Identification of 8-O-4/8-5(Cyclic)- and 8-8(Cyclic)/5-5-Coupled Dehydrotriferulic Acids, Naturally Occurring in Cell Walls of Mono- and Dicotyledonous Plants

Article abstract of DOI:10.1021/acs.jafc.6b02720

Besides ferulate dimers, higher oligomers of ferulic acid such as trimers and tetramers were previously demonstrated to occur in plant cell walls. This paper reports the identification of two new triferulic acids. 8-O-4/8-5(cyclic)-triferulic acid was synthesized from ethyl ferulate under oxidative conditions using copper(II)-tetramethylethylenediamine [CuCl(OH)-TMEDA] as a catalyst, whereas 8-8(cyclic)/5-5-triferulic acid was isolated (preparative size exclusion chromatography, reversed-phase HPLC) from saponified insoluble maize fiber. Structures of both trimers were unambiguously elucidated by high-resolution LC-ToF-MS/MS and one- (¹H) and two-dimensional (HSQC, HMBC, COSY, NOESY) NMR spectroscopy. The newly described trimers were identified by LC-MS/MS in alkaline hydrolysates of insoluble fibers from maize, wheat, and sugar beet, indicating that ferulic acid cross-links between cell wall polymers are more diverse than previously recognized. Saponification experiments also suggest that the newly identified 8-O-4/8-5(cyclic)-triferulic acid is the naturally occurring precursor of the previously identified 8-O-4/8-5(noncyclic)-triferulic acid in plant cell walls.

Full text of DOI:10.1021/acs.jafc.6b02720

Subscriber access provided by Northern Illinois University
Article
Identification of 8 O 4/8 5(cyclic)- and 8 8(cyclic)/5 5 Coupled Dehydrotriferulic
Acids, Naturally Occurring in Cell Walls of Mono- and Dicotyledonous Plants
Martin Waterstraat, Diana Bunzel, and Mirko Bunzel
J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.6b02720 • Publication Date (Web): 19 Aug 2016
Downloaded from http://pubs.acs.org on August 25, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Agricultural and Food Chemistry is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 26
Journal of Agricultural and Food Chemistry
Identification of 8-O-4/8-5(cyclic)- and 8-8(cyclic)/5-5-Coupled Dehydrotriferulic Acids,
Naturally Occurring in Cell Walls of Mono- and Dicotyledonous Plants
†
#
†,*
Martin Waterstraat , Diana Bunzel , and Mirko Bunzel
†
Department of Food Chemistry and Phytochemistry, Karlsruhe Institute of Technology (KIT),
Adenauerring 20a, 76131 Karlsruhe, Germany
#
Department of Safety and Quality of Fruits and Vegetables, Max RubnerꢀInstitut, Federal Research
Institute of Nutrition and Food, HaidꢀundꢀNeuꢀStraße 9, 76131 Karlsruhe, Germany.
*Corresponding author. Post address: Karlsruhe Institute of Technology (KIT), Institute of Applied
Biosciences, Department of Food Chemistry and Phytochemistry, Adenauerring 20a, 76131 Karlsruhe,
Germany; Telephone: (+49) 721 608 42936; Fax: (+49) 721 608 47255; Email: mirko.bunzel@kit.edu.
1
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 2 of 26
1
ABSTRACT
2
3
4
5
6
7
8
9
0
1
2
3
4
Besides ferulate dimers, higher oligomers of ferulic acid such as trimers and tetramers were previously
demonstrated to occur in plant cell walls. Here, we report the identification of two new triferulic acids.
8ꢀOꢀ4/8ꢀ5(cyclic)ꢀtriferulic acid was synthesized from ethyl ferulate under oxidative conditions using
copper(II)ꢀtetramethylethylenediamine
[CuCl(OH)ꢀTMEDA]
as
a
catalyst,
whereas
8ꢀ8(cyclic)/5ꢀ5ꢀtriferulic acid was isolated (preparative size exclusion chromatography, reversedꢀphase
HPLC) from saponified insoluble maize fiber. Structures of both trimers were unambiguously
1
elucidated by high resolution LCꢀToFꢀMS/MS and oneꢀ ( H) and twoꢀdimensional (HSQC, HMBC,
COSY, NOESY) NMR spectroscopy. The newly described trimers were identified by LCꢀMS/MS in
alkaline hydrolysates of insoluble fibers from maize, wheat, and sugar beet, indicating that ferulic acid
crossꢀlinks between cell wall polymers are more diverse than previously recognized. Saponification
experiments also suggest that the newly identified 8ꢀOꢀ4/8ꢀ5(cyclic)ꢀtriferulic acid is the naturally
occurring precursor of the previously identified 8ꢀOꢀ4/8ꢀ5(nonꢀcyclic)ꢀtriferulic acid in plant cell
walls.
1
1
1
1
1
1
1
5
6
KEYWORDS: ferulic acid, ferulate trimers, radical coupling, phenolic crossꢀlinks, plant cell walls
2
ACS Paragon Plus Environment

Page 3 of 26
Journal of Agricultural and Food Chemistry
1
7
INTRODUCTION
1
1
2
2
2
2
2
2
2
2
8
9
0
1
2
3
4
5
6
7
Cell walls are important components of plantꢀbased food products contributing to the texture of plant
foods and being the major source of dietary fiber in most diets. Plant cell wall stability is affected by
numerous structural components such as the celluloseꢀhemicellulose network, structural proteins, and,
depending on the cell type, lignin. In cereals and in certain plants belonging to the order
Caryophyllales, the celluloseꢀhemicellulose network and/or pectins are influenced by polymer bound
ferulic acid (4ꢀhydroxyꢀ3ꢀmethoxycinnamic acid), which can be found in amounts up to 3.1 % in
1
ꢀ3
maize bran and in lower amounts (up to 0.6 %) in dicotyledonous plants such as sugarꢀbeet pulp. In
monocotyledonous plants, ferulic acid predominantly acylates the arabinose residues of arabinoxylans,
whereas it is mostly attached to the pectic arabinan and galactan sideꢀchains of rhamnogalacturonanꢀI
4
in dicotyledonous plants.
2
2
3
3
3
3
3
3
3
3
3
8
9
0
1
2
3
4
5
6
7
8
Oxidative coupling of ferulate monomers by peroxidases and H O results in ferulate dimers, trimers,
2
2
5
and tetramers. Formation of these crossꢀlinks between polysaccharides, polysaccharides and lignin,
6
ꢀ9
and, potentially, polysaccharides and proteins increases cell wall stability against mechanical (and
1
0,11
thermal) stress and fermentability.
Five types of linkages between the ferulate monomers resulting
from oxidative coupling (8ꢀ5ꢀ, 8ꢀOꢀ4ꢀ, 8ꢀ8ꢀ, 5ꢀ5ꢀ, and 4ꢀOꢀ5ꢀcoupling) have been described to date.
However, alkaline hydrolysis results in nine different ferulic acid dimers, which were found in the
alkaline hydrolyzates of cell walls of several plants: three 8ꢀ8ꢀcoupled forms (cyclic, nonꢀcyclic, and
tetrahydrofuran), three 8ꢀ5ꢀcoupled forms (cyclic, nonꢀcyclic, and decarboxylated), and the 5ꢀ5ꢀ,
5
8ꢀOꢀ4ꢀ, and 4ꢀOꢀ5ꢀcoupled structures. It is assumed that all three 8ꢀ8ꢀcoupled dimers naturally occur
in the plant whereas only the 8ꢀ5(cyclic) dimer appears to be native with the other dimers being
1
2
artifacts of the alkaline hydrolysis.
3
4
4
4
4
9
0
1
2
3
The ferulate units of dehydrotriferulic acids (TriFA) and dehydrotetraferulic acids (TetraFA) are
bound through the same types of linkages as found in ferulate dimers, although an additional
13
8ꢀOꢀ4(H
2
O)ꢀtype has been described. The first trimer (5ꢀ5/8ꢀOꢀ4ꢀTriFA) was isolated and
14,15
structurally characterized from maize bran in 2003.
Since then, six additional regioisomers of
ferulic acid trimers and two regioisomers of ferulic acid tetramers have been isolated (all from maize
3
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 4 of 26
4
4
4
4
4
4
5
6
7
8
2
bran) and characterized: 8ꢀOꢀ4/8ꢀOꢀ4ꢀTriFA, 8ꢀ8(cyclic)/8ꢀOꢀ4ꢀTriFA, 5ꢀ5/8ꢀOꢀ4(H O)ꢀTriFA,
8ꢀOꢀ4/8ꢀ5(nonꢀcyclic)ꢀTriFA,
5ꢀ5/8ꢀ5(nonꢀcyclic)ꢀTriFA,
5ꢀ5/8ꢀ8(tetrahydrofuran)ꢀTriFA,
1
3,16,17
4ꢀOꢀ8/5ꢀ5/8ꢀOꢀ4ꢀTetraFA, and 4ꢀOꢀ8/5ꢀ5/8ꢀ5(nonꢀcyclic)ꢀTetraFA.
Here, we report the
structural identification of two novel ferulic acid trimers (Figure 1), occurring in cell walls of both
monocotyledonous and dicotyledonous plants.
4
5
9
0
Materials and Methods
5
5
5
5
5
5
5
5
5
6
6
6
6
6
6
6
6
6
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
Chemicals. Cu(I)Cl, tetramethylethylenediamine (TMEDA), HCl, NaOH, ethyl acetate, and ethanol
were purchased from Carl Roth GmbH (Karlsruhe, Germany). Acetonitrile, methanol, acetone,
petroleum ether, tetrahydrofuran, and Na
USA), acetoneꢀd and ferulic acid from Sigma Aldrich (St. Louis, Missouri, USA), and D
deutero GmbH (Kastellaun, Germany). Acetyl chloride was purchased from Fluka (Buchs,
Switzerland), formic acid from Merck KGaA (Darmstadt, Germany), NaHCO from Riedelꢀde Haën
AG (Seelze, Germany), and carbogen (5 % CO ) from Air Liquide S.A. (Düsseldorf, Germany).
2
SO
4
were from VWR International (Radnor, Pennsylvania,
6
2
O from
3
2
Enzymes. The thermostable αꢀamylase MATS L Classic (from Bacillus licheniformis, 8150 TAU/g),
the peptidase Maxazyme NNP DS (from Bacillus amyloliquefaciens, 184,000 PCU/g), and the
amyloglucosidase Amigase Mega L (from Aspergillus niger) were kindly donated by DSM Food
Specialties (Heerlen, Netherlands) and the thermostable αꢀamylase Termamyl 120 L (from Bacillus
licheniformis, 120 KNU/g), the peptidase Alcalase 2.5 L (from Bacillus licheniformis, 2.5 AU/g), and
the amyloglucosidase AMG 300 L (from Aspergillus niger, 300 AGU/g) were kindly donated by
Novozymes A/S (Bagsvaerd, Denmark). The carbohydrase mixture Driselase (from Basidomycetes,
cellulase activity ≥ 100 U/g, laminarinase activity ≥ 10 U/g, Xylanase activity ≥ 3 U/g) was purchased
from Sigma Aldrich, and the feruloyl esterases EꢀFAERU (from rumen microorganism, 600 U/mL)
and EꢀFAEZCT (from Clostridium thermocellum, 7 U/mL) were from Megazyme (Chicago, Illinois,
USA).
4
ACS Paragon Plus Environment

Page 5 of 26
Journal of Agricultural and Food Chemistry
6
9
0
1
2
3
Plant samples. Popcorn maize (Zea mays L. var. everta.) and wheat grain (Triticum aestivum L.) was
obtained from a local grocery store in Karlsruhe, Germany. Amaranth (Amaranthus hypochondriacus
L.) was field grown, and seeds were harvested in 2012 in Moersingen, Germany. Sugar beet pulp
7
7
7
7
(Beta vulgaris L. var. vulgaris) was kindly donated from Südzucker AG (Mannheim, Germany) in
2013.
7
7
4
5
Synthesis of trimer 1 (5ꢀ[(Z)ꢀ2ꢀcarboxyꢀ2ꢀ(4ꢀ((E)ꢀ2ꢀcarboxyvinyl)ꢀ2ꢀmethoxyphenoxy)vinyl]ꢀ2ꢀ(4ꢀ
hydroxyꢀ3ꢀmethoxyphenyl)ꢀ7ꢀmethoxyꢀ2,3ꢀdihydrobenzofuranꢀ3ꢀcarboxylic acid).
7
7
7
7
8
6
7
8
9
0
Ethyl ferulate was synthesized by adding 5 g (25.8 mmol) of ferulic acid and 2.5 mL (2.75 g,
35 mmol) of acetyl chloride to 50 mL of ethanol and stirring at room temperature overnight. After
evaporation of the solvents, the reaction was repeated by a second addition of acetyl chloride and
ethanol resulting in quantitative conversion. Following evaporation of the solvents, the remaining HCl
18
was removed by washing with 3 x 10 mL of ethanol.
8
8
8
8
8
8
8
8
1
2
3
4
5
6
7
8
Oxidative coupling of ethyl ferulate was carried out by using a catalytic procedure reported by Lu et
1
9
al. (2012). Briefly, 40 mg (0.4 mmol) of CuCl and 60 µL (0.4 mmol) of TMEDA were added to
60 mL of acetonitrile in a 200 mL twoꢀneck flask. After stirring for 5 min, 888 mg (4 mmol) of ethyl
ferulate were added, and the gas atmosphere was changed by attaching a balloon filled with carbogen
(5 % CO in O ) to the twoꢀneck flask. After stirring for 4 h at room temperature, the reaction was
2
2
stopped by adding 24 mL of 1 M HCl. The organic solvents were evaporated under reduced pressure
at 40 °C, and the crude product was extracted into ethyl acetate. The ethyl acetate volume was reduced
by rotary evaporation to 2.5 mL.
8
9
9
9
9
9
9
0
1
2
3
4
Initial fractionation was carried out by flash chromatography (Biotage Horizon flash chromatograph,
Charlottesville, VA, USA) using a prepacked silica column (GötecꢀLabortechnik GmbH, Bickenbach,
Germany, 30 g Si60, 15ꢀ40 µm particle size). Elution solvents were petroleum ether (A) and ethyl
acetate (B) with the following binary elution gradient: initially 30 % B, linear increase to 60 % B
within 1065 mL, followed by an elution with 550 mL of 100 % B. The flow rate was approximately
70 mL/min, and UVꢀdetection was performed at 254 nm. The first fraction (201 mL) was discarded;
5
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 6 of 26
9
9
5
6
subsequently, 48 fractions of 18 mL each were collected automatically. Based on UVꢀdata, fractions
20ꢀ48 were combined.
9
9
9
7
8
9
After evaporation under reduced pressure, the combined fractions were saponified by adding 20 mL of
2
2 M NaOH under N and protected from light for 18 h at room temperature. The alkaline hydrolyzate
was acidified with concentrated HCl to pH < 2, and the precipitate was extracted three times into 5 mL
of ethyl acetate. The organic solvent was evaporated under reduced pressure yielding 156 mg of
solids.
1
1
00
01
1
1
1
1
1
1
1
02
03
04
05
06
07
08
Purification by preparative HPLC was carried out using a Luna C18(2) reversedꢀphase column
(Phenomenex LTD, Aschaffenburg, Germany, 250 x 15 mm, 5 µm particle size, 10 nm pore size) and
UVꢀdetection at 320 nm. A gradient system consisting of 0.01 % formic acid in water (A) and 0.01 %
formic acid in acetonitrile (B) was used for the separation of the fraction (flow rate: 8 mL/min):
initially 15 % B, linear increase to 27.5 % B within 25 min, linear increase to 45 % B within 20 min,
followed by rinsing and equilibration steps. Trimer 1 (1.6 mg) eluted after 45.5 min with sufficient
purity (> 75 %) allowing for its structural characterization.
1
1
1
1
1
1
1
1
1
1
1
1
1
09
10
11
12
13
14
15
16
17
18
19
20
21
Isolation of trimer 2 ((E)ꢀ1ꢀ[5'ꢀ(2ꢀcarboxyvinyl)ꢀ2',6ꢀdihydroxyꢀ3',5ꢀdimethoxyꢀ(1,1'ꢀbiphenyl)ꢀ3ꢀyl]ꢀ
7ꢀhydroxyꢀ6ꢀmethoxyꢀ1,2ꢀdihydronaphthaleneꢀ2,3ꢀdicarboxylic acid).
20
Insoluble maize fiber was isolated according to a previously published method. Popcorn maize was
milled (particle size < 0.5 mm) and defatted with acetone. Aliquots (20 or 40 g) of defatted flour
(435 g) were suspended in phosphate buffer (pH 6.2, 0.08 M, 200 or 400 mL), and thermostable
αꢀamylase (0.5 or 1 mL, DSM Food Specialties) was added. After incubation with occasional shaking
at 92 °C for 30 min, flasks were cooled to room temperature, and the pH was adjusted to 5.2 with
0.325 M HCl. Partial protein removal was performed by adding a peptidase (100 or 200 µL, DSM
Food Specialties and incubation of the suspension in a shaking water bath at 58 °C for 30 min. After
heating the mixture to 92 °C for 5 min and cooling down to room temperature, the pH was adjusted to
4.2 with 0.325 M HCl. Samples were incubated with amyloglucosidase (50 or 100 µL, DSM Food
Specialties) in a shaking water bath at 58 °C for 45 min. Following filtration, the residues were
washed three times with 40 or 80 mL of water (60 °C), twice with 50 or 100 mL of 95 % ethanol and
6
ACS Paragon Plus Environment

Page 7 of 26
Journal of Agricultural and Food Chemistry
1
1
22
twice with 50 or 100 mL of acetone. In total, 90 g (20.6 % yield) of insoluble maize fiber were
recovered after drying overnight at 60 °C.
23
2
0
1
1
1
1
1
1
1
1
1
24
25
26
27
28
29
30
31
32
Alkaline hydrolysis and extraction was carried out as published previously. In brief, insoluble maize
fiber was saponified using 2 M NaOH (3 x 8 g of fiber in 200 mL; 10 x 5 g of fiber in 200 mL; 1 x 4 g
2
of fiber in 200 mL; and 2 x 6 g of fiber in 100 mL) under protection from light and O for 18 h at room
temperature. The suspensions were acidified (pH < 2) with concentrated HCl. Liberated compounds
were extracted into ethyl acetate (100 mL, 3 times), and the combined extracts were concentrated to
200 mL by rotary evaporation. Phenolic acids were extracted into 5 % (w/v) NaHCO
3 times), and, following acidification (pH < 2), reꢀextracted into ethyl acetate (200 mL, 3 times). The
combined ethyl acetate extracts were dried over Na SO and evaporated. In total, approximately 5 g of
3
(200 mL,
2
4
a dark, sticky oil was recovered, which was dissolved in 70 mL of THF.
1
1
1
1
1
1
1
1
1
33
34
35
36
37
38
39
40
41
Semiꢀpreparative size exclusion chromatography was performed on an HPLCꢀsystem consisting of a
gradient pump (System Gold programmable solvent module, Beckmann Coulter, Brea, California), a
column oven (LaChrom column oven Lꢀ7360, Merck), and a UVꢀdetector (System Gold detector
module 166, Beckman Coulter). The solution of phenolic acids in THF was injected manually in
aliquots of 500 µL and separated on a TSK 1000HXL column (Tosoh, Minato, Tokyo, 300 x 21.5 mm,
5 µm particle size, 100 – 1000 Da separation range) at 35 °C with a flow rate of 3.5 mL/min.
Detection was carried out at 355 nm to avoid saturation of the semiꢀpreparative flow cell (2 mm path
length). The fractions containing ferulate trimers (elution between 9.8 and 10.5 min, Figure 2) were
combined and dried by rotary evaporation.
1
1
1
1
1
1
1
42
43
44
45
46
47
48
A twoꢀstep fractionation of ferulate trimers and purification of trimer 2 was carried out using semiꢀ
preparative HPLC on a Luna phenyl hexyl column (Phenomenex, 250 x 10 mm, 5 µm particle size).
The HPLC system consisted of a gradient pump (System Gold 126 solvent module, Beckman Coulter),
a column oven (Beckman Coulter) operated at 35 °C, and a UV detector (System Gold 166 detector
module, Beckman Coulter). Eluents were 1 mM aqueous trifluoroacetic acid (A) and methanol:A (9:1,
v/v) (B). First, samples were dissolved in 1.25 mL of THF:H O (3:2, v/v) and injected manually in
2
aliquots of 80 µL. The following gradient was used: isocratic elution with 50 % B for 10 min, linear
7
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 8 of 26
1
1
1
1
1
1
1
1
49
50
51
52
53
54
55
56
increase to 60 % B within 15 min; isocratic with 60 % B for 10 min, linear increase to 80 % B within
15 min, isocratic with 80 % B for 5 min. Detection was carried out at 270 nm. The fractions eluting
between 16.5 and 20.0 min (Figure 3A) contained trimer 2 with impurities and were combined
2
followed by evaporation. In the final cleanꢀup step, residues were dissolved in 2 mL of THF:H O (1:1,
v/v) and injected in aliquots of 200 µL. The binary gradient started with 13 % B that was increased to
27 % B within 25 min, followed by rinsing and equilibrating steps. Detection was carried out at 260
nm. Trimer 2 eluted after 37 min (Figure 3B); its purity (> 90 %) allowed for unambiguous structural
elucidation.
1
1
1
1
1
1
1
57
58
59
60
61
62
63
NMR Spectroscopy. Both trimers were dissolved in 600 ꢁL of acetoneꢀd
6 2
:D O (10:1, v/v) and were
analyzed at 24.85 °C on a Bruker Ascend 500 MHz NMR spectrometer equipped with a Prodigy
1
1
13
cryoprobe (Bruker, Rheinstetten, Germany). H, phaseꢀsensitive H− CꢀHSQC with gradient
1
13
1
1
selection, H− CꢀHMBC with gradient selection, phaseꢀsensitive H− HꢀCOSY, and phase sensitive
1
1
Hꢀ HꢀNOESY experiments (standard Bruker implementations) were performed. Data were analyzed
using MestReNova software, and spectra were calibrated against the central acetone residual peak
(methyl proton, 2.05 ppm; methyl carbon, 29.84 ppm).
1
1
1
1
1
1
1
1
1
1
1
1
64
65
66
67
68
69
70
71
72
73
74
75
High-resolution mass spectrometry. Determination of exact masses was performed on an LCꢀTime
of Flight (ToF)ꢀMS/MS system consisting of a 1290 Infinity HPLC system (1290 Bin Pump G4220A
with degasser, 1290 Sampler G4226A autosampler, and 1290 DAD G4212A DAD detector, Agilent
Technologies, SantaClara, California, USA) coupled to a Triple ToF 5600 mass spectrometer (AB
Sciex Instruments, Framingham, Massachusetts, USA). The eluents were H O with 0.01 % formic acid
2
(A) and acetonitrile with 0.01 % formic acid (B), and the flow rate was 0.5 mL/min at room
temperature. Injection volume was 20 µL and the elution gradient was as follows: initially 10 % B,
linear to 27 % B within 10 min, linear to 45 % B within 20 min, linear to 60 % B in 2 min, followed
by rinsing and equilibration steps. The Duo Spray ion source was used in electrospray ionization (ESI)
negative mode at 650 °C with curtain gas, 45 psi; ion source gasꢀ1, 70 psi; ion source gasꢀ2, 60 psi;
and declustering potential, −100 V. ToFꢀMS experiments were completed over a scan range of
m/z 100 − 1000 using an accumulation time of 200 ms and a collision energy voltage of −10 V.
8
ACS Paragon Plus Environment

Page 9 of 26
Journal of Agricultural and Food Chemistry
1
1
1
76
ToFꢀMS/MS experiments (product ion scans of m/z 50 − 1000 for m/z 577.1) were performed with the
following parameters, using nitrogen as collision gas: collision energy voltage, −45 V; collision
energy spread, 25 V; and accumulation time, 80 ms.
77
78
1
1
1
1
1
1
1
1
1
1
79
80
81
82
83
84
85
86
87
88
Analysis of maize, wheat, sugar beet, and amaranth. Preparative isolation of insoluble dietary fiber
from maize, wheat, sugar beet, and amaranth was carried out as described for the isolation of trimer 2,
but using 20 g of freezeꢀdried, milled plant material and enzymes from Novozymes A/S (1.5 mL of
αꢀamylase, 600 µL of peptidase, 700 µL of amyloglucosidase) with optimized pHꢀvalues for these
enzymes (pH 6.2 for αꢀamylase, pH 7.5 for peptidase, pH 4.5 for amyloglucoidase). Insoluble dietary
fiber (100 mg) was saponified protected from light and O using 5 mL of 2 M NaOH for 18 h. After
2
acidification (pH < 2) the mixtures were extracted three times with 4 mL of diethyl ether. The extracts
were dried under N
solutions of the extracts from wheat and amaranth formed two layers, they were dried and redissolved
in methanol:H O (1:1, v/v).
2 2
, and the residues were dissolved in 0.5 mL of THF:H O (1:1, v/v). Because initial
2
1
1
1
1
1
1
1
1
89
90
91
92
93
94
95
96
For the enzymatic approaches, 100 mg of insoluble dietary fiber was incubated in duplicate with
6.5 mL of H O, 10 mg of Driselase, 100 µL of feruloyl esterase from Clostridium thermocellum
2
(EꢀFAEZCT) and 100 µL of feruloyl esterase from rumen microorganism (EꢀFAERU) at 40 °C for
24 h under protection from light. To one replicate, 2.16 mL of 4 M NaOH was added (final
concentration: 1 M NaOH), and the suspension was incubated at room temperature for 10 min. All
samples were acidified with concentrated HCl (pH < 2) and extracted three times with 4 mL of diethyl
ether. The organic solvent was evaporated, and the residues were redissolved in 0.5 mL of THF:H O
2
(1:1, v/v).
1
1
1
2
2
2
97
98
99
00
01
02
The samples were analyzed with LCꢀMS/MS, equipped with a 2690 separations module (pumps,
degasser, autosampler and a 996 PDA detector, Waters Corporation, Milford, Massachusetts, USA)
coupled to a Micromass Quattro Micro triple quadrupol mass spectrometer (Waters Corporation) and a
column oven (Jetstream Plus, Beckman Coulter). The sample (20 µL) was injected on a Kinetex
phenyl hexyl column (Phenomenex, 150 x 4.5 mm; 2.6 µm particle size; 10 nm pore size). Gradient
elution was performed with H
2
O with 0.01 % formic acid (A), methanol with 0.01 % formic acid (B),
9
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 10 of 26
2
2
2
2
2
2
2
2
03
04
05
06
07
08
09
10
and acetonitrile with 0.01 % formic acid (C) as eluents. Flow rate was 0.5 mL/min, and oven
temperature was maintained at 23 °C. Two different gradients were applied to detect the newly
discovered triferulic acids. Ternary gradient: 86 % A, 14 % C, linear within 7 min to 81 % A, 19 % C,
linear within 25 min to 69 % A, 9 % B, 22 % C, linear within 1 min to 74 % A, 26 % C, linear within
16 min to 64 % A, 36 % C, followed by rinsing and equilibration steps. Binary gradient: initially 30 %
B, linear within 59 min to 80 % B, followed by rinsing and equilibration steps. MS parameters were:
ionization, ESI negative mode; capillary, 3.0 kV; cone, 30 V; extractor, 2 V; RF lens, 0.1 V; source
temperature, 100 °C; desolvation gas flow, 700 L/h; cone gas flow, 60 L/h, collision energy, 20 eV.
2
2
11
12
RESULTS AND DISCUSSION
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Isolation/synthesis of trimers 1 and 2. Ferulate oligomers can be obtained from biomimetic
2
1,22
12,19
20
synthesis,
chemical synthesis,
or from ferulate rich plant materials. Whereas trimer 1 was
synthesized using a chemical oxidative coupling approach, trimer 2 was isolated from ferulateꢀrich
plant material, popcorn maize, after saponification. Although popcorn insoluble fiber contains
comparably large amounts of esterꢀlinked ferulates, diferulates, and higher oligomers, the alkaline
hydrolyzate contains a variety of matrix components, which need to be separated from the target
molecule. Also, the separation of ferulate dimers and tetramers from individual trimers is a
challenging task, which requires the application of different chromatographic principles. After
liberation of saponifiable components from maize insoluble dietary fiber, partition between ethyl
acetate and 5 % NaHCO is an easy and convenient step to roughly separate large quantities of
3
phenolic acids and other carboxylic acids from phenols and other less acidic compounds. To further
separate phenolic acids and other carboxylic acids, highꢀperformance size exclusion chromatography
was chosen. Because small molecules such as ferulic acid, pꢀcoumaric acid, benzoic acid, and vanillic
2
3
acid are the main phenolic acids in plant cell walls, size exclusion chromatography removes a
dominant fraction of phenolic acids. As shown in Figure 2, these small molecules can easily be
separated from carboxylic acids with higher hydrodynamic volumes. These are mainly ferulic acid
oligomers, which were partially separated into fractions 2ꢀ5, with fraction 3 containing most of the
1
0
ACS Paragon Plus Environment

Page 11 of 26
Journal of Agricultural and Food Chemistry
2
2
30
ferulic acid trimers. This fraction was further separated using two subsequent procedures based on
reversedꢀphase chromatography (Figure 3).
31
2
2
2
2
2
2
2
2
2
2
2
32
33
34
35
36
37
38
39
40
41
42
Different from trimer 2, trimer 1 was synthesized chemically using ethyl ferulate with the copper(II)ꢀ
1
9
TMEDA complex in an oxygenꢀrich atmosphere. Therefore, plantꢀbased matrix compounds do not
interfere; however, because this reaction produces a plethora of radical coupling products of ferulates,
basic chromatographic separation procedures are still required. A single silica flash chromatography
purification step was sufficient to isolate a fraction containing trimer 1 besides two other ferulic acid
1
3
16
trimers (8ꢀOꢀ4/8ꢀ5(nonꢀcyclic)ꢀTriFA, and 8ꢀOꢀ4/8ꢀOꢀ4ꢀTriFA ). These trimers were isolated by
preparative reversedꢀphase chromatography (Figure 4). However, the fraction obtained by rinsing the
flash chromatography column with a more polar solvent, ethyl acetate, was more complex, containing
1
1
17
known (5ꢀ5/8ꢀOꢀ4ꢀTriFA, 5ꢀ5/8ꢀ5(nonꢀcyclic)ꢀTriFA ) and unknown ferulic acid trimers and
2
0
24
tetramers. Compared to isolation procedures and targeted synthesis , synthesis using the copper(II)ꢀ
TMEDA complex is a fast and easy alternative to get ferulic acid trimers as standard compounds.
2
2
2
2
2
2
2
2
2
2
43
44
45
46
47
48
49
50
51
52
Structural elucidation of trimers 1 and 2. The UVꢀspectra of both isolated compounds show
maxima at 323 – 326 nm and shoulders at 290 – 300 nm suggesting ferulic acidꢀcontaining structures
(Figure 5). A broad shoulder as found for trimer 1 has been described earlier as an indicator for the
2
5
involvement of an 8ꢀOꢀ4ꢀlinkage. HPLCꢀESIꢀMS (negative ion mode) analysis produced a high
ꢀ
intensity of m/z 577 [MꢀH] for both compounds, indicating the occurrence of a dehydrotriferulic acid.
ꢀ
The sum formula of C30
H
25
O
12 for the [MꢀH] ion of both compounds was confirmed by highꢀ
ꢀ
resolution mass spectrometry with determined masses of 577.1342 [MꢀH] for trimer 1 and
ꢀ
577.1348 [MꢀH] for trimer 2. This represents an error of ꢀ1.7 ppm and ꢀ0.7 ppm, respectively, in MSꢀ
mode. Interpretation of MS/MSꢀdata resulted in errors of 2.7 ppm (46 fragments considered) and
3.8 ppm (22 fragments), respectively.
2
2
2
2
53
54
55
56
Unambiguous structural elucidation was achieved by interpreting spectra from various 1Dꢀ and 2Dꢀ
1
NMR experiments including Hꢀ, HSQCꢀ, HMBCꢀ, COSYꢀ and NOESYꢀexperiments. As determined
for trimer 1, two proton signals at 6.00 ppm and 4.37 ppm with coupling constants of 8.0 Hz suggest
12
an 8ꢀ5(cyclic)ꢀlinkage within the trimer. Because only two doublets with coupling constants of
1
1
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 12 of 26
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
16.0 Hz (indicative for unsubstituted transꢀpropenyl side chains) were found it was suggested that the
second linkage also involves the 8ꢀposition. Two doublet of doublets with coupling constants of about
8 and 2 Hz (representing protons in position 6 of ferulate units) and two doublets with coupling
constants of about 8 Hz (representing protons in position 5 of ferulate units) indicate that two ferulate
units are not linked in their 5ꢀposition, hinting that the third ferulic acid is coupled via an
8ꢀOꢀ4ꢀlinkage. Theoretically, two triferulate regioisomers that contain 8ꢀOꢀ4ꢀ and 8ꢀ5(cyclic)ꢀlinkages
exist. The proton at B7, which was unambiguously identified by HMBCꢀ and HSQCꢀexperiments, is
represented by a singlet in the proton spectrum thus showing no correlations in the COSYꢀexperiment.
Therefore, it was concluded that proton B7 has no vicinal neighbor demonstrating that the propenylic
side chain of the B unit is involved in the 8ꢀOꢀ4 linkage (Figure 1). All other signals from the different
NMRꢀexperiments are in good agreement with the proposed structure and are comparable to NMR
2
1
data of the previously published 8ꢀOꢀ4/8ꢀ5(cyclic)ꢀdehydrotriferulic acid triethyl ester. Because the
1
13
Hꢀshifts of the protons B2 and B6 overlapped, the corresponding Cꢀshifts at 115.5 and 121.1 ppm
2
1
13
were not independently assigned but based on literature. Also, the Cꢀshift of B1 was not
determined, because long rangeꢀcoupling was not observed using HMBC parameters as applied here.
1
2
2
2
2
2
2
2
72
73
74
75
76
77
78
The Hꢀspectrum of trimer 2 showed signals typical for 8ꢀ8(cyclic), i.e. two doublets at 4.66 ppm (B7)
and 4.04 ppm (B8) with coupling constants of 2.0 Hz and a singlet at 7.70 ppm (A7) (Table 1).
Because the proton spectrum did not show any doublets of doublets or 8 Hz doublets, it was suggested
that the third ferulic acid is coupled via a 5ꢀ5ꢀlinkage. This hypothesis was confirmed by interpretation
of the 2D NMR spectra including the NOESYꢀexperiment, which shows through space interactions of
the protons B6 and C6. Thus, the structure shown in Figure 1 is in agreement with all experimental
data.
2
2
2
2
2
79
80
81
82
83
Natural occurrence of trimers 1 and 2. The existence of trimer 1 and trimer 2 (in other plant
materials than popcorn maize) in plant cell walls was demonstrated by saponifying insoluble dietary
fiber from maize, wheat, sugar beet, and amaranth for 18 h in 2 M NaOH and subsequent LCꢀMS/MS
analysis. Because it was assumed that both newly identified trimers exist in comparably small amounts
in plant cell walls, larger amounts of insoluble dietary fiber (100 mg each) were saponified and run
1
2
ACS Paragon Plus Environment

Page 13 of 26
Journal of Agricultural and Food Chemistry
2
2
2
2
2
2
84
through the cleanꢀup procedure. Based on fragmentation patterns obtained from LCꢀMS/MSꢀanalysis
(Table 2), daughter ions with m/z 193, m/z 311, and m/z 489 for trimer 1 and daughter ions with
m/z 445 and m/z 489 for trimer 2 were chosen (with m/z 577 being the parent ion for both trimers) for
MS/MSꢀanalysis. Comparison of retention times with those of the isolated trimers 1 and 2 as standard
compounds demonstrated the existence of trimer 2 in maize, wheat, and sugar beet pulp, but not in
amaranth.
85
86
87
88
89
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
90
91
92
93
94
95
96
97
98
99
00
01
02
03
04
05
Trimer 1 was not detected in any plant materials after saponification for 18 h. This could be due to the
absence or only very low levels of this trimer in plant materials or due to structural modifications
during saponification and/or the cleanꢀup procedure as suggested for ferulate oligomers containing
5
8ꢀ5(cyclic)ꢀlinkages. It was demonstrated earlier that 8ꢀ5(cyclic)ꢀdehydrodiferulic acid (DFA) is
12
converted to 8ꢀ5(nonꢀcyclic)ꢀDFA and 8ꢀ5(decarboxylated)ꢀDFA under alkaline conditions. More
recently, it was also shown that 8ꢀ5(decarboxylated)ꢀDFA is partially formed from 8ꢀ5(cyclic)ꢀDFA in
2
6
aqueous buffers with pH values around 7. Therefore, less severe methods of hydrolysis were tested.
First, a combination of the carbohydrase mixture Driselase with two different feruloyl esterases was
tested. The carbohydrases of the multi enzyme preparation Driselase were supposed to support the
feruloyl esterases by partially cleaving the (feruloylated) polysaccharides to oligosaccharides.
However, neither of the trimers were detected in the enzymatic hydrolyzates. Whereas some ferulate
dimers were partially liberated from the plant cell wall polysaccharides (data not shown), enzyme
accessibility of areas crossꢀlinked by ferulate trimers may be restricted. However, when the enzymatic
procedure was followed by a brief alkaline hydrolysis (10 min) using less concentrated NaOH (1 M)
trimer 1 was detected in the hydrolyzates from maize, wheat, and sugar beet pulp by using the
MS/MSꢀtransitions described above. As for trimer 2, trimer 1 was also not detected from amaranth.
3
3
3
3
3
06
07
08
09
10
The newly identified trimers add to the complexity of ferulate crossꢀlinks identified in plant materials.
2
5,27
The analysis of ferulate based crossꢀlinks has been a challenging task so far
and will be more
complex by adding these new trimers to the spectrum of ferulate oligomers to be analyzed. A major
challenge in developing new protocols for the analysis of ferulate oligomers is the limited availability
1
9
of standard compounds. Oxidation in organic solvents, as originally developed by Lu et al. and
1
3
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 14 of 26
3
3
3
3
3
11
12
13
14
15
expanded to trimers in this study, is a promising path to synthesize ferulic acid trimers. Although our
studies were of qualitative nature only and do not allow definite conclusions on the quantities of these
newly identified trimers in plant materials, it needs to be kept in mind that even small amounts of
trimers and higher oligomers may have a major impact on the cell wall structure as they potentially
crosslink three or more polymers.
3
3
3
3
3
3
3
3
3
16
17
18
19
20
21
22
23
24
ABBREVATIONS
ESI, electrospray ionization; DFA, dehydrodiferulic acid; TriFA, dehydrotriferulic acid; TMEDA,
tetramethylethylenediamin; ToF, timeꢀofꢀflight
ACKNOWLEDGEMENTS
The authors thank Falco Beer (Max RubnerꢀInstitut) and Pepe Schrimpf (KIT) for technical assistance.
1
4
ACS Paragon Plus Environment

Page 15 of 26
Journal of Agricultural and Food Chemistry
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
25
REFERENCES
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
(1) Saulnier, L.; Marot, C.; Chanliaud, E.; Thibault, J.ꢀF. Cell wall polysaccharide interactions in
maize bran. Carbohydr. Polym. 1995, 26, 279ꢀ287.
(2) Rombouts, F. M.; Thibault, J.ꢀF. Feruloylated pectic substances from sugarꢀbeet pulp. Carbohydr.
Res. 1986, 154, 177ꢀ187.
(3) Hartley, R. D.; Harris, P. J. Phenolic constituents of the cell walls of dicotyledons. Biochem. Syst.
Ecol. 1981, 9, 189ꢀ203.
(4) Colquhoun, I. J.; Ralet, M. C.; Thibault, J.ꢀF.; Faulds, C. B.; Williamson, G. Structure
identification of feruloylated oligosaccharides from sugarꢀbeet pulp by NMR spectroscopy.
Carbohydr. Res. 1994, 263, 243ꢀ256.
(5) Bunzel, M. Chemistry and occurrence of hydroxycinnamate oligomers. Phytochem. Rev. 2010, 9,
47ꢀ64.
(6) Ralph, J.; Grabber, J. H.; Hatfield, R. D. Ligninꢀferulate crossꢀlinks in grasses: active incorporation
of ferulate polysaccharide esters into ryegrass lignins. Carbohydr. Res. 1995, 275, 167ꢀ178.
(7) Saulnier, L.; Crépeau, M.ꢀJ.; Lahaye, M.; Thibault, J.ꢀF.; GarciaꢀConesa, M. T.; Kroon, P. A.;
Williamson, G. Isolation and structural determination of two 5, 5′ꢀdiferuloyl oligosaccharides
indicate that maize heteroxylans are covalently crossꢀlinked by oxidatively coupled ferulates.
Carbohydr. Res. 1999, 320, 82ꢀ92.
(8) Piber, M.; Koehler, P. Identification of dehydroꢀferulic acidꢀtyrosine in rye and wheat: evidence
for a covalent crossꢀlink between arabinoxylans and proteins. J. Agric. Food Chem. 2005, 53,
5276ꢀ5284.
(9) Bunzel, M.; Allerdings, E.; Ralph, J.; Steinhart, H. Crossꢀlinking of arabinoxylans via 8ꢀ8ꢀcoupled
diferulates as demonstrated by isolation and identification of diarabinosyl 8ꢀ8 (cyclic)ꢀ
dehydrodiferulate from maize bran. J. Cereal Sci. 2008, 47, 29ꢀ40.
(10) Saulnier, L.; Thibault, J.ꢀF. Ferulic acid and diferulic acids as components of sugar‐beet pectins
and maize bran heteroxylans. J. Sci. Food Agric. 1999, 79, 396ꢀ402.
1
5
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 16 of 26
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
(11) Grabber, J. H.; Mertens, D. R.; Kim, H.; Funk, C.; Lu, F.; Ralph, J. Cell wall fermentation
kinetics are impacted more by lignin content and ferulate cross‐linking than by lignin
composition. J. Sci. Food Agric. 2009, 89, 122ꢀ129.
(12) Ralph, J.; Quideau, S.; Grabber, J. H.; Hatfield, R. D. Identification and synthesis of new ferulic
acid dehydrodimers present in grass cell walls. J. Chem. Soc., Perkin Trans. 1 1994, 3485ꢀ
3498.
(13) Bunzel, M.; Ralph, J.; Funk, C.; Steinhart, H. Structural elucidation of new ferulic acidꢀ
containing phenolic dimers and trimers isolated from maize bran. Tetrahedron Lett. 2005, 46,
5845ꢀ5850.
(14) Bunzel, M.; Ralph, J.; Funk, C.; Steinhart, H. Isolation and identification of a ferulic acid
dehydrotrimer from saponified maize bran insoluble fiber. Eur. Food Res. Technol. 2003, 217,
128ꢀ133.
(15) Rouau, X.; Cheynier, V.; Surget, A.; Gloux, D.; Barron, C.; Meudec, E.; LouisꢀMontero, J.;
Criton, M. A dehydrotrimer of ferulic acid from maize bran. Phytochemistry 2003, 63, 899ꢀ
903.
(16) Funk, C.; Ralph, J.; Steinhart, H.; Bunzel, M. Isolation and structural characterisation of 8–O–
4/8–O–4ꢀ and 8–8/8–O–4ꢀcoupled dehydrotriferulic acids from maize bran. Phytochemistry
2005, 66, 363ꢀ371.
(17) Bunzel, M.; Ralph, J.; Brüning, P.; Steinhart, H. Structural identification of dehydrotriferulic and
dehydrotetraferulic acids isolated from insoluble maize bran fiber. J. Agric. Food Chem. 2006,
54, 6409ꢀ6418.
(18) Fieser, M., Fiesers' reagents for organic synthesis. John Wiley & Sons: New York, 1967; Vol. 1.
(19) Lu, F.; Wei, L.; Azarpira, A.; Ralph, J. Rapid syntheses of dehydrodiferulates via biomimetic
radical coupling reactions of ethyl ferulate. J. Agric. Food Chem. 2012, 60, 8272ꢀ8277.
(20) Bunzel, M.; Funk, C.; Steinhart, H. Semipreparative isolation of dehydrodiferulic and
dehydrotriferulic acids as standard substances from maize bran. J. Sep. Sci. 2004, 27, 1080ꢀ
1086.
1
6
ACS Paragon Plus Environment

Page 17 of 26
Journal of Agricultural and Food Chemistry
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
78
(21) Bunzel, M.; Heuermann, B.; Kim, H.; Ralph, J. Peroxidaseꢀcatalyzed oligomerization of ferulic
acid esters. J. Agric. Food Chem. 2008, 56, 10368ꢀ10375.
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
(22) Ralph, J.; GarciaꢀConesa, M. T.; Williamson, G. Simple preparation of 8ꢀ5ꢀcoupled diferulate. J.
Agric. Food Chem. 1998, 46, 2531ꢀ2532.
(23) Eraso, F.; Hartley, R. D. Monomeric and dimeric phenolic constituents of plant cell walls —
possible factors influencing wall biodegradability. J. Sci. Food Agric. 1990, 51, 163ꢀ170.
(24) Mouterde, L. M.; Flourat, A. L.; Cannet, M. M.; Ducrot, P. H.; Allais, F. Chemoenzymatic total
synthesis of a naturally occurring (5‐5′)/(8′‐O‐4″) dehydrotrimer of ferulic acid. Eur. J. Org.
Chem. 2013, 2013, 173ꢀ179.
(25) Dobberstein, D.; Bunzel, M. Separation and detection of cell wallꢀbound ferulic acid
dehydrodimers and dehydrotrimers in cereals and other plant materials by reversed phase
highꢀperformance liquid chromatography with ultraviolet detection. J. Agric. Food Chem.
2010, 58, 8927ꢀ8935.
(26) Schendel, R. R.; Karrer, C.; Bunzel, D.; Huch, M.; Hildebrand, A. A.; Kulling, S. E.; Bunzel, M.
Structural transformation of 8–5ꢀcoupled dehydrodiferulates by human intestinal microbiota.
J. Agric. Food Chem. 2015, 63, 7975ꢀ7985.
(27) Jilek, M. L.; Bunzel, M. Dehydrotriferulic and dehydrodiferulic acid profiles of cereal and
pseudocereal flours. Cereal Chem. 2013, 90, 507ꢀ514.
3
96
1
7
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 18 of 26
3
3
3
4
4
4
4
4
97
98
99
00
01
02
03
04
FIGURE CAPTIONS
Figure 1: Structures of the newly identified dehydrotriferulic acids (TriFA) 8ꢀOꢀ4/8ꢀ5(cyclic)ꢀTriFA
(trimer 1) and 8ꢀ8(cyclic)/5ꢀ5ꢀTriFA (trimer 2).
Figure 2: Separation of ferulates (and other acids) from saponified maize fiber with size exclusion
chromatography and UV-detection at 355 nm. Peak 1, ferulic acid and other hydroxycinnamic and
benzoic acids; 2, ferulic acid dimers; 3, ferulic acid trimers; 4ꢀ5, higher ferulic acid oligomers.
4
4
4
4
4
05
06
07
08
09
Figure 3: Chromatograms of the first (A) and second (B) purification steps of trimer 2
(8ꢀ8(cyclic)/5ꢀ5ꢀdehydrotriferulic acid) using semiꢀpreparative phenyl hexylꢀHPLC with UVꢀdetection
at 270 nm (A) and 260 nm (B). Asterisks indicate the fractions that were collected and contained
trimer 2.
4
4
4
4
10
11
12
13
Figure 4: Chromatogram of the purification of trimer 1 (8ꢀOꢀ4/8ꢀ5(cyclic)ꢀTriFA, peak 3) using
preparative C18ꢀHPLC with UVꢀdetection at 320 nm. Peak 1, 8ꢀOꢀ4/8ꢀ5(nonꢀcyclic)ꢀTriFA;
peak 2, 8ꢀOꢀ4/8ꢀOꢀ4ꢀTriFA. TriFA – dehydrotriferulic acid.
4
4
4
14
15
16
Figure 5: UVꢀspectra of trimer 1 (8ꢀOꢀ4/8ꢀ5(cyclic)ꢀdehydrotriferulic acid), and trimer 2
(8ꢀ8(cyclic)/5ꢀ5ꢀdehydrotriferulic acid). The spikes at 363 nm were caused by a detector aberration.
1
8
ACS Paragon Plus Environment

Page 19 of 26
Journal of Agricultural and Food Chemistry
TABLES
Table 1: NMR data for trimer 1 (8ꢀOꢀ4/8ꢀ5(cyclic)ꢀdehydrotriferulic acid), and trimer 2
(
8ꢀ8(cyclic)/5ꢀ5ꢀdehydrotriferulic acid) measured in acetoneꢀd :D O (10:1, v/v). Coupling constants
6
2
are given in Hz.
Trimer 1
H [ppm]
Trimer 2
H [ppm]
1
13
1
13
trimer unit
A1
A2
A3
A4
A5
A6
A7
A8
C [ppm] trimer unit
C [ppm]
ꢀ
130.1
112.2
150.3
148.6
114.3
A1
A2
A3
A4
A5
A6
A7
A8
ꢀ
124.6
113.0
147.5
149.6
117.0
132.5
138.2
124.1
168.5
56.2
7.43 (1H; d; 1.9)
ꢀ
ꢀ
6.83 (1H; d; 8.4)
7.13 (1H; dd; 8.4, 1.9) 122.8
7.59 (1H; d; 16.0)
6.43 (1H; d; 16.0)
ꢀ
7.08 (1H; s)
ꢀ
ꢀ
6.79 (1H; s)
ꢀ
7.70 (1H; s)
ꢀ
145.2
117.5
168.0
56.3
A9
A3ꢀOMe
A9
A3ꢀOMe
ꢀ
4.00 (3H; s)
3.87 (3H; s)
B1
B2
B3
B4
B5
B6
B7
B8
ꢀ
nd
B1
B2
B3
B4
B5
B6
B7
B8
ꢀ
135.1
110.7
148.6
143.6
125.4
122.9
46.1
47.4
173.5
56.2
a
7.50 (1H; br s)
ꢀ
ꢀ
ꢀ
7.50 (1H; br s)
7.46 (1H; s)
ꢀ
ꢀ
115.5
145.3
150.2
127.4
121.1
129.9
138.8
164.7
55.9
6.81 (1H; d; 2.0)
ꢀ
ꢀ
ꢀ
6.61 (1H; d; 2.0)
4.66 (1H; d; 2.0)
4.04 (1H; d; 2.0)
ꢀ
a
B9
B3ꢀOMe
B9
B3ꢀOMe
3.77 (3H; s)
3.80 (3H; s)
C1
C2
C3
C4
C5
C6
C7
C8
C9
ꢀ
131.9
110.5
148.5
147.8
115.7
C1
C2
C3
C4
C5
C6
C7
C8
ꢀ
127.1
109.6
149.2
147.2
126.8
126.1
145.9
7.06 (1H; d; 1.9)
ꢀ
ꢀ
6.82 (1H; d; 8.2)
6.88 (1H; dd; 8.2, 1.9) 120.0
6.00 (1H; d; 8.0)
4.37 (1H; d; 8.0)
ꢀ
7.30 (1H; d; 1.9)
ꢀ
ꢀ
ꢀ
7.05 (1H; d; 1.9)
7.59 (1H; d; 15.9)
6.38 (1H;z d; 15.9) 116.1
88.3
55.8
171.9
56.1
C9
C3ꢀOMe
ꢀ
168.3
56.4
C3ꢀOMe
3.81 (3H; s)
3.94 (3H; s)
a
ꢀ
see text; nd ꢀ not determined
1
9
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 20 of 26
Table 2: Daughter ions of trimer 1 (8ꢀ5(cyclic)/8ꢀOꢀ4ꢀdehydrotriferulic acid), and trimer 2
(
8ꢀ8(cyclic)/5ꢀ5ꢀdehydrotriferulic acid) measured with triple quadrupol mass spectrometry after
fragmentation of m/z 577 as parent ion. MS parameters are given in Materials and Methods.
Trimer 1
Trimer 2
m/z Intensity [%] m/z Intensity [%]
1
4
3
4
5
93
89
11
17
44
100
10.1
9.6
2.3
2.1
489
445
533
271
430
100
29.2
5.5
4.2
2.8
2.3
1.8
515
56
2
2
0
ACS Paragon Plus Environment

Page 21 of 26
Journal of Agricultural and Food Chemistry
FIGURE GRAPHICS
Figure 1
2
1
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 22 of 26
Figure 2
2
2
ACS Paragon Plus Environment

Page 23 of 26
Journal of Agricultural and Food Chemistry
Figure 3
2
3
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 24 of 26
Figure 4
2
4
ACS Paragon Plus Environment

Page 25 of 26
Journal of Agricultural and Food Chemistry
Figure 5
2
5
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 26 of 26
TOC graphic
2
6
ACS Paragon Plus Environment