Stereoselective Formation of β-O-4 Structures Mimicking Softwood Lignin Biosynthesis: Effects of Solvent and the Structures of Quinone Methide Lignin Models

Article abstract of DOI:10.1021/acs.jafc.9b01968

p-Quinone methide (QM) is formed as an intermediate during lignin biosynthesis. The aromatization of the QM by the attack of a nucleophile at the α-position of its side chain generates a phenolic hydroxy group in a growing polymer and creates stereoisomeric forms in the side chain. A series of β-O-4-aryl ether QMs was reacted with water at 25 °C to replicate the formation of p-hydroxyphenyl (H) and guaiacyl (G) β-O-4 structures in plant cell walls. Water addition occurred in 3-methoxy-substituted QMs (G-type QMs) with half-lives (t_1/2) between 13 and 15 min, at pH 7, in 50% water solution (dioxane-water, 1:1). The rate increased as the water concentration increased to 99% (t_1/2, 1.2-1.4 min). Similar solvent effects were observed for more reactive nonsubstituted QMs (H-type QMs with t_1/2 of <1 min). Consequently, t_1/2 of the H-type QMs was shorter than that of the G-type QMs under every solvent condition. Upon increasing the water concentration, the variation in the erythro/threo ratios of the four dimeric β-O-4 products increased. Interestingly, the effect of pH on the stereopreference, which was observed in 50% water solution, was small and became imperceptible as the water concentration increased to 99%, suggesting that the effect of the solvent, as well as the effect of the pH, plays an important role in understanding the reaction conditions in cell walls during lignin biosynthesis. The threo isomer was preferentially formed in the four dimeric β-O-4 structures, which is inconsistent with the structural features of compression wood lignin rich in H-units. However, the erythro-selective formation was attained in an H-type QM at every pH studied (pH 3.5-7) by introducing a biphenyl structure into the β-etherified ring moiety.

Full text of DOI:10.1021/acs.jafc.9b01968

Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC
Agricultural and Environmental Chemistry
Stereoselective formation of #–O–4 structures mimicking
softwood lignin biosynthesis: Effects of solvent,
and the structures of quinone methide lignin models
Xuhai Zhu, Takuya Akiyama, Tomoya Yokoyama, and Yuji Matsumoto
J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.9b01968 • Publication Date (Web): 31 May 2019
Downloaded from http://pubs.acs.org on June 4, 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 34
Journal of Agricultural and Food Chemistry
1
2
3
4
Stereoselective formation of β–O–4 structures
mimicking softwood lignin biosynthesis: Effects of
solvent, and the structures of quinone methide
lignin models
5
Zhu Xuhai,† Takuya Akiyama,*, ‡ Tomoya Yokoyama† and Yuji Matsumoto†
6
7
†Wood Chemistry Laboratory, Department of Biomaterial Sciences, Graduate School of
Agricultural and Life Sciences, The University of Tokyo, Bunkyo-ku, Tokyo 113-8657, Japan
8
9
‡Wood Chemistry Laboratory, Department of Forest Resource Chemistry, Forestry and Forest
Products Research Institute, 1 Matsunosato, Tsukuba, Ibaraki 305-8687, Japan
10 KEYWORDS: Lignification, para-quinone methide, radical coupling, erythro and threo,
11 gymnosperm, p-hydroxyphenyl, guaiacyl
12 ABSTRACT
13 p-Quinone methide (QM) is formed as an intermediate during lignin biosynthesis. The
14 aromatization of the QM by the attack of a nucleophile at the α-position of its side-chain
15 generates a phenolic hydroxy group in a growing polymer and creates stereoisomeric forms in
16 the side-chain. A series of β–O–4-aryl ether QMs was reacted with water at 25 °C to replicate
17 the formation of p-hydroxyphenyl (H) and guaiacyl (G) β–O–4 structures in plant cell walls.
18 Water addition occurred in 3-methoxy-substituted QMs (G-type QMs) with half-lives (t1/2)
19 between 13 and 15 min, at pH 7, in 50% water solution (dioxane-water, 1:1). The rate
20 increased as the water concentration increased to 99% (t1/2, 1.2–1.4 min). Similar solvent
21 effects were observed for more reactive nonsubstituted QMs (H-type QMs with t1/2 of <1 min).
22 Consequently, t1/2 of the H-type QMs was shorter than that of the G-type QMs under every
23 solvent condition. Upon increasing the water concentration, the variation in the erythro/threo
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 2 of 34
24 ratios of the four dimeric β–O–4 products increased. Interestingly, the effect of pH on the
25 stereo-preference, which was observed in 50%-water solution, was small and became
26 imperceptible as the water concentration increased to 99%, suggesting that the effect of the
27 solvent as well as the effect of the pH, play an important role in understanding the reaction
28 condition in cell walls during lignin biosynthesis. The threo isomer was preferably formed in
29 the four dimeric β–O–4 structures, which is inconsistent with the structural features of
30 compression wood lignin rich in H-units. However, the erythro-selective formation was
31 attained in an H-type QM at every pH studied (pH 3.5–7) by introducing a biphenyl structure
32 into the β-etherified ring moiety.
33
34 INTRODUCTION
35 Lignins are aromatic polymers that are deposited in vascular-plant cell walls, providing
36 rigidity and structural reinforcement to the cell walls. They are primarily derived from three
37 monolignols, p-coumaryl, coniferyl, and sinapyl alcohols that give rise to p-hydroxyphenyl
38 (H), guaiacyl (G), and syringyl (S) units, respectively, in the lignin structures.1-3 Lignins differ
39 greatly in their chemical structures, which is largely due to the compositions of the
40 monolignols, and their polymerization mechanisms during lignin biosynthesis.
41
p-Quinone methides (QMs) are formed as transient intermediates during lignin
42 biosynthesis. They are generated by a β–O–4, β–β, β–5, or β–1 coupling between two
43 radicals arising from monolignols or the phenolic end of the growing polymer. The QMs are
44 rearomatized by the addition of water at the exocyclic methylene-group site. The
45 aromatization, i.e. the regeneration of a phenolic hydroxy group, is essential for lignin
46 polymerization. A phenolic hydroxy group is regenerated by the aromatization to form a
47 phenolic end in the growing lignin, thereby being prepared for further oxidation into a
48 phenoxyl radical that couples with another radical to elongate the lignin chain.
49
The structural variation of QMs and their chemical properties are thought to
50 influence the polymerization efficiency and the structure of the resulting lignins.4-8 Kobayashi
51 et al.9 reported in their study of monolignols oxidizability, that coniferyl alcohol (G-type) was
52 consumed faster than p-coumaryl alcohol (H-type) in a peroxidase−H2O2 system. The
ACS Paragon Plus Environment

Page 3 of 34
Journal of Agricultural and Food Chemistry
53 oxidation potentials determined by cyclic voltammetry indicated that p-coumaryl alcohol can
54 be less oxidized than coniferyl alcohol.9 Therefore, p-coumaryl alcohol does not seem to have
55 an advantage over coniferyl alcohol in driving the lignin-polymerization cycle efficiently. In
56 contrast, the results from our previous model study8 indicated that water addition occurred
57 much faster in nonsubstituted QMs (H-type QMs) than in 3-methoxy-substituted QMs
58 (G-type QMs) when the water addition experiments were conducted in dioxane-water mixture
59 (1:1, v/v), at pH 3−7 and 25 °C. This result suggested that once a radical-coupling reaction
60 occurs, generating a β–O–4-bonded QM at the growing end of a lignin chain, the subsequent
61 aromatization proceeds faster in the H-type QMs than in the G-type QMs. Therefore, we
62 postulated that the rapid aromatization in H-type QMs might provide an advantage over
63 G-type QMs for efficiently driving the lignin-polymerization cycle, which may possibly play
64 a certain role in the rapid lignin deposition, and the development of highly lignified
65 compression wood in response to growing stress.
66
Water addition to the β–O–4-aryl ether QM is a stereo-differentiating reaction,
67 which cause an increase in the structural variety of lignins. During the formation of
68 arylglycerol-β-aryl ether structures (β-O-4 structures), a β–O–4-linkage is formed by a radical
69 coupling reaction between a monolignol and the phenolic end of a growing lignin to produce
70 a QM intermediate, which houses a chiral center: the β-carbon atom (Scheme 1).10 The β–O–
71 4-aryl ether QM bearing the β-asymmetric carbon is aromatized by the water addition at the
72 prochiral α-carbon, affording either the erythro or threo forms of the β–O–4 structure. The
73 generation of erythro or threo forms depends on which face of the QM reacts faster with
74 water.4,6,11 The erythro/threo ratio is well known to influence the lignin degradation efficiency
75 under alkaline delignification,12-15 and enzymatic and chemical oxidation.16-19 Therefore, the
76 erythro/threo ratio of β–O–4 structures have been recognized as an important structural and
77 chemical characteristic of lignins.
78
The proportion of the erythro form to the threo form of β–O–4 structures was
79 nearly 50:50 in softwood lignins mostly composed of G-units, i.e., the proportion of erythro
80 form determined by ozonation method was within the 50.0 ± 0.6% range for five
81 gymnosperm species.20 In contrast, the erythro form dominates in the β–O–4 structures in
82 hardwood lignins primarily composed of G- and S-units.20-24 A positive relationship between
83 the erythro/threo ratio and S/G ratio has been found among various hardwood species.20,25-27
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 4 of 34
84 Brunow et al.6 conducted an in vitro water addition reaction to G-type and
85 3,5-dimethoxy-substituted (S-type) QMs for replicating the formation of β–O–4 structures in
86 hardwood lignin biosynthesis; they revealed that an erythro-preferential reaction occurred for
87 the formation of arylglycerol-β-syringyl ethers at pH 3 (GS, and SS), whereas the
88 stereopreferential formation of erythro isomers was not observed for a β-guaiacyl ether (GG)
89 at the same pH.
90
Lignin structural studies of the reaction wood in softwood revealed that the
91 erythro-preferential formation occurs in the compression wood, of which lignins contain a
92 substantial amount of H-units. The β–O–4 structures in the compression wood lignin of
93 softwoods were slightly, but clearly dominantly in the erythro form.26,28 A positive correlation
94 was found between the erythro/threo ratio and H/G ratio in the compression wood lignins.26
95 This result implied that the erythro-selective formation reaction can occur under the influence
96 of the H-units in lignins.
97
Previously, as described above, we conducted water addition experiments similar to
98 those by Brunow’s group,6 for mimicking the formation of H-type β–O–4 structures in
99 compression wood lignin biosynthesis.8 The threo-preferential water addition was observed
100 both for H- and G-types QMs examined at pH 3.5–7. An erythro-preferential reaction was not
101 observed in the model experiments, whereas the erythro/threo ratio of the β–O–4 structures in
102 compression wood, which contains a substantial amount of H-units, is slightly but clearly
103 higher than 1.0.26,28
104
Herein, we investigated the gap between the model experiment and the lignin
105 structures in actual wood cell walls. Water addition experiments were conducted focusing on
106 the effect of the solvent to advance our understanding of the reaction condition that facilitates
107 erythro-preferential reactions with H-type QMs. In addition to previously reported dimeric
108 QM models, a novel trimeric QM compound bearing a biphenyl structure on the β-etherified
109 ring moiety was examined (Scheme 2). Dehydrogenation polymers (DHPs) were also
110 prepared from p-coumaryl alcohol in an in vitro peroxidase−H2O2 system under various
111 conditions. The DHPs were characterized by SEC and ozonation method29 to investigate the
112 relationship between their molecular weights and the erythro/threo ratio of the β–O–
113 4-structures.
ACS Paragon Plus Environment

Page 5 of 34
Journal of Agricultural and Food Chemistry
lignin
1
R
H,
R'
5
R
H : p-Hydroxyphenyl
H, OMe: Guaiacyl
OMe, OMe: Syringyl
4
3
HO
R'
growing lignin polymer
OH
γ
β
α
1
5
3
R''
R'''
4
OH
Monolignol
R'' R'''
H, H : p-Coumaryl alcohol
1. radical coupling
reaction
H, OMe: Coniferyl alcohol
OMe, OMe: Sinapyl alcohol
lignin
HO
R
H
4
β
O
R'
H
Left side
α
Right side
(Si face)
(Re face)
R'''
R''
O
Quinone methide
(racemic)
2. water addition
from left side
2. water addition
from right side
HO
HO
R
R
HO
H
H
H
O
H
β
β
HO
O 4
4
α
α
R'
R'
R''
R'''
R''
R'''
OH
OH
erythro form
threo form
β-O-4 structures in lignin
114
115 Scheme 1. Formation of arylglycerol-β-aryl ether (β–O–4) structures during lignin
116 biosynthesis.
117
118
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 6 of 34
119 EXPERIMENTAL
120
121 General. The reagents and solvents used were purchased from Fujifilm Wako Pure Chemical
122 Co. (Osaka, Japan), Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan), or Sigma-Aldrich
123 (Tokyo, Japan). The pH levels of the buffers were recorded with a Horiba F-52 pH meter
124 equipped with a JF15 electrode (Horiba, Kyoto, Japan). The UV spectra were recorded on a
125 UV–visible spectrometer (Jasco V-660, Jasco International Co. Ltd., Tokyo, Japan) equipped
126 with a circulating water bath (cooling-circulator CB-15, Iuchi Seieido Co. Ltd., Japan). The
127 NMR spectra were recorded with a JEOL JNM-A500 500 MHz spectrometer for structural
128 elucidation, and assignment of newly synthesized compounds. The standard JEOL programs
129 of one- and two-dimensional (proton, carbon, DEPT-135, COSY, HSQC, and HMBC) NMR
130 experiments were used. The central peak of the residual solvent was used as an internal
131
reference (δH 7.26, δc 77.0 ppm for CDCl3; δH 2.04, δc 29.8 ppm for acetone-d6). The
132 traditional numbering system for lignins1,30 was used rather than the systematic IUPAC
133 numbering system.
134 Synthesized model compounds.
135 Four β–O–4-type lignin model compounds, viz. guaiacylglycerol-β-guaiacyl ether (GG),
136 guaiacylglycerol-β-p-hydroxyphenyl ether (GH), p-hydroxyphenylglycerol-β-guaiacyl ether
137 (HG), and p-hydroxyphenylglycerol-β-p-hydroxyphenyl ether (HH), bearing H- and/or
138 G-type nuclei on two aromatic rings (Scheme 3), were synthesized according to Adler’s
139 method.31 The erythro and threo isomers of the β–O–4 model (GG, GH, HH, or HG) were
140 separated from each other,32-34 and were used as the authentic compounds for the structural
141 determination of products in the water addition experiment. NMR data were described in our
142 previous study,8 together with the configuration of each isomer, determined by the ozonation
143 method.29 p-Coumaryl alcohol and coniferyl alcohol were prepared according to the method
144 of Quideau and Ralph.35
145
146 A β–O–4-biphenyl ether compound (H-HHbiphenyl).
ACS Paragon Plus Environment

Page 7 of 34
Journal of Agricultural and Food Chemistry
147 As a trimeric β–O–4 model, p-hydroxyphenylglycerol-β-biphenyl ether (H-HHbiphenyl),
148 bearing a 5–5-linked biphenyl in its β-etherified structural moiety, was prepared as shown in
149 Scheme 2. This β–O–4 model was synthesized from 2,2'-biphenol and acetoguaiacone by
150 Alder’s method31 as described below.
2
2
3
4
1
6
1
6
3
4
i
HO
HO
5
5
2
2
5
5
HO
OH
OMe
1
6
3
4
1
6
3
4
6
1
6
1
R
γ
4
3
4
3
B
B
O
HO
β
β
O
O
α
5
α
5
2
2
1
5
1
5
OMe
OMe
1
6
5
2
3
6
1
iv
6
2
6
1
vi
4
3
4
3
A
A
C
C
5
3
R
4
4
2
2
O
O
CH3
O
OR
β
α
1
1
6 (R = Bn)
H-HHbiphenyl (R = H)
6
5
2
3
6
2
3
vii
ii
5
4
4
OBn
OH
4 (R = H)
5 (R = CH2OH)
2 (R = H)
3 (R = Br)
v
iii
2
2
HO
HO
3
4
1
6
3
4
1
6
γ
γ
B
B
HO
H
β
β
α
O
O
α
5
5
viii
1
5
1
5
OMe
OMe
6
5
2
3
6
1
6
2
3
6
1
4
3
4
3
A
A
C
C
5
4
4
2
2
OH
O
QM-H-HHbiphenyl
H-HHbiphenyl
151
152 Scheme 2. Synthetic route of β–O–4 model compound H-HHbiphenyl and its quinone methide,
153 QM-H-HHbiphenyl: (i) MeI, K2CO3; (ii) BnCl, KI, K2CO3; (iii) Br2; (iv) K2CO3; (v) HCHO,
154 K2CO3; (vi) NaBH4; (vii) H2, Pd/C; and (viii) 1, TMSiBr and 2, NaHCO3.
155
156
Compound 1 (2'-methoxy-biphenyl-2-ol). To a solution of 2,2'-biphenol (12.02 g,
157 64 mmol) in acetone (dried over MgSO4, 200L), methyl iodide (10.98 g, 77 mmol) and
158 potassium carbonate (10.67 g, 77 mmol) were added. The reaction suspension was stirred at
159 room temperature for 20 h. The reaction was monitored by thin-layer chromatography [TLC,
160 hexane-EtOAc, 2:1; Rf: 0.34 (starting material) and 0.38 (target compound)]. The mixture
161 was partially concentrated under reduced pressure and extracted with dichloromethane. The
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 8 of 34
162 organic layer was washed with water and saturated NaCl, successively, and dried over Na2SO4.
163 After filtration, the filtrate was concentrated under vacuum pressure. The residue was
164 separated by silica gel chromatography with hexane-EtOAc (2:1, v/v) to afford
165 2'-methoxy-biphenyl-2-ol 1 in syrup form (11.3 g, 88% yield). 1H NMR (CDCl3, 500 MHz),
166 δH: 3.91 (3H, s, OMe), 6.24 (0.9H, s, B4-OH), 7.04 (1H, broad-td, J = 7, 1 Hz, B1), 7.06 (1H,
167 br-dd, J = 7, 1 Hz, B3), 7.07 (1H, br-dd, J = 7, 1 Hz, C3), 7.14 (1H, br-td, J = 7, 1 Hz, C1),
168 7.28 (1H, br-dd, J = 7, 2 Hz, B6), 7.32 (1H, br-td, J = 7, 2 Hz, B2), 7.36 (1H, br-dd, J = 7, 2
13
169 Hz, C6), 7.41 (1H, br-td, J = 7, 2 Hz, C2). C NMR (CDCl3, 125 MHz), δC: 56.2 (OMe),
170 111.6 (C3), 117.4 (B3), 120.9 (B1), 122.2 (C1), 126.2 (B5), 127.1 (C5), 129.2 (B2), 129.3
171 (C2), 131.3 (B6), 132.5 (C6), 153.7 (B4), 155.5 (C4).
172
Compound 2 (4-benzyloxyacetophenone). Potassium iodide (2.32 g, 14 mmol),
173 potassium carbonate (38.70 g, 280 mmol), 4-hydroxyacetophenone (27.23 g, 200 mmol), and
174 200 mL of dimethylformamide (dried over MgSO4) were placed in a 500-mL round-bottom
175 flask equipped with a reflux condenser. The mixture was heated to 90 °C in an oil bath while
176 stirring. Benzyl chloride (35.44 g, 280mmol) was added in small portions. The reaction
177 mixture was continuously stirred at 90 °C for an additional 1.5 h. The reaction progress was
178 monitored by TLC [hexane-EtOAc, 2:1; Rf: 0.17 (starting material) and 0.37 (target
179 compound)]. The reaction mixture was cooled to room temperature, poured onto crushed ice
180 (approx. 150 g), and stirred until the ice melted. The white solid formed was isolated by
181 suction filtration, after which it was washed with water, and once with hexane. The solid was
182 crystalized from EtOH to afford 4-benzyloxyacetophenone (40.6 g, 91% yield). 1H NMR
13
183 (acetone-d6, 500 MHz), δH: 2.50 (3H, s, β), 5.21 (2H, s, PhCH2), 7.09–7.96 (9H, m, Ar). C
184 NMR (acetone-d6, 125 MHz), δC: 26.5 (β), 70.7 (PhCH2), 115.5, 128.6, 128.9, 129.4 and
185 131.3, (Ar, -CH), 131.6, 137.8 and 163.6, (Ar, quaternary C), 196.4 (α).
186
Compound
3
(4-benzyloxy-α-bromoacetophenone).
The
compound,
187 4-benzyloxyacetophenone 2 (12.32 g, 54 mmol), was dissolved in EtOH (390 mL) while
188 stirring under nitrogen atmosphere. Bromine (10.55 g, 66 mmol) was quickly added to the
189 stirred mixture at room temperature. The reaction was monitored by TLC [hexane-EtOAc,
190 2:1; Rf: 0.44 (starting material) and 0.52 (target compound)]. After 25 min of continuous
191 stirring, the color of the solution faded from dark red to pale yellow accompanied by the
192 disappearance of the starting material. The reaction mixture was poured onto approx. 200 g of
ACS Paragon Plus Environment

Page 9 of 34
Journal of Agricultural and Food Chemistry
193 ices and a white solid substance began to form during stirring; the mixture was stirred until
194 the ice melted. The solid was collected by suction filtration, after which it was washed with
195 water, and once with hexane. The solid was crystalized from EtOH to afford
196 4-benzyloxy-α-bromoacetophenone (16.5 g, 98% yield). 1H NMR (acetone-d6, 500 MHz), δH:
197 3.88 (2H, s, β), 4.46 (2H, s, PhCH2), 6.36–7.25 (9H, m, Ar). 13C NMR (acetone-d6, 125 MHz),
198 δC: 32.7 (β), 70.9 (PhCH2), 115.8, 128.66, 129.0, 129.5 and 132.1 (Ar, -CH), 12.2, 137.7 and
199 164.3, (Ar, quaternary C), 196.4 (α).
200
Compound
4
201 (1-(4-benzyloxyphenyl)-2-(2'-methoxy-[1,1'-biphenyl]-2-yloxy)ethanone).
Biphenol
202 monomethylated 1 (8.20 g, 41 mmol), bromoacetophenone 3 (11.02 g, 36 mmol), powdered
203 potassium carbonate (6.03 g, 44 mmol), and 180 mL of acetone (dried over MgSO4) were
204 placed in a 500-mL round-bottom flask. The mixture was heated at 50 °C for 15 h while
205 stirring. The reaction was monitored by TLC [hexane-EtOAc, 2:1; Rf: 0.44 (target compound)
206 and 0.48 (starting material)]. The mixture was partially concentrated under reduced pressure,
207 and extracted with dichloromethane. The organic layer was washed with 0.1 M NaOH (× 2),
208 water, and saturated NaCl, successively, and dried over Na2SO4. After filtration, the filtrate
209 was concentrated under vacuum pressure. The crude product was purified by crystallization
210 from EtOH to afford compound 4 (14.2 g, 93% yield). 1H NMR (CDCl3, 500 MHz), δH: 3.71
211 (3H, s, OMe), 5.08 (2H, s, β), 5.13 (2H, s, PhCH2), 6.93 (2H, d, J = 8.9 Hz, A3 and A5), 6.93
212 (2H, broad-d, J = 7 Hz, C3 and B3), 6.99 (1H, br-td, J = 7, 1 Hz, C1), 7.05 (1H, br-td, J = 7, 1
213 Hz, B1), 7.27 (1H, br-dd, J = 7, 2 Hz, B6), 7.28 (1H, br-td, J = 7, 2 Hz, B2), 7.29 (1H, br-dd,
214 J = 7, 2 Hz, C6), 7.32 (1H, br-td, J = 7, 2 Hz, C2), 7.36 (1H, m, Bn4), 7.42 (2H, m, Bn3 and
215 Bn5), 7.42 (2H, d, J = 8.6 Hz, Bn2 and Bn6), 7.89 (2H, d, J = 8.9 Hz, A2 and A6). 13C NMR
216 (CDCl3, 125 MHz), δC: 55.6 (OMe), 70.1 (PhCH2), 72.3 (β), 110.9 (C3), 113.1 (B3), 114.6
217 (A3 and 5), 120.2 (C1), 121.5 (B1), 127.4 (Bn2 and Bn6), 127.5 (A1 and C5), 128.1 (B5),
218 128.2 (Bn4), 128.6 (B2), 128.7 (C2), 128.7 (Bn3 and Bn5), 130.9 (A2,6), 131.6 (C6), 131.7
219 (B6), 136.1 (Bn1), 155.7 (B4), 157.0 (C4), 162.9 (A4), 193.8 (α).
220
Compound
5
221 (1-(4-benzyloxyphenyl)-3-hydroxy-(2'-methoxy-[1,1'-biphenyl]-2-yloxy)propan-1-one).
222 To a solution of compound 4 (9.67 g, 23 mmol) in THF (115 mL), a 37 wt% aqueous solution
223 of formaldehyde (6.84 g, 229 mmol), and potassium carbonate (3.16 g, 23 mmol) were added
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 10 of 34
224 while stirring. The mixture was heated at 40 °C, and continuously stirred for 16 h. The
225 reaction was monitored by TLC [hexane-EtOAc, 2:1; Rf: 0.31 (target compound) and 0.46
226 (starting material)]. The reaction mixture was neutralized with 1M HCl, and extracted with
227 dichloromethane. The organic layer was washed with water and saturated NaCl, successively,
228 and dried over Na2SO4. After filtration, the filtrate was concentrated under vacuum pressure.
229 The crude product was purified by silica gel chromatography with hexane-EtOAc (2:1, v/v) to
230 afford compound 5, a pale-yellow syrup (8.86 g, 86% yield). 1H NMR (CDCl3, 500 MHz), δH:
231 3.76 (3H, s, OMe), 3.85 (1H, broad-dd, J = 12, 7 Hz, γ1), 3.91 (1H, br-dd, J = 12, 3 Hz, γ2),
232 5.12 (2H, s, PhCH2), 5.37 (1H, br-dd, J = 7, 3 Hz, β), 6.77 (1H, br-d, J = 8 Hz, C3), 6.93 (2H,
233 d, J = 8.9 Hz, A3 and A5), 7.02 (1H, br-d, J = 8 Hz, B3), 7.03 (1H, br-t, J = 8 Hz, C1), 7.07
234 (1H, br-t, J = 8 Hz, B1), 7.23 (1H, br-td, J = 8, 2 Hz, C2), 7.24 (1H, br-dd, J = 8, 2 Hz, C6),
235 7.27 (1H, br-dd, J = 8, 2 Hz, B6), 7.37 (1H, br-td, J = 8, 2 Hz, B2), 7.34-7.42 (5H, m, PhCH2),
236 7.95 (2H, d, J = 8.9 Hz, A2 and A6). 13C NMR (CDCl3, 125 MHz), δC: 56.5 (OMe), 63.5 (γ),
237 70.1 (PhCH2), 83.4 (β), 112.2 (B3), 113.1 (C3), 114.7 (A3 and A5), 121.3 (B1), 121.7 (C1),
238 127.4, 127.8 (A1), 128.2 (C5), 128.3 (Bn4), 128.6 (B5), 128.7, 128.8 (C2), 129.0 (B2), 131.4
239 (A2 and A6), 131.6 (B6 and C6), 136.0 (Bn1), 154.7 (B4), 156.7 (C4), 163.1 (A4), 195.0 (α).
240
Compound
6
241 (1-(4-benzyloxyphenyl)-2-(2'-methoxy-[1,1'-biphenyl]-2-yloxy)propane-1,3-diol). To
a
242 stirred solution of compound 5 (10.29 g, 22.76 mmol) in THF-EtOH (180 mL, 1:1, v/v),
243 excess NaBH4 (1.72 g, 45.51 mmol) was added slowly over 5 min. The reaction suspension
244 was stirred at room temperature for 11 h. The completion of the reaction was monitored by
245 TLC [hexane-EtOAc, 1:1, v/v; Rf: 0.39 (erythro, target compound), 0.48 (threo, target
246 compound), and 0.53 (starting material)]. To deplete the excess NaBH4, 1 M HCl was added
247 dropwise. The reaction mixture was extracted with dichloromethane, washed twice with H2O,
248 once with saturated NaCl, and dried over Na2SO4. After filtration, the filtrate was
249 concentrated under vacuum pressure to afford compound 6 (9.74 g, 94% yield) as a mixture
250 of erythro and threo isomers (6e and 6t, respectively). Each isomer was separated from the
251 mixture by silica-gel chromatography (hexane-EtOAc, 9:1 to 3:7). Compound 6e (erythro
252 isomer). 1H NMR (CDCl3, 500 MHz), δH: 3.60 (1H, m, γ1), 3.72 (1H, broad-dd, J = 12, 6 Hz,
253 γ2), 3.74 (3H, s, OMe), 4.42 (1H, m, β), 4.90 (1H, t, J = 4.0 Hz, α), 5.04 (2H, s, PhCH2), 6.87
254 (2H, d, J = 8.6 Hz, A2 and A6), 7.02 (1H, br-d, J = 7 Hz, B3 and C3), 7.06 (1H, br-td, J = 7, 1
255 Hz, B1), 7.07 (1H, br-td, J = 7, 1 Hz, C1), 7.11 (2H, d, J = 8.6 Hz, A3 and A5), 7.21 (1H,
ACS Paragon Plus Environment

Page 11 of 34
Journal of Agricultural and Food Chemistry
256 br-dd, J = 7, 2 Hz, C6), 7.25 (1H, br-dd, J = 7, 2 Hz, B6), 7.32 (1H, br-td, J = 7, 2 Hz, B2),
257 7.33 (1H, m, Bn4), 7.39 (1H, br-td, J = 7, 2 Hz, C2), 7.39 (2H, br-t, J = 7 Hz, Bn3 and Bn5),
13
258 7.43 (2H, br-d, J = 7 Hz, Bn2 and Bn6). C NMR (CDCl3, 125 MHz), δC: 56.5 (OMe), 60.3
259 (γ), 70.1 (PhCH2), 72.3 (α), 82.2 (β), 112.0 (C3), 113.9 (B3), 114.7 (A2 and A6), 121.2 (C1),
260 121.5 (B1), 127.3 (A3 and A5), 127.4 (Bn2 and Bn6), 127.9 (Bn4), 128.3 (C5), 128.6 (Bn3
261 and Bn5), 128.8 (B2), 129.0 (C2), 129.4 (B5), 131.5 (C6), 131.5 (B6), 132.2 (A1), 136.9
1
262 (Bn1), 154.7 (B4), 156.7 (C4), 158.2 (A4). Compound 6t (threo isomer). H NMR (CDCl3,
263 500 MHz), δH: 3.34 (1H, broad-dd, J = 12, 3 Hz, γ1), 3.64 (1H, br-dd, J = 12, 3 Hz, γ2), 3.78
264 (3H, s, OMe), 4.37 (1H, m, β), 4.75 (1H, d, J = 8.0 Hz, α), 5.06 (2H, s, PhCH2), 6.94 (2H,
265 br-d, J = 9 Hz, A2 and A6), 7.05 (1H, br-d, J = 7 Hz, B3 and C3), 7.07 (1H, br-td, J = 7, 1 Hz,
266 B1), 7.09 (1H, br-td, J = 7, 1 Hz, C1), 7.24 (1H, br-dd, J = 7, 2 Hz, C6), 7.25 (1H, br-dd, J =
267 7, 2 Hz, B6), 7.29 (2H, br-d, J = 9 Hz, A3 and A5), 7.32 (1H, br-td, J = 7, 2 Hz, B2), 7.33 (1H,
268 br-t, J = 7 Hz, Bn4), 7.39 (2H, br-t, J = 7 Hz, Bn3 and Bn5), 7.39 (1H, br-td, J = 7, 2 Hz, C2),
13
269 7.43 (2H, br-d, J = 7 Hz, Bn2 and Bn6). C NMR (CDCl3, 125 MHz), δC: 56.4 (OMe), 60.3
270 (γ), 70.0 (PhCH2), 73.2 (α), 83.5 (β), 111.9 (C3), 113.5 (B3), 114.9 (A2 and A6), 121.4 (C1),
271 121.5 (B1), 127.4 (Bn2 and Bn6), 127.9 (Bn4), 128.3 (A3 and A5), 128.5 (Bn3 and Bn5),
272 128.9 (B2), 129.0 (C2), 129.1 (C5 and B5), 131.4 (C6), 131.6 (B6), 131.7 (A1), 136.9 (Bn1),
273 155.1 (B4), 156.6 (C4), 158.7 (A4).
274
Compound
H-HHbiphenyl-erythro
(erythro
isomer
of
275 1-(4-hydroxyphenyl)-2-(2'-methoxy-[1,1'-biphenyl]-2-yloxy)propane-1,3-diol). A stirred
276 suspension of compound 6e (1.65 g, 3.6 mmol) and 10% Pd/C (0.17 g) in THF (73 mL) was
277 fitted with a hydrogen-gas-filled balloon. The reaction mixture was continuously stirred
278 overnight, at room temperature. The completion of the reaction was monitored by TLC
279 [hexane-EtOAc=1:1; Rf: 0.35 (target compound) and 0.45 (starting material)]. After the
280 reaction, Pd/C was passed through a glass filter using celite and washed with acetone. The
281 filtrate was concentrated under vacuum pressure. The residue was purified by silica gel
282
chromatography (hexane-EtOAc, 1:1) to afford the erythro isomer of compound H-HHbiphenyl
283 in syrup form (1.3 g, 98% yield). The configuration of the compound was determined by
284 ozonation in AcOH-H2O-MeOH (80:15:5, v/v/v) solvent, which afforded only erythronic acid.
285 1H NMR (CDCl3, 500 MHz), δH: 3.55 (1H, broad-dd, J = 12, 3 Hz, γ1), 3.69 (1H, br-dd, J =
286 12, 5 Hz, γ2), 3.71 (3H, s, OMe), 4.39 (1H, m, β), 4.84 (1H, d, J = 4.3 Hz, α), 6.62 (2H, br-d,
287 J = 8 Hz, A2 and A6), 6.98 (2H, br-d, J = 8 Hz, A3 and A5), 6.99 (1H, br-d, J = 8 Hz, B3),
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 12 of 34
288 7.01 (1H, br-d, J = 8 Hz, C3), 7.05 (1H, br-t, J = 8 Hz, B1), 7.05 (1H, br-t, J = 8 Hz, C1), 7.19
289 (1H, br-dd, J = 8, 2 Hz, C6), 7.23 (1H, br-dd, J = 8, 2 Hz, B6), 7.30 (1H, br-td, J = 8, 2 Hz,
290
B2), 7.36 (1H, br-td, J = 8, 2 Hz, C2). 13C NMR (CDCl3, 125 MHz), δC: 56.5 (OMe), 60.1 (γ),
291 72.2 (α), 82.0 (β), 112.2 (C3), 114.0 (B3), 115.2 (A2 and A6), 121.4 (C1), 121.6 (B1), 127.4
292 (A3 and A5), 128.3 (C5), 128.9 (B2), 129.0 (C2), 129.4 (B5), 131.3 (A1), 131.5 (C6), 131.6
293 (B6), 154.6 (B4), 155.3 (A4), 156.7 (C4).
294
Compound
H-HHbiphenyl-threo
(threo
isomer
of
295 1-(4-hydroxyphenyl)-2-(2'-methoxy-[1,1'-biphenyl]-2-yloxy)propane-1,3-diol). Following
296 the procedure described above for the erythro isomer of H-HHbiphenyl-erythro, compound 6t
297 (5.57 g, 12.2 mmol) was converted to the title compound (H-HHbiphenyl-threo) in syrup form
298 (4.4 g, 98% yield). Ozonation of this compound in AcOH-H2O-MeOH (80:15:5, v/v/v)
299 solvent afforded only threonic acid. The Rf values on TLC (hexane-EtOAc, 1:1, v/v) were
300 0.42 (target compound) and 0.56 (starting material). 1H NMR (CDCl3, 500 MHz), δH: 3.29
301 (1H, broad-dd, J = 13, 3 Hz, γ1), 3.61 (1H, br-dd, J = 13, 3 Hz, γ2), 3.74 (3H, s, OMe), 4.36
302 (1H, m, β), 4.68 (1H, d, J = 8.0 Hz, α), 6.63 (2H, d, J = 8.6 Hz, A2 and A6), 7.04 (1H, br-d, J
303 = 8 Hz, B3), 7.05 (1H, br-d, J = 8 Hz, C3), 7.06 (1H, br-t, J = 8 Hz, B1), 7.08 (1H, br-t, J = 8
304 Hz, C1), 7.13 (2H, d, J = 8.6 Hz, A3 and A5), 7.22 (1H, br-dd, J = 8, 2 Hz, C6), 7.23 (1H,
305 br-dd, J = 8, 2 Hz, B6), 7.32 (1H, br-td, J = 8, 2 Hz, B2), 7.38 (1H, br-td, J = 8, 2 Hz, C2). 13C
306 NMR (CDCl3, 125 MHz), δC: 56.6 (OMe), 60.3 (γ), 73.4 (α), 83.4 (β), 112.3 (C3), 113.4 (B3),
307 115.6 (A2 and A6), 121.6 (C1 and B1), 128.4 (B5 and C5), 128.5 (A3 and A5), 129.0 (B2),
308 129.1 (C2), 130.6 (A1), 131.4 (C6), 131.7 (B6), 155.1 (B4), 156.0 (A4), 156.6 (C4).
309 Preparation of QMs. Dimeric β–O–4-aryl ether QM compounds (QM-HH, -HG, -GH, and
310 -GG) were synthesized from the corresponding dimeric β–O–4 model compounds (HH, HG,
311 GH, and GG, respectively) by mild alkaline treatment of the benzyl bromides,4 which were
312 prepared using trimethylsilyl bromide (TMSiBr).5, 36 The detailed procedure was described in
313 our previous report.8 The β–O–4-biphenyl ether QM (QM-H-HHbiphenyl) was also prepared
314 according to this method, as described below.
315 The β-O-4 compound H-HHbiphenyl (0.1 mmol) was dissolved in chloroform (1 mL) in a 4-mL
316 vial with a screw cap. TMSiBr (0.2 mmol) was added to this solution and shaken for 90 s at
317 room temperature to generate benzyl bromide. The reaction mixture was shaken with
ACS Paragon Plus Environment

Page 13 of 34
Journal of Agricultural and Food Chemistry
318 saturated NaHCO3 (1 mL) for 10 s. The collected organic layer was washed with saturated
319 NaCl and dried over Na2SO4. The resulting QM-H-HHbiphenyl (pale-yellow) was diluted with
320 dioxane into desired concentrations (approx. 0.1 mM or 8 mM). These QM solutions were
321 used for the subsequent water addition experiments without further purification. For NMR
322 structural determination, the QM was prepared in a separate experiment, the same way, using
323 deuterated chloroform as a reaction solvent. The chloroform-d solution of the crude QM was
324 transferred into an NMR tube and kept in liquid N2 before the NMR measurement. No
325 residual peak from the β–O–4 dimer was found on the 1H NMR spectrum of QM-H-HHbiphenyl
326 (Figure S1) as well as on those of the other QMs.8
1
Compound QM-H-HHbiphenyl. H NMR (CDCl3, 500 MHz), δH: 3.68 (2H, m, γ),
327
328 3.77 (3H, s, C4-OMe), 5.25 (1H, m, β), 6.33 (1H, d, J = 7.2 Hz, α), 6.37 (1H, broad-dd, J =
329 10, 2 Hz, A3), 6.40 (1H, br-dt, J = 10, 2 Hz, A5), 6.84 (1H, br-d, J = 8 Hz, B3), 7.00 (1H,
330 br-dd, J = 10, 2 Hz, A2), 7.04 (1H, br-d, J = 8 Hz, C3), 7.08 (2H, br-t, J = 7 Hz, B1 and C1),
331 7.23 (1H, br-dd, J = 7, 2 Hz, C6), 7.26 (1H, br-dd, J = 7, 2 Hz, B6), 7.28 (1H, m, B2), 7.38
13
332 (1H, br-td, J = 7, 8, 2 Hz, C2), 7.46 (1H, br-dd, J = 10, 2 Hz, A6). C NMR (CDCl3, 125
333 MHz) δC: 56.5 (C4-OMe), 64.6 (γ), 77.4 (β), 112.2 (C3), 114.1 (B3), 121.4 (C1), 122.2 (B1),
334 128.0 (B5 and C5), 128.9 (A3), 129.1 (B2 and C2), 130.4 (A5), 131.5 (C6), 131.8 (B6), 133.0
335 (A1), 133.1 (A6), 141.5 (A2), 144.9 (α), 154.7 (B4), 156.6 (C4), 186.9 (A4).
336 Water addition to QMs in different solvent conditions at pH 5 or 7. A dioxane solution of
337 QM prepared as described above (approx. 8 mM × 75 μL, which originates from 0.6 μmol of
338 β-O-4 dimer), was further diluted with a certain volume of dioxane (0–3 mL) in a small vial
339 with a screw cap; after mixing with a buffer, the volume ratio of dioxane-water (v/v) became
340 1:79, 1:7, 1:5, 1:1.5, or 1:1 with a total volume of approx. 6 mL (Therefore, water addition
341 reaction was conducted at the concentration of approx. 0.1 mM QM). The water addition
342 reaction was initiated by the addition of the buffer to the dioxane solution of QM. This
343 marked the reaction time of 0 s. After the immediate shaking of the reaction mixture for a few
344 seconds, an aliquot of the mixture (approx. 3 mL) was transferred into a quartz UV cell (1 cm
345 square, TOS-UV-10; Toshin Riko Co., Ltd., Japan), which was placed in a UV–visible
346 spectrometer equipped with a circulating water bath. The reaction mixture was maintained at
347 25 °C, and monitoring of the reaction progress began at 20 s. The disappearance of QMs and
348 their half-life (t1/2) in the buffered dioxane–water mixture was determined on the basis of the
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 14 of 34
349 decrease in UV absorbance.8,37 Absorbance data at 304 nm was collected every 0.2 s (or 1 s)
350 from the reaction time of 20 s to 10 min (or 120 min) depending the reaction rates of QMs.
351 The collected data were fitted with an exponential function to calculate the pseudo-first-order
352 reaction rate constants (kobs) for the disappearance of the QMs (Table S1). The citrate–
353 phosphate buffer at pH 4.8, 4.5, 4.4, 3.8, or 3.2 was used to prepare the 1:79, 1:7, 1:5, 1:1.5,
354 or 1:1 ratio of dioxane-water mixture, respectively; after mixing with dioxane, the resulting
355 buffered dioxane–water mixture indicated a pH of 5.0. Similarly, the buffer at pH 6.8, 6.6, 6.4,
356 5.5, or 5.3 was used to conduct the water addition experiment at pH 7.0 in the dioxane–buffer
357 mixtures (1:79, 1:7, 1:5, 1:1.5, or 1:1, v/v). The buffer at pH 3.4 was used for the water
358 addition experiment in dioxane–water mixture (1:79), at pH 3.5.
1
359
The H NMR spectra of the QMs indicated that each crude QM solution was
360 essentially pure and contained only small amounts of unknown products. Additionally, the
361 1H NMR spectra of QM-H-HHbiphenyl (Figure S1) and of the other QMs8 did not exhibit any
362 residual peak of the substrate (β–O–4 model). The β–O–4 yield was expressed as the yield of
363 the succeeding two reaction steps consisting of the QM formation (from a β–O–4 model) and
364 water addition (from QM to the β–O–4 products).
365
The yields of erythro and threo β–O–4 products were determined by HPLC
366 analyses. After the complete consumption of QM, an aliquot of the reaction mixture (1 mL)
367 was mixed with a MeOH solution of 3,4-dimethoxyacetophenone (0.05 μmol), which was
368 used as an internal standard (IS). The mixture was passed through a hydrophilic PTFE
369 membrane filter (Millex-LG, 0.2 μm, Millipore), and analyzed by a high-performance liquid
370 chromatograph (HPLC, LC-10A, Shimadzu Co., Kyoto, Japan) equipped with an SPD-M10A
371 detector (280 nm, Shimadzu Co.).
372
The conditions for the HPLC analyses of the β–O–4 dimers GG, GH, HG and HH
373 were described in our previous report.8 The conditions for the HPLC analysis of the β–O–4
374 product H-HHbiphenyl were as follows: column, Luna 5u C18 (2) 100 A (150 mm × 4.6 mm;
375 Phenomenex, Inc., Torrance, CA, USA); oven temperature, 40 °C; flow rate, 1.0 mL min−1.
376 The erythro and threo diastereomers of compound H-HHbiphenyl in the reaction products
377 mixture were separated using a binary gradient system consisting of CH3OH and H2O initially
378 at 15:85 ratio (v/v), and subsequently increased linearly over 7.5 min to 45:55, and again over
ACS Paragon Plus Environment

Page 15 of 34
Journal of Agricultural and Food Chemistry
379 27.5 min to 75:25 (35 min total), following an equilibration step. The retention times of the
380 products were 13.2 min for 3,4-dimethoxyacetophenone (IS), and 21.7 and 22.8 min for
381 H-HHbiphenyl (erythro and threo isomers, respectively). The calibration curves of H-HHbiphenyl
382 were as follows: Y = 1.93X + 0.008 (for the erythro isomer) and Y = 1.83X + 0.026 (for the
383 threo isomer). Y is the molar ratio of each isomer to IS, and X is the peak area of each isomer
384 to IS.
385 Water addition to QM-H-HHbiphenyl at different pH levels in dioxane-water (1:1, v/v). The
386 effect of pH on the erythro/threo ratio of β–O–4 products in the water addition to
387 QM-H-HHbiphenyl was examined using the same procedure as that used in our previous study
388 on QM-HH, QM-HG, QM-GH, and QM-GG, in which the water addition reaction was
389 conducted at a concentration of approx. 0.05 mM QM. A dioxane solution of QM prepared as
390
described above (approx. 0.1 mM × 3 mL, which originates from 0.3 μmol of β–O–4 dimer)
391 was placed in a small vial with a screw cap. The reaction was initiated by mixing with the
392 same volume of citrate–phosphate buffer at pH 2.4, 3.2, 3.9, 4.5, or 5.3 (3 mL). Therefore, the
393 water addition reaction was conducted at the concentration of approx. 0.05 mM QM. This
394 marked the reaction time of 0 s. The citrate–phosphate buffers mentioned above were
395 prepared by mixing 0.01 M citric acid and 0.02 M disodium phosphate in different
396 proportions; after mixing with an equal volume of dioxane, the resulting buffered dioxane–
397 water mixture (1:1, v/v) indicated the desired pH values (3.5, 4.5, 5.5, 6.0, and 7.0). After the
398 immediate shaking of the reaction mixture for 2–3 s, an aliquot of the mixture (approx. 3 mL)
399 was transferred into a quartz UV cell, and UV absorbance data was measured using the same
400 procedure as that in “Water addition to QMs in different solvent condition at pH 5 or 7”
401 described above.
402 Preparation of DHP of p-coumaryl alcohol (H-DHP) by dropwise addition
403 (Zutropfverfahren38). p-Coumaryl alcohol (200 mg in 100 mL of dioxane-water (3:7, v/v)),
404 hydrogen peroxide (1.77 mmol in 20 mL water), and horseradish peroxidase (3.3 mg in 20
405 mL of water; 454 units/mg, Oriental yeast CO., LTD) were separately added dropwise over
406 25 h, using three syringe pumps (YSP-202, YMC, Japan), to a 500-mL three-necks flask
407 containing a 0.5 M citrate-phosphate buffer at pH 3.5, 4.5, 5.0, 6.0 or 7.0 (100 mL), while
408 stirring at room temperature. Therefore, the proportion of dioxane in the reaction mixture was
409 changed during the addition, and the final proportion of dioxane-water was 1:7 when the
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 16 of 34
410 addition was complete. The flask was covered with aluminum foil to limit light penetration
411 during the reaction. When the addition was finished, the reaction was brought to a stop by the
412 addition of an aqueous sodium thiosulfate solution (1 mmol) and NaCl powder (36 g).
413 Subsequently, the insoluble part was collected by filtration with a hydrophilic PTFE
414 membrane filter (ADVANTEC, 0.2 μm), washed with distilled water, and dried under
415 reduced pressure to afford light-brown colored powder (H-DHPs, 132–202 mg, 66–101 wt%
416 yield). A part of the H-DHP (20 mg) was acetylated with acetic anhydride (0.5 mL) and
417 pyridine (1.5 mL), at room temperature overnight. The reaction mixture was concentrated
418 under reduced pressure at 45 °C. The remaining solvent was removed by repetition of the
419 evaporation with EtOH to afford acetylated DHP.
420 Preparation of DHP of coniferyl alcohol (G-DHP) by dropwise addition
421 (Zutropfverfahren38). Coniferyl alcohol (200 mg in 100 mL of dioxane-water (3:7, v/v)),
422 hydrogen peroxide (1.77 mmol in 20 mL water), and horseradish peroxidase (3.3 mg in 20
423 mL of water; 454 units/mg) were separately added dropwise over 25 h using three syringe
424 pumps to a 500 mL three-necks flask containing a 0.5 M citrate-phosphate buffer at pH 6.0
425 (100 mL), while stirring at room temperature. The same work-up procedure as that of the
426 DHPs mentioned above was used to collect an insoluble part from the reaction suspension
427 (G-DHP-6, 151 mg, 76 wt% yield). A part of G-DHP-6 (20 mg) was acetylated with acetic
428 anhydride and pyridine.
429 Preparation of Zulauf-DHPs by one-time addition (Zulaufverfahren38). Monolignol (200
430 mg of p-coumaryl alcohol (H-type monolignol) or coniferyl alcohol (G-type monolignol) in
431 100 mL of dioxane-water (3:7, v/v)), hydrogen peroxide (1.77 mmol in 20 mL water), and
432 horseradish peroxidase (3.3 mg in 20 mL of water; 454 units/mg) were added simultaneously
433 to a 0.5 M citrate-phosphate buffer at pH 6.0 (100 ml), and stirred for 1 h at room temperature.
434 The same work-up procedure as that of the DHPs mentioned above was used to collect an
435 insoluble part from the reaction suspension (H-Zulauf-DHP, 174 mg, 87 wt% yield;
436 G-Zulauf-DHP, 120 mg, 60 wt%). A part of the Zulauf-DHP (20 mg) was acetylated with
437 acetic anhydride and pyridine.
438 Ozonation method. The stereo configurations of the synthesized β–O–4 model compounds
439 and the β–O–4 structures in DHPs were determined by the ozonation method according to
ACS Paragon Plus Environment

Page 17 of 34
Journal of Agricultural and Food Chemistry
440 Akiyama et al.29 A β–O–4 compound (10 μmol) or DHP (20 mg) was treated with ozone in
441 AcOH-H2O-MeOH solvent (80:15:5, v/v/v). The erythronic acid (E) and threonic acid (T)
442 obtained as ozonation products were determined by gas chromatography as trimethylsilyl
443 derivatives. The erythro/threo ratio was defined as the E/T ratio.
444 Size-exclusion chromatographic (SEC) analyses. SEC analyses of the acetylated DHP
445 samples were performed on a Shimadzu HPLC/GPC system with an LC-20AD pump and an
446 SPD-20A UV-Vis detector. The conditions for the SEC analyses were as follows: column,
447 Shodex GPC KF-802 and KF-802.5 (Showa Denko, Japan) in series; column oven
448 temperature, 40 °C; eluent, THF; flow rate, 1.0 mL min-1. The retention time of the column
449 was calibrated with polystyrene standards (Shodex Standards, SL-105, Showa Denko, Japan)
450 in the MW range of 482–0.94 × 103 g mol-1, propylbenzene (Mw, 120), and benzene (Mw, 78).
451 Approximately 2.5 mg of acetylated DHPs (H-DHP-3.5, -4.5, and -5, H-zulauf-DHP-6, and
452 G-zulauf-DHP-6) were dissolved in THF (5 mL) and kept overnight. The solution was passed
453 through a 0.45-μm filter, and 20 μL of the solution was injected into the columns. Other
454 acetylated DHPs (H-DHP-6, and -7, and G-DHP-6) were only partially dissolved in THF and
455 could not be analyzed. Shimadzu LC solution (ver. 5.73) software was used for the data
456 acquisition and the data processing at 280 nm to obtain the peak MW (Mp), weight average
457 MWs (Mn and Mw, respectively), and polydispersity (Mw/Mn).
458
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 18 of 34
459 RESULTS AND DISCUSSION
460
In our previous study,8 water addition experiments were conducted in 50% water
461 solution (dioxane/water, 1:1, v/v) to investigate the reactivity of QMs and the stereoselectivity
462 of the formation reaction of β–O–4 structures from the QMs. We followed the experimental
463 conditions developed by Nakatsubo’s group4 and Brunow’s group,6 and the experiments were
464 conducted at a wide pH range, under neutral and mildly acidic conditions (pH 3.5–7), to cover
465 the possible pH conditions assumed in cell walls since the pH condition at the lignification
466 site remains unknown. Similarly, the solvent conditions or circumstances that model the
467 surroundings of the quinone methide structure, located at the end of growing lignins in the
468 cell walls where cellulose and hemicellulose are already present, remains unknown.
469 Nakatsubo et al.4 conducted water addition experiments to QM-GG and found that the
470 erythro/threo ratio of the β–O–4 product GG was 0.4 in dioxane/water (1:1, v/v), and 0.5 in
471 dioxane/water (1:9, v/v), respectively. In the present study, we investigated the details of the
472 solvent effect. The water concentration of buffered dioxane-water solvent was changed from
473 50 vol% to approximately 99 vol% (dioxane/water, 1:1 to 1:79, v/v), and the water addition
474 experiments were conducted for the various types of QMs at 25 °C (Scheme 3).
475
476
HO
HO
HO
γ
γ
γ
B
B
B
4
4
4
R
OMe,
H,
OMe,
H,
R'
OMe
OMe
H
HO
HO
β
β
β
O
R
: GG
: HG
: GH
: HH
O
R
α
O
R
α
α
1
1
1
R'
R'
R'
dioxane/water (1:1 – 1:79, v/v)
pH 3.5, 5, or 7 at 25 ºC
A
A
A
H
3
3
3
4
4
4
: -H-HHbiphenyl
H,
O
OH
erythro isomer
Arylglycerol-β-aryl ethers
OH
threo isomer
OMe
Quinone methides
R'
R
OMe,
H,
OMe,
H,
β-guaiacyl
ethers
: QM-GH β-p-hydroxyphenyl
: QM-HH ethers
: QM-GG
: QM-HG
OMe
OMe
H
(β-O-4 products)
H
: QM-H-HHbiphenyl
H,
OMe
477
478 Scheme 3. Water addition experiments to replicate the formation of H- and G-type β–O–
479 4-structures from the quinone methides (QMs) during lignin biosynthesis.
480
ACS Paragon Plus Environment

Page 19 of 34
Journal of Agricultural and Food Chemistry
481
482
483 Table 1. Half-lives of quinone methides (QMs)a in buffered dioxane-water solutionsb at 25 °C
484 and the yields of erythro and threo stereoisomers of β–O–4-products.
Entry
Dioxane/water
solvent (v/v)
Half-life
(t1/2)
pH 5
＜20 s
＜20 s
25 s
32 s
45 s
28 s
29 s
Product
Yield
(mol%)
erythro/threo
pH 5
pH 7
＜20 s
24 s
24 s
36 s
51 s
26 s
28 s
29 s
pH 5
pH 7
87
86
87
82
78
84
90
87
85
79
86
83
83
68
66
76
71
80
60
74
--
pH 7
19:81
20:80
21:79
23:77
23:77
30:70
29:71
28:72
27:73
27:73
23:77
24:76
24:76
26:74
27:73
43:57
39:61
38:62
35:65
34:66
--
1:79
1:7
1:5
1:1.5
1:1
1:79
1:7
1:5
1:1.5
1:1
1:79
1:7
1:5
1:1.5
1:1
1:79
1:7
1:5
1:1.5
1:1
88
86
87
86
87
85
89
84
83
84
81
77
79
75
74
73
63
66
67
75
(86) c
65
78
78
87
85
20:80
21:79
22:78
25:75
26:74
31:69
30:70
30:70
31:69
31:69
23:77
25:75
26:74
28:72
29:71
44:56
40:60
40:60
39:61
38:62
(49:51) c
55:45
58:42
59:41
61:39
60:40
QM-HH
HH
HH
HH
HH
HH
HG
HG
HG
HG
HG
GH
GH
GH
GH
GH
GG
GG
GG
GG
GG
GG
QM-HG
QM-GH
QM-GG
29 s
36 s
49 s
77 s
41 s
59 s
72 s
126 s
130 s
398 s
759 s
81 s
146 s
157 s
448 s
839 s
(47 s) c
123 s
133 s
426 s
795 s
84 s
150 s
182 s
501 s
978 s
---
(1:79, pH3.5)c
1:79
1:7
QM-H-HHd
nde
nde
nde
nde
nde
nde H-HHd
nde H-HHd
nde H-HHd
nde H-HHd
nde H-HHd
64
78
82
83
78
54:46
55:45
56:44
56:44
54:46
1:5
1:1.5
1:1
485 aSee Table S1 for the pseudo-first-order-reaction-rate constants (kobs) for the disappearance of
b
486 the QMs. Dioxane-water ratio of 1:79, 1:7, 1:5, 1:1.5, and 1:1; pH 3.5, 5, and 7.
487 cDioxane-water ratio of 1:79 at pH 3.5. dQM-H-HHbiphenyl and H-HHbiphenyl. eNot determined.
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 20 of 34
2
2
(a)
(b)
QM-HH, 1:1
QM-HH, 1:1.5
QM-HH, 1:5
QM-HH, 1:7
QM-HH, 1:79
QM-HG, 1:1
QM-HG, 1:1.5
QM-HG, 1:5
QM-HG, 1:7
1.5
1
1.5
QM-HG, 1:7
1
0.5
0
0.5
0
0
1
2
3
4
5
6
0
1
2
3
4
5
6
Reaction time (min)
Reaction time (min)
2
2
(c)
(d)
QM-GH, 1:1
QM-GH, 1:1.5
QM-GH, 1:5
QM-GH, 1:7
QM-GH, 1:79
QM-GG, 1:1
QM-GG, 1:1.5
QM-GG, 1:5
QM-GG, 1:7
QM-GG, 1:79
1.5
1
1.5
1
0.5
0.5
0
0
0
10 20 30 40 50 60
Reaction time (min)
0
10 20 30 40 50 60
Reaction time (min)
488
489 Figure 1. Effect of solvent on the disappearance of quinone methides (QMs) in pH 7 buffered
490 dioxane-water solutions, at 25 °C. Changes in the absorbance at 304 nm for (a) QM-HH, (b)
491 QM-HG, (c) QM-GH, and (d) QM-GG. Dioxane—water ratio of 1:1 to 1:79 (v/v). See
492 Figure S2 for the UV spectra of the QMs under at pH 7 condition.
493
494
ACS Paragon Plus Environment

Page 21 of 34
Journal of Agricultural and Food Chemistry
495 Reactivities of QMs with water.
496
The effect of the solvent on the half-lives of QMs was investigated. The water
497 addition reaction proceeded fast in the order of QM-HH > -HG > -GH > -GG in the
498 dioxane-water (1:1, v/v) solvent, at both pH 5 and 7 (Table 1). These results confirmed the
499
reproducibility of our previous result obtained in 50% water solution8. As shown in Table 1
500 and Figure 1, the reaction rate of QMs increased when the proportion of the water in a
501 dioxane-water (1:1, 1:1.5, 1:5, 1:7, and 1:79, v/v) solvent was increased from 50 % to approx.
502 99 %. For example, QM-GG had the longest half-life (t1/2) of 978 s in the 50%-dioxane
503 solution (dioxane-water, 1:1, v/v) at pH 7. Increasing the proportion of water shortened the t1/2
504 by up to 12-times (t1/2: 84 s in 99 %-water) at the same pH. The solvent effect was observed
505 for another G-type QM (QM-GH). More so, the reaction rate of the less stable H-type QMs
506 (QM-HH and -HG) increased with the proportion of the water (e.g., t1/2 of QM-HG at pH 7:
507 59 s and 26 s in 50%- and 99%-water solutions, respectively). Consequently, t1/2 was always
508 short in the order of QM-HH < -HG < -GH < -GG, regardless of the solvent conditions.
509 Moreover, this order of t1/2 observed at pH 7 in different solvent systems did not change at pH
510 5.
511
The half-life of QM-H-HHbiphenyl could not be measured because its UV absorbance
512 peak intensively overlapped with the shoulder of the peak of the β–O–4 products H-HH
513 (Figure S2-c). However, it was clear that QM-H-HHbiphenyl disappeared within 4 min in
514 dioxane-water (1:1, v/v) at pH 7 (Figure S2-c). This period (4 min) was even shorter than the
515 half-life of QM-GH and -GG (t1/2, 795 s for QM-GH, and 978 s for QM-GG) under the same
516 reaction condition. In addition, QM-H-HHbiphenyl afforded the β–O–4 products, H-HHbiphenyl
,
517 in high yields (82% as a total of erythro and threo isomers), suggesting that the water addition
518 reaction was the main reaction in this experiment. Therefore, it can be safely concluded that
519 the reaction rate of QM-H-HHbiphenyl with water was higher than that of methoxy-substituted
520 QMs (QM-GH and -GG), and that rapid rearomatization occurred to generate the β–O–4
521 structure similar to other non-substituted QMs (QM-HH and -HG).
522
Previously,8 as an implication for the lignin biosynthesis, we proposed that when a
523 new β–O–4-linkage is introduced at the end of a growing lignin by a radical-coupling reaction
524 with a monolignol, a nonsubstituted QM moiety (H-type), derived from p-coumaryl alcohol,
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 22 of 34
525 will undergo rearomatization faster than the methoxy-substituted QM (G-type) from coniferyl
526 alcohol will. This proposal was based on the results of t1/2 values obtained for QM models in
527 the 50% water system. The results obtained in the present study using different solvent
528 condition also supported this proposal.
529 Effect of solvent on the stereopreferential formation of the β–O–4 structure
530
We previously reported8 that the four kinds of QMs bearing a guaiacyl or a
531 p-hydroxyphenyl nuclei in their β-etherified ring (QM-HH, -HG, -GH, and -GG) showed
532 threo-preferential water addition for the formation of dimeric β–O–4-products. The resultant
533 erythro/threo ratio was always high in the order of GG > HG > GH > HH, at every pH
534 condition within the range of 3.5–7, when the water addition experiments were conducted in
535 the 50%-water solvent (dioxane-water, 1:1, v/v). The solvent effect on the erythro/threo ratio
536 was examined at pH 5 and 7 in the present study. As shown in Figures 2a1, and a2 and Table
537 1, the order did not change when the water content in the solvent changed from 50 to 99%
538 (erythro/threo ratio: GG > HG ≥ GH > HH).
539
However, the effect of the change in the solvent water content on the erythro/threo
540 ratio appeared in an opposite way between the β–O–4-guaiacyl ether QMs (QM-GG, and
541 -HG) and the β–O–4-p-hydroxyphenyl ether QMs (QM-GH, and -HH). The erythro/threo
542 ratio was high for the formation of β–O–4-products, GG and HG, with an increase in the
543 water concentration, whereas the threo isomers of GH and HH were more selectively formed
544 under a high water concentration condition. This trend was observed at both pH 5 and 7
545 (Figures 2a1 and a2). Consequently, the variation in the erythro/threo ratios of the four
546 dimeric β–O–4-products obtained in 99%-water solution was much wider than that in
547 50%-water solution (the proportion of erythro isomer obtained at pH 7: 19–43% (∆24%) in
548 99%-water solution, and 23–34% (∆11%) in 50%-water solution). In our previous study, a
549 wider variation in the erythro/threo ratio was obtained under a more acidic condition, in
550 50%-water solution (proportion of erythro isomer: 30–64% (∆26%) at pH 3.5). More so,
551 under the neutral pH condition, a comparatively wide variation in the erythro/threo ratio was
552 observed as the water concentration increased to 99% (Figure 2a2).
553
ACS Paragon Plus Environment

Page 23 of 34
Journal of Agricultural and Food Chemistry
554
70
60
50
40
30
20
10
70
(a1) pH 5
(a2) pH 7
60
50
40
30
20
10
QM-H-HH pH5
QM-HHH pH7
QM-GG pH5
QM-HG pH5
QM-GH pH5
QM-HH pH5
QM-GG pH7
QM-HG pH7
QM-GH pH7
QM-HH pH7
50 60 70 80 90 100 50 60 70 80 90 100
Water (vol%)
Water (vol%)
70
60
50
40
30
20
10
70
60
50
40
30
20
10
(b1) H-type QM
(b2) G-type QM
QM-GG pH5
QM-GG pH7
QM-GH pH5
QM-GH pH7
QM-H-HH pH5
QM-H-HH pH7
QM-HG pH5
QM-HG pH7
QM-HH pH5
QM-HH pH7
50 60 70 80 90 100 50 60 70 80 90 100
Water (vol%) Water (vol%)
555
556 Figure 2 Effect of solvent on the erythro/threo ratio of the β–O–4-products obtained by water
557 addition to quinone methides (QM) in dioxane-water solutions, at 25 °C, at pH 5 and 7. The
558 proportion of erythro isomer observed for (a1) all QMs at a pH 5, (a2) all QMs at pH 5, (b1)
559 H-type QMs, (b2) G-type QM. Water (vol%): proportions of water in the dioxane–water
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 24 of 34
560 solutions (1:1, 1:1.5, 1:5, 1:7, and 1:79, v/v). erythro isomer (%): proportion of erythro
561 isomer to the total of erythro and threo isomers.
562
In a 50%-water solution (dioxane-water, 1:1, v/v), the erythro/threo ratio observed
563 at pH 5, (solid lines in Figures 2b1 and 2b2) was significantly higher than that observed at pH
564 7 (dotted lines). The reproducibility of the pH dependence observed in our previous study was
565 confirmed by this result. However, interestingly, the difference in the erythro/threo ratios
566 caused by the difference in pH gradually ceased as the solvent water concentration increased.
567 Almost no difference was found between the erythro/threo ratios obtained at pH 5 and 7, in
568 the 99%-water solvent (dioxane-water, 1:79). This trend was observed for all five QMs
569 examined here, regardless of whether the erythro/threo ratios increase or decrease as the
570 water proportion in the dioxane-water solvent increased. In addition, as shown in Table 1, a
571 relatively high erythro/threo ratio of GG was obtained under a more acidic condition, in the
572 order of pH 3.5 > 5 > 7, in 99%-water solution. The same order was observed in the
573 50%-water solution in the previous study;8 however, the range of the erythro/threo ratio was
574 narrower in 99%-water solution than in the 50%-water solution (the proportion of erythro
575 isomer of GG at pH 3.5–7: 43–49% (∆6%) in 99%-water solution, and 35–50% (∆15%) in
576 50%-water solution). Apparently, the effect of pH on the stereo-preference of the water
577 addition reaction was smaller in the 99%-water solution than in 50%-water solution.
578
The increase or decrease in the erythro/threo ratio with the increase in the water
579 proportion depends on the chemical structures of QMs. The trend of the erythro/threo ratio
580 seems to be largely dependent on the β-etherified ring-type. As shown in Figures 2a1 and a2,
581 the erythro/threo ratio of β–O–4 products obtained from β-p-hydroxyphenyl ether QMs
582 (QM-HH and -GH) decreased as the water proportion in the solvent increased. This trend
583 was observed at both pH 5 and 7. In contrast, for β-guaiacyl ether QMs, the erythro/threo
584 ratio obtained for QM-HG did not significantly change, while that of QM-GG increased as
585 the water proportion increased (Figures 2a1 and a2), indicating that, as well as the structures
586 of β-etherified ring, the type of quinone methide moiety also influences, to some extent, the
587 erythro/threo ratio as the water proportion in the solvent increases.
588 The erythro-preferential formation of β–O–4 structure
ACS Paragon Plus Environment

Page 25 of 34
Journal of Agricultural and Food Chemistry
589
The erythro/threo ratio of β–O–4 structures in compression wood lignin are slightly
590 but clearly higher than 1.0, which was determined by the ozonation analyses in the reaction
591 woods of Loblolly pine (Pinus taeda)28 and Sumatran pine (Pinus merkusii).26 Τhe
592 erythro/threo ratio was 50:50 in the upper side of the leaning stem and increased towards the
593 lower side, i.e., the compression wood side with an increased content of H-units (the erythro
594 form 49.7% and 53.5% in the upper and lower side, respectively).26 However, as described
595 above, the water addition to the QMs stereo-preferentially afforded the threo isomer during
596 the formation of all the H- and G-type dimeric β–O–4-products, although the variation in the
597 erythro/threo ratio widened as the water concentration increased.
598
Conversely, Brunow et al.6 conducted a model study using β-syringyl ether QMs to
599 replicate the hardwood lignin biosynthesis and evidenced that erythro-selective formation
600 reaction occurs in vitro water addition to QMs. To investigate whether erythro-selective
601 formation reaction occurs in the absence of the syringyl unit, we examined an H-type QM
602 bearing a biphenyl structure, which serves to increase the steric bulkiness of the aryl group
603 etherified at the β-position of QM, which can possibly affect the preferable stereo
604 conformation of the functional groups attached to the β-asymmetric carbon of the QM, and
605 may influence the stereo-preference of the water addition reaction. We synthesized an H-type
606 QM-β-biphenyl ether compound (QM-H-HHbiphenyl in Scheme 2) from the corresponding
607 trimeric β–O–4-compound (H-HHbiphenyl) that is composed of only H-units. This structure can
608 potentially be formed during biosynthesis, although no structural evidence has been obtained
609 (Scheme 3). The HH-type biphenyl present in the structure was reported to be that of
610 dibenzodioxocin structure,39,40 however, this 8-membered ring structure can be formed by the
611 intramolecular cyclization of a β-O-4-biphenyl ether QM, instead of the water addition
612 reaction. The phenolic hydroxy group was protected by methylation in the QM-H-HHbiphenyl
613 to avoid the cyclization reaction and to make the water addition reaction possible.
614
As shown in Table 2, the erythro/threo ratios of the β–O–4 products were higher
615 than 1.0 under all the condition examined. It was evidenced that erythro-preferential water
616 addition can occur to a quinone methide composed of only H-units. Notably, an
617 erythro-preferential formation was observed even under neutral conditions in a water addition
618 experiment to a lignin-related QM, although the mechanism of the stereo-preference remains
619 unknown.
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 26 of 34
620
621 Table 2. Effect of pH on the erythro/threo ratio of the β–O–4-product (H-HHbiphenyl) obtained
622 by the water addition reaction to QM-H-HHbiphenyl, in buffered dioxane-water solutions (1:1),
623 at 25 °C.
Entry
pH
Product
Yield
(mol%)
93
erythro/threo
QM-HHbiphenyl
3.5
4.5
5.5
6
H-HHbiphenyl
H-HHbiphenyl
H-HHbiphenyl
H-HHbiphenyl
H-HHbiphenyl
70:30
66:34
57:43
56:44
56:44
82
76
75
82
7
624
625 The erythro/threo ratio of the β–O–4 structures in DHPs
626 In the water addition experiments, the threo preference observed in the H- and
627 G-type dimeric β–O–4-products turned to erythro preference in the trimeric biphenyl ether β–
628 O–4-products, implying that an increase in the steric bulkiness of the structure etherified at
629 the β-position of QM induced the erythro preference. Dehydrogenation polymers (DHPs)
630 were prepared from p-coumaryl alcohol by enzymatic oxidative coupling reaction with
631 horseradish peroxidase and hydrogen peroxide, and the erythro/threo ratios of the β–O–4
632 structures in the DHP were analyzed by ozonation method. We attempted to investigate the
633 size effect of the structure attached to the β-etherified ring of QM on the stereopreference in
634 the water addition reaction.
635
We focused on the structural differences between two types of DHPs prepared by
636 the gradual addition of a monomer to the dehydrogenation system (Zutropfverfahren), and the
637 one-time addition of the monomer (Zulaufverhahren).38 Unless otherwise noted, the DHPs
638 were prepared by the gradual addition of the monomer. The DHP prepared by the one-time
639 addition of the monomer was named “Zulauf-DHP,” in the present study (see the footnote of
640 Table 3 for the details). The yield of ozonation products (E + T) obtained from the G-type
641 DHP, prepared from coniferyl alcohol at pH 6 (G-DHP-6), was 977 μmol/g-DHP (Table 3),
642 which was not significantly lower than those obtained from softwood lignins20 composed
643 mostly of G-unit in the previous study (E + T for six softwood species, 1180–1380
644 μmol/g-lignin, on the Klason lignin content basis). This result suggested that the β–O–4
ACS Paragon Plus Environment

Page 27 of 34
Journal of Agricultural and Food Chemistry
645 content in the G-DHP-6 was similar to those in softwood lignins. The erythro/threo ratio of
646 the G-DHP-6 was 55:45, which is closer to the ratios reported for softwood lignins (1:1)20,22,41
647 than it is to that for H-DHPs.
648
649 Table 3. The erythro/threo ratioa of β–O–4-structures in dehydrogenation polymers (DHPs)b
650 prepared from p-coumaryl or coniferyl alcohol in peroxidase-H2O2 system and the molecular
651 weight distributionsc of the DHPs.
pHb
Product name
(DHP)c
Mw/Mn
Ozonation
Mnd Mwd Mpd
Yield
Entry
E + T
E/T
(wt%)
(µmol/g)
p-Coumaryl 3.5 H-DHP-3.5
66
98
99
100
101
87
1642 4715 1866
2022 5959 2437
2780 11099 2455
NAd NAd NAd
NAd NAd NAd
1278 2250 1655
NAd NAd NAd
1507 2460 1839
2.9
3.0
288
313
346
253
265
94
37:63
37:63
36:64
36:64
38:62
32:68
55:45
65:35
alcohol
4.5 H-DHP-4.5
5
6
7
6
6
6
H-DHP-5
H-DHP-6
H-DHP-7
H-Zulauf-DHP-6
G-DHP-6
4.0
NAd
NAd
1.8
NAd
1.6
Coniferyl
alcohol
76
60
977
613
G-Zulauf-DHP-6
652 aThe erythro/threo ratio of DHP (E/T) based on the yields of erythronic acid (E) and threonic
653 acid (T), obtained by ozonation method.29 bDHP was prepared in dioxane-water systems
654 buffered at pH 3.5, 4.5, 5, 6, or 7. DHPs were prepared by the gradual addition of the
c
655 monomer to the dehydrogenation system (Zutropf method) unless otherwise noted. The DHP
656 prepared by the one-time addition of the monomer (Zulauf method) was named “Zulauf-DHP.”
657 The letters H and G are used to respectively designate the DHPs prepared from p-coumaryl
658 and coniferyl alcohols. Numbers 3.5, 4.5, 5, 6, and 7 represent the pH conditions during the
d
659 preparation of the DHPs. Molecular weight distributions of the DHP (acetylated) measured
660 by size exclusion chromatography. Mn and Mw (weight average MWs), Mp (peak MW), and
661 Mw/Mn (polydispersity) were calculated using a calibration curve of polystyrene standards.
662 NA: Not available due to the limited solubility of the DHP (acetylated).
663
664
Based on the ozonation products yields, the H-type DHP prepared from p-coumaryl
665 alcohol at pH 6 (H-DHP-6), contains a much lower amount of β-O-4-structures than that
666 contained in the G-type DHP prepared by the same method (G-DHP-6). A similar trend in the
667 ozonation-products yields has been found between H- and G-type lignins. Namely, the yields
668 of ozonation products were significantly lower in the compression wood containing a
669 substantial amount of H-units (E+T, 711–930 μmol/g-lignin) than in the opposite wood
670 containing a negligible amount of H-unit (912–1033 μmol/g-lignin).26 The erythro/threo ratio
671 of the H-DHP determined by ozonation method suggested that the threo-preferential water
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 28 of 34
672 addition occurred during the formation of the β–O–4-structures, similarly to the results during
673 the formation of dimeric β–O–4-products, HH, by water addition to model QMs. However,
674 the erythro/threo ratios of H-DHPs were higher than that of HH (the erythro proportion:
675 32-38% for all H-DHPs and H-Zulauf-DHP, and 19-30% for HH obtained in the present and
676 previous studies8).
677
A significant difference in the erythro/threo ratio was also found between the
678 H-DHP prepared by the dropwise addition method and H-Zulauf-DHP prepared by the
679 one-time addition method. Although the H-DHPs (dropwise addition method) were prepared
680 under different pH conditions in the pH range of 3.5–7, the variation in the erythro/threo ratio
681 was small (erythro proportion: 36–38%). Τhese erythro/threo ratios were higher than that
682 obtained for H-Zulauf-DHP-6 (erythro proportion: 32%). The molecular weight distributions
683 of acetylated samples of DHPs were measured by SEC, and it was indicated that the H-DHPs
684 prepared at pH of 3.5, 4.5, and 5 (H-DHP-3.5, -4.5, and -5) are composed of higher molecular
685 weight polymers relative to the polymer composition of the DHP prepared by one-time
686 addition (H-Zulauf-DHP-6). The Mw and Mn values of the H-DHP (in acetate form) were
687 slightly relatively high when the polymer was prepared under high pH condition (Mn and
688 Mw: pH 5 > 4.5 > 3.5). The H-DHPs prepared at pH 6 and 7 could not be analyzed by SEC
689 because their acetylated samples were hardly dissolved in THF solvent, which was used as a
690 column eluent. It was suggested that the H-DHPs prepared by gradual addition method, which
691 seem to have a higher molecular weight distribution than H-zulauf-DHP, contain β–O–4
692 structures with a relatively high erythro/threo ratio, although the difference in molecular
693 weights between the two DHPs prepared by gradual-addition and one-time addition methods,
694 under the same pH condition (H-DHP-6 and H-Zulauf-DHP-6), could not be compared. This
695 result was in line with the results of the water addition to the H-type QMs, i.e.,
696 QM-H-HHbiphenyl bearing a bulky structure in its β-etherified moiety showed a higher
697 erythro/threo ratio compared to that exhibited by simple β-p-hydroxylphenyl ether QM-HH.
698 Moreover, the erythro/threo ratio of the H-DHP prepared by the gradual-addition method was
699 closer to the values reported for compression wood lignins in comparison with the ratio of the
700 H-Zulauf-DHP. Conversely, an opposite trend in the erythro/threo ratio was observed for
701 G-type DHP, in which the G-DHP prepared by the gradual-addition method (G-DHP-6) had
702 an erythro/threo ratio lower than that of the one prepared by Zulauf method
ACS Paragon Plus Environment

Page 29 of 34
Journal of Agricultural and Food Chemistry
703 (G-zulauf-DHP-6), although their molecular weights could not be compared due to the low
704 solubility of the acetylated of G-DHP sample.
705
In the present study, it was found that solvent and pH conditions significantly
706 influenced the reaction rates and the stereo-preference of QM lignin models in the water
707 addition reaction. The reaction rates of the nonsubstituted QMs were always higher than that
708 of 3-methoxy substituted QM regardless of the reaction conditions. This result supports our
709 previous proposal: during lignin biosynthesis, p-hydroxyphenyl-type QMs derived from
710 p-coumaryl alcohol by radical-coupling reaction with a growing lignin chain, can undergo
711 rearomatization faster than guaiacyl-type QMs can. On the other hand, threo-preferential
712 formation observed for most QMs differed from the erythro preference naturally found in
713 compression wood lignins. This model experiment may not yet perfectly mimic the water
714 addition in lignin biosynthesis. However, it was found that an erythro-preferential formation
715 was able to occur even in an H-type QM. The stereoselectivity was greatly influenced by the
716 β-etherified aromatic ring structure. The erythro-preferential formation was attained by
717 introducing a biphenyl structure into the β-etherified aromatic-ring moiety.
718
Solvent and pH conditions significantly influenced the stereo-preference.
719 Interestingly, the effect of pH on the stereopreference, which was observed in the
720 dioxane-water solvent (1:1), almost disappeared upon increasing the proportion of water to
721 99%. The variation of the erythro/threo ratio between the different QMs widened as the water
722 concentration increased. The effect of solvent and also substances coexisting in the cell wall
723 may deserve more attention in understanding and reproducing the condition for lignification
724 in cell walls.
725
726 Supporting Information. 1H NMR spectra of H-HHbiphenyl and QM-H-HHbiphenyl (Figure S1),
727
pseudo-first-order reaction rate constants (kobs) for the disappearance of β–O–4-aryl ether
728 QMs (Table S1), the UV absorbance spectra of QMs (Figure S2), the erythro/threo ratios of
729 β–O–4-products obtained from QMs at pH 3.7–7 (Figure S3).
730 Corresponding Author
731 * Takuya Akiyama. * Wood Chemistry Laboratory, Department of Forest Resource
732 Chemistry, Forestry and Forest Products Research Institute, 1 Matsunosato, Tsukuba, Ibaraki
ACS Paragon Plus Environment