Effects of Temperature on the Composition and Xanthine Oxidase Inhibitory Activities of Caffeic Acid Roasting Products

Article abstract of DOI:10.1021/acs.jafc.9b03633

The high-temperature treatment of caffeic acid by a model reaction for the processing of foods by roasting enhanced its xanthine oxidase (XO) inhibitory activity. The thermal reaction products included various oligomeric compounds, whose structures were determined as being produced via the intermediate 4-vinylcatechol. Measurements of their XO inhibitory activities were also carried out. Among the identified oligomers, the coupling products of caffeic acid and vinylcatechol, which were mainly produced at 140-170 °C, presented stronger XO inhibitory activities than the other types of oligomers produced. Further reacted compounds, which were mainly formed at 200 °C by the addition or elimination of catechol unit in the oligomers, displayed weaker activities. These results indicated that thermal enhancement of the XO inhibitory activity of caffeic acid can be explained by the differences in the XO inhibitory activities of the various constituents of the thermal reaction products. Caffeic acid and its derivatives are polyphenols found widely distributed in foods. Moreover, XO inhibition is closely related to the prevention of the life-style-related disease gout. The results suggest that a simple roasting process (170 °C) can lend useful human-health-related functionalities to caffeic acid containing foods such as coffee.

Full text of DOI:10.1021/acs.jafc.9b03633

Subscriber access provided by BUFFALO STATE
Food and Beverage Chemistry/Biochemistry
Temperature Effects on the Composition and Xanthine
Oxidase Inhibitory Activity of Caffeic Acid Roasting Products
Toshiya Masuda, Yuya Fukuyama, Sayaka Doi, Akiko Masuda, Sinichiro Kurosawa, and Shigeyoshi Fujii
J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.9b03633 • Publication Date (Web): 23 Jul 2019
Downloaded from pubs.acs.org on July 23, 2019
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 45
Journal of Agricultural and Food Chemistry
1
2
3
4
5
6
7
8
9
Temperature Effects on the Composition and Xanthine Oxidase Inhibitory
Activity of Caffeic Acid Roasting Products
Toshiya Masuda,†,* Yuya Fukuyama,† Sayaka Doi,† Akiko Masuda,§
Shinichiro Kurosawa,‡ and Shigeyoshi Fujii‡
† Graduate School of Human Life Science, Osaka City University, Sumiyoshi, Osaka
558-8585, Japan.
§ Faculty of Human Life Science, Shikoku University, Tokushima 771-1192, Japan.
‡Technical Research and Development Institute, Ajinomoto AGF, Inc., Sanken Bldg., 3-
25-1, Hyakunin-cho Shinjuku-ku, Tokyo 169-0073 Japan
10
11
12
13
Running Head: Temperature effects on the roasting products of caffeic acid
14
15
*Corresponding Author
16
17
(Phone & FAX: +81-6-6605-2813; E-mail: masuda_t@life.osaka-cu.ac.jp)
1
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 2 of 45
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
ABSTRACT
The high-temperature treatment of caffeic acid by a model reaction for the roasting
processing of foods enhanced its xanthine oxidase (XO) inhibitory activity. The thermal
reaction products included various oligomeric compounds, whose structures were
determined as produced via the intermediate 4-vinylcatechol. And their XO inhibitory
activity measurement were also carried out. Among the identified oligomers, the
coupling products of caffeic acid and vinylcatechol, which were mainly produced at
140–170 ℃, presented stronger XO inhibitory activity than the other types of produced
oligomers. Further reacted compounds, which were mainly formed at 200 ℃ by the
addition or elimination of a catechol unit in the oligomers, displayed weaker activity.
These results indicated that thermal enhancement of the XO inhibitory activity of
caffeic acid can be explained by the difference in the XO inhibitory activity of the
various constituents of the thermal reaction products. Caffeic acid and its derivatives are
polyphenols found widely distributed in foods. Moreover, XO inhibition is closely
related to the prevention of the life-style related disease, gout. The results suggest that a
simple roasting process (170 ℃) can lend useful human health-related functionalities to
caffeic acid-containing foods such as coffee.
2
ACS Paragon Plus Environment

Page 3 of 45
Journal of Agricultural and Food Chemistry
35
Keywords: Thermal reaction, Caffeic acid, Xanthine oxidase inhibition, Roasting
36
37
temperature, Structure identification
3
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 4 of 45
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
INTRODUCTION
Roasting is a high-temperature cooking method (130–230 ℃),1 that is applied to grains,
vegetables, meats, and fish in both industrial food processing and home cooking.2
Roasting can cause various thermal reactions of the food chemical constituents, thereby
developing the flavors and color.1,3 Caffeic acid is a polyphenol that is widely
distributed in plant foods in its free or bound forms. Coffee beans contain free caffeic
acid as well as a large amount of chlorogenic acid, a major caffeic acid derivative.4,5
The roasting of coffee beans at approximately 200 ℃ is indispensable for making coffee
beverages. During the roasting process, caffeic acid derivatives are degraded into
various compounds including disassembled caffeic acid.6,7 The degraded products react
with each other to afford more complex substances,8 which affect the flavor and taste as
well as human health functions of roasted coffee beverages.9 To date, many thermal
reaction products have been identified in roasted coffee beans, including volatile
flavoring compounds,10 Maillard reaction products,11 the bitter-tasting chlorogenic
lactones,12 and phenylindane derivatives.13 Recently, several epidemiological studies by
Choi et al. 14 and other groups,15 have reported that coffee drinking reduces the risk of
gout. Additionally, we recently discovered that roasted coffee bean extract exhibits
4
ACS Paragon Plus Environment

Page 5 of 45
Journal of Agricultural and Food Chemistry
56
xanthine oxidase (XO) inhibitory activity.16 We also reported the presence of several
inhibitors in roasted coffee beans.17 XO is a key enzyme in the catabolism of purine to
uric acid, the substance that causes gout when it over-accumulates in plasma.18 Stadler
et al.19 were first to report that phenylindanes were produced from the thermal treatment
of caffeic acid via a vinylcatechol intermediate. These compounds were observed in
roasted coffee beans and play a role in the bitter taste of coffee.13 We also reported
moderate XO inhibitory activity for some phenylindanes.20
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
In this investigation, we found maximum XO inhibitory activity for the caffeic
acid products of thermal treatment at around 170 ℃. These conditions were used as a
model reaction of the roasting process and, notably, differences in the thermal reaction
products were found between 110 and 290 ℃. Herein, we report the structures of the
constituent compounds of the thermal reaction products (140–200 ℃) and their XO
inhibitory activities.
MATERIALS AND METHODS
Chemicals and instruments. Caffeic acid was obtained from Kanto Chemicals (Tokyo,
Japan). Xanthine oxidase (XO) was purchased from Wako Pure Chemicals (Osaka,
Japan). Allopurinol was obtained from Sigma-Aldrich (St. Louis, MO). All other
5
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 6 of 45
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
reagents and solvents were purchased from Nacalai Tesque (Kyoto, Japan). NMR
spectra were obtained from a JNM-ECZ400S (400 MHz, JEOL, Tokyo, Japan) or
ARX600 (600 MHz, Bruker Japan, Yokohama, Japan). MS spectra were measured
using a JMS-T100 spectrometer (JEOL) with direct analysis at real time (DART) or
electro spray ionization (ESI), and time of flight (TOF) measuring mode under the
following conditions: DART (positive mode), orifice 1, 10 V; orifice 2, 5 V; ring lens,
10 V; helium gas pressure, 0.5 MPa; flow rate of 10 L/min; ion resource temperature,
350 ºC; grid voltage, 400 V and ESI (positive mode), orifice 1, 85 V and 80 ℃; orifice
2, 5 V; ring lens, 15 V; helium gas pressure, 0.7 MPa; needle, 2000 V; desolvating
chamber temperature, 250 ℃. Elemental compositions of compounds were calculated
using MassLynx software, version 4.1 (Waters Japan, Tokyo, Japan) from high
resolution MS (HR-MS) data. Analytical HPLC was performed using a PU-4180
RHPLC quaternary gradient pump (JASCO, Tokyo, Japan) equipped with an MD-4015
photodiode array detector (JASCO) and a CTO-10AS column oven (Shimadzu, Kyoto,
Japan). Data were analyzed using the ChromNAV software (v. 2, JASCO). Preparative
HPLC was performed using an LC-6AD pump (Shimadzu) equipped with an SPD-6A
UV detector (Shimadzu).
6
ACS Paragon Plus Environment

Page 7 of 45
Journal of Agricultural and Food Chemistry
92
Thermal reaction of caffeic acid and HPLC analysis of the reaction products
In a screw-capped test tube (i.d. 8 mm and length, 100 mm) were placed 10 mg
caffeic acid in methanol (200 µL) and 400 µL phosphate buffer (500 mmol/L, pH 6.0,
Na2HPO4 and KH2PO4). After removing the solvent in vacuo, the tube was heated in a
metal block bath (MB‒1H‒UII, KPI, Itami, Japan) at the appropriate temperatures
(110–290 ℃) for 0.5 h. After cooling the tube, methanol (1 mL) was added and the
mixture was centrifuged at 2000 rpm for 5 min at 25 ℃. Subsequently, the supernatant
was collected and the solvent was removed by evaporation to give the thermal reaction
products. A portion of the reaction products was used for the XO inhibitory assay. The
constituents in the products were analyzed by HPLC [Cosmosil 5C18-AR-II column
(Nacalai, Kyoto, Japan), 250 mm × 4.6 mm i.d., 5 μm; flow rate, 1.0 mL/min; solvent
A, 1% acetic acid in H2O; solvent B, CH3CN; linear gradient from 5% solvent B to 45%
solvent B for 40 min, then from 45% solvent B to 100% solvent B for 10 min, and
100% solvent B for 5 min; detection at 280 and 320 nm].
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
Gel Permeation Chromatography (GPC) analysis of the thermal reaction products
The thermal reaction products (weight corresponding to 6 mg of initial caffeic
acid) were treated with acetic anhydride (0.5 mL) and pyridine (0.5 mL) at 25 ℃ for 1
7
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 8 of 45
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
h. After removal of acetic acid and pyridine in vacuo, the residue was dissolved in THF
to produce an analytical solution (0.4 mg/mL, corresponding to the initial caffeic acid
concentration). The solution (15 μL) was injected into a GPC column (GPC KF-803,
300 mm × 8 mm i.d., Shodex, Tokyo, Japan) for analysis (solvent, HPLC grade THF;
flow rate, 1.0 mL/min; temperature, 30 ℃; detection at 264 nm). Molecular size
calibration was performed under the same conditions using polystyrene standards [266,
2280, 6540, and 25500 Da polystyrene; Ready Cal-Kit Poly(styrene) Low, Sigma-
Aldrich, St. Louis, MO].
Isolation of the thermal reaction product constituents from the large-scale thermal
reactions of caffeic acid at 140–200 ℃
For thermal treatment at 140 ℃, caffeic acid (10 g) was dissolved in 100 mL
methanol and 400 mL of 500 mmol/L Na2HPO4–KH2PO4 (pH 6.0) and subsequently
evaporated to dryness. The solid residue was heated at 140 ℃ for 90 min. After cooling,
the reaction mixture was extracted twice with 1 L methanol and this procedure was
repeated ten times. The methanol extracts were combined and the solvent was removed.
The resultant residue was further purified to isolate the constituent compounds. A
portion (84 g) of the residue was subjected to Amberlite XAD-7 (Organo, Tokyo,
8
ACS Paragon Plus Environment

Page 9 of 45
Journal of Agricultural and Food Chemistry
128
Japan) column chromatography and eluted with increasing concentrations of methanol
(50 to 100%) in water to produce eight separate fractions. Fraction 3 (208 mg of 6.5 g),
which was eluted with 60% methanol in water, was purified by preparative HPLC
(Cosmosil 5C18-AR-II column, 250 × 20 mm i.d.; solvent, 1% acetic acid in 85:15
H2O–CH3CN; flow rate, 9.6 mL/min; detection at 280 nm) to yield compounds 1–4 in 2,
6, 3, and 3 mg yields, respectively. Fraction 6 (14.3 g), which was eluted with 75%
methanol in water from the XAD-7 column, was subjected to Sephadex LH-20 (GE
Healthcare, Hino, Japan) column chromatography (1.5 L column) and eluted with
methanol to produce 5 g residue, which was collected from 1.5 to 1.65 L of the total
elution volume. A part of the residue (656 mg) was purified by preparative HPLC
(Cosmosil 5C18-AR-II column, 250 mm × 20 mm i.d.; solvent, 1% acetic acid in 75:25
H2O–CH3CN; flow rate, 9.6 mL/min; detection at 280 nm) to produce compounds 5–8
in 5, 4, 87, and 134 mg yields, respectively, along with 52 mg of a mixture of
compounds 9 and 10. This mixture was separated by HPLC in recycle mode (eleven
times) under the following conditions: Cosmosil 5C18-AR-II column, 250 mm × 20 mm
i.d.; solvent, 1% acetic acid in 45:55 H2O–CH3OH; flow rate, 9.6 mL/min; detection at
280 nm, to produce pure 9 and 10 in 5 mg yields.
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
For thermal treatment at 200 ℃, 40 mL phosphate buffer (pH 6.0; 0.5 mol/L)
9
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 10 of 45
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
was added to caffeic acid (1 g) in methanol (20 mL) and mixed well. After removal of
the solvent from the mixture in vacuo, the residual solid was heated at 200 ℃ for 30min.
The resulting dark-colored residue was extracted twice with methanol (50 mL) to
produce 0.76 g extract after evaporation. The extract was fractionated by XAD-7
column chromatography eluting with increasing concentrations of methanol (10 to
100%) in water. Two fractions, eluted with 50 and 75% methanol, were evaporated to
produce fractions I and II in 74 and 86 mg yields, respectively. A portion (32 mg) of the
residue of fraction I was purified by HPLC (Cosmosil 5C18-AR-II column, 250 mm ×
20 mm i.d.; solvent, 1% acetic acid in 90:10 H2O–CH3CN; flow rate, 9.6 mL/min;
detection at 280 nm) to give compound 11 in 1 mg yield. Likewise, the residue of
fraction II (37 mg) was separated by HPLC (Cosmosil 5C18-AR-II column, 250 mm ×
20 mm i.d.; solvent, 1% acetic acid in 70:30 H2O–CH3CN; flow rate, 9.6 mL/min;
detection at 280 nm) to produce compounds 5–8 and 12 in 1, 5 , 1 , 2 , and 2 mg yields,
respectively. Another reaction product (570 mg), which was obtained from the thermal
treatment of caffeic acid (1 g) under the same conditions, was fractionated by XAD-7
column chromatography using the same protocol as described above. The fraction
eluted with 90% methanol (195 mg) was separate d by HPLC (Cosmosil 5C18-AR-II
column, 250 mm × 20 mm i.d.; solvent, 1% acetic acid in 45:55 H2O–CH3CN; flow
10
ACS Paragon Plus Environment

Page 11 of 45
Journal of Agricultural and Food Chemistry
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
rate, 9.6 mL/min; detection at 280 nm) to produce compounds 13, 14, and 16–18 in 5, 4,
3, 2, and 2 mg yields, respectively. The product (17. 1 g) of a third thermal reaction of
caffeic acid (20 g) was also fractionated into five fractions by XAD-7 column
chromatography under the same conditions. The fraction eluted with 70% methanol (2
g) produced the subsequent four fractions 4-1 to 4-4 in 523, 61, 80, and 47 mg yields,
respectively, (retention times: fraction 4-1, 9–24 min, fraction 4-2, 27 min, fraction 4-3,
33 min, and fraction 4-4, 40 min) under the following conditions: Cosmosil 5C18-AR-II
column, 250 mm × 20 mm i.d.; solvent, 1% acetic acid in 70:30 H2O–CH3CN; flow
rate, 9.6 mL/min; detection at 280 nm. Compound 13 (15 mg) was isolated from
fraction 4-1 (487 mg) by HPLC with a retention time of 41 min (Cosmosil 5C18-AR-II
column, 250 mm × 20 mm i.d.; solvent, 1% acetic acid in 60:40 H2O–CH3OH; flow
rate, 6.7 mL/min; detection at 280 nm). Compound 15 (7 mg) was isolated from fraction
4-2 (20 mg) by HPLC with a retention time of 37 min (Cosmosil 5C18-AR-II column,
250 mm × 20 mm i.d.; solvent, 1% acetic acid in 45:55 H2O–CH3OH; flow rate, 4.8
mL/min; detection at 280 nm). Compounds 14 (10 mg) and 17 (14 mg) were obtained
from fraction 4-3 (70 mg) by HPLC with retention times of 43 and 45 min, respectively
(Cosmosil 5C18-AR-II column, 250 mm × 20 mm i.d.; solvent, 1% acetic acid in 45:55
H2O–CH3OH; flow rate, 4.8 mL/min; detection at 280 nm). Finally, compound 18 (6
11
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 12 of 45
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
mg) was isolated from fraction 4-4 (33 mg) by HPLC with a retention time 50 min
(Cosmosil 5C18-AR-II column, 250 mm × 20 mm i.d.; solven, 1% acetic acid in 45:55
H2O–CH3OH; flow rate, 4.8 mL/min; detection at 280 nm).
Analytical data for the structures of compounds 5, 6, and 11–18.
Chemical structures of the isolated products, except previously identified
phenylindanes (compounds 1–4, 7, and 8), were elucidated by NMR and MS data of
their intact or acetylated forms, which were derived by acetylation with acetic anhydride
in pyridine. The data were listed as follows and position numbering of each compound
is tentative as shown in the structures in Fig. 3.
5. HR-ESIMS (m/z) [M+Na]+ calcd for C16H16O4Na 295.0946 , found
298.0969, and [2M+Na]+ calcd for C32H32O8Na 567.1995, found 567.2003; 1H NMR
(CD3OD)  6.69 (4H, s, H3, H6, H3', and H6'), 3.75 (2H, q, J=7.4 Hz, H7 and H7'), and
1.43 (6H, d, J=7.4 Hz, H8 and H8').
6. HR-ESIMS (m/z) [M+Na]+ calcd for C14H14O4Na 269.0790, found 269.0817
and [2M+Na]+ calcd for C28H28O8Na 515.1682, found 515.1690; 1H NMR (CD3OD) 
6.65 (2H, d, J=1.6 Hz, H3 and H2'), 6.70 (2H, d, J=8.2 Hz, H6 and H5'), 6.57 (2H, dd,
J=8.2 and1.6 Hz, H5 and H6'), 3.87 (1H, q, J=7.2 Hz, H7), and 1.52 (3H, d, J=7.2 Hz,
12
ACS Paragon Plus Environment

Page 13 of 45
Journal of Agricultural and Food Chemistry
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
H8).
9. HR-DARTMS (m/z) [M+H]+ calcd for C24H25O6 409.1651, found 409.1656;
NMR, See Table S1 in supporting information.
10. HR-DARTMS (m/z) [M+H]+ calcd for C24H25O6 409.1651, found
409.1652; NMR, See Table S1 in supporting information.
11. DARTMS (m/z) [M+H]+ 111; 1H NMR (acetone-d6)  6.65 (2H, AA'BB'
coupling, H3 and H6) and 6.79 (2H, AA'BB' coupling, H4 and H5).
12. DARTMS (m/z) [M+H]+ 139; 1H NMR (CD3OD)  6.65 (1H, d, J=1.9 Hz,
H3), 6.69 (1H, d, J=8.0 Hz, H5), 6.53 (1H, dd, J=8.0 and1.9 Hz, H6), 2.52 (2H, q, J=7.4
Hz, 4-CH2-CH3), 1.20 (3H, t, J=7.4 Hz, 4-CH2-CH3).
13. HR-ESIMS [M+Na]+ calcd for C14H14O4Na 269.0790, found 269.0816, and
[2M+Na]+ calcd for C28H28O8Na 515.1682, found 515.1667; 1H-NMR (CDCl3) of
tetraacetate (13a)  6.99–7.03 (2H, complex, H3 and H5), 7.06–7.12 (3H, complex, H6,
H4′, and H6′), 4.25 (1H, q, J=7.0 Hz, H7), 1.57 (3H, d, J=7.0 Hz, H8), 7.23 (1H, dd,
J=8.6 and 8.2 Hz, H5'), 2.18 (3H, s, 1, 2, 2' or 3'-CH3CO), 2.25 (3H, s, 1, 2, 2' or 3'-
CH3CO), 2.26 (6H, s, 1, 2, 2' or 3'-CH3CO).
14. HR-DARTMS [M+H]+ calcd for C10H13O2 165.0916, found 165.0914; 1H-
NMR (CDCl3) of diacetate of 14,  3.17 (1H, sext, J=7.6 Hz, H1), 1.64 (1H, dq, J=12.4
13
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 14 of 45
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
and 7.6 Hz, H2a), 2.29 (1H, ddt, J=12.4, 7.6, and 4.0 Hz, H2b), 2.77–2.92 (2H, m, H3),
6.99 (1H, s, H4 or H7), 6.95 (1H, s, H4 or H7), 1.26 (3H, d, J=7.6 Hz, 1-CH3), 2.28
(6H, s, 5- and 6-CH3CO ).
15. HR-ESIMS [M+Na]+ calcd for C16H16O4Na 295.0946, found 295.0986, and
[2M+Na]+ calcd for C32H32O8Na 567.1995, found 567.2003; 1H NMR (CD3OD)  3.21
(1H, dquint, J=8.8 and 7.0 Hz, H1), 1.99 – 2.08 (1H, m, H2a), 2.20 (1H, ddd, J=12.0,
7.0 and 1.2 Hz, H2b), 4.38 (1H, brd, J=8.0 Hz, H3), 6.60 (1H, d, J=8.0 Hz, H6), 6.73
(1H, d, J=8.0 Hz, H7), 1.25 (1H, d, J=7.0 Hz, 1-CH3), 6.49 (1H, d, J=1.7 Hz, H2'), 6.66
(1H, d, J=8.0 Hz, H5'), 6.44 (1H, dd, J=8.0 and 1.7 Hz, H6').
16. HR-ESIMS [M+Na]+ calcd for C16H16O4Na 295.0946, found 295.0964; 1H
NMR (CD3OD)  3.08 (1H, sext, J=7.5 Hz, H1), 1.55 (1H, dt, J=12.4 and 7.5 Hz, H2a),
2.75 (1H, dt, J=12.4 and 7.5 Hz, H2b), 4.24 (1H, t, J=7.5 Hz, H3), 6.61 (1H, d, J=8.0
Hz, H6), 6.73 (1H, d, J=8.0 Hz, H7), 1.26 (3H, d, J=7.5 Hz, 1-CH3), 6.60 (1H, d, J=2.0
Hz, H2'), 6.70 (1H, d, J=8.0 Hz, H5'), 6.55 (1H, dd, J=8.0 and 2.0 Hz, H6').
17. HR-ESIMS [M+Na]+ calcd for C18H18O4Na 321.1103, found 321.1137, and
[2M+Na]+ calcd for C36H36O8Na 619.2308, found 619.2301; 1H-NMR (CDCl3) of
tetraacetate of 17  2.81 (1H, dq, J=7.8 and 6.8 Hz, H1), 2.55 (1H, dt, J=1.6 and 7.8 Hz,
H2), 6.92 (1H, s, H4), 6.90 (1H, s, H7), 1.35 (3H, d, J=6.8 Hz, 1-CH3), 1.54 (3H, s, 3-
14
ACS Paragon Plus Environment

Page 15 of 45
Journal of Agricultural and Food Chemistry
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
CH3), 7.15 (1H, s, H2'), 7.00 (1H, s, H5'), 3.23 (1H, dd, J=16.2 and 7.8 Hz, 7′-CH2),
2.82 (1H, brd, J=16.2 Hz, 7′-CH2), 2.25 (6H, s, 5-, 6-, 3'- or 4'-CH3CO), 2.27 (3H, s,
5-, 6-, 3'- or 4'-CH3CO), 2.30 (3H, s, 5-, 6-, 3'- or 4'-CH3CO). Observed NOESY
correlations from H2 to 1-CH3, 3-CH3 7′α-CH2α, and 7′β-CH2, from H7 to 1-CH3, from
1-CH3 to H2, and H7, from 7′-CH2 to 3-CH3, from 3-CH3 to H2, 7′-CH2, H4, and
H2', and from H2′ to 3-CH3.
18. HR-ESIMS [M+Na]+ calcd for C18H18O4Na 321.1103, found 321.1133, and
[2M+Na]+ calcd for C36H36O8Na 619.2308, found 619.2295; 1H-NMR (CDCl3) of
tetraacetate of 18, 3.04 (2H, dq, J=4.8 and 7.0 Hz, H1 and H7′), 2.61 (1H, dt, J=8.6 and
4.8 Hz, H2), 4.64 (1H, d, J=8.6 Hz, H3), 6.95 (2H, s, H4 and H2'), 7.11 (2H, s, H7 and
H5'), 1.32 (6H, d, J=7.0 Hz, 1-CH3 and 7′-CH3), 2.26 (6H, s, 5-, 6-, 3'-, or 4'-CH3CO),
2.28 (6H, s, 5-, 6-, 3'-, or 4'-CH3CO). Observed NOE from 1- and 7′-CH3 to H2 and H3
in NOE difference spectra.
Measurement of the xanthine oxidase inhibitory activity of the isolated compounds
The xanthine oxidase inhibition assay was performed according to a
previously reported protocol.17 Briefly, a reaction medium consisting of 10 µL of 1
mmol/L xanthine in DMSO, 10 µL of the test sample (6 mg/mL for thermal products
15
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 16 of 45
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
and 4 mmol/L for isolated compounds) in DMSO, and 160 µL of 12.5 mmol/L
phosphate buffer (pH 7.4) was pre-incubated at 37 °C for 5 min. To this solution, 20 µL
of 0.027 unit/mL XO buffer solution was added. After incubation at 37 °C for 10 min,
3% HClO4 aq (25 µL) was added to quench the reaction. An aliquot of the solution (20
µL) was injected into the HPLC column to quantify the uric acid production [Mightysil
RP-18 GP Aqua column (Kanto Chemical, Tokyo Japan), 250 mm × 4.6 mm i.d., 5 μm;
flow rate, 1.0 mL/min; solvent, 2.5:97.5 methanol–0.1% phosphoric acid in water;
detection at 290 nm; temperature, 35 °C]. The percent inhibition was calculated
according to the following equation: inhibition (%) = [(peak area of uric acid in control
experiment) – (peak area of uric acid in sample experiment)] × 100 / (peak area of uric
acid in control experiment).
Statistical analysis
Activity data obtained from the experiments in triplicate are presented as mean ±
standard error (SE). The differences of the data between sample and control
experiments were evaluated by the Dunnett's test. A p-value < 0.05 was considered
statistically significant.
16
ACS Paragon Plus Environment

Page 17 of 45
Journal of Agricultural and Food Chemistry
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
RESULTS AND DISCUSSION
HPLC analysis of the thermal reaction products of caffeic acid at various
temperatures
In our previous work, we observed the XO inhibitory activity of a model
roasting reaction product of chlorogenic acid while intact chlorogenic acid did not
inhibit XO activity. We also examined the XO inhibitory activity of the thermally
treated caffeic and quinic acids, which are both components of chlorogenic acid.20 The
results revealed that the XO inhibitory activity of the thermal reaction product of caffeic
acid was greater than that of chlorogenic acid, while no XO inhibitory activity was
observed for the product of quinic acid. Therefore, in this investigation the thermal
reaction of caffeic acid was further investigated for the production of XO inhibitors and
enhanced XO inhibitory activity. The high-temperature thermal treatment of caffeic acid
was carried out through a model roasting reaction according to a previously reported
method [caffeic acid : solid support (a phosphoric salt)] = 10:28.20 The XO inhibitory
activities of the thermal treatment products at different temperatures are illustrated in
Fig. 1. Intact caffeic acid has weak XO inhibitory activity (22% inhibition) at the
examined concentration (0.3 mg/mL), while the XO inhibitory activities of the thermal
17
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 18 of 45
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
reaction products were greater than that of caffeic acid. The activity increased with
increasing reaction temperature up to 170 ℃, after which it decreased gradually up to
290 ℃. Figure 2A (Figure S1A in supporting information) displays the HPLC data
representing the constituents of the thermal reaction products obtained by gradient
elution on an octadecylsilyl silica gel (ODS) analytical column with 5 to 100%
acetonitrile. At the reaction temperature 110 ℃, the only noticeable peak was that of the
starting material at the retention time 15.5 min. When the temperature was increased to
140 and 170 ℃, numerous peaks were observed in the retention time range of 19 to 42
min, along with the peak corresponding to caffeic acid (retention time 15.5 min), with
very complex peaks observed at 28–42 min. With a further increase in temperature to
200 ℃, the caffeic acid peak and peaks at 19–22 min disappeared. Moreover, the
intensity of the complex peaks at 30–42 min were lower than those observed at 170 ℃.
On the other hand, a new peak appeared at the retention time 11.6 min for the reactions
at temperatures of 200 ℃ or greater. At temperatures over 230 ℃, almost all peaks
gradually disappeared, except for that of the major peak at 11.6 min and a minor peak at
27.0 min. Figure 2B (Figure S1B in supporting information) presents GPC data of the
acetylated products from the thermal reactions of caffeic acid. Under the employed GPC
conditions, acetylated caffeic acid (0.26 kDa), which was obtained from caffeic acid by
18
ACS Paragon Plus Environment

Page 19 of 45
Journal of Agricultural and Food Chemistry
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
treatment with acetic anhydride and pyridine, was detected at the retention time 10.8
min. The area of the earlier eluting peak increased with increasing reaction temperature
up to 200 ℃. This peak was clearly observed at 10.1 min (corresponding to 0.7 kDa:
calculated from the top of the peak) for the reaction at 170 ℃ and broadened to the
retention time 9 min at 200 ℃, which indicates that higher weight molecules (up to ca.
0.3 kDa) were produced by the elevation of the reaction temperature. In the GPC data of
the reactions at 230 ℃ or above, the peak was further broadened between the retention
times 8.5 and 12 min, while two new peaks at retention times 11.1 and 11.5 min were
also observed, which indicates that low-molecular-weight degradation products (below
0.3 kDa) were produced along with higher weight compounds (up to ca. 6 kDa).
Considering the data of constituent analysis (Fig. 2A and S1A), molecular size analysis
(Fig. 2 B and S1B), and XO inhibitory activities of the products from the reactions at
various temperatures (Fig. 1), it was determined that the XO inhibitory activity of
moderate-molecular-weight compounds (3–0.3 kDa by GPC analysis), which may be
coupling products of caffeic acid, should exhibit stronger than that of caffeic acid.
Additionally, these compounds might contribute towards the enhanced XO inhibitory
activity of the products obtained at 170 ℃. On the other hand, higher molecular weight
products, produced in the reactions at above 200 ℃ and eluted earlier from the GPC
19
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 20 of 45
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
column, might exhibit weak XO inhibition. Finally, the low-molecular-weight products,
observed as peaks eluted later from the GPC column of the reactions at high
temperatures, may also present weak XO inhibitory activity. Thus, the accumulation of
both the higher and lower molecular weight products should reduce the XO inhibitory
activity of the reaction products produced at above 200 ℃. From the constituent
analysis data of the products obtained at various temperatures (Fig. 2A and S1A),
moderate-molecular-weight products, which are predicted to be strong XO inhibitors,
should correspond to the peaks in the retention time range of 19 to 42 min in the HPLC
data of the reactions at 140–200 ℃.
Isolation and structure identification of the constituent compounds obtained from
thermal reactions at 140–200 ℃
To identify the compounds that correspond to the peaks observed in the HPLC
constituent analysis of the products obtained at 140–200 ℃, isolation and structure
identification of the constituents of the thermal reaction products were carried out.
Multiple large scale thermal reactions of caffeic acid were performed at 140 and 200 ℃;
in total, 18 compounds were purified from the reaction products as described in the
Materials and Methods section. The isolated compounds 1, 2, 3, 4, 7, and 8 were
20
ACS Paragon Plus Environment

Page 21 of 45
Journal of Agricultural and Food Chemistry
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
identical to the previously identified phenylindanes according to their NMR and MS
data (Fig. 3.)20 Compounds 15 and 16 were determined to be other phenylindane
compounds that had been previously identified by Frank et al.13 Compounds 11 and 12
were determined to be catechol and 4-ethylcatechol, which are reportedly found in
roasted coffee.21 The structures of compounds 5, 6, and 14 were also identified as shown
in Fig. 3 by comparing the NMR and MS data to previously reported data.22–24 Notably,
compounds 9, 10, 13, 17, and 18 were compounds newly identified in this work and
their structures were elucidated by NMR and MS analyses.
High resolution mass spectrometry (HR-MS) data of compounds 9 and 10
indicated that they were constitutional isomers with the molecular formula C24H24O6 (9:
m/z 409.1656 [M+H]+ and 10: m/z 409.1652 [M+H]+). The 1H-NMR spectra of 9 and
10 were also very similar, indicating that they were stereoisomers as summarized in
Table S1 (supporting information). The 1H-NMR spectrum of 9 revealed seven aromatic
proton signals at 6.60, 6.67, and 6.72 ppm in the same coupling system and at 6.19,
6.20, 6.66, and 6.77 ppm as singlet shape. The former peaks indicated the existence of a
tri-substituted benzene ring with protons arranged in a 1,2,4 configuration while the
latter peaks revealed the presence of two tetra-substituted benzene rings with protons
oriented para to one other; these systems were derived from the aromatic moiety of
21
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 22 of 45
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
caffeic acid. The HH-COSY data of 9 clarified two coupling systems, consisting of CH3
(δ 1.21, 1-CH3)–CH (δ 3.19, H1)–CH2 (δ 2.05, H2)–CH (δ 4.49, H3), and CH3 (δ 1.56,
1″-CH3)–CH (δ 4.29, H1″). The HMBC spectrum of 9 showed correlations from C1a to
H1, from C3a to H3, from C1 to H7, from C3 to H4 and H2′, from C4 to H3, from C7 to
H1, and from C1′ to H3 as shown in Fig. 4, indicating the existence of a phenylindane
structure consisting of two tetra-substituted benzene rings and the CH3–CH–CH2–CH
system. Other C–H long-range correlations were observed from C1″ to H5′, H3″, H7″,
and 1″-CH3; from C2″ to H1″ and 1″-CH3; from C7″ to H1″; and from 1″-CH3 to H1″
between the CH3-CH system and a tri-substituted benzene. Correlations were also
observed from C1′ to H1″; and C6′ to H1″ and 1″-CH3 between the CH3–CH system and
one of the tetra-substituted benzenes as shown in Fig. 4. These results indicated that an
ethylcatechol structure was attached to the 6′-position of the tetra-substituted benzene at
the 3-position of the indane sub-structure. The stereochemistry of the three chiral
carbons was deduced from the observed NOE correlations in the NOESY spectrum of 9.
Strong correlations between H1 and H2′ and between H3 and 1-CH3 revealed the trans-
relationship of the methyl group at the 1-position and the phenyl group at the 3-position.
The stereochemistry of C1″ could not be deduced from its NMR data due to its flexible
structure. Thus the structure of 9 was determined as depicted in Fig. 3.
22
ACS Paragon Plus Environment

Page 23 of 45
Journal of Agricultural and Food Chemistry
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
The 1H and 13C NMR data, including correlations in the COSY and HMBC, of
10 were very similar to those of 9 (Table S1 in supporting information and Fig. 4),
indicating that the planar structure of 10 was the same as that of 9. Differences in the
stereochemistry of 10 were deduced by NOESY. The observed NOE between H1 and
H3 in the NOESY spectrum of 10 revealed that the methyl group at the 1-position and
the phenyl group at the 3-position were cis relative to one another. Similar to 9, the
stereochemistry at the 1″-position could not be elucidated; thus, the structure of
compound 10 was determined as depicted in Fig. 3.
The HR-MS data of compound 13 revealed a molecular-related ion at m/z
269.0816 (Calcd for C14H14O4Na, 269.0790), indicating the molecular formula
C14H14O4, which is the same as that of compound 6. The NMR data of the tetra-
acetylated derivative (13a) of 13 showed the presence of a CH3–CH unit [4.25 (1H, q, J
= 7.0 Hz), 1.57 (3H, d, J = 7.0 Hz)], and six aromatic protons, which were very similar
to those of 6. A signal for an aromatic proton having two ortho-couplings [7.23 (1H, dd,
J = 8.6 and 8.2 Hz)] and signals of four phenolic acetyl groups indicated that 13 was a
regioisomer of 6 in reference to the attached position of the CH3–CH unit as shown in
Fig. 3.
The HR-MS data of compounds 17 and 18 indicated that both molecular
23
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 24 of 45
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
formulas were the same C18H18O14 (17: m/z 321.1137 [M+Na]+, 18: m/z 321.1133
[M+Na]+, both calculated for C18H18O4Na, 321.1103). The 1H-NMR spectrum of 17a,
which was obtained from 17 by reaction with acetic anhydride in pyridine, showed four
signals for phenolic acetyl groups at 2.25 (3H, s × 2), 2.27 (3H, s), and 2.30 (3H, s) ppm
and four singlet aromatic proton signals at 6.90, 6.92, 7.00 and 7.15 ppm, indicating the
presence of two sets of 1,2,4,5-tetra-substituted benzene rings derived from two
molecules of caffeic acid. The HH-COSY of 17a revealed the presence of a CH3 (1.35
ppm)–CH (2.81 ppm)–CH (2.55 ppm)–CH2 (2.82 and 3.23 ppm) coupling system,
which was adjacent to the aromatic rings. Another singlet signal (1.54 ppm) due to a
methyl group adjacent to a quaternary carbon was also observed. These data indicated
that 17 comprosed a bicyclo[3.3.0]octane structure appended with two
dihydroxybenzene rings and two methyl groups. The stereochemistry around the
bicylooctane rings was deduced by NOE correlations from H2 to the 1- and 3-methyl
protons in the NOESY spectrum of 17a, revealing the presence of cis relations. Thus,
the structure of 17 was determined as depicted in Fig. 3.
Although 17 and 18 presented the same molecular-related peak in their MS
data, the 1H NMR data of 18a (acetylated derivative of 18) were much simpler than
those of 17a, indicating that 18 was a symmetrical isomer of 17. The HH-COSY of 18a
24
ACS Paragon Plus Environment

Page 25 of 45
Journal of Agricultural and Food Chemistry
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
indicated the presence of a symmetrical coupling system, [CH3 (1.32 ppm)–CH (3.04
ppm)]2–CH (2.61 ppm)–CH (4.64 ppm), which was attached to two tetra-substituted
benzene rings. In addition to the above data, proton signals due to four phenolic acetyl
groups [2.26 (3H × 2) and 2.28 (3H × 2) ppm] and two sets of para-substituted aromatic
protons [6.95 (1H × 2) and 7.11 (1H × 2)] indicated that 18 had a symmetrical
bicyclo[3.3.0]octane structure similar to that of 17. NOEs were observed from 1- and 7′-
CH3 to H2 and H3 in the NOE difference spectra of 18. Thus, the structure of 18 was
determined as depicted in Fig. 3.
Structure-based classification of the thermal reaction products
The high-temperature treatment of caffeic acid stimulates its decarboxylation to
afford 4-vinylcatechol, which is significantly more reactive than the parent caffeic
acid.25 Therefore, 4-vinylcatechol presents an important intermediate of the thermally
mediated reactions of caffeic acid. By considering the structural differences of the
constituent compounds obtained from the thermal reactions, the identified compounds
were classified into three categories based on their vinylcatechol substructure: category
I (compounds 1–4) consists of the coupling products of vinylcatechol and caffeic acid,
category II (compounds 5, 7, 8, 9, 10, 15, and 16) consists of the coupling products of
25
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 26 of 45
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
two or three vinylcatechols, and category III (compounds 6, 13, 14, 17, and 18) is
composed of compounds formed by coupling with or by elimination of the catechol
moiety (Fig. 3). All the categorized compounds were assigned to the observed peaks in
the HPLC chromatograms from the reactions at 140, 170, and 200 ℃ as shown in Fig. 5.
The category I compounds were observed in the 140 ℃ reaction product along with
caffeic acid and the category II compounds. The HPLC data of the 170 ℃ reaction
product indicated a decrease in the intensities of caffeic acid and the category I
compounds. Additionally, the region representing the category II compounds (retention
time 30–40 min) became more complex compared to the same region in the
chromatogram of the 140 ℃ reaction product. This indicated that at the reaction
temperature 170 ℃, various vinylcatechol oligomers were additionally produced instead
of the caffeic acid-vinylcatechol compounds. Through consideration of the GPC
analytical data of the 170 ℃ reaction product (Fig. 2 B), it was determined that the
oligomerization degree of the newly produced compounds might be higher than that of
the compounds observed in the 140 ℃ reaction product. At the reaction temperature
200 ℃ the category I compounds disappeared, while the category III compounds were
observed in addition to the vinylcatechol oligomers (category II).
26
ACS Paragon Plus Environment

Page 27 of 45
Journal of Agricultural and Food Chemistry
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
XO inhibitory activities of the oligomeric compounds from the thermal reactions of
caffeic acid
Figure 6 illustrates the XO inhibitory activities of the isolated constituents (200
μmol/L) of the thermal reaction products. Although the starting material, caffeic acid,
showed very weak activity at the examined concentration, the isolated compounds
exerted stronger activities in the order 3 > 4 > 8, 10, 17 > 7, 9, 2, 5, 1, 6 > 16, 15, 14,
and 13, based on their percent inhibition values. The compounds with the highest
inhibition levels, caffeic acid-vinylcatechol compounds 3 and 4, were detected in the
products of the 140 and 170 ℃ reactions but were absent in the product of the 200 ℃
reaction. There were followed by compounds 7, 8, (vinylcatechol dimers) and 10
(trimer), which were mainly observed in the products of the reactions at 140 and 170 ℃.
Conversely, the weakest inhibitory compounds 13 and 14, catechol-eliminated
compounds belonging to category III, were both observed solely as the constituents of
the product of the 200 ℃ reaction.
A comparison of the results of the XO inhibitory activity of the identified
compounds with those of the thermal reaction products (110–290 ℃, Fig. 1) indicated
that caffeic acid was initially converted to caffeic acid-vinylcatechol compounds, some
of which contributed to the enhanced inhibitory activity of the products produced at
27
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 28 of 45
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
170 ℃ or less. Subsequently, caffeic acid was converted to the moderately active
vinylcatechol oligomers followed by their transformation to weakly active catechol-
reacted compounds, the production of which decreases the XO inhibitory activity of the
thermal reaction products produced at the temperatures above 170 ℃. Although not all
the produced constituents of the thermal reaction products could be identified due to
their complexity, the observed structural differences of the constituents might induce
the enhanced XO inhibitory activity of thermal reaction products. Notably, the thermal
treatment at 170 ℃, which exhibited the maximum XO inhibitory activity, coincides
with the general roasting conditions of food processing.26 Caffeic acid and its
derivatives including chlorogenic acid, are widely distributed in plant-derived
foodstuffs.27 Therefore, the roasting processing of foodstuffs might present significant
XO inhibitory activity by the thermal reaction of caffeic acid derivatives. In terms of
coffee beverages, we previously reported that only roasted coffee beans exhibit XO
inhibitory activity,16 which enhances the ability of coffee to prevent gout.28 Frank et
al.13 reported the existence of various vinylcatechol oligomers in roasted coffee, as
bitter-tasting compounds, by using LC-MS techniques. Although pyrogallol and
chlorogenic lactones have been identified as significant XO inhibitors present in roasted
coffee beans,16,17 vinylcatechol oligomers and related compounds might also play a role
28
ACS Paragon Plus Environment

Page 29 of 45
Journal of Agricultural and Food Chemistry
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
in the XO inhibitory activity. Differences in the inhibitory ability due to the different
reaction temperatures should also be considered when determining the conditions of the
roasting process of raw coffee beans. Optimized roasting conditions should be useful for
the production of health-functional coffee beverages that exhibit XO inhibitory activity.
Although the bioavailability of the identified phenylindanes and vinylcatechol
oligomers has not been examined to date, the ability of caffeic acid and the related
phenolic acids in coffee has been well studied.29,30 These compounds can be absorbed in
humans in their conjugated and non-conjugated forms. Ota et al. 31 examined the
interaction of these phenolic acids and corresponding vinylcatechol derivatives with
lipid membranes. The results revealed that the more non-polar vinylcatechols displayed
stronger interactions with the membrane, which were related to their bioavailability.
The identified thermal reaction products are also non-polar substances compared to
caffeic acid; therefore, higher bioavailability is expected for the product constituents.
When the non-polar thermal reaction products of caffeic acid are absorbed by humans,
they might exhibit characteristic functions in the human body. XO is localized in the
liver and intestine and produces uric acid, 32,33 the over accumulation of which is
responsible for the development of gout.28 The absorbed thermal products might play a
role in inhibiting XO from overworking when reaching liver or intestine, thereby
29
ACS Paragon Plus Environment