Hydroxycinnamoyl Glucose and Tartrate Esters and Their Role in the Formation of Ethylphenols in Wine

Article abstract of DOI:10.1021/acs.jafc.6b04074

Synthesized p-coumaroyl and feruloyl l-tartrate esters were submitted to Brettanomyces bruxellensis strains AWRI 1499, AWRI 1608, and AWRI 1613 to assess their role as precursors to ethylphenols in wine. No evolution of ethylphenols was observed. Additionally, p-coumaroyl and feruloyl glucose were synthesized and submitted to B. bruxellensis AWRI 1499, which yielded both 4-ethylphenol and 4-ethylguaiacol. Unexpected chemical transformations of the hydroxycinnamoyl glucose esters during preparation were investigated to prevent these in subsequent synthetic attempts. Photoisomerization gave an isomeric mixture containing the trans-esters and undesired cis-esters, and acyl migration resulted in a mixture of the desired 1-O-β-ester and two additional migrated forms, the 2-O-α- and 6-O-α-esters. Theoretical studies indicated that the photoisomerization was facilitated by deprotonation of the phenol, and acyl migration is favored during acidic, nonaqueous handling. Preliminary LC-MS/MS studies observed the migrated hydroxycinnamoyl glucose esters in wine and allowed for identification of feruloyl glucose in red wine for the first time.

Full text of DOI:10.1021/acs.jafc.6b04074

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)
Article
Hydroxycinnamoyl Glucose and Tartrate Esters and
Their Role in the Formation of Ethylphenols in Wine
Josh Lee Hixson, Yoji Hayasaka, Chris D. Curtin, Mark A Sefton, and Dennis Kenwyn Taylor
J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.6b04074 • Publication Date (Web): 19 Nov 2016
Downloaded from http://pubs.acs.org on November 24, 2016
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 free 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 accessible to all
readers and 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.
Journal of Agricultural and Food Chemistry 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 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 1
Hydroxycinnamoyl Glucose and Tartrate Esters and Their Role in the
Formation of Ethylphenols in Wine
Josh L. Hixson,†∞* Yoji Hayasaka,¶ Christopher D. Curtin,¶ Mark A. Sefton† and Dennis K.
Taylor,†
† School of Agriculture, Food and Wine, The University of Adelaide, Waite Campus, PMB 1,
Glen Osmond, SA 5064, Australia.
¶ The Australian Wine Research Institute, P.O. Box 197, Glen Osmond, Adelaide, 5064,
Australia.
∞ Current address: The Australian Wine Research Institute.
Corresponding Author
Ph: +61ꢀ8ꢀ8313 6600; Fax +61ꢀ8ꢀ8313 6601; email josh.hixson@awri.com.au.
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 2 of 37
Hydroxycinnamoyl Glucose Esters
Page 2
1
2
3
4
5
6
7
8
9
ABSTRACT
Synthesized pꢀcoumaroyl and feruloyl
L
ꢀtartrate esters were submitted to Brettanomyces
bruxellensis strains AWRI 1499, 1608 and 1613 to assess their role as precursors to ethylphenols in
wine. No evolution of ethylphenols was observed. Additionally, pꢀcoumaroyl and feruloyl glucose
were synthesized and submitted to B. bruxellensis AWRI 1499 which yielded both 4ꢀethylphenol
and 4ꢀethylguaiacol. Unexpected chemical transformations of the hydroxycinnamoyl glucose esters
during preparation were investigated to prevent these in subsequent synthetic attempts.
Photoisomerization gave an isomeric mixture containing the transꢀesters and undesired cisꢀesters,
and acyl migration resulted in a mixture of the desired 1ꢀO-
βꢀester and two additional migrated
10
11
12
13
14
15
16
17
forms, the 2ꢀO- ꢀ and 6ꢀO- ꢀesters. Theoretical studies indicated that the photoisomerization was
α
α
facilitated by deprotonation of the phenol, and acyl migration is favored during acidic, nonꢀaqueous
handling. Preliminary LCꢀMS/MS studies observed the migrated hydroxycinnamoyl glucose esters
in wine, and allowed for identification of feruloyl glucose in red wine for the first time.
KEYWORDS:
Brettanomyces, hydroxycinnamate ester, ethylphenol,
photoisomerization, acyl migration.
ACS Paragon Plus Environment

Page 3 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 3
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
INTRODUCTION
The formation of 4ꢀethylphenol (7) and 4ꢀethylguaiacol (8) in red wine has previously been
linked to the breakdown of free hydroxycinnamic acids (3 and 4) by Brettanomyces yeast.1
Recently, we have shown that some commonly isolated strains of B. bruxellensis from Australia
also possess the ability to convert ethyl hydroxycinnamates (1 and 2) to ethylphenols to varying
extents (Figure 1),2 and that these esters can therefore contribute to the undesirable aromas and
flavors associated with wines effected by Brettanomyces. Presumably, this occurs via esterase
mediated release of hydroxycinnamic acids from the ethyl esters and this supposition has lead us to
investigate the possibility that other hydroxycinnamoyl esters present during wine maturation could
be metabolized by B. bruxellensis and contribute to the accumulation of ethylphenols in red wine.
Hydroxycinnamoyl Lꢀtartaric acid esters, including pꢀcoumaroyl and feruloyl Lꢀtartrate (9
and 10), have been quantified during the vinification process,3, 4 and concentration of the tartrates in
wine has been shown to be greatly reduced from that of the initial must concentrations with most of
this loss seen during vinification, and a smaller amount lost during the ageing process.4, 5 The
decrease in tartrate concentration during winemaking has been linked to fermentation and the yeast
Saccharomyces cerevisiae,6 malolactic fermentation and the bacteria Oenococcus oeni,7 as well as
wine ageing,5 and commercial enzyme preparations added to must.8, 9 One study linked a decrease
in tartrate esters to the yeast genus Brettanomyces but did not correlate these loses with an increase
in volatile phenol production,10 while another implied that Brettanomyces is unable to metabolize
the tartrate esters of hydroxycinnamic acids.11 Model fermentation experiments with synthesized
hydroxycinnamoyl tartrates would definitively assess the ability of Brettanomyces to produce
ethylphenols from the tartrate esters.
In addition to the tartrate esters, the existence of pꢀcoumaroyl and feruloyl glucose esters (11
and 12) was confirmed in white grapes12 and Riesling wine,13 followed by the identification of pꢀ
coumaroyl glucose esters in red wines.14 The identification of two separate pꢀcoumaroyl glucose
esters in several red wines was achieved via liquidꢀliquid extraction,15 with the authors presuming
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 4 of 37
Hydroxycinnamoyl Glucose Esters
Page 4
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
that the two esters differed by the position of attachment to glucose. In subsequent quantification of
these two esters over 26 months of ageing in three wines the glucose ester concentration was in the
range of 0.1ꢀ1.0 mg/L and only fluctuated slightly over time,14 indicating that the glucose esters are
stable over long periods of storage. Another group, using a similar extraction procedure identified
two pꢀcoumaroyl hexoses with initial concentrations of 1.23 and 1.51 mg/L, increasing throughout
malolactic fermentation.7 Subsequent studies have also identified two pꢀcoumaroyl glucose esters in
red grapes or wine, without providing further structural information.16, 17 To date there is no study
linking hydroxycinnamoyl glucose esters to metabolism by Brettanomyces.
Hydroxycinnamates can undergo photoisomerization via ultraꢀviolet radiation which is
dependent on the substrate, solvent, pH, wavelength of incident light and is susceptible to
concentration effects.18-21 The cis/transꢀisomerization of pꢀcoumaric acid, and related derivatives,
has been the subject of numerous theoretical studies,22-27 but the isomerization, and even the
different isomers of the hydroxycinnamoyl glucose esters are yet to be studied experimentally or
theoretically. The stereochemistry of hydroxycinnamates is of importance in the accumulation of
ethylphenols as the decarboxylase of B. bruxellensis was recently observed to only be active
towards the transꢀhydroxycinnamic acids.28
Additionally, acyl migrations, or intraꢀmolecular transesterifications, within glucose
derivatives are widely known. Examples of migrations around glucose rings by small or simple
moieties have been observed for decades,29, 30 and can yield different outcomes under changing
conditions,31 including catalysis by silica.32 The products and final outcomes of acyl migrations are
determined by thermodynamic influences,33 the speed at which these products can be formed are
governed by mechanistic restraints such as the ringꢀsize of the intermediates formed34 and the
electronic characteristics of the esters involved.35 Although acyl migration has not previously been
described for hydroxycinnamoyl glucose esters in these mechanistic studies,33, 36 they provide a
starting point for examining the possibility that multiple glucose esters previously observed in wine
extracts could be products of acyl migration.
ACS Paragon Plus Environment

Page 5 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 5
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
This paper details the synthesis of hydroxycinnamoyl tartrate and glucose esters, and role of
both classes of hydroxycinnamoyl ester as potential precursors to ethylphenols was examined by
submitting them to fermentation experiments with B. bruxellensis. Computational studies into the
observed photoisomerization and acyl migrations during synthesis of the hydroxycinnamoyl glucose
esters, and preliminary studies into the presence of migrated forms of the glucose esters in wine
were both undertaken to prevent these unwanted transformations in future synthetic attempts.
MATERIALS AND METHODS
Chemicals. Dry organic solvents were purchased and dispensed using a Puresolv™ solvent
purification system (Innovative Technologies, Massachusetts, USA). General organic solvents were
obtained (Chemsupply, Adelaide, Australia) and distilled where needed; other chemicals were
purchased (Sigma-Aldrich Chemical Company, Castle Hill, Australia).
General. Reverseꢀphase C18 chromatography was performed using preꢀpacked cartridges
designed for use with automated flash chromatography systems (Teledyne Isco, Nebraska, USA).
The cartridge was loaded with compound dissolved in a minimal volume of methanol, before being
washed with water, eluted with acetonitrile/water/formic acid (30:69:1), flushed with
acetonitrile/formic acid (99:1), and reꢀequilibrated with water. The 1H NMR spectra were acquired
with a Bruker Ultrashield Plus 400 MHz or 600 MHz spectrometer as where specified.
Synthesis of substrates and standards. Detailed synthetic preparations and additional
characterization data can be found in the supporting information.
1-O-Chloroacetyl hydroxycinnamates (13 and 14). Preparation of 13 and 14 were achieved
using a literature procedure.37 Characterization data of 13 was as previously described,37 NMR
assignment for 14 was made based on 13.
Tartrate coupling reactions. Synthesis of diꢀtertꢀbutyl
was achieved via O,O’ꢀdiacetyl ꢀtartaric anhydride (15), O,O’ꢀdiacetyl
O,O’ꢀdiacetylꢀdiꢀtertꢀbutyl
L
ꢀtartrate (18) from
Lꢀtartaric acid
L
Lꢀtartaric acid (16) and
L
ꢀtartrate (17) based on literature procedures.38-42 Characterization data
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 6 of 37
Hydroxycinnamoyl Glucose Esters
matched literature reports for 15,40 16,40 1742 and 18.42 Coupling reactions of chloroacetyl
Page 6
96
97
hydroxycinnamate (13 or 14) with 18 to give 19 or 21were achieved in an analogous fashion to
98
previous reports.43
99
p-Coumaroyl di-tert-butyl L δ: 7.60 (d, 1H, J =
-tartrate (19). 1H NMR: (400 MHz, CDCl3)
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
15.9 Hz, H7), 7.29 (app. d, 2H, J = 8.6 Hz, H3,5), 6.84 (app. d, 2H, J = 8.6 Hz, H2,6), 6.18 (d, 1H, J =
15.9 Hz, H8), 5.51 (d, 1H, J = 2.3 Hz, H2’), 4.68 (d, 1H, J = 2.3 Hz, H3’), 3.34 (br. s, 1H, OH), 1.52
(s, 9H, tꢀBu4), 1.44 (s, 9H, tꢀBu1).
Feruloyl di-tert-butyl L δ: 7.68 (d, 1H, J = 15.9
-tartrate (21). 1H NMR: (600 MHz, CDCl3)
Hz, H7), 7.07 (dd, 1H, J = 8.1 and 1.8 Hz, H5), 7.04 (d, 1H, J = 1.8 Hz, H3), 6.92 (d, 1H, J = 8.1 Hz,
H6), 6.35 (d, 1H, J = 15.9 Hz, H8), 5.50 (d, 1H, J = 2.3 Hz, H2’), 4.67 (dd, 1H, J = 6.9 and 2.3 Hz,
H3’), 3.93 (s, 3H, OCH3), 3.20 (d, 1H, J = 6.9 Hz, OH), 1.51 (s, 9H, tꢀBu4), 1.44 (s, 9H, tꢀBu1).
p-Coumaroyl L-tartrate (9). pꢀCoumaroyl tertꢀbutyl Lꢀtartrate (19) (46.2 mg, 0.11 mmol)
was dissolved in dry dichloromethane (5 mL) followed by the addition of trifluoroacetic acid (0.18
mL, 2.29 mmol) and the mixture was stirred at room temperature under a nitrogen atmosphere for
24 hours before the solvent was removed in vacuo. Purification by reversedꢀphase chromatography
on C18 gave 9 (27.5 mg, 82%) as an amorphous solid. Rf (20% MeOH/CH2Cl2): 0.00. 1H NMR:
(400 MHz, CD3OD) δ: 7.74 (d, 1H, J = 15.9 Hz, H7), 7.48 (app. d, 2H, J = 8.7 Hz, H3,5), 6.81 (app.
d, 2H, J = 8.7 Hz, H2,6), 6.38 (d, 1H, J = 15.9 Hz, H8), 5.55 (d, 1H, J = 2.3 Hz, H2’), 4.77 (d, 1H, J =
2.3 Hz, H3’). 13C NMR spectral properties were as previously described.44
Feruloyl L-tartrate (10). Feruloyl tertꢀbutyl Lꢀtartrate (21) (35.9 mg, 0.082 mmol) was
submitted to the same reaction conditions as described for 19 (above). This gave 10 (11.0 mg, 41%)
as an offꢀwhite amorphous solid. Rf (20% MeOH/CH2Cl2): 0.00. 1H NMR: (400 MHz, CD3OD)
δ:
7.73 (d, 1H, J = 16.0 Hz, H7), 7.20 (d, 1H, J = 1.9 Hz, H3), 7.10 (dd, 1H, J = 8.2 and 1.9 Hz, H5),
6.82 (d, 1H, J = 8.2 Hz, H6), 6.41 (d, 1H, J = 16.0 Hz, H8), 5.57 (d, 1H, J = 2.4 Hz, H2’), 4.78 (d,
1H, J = 2.4 Hz, H3’), 3.89 (s, 3H, OCH3). 13C NMR spectral properties were as previously
reported.44, 45
ACS Paragon Plus Environment

Page 7 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Glycosylation reactions. The preparation of 2,3,4,6ꢀtetraꢀOꢀchloroacetylꢀαꢀDꢀ
Page 7
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
glucopyranosyltrichloroacetimidate (25) from Dꢀglucose and the method of glycosylation were as
previously reported.46 The characterization data for the product and intermediates (23ꢀ25) were
analogous to previous reports although anomeric ratios tended to vary.37, 46, 47
2,3,4,6-Tetra-O-chloroacetyl-β-D-glucopyranosyl 1-O-chloroacetyl coumarate (26). From
13 (210.6 mg, 0.88 mmol), gave 0.30 g (48%) of 26 as a white honeycomb. 1H NMR: (400 MHz,
CDCl3) δ: 7.73 (d, 1H, J = 16.0 Hz, H7), 7.58 (app. d, 2H, J = 8.7 Hz, H3,5), 7.20 (app. d, 2H, J =
8.7 Hz, H2,6), 6.37 (d, 1H, J = 16.0 Hz, H8), 5.90 (d, 1H, J = 8.2 Hz, H1’), 5.44 (dd, 1H, J = 9.6 and
9.5 Hz, H3’), 5.33 (dd, 1H, J = 9.5 and 8.2 Hz, H2’), 5.26 (dd, 1H, J = 9.7 and 9.6 Hz, H4’), 4.42 (dd,
1H, J = 12.5 and 4.3 Hz, H6a’), 4.33 (m, 3H, ArOCOCH2Cl and H6b’), 4.12 (app. s, 2H, OCH2Cl),
4.04ꢀ4.01 (m, 7H, 3 x OCH2Cl and H5’).
2,3,4,6-Tetra-O-chloroacetyl-β-D-glucopyranosyl 1-O-chloroacetyl ferulate (27). From 14
(224.0 mg, 0.83 mmol), afforded 0.38 g (64%) of 27 as a paleꢀyellow honeycomb. NMR and mass
spectral properties were as previously reported.46
General procedure for de-chloroacetylation (causing cis/trans-isomerization). 2,3,4,6ꢀTetraꢀ
OꢀchloroacetylꢀβꢀDꢀglucopyranosyl hydroxycinnamate (26 or 27) (100.0 mg) was dissolved in
pyridine/water (1:1, 10 mL) and stirred at room temperature for 6 hours. The reaction mixture was
concentrated and the crude mixture purified using XADꢀ8 resin (eluted with 60% MeOH/H2O) to
give a mixture of cisꢀ and transꢀβꢀDꢀglucopyranosyl hydroxycinnamate (11 or 12) as a colorless
residue. The spectrum of the cisꢀisomer was extracted from the mixture. See below for
characterization data for the transꢀisomers.
cisꢀp-Coumaroyl glucose. 1H NMR: (400 MHz, CD3OD)
δ: 7.73 (m, 2H, H3,5), 6.94 (d, 1H,
J = 12.9 Hz, H7), 6.82 (app. d, 2H, J = 8.8 Hz, H2,6), 5.82 (d, 1H, J = 12.9 Hz, H8), 5.55 (d, 1H, J =
8.0 Hz, H1’), 3.85 (m, 1H, H6a’), 3.68 (m, 1H, H6b’), 3.47ꢀ3.32 (m, 4H, H2’, 3’, 4’, 5’). Assignment and
identification of the cisꢀisomer was performed using the known transꢀisomer (transꢀ11) and the data
for the cisꢀaglycone (cisꢀ3).48, 49
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 8 of 37
Hydroxycinnamoyl Glucose Esters
Page 8
1
148
149
150
151
152
153
154
155
156
157
158
cisꢀFeruloyl glucose. H NMR: (400 MHz, CD3OD)
δ
: 7.87 (d, 1H, J = 1.9 Hz, H3), 7.17
(dd, 1H, J = 8.3 and 1.9 Hz, H5), 6.94 (d, 1H, J = 13.0 Hz, H7), 6.77 (d, 1H, J = 8.3 Hz, H6), 5.83 (d,
1H, J = 13.0 Hz, H8), 5.56 (d, 1H, J = 7.8 Hz, H1’), 3.88 (s, 3H, OCH3), 3.88ꢀ3.84 (m, 1H, H6a’),
3.72ꢀ3.66 (m, 1H, H6b’), 3.49ꢀ3.35 (m, 4H, H2’,3’,4’,5’). Assignment and identification of the cisꢀ
isomer was performed using the known transꢀisomer (transꢀ12) and the data for the cisꢀaglycone
(cisꢀ4).48, 50
General procedure for de-chloroacetylation (causing migration). 2,3,4,6ꢀTetraꢀOꢀ
chloroacetylꢀβꢀDꢀglucopyranosyl hydroxycinnamate (26 and 27) (300.0 mg) was dissolved in
pyridine/water (1:1, 20 mL) and stirred at room temperature in the dark for 6 hours. Only being
exposed to red light, the reaction mixture was concentrated and the crude mixture purified using
column chromatography (10% MeOH/CH2Cl2) to give migrated mixtures of the transꢀ
βꢀDꢀ
159
glucopyranosyl hydroxycinnamate, but largely consisting of the 1ꢀOꢀ ꢀester (approx. 80% for 11
β
160
161
162
163
164
165
166
167
168
169
170
171
172
173
and 90% for 12). The migrated mixture was found to revert back to the 1ꢀOꢀβꢀester after standing in
buffered water at wine pH (3.5).
p-Coumaroyl glucose (11). From 26 (261.7 mg, 0.37 mmol), gave 51.5 mg (43%) of 11 as a
white residue. Rf (20 % MeOH/CH2Cl2): 0.29. MS (ꢀEI) m/z (%): 325.7 (Mꢀ, 100), 265.5 (7), 187.7
1
(8), 163.4 (21), 145.2 (44). H NMR: (400 MHz, CD3OD) δ: 7.73 (d, 1H, J = 15.9 Hz, H7), 7.48
(app. d, 2H, J = 8.5 Hz, H3,5), 6.82 (app. d, 2H, J = 8.5 Hz, H2,6), 6.37 (d, 1H, J = 15.9 Hz, H8), 5.57
(d, 1H, J = 7.9 Hz, H1’), 3.85 (dd, 1H, J = 12.1 and 1.8 Hz, H6a’), 3.69 (dd, 1H, J = 12.1 and 4.6 Hz,
H6b’), 3.45ꢀ3.38 (m, 4H, H2’,3’,4’,5’). 13C NMR spectral properties for the 1ꢀO-
βꢀester were as
previously reported.13
Feruloyl glucose (12). From 27 (501.8 mg, 0.68 mmol), 47.5 mg (20%) of 12 as an offꢀ
white residue. Rf (20% MeOH/CH2Cl2): 0.32. MS (ꢀEI) m/z (%): 355.3 (Mꢀ, 100), 295.5 (8), 217.2
(20), 193.6 (25), 175.4 (32). 1H NMR: (400 MHz, CD3OD) δ: 7.73 (d, 1H, J = 15.9 Hz, H7), 7.21
(d, 1H, J = 1.9 Hz, H3), 7.10 (dd, 1H, J = 8.2 and 1.9 Hz, H5), 6.82 (d, 1H, J = 8.2 Hz, H6), 6.41 (d,
1H, J = 15.9 Hz, H8), 5.58 (d, 1H, J = 7.5 Hz, H1’), 3.90 (s, 3H, OCH3), 3.86 (dd, 1H, J = 12.1 Hz,
ACS Paragon Plus Environment

Page 9 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 9
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
H6a’), 3.70 (dd, 1H, J = 12.1 and 4.5 Hz, H6b’), 3.49ꢀ3.35 (m, 4H, H2’, 3’, 4’, 5’). 13C NMR spectral
properties for the 1ꢀO-β
ꢀester were as previously reported.13
Migrated glucose esters. 1H NMR shifts of migrated glucose esters were partially
determined for glucose proton shifts that were significantly different from those of the 1ꢀOꢀβꢀester,
with assignments made using coupling patterns and changes in chemical shifts analogous to
literature data.51
Computational studies. Theoretical calculations were performed using the Spartan ’08
package (Wavefunction Inc., California, USA) with geometry optimizations and final energies
determined using the DFT B3LYP 6ꢀ31G* basis set. Full experimental conditions for the
computational studies can be found in the supporting information.
Wine samples for analysis. One white wine (Stanley Classic Dry White) and one red wine
(Yalumba 1997 Shiraz) were used throughout the study. Concentrated wine samples were prepared
from 50 mL of wine at 30 oC under reduced pressure until the volume had reduced to 5 mL.
Unconcentrated and concentrated wine samples were passed through a 45 ꢀm syringe filter and
analyzed directly. Standards of pꢀcoumaroyl glucose (11) and feruloyl glucose (12) were prepared
(in methanol, 10 mg/L) to determine retention times and obtain reference mass spectral data.
HPLC-MS analysis of wine samples. HPLCꢀMS or MS/MS analysis was carried out using
a 4000 Q TRAP hybrid tandem mass spectrometer interfaced with a Turbo V ion source for
elecrospray ionization (AB Sciex AB Sciex, Foster City, CA), combined with an Agilent 1200
HPLC system equipped with a binary pump, degasser, autosampler, column oven, and photodiode
array (PDA) detector.
HPLC conditions. A 10 µL aliquot of the samples was injected and chromatographed using
a 250 x 4.6 mm, 3 ꢀm, 100 Å Luna C18 column, operated at 25 ºC and protected by a C18 guard
column (4 x 2 mm) (Phenomenex, Lane Cove, NSW, Australia). The eluents were formic
acid/water (0.5:99.5 v/v, Eluent A) and formic acid/acetonitrile/water (0.5:25.0:74.5 v/v, Eluent B)
with a flow rate of 1 mL/min. A gradient was applied as follows: 20% to 30% B linear from 0 to 20
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 10 of 37
Hydroxycinnamoyl Glucose Esters
minutes; 30% to 50% B linear from 20 to 50 minutes; 50% B to 100% B linear from 50 to 60
Page 10
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
minutes; 100% B to 20% B from 60 to 65 minutes. The column was equilibrated with 20% B for 10
minutes prior to an injection. The eluent from the HPLC was split by use of a splitter (a tee) and
delivered at a follow rate of 0.45 mL/min to the mass spectrometer and at 0.55 mL/min to the PDA
detector with monitoring wavelengths at 290, 320 and 370 nm with a slit width of 4 and a
bandwidth of 16 nm.
Electrospray and mass spectrometric conditions. All mass spectrometric data were
obtained in negative ion mode. Nitrogen gas was used for the curtain, nebulizer, turbo and collision
gases. The Turbo V ion source parameter were set at ꢀ3500 V for the ion spray potential, ꢀ60 V for
the declustering potential, ꢀ10 V for the entrance potential, 50 psi for gas 1 (nebulizer) and gas 2
(turbo), 15 psi for the curtain gas, and 500 °C for the turbo gas (gas 2) temperature.
For tandem mass spectrometry, the collision potential was set in an appropriate range from ꢀ
15 to ꢀ25 V and the collision gas pressure was set at high. Product ion spectra of m/z 325 for pꢀ
coumaroyl glucose and m/z 355 for feruloyl glucose were recorded in a mass range from m/z 50 to
400 with a scan time of 1 s and a step mass of 0.1. For multiple reaction monitoring (HPLCꢀMRM),
the following mass transitions were monitored with a dwell time of 50 ms; m/z 325 119, 145, 163
and 187 for pꢀcoumaroyl glucose, and m/z 355 119, 175, 193 and 217 for feruloyl glucose.
Yeast and growth/Fermentation experiments/Ethylphenol analyses. These were as
previously described.2 Briefly, starter cultures of B. bruxellensis were prepared in YPD (Yeast
Peptone Dextrose). Model fermentation experiments were performed in triplicate in YNB (Yeast
Nitrogen Base) media (supplemented with 20 g/L glucose, pH 3.5). Ferments were spiked with
either 9 and 10 or 11 and 12 at 10 mg/L, inoculated with a single strain of B. bruxellensis and
incubated at 28 oC. Control experiments were run concurrently, performed under analogous
conditions without yeast inoculation. Samples (5 mL) were collected every second day, centrifuged
(4000 r.p.m. for 5 mins) and the supernatant decanted from the yeast pellet and stored at ꢀ20 oC
until required for analysis. Ethylphenol analysis was performed as previously described.52
ACS Paragon Plus Environment

Page 11 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 11
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
RESULTS AND DISCUSSION
Synthesis of hydroxycinnamoyl tartrate esters (Figure 2). Preparation of diꢀtertꢀbutyl
tartrate (18) from ꢀtartaric acid was achieved as shown. The acid chlorides of 13 and 14 were
L
formed, concentrated and used directly, being added dropꢀwise in dry hexane to 18 in dry
pyridine,43 which for short reaction times gave the chloroacetylated products (20 or 22) with minor
dechloroacetylation. However, longer exposure to pyridine gave further deprotection to give mainly
the desired products (19 or 21), but always with some intermediate 20 or 22, that could be isolated
and deprotected further. Hydrolysis of the tertꢀbutyl esters and purification with reverseꢀphase
chromatography yielded pure hydroxycinnamoyl tartrate esters (9 or 10).
Synthesis of hydroxycinnamoyl glucose esters (Figure 3). The synthesis of pꢀcoumaroyl
and feruloyl glucose was based on the glucosyl donor (25) used previously to prepare
hydroxycinnamoyl glucose esters,37, 47 and the synthetic pathway has since been validated in
preparation of feruloyl and sinapoyl glucose.46
Glycosylation between the glycosyl donor, 25, and acids (13 or 14) gave pentaꢀchloroacetyl
protected esters 26 or 27. Removal of the chloroacetyl groups gave a mixture of the desired
products (11 or 12) and 6ꢀOꢀchloroacetyl glucopyranosyl hydroxycinnamates (28 or 29). Under
ambient light conditions an initial attempt to separate the product resulted in rapid
photoisomerization to give cisꢀ11 or 12, which could be controlled by handling in the dark, or under
red light. Using a modified purification method, flash chromatography to remove the monoꢀ
protected species, 28 and 29, using 10% methanol in dichloromethane, and indeed multiple solvent
combinations, resulted in acyl migration. This gave a combination of esters mostly consisting of the
desired 1ꢀO-
β
-ester as well as significant amounts of 2ꢀO-αꢀ and 6ꢀO-αꢀesters and trace amounts of
249
250
251
the 2ꢀO- ꢀ, 6ꢀO-
β
β
− and 3ꢀO- ꢀesters, which could be partially characterized. In these minor
α/β
esters the chemical shifts for hydroxycinnamate alkene protons (H7 and H8) were moved upfield a
negligible amount, and for each positional isomer the glucose proton shifts were identical between
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 12 of 37
Hydroxycinnamoyl Glucose Esters
Page 12
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
feruloyl and pꢀcoumaroyl derivatives. In the mixtures a greater extent of migration for the pꢀ
coumaroyl ester was observed. It was found that when migrated mixtures were stored under
aqueous acidic conditions glucose esters reverted back to the 1ꢀO-β-esters, which will be detailed
below.
cis/trans-Isomerization of glucose esters of p-coumaric and ferulic acids. The ease of
cis/transꢀisomerization that occurred in the glucose esters, and comparative lack thereof in other
hydroxycinnamates prepared synthetically in this work, led to the investigation of the nature of the
isomerization and the factors that influenced this phenomenon in the glucose esters. Initial
investigations into excitation to the first excited triplet state (T1), calculated in a vacuum, gave
energy profiles for p-coumaric acid and pꢀcoumaroyl glucose that explained the propensity for
photoisomerization, with the lowest energy conformations possessing a C6ꢀC7ꢀC8ꢀC9 dihedral angle
of approximately 90o (Figures 4a and 4b). However, the energy differences between the singlet
ground state (S0) and T1 for the transꢀisomers (180o dihedral) for pꢀcoumaric acid and pꢀcoumaroyl
glucose were found to be 229.4 and 226.2 kJ/mol, respectively, and not considered large enough to
explain the difference in isomerization away from the transꢀisomer based on substrate alone.
During the preparation of numerous hydroxycinnamate derivatives, changes in solubility
required different compounds to be handled in various solvents, with the glucose esters handled in
more polar solvents than the free hydroxycinnamic acid were. To understand the role that solvents
may play in hydroxycinnamate isomerization, solvents of differing polarities (expressed as ET30)53
were introduced into the calculations. The vertical excitation energy of the transꢀsubstrate, or the
energy required to achieve an excited state and facilitate isomerization away from the trans-
isomers, showed reduction with increasing polarity (Figure 4c). Additionally, the vertical excitation
energies determined for pꢀcoumaroyl glucose were lower than those for pꢀcoumaric acid for all
solvents calculated. This trend was also mirrored in the HOMOꢀLUMO gap for each S0 species
(Figure 4d). The minor changes in vertical excitation energy and HOMOꢀLUMO gap with solvents
of differing polarities, along with minor changes also due to substrate, start to provide an
ACS Paragon Plus Environment

Page 13 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 13
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
explanation as to the ease of isomerization observed for the glucose esters, assisting in lowering the
energy required to achieve an excited state and hence, isomerization.
The mirrored trends in S0ꢀT1 vertical excitation energy and S0 HOMOꢀLUMO gap removed
the need to calculate the energy of both the S0 and T1 state and allowed for rapid determination of
energy barriers simply from the HOMOꢀLUMO gap of the S0 state. Hence a representation of ease
of isomerization for multiple hydroxycinnamates could be rapidly studied by examining the
relationship between the HOMOꢀLUMO gap and structural features. With phenolic deprotonation
shown to lower excitation energy in pꢀcoumaric acid,25 as well as pH effecting isomeric ratios in
neutral hydroxycinnamates,18 the structures studied were chosen because of their differing
electronic configurations with respect to their ability to become resonance stabilized by movement
of electrons from the phenol through the alkene and towards the carboxyl group. This included
structures possessing carboxyl anions and phenolic protection that are expected to have very little
resonance contribution resulting in an alkene with increased double bond character (Table 1).
Across the 11 structures, the HOMOꢀLUMO gap correlated with both the charge ratio
between oxygens 1 and 3 (rꢀvalue = ꢀ0.89, P = 0.0002) and the alkene double bond length (rꢀvalue =
ꢀ0.9677, P < 0.0001), and adequately explained what was observed experimentally. By altering the
electron density at the extremities of the molecules (oxygens 1 and 3), the double bond character of
the C7=C8 alkene was being affected (as shown by double bond length) as was the energy required
to excite an electron to the first excited state (HOMOꢀLUMO gap) and facilitate isomerization. As
such, hydroxycinnamate esters that are handled under basic conditions, producing a phenolate
anion, will possess a lower HOMOꢀLUMO gap and a higher propensity to isomerize. Substrate,
solvent, pH, and electronic environment influence the energy required for isomerization between
hydroxycinnamates, and these results support what was observed experimentally. Environments that
contribute to increasing the electron density on the phenolic oxygen, or even diminish electron
density at the carboxyl oxygen should be avoided under most lighting conditions.
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 14 of 37
Hydroxycinnamoyl Glucose Esters
Page 14
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
Migration of glucose esters. To better understand the acyl migration of the glucose esters,
to explain the ratios of products observed and to investigate the possibility of this occurring in wine
or organic solvent wine extracts, theoretical studies were first performed. Water and
dichloromethane were used to study the positional isomers, the most common solvents that the
glucose esters experienced, along with ethanol and toluene which were included as examples of
polar protic and apolar aprotic solvents. The relative energies of the esters in water (Figure 5)
justified the predominance of the 1ꢀOꢀβꢀesters in aqueous environments, being the
thermodynamically most stable, though this is not the case for any of the other solvents studied
(Figure 5). In dichloromethane, ethanol and toluene a thermodynamic preference for migration to
the 6ꢀOꢀαꢀester was observed. The 2ꢀOꢀαꢀester was also lower in energy than the 1ꢀOꢀβꢀester in
dichloromethane and ethanol. For NMR characterization of the glucose esters, analysis in d4ꢀ
methanol may have influenced the ratio of esters present.
Experimentally, the extent of migration differed between the two glucose esters with greater
amounts of 2ꢀO-αꢀ and 6ꢀO-α
-esters observed in the 1H NMR spectrum of pꢀcoumaroyl glucose
than for feruloyl glucose, which is not evident based on the thermodynamic calculations. While the
exact route to migration is not understood, we have mechanistically investigated the plausible
migrations away from the 1ꢀO-βꢀesters to explain the relative extents of migrations observed
between the two esters. During synthesis, the migration was thought to be facilitated by silica gel,
and it was also reversed in wineꢀlike conditions. Therefore, the mechanism applied here (as
described by Horrobin33) is that which would occur in aqueous acidic environments, and reflect the
possibilities in wine. The key intermediates employed in calculation of the 1ꢀO-
coumaroyl glucose migration are shown in Figure 6 (with analogous intermediates used for
subsequent migrations) along with relative energies. Although the migrated mixtures contained
anomers, 6ꢀO- ꢀ and 2ꢀO- ꢀesters, migration away from the 1ꢀO- ꢀester was proposed to have
occurred before anomerization due to the mechanistic challenges faced with direct anomerization of
βꢀ to 2ꢀO-βꢀpꢀ
αꢀ
α
α
β
the 1ꢀO-
β
ꢀ to 1ꢀ O-
α
ꢀesters, with previous evidence supporting this.51
ACS Paragon Plus Environment

Page 15 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 15
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
For both the 1ꢀOꢀ to 2ꢀOꢀ and 1ꢀOꢀ to 6ꢀOꢀmigrations were less favored, as energy barriers
were increased, in an aqueous environment for both glucose esters when compared with
dichloromethane. This supports the rapid migration seen in 10 % methanol/dichloromethane on
silica gel (Figure 6). Not only are the 2ꢀO-αꢀ and 6ꢀO-α-esters thermodynamically favored in
dichloromethane (Figure 5), but migration directly to these positions is likely to happen more
rapidly also. For a 1ꢀOꢀ to 2ꢀOꢀmigration, the difference in energy of intermediate 2 in
dichloromethane between feruloyl and pꢀcoumaroyl (with an energy barrier between intermediates
1 and 2 of 30 kJ/mol for the pꢀcoumaroyl and 65 kJ/mol for the feruloyl) describes a kinetically
more favorable migration for the pꢀcoumaroyl glucose. If the mixtures had not yet reached
equilibrium and were not yet representing the thermodynamic products, the energy barriers for
migration directly to the 2ꢀOꢀposition helps to explain why a greater extent of migration was
observed for pꢀcoumaroyl glucose (11) than in feruloyl glucose (12) at the time of characterization.
In studying direct migrations to both the 6ꢀOꢀ and 3ꢀOꢀpositions a ringꢀflipped glucose was
produced, giving 1ꢀOH, 3ꢀOH and 6ꢀOH on the same side of the ring. As such, the bicyclic
intermediate 2 in the 1ꢀOꢀ to 3ꢀOꢀmigration is stabilized by the ringꢀflip and has a similar, or lower
energy than intermediate 1 for both esters in all conditions investigated (Figure 6). Additionally, the
1ꢀOꢀ to 3ꢀOꢀmigrations involve a favorable 6ꢀmembered cyclic intermediate.34, 54Although the ringꢀ
flipped glucose allows for facile migration to the 3ꢀOꢀposition, subsequent migration to the
thermodynamically favored 6ꢀOꢀposition is also facilitated,55 and most likely why the 3ꢀOꢀesters
were not prevalent in the migrated ester mixtures.
However, these results do indicate that in largely aqueous conditions that the 1ꢀO-β-esters
will be thermodynamically favored, and that most migrations will occur more slowly due to higher
energy barriers to migration, with the only exception being the migration to the 3ꢀOꢀposition which
requires a ring flip of the glucose ring before the migration can take place.
In synthesis or isolation of the hydroxycinnamoyl glucose esters, the use of solvents other
than water should be limited, especially under acidic conditions conducive to migration. If organic
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 16 of 37
Hydroxycinnamoyl Glucose Esters
Page 16
355
356
solvents are employed, they should be done so under neutral conditions, or in the presence of a
buffer. In the event of migration away from the desired 1ꢀO-
β
-esters, it was found that for storage
-esters, with the spectral
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
under aqueous acidic conditions, migrated mixtures will revert to the 1ꢀO-
β
data of each ester corresponding to literature data, as reported in the experimental section.15, 56
(extracted ion chromatograms of glucose esters stored in pH = 3.5 model wine and mass spectra of
identified peaks can be found in supporting information).
HPLC-MS analysis of glucose esters in wine. The ease of migration of the glucose esters
in nonꢀaqueous solvents brings into question whether multiple esters observed in previous
quantifications7, 14, 15 are an artefact of extraction using nonꢀaqueous solvents promoting migration
and multiple positional isomers, rather than being present in grape or wine products.
When the synthetically prepared glucose esters were stored in model wine, subsequent
analysis by HPLCꢀMS gave extracted ion chromatograms that showed largely single esters. This
analysis, without a solvent extraction or purification step, suggests that migration will not be greatly
facilitated by the method of analysis solely.
Unconcentrated red and white wine along with concentrated samples (5 times concentrated
under reduced pressure) were submitted to analysis by HPLCꢀMS through direct injection rather
than extraction by organic solvent, to avoid solvent induced migration. While pꢀcoumaroyl glucose
and feruloyl glucose could be identified in concentrated white wine (extracted ion chromatograms
and mass spectra in supporting information), the HPLCꢀMS analysis of red wine failed to identify
feruloyl glucose due to insensitivity caused by coꢀeluting species (tentatively identified as pꢀ
coumaroyl anthocyanin derivatives). This led to decreased resolution of the m/z 355 extracted ion
chromatogram and no fragmentations were found matching that of the reference sample, which may
explain why feruloyl glucose has yet to be identified in red wine. Increased sensitivity was achieved
using HPLCꢀMRM with the predominant fragmentations, those from the precursor ions (m/z 325 or
355) to the aglycone (m/z 163 or 193) and to the aglycone minus water (m/z 145 or 175), displayed
in Figure 7.
ACS Paragon Plus Environment

Page 17 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Multiple pꢀcoumaroyl glucose esters were observed in unconcentrated white wine,
Page 17
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
concentrated white wine and concentrated red wine (29.7 and 30.6 mins, Figures 7b, c and d, left
side). The presence of a second feruloyl glucose ester was not obvious in unconcentrated white
wine (Figure 7b), but slightly better resolution in concentrated white wine facilitated identification
of a second feruloyl glucose peak (37.7 mins, Figure 7c). As near identical chromatograms were
observed for unconcentrated and concentrated white wine, the concentration step was not
considered to alter the compounds present, only providing improved resolution. As such,
unconcentrated red wine was not analyzed, instead only concentrated red wine was used to obtain
that improved sensitivity while providing information that was representative of an unconcentrated
sample. Although the concentration of feruloyl glucose in concentrated red wine appears to be
somewhat lower than in white wine, evidence of a second peak in the concentrated red wine sample
was seen (37.6 mins, Figure 7d). These data confirm the previous findings that there are multiple
glucose esters in wine,3, 14, 16 but the theoretical studies show that the extraction method can
contribute to the extent of migration. These results also suggest why the feruloyl glucose ester was
not observed in previous studies, as the presence of what are likely to be pꢀcoumaroyl anthocyanin
derivatives in red wine coꢀelute, and prevent identification unless HPLCꢀMRM is used.
Metabolism of p-coumaroyl and feruloyl glucose by Brettanomyces yeast. The synthetic
samples of the glucose esters were spiked as migrated mixtures containing mainly the 1ꢀO-
with the knowledge that in wineꢀlike environments the 1ꢀO- ꢀesters will prevail, or will at least
reside in a ratio that is representative of what occurs in wine. As acyl migration is a dynamic
β-esters,
399
400
401
402
403
404
405
406
β
process, isolating single positional isomers and submitting to fermentation experiments would
simply result in a mixture governed by the stability of each positional isomer in a given medium.
Regardless, during fermentation, to release the free hydroxycinnamic acid the esterase of
Brettanomyces must be active towards a hydroxycinnamoyl glucose ester.
The evolution of ethylphenols during fermentation (Figure 8) with AWRI 1499 gave a
similar percentage conversion for 11 and 12 of approximately 36%. The uninoculated control
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 18 of 37
Hydroxycinnamoyl Glucose Esters
Page 18
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
ferments showed no evolution of ethylphenols, and the stability of the glucose esters has been
shown by Monagas14 and was proven by storage in fermentationꢀlike conditions for the length of
the experiment, so it can be confirmed that the production of ethylphenols was caused by enzymatic
cleavage of the glucose esters. The intermediate hydroxycinnamic acids (3 and 4) were not
quantified in this work as, in our experience, once the acid has been liberated the conversion to the
ethylphenol is good,28 and recovery of both the ester and ethylphenol account for a significant
proportion of the spiked material.2
This experiment shows for the first time that the hydroxycinnamoyl glucose esters can be
metabolized by B. bruxellensis and can contribute to the formation of ethylphenols in wine. With
moderate conversions from the glucose esters, the spoilage potential during barrel ageing will
largely be determined by the concentration present in the wine, which as described previously can
range from trace to several mg/L.3, 14, 16 Assuming a preꢀmalolactic fermentation concentration of pꢀ
coumaroyl glucose, as observed by Hernandez et al., of 2.74 mg/L,3 a 36% molar conversion would
result in a 4ꢀethylphenol concentration of around 370 ꢀg/L. This is more than half the detection
threshold,1 suggesting hydroxycinnamoyl glucose esters should be considered when assessing the
potential for ethylphenol accumulation.
Metabolism of p-coumaroyl and feruloyl tartrate by Brettanomyces yeast. The
hydroxycinnamoyl tartrate esters (9 and 10), when fermented with AWRI 1499, AWRI 1608 and
AWRI 1613 (the three strains tested belonging to the three main genetic groups and being
representative of 98% of Australian Brettanomyces isolates57) gave no conversion to 4ꢀethylphenol
or 4ꢀethylguaiacol, and the result was confirmed in repeat experiments. This is consistent with that
detailed by Schopp et al. in red wine ferments with B. bruxellensis.11
The potential for the tartrate esters to contribute to spoilage during barrel ageing is limited,
with B. bruxellensis lacking the enzymatic ability to release hydroxycinnamic acids and
subsequently form ethylphenols. However, previous studies showed that the hydroxycinnamoyl
tartrates could be hydrolyzed using commercial enzyme preparations.8, 9 The resulting
ACS Paragon Plus Environment

Page 19 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 19
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
hydroxycinnamic acids would be metabolized to yield ethylphenols in the presence of B.
bruxellensis, and so the tartrate esters might act as indirect precursors to ethylphenols, if first
exposed to specific enological conditions, regardless of the ability of B. bruxellensis to directly
metabolize them.
While the hydroxycinnamoyl tartrates were not metabolized, this work shows that the
glucose esters can contribute to the accumulation of ethylphenols, adding to that already known for
the free acids and the ethyl esters,1, 2 and expanding the pool of ethylphenol precursors.
ABBREVIATIONS USED
HPLC, high performance liquid chromatography; MS, mass spectrometry; AWRI,
Australian Wine Research Institute; NMR, nuclear magnetic resonance; FTIR, Fourier
transform infra-red; HRMS. High resolution mass spectrometry; DFT, density functional
theory; MMFF, Merck molecular force field; HOMO, highest occupied molecular orbital;
LUMO. Lowest unoccupied molecular orbital; PDA, photodiode array; S0, singlet ground
state; T1, first excited triplet state; MRM, multiple reaction monitoring.
449
450
ACKNOWLEDGEMENT
The School of Agriculture, Food and Wine, the University of Adelaide and The Australian
451
452
453
454
455
456
457
458
459
Wine Research Institute, members of the Wine Innovation Cluster in Adelaide, are supported by
Australian grape growers and winemakers through their investment body, Wine Australia, with
matching funds from the Australian Government.
ASSOCIATED CONTENT
Supporting information. synthetic methodologies and characterization data for preparation
of hydroxycinnamoyl esters; computational methodologies and absolute values generated in
computational studies of hydroxycinnamoyl isomerization and glucose ester migration;
chromatographic data from HPLCꢀMS studies of glucose esters; absolute values of ethylphenol
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 20 of 37
Hydroxycinnamoyl Glucose Esters
Page 20
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
evolution from hydroxycinnamoyl glucose esters. This material is available free of charge via the
internet at http://pubs.acs.org.
REFERENCES
1.
wines. J. Sci. Food Agric. 1992, 60, 165ꢀ178.
2. Hixson, J. L.; Sleep, N.; Capone, D. L.; Elsey, G. M.; Curtin, C.; Sefton, M. A.; Taylor, D.
Chatonnet, P.; Dubourdieu, D.; Boidron, J. N.; Pons, M., The origin of ethylphenols in
K., Hydroxycinnamic Acid Ethyl Esters as Precursors to Ethylphenols in Wine. J. Agric. Food
Chem. 2012, 60, 2293ꢀ2298.
3.
Hernandez, T.; Estrella, I.; Carlavilla, D.; MartinꢀAlvarez, P. J.; MorenoꢀArribas, M. V.,
Phenolic compounds in red wine subjected to industrial malolactic fermentation and ageing on lees.
Anal. Chim. Acta 2006, 563, 116ꢀ125.
4.
Nagel, C. W.; Wulf, L. W., Changes in the anthocyanins, flavanoids and hydroxycinnamic
acid esters during fermentation and aging of Merlot and Cabernet Savignon. Am. J. Enol. Vitic.
1979, 30, 111ꢀ116.
5.
Somers, T. C.; Verette, E.; Pocock, K. F., Hydroxycinnamate esters of Vitis vinifera ꢀ
Changes during white vinification, and effects of exogenous enzymaticꢀhydrolysis. J. Sci. Food
Agric. 1987, 40, 67ꢀ78.
6.
Monagas, M.; GomezꢀCordoves, C.; Bartolome, B., Evaluation of different Saccharomyces
cerevisiae strains for red winemaking. Influence on the anthocyanin, pyranoanthocyanin and nonꢀ
anthocyanin phenolic content and colour characteristics of wines. Food Chem. 2007, 104, 814ꢀ823.
7.
Hernandez, T.; Estrella, I.; PerezꢀGordo, M.; Alegria, E. G.; Tenorio, C.; RuizꢀLarrrea, F.;
MorenoꢀArribas, M. V., Contribution of malolactic fermentation by Oenococcus oeni and
Lactobacillus plantarum to the changes in the nonanthocyanin polyphenolic composition of red
wine. J. Agric. Food Chem. 2007, 55, 5260ꢀ5266.
ACS Paragon Plus Environment

Page 21 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 21
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
8.
Dugelay, I.; Gunata, Z.; Sapis, J. C.; Baumes, R.; Bayonove, C., Role of cinnamoyl esteraseꢀ
activities from enzyme preparations on the formation of volatile phenols during winemaking. J.
Agric. Food Chem. 1993, 41, 2092ꢀ2096.
9.
on the content of volatile phenols in Pinot noir wines. Am. J. Enol. Vitic. 2002, 53, 131ꢀ137.
10. Silva, L. R.; Andrade, P. B.; Valentao, P.; Seabra, R. M.; Trujillo, M. E.; Velazquez, E.,
Gerbaux, V.; Vincent, B.; Bertrand, A., Influence of maceration temperature and enzymes
Analysis of nonꢀcoloured phenolics in red wine: Effect of Dekkera bruxellensis yeast. Food Chem.
2005, 89, 185ꢀ189.
11.
Schopp, L. M.; Lee, J.; Osborne, J. P.; Chescheir, S. C.; Edwards, C. G., Metabolism of
nonesterified and esterified hydroxycinnamic acids in red wines by Brettanomyces bruxellensis. J.
Agric. Food Chem. 2013, 61, 11610ꢀ11617.
12.
Reschke, A.; Herrmann, K., Occurrence of 1ꢀOꢀhydroxycinnamylꢀβꢀDꢀglucoses in fruits.
15. Phenolics of fruits. Zeitschrift für Lebensmitteluntersuchung und -Forschung A 1981, 173, 458ꢀ
463.
13.
Baderschneider, B.; Winterhalter, P., Isolation and characterization of novel benzoates,
cinnamates, flavonoids, and lignans from Riesling wine and screening for antioxidant activity. J.
Agric. Food Chem. 2001, 49, 2788ꢀ2798.
14.
from Vitis vinifera L. during aging in the bottle. Eur. Food Res. Technol. 2005, 220, 331ꢀ340.
15. Monagas, M.; Suarez, R.; GomezꢀCordoves, C.; Bartolome, B., Simultaneous determination
Monagas, M.; Bartolomé, B.; GómezꢀCordovés, C., Evolution of polyphenols in red wines
of nonanthocyanin phenolic compounds in red wines by HPLCꢀDAD/ESIꢀMS. Am. J. Enol. Vitic.
2005, 56, 139ꢀ147.
16.
PerezꢀMagarino, S.; OrtegaꢀHeras, M.; CanoꢀMozo, E., Optimization of a solidꢀphase
extraction method using copolymer sorbents for isolation of phenolic compounds in red wines and
quantification by HPLC. J. Agric. Food Chem. 2008, 56, 11560ꢀ11570.
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 22 of 37
Hydroxycinnamoyl Glucose Esters
Page 22
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
17.
Boido, E.; GarcíaꢀMarino, M.; Dellacassa, E.; Carrau, F.; RivasꢀGonzalo, J. C.; Escribanoꢀ
Bailón, M. T., Characterisation and evolution of grape polyphenol profiles of Vitis vinifera L. cv.
Tannat during ripening and vinification. Australian Journal of Grape and Wine Research 2011, 17,
383ꢀ393.
18.
solutions at different pH. Phytochemistry 1967, 6, 755ꢀ&.
19. Kahnt, G., Uber das gleichgewicht zwischen den stereoisomeren einiger zimtꢀsaurederivate
Kahnt, G., trans-cisꢀEquilibrium of hydroxycinnamic acids during irradiation of aqueous
in abhangigkeit von der molaren konzentration und ihre quantitative spektrophotometrische
messung bei pflanzenanalysen. Biologisches Zentralblatt 1966, 85, 545ꢀ554.
20.
Hartley, R. D.; Jones, E. C., Effect of ultravioletꢀlight on substituted cinnamic acids and
estimation of their cis and trans isomers by gasꢀchromatography. Journal of Chromatography 1975,
107, 213ꢀ218.
21.
derivatives. Journal of Chromatography 1978, 152, 517ꢀ522.
22. Li, Q. S.; Fang, W. H., Ab initio study on the structures and properties of trans-pꢀcoumaric
acid in lowꢀlying electronic states. Chem. Phys. 2005, 313, 71ꢀ75.
23. Yoda, M.; Houjou, H.; Inoue, Y.; Sakurai, M., Spectral tuning of photoactive yellow
Fenton, T. W.; Mueller, M. M.; Clandinin, D. R., Isomerization of some cinnamic acidꢀ
protein. Theoretical and experimental analysis of medium effects on the absorption spectrum of the
chromophore. J. Phys. Chem. B 2001, 105, 9887ꢀ9895.
24.
yellow protein: A quantumꢀchemical analysis. Proc. Natl. Acad. Sci. U. S. A. 2001, 98, 4299ꢀ4304.
25. Sergi, A.; Gruning, M.; Ferrario, M.; Buda, F., Density Functional study of the photoactive
yellow protein's chromophore. J. Phys. Chem. B 2001, 105, 4386ꢀ4391.
26. Ko, C.; Levine, B.; Toniolo, A.; Manohar, L.; Olsen, S.; Werner, H. J.; Martinez, T. J., Ab
Molina, V.; Merchan, M., On the absorbance changes in the photocycle of the photoactive
initio excitedꢀstate dynamics of the photoactive yellow protein chromophore. J. Am. Chem. Soc.
2003, 125, 12710ꢀ12711.
ACS Paragon Plus Environment

Page 23 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 23
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
27.
Yamada, A.; Yamamoto, S.; Yamato, T.; Kakitani, T., Ab initio MO study on potential
energy surfaces for twisting around C7=C8 and C4ꢀC7 bonds of coumaric acid. Theochem-J. Mol.
Struct. 2001, 536, 195ꢀ201.
28.
Hixson, J. L.; Curtin, C. D.; Sefton, M. A.; Taylor, D. K., Stereospecificity of D.
bruxellensis in the production of ethylphenol offꢀflavour in wine. In Proceedings from XIII
Weurman Flavour Research Symposium, Ferreira, V.; Lopez, R., Eds. Academic Press: 2013; pp
pp. 183ꢀ187.
29.
acetylꢀDꢀglucopyranoses. J. Org. Chem. 1959, 24, 1388ꢀ1390.
30. Wood, H. B.; Fletcher, H. G., Migration of a mesitoyl group from C1 to C2 in alphaꢀDꢀ
glucopyranose ꢀ derivatives of 2ꢀOꢀmesitoylꢀDꢀglucose. J. Am. Chem. Soc. 1956, 78, 2849ꢀ2851.
31. Molinier, V.; Wisnienwski, K.; Bouchu, A.; Fitremann, J.; Queneau, Y., Transesterification
Bonner, W. A., C1ꢀC2 acetyl migration on methylation of the anomeric 1,3,4,6ꢀtetraꢀOꢀ
of sucrose in organic medium: Study of acyl group migrations. J. Carbohydr. Chem. 2003, 22, 657ꢀ
669.
32.
from a sulfur to an oxygen atom. Carbohydr. Res. 1973, 31, 237ꢀ243.
33. Horrobin, T.; Tran, C. H.; Crout, D., Esteraseꢀcatalysed regioselective 6ꢀdeacylation of
Whistler, R. L.; Anisuzza.Ak; Kim, J. C., Silica gel catalyzed migration of acetyl groups
hexopyranose perꢀacetates, acidꢀcatalysed rearrangement to the 4ꢀdeprotected products and
conversions of these into hexose 4ꢀ and 6ꢀsulfates. J. Chem. Soc.-Perkin Trans. 1 1998, 1069ꢀ1080.
34.
migration of methyl monoꢀOꢀmyristoylꢀα and βꢀ
glucopyranoses. Chem. Pharm. Bull. 1983, 31, 4324ꢀ4334.
35. Roslund, M. U.; Aitio, O.; Warna, J.; Maaheimo, H.; Murzin, D. Y.; Leino, R., Acyl group
migration and cleavage in selectively protected ꢀgalactopyranosides as studied by NMR
Yoshimoto, K.; Tsuda, Y., General path of Oꢀacyl migration in
Dꢀglucose derivatives ꢀ Acyl
D
ꢀglucopyranosides and monoꢀOꢀmyristoylꢀ
Dꢀ
βꢀD
spectroscopy and kinetic calculations. J. Am. Chem. Soc. 2008, 130, 8769ꢀ8772.
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 24 of 37
Hydroxycinnamoyl Glucose Esters
Page 24
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
36.
Iddon, L.; Richards, S. E.; Johnson, C. H.; Harding, J. R.; Wilson, I. D.; Nicholson, J. K.;
Lindon, J. C.; Stachulski, A. V., Synthesis of a series of phenylacetic acid 1ꢀβꢀOꢀacyl glucosides
and comparison of their acyl migration and hydrolysis kinetics with the corresponding acyl
glucuronides. Org. Biomol. Chem. 2011, 9, 926ꢀ934.
37.
glycopyranosyl donors. J. Carbohydr. Chem. 1993, 12, 357ꢀ370.
38. Barros, J. C.; da Silva, J. M.; Calazans, A. R.; Tanuri, A.; Brindeiro, R. D. M.; Williamson,
Ziegler, T.; Pantkowski, G., Preparation of 1ꢀOꢀacylꢀDꢀglycopyranoses via chloroacetylated
J. S.; Antunes, O. A. C., Synthesis of pseudopeptides derived from (R,R)ꢀtartaric acid as potential
inhibitors of HIVꢀprotease. Lett. Org. Chem. 2006, 3, 882ꢀ886.
39.
Synoradzki, L.; Ruskowski, P.; Bernas, U., Tartaric acid and its Oꢀacyl derivatives. Part 1.
Synthesis of tartaric acid and Oꢀacyl tartaric acids and anhydrides. Org. Prep. Proced. Int. 2005, 37,
37ꢀ63.
40.
for secondary interactions. Organometallics 1996, 15, 483ꢀ485.
41. Wright, S. W.; Hageman, D. L.; Wright, A. S.; McClure, L. D., Convenient preparations of
tꢀbutyl esters and ethers from tꢀbutanol. Tetrahedron Lett. 1997, 38, 7345ꢀ7348.
42. Uray, G.; Lindner, W., tertꢀButyl esters and ethers of (R,R)ꢀtartaric acid. Tetrahedron 1988,
44, 4357ꢀ4362.
MacFarland, D. K.; Landis, C. R., Synthesis and characterization of novel ligands designed
43.
Zhao, H.; Burke, T. R., Facile syntheses of (2R,3R)ꢀ(ꢀ)ꢀ and (2S,3S)ꢀ(+)ꢀchicoric acids.
Synth. Commun. 1998, 28, 737ꢀ740.
44.
aus Weisswein. Ph.D. Thesis, Universitat Braunschweig, Germany, 2000.
45. Opitz, S.; Schnitzler, J. P.; Hause, B.; Schneider, B., Histochemical analysis of
Baderschneider, B. Isolierung und Strukturaufklarung antioxidativ wirksamer Verbindungen
phenylphenalenoneꢀrelated compounds in Xiphidium caeruleum (Haemodoraceae). Planta 2003,
216, 881ꢀ889.
ACS Paragon Plus Environment

Page 25 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Zhu, Y. M.; Ralph, J., Stereoselective synthesis of 1ꢀOꢀ ꢀferuloyl and 1ꢀOꢀβꢀsinapoyl
Page 25
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
46.
β
glucopyranoses. Tetrahedron Lett. 2011, 52, 3729ꢀ3731.
47. Galland, S.; Mora, N.; AbertꢀVian, M.; Rakotomanomana, N.; Dangles, O., Chemical
synthesis of hydroxycinnamic acid glucosides and evaluation of their ability to stabilize natural
Colors via anthocyanin copigmentation. J. Agric. Food Chem. 2007, 55, 7573ꢀ7579.
48.
Salum, M. a. L.; Robles, C. J.; ErraꢀBalsells, R., Photoisomerization of ionic liquid
ammonium cinnamates: Oneꢀpot synthesis−isolation of Zꢀcinnamic acids. Org. Lett. 2010, 12,
4808ꢀ4811.
49.
Kort, R.; Vonk, H.; Xu, X.; Hoff, W. D.; Crielaard, W.; Hellingwerf, K. J., Evidence for
trans-cis isomerization of the pꢀcoumaric acid chromophore as the photochemical basis of the
photocycle of photoactive yellow protein. FEBS Lett. 1996, 382, 73ꢀ78.
50.
Machida, K.; Kikuchi, M., Norisoprenoids from Viburnum dilatatum. Phytochemistry 1996,
41, 1333ꢀ1336.
51.
Brecker, L.; Mahut, M.; Schwarz, A.; Nidetzky, B., In situ proton NMR study of acetyl and
formyl group migration in monoꢀOꢀacyl Dꢀglucose. Magnetic Resonance in Chemistry 2009, 47,
328ꢀ332.
52.
ethylguaiacol in red wine. J. Chromatogr. A 2000, 874, 101ꢀ109.
53. Lowry, T. H.; Richardson, K. S., Mechanism and Theory in Organic Chemistry. 3rd ed.;
Harper and Row: New York, 1987; p 1090.
54. Rowell, R. M., Acil migrations in the synthesis of ethyl 4ꢀOꢀmethylꢀβꢀ
Carbohydr. Res. 1972, 23, 417ꢀ424.
Pollnitz, A. P.; Pardon, K. H.; Sefton, M. A., Quantitative analysis of 4ꢀethylphenol and 4ꢀ
Dꢀglucopyranoside.
55.
Yoshimoto, K.; Tsuda, Y., On the possibility of direct Oꢀ1ꢀ
ꢀglucose derivatives. Chem. Pharm. Bull. 1983, 31, 4335ꢀ4340.
Lunkenbein, S.; Bellido, M.; Aharoni, A.; Salentijn, E. M. J.; Kaldenhoff, R.; Coiner, H. A.;
βꢀ to ꢀ6 acyl migration in 1ꢀOꢀ
acylꢀb-
D
56.
MunozꢀBlanco, J.; Schwab, W., Cinnamate metabolism in ripening fruit. Characterization of a
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 26 of 37
Hydroxycinnamoyl Glucose Esters
Page 26
612
613
614
615
616
617
UDPꢀglucose: Cinnamate glucosyltransferase from strawberry. Plant Physiol. 2006, 140, 1047ꢀ
1058.
57.
Curtin, C. D.; Bellon, J. R.; Henschke, P. A.; Godden, P. W.; de Barros Lopes, M. A.,
Genetic diversity of Dekkera bruxellensis yeasts isolated from Australian wineries. Federation of
European Microbiological Societies Yeast Research 2007, 7, 471ꢀ481.
ACS Paragon Plus Environment

Page 27 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 27
FIGURE CAPTIONS
Figure 1. Presumed pathway for the breakdown of ethyl hydroxycinnamates by B. bruxellensis.
Figure 2. Preparation of hydroxycinnamoyl tartrate esters. i) acetyl chloride, reflux ii) water,
acetone iii) MgSO4, H2SO4, tertꢀBuOH iv) KOH, MeOH v) SOCl2, reflux vi) 18, pyridine vii)
trifluoroacetic acid, C18ꢀRP chromatography.
Figure 3. Preparation and transformations of hydroxycinnamoyl glucose esters. i) chloroacetyl
chloride, pyridine, reflux ii) hydrazine acetate iii) trichloroacetonitrile, DBU iv) 25, TMSOTf,
molecular sieves v) pyridine/water vi) XADꢀ8, MeOH/water vii) silica gel, 10% MeOH/CH2Cl2.
Figure 4. Energy profiles for isomerization of hydroxycinmmates a) pꢀcoumaric acid, and b) pꢀ
coumaroyl glucose, calculated using DFT B3LYP 6ꢀ31G* basis set in a vacuum. Excitation
energies for pꢀcoumaric acid and pꢀcoumaroyl glucose in solvents of differing polarities, c) vertical
excitation energy of optimized S0 configuration to T1 state, and d) HOMOꢀLUMO gap energy of S0
state. Absolute values can be found in the supporting information.
Figure 5. Energies of hydroxycinnamoyl glucose esters, relative to 1ꢀO-βꢀester, calculated using
DFT B3LYP 6ꢀ31G* basis set in water, dichloromethane, ethanol and toluene. Absolute values can
be found in the supporting information.
Figure 6. Key intermediates in acidꢀcatalyzed 1ꢀO-βꢀ to 2ꢀO-β acyl migration for pꢀcoumaroyl
glucose, and energies of key intermediates in acyl migrations, relative to Intermediate 1 determined
using the DFT B3LYP 6ꢀ31G* basis set. Absolute values can be found in the supporting
information.
Figure 7. HPLCꢀMRM chromatograms (aglycone ꢀ blue, aglycone minus water ꢀ red) of
hydroxycinnamoyl glucose esters. a) Pure glucose esters. b) Unconcentrated white wine. c)
Concentrated white wine. d) Concentrated red wine.
Figure 8. Percentage molar conversion of glucose esters to volatile phenols in model ferments, data
expressed as average of triplicates ± standard deviation. Absolute values can be found in the
supporting information.
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 28 of 37
Hydroxycinnamoyl Glucose Esters
Page 28
TABLES
Table 1. Numbering of key atoms in hydroxycinnamate skeleton and the structural features
obtained from B3LYP 6ꢀ31G* optimized equilibrium geometries.
Hydroxycinnamate
skeleton
C7=C8
O1
O3
O1/O3
HOMO-LUMO gap
(kJ/mol)
R
H
anion
anion
R'
H
H
length (Å) charge charge charge ratio
1.351
1.346
1.354
ꢀ0.573
ꢀ0.589
ꢀ0.774
ꢀ0.597
ꢀ0.764
ꢀ0.777
0.960
0.771
0.996
400.95
437.46
372.07
anion
CH2CH3
H
1.354
ꢀ0.58
ꢀ0.365
1.589
400.47
CH2CH3
Glucose
Glucose
anion
H
anion
1.369
1.355
1.373
ꢀ0.739
ꢀ0.573
ꢀ0.728
ꢀ0.369
ꢀ0.315
ꢀ0.317
2.003
1.819
2.297
329.87
385.99
318.8
H
C(O)CH3
C(O)CH3
1.349
1.345
1.348
1.344
ꢀ0.476
ꢀ0.483
ꢀ0.469
ꢀ0.473
ꢀ0.562
ꢀ0.754
ꢀ0.561
ꢀ0.755
0.847
0.641
0.836
0.626
434.23
449.43
437.19
445.27
anion
H
C(O)CH2Cl
C(O)CH2Cl
anion
ACS Paragon Plus Environment

Page 29 of 37
Journal of Agricultural and Food Chemistry
Hydroxycinnamoyl Glucose Esters
Page 29
FIGURE GRAPHICS
Figure 1
ACS Paragon Plus Environment