NMR Assignment for Diaryl Ether Structures (4-O-5 Structures) in Pine Wood Lignin

Article abstract of DOI:10.1021/acs.biomac.6b00257

A 4-O-5-tetramer lignin model compound carrying β-O-4 linkages on each of the side-chain moieties was synthesized, as well as 4-O-5-coupled dehydrodiconiferyl alcohol. By comparison with their NMR data, two cross-signals in the HSQC spectrum of pine milled wood lignin recorded in DMSO-d₆ were assigned to H2/C2 and H6/C6 correlations on the aromatic rings of 4-O-5-linked units. Although the H2/C2 correlation peak appeared in the same region as syringyl units, nitrobenzene oxidation of the pine lignin did not yield any syringyl-type product, but did release a 4-O-5-type product.

Full text of DOI:10.1021/acs.biomac.6b00257

Subscriber access provided by UNIV OF LETHBRIDGE
Article
NMR assignment for diaryl ether structures
(4–O–5 structures) in pine wood lignin
Yanding Li, Takuya Akiyama, Tomoya Yokoyama, and Yuji Matsumoto
Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.6b00257 • Publication Date (Web): 14 Apr 2016
Downloaded from http://pubs.acs.org on April 21, 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.
Biomacromolecules 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 33
Biomacromolecules
1
2
3
4
5
6
7
8
9
NMR assignment for diaryl ether structures (4–O–5
structures) in pine wood lignin
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Yanding Li,† Takuya Akiyama,*,†,‡ Tomoya Yokoyama,† Yuji Matsumoto†
†Wood Chemistry Laboratory, Department of Biomaterial Sciences, Graduate School of
Agricultural and Life Sciences, The University of Tokyo, Bunkyoꢀku, Tokyo 113ꢀ8657, Japan
‡Japan Science and Technology Agency (JST), PRESTO, Kawaguchi, Saitama 332ꢀ0012, Japan
Keywords: diaryl ether, gymnosperm, lignins, 4–O–5 linkage, plant cell wall, softwood
Abstract
A 4–O–5ꢀtetramer lignin model compound carrying β–O–4 linkages on each of the sideꢀchain
moieties was synthesized, as well as 4–O–5ꢀcoupled dehydrodiconiferyl alcohol. By comparison
with their NMR data, two crossꢀsignals in the HSQC spectrum of pine milled wood lignin
recorded in DMSOꢀd6 were assigned to H2/C2 and H6/C6 correlations on the aromatic rings of
4–O–5ꢀlinked units. Although the H2/C2 correlation peak appeared in the same region as
syringyl units, nitrobenzene oxidation of the pine lignin did not yield any syringyl type product,
but did release a 4–O–5ꢀtype product.
ACS Paragon Plus Environment
1

Biomacromolecules
Page 2 of 33
1
2
3
4
Introduction
5
6
7
8
Lignin is a natural aromatic polymer that is abundant in the secondary cell walls of woody plants
(around 20ꢀ35 wt% of wood).1 Lignins are derived from three monolignols, pꢀcoumaryl,
coniferyl, and sinapyl alcohols, that give rise to pꢀhydroxyphenyl (H), guaiacyl (G), and syringyl
(S) units during their polymerization process by radical coupling reactions followed by
rearomatization steps. Whereas softwood species contain lignins primarily composed of G units
together with much lower amount of H units, hardwood lignins additionally contain S units in
different amounts in various wood species.1 In the radical coupling step, G units can be coupled
at the 5ꢀposition on the aromatic ring to form β–5ꢀ, 5–5ꢀ and 4–O–5ꢀlinked units, none of which
can be formed solely from syringyl units. Consequently, the variation in syringyl/guaiacyl
compositions observed in different wood lignins is reflected in the composition of different
interunit linkage types in many cases.2ꢀ5
9
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The 5ꢀlinked G unit is a key to understanding the polymer structure of lignins. Lignins are
considered to be not linear but branched polymers as biphenyl (5–5ꢀlinked) structures and diaryl
ether (4–O–5ꢀlinked) structures can be found in chemical degradation products of lignin.6ꢀ11
They are recognized as candidate linkages for branchꢀpoints in a lignin polymer.12 As well as the
other major units in lignin such as those containing β–O–4, β–5, β–β linkages, all of which have
been evidenced by both chemical degradation and NMR methods, the existence of 5–5 and 4–O–
5 units also have been investigated by both analytical approaches.13ꢀ16
Among the major structural units in lignin, β–O–4, β–5, β–β linkages are readily differentiated
1
by using shortꢀrange H–13C correlation NMR spectroscopy (heteronuclear singleꢀquantum
correlation (HSQC) experiment), in which the correlation peaks from their sideꢀchain parts are
well separated from each other, especially correlation peaks at βꢀ and αꢀpositions. A large
ACS Paragon Plus Environment
2

Page 3 of 33
Biomacromolecules
1
2
3
4
5
6
7
8
9
fraction of 5–5 units also can be identified from correlation peaks in the sideꢀchain region
(aliphatic region) of the HSQC spectrum as dibenzodioxocin structures that have an eight
membered ring involving 5–5 linkage.14ꢀ15 Although the atoms involved in the 5–5 linkage itself
do not show a correlation peak in the HSQC spectrum due to the lack of protons attached directly
to the carbons at the 5ꢀpositions, dibenzodioxocins exhibit two particular correlation peaks in the
sideꢀchain region that originate from the αꢀ and βꢀpositions in the dibenzodioxocin ring.
Whereas the structural evidence has been obtained for the biphenyl structure, limited
information is available for the diaryl ether structure (4–O–5 linkage). As well as the 5–5
linkages, 4–O–5 linkages also potentially can be branchꢀpoints in lignin, and are therefore crucial
structures to understand the shape of the polymer. Although 4–O–5ꢀlinked products have been
obtained from lignin by chemical degradation methods, e.g., permanganate oxidation,9
thioacidolysis,10 and DFRC methods,11 4–O–5 units have not been fully assigned in the 2D
HSQC spectra of lignins.
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The 4–O–5 units are different from the other unit in that their correlation peaks are not
expected to appear in the aliphatic region (sideꢀchain region) of lignin HSQC spectra. However,
this structure potentially can be identified in the aromatic region if crosspeaks from aromatic
rings involved in 4–O–5 linkages show correlation peaks distinguishable from the others. This
methodology has been proposed in a detailed 2D NMR study of pine lignin by the comparison of
the chemical shifts of 4–O–5ꢀmodel compounds carrying simple sideꢀchain structures.16 In the
study, the author carefully avoided the assignment of candidate signals on the HSQC spectrum
considering the overlapping signals from possible syringyl units in the lignin and the imperfect
matching of chemical shifts between the lignin and the model compounds.
ACS Paragon Plus Environment
3

Biomacromolecules
Page 4 of 33
1
2
3
4
5
6
7
8
9
In the present study, a 4–O–5ꢀdimer model compound and a tetrameric 4–O–5ꢀlinked dimeric
model carrying β–O–4 linkages on each of the sideꢀchain moieties were synthesized for the
structural elucidation of softwood lignin. We investigated whether any crossꢀsignal in the HSQC
spectra of the softwood lignin can be assigned for 4–O–5 units based on the comparison with the
NMR data of the model compounds. The result of the assignment is discussed together with
nitrobenzene oxidation products of the lignin and the model compounds.
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental Section
General
Dioxane was distilled over Na metal. All other chemicals were purchased from Wako Pure
Chemical Industries, Ltd. (Osaka, Japan), Tokyo Chemical Industry, Co., Ltd. (Tokyo, Japan) or
SigmaꢀAldrich (Tokyo, Japan), and used without further purification. The NMR spectra of
synthesized compounds were taken on a JEOL JNMꢀA500 500 MHz spectrometer. The standard
JEOL programs of oneꢀ and twoꢀdimensional (proton, carbon, DEPTꢀ135, 1H–1H COSY, 1H–13C
1
HSQC, and H–13C HMBC) NMR experiments were used for the structural assignments of
newly synthesized compounds. Electrospray ionization (ESI) mass spectra were recorded on a
timeꢀofꢀflight mass spectrometer (JEOL, AccuTOF JMSꢀT100LC). The NMR spectra of lignin
samples were acquired by a Bruker Avance 600 MHz spectrometer fitted with a 5 mm TCI
gradient cryoprobe. The central peak of the residual solvent was used as an internal reference (δH
7.26, δC 77.0 ppm for CDCl3, δH 2.04, δC 29.8 ppm for acetoneꢀd6, δH 2.49, δC 39.5 ppm for
DMSOꢀd6). The traditional numbering system for lignins17 was followed rather than the
systematic IUPAC numbering scheme.
ACS Paragon Plus Environment
4

Page 5 of 33
Biomacromolecules
1
2
3
4
Isolation of milled wood lignins
5
6
7
8
Milled wood lignin (MWL) was isolated according to the Björkman method,18 with minor
modifications in the following procedure. Wood meal (80 mesh pass) from Japanese red pine
(Pinus densiflora Siebold & Zucc.) and birch (Betula maximowiczii Regel) were prepared from
the sapwood part of their wood stem disks by using a Wiley mill, and preꢀextracted with
ethanol/benzene (1:2, v/v) using a Soxhlet apparatus.
9
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The preꢀextracted pine wood meal (20 g for each run) was placed in a zirconium dioxide jar
(500 mL) with ZrO2 balls (10 mm diameter, 80 pieces), and ground further in a ball mill
(planetary mill; Pulverisette 5, Fritsch Japan Co., Ltd.; 300 rpm) in an environment at 4 °C for 11
h: the milling was interrupted after every 15 min for 15 min to avoid excessive heat
development. The ballꢀmilled wood (BM) was extracted twice with with dioxane/water (96:4,
v/v, 4 mL per 1 g BM) overnight at r.t. The extracts were combined, and freezeꢀdried to give
crude MWL (11.5 g, 3.8 wt% yield from BM). The crude MWL dissolved in a small amount of
dioxane/water (96:4, v/v, 2 mL per 1 g of crude MWL) was poured into 100 times the volume of
water, and then filtered through a hydrophilic PTFE membrane filter (0.2 ꢁm pore size;
Advantec Toyo Kaisha, Ltd., Japan). The residue was dissolved again in a small amount of
dioxane/water (96:4, v/v), then poured into a 10 times volume of diethyl ether to precipitate
purified pine MWL (7.9 g, 74.3 wt% yield from the crude MWL). A part of the pine MWL (400
mg) was acetylated overnight at r.t. with Ac2O (2 mL) and pyridine (6 mL).
Crude birch MWL was isolated from the preꢀextracted wood meals, and purified in the same
manner. The purified birch MWL was acetylated with Ac2O and pyridine.
Alkaline Nitrobenzene oxidation
ACS Paragon Plus Environment
5

Biomacromolecules
Page 6 of 33
1
2
Alkaline nitrobenzene oxidation (NBO)19 of pine MWL (10 mg) or model compound (5 ꢁmol)
was carried out using nitrobenzene (0.4 mL) and 2M NaOH (7 mL) at 170 ºC for 2 h as
described previously.8 The NBOꢀproducts were analyzed as trimethylsilyl derivatives by gas
chromatographyꢀmass spectroscopy (GCꢀMS, Shimadzu GC2010/PARVUM2, ICꢀ1 column)
equipped with a fusedꢀsilica capillary column (ICꢀ1, 0.25 mm i.d. × 30 m, 0.4 ꢁm thickness, GL
Science Inc) to identify the products by the comparison with the peak retention time and mass
spectra of the authentic compounds (vanillin, vanillic acid, syringaldehyde, syringic acid,
compound 2, and compound 3). The identified products were quantified by gas chromatography
with flame ionization Detector (GCꢀFID, Shimadzu GCꢀ2014) using the same column and the
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
following calibration curves: YV = 1.0957 XV + 0.0298 (R2 = 1.00) (0.55 < XV< 5.65), YVA
0.08395 XVA + 0.0035 (R2 = 0.99) (0.0045 < XVA< 0.608), Y4–O–5ꢀVVA = 0.71483 X4–O–5ꢀVVA
=
+
0.00864 (R2 = 0.99) (0.007 < X4–O–5ꢀVVA< 0.13). Where “Y” is the molar ratio of each authentic
compound to EV, and “X” is the GC area ratio of each authentic compound to EV. Initial column
temperature: 150 ºC (held for 25 min), raised at 3 ºC/min to 190 ºC, raised at 10 ºC/min to 230 ºC
(held for 20 min), raised at 3 ºC/min to 280 ºC (held for 15 min). Total running time 94 min.
Retention times: 8.2 min, vanillin (V); 10.6 min, 3ꢀethoxyꢀ4ꢀhydroxybenzaldehyde (EV, internal
standard); 15.9 min, syringaldehyde (S); 23.0 min, vanillic acid (VA); 33.9 min, syringic acid
(SA); 56.7 min, compound 2 (4–O–5ꢀdiV); 68.5 min, compound 3 (4–O–5ꢀVVA).
HSQC experiments of lignin samples
Acetylated or nonꢀacetylated milled wood lignin prepared from pine wood or birch wood (50
mg) was dissolved in 0.6 mL of CDCl3, acetoneꢀd6, or DMSOꢀd6. In the HSQC experiments,
1846 data points were acquired from 10.5 to ꢀ0.5 ppm in F2 (1H), with an acquisition time of 140
ACS Paragon Plus Environment
6

Page 7 of 33
Biomacromolecules
1
2
ms, and from 200 to 0 ppm in F1 (13C) with 512 increments, 16ꢀ48 scans, and a 1.0 s interscan
delay, with a total acquisition time of 2 h 39 min to 7 h 55 min depending on the number of scans
used. Processing the final matrix to 2 k by 1 k data points was performed by means of a 90ꢀ
degreeꢀshifted squared sineꢀbell in both F2 and F1. Correlation peaks appearing in the aliphatic
region of the HSQC spectrum of acetylated MWL (CDCl3) were assigned based on the NMR
data of acetylated lignin model compounds (CDCl3) reported in the NMR database of lignin
model compounds20 and previous studies.14, 21ꢀ23 The # refers to the library number of the NMR
database20: in the case of βꢀOꢀ4 (#3, #74, #29, #97, #98, #214, Hauteville et al.,21 and Sipilä et
al.22), βꢀβ (#109 and #123), βꢀ5 (#2005 and Li et al.23), and dibenzodioxocin (#278 and Karhunen
et al.14).
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Synthetic route of the 4–O–5 lignin model compounds. i) Ag2O; ii) DDQ; iii) Ag2O,
NaOH; iv) ethyl vinyl ether, PPTS; v) triethyl phosphonoacetate, NaH; vi) DIBALꢀH; vii) Ac2O,
Py; viii) ethyl chloroacetate, K2CO3; ix) BnBr, K2CO3; x) LDA; xi) NaBH4; xii) H2, Pd/C.
ACS Paragon Plus Environment
7

Biomacromolecules
Page 8 of 33
1
2
3
4
5
6
7
8
9
Synthesized model compounds
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As shown in Figure 1, 4–O–5ꢀcoupled dehydrodiferulic acid 5 was chemically synthesized
starting from vanillyl alcohol via 4–O–5ꢀcoupled dehydrodivanillin 2 as described previously.24
The 4–O–5ꢀdimer model 6 was prepared by the reduction of compound 5 with DIBALꢀH.25
Compound 2 was oxidized to 4–O–5ꢀcoupled dehydrovanillinꢀvanillic acid 3 using silver oxide
(I).26 The 4–O–5ꢀtetramer models 10 and 11 were synthesized by Nakatsubo’s method27 starting
from compound 2 and ethyl 2ꢀmethoxyphenoxyacetate 7 as described below. Compounds 6, 10,
and 11 were acetylated at r.t. with Ac2O and pyridine.
4–O–5-coupled dehydrovanillin-vanillic acid (3)
To a stirred solution of compound 2 (3.02 g, 10 mmol) in 2.4 M NaOH (50 mL), Ag2O (2.78 g,
12 mmol)26 was added slowly. The suspension was stirred vigorously at room temperature for 30
min. The reaction progress was monitored by TLC (hexaneꢀEtOAc, 1:2, with one drop AcOH).
The reaction mixture was filtrated off through a Buchner funnel. The filtrate was acidified with 2
M HCl and stirred for 10 min. The resulting suspension was filtered to collect the insoluble crude
product. The crude product was purified by crystallization from hexaneꢀacetone (3:1) to afford
1
compound 3 as white powder (0.31 g, 30%). HꢀNMR (DMSOꢀd6, 500 MHz) δH: 3.86 (3H, s,
B3ꢀOMe), 3.90 (3H, s, A3ꢀOMe), 6.76 (1H, d, J = 8.0 Hz, B5), 7.11 (1H, s, A6), 7.35 (1H, s,
A2), 7.50 (1H, d, J = 8.0 Hz, B6), 7.60 (1H, s, B2), and 9.73 (1H, s, AꢀCHO). 13CꢀNMR
(DMSOꢀd6, 125 MHz) δC: 55.78 (B3ꢀOMe), 56.24 (A3ꢀOMe), 107.89 (A2), 113.36 (B2), 115.62
ACS Paragon Plus Environment
8

Page 9 of 33
Biomacromolecules
1
2
3
4
5
6
(A6), 116.65 (B5), 122.87 (B6), 126.03 (B1), 127.53 (A1), 143.18 (A5), 144.37 (B4), 149.28,
149.42 and 149.52 (A4, A3 and B3), 166.87 (BꢀCOOH), and 190.93 (AꢀCHO).
7
8
9
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4–O–5-coupled dehydrodiconiferyl alcohol (6)
A 200 mL dried flask equipped with a stirrer bar was charged with compound 5 (1.33 g, 3
mmol), and sealed with a rubber cap. Anhydrous toluene (50 mL) was injected through the
septum. To the stirred solution over an ice bath at 0 ºC, 1.5 M diisobutylaluminium hydride
(DIBALꢀH)25 in toluene (18 mmol) was added slowly by using a syringe over 30 min. The
reaction mixture was stirred for additional 30 min. The reaction progress was monitored by TLC
(hexane:EtOAc, 1:3) The resulting suspension was diluted with cold EtOAc (50 mL) and crushed
ice (100 mL), and then acidified with 4 M HCl until the aqueous layer was pH<2. The organic
layer was collected, and filtered through a silica gel column to remove insoluble materials. The
aqueous layer was extracted with EtOAc (3 × 50 mL). These organic layers and filtrate above
were combined, washed with brine several times until the pH of the aqueous layer was over 5,
dried over Na2SO4, concentrated under the vacuum. The residue was purified by silica gel
chromatography with hexaneꢀEtOAc (1:3) to afford compound 6 as pale yellow oil (0.96 g,
1
90%). Mass (EI, 70eV) calculated for C29H46O6Si3 574 (TMS derivative), found 574 m/z. Hꢀ
NMR (acetoneꢀd6, 500 MHz) δH: 3.73 (0.8H, broadꢀt, J = 5.5 Hz, AγꢀOH), 3.85 (4H, s, B3ꢀOMe
and BγꢀOH), 3.89 (3H, s, A3ꢀOMe), 4.14 (2H, d, J = 5.0 Hz, Aγ), 4.22 (2H, d, J = 5.5 Hz, Bγ),
6.15 (1H, dt, J = 16.0, 5.5 Hz, Aβ), 6.34 (1H, dt, J = 16.0 Hz, 5.5 Hz, Bβ), 6.41 (1H, d, J = 16.0
Hz, Aα), 6.51 (1H, s, A6), 6.57 (1H, d, J = 16.0 Hz, Bα), 6.78 (1H, d, J = 8.0 Hz, B5), 6.87 (1H,
13
s, A2), 6.93 (1H, d, J = 8.0 Hz, B6), 7.18 (1H, s, B2), and 7.67 (0.6H, brꢀs, A4ꢀOH). CꢀNMR
(acetoneꢀd6, 125 MHz) δC: 56.24 (B3ꢀOMe), 56.55 (A3ꢀOMe), 63.20 (Aγ and Bγ), 105.95 (A2),
ACS Paragon Plus Environment
9

Biomacromolecules
Page 10 of 33
1
2
3
4
110.67 (A6), 111.41 (B2), 119.49 (B5), 119.97 (B6), 128.95 (Aβ), 129.32 (A1), 129.61 (Bα),
5
6
7
8
9
129.95 (Aα), 130.25 (Bβ), 134.32 (B1), 138.26 (A4), 145.45 (A5), 146.52 (B4), 149.75 (A3),
and 151.58 (B3). After adding 1 drop of D2O, the peak at δH: 3.73 and 7.67 disappeared. The
integration of δH: 3.85 decreased from 4H to 3H.
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
Compound 6 in DMSOꢀd6: HꢀNMR (DMSOꢀd6, 500 MHz) δH: 3.81 (3H, s, B3ꢀOMe), 3.82
(3H, s, A3ꢀOMe), 4.02 (2H, broadꢀs, Aγ), 4.10 (2H, brꢀs, Bγ), 4.73 (0.8H, brꢀs, AγꢀOH), 4.83
(0.8H, brꢀs, BγꢀOH), 6.11 (1H, dt, J = 16.0, 4.5 Hz, Aβ), 6.31 (1H, dt, J = 16.0, 4.5 Hz, Bβ), 6.34
(1H, d, J = 16.0 Hz, Aα), 6.41 (1H, s, A6), 6.49 (1H, d, J = 16.0 Hz, Bα), 6.63 (1H, d, J = 8.5
Hz, B5), 6.84 (1H, s, A2), 6.89 (1H, d, J = 8.5 Hz, B6), 7.16 (1H, s, B2), and 8.80 (0.8H, brꢀs,
A4ꢀOH). 13CꢀNMR (DMSOꢀd6, 125 MHz) δC: 55.68 (B3ꢀOMe), 56.01 (A3ꢀOMe), 61.52 and
61.54 (Aγ and Bγ), 105.27 (A2), 109.78 (A6), 110.39 (B2), 117.82 (B5), 118.97 (B6), 127.75
(A1), 128.20 (Bα), 128.46 (Aβ), 128.56 (Aα), 129.81 (Bβ), 132.65 (B1), 137.06 (A4), 144.23
(A5), 145.16 (B4), 149.08 (A3), and 149.94 (B3). After adding 1 drop of D2O, the peak at δH:
4.73, 4.83 and 8.80 disappeared. The peak shape at δH: 4.02 and 4.10 became doublet with a
coupling constant J = 4.5 Hz.
1
Acetates of compound 6 (6Ac) in CDCl3: HꢀNMR (CDCl3, 500 MHz) δH: 2.05 (3H, s, Aγꢀ
OCOMe), 2.08 (3H, s, BγꢀOCOMe), 2.22 (3H, s, A4ꢀOCOMe), 3.82 (3H, s, B3ꢀOMe), 3.83 (3H,
s, A3ꢀOMe), 4.63 (2H, dd, J = 6.5, 1.0 Hz, Aγ), 4.70 (2H, dd, J = 6.5, 1.0 Hz, Bγ), 6.09 (1H, dt, J
= 16.0, 6.5 Hz, Aβ), 6.21 (1H, dt, J = 16.0, 6.5 Hz, Bβ), 6.43 (1H, d, J = 2.0 Hz, A6), 6.46 (1H,
d, J = 16.0 Hz, Aα), 6.60 (1H, d, J = 16.0 Hz, Bα), 6.70 (1H, d, J = 2.0 Hz, A2), 6.87 (1H, d, J =
7.5 Hz, B5), 6.90 (1H, dd, J = 7.5, 1.5 Hz, B6), and 6.99 (1H, d, J = 2.0 Hz, B2). 13CꢀNMR
(CDCl3, 125 MHz) δC: 20.34 (AγꢀOCOMe), 20.92 (BγꢀOCOMe), 20.96 (A4ꢀOCOMe), 56.06
(OMe), 56.17 (OMe), 64.65 (Aγ), 64.96 (Bγ), 104.72 (A2), 108.99 (A6), 110.59 (B2), 119.86
ACS Paragon Plus Environment
10

Page 11 of 33
Biomacromolecules
1
2
3
4
5
6
7
8
9
(B6), 120.63 (B5), 122.84 (Bβ), 123.88 (Aβ), 130.24 (A4), 133.19 (B1), 133.31 (Aα), 133.67
(Bα), 134.61 (A1), 145.10 (B4), 150.28 (A5), 151.03 (B3), 152.62 (A3), 168.37 (A4ꢀOCOMe),
170.73 (AγꢀOCOMe), and 170.82 (BγꢀOCOMe).
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Acetates of compound 6 (6Ac) in acetoneꢀd6: 1HꢀNMR (acetoneꢀd6, 500 MHz) δH: 1.99 (3H, s,
AγꢀOCOMe), 2.03 (3H, s, BγꢀOCOMe), 2.18 (3H, s, A4ꢀOCOMe), 3.83 (3H, s, B3ꢀOMe), 3.87
(3H, s, A3ꢀOMe), 4.63 (2H, dd, J = 6.5, 1.0 Hz, Aγ), 4.70 (2H, dd, J = 6.5, 1.0 Hz, Bγ), 6.25
(1H, dt, J = 16.0, 6.5 Hz, Aβ), 6.35 (1H, dt, J = 16.0, 6.5 Hz, Bβ), 6.49 (1H, d, J = 2.0 Hz, A6),
6.56 (1H, d, J = 16.0 Hz, Aα), 6.69 (1H, d, J = 16.0 Hz, Bα), 6.89 (1H, d, J = 9.0 Hz, B5), 6.96
(1H, d, J = 1.5 Hz, A2), 7.02 (1H, dd, J = 8.5, 2.0 Hz, B6), and 7.26 (1H, d, J = 2.5 Hz, B2). 13Cꢀ
NMR (acetoneꢀd6, 125 MHz) δC: 20.18 (AγꢀOCOMe), 20.72 (BγꢀOCOMe), 20.77 (A4ꢀOCOMe),
56.31 (B3ꢀOMe) and 56.59(A3ꢀOMe), 64.93 (Aγ), 65.21 (Bγ), 105.43 (A2), 109.12 (A6), 111.83
(B2), 120.55 (B6), 121.63 (B5), 124.44 (Bβ), 125.43 (Aβ), 130.95 (A4), 133.33 (Aα), 133.76
(Bα), 134.74 (B1), 135.72 (A1), 145.67 (B4), 151.51 (A5), 152.35 (B3), 153.90 (A3), 168.38
(A4ꢀOCOMe), 170.66 (AγꢀOCOMe), and 170.73 (BγꢀOCOMe).
Benzyl ether of 4–O–5-coupled dehydrodivanillic acid (8)
To a stirred suspension of compound 2 (3.02 g, 10 mmol), K2CO3 (2.78 g, 20 mmol) in DMF (20
mL), benzyl bromide (1.3 mL) was added dropwise. The reaction mixture was kept stirring at
room temperature for 1.5 h. The progress of the reaction was monitored by TLC (Hexaneꢀ
EtOAc, 2:1). The reaction mixture was diluted with water (50 mL), and extracted with CH2Cl2 (3
× 50 mL). The combined organic layer was washed with water (20 mL), brine (20 mL), dried
over Na2SO4, and concentrated under vacuum. The residue was purified by silica gel
chromatography with hexaneꢀEtOAc (4:1) to afford compound 8 as white powder (3.60 g, 92%).
ACS Paragon Plus Environment
11

Biomacromolecules
Page 12 of 33
1
2
1HꢀNMR (acetoneꢀd6, 500MHz) δH: 3.93 (3H, s, OMe), 4.01 (3H, s, OMe), 5.16 (2H, s, PhCH2),
6.97 (1H, d, J = 9.0 Hz, B5), 7.17 (1H, s), 7.24ꢀ7.36 (5H, PhCH2), 7.45 (1H, s), 7.49 (1H, d, J =
8.0 Hz, B6), 7.58 (1H, s, B2), 9.87 (1H, s, CHO), and 9.93 (1H, s, CHO).
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compound 9
An ovenꢀdried 200 mL threeꢀnecked flask equipped with a stirrer bar was sealed with rubber
caps, equipped with a nitrogenꢀgasꢀfilled balloon. The flask was evacuated and filled with
nitrogen gas. To the lithium diisopropylamide27 (2 M LDA in THFꢀheptaneꢀethylbenzene from
Aldrich Co., 12 mmol) was added through the septum by using a syringe, and diluted with
anhydrous THF (6 mL). To the stirred suspension under a cold bath at ꢀ78 °C, a solution of ethyl
2ꢀmethoxyphenoxyacetate 7 (2.31 g, 11 mmol)27 in anhydrous THF (10 mL) was added drop
wise over a period of 1 h, and kept stirring for another 30min at ꢀ78 °C. To the suspension, a
solution of compound 8 (1.96 g, 5 mmol) in anhydrous THF (20 mL) was added through a
syringe during 1h. The reaction mixture was kept stirring for another 2 h at this temperature. The
progress of the reaction was monitored by TLC (CH2Cl2ꢀEtOAc, 4:1). Water (60 mL) was added
dropwise to the reaction mixture, and organic layer was collected. The aqueous layer was
extracted with EtOAc (3 × 50 mL). All organic layers were combined, washed with 4 M HCl (20
mL), water (60 mL), brine (60 mL), dried over Na2SO4, and then concentrated under vacuum.
The residue was purified by silica gel chromatography with hexaneꢀEtOAc (3:2) at multiple
times to afford two fractions composed of the mixture of diastereomers of compound 9 (major
fraction 2.8 g, minor fraction 0.9 g, total 91% yield). Each fraction contained two diastereomers
of the theoretically possible 8 diastereomers for compound 9 (two diastereomers for each
erythroꢀerythro, threoꢀerythro, threoꢀerythro, and threoꢀthreo isomers). The two diastereomers
ACS Paragon Plus Environment
12

Page 13 of 33
Biomacromolecules
1
2
1
3
4
5
6
7
8
9
in each fraction exhibited almost identical peaks in the HꢀNMR spectrum, but some carbon
chemical shifts of the isomers’ peaks were slightly different in the 13CꢀNMR spectrum.
Compound 9 (major isomers): 1HꢀNMR (acetoneꢀd6, 500 MHz) δH: 1.13~1.20 (6H, m, Aγꢀ and
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
BγꢀCOOCH2CH3), 3.70/3.71 (3H, s, OMe of two isomers), 3.76 (3H, s, OMe), 3.80 (3H, s, B3ꢀ
OMe), 3.89 (3H, s, A3ꢀOMe), 4.02~4.17 (4H, m, Aγꢀ and BγꢀCOOCH2CH3), 4.61 (1H, dd, J =
6.5, 1.5 Hz, Aβ), 4.71 (1H, dd, J = 7.0, 1.0 Hz, Bβ), 4.87 (0.8H, d, J = 5.0 Hz, AαꢀOH), 4.90
(0.8H, d, J = 4.5 Hz, BαꢀOH), 4.97 (1H, dd, J = 6.5, 5.0 Hz, Aα), 5.03 (2H, s, PhCH2), 5.10 (1H,
dd, J = 7.0, 4.5 Hz, Bα), 6.61/6.62 (1H, d, J = 1.5 Hz, A6), 6.77 (1H, m, B5), 6.73~6.81 and
6.91~6.94 (4H+4H, m, Cꢀ and Dꢀrings), 7.01 (1H, s, B6), 7.04 (1H, brꢀs, A2), 7.03~7.29 (3H, m,
13
PhCH2), 7.34/7.35 (1H, d, J = 1.5 Hz, B2), and 7.43~7.44 (2H, m, PhCH2). CꢀNMR (acetoneꢀ
d6, 125 MHz) δC: 14.46 (COOCH2CH3) and 14.50 (COOCH2CH3), 56.24 (OMe), 56.28 (OMe),
56.37 (B3ꢀOMe), 56.47 (A3ꢀOMe), 61.28 (Aγꢀ and BγꢀCOOCH2CH3), 74.60 (Aα and Bα),
75.30 (PhCH2), 83.65 and 83.68 (Aβ and Bβ), 107.47 (A2), 110.68/110.72 (A6 of two isomers),
113.28/113.31 (B2), 113.82 and 113.89 (C2 and D2), 117.54, 117.58, 117.60 and 117.63 (Cꢀ or
Dꢀring), 119.44/119.47 (B5), 120.44/120.45 (B6), 121.52, 121.61, 123.61 and 123.65 (Cꢀ or Dꢀ
ring), 128.32, 128.82 and 128.91 (PhCH2), 137.91 (A1), 138.29/138.31 (B1), 139.04 (A4),
139.23 (PhCH2), 146.25/146.27 (B4), 148.26 and 148.40 (C4 and D4), 151.26 and 151.29 (C3
and D3), 151.32 (B3), 151.34 (A5), 154.67 (A3), 170.14 (AγꢀCOOEt), and 170.26 (BγꢀCOOEt).
MS (ESI) calculated for C45H48O14Na [M+Na]+ 835.29, found 835.29.
1
Compound 9' (minor isomers): HꢀNMR (acetoneꢀd6, 500 MHz) δH: 1.07 (3H, t, J = 7.0 Hz,
AγꢀCOOCH2CH3), 1.18 (3H, t, J = 6.5 Hz, BγꢀCOOCH2CH3), 3.74 (3H, s, OMe), 3.76 (3H, s,
OMe), 3.81 (3H, s, B3ꢀOMe), 3.87 (3H, s, A3ꢀOMe), 4.01 (2H, m, AγꢀCOOCH2CH3), 4.14 (2H,
q, J = 7.0 Hz, BγꢀCOOCH2CH3), 4.64 (1H, m, Aβ), 4.71 (1H, m, Bβ), 4.80 (0.3H, d, J = 5.0 Hz,
ACS Paragon Plus Environment
13

Biomacromolecules
Page 14 of 33
1
2
3
4
AαꢀOH), 4.92 (0.8H, d, J = 5.5 Hz, BαꢀOH), 4.99 (1H, d, J = 5.5 Hz, Aα), 5.02 (2H, s, PhCH2),
5
6
7
8
9
5.08 (1H, d, J = 7.0 Hz, Bα), 6.57 (1H, d, J = 2 Hz, A6), 6.75 (1H, d, J = 8.0 Hz, B5), 6.78~6.83
and 6.90~6.95 (4H+4H, m, Cꢀ and Dꢀrings), 6.98 (1H, d, J = 2 Hz, A2), 7.01 (1H, dd, J = 8.0, 1.5
Hz, B6), 7.21~7.29 (3H, m, PhCH2), 7.34 (1H, d, J = 1.5 Hz,, B2), and 7.41~7.42 (2H, m,
PhCH2). 13CꢀNMR (acetoneꢀd6, 125 MHz) δC: 14.38 (AγꢀCOOCH2CH3), 14.50 (Bγꢀ
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
COOCH2CH3), 56.02 (OMe), 56.31 (OMe), 56.34 (B3ꢀOMe), 56.49 (A3ꢀOMe), 61.27 (Aγꢀ
COOCH2CH3), 61.31 (BγꢀCOOCH2CH3), 74.40/74.56 (Aα of two isomers), 75.12/75.26 (Bα),
75.27 (PhCH2), 83.64 (Aβ), 83.68 (Bβ), 107.29 (A2), 110.62 (A6), 113.22 (B2), 113.85 (C2 and
D2), 117.48, 117.51 and 117.71 (Cꢀ or Dꢀring), 119.33 (B5), 120.41 (B6), 121.58, 123.57 and
123.62 (Cꢀ or Dꢀring), 128.34, 128.82 and 128.92 (PhCH2), 136.74 (A1), 138.27 (B1), 139.09
(A4), 139.15 (PhCH2), 146.21 (B4), 148.37 (D4), 148.65 (C4), 151.12 and 151.25 (C3 and D3),
151.25 (B3), 151.37 (A5), 154.75 (A3), 169.69 (AγꢀCOOEt), and 170.25 (BγꢀCOOEt). MS (ESI)
calculated for C45H48O14Na [M+Na]+ 835.29, found 835.32.
Benzylated 4–O–5-tetramer model (10)
To a stirred solution of major isomers of compound 9 (2.5 g, 3.01 mmol) in THF (30 mL) under
an iced bath at 0 ºC, excess of NaBH4 (1.13 g, 30 mmol, 10 eq.)28 was added slowly over 5 min.
After 55 min, the iceꢀwater bath was removed and the reaction suspension was stirred at room
temperature for 1 day. The completion of the reaction was monitored by TLC (EtOAc, 100%).
Water (30 mL) and AcOH (10 mL) was added under an iceꢀwater bath to quench the excess of
NaBH4. The reaction mixture was extracted with EtOAc (3 × 50 mL). The combined organic
layers were washed with saturated NaHCO3 (20 mL), water (30 mL), brine (30 mL), and
concentrated under vacuum. The residue was separated on silicaꢀgel chromatography with
ACS Paragon Plus Environment
14

Page 15 of 33
Biomacromolecules
1
2
3
4
hexaneꢀEtOAc (1:9) to obtain a fraction containing compound 10. The product was further
5
6
purified by preparative TLC (1 mm thickness, CHCl3ꢀEtOAc, 1:2, 5 time developments) to
7
8
9
1
afford compound 10 (0.47 g, 21%) as colorless syrup. HꢀNMR (DMSOꢀd6, 500 MHz) δH:
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.57~3.66 (4H, m, Aγ and Bγ), 3.62 (3H, s, OMe), 3.68 (3H, s, OMe), 3.70 (3H, s, B3ꢀOMe),
3.78(3H, s, A3ꢀOMe), 4.21 (1H, m, Aβ), 4.34 (1H, m, Bβ), 4.61~4.67 (3H, m, AγꢀOH, BγꢀOH
and Aα), 4.80 (1H, m, Bα), 4.88 (2H, s, PhCH2), 5.47 (1H, d, J = 5.0 Hz, AαꢀOH), 5.51 (1H, dd,
J = 5.0, 3.0 Hz, BαꢀOH), 6.41 (1H, d, J = 1.0 Hz, A6), 6.57 (1H, d, J = 8.2 Hz, B5), 6.77~6.91
(7H, m, Cꢀ and Dꢀrings), 6.83 (1H, brꢀs, A2), 6.87 (1H, m, B6), 6.99 (1H, m, Cꢀ or Dꢀring), 7.18
(1H, brꢀs, B2), and 7.25~7.34 (5H, m, PhCH2). 13CꢀNMR (DMSOꢀd6, 125 MHz) δC: 55.48
(OMe), 55.52 (OMe), 55.60 (OMe), 55.88 (OMe), 60.04 (Bγ), 60.11 (Aγ), 71.52 (Aα and Bα),
74.21 (PhCH2), 83.51 and 83.56 (Aβ and Bβ), 106.52 (A2), 109.30 (A6), 112.23/112.25 (B2 of
two isomers), 112.58 (C2), 112.68 (D2), 115.94, 115.99 and 116.04 (Cꢀ or Dꢀring),
118.02/118.04 (B5), 119.49 (B6), 120.58, 120.69, 121.08, 121.10, and 121.16 (Cꢀ and Dꢀrings),
127.71, 127.99 and 128.08 (PhCH2), 136.80/136.82 (A4), 137.69 (PhCH2), 138.46 (A1),
138.49/138.52 (B1), 144.04 (B4), 147.92 (C4), 148.02 (D4), 149.57 (A5), 149.64/149.66 (B3),
149.69 and 149.79 (C3 and D3), and 152.98 (A3). MS (ESI) calculated for C41H44O12Na
[M+Na]+ 751.27, found 751.30.
1
Acetates of compound 10 (10Ac) in CDCl3: HꢀNMR (CDCl3, 500 MHz) δH: 1.98 (3H, s,
OCOMe), 2.00 (3H, s, OCOMe), 2.01/2.02 (3H, s, AαꢀOCOMe of two isomers), 2.06/2.07 (3H,
s, BαꢀOCOMe), 3.70/3.71 (3H, s, OMe), 3.75 (3H, s, OMe), 3.76 (3H, s, B3ꢀOMe), 3.82 (3H, s,
A3ꢀOMe), 4.15~4.21 (2H, m, Aγ1 and Bγ1), 4.37~4.45 (2H, m, Aγ2 and Bγ2), 4.56 (1H, m, Aβ),
4.65 (1H, m, Bβ), 5.00 (2H, s, PhCH2), 5.93 (1H, d, J = 6.0 Hz, Aα), 6.03 (1H, d, J = 5.0 Hz,
Bα), 6.58 (1H, d, J = 2.0 Hz, A6), 6.62 (1H, m, B5), 6.72~6.85 (6H, m, Cꢀ and Dꢀrings), 6.80
ACS Paragon Plus Environment
15

Biomacromolecules
Page 16 of 33
1
2
3
4
(1H, brꢀs, A2), 6.83 (1H, m, B6), 6.92~6.97 (2H, m, Cꢀ and/or D ring), 7.04 (1H, d, J = 1.5 Hz,
5
6
13
B2), and 7.17~7.30 (5H, m, PhCH2). CꢀNMR (CDCl3, 125 MHz) δC: 20.74/20.76 (OCOMe),
7
8
9
20.90 (OCOMe), 20.93 (OCOMe), 21.02 (OCOMe), 55.64 (OMe), 55.71 (OMe), 55.90 (OMe),
56.14 (OMe), 62.62 and 62.69 (Aγ and Bγ), 73.65/73.70 (Aα of two isomers), 73.86 (Bα), 74.97
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(PhCH2), 80.15 and 80.19/80.21 (Aβ and Bβ), 107.33/107.44 (A2), 111.20/111.32 (A6), 112.01
(B2), 112.45 (C2), 112.48 (D2), 117.55/117.69 (B5), 119.28 and 119.34 (C ring or D ring),
119.82 (B6), 120.85, 120.87, 121.92, 123.46 and 123.51 (Cꢀ and Dꢀring), 127.65, 128.03, 128.06
and 128.14 (PhCH2), 131.84/131.90 (B1), 132.46/132.51 (A1), 137.51 and 137.54 (PhCH2),
139.05/139.10 (A4), 146.21/146.30 (B4), 147.14 (C4 and D4), 149.61/149.68 (A5),
149.76/149.83 (B3), 150.92 and 151.03/151.05 (C3 and D3), 153.92 (A3), 169.47/169.51 and
169.55 (Aαꢀ and BαꢀOCOMe), 170.71/170.73 and 170.74 (Aγꢀ and BγꢀOCOMe).
1
Acetates of compound 10 (10Ac) in acetoneꢀd6: HꢀNMR (acetoneꢀd6, 500 MHz) δH:
1.912/1.914 (3H, s, OCOMe of two isomers), 1.947/1.949 (3H, s, OCOMe), 1.98/1.99 (3H, s,
AαꢀOCOMe), 2.07 (3H, s, BαꢀOCOMe), 3.72/3.73 (3H, s, OMe), 3.79 (3H, s, B3ꢀOMe), 3.81
(3H, s, OMe), 3.89 (3H, s, A3ꢀOMe), 4.15~4.22 (2H, m, Aγ1 and Bγ1), 4.31~4.44 (2H, m, Aγ2
and Bγ2), 4.74 (1H, m, Aβ), 4.85 (1H, m, Bβ), 5.00 (2H, s, PhCH2), 5.93 (1H, d, J = 5.5 Hz, Aα),
6.03 (1H, d, J = 5.0 Hz, Bα), 6.56 (1H, d, J = 2.0 Hz, A6), 6.76 (1H, dd, J = 8.0, 1.5 Hz, B5),
6.80~6.87 (2H, m, Cꢀ and/or Dꢀring), 6.91~7.03 (6H, m, Cꢀ and Dꢀring), 6.96 (1H, brꢀs, A2),
6.98 (1H, m, B6), 7.22~7.29 (3H, m, PhCH2), 7.26 (1H, m, B2), and 7.37~7.39 (2H, m, PhCH2).
13CꢀNMR (acetoneꢀd6, 125 MHz) δC: 20.63 (OCOMe), 20.64 (OCOMe), 20.79/20.81 (OCOMe),
20.90 (OCOMe), 56.11 (OMe), 56.18 (OMe), 56.30 (OMe), 56.56 (A3ꢀOMe), 63.12/63.14 and
63.17 (Aγ and Bγ), 74.57 (Aα), 74.67 (Bα), 75.27 (PhCH2), 80.29 and 80.32 (Aβ and Bβ),
107.67 (A2), 110.89 (A6), 113.28 (B2), 113.68 and 113.74 (C2 and D2), 119.55 (B5), 119.67,
ACS Paragon Plus Environment
16

Page 17 of 33
Biomacromolecules
1
2
3
4
5
6
7
8
9
119.74 and 119.79 (Cꢀ and/or Dꢀring), 120.79/120.82 (B6 of two isomers), 121.58, 121.62,
123.97, 123.99 and 124.03 (Cꢀ and Dꢀrings), 128.39 and 128.84 (PhCH2), 133.65 (A1), 133.93
(B1), 139.00 (PhCH2), 139.34 (A4), 146.51/146.53 (B4), 148.30 and 148.31 (C4 and D4),
151.23 (A5), 151.42 (B3), 151.88 and 151.99/152.02 (C3 and D3), 154.99 (A3), 169.80 and
169.82 (AαꢀOCOMe), 169.92 (BαꢀOCOMe), 170.72 (OCOMe), and 170.75 (OCOMe).
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4–O–5-tetramer model (11)
A stirred suspension of compound 10 (350 mg, 0.45 mmol) and Pd/C (35 mg) in THF (20 mL)
was fitted with a hydrogenꢀgasꢀfilled balloon. The reaction mixture was kept stirring overnight at
room temperature. The completion of the reaction was monitored by TLC (EtOAc, 100%). After
the reaction, Pd/C was filtered off through a glass filter with an aid of celite, and washed with
acetone (20 mL). Filtrate was concentrated under vacuum to afford compound 11 as white
powder (308 mg, 99%). 1HꢀNMR (DMSOꢀd6, 500 MHz) δH: 3.58~3.65 (4H, m, Aγ and Bγ), 3.62
(3H, s, OMe), 3.69 (3H, s, OMe), 3.73 (3H, s, B3ꢀOMe), 3.76 (3H, s, A3ꢀOMe), 4.20 (1H, dt, J =
5.2, 4.7 Hz, Aβ), 4.30 (1H, m, Bβ), 4.56 (1H, brꢀt, J = 5.0 Hz, AγꢀOH), 4.61 (1H, dd, J = 5.0, 4.5
Hz, Aα), 4.64 (1H, brꢀt, J = 5.0 Hz, BγꢀOH), 4.77 (1H, dd, J = 5.0, 4.5 Hz, Bα), 5.35 (1H, d, J =
4.5 Hz, AαꢀOH), 5.46 (1H, dd, J = 4.5, 2.5 Hz, BαꢀOH), 6.42 (1H, brꢀs, A6), 6.45 (1H, dd, J =
8.0, 1.0 Hz, B5), 6.76 (1H, d, J = 7.5 Hz, Cꢀ or Dꢀring), 6.79~6.83 (2H, m, Cꢀ and/or Dꢀring),
6.81 (1H, brꢀs, A2), 6.82 (1H, m, B6), 6.84~6.88 (3H, m, Cꢀ and/or D ring), 6.91 (1H, dd, J =
8.0, 2.0 Hz, Cꢀ or Dꢀring), 6.98 (1H, ddd, J = 8.0, 4.0, 2.0 Hz, Cꢀ or Dꢀring), 7.14 (1H, brꢀs, B2),
and 8.51 (0.8H, s, A4ꢀOH). 13CꢀNMR (DMSOꢀd6, 125 MHz) δC: 55.48 (OMe), 55.63 (OMe),
55.83 (OMe), 55.87 (OMe), 60.05 (Bγ), 60.21 (Aγ), 71.55 (Aα and Bα), 83.58 (Aβ), 83.63 (Bβ),
106.53 (A2), 110.50/110.52 (A6 of two isomers), 112.07/112.09 (B2), 112.61 and 112.71 (C2
ACS Paragon Plus Environment
17

Biomacromolecules
Page 18 of 33
1
2
3
4
5
6
7
8
9
and D2), 115.80, 115.83 and 116.00 (Cꢀ and/or Dꢀring), 116.63/116.67 (B5), 119.32 (B6),
120.59, 120.69, 120.98 and 120.12 (Cꢀ and/or Dꢀring), 132.80 (A1), 136.43/136.45 (A4),
137.38/137.45 (B1), 143.54/143.56 (A5), 144.94 (B4), 148.03 (C4 and D4), 148.27 (A3),
149.12/149.15 (B3), 149.68 and 149.80 (C3 and D3). MS (ESI) calculated for C34H38O12Na
[M+Na]+ 661.23, found 661.26.
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Acetates of compound 11 (11Ac) in CDCl3: 1HꢀNMR (CDCl3, 500 MHz) δH: 1.96/1.97 (3H, s,
OCOMe of two isomers), 1.98/2.00 (3H, s, AαꢀOCOMe), 2.01 (3H, s, OCOMe), 2.06/2.07 (3H,
s, BαꢀOCOMe), 2.15/2.16 (3H, s, A4ꢀOCOMe), 3.68/3.69 (3H, s, OMe), 3.77~3.78 (6H, m, 2 ×
OMe), 3.80/3.81 (3H, s, A3ꢀOMe), 4.15~4.20 (2H, m, Aγ1 and Bγ1), 4.36~4.44 (2H, m, Aγ2 and
Bγ2), 4.53~4.55 (1H, m, Aβ), 4.66 (1H, dt, J = 5.2, 4.7 Hz, Bβ), 5.94 (1H, d, J = 5.0 Hz, Aα),
6.02 (1H, d, J = 5.0 Hz, Bα), 6.50/6.51 (1H, d, J = 1.8 Hz, A6), 6.72~6.74 (1H, m, Cꢀ or Dꢀring),
6.75 (1H, m, B5), 6.78~6.87 (5H, m, Cꢀ and Dꢀrings), 6.81 (1H, m, A2), 6.86 (1H, m, B6),
13
6.93~6.99 (2H, m, Cꢀ and/or Dꢀring), and 7.05/7.06 (1H, d, J = 1.5 Hz, B2). CꢀNMR (CDCl3,
125 MHz) δC: 20.26 (A4ꢀOCOMe), 20.70/20.71(OCOMe of two isomers), 20.75 (OCOMe),
20.86/20.90 (AαꢀOCOMe), 21.01 (BαꢀOCOMe), 55.65 (OMe), 55.74 (OMe), 56.04 (B3ꢀOMe),
56.24 (A3ꢀOMe), 62.55 and 62.58 (Aγ and Bγ), 73.59/73.63 (Aα), 73.83 (Bα), 80.06,
80.15/80.18 (Aβ and Bβ), 106.37/106.45 (A2), 109.91/110.03 (A6), 112.31/112.34 (B2), 112.51
and 112.54 (C2 and D2), 119.21, 119.31, 119.46 and 119.54 (Cꢀ and/or Dꢀring), 119.59 (B5),
119.89/119.93 (B6), 120.93, 120.95, 123.51, 123.56 and 123.61 (Cꢀ and Dꢀrings), 130.49/130.54
(A4), 132.96/132.98 (B1), 135.08/135.12 (A1), 145.26/145.33 (B4), 147.09 and 147.12 (C4 and
D4), 149.52/149.57 (A5), 150.43/150.47 (B3), 151.01 and 151.03/151.06 (C3 and D3), 152.50
(A3), 168.15 (A4ꢀOCOMe), 169.34 (AαꢀOCOMe), 169.56 (BαꢀOCOMe), and 170.72 (Aγꢀ
OCOMe and BγꢀOCOMe).
ACS Paragon Plus Environment
18

Page 19 of 33
Biomacromolecules
1
2
Acetates of compound 11 (11Ac) in acetoneꢀd6: 1HꢀNMR (acetoneꢀd6, 500 MHz) δH: 1.89/1.90
(3H, s, AγꢀOCOMe of two isomers), 1.95/1.96 (3H, s, BγꢀOCOMe), 1.98/1.99 (3H, s, Aαꢀ
OCOMe), 2.07 (3H, s, BαꢀOCOMe), 2.14 (3H, s, A4ꢀOCOMe), 3.71/3.72 (3H, s, OMe),
3.78/3.79 (3H, s, B3ꢀOMe), 3.82 (3H, s, OMe), 3.85 (3H, s, A3ꢀOMe), 4.16 (1H, dd, J = 12.0,
1.5 Hz, Aγ), 4.21 (1H, dd, J = 12.0, 3.0 Hz, Bγ), 4.33 (1H, dd, J = 12.0, 1.5 Hz, Aγ), 4.39 (1H,
dd, J = 12.0, 3.0 Hz, Bγ), 4.72 (1H, m, Aβ), 4.85 (1H, m, Bβ), 5.93 (1H, d, J = 5.0 Hz, Aα), 6.06
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(1H, d, J = 5.0 Hz, Bα), 6.52/6.53 (1H, d, J = 2.0, A6), 6.80~6.88 (2H, m, Cꢀ and/or Dꢀring),
6.84 (1H, dd, J = 8.3, 2.0 Hz, B5), 6.90~6.92 (1H, m, Cꢀ or Dꢀring), 6.94~7.04 (5H, m, Cꢀ and
Dꢀrings), 6.97 (1H, s, A2), 6.98~7.02 (2H, m, Cꢀ and/or Dꢀring), 7.01 (1H, m, B6), and 7.27 (1H,
brꢀs, B2). 13CꢀNMR (acetoneꢀd6, 125 MHz) δC: 20.14 (A4ꢀOCOMe), 20.60 (OCOMe), 20.64
(OCOMe), 20.75/20.77 (AαꢀOCOMe of two isomers), 20.88 (BαꢀOCOMe), 56.12 (OMe), 56.20
(OMe), 56.33 (OMe), 56.62 (OMe), 62.98 (Aγ), 63.10/63.12 (Bγ), 74.53 (Aα), 74.61 (Bα), 80.26
and 80.29 (Aβ and Bβ), 106.69 (A2), 109.82 (A6), 113.41/113.42 (B2), 113.71 and
113.75/113.77 (C2 and D2), 119.59, 119.71 and 119.98 (Cꢀ and/or Dꢀring), 120.84 and 120.89
(B5 and B6), 123.99, 124.04 and 124.10 (Cꢀ and/or Dꢀring), 130.87 (A4), 134.86 (B1), 136.29
(A1), 145.60 (B4), 148.21 and 148.29 (C4 and D4), 150.84 (A5), 151.81 (B3), 151.94 and
151.99 (C3 and D3), 153.60 (A3), 168.34 (A4ꢀOCOMe), 169.79/169.82 (AαꢀOCOMe), 169.95
(BαꢀOCOMe), 170.74 (OCOMe), and 170.77 (OCOMe).
ACS Paragon Plus Environment
19

Biomacromolecules
Page 20 of 33
1
2
3
4
Results and Discussion
5
6
7
8
As model compounds for the 4–O–5 structure in lignin, three types of diaryl ethers were
synthesized: 4–O–5ꢀcoupled dehydrodiconiferyl alcohol (6) that is a model for a coniferyl
alcohol endꢀgroup of lignin, nonꢀphenolic tetramer model (10) for 4–O–5 structures at the
internal unit of lignin, and phenolic tetramer (11) for 4–O–5 structures at the phenolic end of
lignin (Figure 1). As well as pine lignin sample (MWL), these model compounds were examined
for a chemical degradation method and NMR spectroscopy, and their results were compared for
the analysis of 4–O–5 structures in lignin.
9
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Nitrobenzene oxidation products released from 4–O–5 structures
Among the 4–O–5 models, compound 6 and 11 in Figure 1 were subjected to nitrobenzene
oxidation method (NBO) to investigate whether this analysis method is applicable to the 4–O–5
structures in lignin or not. Unlike β–O–4 and 5–5 structures that give simple benzaldehydes,
vanillin and 5–5ꢀcoupled dehydrodivanillin, respectively,8 both model compounds 6 and 11 did
not yield 4–O–5ꢀcoupled dehydrodivanillin (2). Instead, 4–O–5ꢀcoupled dehydrovanillinꢀvanillic
acid (3) was obtained as their degradation product accompanied by larger amounts of vanillin
and vanillic acid (Table 1). Compound 2 and 3 were subjected to the NBO method to check their
stability under the reaction condition.
Compound 3 remained fairy stable under the harsh oxidation condition; 74% of this compound
was recovered (Table 1). On the other hand, compound 2 was not recovered under the same
condition, and degraded into compound 3, vanillin and vanillic acid. This result implied the
possibility that a part of vanillin and vanillic acid obtained from 4–O–5 model compounds 6 and
11 were formed via compound 2. The low yield of 4–O–5ꢀtype products from the model
ACS Paragon Plus Environment
20

Page 21 of 33
Biomacromolecules
1
2
3
4
5
6
7
8
9
compounds 6 and 11 limited the opportunity for the quantification of 4–O–5 structures in lignin
by the NBO method. However, it was suggested that compound 3 released from the 4–O–5 units
by the NBO method is useful for investigating the existence of 4–O–5 structures in a lignin
sample.
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The NBO of pine MWL released a small amount of compound 3, suggesting that at least
0.56% of aromatic ring in the pine lignin was detected as 4–O–5 structure based on the
calculation that molecular weight of a phenylpropanoid unit is 200 although the actual 4–O–5
content in the MWL must be higher than this value considering the low yield of compound 3
obtained from the model compounds 6 and 11 (Table 1). In addition, neither syringaldehyde nor
syringic acid was found in the NBO products of the pine MWL at the detectable level by GCꢀ
FID analysis, indicating that the pine lignin sample does not contain syringyl units, and that there
was no accidental contamination by hardwood meals during the grinding process to make wood
meal from pine wood chips using a Wileyꢀmill.
Table 1. Yields of nitrobenzene oxidation products from 4–O–5 lignin model compounds and
pine MWL.
Sample
NBO products from model compounds (mol %) and pine MWL
(ꢁmol gꢀ1)
4–O–5ꢀ
4–O–5ꢀ
V
VA
S
SA
diV 2
VVA 3
4–O–5ꢀdiV 2
4–O–5ꢀVVA 3
4–O–5ꢀdimer 6
n.d.
n.d.
n.d.
18.1 ± 1.7 56.2 ± 5.8 9.9 ± 1.5
74.0 ± 0.2 1.4 ± 0.1 1.0 ± 0.1
ꢀꢀꢀ
ꢀꢀꢀ
ꢀꢀꢀ
ꢀꢀꢀ
6.7 ± 2.8
5.3 ± 1.1
0.56a
36.4 ± 4.4 10.8 ± 3.1 ꢀꢀꢀ
17.0 ± 1.6 19.0 ± 2.7 ꢀꢀꢀ
ꢀꢀꢀ
4–O–5ꢀtetramer 11 n.d.
Pine MWL n.d.
ꢀꢀꢀ
33.29 a
5.05 a
n.d.
n.d.
ACS Paragon Plus Environment
21

Biomacromolecules
Page 22 of 33
1
2
(13.95b)
(1664.3 b) (252.3 b)
3
4
5
6
7
8
a Yield was expressed in molar yield (%) based on the assumption that the molecular weight of
phenyl propane (C6ꢀC3) unit is 200. Yield was expressed in ꢁmol per one gram MWL.
Chemical structures of compound 2, 3, 6 and 11 were shown in Figure 1. V: vanillin, VA:
b
9
vanillic acid, S: syringaldehyde, SA: syringic acid.
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The HSQC spectra of pine MWL
The HSQC spectrum of the acetylated pine MWL (CDCl3 solvent) showed different correlation
peaks that can be assigned for β–O–4, β–β (resinol), β–5 (phenylcoumaran) in its aliphatic region
(Figure 2).20ꢀ23 In addition, the correlation peaks for dibenzodioxocins (5–5/4–O–β) were readily
found in the aliphatic region.14ꢀ15, 29 It was confirmed that the pine MWL sample prepared in the
present study had an interꢀunit linkage distribution typical of guaiacyl lignin.14 In the aromatic
region of the HSQC spectrum, two correlation peaks at 6.70/104.2 and 6.75/105.8 ppm were
assigned to the H2/C2 correlation of 4–O–5 structures, which were found at a low contour level
close to the noise level of the spectrum (Figure 3Bꢀ2), although these correlation peaks were not
visible at the higher contour level (Figure 3Bꢀ1) that is used for the detection of the major signals
1
13
from lignins (Figure 2, and 3A). The H and C chemical shifts of the two correlations were in
good agreement with those of the model compounds obtained as acetate forms in CDCl3 solvent
(6.70/104.72 ppm in 6Ac, 6.80/107.40 in 10Ac, and 6.81/106.37 in 11Ac) as shown in Table 2. The
H2/C2 correlations were within the aromatic region that is occupied by syringyl units when they
are present (Figure 3A and 3Bꢀ2). However, it can be safely concluded that these signals do not
originate from syringyl units but from 4–O–5ꢀstructures as the lack of syringyl units in the MWL
was indicated by the results of nitrobenzene oxidation mentioned above. Although the
correlation around 6.5/110 ppm in the MWL can possibly be assigned to H6/C6 correlation of 4–
O–5 structures based on the NMR data of lignin models, the crossꢀsignal is obscured by
ACS Paragon Plus Environment
22

Page 23 of 33
Biomacromolecules
1
2
3
4
5
6
7
8
9
correlations from other types of guaiacyl units. To obtain more reliable assignments for 4–O–5
structures, the NMR data of the model compounds were collected under different sample and
solvent conditions.
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
50
Pine MWL (acetylated, CDCl3)
B
β
C
β
Methoxyl
60
70
80
A
+ ...
γ
B
γ
A
α
C
γ
A
β
D
β
D
α
13C
C
α
90
B
α
6.0
5.5
5.0
4.5
4.0
3.5
3.0
1H
ppm
OCH3
O
HO
γ
γ
5
O
β
γ
OCH3
H3CO
5
5
OCH3
HO
Methoxyl
HO
β
O
4
α
O
O
O
β
α
OCH3
OH
α
β
α
O
γ
Unresolved,
Unassigned,
OCH3
O
O
O
OCH3
O
OCH3
saccharides etc.
OCH3
A
Β
C
D
βꢀaryl ether phenylcoumaran
resinol
dibenzodioxocin
(5ꢀ5/βꢀOꢀ4)
(βꢀΟꢀ4)
(βꢀ5)
(βꢀβ)
Figure 2. Partial shortꢀrange 1H–13C (HSQC) NMR spectrum (aliphatic regions) of milled wood
lignin isolated from pine wood trunk (acetylated, CDCl3). MWL: milled wood lignin.
ACS Paragon Plus Environment
23

Biomacromolecules
Page 24 of 33
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. HSQC spectra (aromatic regions) of milled wood lignins isolated from pine wood and
birch wood trunks (acetylated, CDCl3). A) birch MWL; B1) pine MWL and B2) pine MWL at
the lower contour level than for B1. MWL: milled wood lignin, S: syringyl region, G: guaiacyl
region.
An NMR solvent influences the chemical shift values, a feature that can possibly be
utilized for obtaining more reliable assignments of lignin spectra. Although the choice of NMR
solvent is limited to DMSOꢀd6 and some uncommon NMR solvents, e.g., DMSOꢀd6, Pyridineꢀd5,
DMSOꢀd6/Pyridineꢀd5 (4:1), or dioxaneꢀd8/D2O (9:1),9,30ꢀ36 for an underivatized lignin sample
due to its solubility limitations, acetylated lignins can be solubilized in common NMR solvents
such as CDCl3, acetoneꢀd6 or DMSOꢀd6.14,21,29,30,37,38 There is a large difference in the chemical
shift values between underivatized and acetylated lignins.39 For example, acetylation of lignin
causes a large downfield shift in the αꢀproton from β–O–4 structures (ꢂδH: +1 ppm) that results
in a wellꢀdispersed 1H NMR lignin spectrum, and makes assignments of the correlation signals in
the aliphatic regions of HMBC and HSQCꢀTOCSY spectra of lignins more reliable. The NMR
ACS Paragon Plus Environment
24

Page 25 of 33
Biomacromolecules
1
2
3
4
5
6
7
8
9
spectra of acetylated lignin samples have typically been recorded using CDCl3 as
solvent,14,21,23,29,37,38 and the same sample condition was employed in our study, as described
above (Figure 2 and 3). However, the NMR spectra of underivatized lignins are required for
investigating acylating group that are naturally present in some lignins, and are often recorded in
DMSOꢀd632,36,39 or DMSOꢀd6/pyridineꢀd5 (4:1).33,34
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The NMR sample conditions for lignins have been converging on the following three
conditions:9,14,21,23,29ꢀ32,35ꢀ38 the acetatylated lignin using CDCl3 or acetoneꢀd6, and the
underivatized lignin using DMSOꢀd6. For the assignment of lignin spectra, the NMR data from a
large number of lignin model compounds have been collected under the three conditions13,20ꢀ23,40ꢀ
42
as the chemical shift values, especially in the aliphatic region, vary under the different
conditions. Although the effects on the chemical shift values have not been investigated for the
aromatic region in detail, such an effect could be utilized for the assignment of 4–O–5 structures.
In order to evaluate the influence of sample and solvent conditions on the chemical shift values,
the NMR data of the 4–O–5 models 6, 10 and 11 were collected under the three conditions:
acetate form in CDCl3 or acetoneꢀd6, and nonꢀacetate form in DMSOꢀd6 (Table 2).
Table 2. Selected NMR chemical shifts for 4–O–5ꢀmodel compounds at H2/C2 and H6/C6
positions on the 5ꢀlinked aromatic rings (Aꢀrings) measured in three sample conditions.
Model
In DMSOꢀd6 (ppm)
Model
In CDCl3 (ppm)
H2/C2 H6/C6
In acetoneꢀd6 (ppm)
H2/C2 H6/C6
(acetates)
H2/C2
H6/C6
6.84/105.27 6.41/109.78
6.83/106.52 6.41/109.30
6.81/106.53 6.42/110.51
6.70/104.72 6.43/108.99 6.96/105.43 6.49/109.12
6.80/107.40 6.58/111.30 6.96/107.67 6.56/110.89
6.81/106.37 6.50/109.91 6.97/106.69 6.53/109.82
6
6Ac
10
10Ac
11Ac
11
a
a
ꢂδH/ꢂδC
ꢂ0.03/ꢂ1.26
ꢂ0.01/ꢂ1.21
ꢂδH/ꢂδC
ꢂ0.11/ꢂ2.68
ꢂ0.15/ꢂ2.31
ꢂ0.01/ꢂ2.24
ꢂ0.07/ꢂ1.77
ACS Paragon Plus Environment
25

Biomacromolecules
Page 26 of 33
1
2
aDifferences in the chemical shifts among H2/C2 or H6/C6 signals on the Aꢀrings of 4–O–5ꢀ
dimer 6, and 4–O–5ꢀtetramers 10 and 11. Their chemical structures were shown in Figure 1.
3
4
5
6
7
8
9
B) MWL in DMSOꢀd6
C) Acetylated MWL in CDCl3
D) Acetylated MWL in acetoneꢀd6
A) 4–O–5 model compounds
H2/C2
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
104
106
108
110
112
114
104
106
108
110
112
114
104
106
108
110
112
114
104
6
Ac
6
Ac
6
Ac
6Ac
6
6
106
108
110
112
114
11Ac
11
10Ac
11Ac
10Ac
11Ac
10Ac
11Ac
10Ac
10
1011
H6/C6
6
Ac
6
Ac
6
Ac
6Ac
10
6
11
10
6
11
11Ac
11Ac
10Ac
11Ac
11Ac
10Ac
10Ac
10Ac
DMSOꢀd6 (nonꢀacetate)
CDCl3 (acetate)
acetoneꢀd6 (acetate)
13C
13C
13C
13C
1
1
1
1
ppm
ppm
ppm
ppm
H
H
H
H
7.0
6.5
7.0
6.5
7.0
6.5
7.0
6.5
Figure 4.Selected NMR data plot of 4–O–5 lignin model compounds, and partial HSQC spectra
(δH/δC 6.1ꢀ7.2/100ꢀ115) of pine MWL measured in different sample conditions. A) H2/C2 and
H6/C6 positions on the 5ꢀlinked aromatic rings (Aꢀring) of a 4–O–5ꢀdimer 6, 4–O–5ꢀtetramers
10 and 11, and their acetates 6Ac, 10Ac, 11Ac; B) MWL in DMSOꢀd6 (H2/C2 correlation at
6.76/105.2 ppm and H6/C6 correlation at 6.40/109.4 ppm); C) acetylated MWL in CDCl3
(H2/C2 correlations at 6.70/104.2 and 6.75/105.8 ppm); D) acetylated MWL in acetoneꢀd6
(H2/C2 correlations at 6.84/105.1 and 6.95/106.7 ppm). NMR data of the models were plotted
over the lignin spectrum recorded in the same sample and solvent condition. See Figure 1 for the
chemical structures of model compounds, Table 2 for chemical shift values of model
compounds, and Figure 3 for the whole aromatic region of the lignin spectra.
Although the H2/C2 and H6/C6 positions had the similar δH values among three model
compounds 6Ac, 10Ac, and 11Ac (Table 2), they showed a wide range of C2 and C6 carbon shifts
in CDCl3 solvent (ꢂδC: 2.68 and 2.31 ppm for C2 and C6 carbons, respectively). The wide
ACS Paragon Plus Environment
26

Page 27 of 33
Biomacromolecules
1
2
3
4
5
6
7
8
9
variation in carbon chemical shifts can be utilized to distinguish the structural difference of 4–O–
5 structures in a lignin sample, such as differences in their sideꢀchain parts and/or the phenolic
end type from the others. However, as already shown in Figure 2, the pine MWL used in this
study showed a weak obscure signal for its H6/C6 correlation presumably due to the low content
and/or the structural variation in the 4–O–5 structures. Although two kinds of the H2/C2
correlations were observed under the sample and solvent condition, further interpretation of these
correlations could not be attained because of the lack of H6/C6 signal information and an
insufficient number of model compounds.
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A solvent change from CDCl3 to acetoneꢀd6 caused substantial downfield shifts to proton at H2
position of the models 6Ac, 10Ac, and 11Ac (approximately +0.2 ppm), whereas the chemical shifts
of H6 proton, and C2 and C6 carbons were similar between two solvents (Fig 4, Table 2).
Consequently, similarly to the case of CDCl3 solvent, both at 2 and 6ꢀpositions, the δH and δC
values of the three models showed narrow and wide variations in acetoneꢀd6 solvent,
respectively. Also in acetoneꢀd6 solvent, the chemical shifts of H2/C2 correlation peaks of the
acetylated pine MWL were in good agreement with those of model compounds. The chemical
shifts of H2 protons were approximately 0.2 ppm downfield from those observed in the
acetylated lignin using CDCl3 solvent, as reflected by the results of model compounds.
In contrast, in the case of DMSOꢀd6 solution, the chemical shifts at the H2/C2 and H6/C6
positions were less influenced by the structural differences among the three nonꢀacetate models
6, 10 and 11 (Table 2). Their H2/C2 and H6/C6 positions appeared at 6.83/105.90 ppm (±
0.02/0.63) and 6.41/109.90 ppm (± 0.01/0.61), respectively. The HSQC spectrum of nonꢀ
acetylated pine MWL showed not only H2/C2 correlation peak (6.76/105.2 ppm), but a clear
single correlation peak assigned to the H6/C6 position on 4–O–5 structures (6.40/109.4 ppm).
ACS Paragon Plus Environment
27

Biomacromolecules
Page 28 of 33
1
2
3
4
5
6
7
8
9
This observation suggests that the detection of the H6/C6 correlation peak in the MWL was
accomplished by suppressing the broadening of the peak attributed in part to the structural
variation of 4–O–5 structures.
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Conclusion
Diaryl ether structures (4–O–5ꢀlinked units) were found in pine MWL through the use of the
HSQC experiment. The underivatized lignin spectrum recorded in DMSOꢀd6 revealed the H2/C2
and H6/C6 correlation signals from the 5ꢀlinked moiety of the 4–O–5ꢀlinked units’ aromatic
ring; correlation peaks were assigned based on the chemical shift values of 4–O–5ꢀdimeric and
tetrameric model compounds. In the case of acetylated lignin spectra recorded in CDCl3 and
acetoneꢀd6, only the H2/C2 correlation signals can be assigned as the candidate signals for
H6/C6 correlations are obscured by correlation signals from the other types of guaiacyl units.
The assignments were possible by exploiting the absence of syringyl units in the lignin sample,
as evidenced by the complete absence of syringyl type products from nitrobenzene oxidation of
the pine lignin. In addition, by this chemical degradation method, a 4–O–5ꢀtype product was
released from both the pine lignin and from 4–O–5ꢀmodel compounds, indicating that
nitrobenzene oxidation is a useful method for the detection of 4–O–5 structures in a lignin
sample.
Corresponding Author
Takuya Akiyama
Wood Chemistry Laboratory, Department of Biomaterial Sciences, Graduate School of
Agricultural and Life Sciences, The University of Tokyo, Bunkyoꢀku, Tokyo 113ꢀ8657, Japan
ACS Paragon Plus Environment
28

Page 29 of 33
Biomacromolecules
1
2
*Eꢀmail: akiyama@woodchem.fp.a.uꢀtokyo.ac.jp
3
4
5
6
7
8
9
Author Contributions
10
11
12
13
14
15
16
17
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript. These authors contributed equally.
†
Acknowledgments
This work was supported by the PRESTO program from the Japan Science and Technology
Agency (JST). The authors would like to thank Dr. Y. Hori, Technology Advancement Center,
the University of Tokyo for collection of the mass spectra. NMR experiments were carried out at
the Yokohama City University in Japan, with support from the Project for Creation of Research
Platforms and Sharing of Advanced Research Infrastructure by the Ministry of Education,
Culture, Sports, Science, and Technology of Japan (MEXT).
Abbreviations
BnBr, benzyl bromide; COSY, (1H–1H) correlation spectroscopy; DEPTꢀ135, distortionless
enhancement by polarization transfer; DIBALꢀH, diisobutylaluminium hydride; HSQC, (1H–13C)
heteronuclear singleꢀquantum coherence (NMR spectroscopy); HMBC, (1H–13C) heteronuclear
multipleꢀbond correlation (NMR spectroscopy); LDA, lithium diisopropylamide; PPTS,
pyridinium pꢀtoluenesulfonate; Py, pyridine.
References
ACS Paragon Plus Environment
29