation
of 6-(hydroxymethyl)pyridin-3-ol (). HMF was heated in the presence of ammonium
chloride (30 mol) and 0.3 M sodium phosphate buffer, pH 8, for 22 h at 100 ºC.
Figure 3. Effect of amine compound concentration on the formation of 6-
(hydroxymethyl)pyridin-3-ol ( and ) and on the disappearance of 5-
hydroxymethylfurfural (HMF,  and ). HMF (10 mol) was heated in the presence of
the amine compound and 0.3 M buffer for 22 h at 100 ºC. Two amine compounds were
assayed: ammonia ( and ) and ammonium chloride ( and ). Data are given in
nmol of the produced 6-(hydroxymethyl)pyridin-3-ol per mol of the amount of HMF
added to the reaction mixture.
Figure 4. Effect of time and temperature on the formation of 6-(hydroxymethyl)pyridin-
3-ol. 5-Hydroxymethylfurfural (HMF, 10 mol) was heated in the presence of ammonium
chloride (30 mol) and 0.3 M sodium phosphate buffer, pH 8. The assayed temperatures
were: 100 (), 120 (), 140 (), and 160 ( ) ºC. Data are given in nmol of the
21
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 22 of 34
produced 6-(hydroxymethyl)pyridin-3-ol per mol of the amount of HMF added to the
reaction mixture.
Figure 5. Arrhenius plot obtained for the formation of 6-(hydroxymethyl)pyridin-3-ol
() and the disappearance of 5-hydroxymethylfurfural (HMF, ). HMF (10 mol) was
heated in the presence of ammonium chloride (30 mol) and 0.3 M sodium phosphate
buffer, pH 8, at the temperatures indicated in Figures 4 and 6.
Figure 6. Effect of time and temperature on the disappearance of 5-
hydroxymethylfurfural (HMF). HMF (10 mol) was heated in the presence of ammonium
chloride (30 mol) and 0.3 M sodium phosphate buffer, pH 8. The assayed temperatures
were: 100 (), 120 (), 140 (), and 160 ( ) ºC.
Figure 7. Formation of pyridin-3-ols [6-(hydroxymethyl)pyridin-3-ol (stripped bars),
pyridin-3-ol (open bars), and 2-methylpyridin-3-ol (crosshatched bars)] from 2-oxofurans
[5-hydroxymethylfurfural (HMF), furfural, and 2-acetylfuran, respectively]. The 2-
oxofuran (10 mol) was heated in the presence of ammonium chloride (30 mol) and 0.3
M sodium phosphate buffer, pH 8, at 100 ºC for the indicated times. Data are given in
nmol of the produced pyridin-3-ol per mol of the amount of 2-oxofuran added to the
reaction mixture.
Figure 8. Proposed pathway for the formation of pyridine-3-ols from the corresponding
2-oxofurans. For furfural and pyridin-3-ol, R1 = H and R2 = H. For 2-acetylfuran and 2-
methylpyridin-3-ol, R1 = H and R2 = CH3. For 5-hydroxymethylfurfural (HMF) and 6-
(hydroxymethyl)pyridin-3-ol, R1 = CH2OH and R2 = H.
22
ACS Paragon Plus Environment

Page 23 of 34
Journal of Agricultural and Food Chemistry
Figure 9. Effect of heating time on the formation of 6-(hydroxymethyl)pyridin-3-ol ()
and 5-hydroxymethylfurfural (HMF, ) in honey heated at, A, 60, and B, 100 ºC. Honey
(1 g) was heated in closed test tubes for the indicated times and temperatures.
Figure 10. Effect of heating time on the formation of 6-(hydroxymethyl)pyridin-3-ol (),
5-hydroxymethylfurfural (HMF, ), pyridin-3-ol (), and 2-methylpyridin-3-ol () in
sugarcane honey heated at, A, 60, and B, 100 ºC. Sugarcane honey (1 g) was heated in
closed test tubes for the indicated times and temperatures.
23
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 24 of 34
75
50
25
12
9
6
3
3
4
5
6
7
8
9 10
pH
Figure 1
24
ACS Paragon Plus Environment

Page 25 of 34
Journal of Agricultural and Food Chemistry
120
80
40
0
60
40
20
0
0
5
10
15
20
HMF (mol)
Figure 2
25
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 26 of 34
100
75
50
25
0
12
9
6
3
0
0
10
20
30
40
100
Amine compound (mol)
Figure 3
26
ACS Paragon Plus Environment

Page 27 of 34
Journal of Agricultural and Food Chemistry
15
12
9
6
3
0
0
50
100
150
200
Time (min)
Figure 4
27
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 28 of 34
1
0
-1
-2
-3
-4
0.00234
0.00252
1/T (K-1)
0.00270
Figure 5
28
ACS Paragon Plus Environment

Page 29 of 34
Journal of Agricultural and Food Chemistry
100
75
50
25
0
15
30
45
60
Time (min)
Figure 6
29
ACS Paragon Plus Environment