Identification of dehydro-ferulic acid-tyrosine in rye and wheat: Evidence for a covalent cross-link between arabinoxylans and proteins

Article abstract of DOI:10.1021/jf050395b

To monitor chemical reactions between ferulate and proteins during breadmaking, 8-¹⁴C-(E)-ferulic acid-(D-galactopyranose-6′-yl) ester was synthesized as a radiotracer and added to wheat and rye flour prior to breadmaking. Breads were lyophiliz

Full text of DOI:10.1021/jf050395b

5276 J. Agric. Food Chem. 2005, 53, 5276−5284
Identification of Dehydro-Ferulic Acid-Tyrosine in Rye and
Wheat: Evidence for a Covalent Cross-Link between
Arabinoxylans and Proteins
MICHAEL PIBER AND PETER KOEHLER*
Deutsche Forschungsanstalt fu¨r Lebensmittelchemie, Lichtenbergstrasse 4, 85748 Garching, Germany
To monitor chemical reactions between ferulate and proteins during breadmaking, 8-14C-(E)-ferulic
acid-(D-galactopyranose-6′-yl)ester was synthesized as a radiotracer and added to wheat and rye
flour prior to breadmaking. Breads were lyophilized, extracted by means of a modified Osborne
fractionation, and the radioactivity of the fractions was determined by scintillation analysis. The major
portion of the radioactivity remained in the water-soluble fraction. However, a significant enrichment
of the tracer was also detected in the prolamin and glutelin fractions in comparison to the control
experiment. Separation of the prolamin fraction by RP-HPLC and scintillation measurement of the
fractions gave evidence for a chemical modification of the tracer. To determine the structure of the
reaction product, the prolamin fractions were completely hydrolyzed to free amino acids by means of
an enzyme cocktail, and the digests were studied by LC-MS. In one fraction, a newly formed
compound was detected. Comparison of its chromatographic and mass spectrometric behavior with
a synthetic reference compound gave evidence that the newly identified compound was a
dehydroferulic acid-tyrosine cross-link. It is likely that this cross-link represented a covalent linkage
between arabinoxylans and cereal proteins. The newly identified cross-link was also identified in
wheat and rye flour doughs, which had been prepared without addition of the ferulate tracer. The
relative concentration of the dehydroferulic acid-tyrosine cross-link increased during wheat dough
preparation.
KEYWORDS: Arabinoxylans; pentosans; ferulic acid; tyrosine; cross-link; bread; wheat; rye
INTRODUCTION
Although there are possible physical explanations for the
effects of AX during breadmaking, like the strong gel formation
potential and high water holding capacity (14), there are several
effects that can only be explained by chemical interactions of
AX during dough and bread preparation. While the addition of
rye and wheat AX results in dough with improved viscoelastic
properties, addition of dephenolated AX results in practically
unchanged dough (15). From flour more AX an be isolated than
from dough, and mixing of dough in the presence of air oxygen
leads to a decrease of the ferulate content of AX (16). Along
with overmixing of wheat dough, ferulate is lost, whereas large
amounts of exogenous ferulic acid can induce breakdown of
the dough (17). AX interfere with gluten formation, because
they have a negative effect on gluten yield and lead to a less
extensible gluten (18-20). AX can also affect the re-agglomera-
tion of gluten-network. This effect can be weakened by addition
of free ferulic acid suggesting a ferulate related chemical
interaction (21). Covalent cross-linking of proteins and ferulate
containing AX induced by peroxidase has been proposed to
explain these results (22-24).
Arabinoxylans (AX) are important components of cereal cell
walls. As nondigestible constituents, they are by definition part
of the dietary fiber in cereals (1), where they make up the main
portion of the so-called pentosans. AX consist of linear (1 f
4)-â-D-xylopyranosyl-chains, which can be substituted at the
O-2 and/or O-3-positions with R-L-arabinofuranose (2). A
particular minor component of AX is ferulate, which is bound
to arabinose as an ester at the O-5-position (3). The AX of
different cereals can vary substantially in content, substitution
pattern, and molecular weight (4-7).
AX from rye and wheat flour have been the subject of many
investigations since they have been shown to have significant
influence on the formation and properties of dough and bread.
The addition of AX from rye and wheat to flour leads to an
increase in water absorption and loaf volume, and to a
retardation of bread staling (8-10). Addition of AX strongly
affects rheological parameters of wheat gluten (11). The content
of xylanase in flour, which can degrade AX, can have a strong
effect on the quality of dough and bread, stressing the influence
of AX on breadmaking, again (12, 13).
Despite this large number of investigations, no clear structure
function relationship of AX during breadmaking has been found.
The aim of this study was the investigation of chemical reactions
of ferulate present in wheat and rye flours during breadmaking.
* Author to whom correspondence should be addressed [telephone
++49 89 289 13372; fax ++49 89 289 14183; e-mail peter.koehler@
ch.tum.de].
10.1021/jf050395b CCC: $30.25
© 2005 American Chemical Society
Published on Web 06/02/2005

Protein−Carbohydrate Cross-Links in Grain-Based Products
J. Agric. Food Chem., Vol. 53, No. 13, 2005 5277
For this reason, a radioactive, [14C]-labeled ferulic acid ester
was synthesized as a tracer to monitor chemical modifications.
mL). The combined organic phases were washed with half concentrated
aqueous sodium chloride solution (2 × 20 mL). The organic phase
was dried over sodium sulfate and concentrated in vacuo. The crude
material was purified by flash chromatography (silica gel 60, hexane/
ethyl acetate 60/40 (v/v)) to give the desired product (667 mg, 1.53
mmol, 76%) as a white solid.
[1H] NMR (360 MHz, CDCl3): δH 7.63 (d, J ) 16.0 Hz, 1H), 7.05-
7.03 (m, 2H), 6.91 (d, J ) 8.0 Hz, 1H), 6.34 (d, J ) 16.0 Hz, 1H),
5.56 (d, J ) 5.0 Hz, 1H), 4.64 (dd, J ) 7.9 und 2.5 Hz, 1H), 4.44 (dd,
J ) 11.6 and 4.6 Hz, 1H), 4.34 (dd, J ) 5.0 and 2.6 Hz, 1H), 4.31 (t,
J ) 1.7 Hz, 1H), 4.29 (t, J ) 1.9 Hz, 1H), 4.13-4.09 (m, 1H), 3.93 (s,
3H), 1.53 (s, 3H), 1.47 (s, 3H), 1.36 (s, 3H), 1.34 (s, 3H). [13C] NMR
(63 MHz, CDCl3): δC 167.1, 148.0, 146.8, 145.2, 126.9, 123.3, 115.1,
114.7, 109.6,109.2, 108.8, 96.3, 71.1, 70.7, 70.4, 66.1, 63.3, 55.9, 25.5,
24.9, 24.8, 24.4.
MATERIALS AND METHODS
Chemicals. All reagents were obtained from commercial suppliers
in the highest purity available and used without further purification
unless otherwise noted. [1H] and [13C] NMR spectra were obtained on
a BRUKER AM-360 spectrometer, using tetramethylsilane as an
internal standard.
Radioactivity Analysis. Radioactivity of [14C]-decay was measured
by using a 1219 Rack Beta Liquid Scintillation Counter (LKB Wallac,
Sweden) with Emulsifier-SAFE Liquid Scintillation Cocktail (Packard
BioScience, Boston, MA; 4 mL scintillation cocktail/mL sample
solution). Counting time for each measurement was 60 s; standard
deviation was (2.3 dpm.
Synthesis of 8-[14C]-(E)-Ferulic Acid-(D-galactopyranose-6-yl)-
ester. To a solution of 8-[14C]-(E)-ferulic acid-(1,2:3,4-di-isopro-
pylidene-D-galactopyranose-6-yl)ester in 1,4-dioxane (11.4 mL) was
added hydrochloric acid (c ) 1 mol/L, 2.8 mL). The solution was stirred
under nitrogen atmosphere in a sealed tube for 1 h at 100 °C until
TLC analysis showed complete reaction. The solvent and hydrochloric
acid were removed under a nitrogen gas flow and finally dried in vacuo
to give the unprotected sugar ester (1.5 mmol, quantitative) as an orange
oil. For more convenient handling, the product was dissolved in DMSO
(10.0 mL).
Synthesis of 8-[14C]-(E)-Ferulic Acid. To a suspension of 2-[14C]-
disodiummalonate (1.9 mg, 250 µCi, 12.6 µmol; Amersham, Backing-
hamshire, UK) in pyridine (4 mL) were added malonic acid (1.25 g,
12.0 mmol), vanilline (1.52 g, 10.0 mmol), and piperidine (98 µL, 1.0
mmol). The reaction mixture was stirred under nitrogen atmosphere at
70 °C for 16 h until TLC-analysis indicated complete reaction. The
mixture was poured into hydrochloric acid (c ) 6 mol/L, 60 mL),
whereby a precipitate formed. The precipitate was centrifuged off,
washed with water (2 × 30 mL), and dried in vacuo to give the desired
product (1.53 g, 7.87 mmol, 78%) as an offwhite, crystalline solid.
[1H] NMR (360 MHz, methanol-d4): δH 7.59 (d, J ) 15.9 Hz, 1H),
7.16 (s, 1H), 7.05 (d, J ) 8.2 Hz, 1H), 6.81 (d, J ) 8.2 Hz, 1H), 6.31
(d, J ) 15.9 Hz, 1H), 4.92 (br. s, 2H), 3.88 (s, 3H). [13C] NMR (90
MHz, methanol-d4): δC 171.3, 150.6, 149.5, 147.2, 128.0, 124.2, 116.7,
116.1, 111.9, 56.7. Activity: 1.2 kBq/mg.
Synthesis of 4-O-tert-Butyldimethylsilyl-8-[14C]-(E)-ferulic Acid.
To a solution of 8-[14C]-(E)-ferulic acid (408 mg, 2.1 mmol, 492 kBq)
in DMF (7 mL) were added tert-butyldimethysilyl chloride and
imidazole (429 mg, 6.3 mmol). The solution was stirred under nitrogen
atmosphere at room temperature for 18 h. The reaction mixture was
poured into ice water/phosphoric acid (c ) 1 mol/L, 25 mL) and
extracted with ethyl acetate (3 × 10 mL). The combined organic phases
were washed with water (2 × 10 mL) and saturated sodium chloride
solution (10 mL). The organic phase was dried over sodium sulfate
and concentrated in vacuo to give the disilylated product. The disilylated
product was suspended in acetic acid/water/THF (3+1+1 (v+v+v), 7
mL), and the suspension was stirred at room temperature for 80 min.
The reaction was poured into ice water (20 mL) and extracted with
ethyl acetate (3 × 20 mL). The combined organic phases were washed
with water (2 × 10 mL) and saturated aqueous sodium chloride solution
(10 mL). The organic phase was dried over sodium sulfate and
concentrated in vacuo to give the desired product (626 mg, 2.03 mmol,
97%) as a reddish solid.
MS (ESI pos) (m/z, rel intensity): 357.1 ([M + H]+, 100%), 177.0
([M - C6H11O6]+, 88%). MS (ESI neg) (m/z): 355.0 ([M - H]-).
Activity: 24.6 kBq/mL.
Wheat Breadmaking Procedure. Baking experiments were per-
formed following the 10 g micro-version of the rapid-mix-test developed
by Kieffer et al. (25). To wheat flour (10 g, cv. Soissons, ash content
0.55% in dry mass), sodium chloride (0.2 g), sucrose (0.1 g), baker’s
yeast (0.7 g, Wieninger, Passau, Germany), and water (6.15 mL) was
added 8-[14C]-(E)-ferulic acid-(D-galactopyranose-6-yl)ester-solution
(100 µL, 2.5 kBq). The compounds were mixed in a micro-rapid-mixer
for 45 s at 1250 rpm. The dough was fermented for 20 min at 30 °C
and 90% rel. humidity. Afterward, a spherical dough piece was formed,
and proofed for 45 min at 30 °C and 90% rel. humidity. Baking was
performed at 230 °C for 10 min. The bread was lyophilized, ground,
and defatted by extraction with hexane (2 × 30 mL).
Rye Breadmaking Procedure. For the preparation of sour dough,
rye flour (100 g, cv. Nikita, ash content 1.15% in dry mass), sour dough
deep freeze starter (10 g, Bo¨cker, Minden, Germany), and water (90
mL) were mixed in a 500 mL beaker for 5 min and proofed at 28 °C
and 90% rel. humidity for 18 h to give sour dough. For the preparation
of the bread, to rye flour (5.72 g, cv. Nikita, ash content 1.15% in dry
mass), sour dough (4.29 g), sodium chloride (0.7 g), and water (4.8
mL) was added 8-[14C]-(E)-ferulic acid-(D-galactopyranose-6-yl)ester-
solution (300 µL, 7.4 kBq). The compounds were mixed in a micro-
rapid-mixer for 45 s at 1250 rpm. The dough was fermented for 15
min at 29 °C and 90% rel. humidity. The dough was given into a micro
baking tin (2.5 × 5 cm), and proofed for 50 min at 29 °C and 90% rel.
humidity. Baking was performed at 220 °C for 10 min. The bread was
lyophilized, ground, and defatted by extraction with hexane (2 × 30
mL).
[1H] NMR (360 MHz, CDCl3): δH 7.73 (d, J ) 15.7 Hz, 1H), 7.07-
7.03 (m, 2H), 6.85 (d, J ) 8.6 Hz, 1H), 6.31 (d, J ) 15.8 Hz, 1H),
3.84 (s, 3H), 0.99 (s, 9H), 0.18 (s, 6H). [13C] NMR (90 MHz, CDCl3):
δC 172.9, 151.3, 148.0, 147.1, 127.9, 122.7, 121.1, 115.0, 111.1, 55.4,
25.6, 18.5, -4.6.
Synthesis of 8-[14C]-(E)-Ferulic Acid-(1,2:3,4-di-isopropylidene-
D-galactopyranose-6-yl)ester. To a solution of 4-O-tert-butyldimeth-
ylsilyl-8-[14C]-(E)-ferulic acid (616 mg, 1.99 mmol) in dichloromethane
(20 mL) were added 1,2:3,4-di-isopropylidene-D-galactopyranose (551
mg, 2.11 mmol), dicyclohexylcarbodiimide (433 mg, 2.10 mmol), and
4-(dimethylamino)pyridine (24 mg, 0.20 mmol). The solution was
stirred at room temperature under nitrogen atmosphere for 20 h until
TLC analysis showed complete reaction; meanwhile a white precipitate
formed. The mixture was filtered, and the filtrate was washed with
aqueous acetic acid (w ) 5% (m/m), 2 × 15 mL), and half saturated
aqueous sodium chloride solution (15 mL). The organic phase was dried
over sodium sulfate and concentrated in vacuo to obtain the silylated
ester. To a solution of the silylated ester in DMF (15 mL) were added
potassium fluoride (233 mg, 4.0 mmol) and hydrobromic acid (w )
47% (m/m), 46 µL, 0.4 mmol). The mixture was stirred at room
temperature for 1 h. The reaction was poured into aqueous acetic acid
(w ) 5% (m/m), 60 mL) and extracted with ethyl acetate (3 × 30
Modified Osborne Fractionation. Osborne fractionations were
performed following the method developed by Wieser et al. (26) with
some modifications. Lyophilized and defatted bread samples (1000 mg)
were stepwise extracted with 0.4 mol/L NaCl/0.067 mol/L HKNaPO4
(pH 7.6) (3 × 10 mL), with 60% (v/v) aqueous ethanol (3 × 10 mL),
with 0.05 mol/L NaH2PO4/1% (w/v) SDS (pH 6.9) (3 × 10 mL), and
finally with 50% (v/v) aqueous 1-propanol/2 mol/L urea/0.05 mol/L
Tris-HCl (pH 7.5)/1% (w/v) dithioerythritol under nitrogen and
increased temperature (60 °C) (3 × 10 mL).
HPLC Analysis of Bread Prolamins. HPLC analyses of the bread
prolamins were performed on a Kontron Instruments System, which
was equipped with a data system D450, two HPLC 420 pumps, and an
HPLC 432 detector (Kontron Biotek, Neufahrn, Germany). The
chromatographic separation of the hydrolyzates was carried out on a
Nucleosil 300-5-C8 (250 × 4.6 mm, 5 µm) column (Macherey-Nagel,
Du¨ren, Germany) at 50 °C. The mobile phase was as follows: eluent

5278 J. Agric. Food Chem., Vol. 53, No. 13, 2005
Piber and Koehler
A, TFA in water (0.1% (v/v)); eluent B, TFA in acetonitrile (0.1%
(v/v)). A linear gradient (0-50 min, 24-56% B) was applied, flow
rate was 1.0 mL/min, UV-detection was at 210 nm. Injection volume
of the sample was 50 µL.
The sample (10 µL) was separated at a flow rate of 0.2 mL/min. The
solvent system was composed of (A) formic acid in water (0.1% (v/
v)) and (B) formic acid in acetonitrile (0.1% (v/v)). A linear gradient
(3-20 min, 10-42% B) was applied. The effluent between 1.5 and
10.0 mL was directed into the electrospray interface. The mass
spectrometer was operated in the positive electrospray ionization mode
(ESI+) with a spray needle voltage of 3.0 kV and a spray current of 5
µA. The temperature of the capillary was 300 °C, and the capillary
voltage was 35 V. The sheath and auxiliary gas (nitrogen) was adjusted
to 35 and 7 arbitrary units, respectively. The collision cell was operated
at a collision gas (argon) pressure of 6.7 × 10-2 Pa. The scan time for
each transition and single reaction monitoring was 0.60 s.
Enzymatic Hydrolysis of Bread Samples. The enzymatic hydrolysis
of the bread samples was performed with some modifications following
a procedure developed by Schmitz et al. for the hydrolysis of milk
proteins (25). Sample material (150-200 mg) was weighed into test
tubes (20 × 100 mm, Pyrex, UK) and suspended in hydrochloric acid
(5 mL, c ) 20 mmol/L). A solution of pepsin (200 µL, 6.5 mg/mL;
Roche, Basel, Switzerland) and a small crystal of thymol were added.
Incubation was carried out in a water bath at 37 °C for 24 h. After
addition of sodium hydroxide (200 µL, c ) 5 mol/L), the mixture was
incubated at room temperature for 2 h to hydrolyze ester bonds.
Hydrochloric acid (200 µL, c ) 5 mol/L) and Tris/HCl-buffer (600
µL, c ) 2 mol/L, pH 8.2) were added, followed by an R-amylase
solution (200 µL; 2.8 mg/mL, from bacillus licheniformis, 631 units/
mg). The mixture was heated at 90 °C for 5 min in a water bath and
incubated at room temperature for another 1.5 h. After addition of an
amyloglucosidase-solution (200 µL, 2.8 mg/mL, from aspergillus niger,
7.9 units/mg) and a â-xylanase-suspension (10 µL, in 3.2 mol/L
ammoniumsulfate, from T. longibrachiatum, 2740 units/mL, Mega-
zymes, Bray, Ireland), the mixture was incubated at 37 °C for 24 h,
after which a solution of Pronase E (200 µL, protease from strepto-
myceus griseus, 5.9 mg/mL, 5.6 units/mg) was added. Incubation was
carried out for another 24 h at 37 °C. A prolidase solution (125 µL,
from porcine kidney, 2.0 mg/mL, 88 units/mg) was added and the
incubation carried out another 24 h at 37 °C. Finally, a leucine
aminopeptidase suspension (70 µL, microsomal, from porcine kidney,
25 units/180 µL suspension (in 3.5 mol/L ammoniumsulfate, 10 mol/L
MgCl2, pH 7.7)) was added to the hydrolyzate mixture and incubated
for 24 h at 37 °C. After hydrolysis, samples were lyophilized, dissolved
in acetonitrile/water (15/85 (v/v)), and filtered (0.45 µm membrane
filter) prior to HPLC analysis.
Solid-Phase Synthesis of Dehydro-Ferulic Acid-Tyrosine. To a
suspension of Tentagel S-OH (500 mg, 90 µm, Rapp Polymere,
Tuebingen, Germany) in dichloromethane (5 mL) were added 4-O-
tert-butyldimethylsilyl-(E)-ferulic acid (10.2 mg, 33 µmol), diisopro-
pylcarbodiimide (13 µL, 83 µmol), and 4-(dimethylamino)pyridine (4.0
mg, 33 µmol). The mixture was shaken under nitrogen atmosphere at
room temperature for 20 h. The suspension was centrifuged off, and
the solvent was removed. The solid phase was washed with dichlo-
romethane (2 × 5 mL) and THF (2 × 5 mL). The solid phase was
suspended in THF (3 mL). Glacial acetic acid (50 µL) and a solution
of tetrabutylammoniumfluoride (330 µL, c ) 1 mol/L in THF) were
added and stirred for 1 h at 0 °C to deprotect the ferulate. The
suspension was centrifuged off, and the solvent was removed. The solid
phase was washed with THF (3 × 5 mL) and dried under a nitrogen
gas flow. The solid phase was washed with ammonium formiate-buffer
(5 mL, c ) 135 mmol/L, pH 9.2) and afterward suspended again in
ammonium formiate-buffer (5 mL, c ) 135 mmol/L, pH 9.2).
L-Tyrosine (50 mg) and peroxidase (from horseradish) (1.1 mg) were
added. The suspension was shaken for 2 h at room temperature.
Hydrogen peroxide solution (200 µL, c ) 0.3 mol/L) was added, and
the mixture was shaken for 3 h at room temperature. The mixture was
centrifuged off, and the solvent was removed. The solid phase was
washed with 0.1% aqueous formic acid (3 × 5 mL) and water (5 mL).
To the solid phase was added acetonitrile/water (3 mL, 1/1 (v/v))/1%
sodium hydroxide (m/m) to hydrolyze the ester. The suspension was
shaken for 2 h at room temperature. The suspension was centrifuged
off, and the supernatant was collected. The solid phase was washed
with acetonitrile/water (1/1 (v/v))/1% (m/v) sodium hydroxide (2 × 3
mL) and centrifuged off. To the combined supernatants was added
concentrated formic acid (100 µL). The solution was dried under a
nitrogen gas flow to give a white solid, which was dissolved in 0.1%
aqueous formic acid (1.0 mL). For further purification, a solid-phase
extraction workup was carried out.
HPLC Analysis of Bread Hydrolyzates. HPLC analyses of the
bread hydrolyzates were performed on a Kontron Instruments System,
which was equipped with a System 522 pump and an HPLC 535
detector (Kontron Biotek, Neufahrn, Germany). The chromatographic
separation of the hydrolyzates was carried out on a HyperClone 5 µ
ODS (C18) (250 × 4.6 mm, 5µm) column (Phenomenex, Aschaffenburg,
Germany) at room temperature. The mobile phase was as follows:
eluent A, TFA in water (0.1% (v/v)); eluent B, TFA in acetonitrile
(0.1% (v/v)). A linear gradient (1-40 min, 5-95% B) was applied;
flow rate was 0.8 mL/min, UV-detection was at 310 nm. Injection
volume of the sample was 100 µL.
A Strata C18-T SPE-tube (500 mg, 3 mL) (Phenomenex, Aschaffen-
burg, Germany) was conditioned by elution with formic acid in
acetonitrile (0.1% (v/v)) (2 mL), Na2HPO4-buffer (c ) 0.05 moL/L,
pH 7.4) (3 × 2 mL), and formic acid in water (0.1% (v/v)) (3 × 2
mL). Afterward, the sample solution was applied onto the SPE tube
and eluted with formic acid in water/acetonitrile (99/1 (v/v))/0.1% (v/
v) formic acid to remove polar compounds. The desired mixture
containing several tyrosine-ferulic acid products was eluted with water/
acetonitrile (55/45 (v/v))/0.1% (v/v) formic acid. The eluent was
separated by RP-HPLC, and the desired product (Figure 6, peak #6)
was detected by MS and MS-MS.
HPLC/MS Analysis. Mass spectra were recorded by means of an
ion trap mass spectrometer (LCQ classic, Finnigan MAT, Dreieich,
Germany) coupled to a liquid chromatography system (Spectra System
P4000 HPLC pump and Spetra System UV1000 Detector, Finnigan
MAT, Dreieich, Germany) equipped with an autosampler and a
HyperClone 5 µ ODS (C18) (250 × 4.6 mm, 5 µm) column (Phenom-
enex, Aschaffenburg, Germany). The mobile phase was composed as
follows: eluent A, formic acid in water (0.1% (v/v)); eluent B, formic
acid in acetonitrile (0.1% (v/v)). A linear gradient (1-21 min, 15-
42% B) was applied; flow rate was 0.8 mL/min, UV-detection was at
310 nm. The effluent between 4.0 and 19.2 mL was directed into the
electrospray interface. The mass spectrometer was operated in the
positive electrospray ionization mode (ESI+) with a spray needle voltage
of 5.0 kV and a spray current of 80 µA. The temperature of the capillary
was 200 °C, and the capillary voltage was 10 V. The sheath and
auxiliary gas (nitrogen) was adjusted to 80 and 20 arbitrary units,
respectively. The compounds in the hydrolyzed samples were character-
ized by means of their molecular masses obtained in the full scan mode
(range m/z 105-1000).
MS (ESI pos.)(m/z, rel intensity): 769.1 ([2M + Na]+, 85%), 747.0
([2M + H]+, 100%), 396.1 ([M + Na]+, 22%), 374.0 ([M + H]+, 87%).
MS-MS (374.0; ESI pos., collision energy 27%) (m/z, rel intensity)
(315.3, 5%; 310.0, 20%; 292.9, 50%; 283.8, 80%; 271.0, 100%).
Synthesis of 3-OMe-[2H]3-(E)-Ferulic Acid. To a solution of
3-[2H]3-methoxy-4-hydroxybenzaldehyd (MR(155)/MR(152) ) 99.67%)
(donation of Ms. C. Scheidig) (200 mg, 1.29 mmol) and malonic acid
(105 mg, 1.55 mmol) in pyridine (2 mL) was added piperidine (13
µL). The reaction mixture was stirred under nitrogen atmosphere at 70
°C for 18 h until TLC analysis indicated complete reaction. The mixture
was poured into hydrochloric acid (c ) 6 mol/L, 30 mL) and extracted
with ethyl acetate (3 × 10 mL). The combined organic phases were
washed with 1 mol/L hydrochloric acid (2 × 15 mL) and saturated
aqueous sodium chloride solution (15 mL). The organic phase was dried
over sodium sulfate and concentrated in vacuo. The crude material was
HPLC/MS-MS Analysis. Mass spectra were recorded by means
of a triple quadrupole tandem mass spectrometer (TSQ Quantum
Discovery, Thermo Electron, Dreieich, Germany) coupled to a Surveyor
HPLC system (Thermo Finnigan, Dreieich, Germany) equipped with
a thermostated (20 °C) autosampler and a Synergi Polar-RP 80 Å HPLC
column (150 × 2.0 mm, 4 µm, Phenomenex, Aschaffenburg, Germany;
kept at 30 °C) and a Polar-RP precolumn (4 × 2.0 mm, Phenomenex).

Protein−Carbohydrate Cross-Links in Grain-Based Products
J. Agric. Food Chem., Vol. 53, No. 13, 2005 5279
Figure 1. Synthesis of 8-[14C]-(E)-ferulic acid-( 14C label.
D-galactopyranose-6-yl)ester. /:
J ) 8.2 Hz, 1H), 6.30 (d, J ) 15.9 Hz, 1H), 4.86 (br. s, 2H). [13C]
NMR (90 MHz, methanol-d4): δC 171.3, 150.8, 149.6, 147.2, 128.1,
124.3, 116.7, 116.2, 112.0.
MS (ESI pos) (m/z, rel intensity): 198.0 ([M + H]+, 100%), 180.2
([M - H2O + H+], 44%).
Solid-Phase Synthesis of [2H]3-Dehydro-(E)-Ferulic Acid-Ty-
rosine. Synthesis of [2H]3-4-O-tert-Butyldimethylsilyl-8-(E)-Ferulic
Acid. The synthesis was performed as described for the undeuterated
compound. 3-OMe-[2H]3-(E)-ferulic acid (191 mg, 0.97 mmol) was
transformed after reaction with tert-butyldimethylsilyl chloride (377
mg) and imidazole (68 mg) to the desired product (277 mg, 0.89 mmol,
92%) as an off-white solid.
[1H] NMR (360 MHz, CDCl3): δH 7.75 (d, J ) 15.9 Hz, 1H), 7.07-
7.04 (m, 2H), 6.86 (d, J ) 8.6 Hz, 1H), 6.32 (d, J ) 15.9 Hz, 1H),
1.00 (s, 9H), 0.18 (s, 6H). [13C] NMR (90 MHz, CDCl3): δC 172.6,
151.3, 148.0, 147.1, 128.0, 122.7, 121.1, 115.0, 111.1, 25.6, 18.5, -4.6.
Synthesis of [2H]3-Dehydro-(E)-Ferulic Acid-Tyrosine. The synthesis
was performed as described for the undeuterated compound. [2H]3-4-
O-tert-Butyldimethylsilyl-8-(E)-ferulic acid (40.7 mg, 0.13 mmol) was
coupled to Tentagel S-OH (2.0 g) with diisopropylcarbodiimide (51
µL) and (dimethylamino)pyridine (4.0 mg). The ester was desilylated
with tetrabutylammoniumfluoride (1.32 mmol) and afterward brought
to reaction with L-tyrosine (200 mg), horseradish peroxidase (3.0 mg),
and hydrogen peroxide (0.13 mmol). After alkaline release and solid-
phase extraction workup, a solution of [2H]3-dehydro-(E)-ferulic acid-
tyrosine and other deuterated tyrosine-ferulic acid products (12 mL)
was obtained. The concentration of the desired isomer (Figure 6, peak
# 6) was estimated by HPLC/UV-detection at 310 nm. Because no
identical standard compound was available, ferulic acid was used for
the generation of a calibration curve. Under the assumption of an
identical molar absorption coefficient of ferulic acid and DFT at 310
nm (as tyrosine does not absorb light at 310 nm), the concentration of
the desired isomer (Figure 6, peak # 6) was 978 ng/mL.
MS (ESI pos.) (m/z, rel intensity): 377.0 ([M + H]+, 100%).
MS/MS (377.0; ESI pos., collision energy 21%) (m/z, rel intensity)
(377.0, 12%; 359.1, 45%; 313.1, 56%; 296.2, 100%).
Quantitation of Dehydro-Ferulic acid-Tyrosine in Wheat and
Rye Flour, Dough, and Bread. Wheat and rye dough and bread were
made as described in the baking procedures, except no ferulic acid was
added. Flour, dough, and bread samples were lyophilized and ground
(<0.2 mm). 10-20 mg of the sample was weighed into a sealed
tube. Hydrochloric acid/phenol (c(HCl) ) 6 mol/L, w(phenol) ) 0.1%
(m/v)) was added, and the material was hydrolyzed under nitrogen
atmosphere at 110 °C for 24 h. Hydrochloric acid was removed at room
Figure 2. RP-HPLC/scintillation chromatogram of secalins and gliadins
isolated by modified Osborne fractionation of rye and wheat bread. Each
data point corresponds to the radioactivity present in the eluate collected
during 1 min of elution time. Mean coefficient of variation <3%.
purified by flash chromatography (silica gel 60, ethyl acetate/hexane/
acetic acid, 80+20+1 (v+v+v)) to give the desired product (235 mg,
1.19 mmol, 93%) as an off-white solid.
[1H] NMR (360 MHz, methanol-d4): δH 7.59 (d, J ) 15.9 Hz, 1H),
7.16 (d, J ) 1.8 Hz, 1H), 7.05 (dd, J ) 8.2 and 1.9 Hz, 1H), 6.81 (d,

5280 J. Agric. Food Chem., Vol. 53, No. 13, 2005
Piber and Koehler
Figure 3. RP-HPLC/scintillation chromatogram of rye and wheat bread enzymic hydrolyzates. Each data bar corresponds to the radioactivity present in
the eluate collected during 1 min of elution time. Mean coefficient of variation <3%.
Figure 4. RP-HPLC/ESI-MS of dehydro-ferulic acid-tyrosine present in a wheat bread enzymic hydrolyzate. Signals are assigned to the structure of the
ions. M: molecular mass of dehydro-ferulic acid-tyrosine (373 g/mol).
temperature under nitrogen gas flow. To the dry hydrolyzate were added
the identified isomer of [2H]3-dehydro-(E)-ferulic acid-tyrosine (9.8 ng)
and aqueous formic acid (1.0 mL, 0.1% (v/v)). The mixture was
homogenized using an ultrasonic bath for 15 min, followed by a solid-
phase extraction workup as described for the synthesis of DFT. The
eluent of the workup was lyophilized and dissolved in acetonitrile/
water/formic acid (250 µL, 10/90/0.1 (v/v/v)). The concentration of
DFT was determined by means of HPLC/MS-MS analysis.

Protein−Carbohydrate Cross-Links in Grain-Based Products
J. Agric. Food Chem., Vol. 53, No. 13, 2005 5281
Figure 5. Solid-phase synthesis of dehyro-ferulic acid tyrosine.
the control experiment was found in the aqueous extract, and
less than 4% in the ethanol extract. In comparison to the control
experiment, the prolamin fraction (ethanol extract) of the rye
and wheat bread was strongly enriched with the tracer. The
glutelin fractions (SDS extract without and with reducing agent
added) of the wheat bread also contained significantly higher
amounts of the tracer than the corresponding extracts of the
control experiment. These findings indicated that ferulate
underwent chemical modifications during breadmaking, which
led to an enrichment of ferulate in the protein fractions of the
bread.
HPLC/Scintillation of Prolamins. As higher radioactivity
was found in the prolamin fractions than in the glutelin fractions,
the prolamins of rye and wheat bread were further investigated
by HPLC/scintillation measurements. First, the isolated secalins
(rye prolamins) and gliadins (wheat prolamins) of the tracer-
containing breads were separated by means of reversed phase
HPLC. As a control experiment, the unmodified tracer was
examined in the same manner. Fractions of the HPLC separation
were further analyzed by scintillation measurements to obtain
HPLC/scintillation chromatograms. Both the chromatograms of
secalins and gliadins showed signals that were not present in
the control experiment (Figure 2). These results confirmed the
assumption that ferulate was chemically modified during rye
and wheat bread preparation.
Figure 6. RP-HPLC chromatogram of the solid-phase product mixture
after alkaline release. 1, tyrosine; 2, 3, 4, and 6, isomers of dehydro-
ferulic acid-tyrosine; 5, ferulic acid; 7, isomers of dehydro-diferulic acid-
tyrosine and -triferulic acid-tyrosine; 8 and 9, isomers of decarboxylated
dehydro-triferulic acid-tyrosine.
Table 1. Retention of the Ferulate Radio Tracer within Protein
Fractions
protein fraction
albumins/
globulins
SDS-soluble
glutelins
SDS-insoluble
glutelins
prolamins
controla
95.8%
70.5%
78.6%
3.4%
24.5%
19.8%
0.5%
1.2%
0.7%
0.3%
3.8%
0.9%
HPLC/Scintillation of Bread Hydrolyzates. To isolate the
chemically modified ferulate compound, the bread samples had
to be hydrolyzed. Whole bread samples were used rather than
Osborne fractions because the former had a higher radioactivity
than the prolamin and glutelin fractions. As described above, a
modified enzymatic procedure by Schmitz et al. (27) was used
to hydrolyze the bread samples and, therefore, to release the
chemically modified tracer. Again, unmodified ferulate tracer
was used in the control experiment. The hydrolyzed samples
were separated by means of reversed phase HPLC, and fractions
of the separation were examined by scintillation measurement
to monitor the ferulate tracer. The results are shown in Figure
3. The largest signals in all measurements at t ) 21 min were
due to unchanged ferulate tracer. Additionally to the unchanged
tracer, both rye and wheat bread samples showed numerous
signals that were not present in the control experiment. All of
these signals originated from ferulate modifications formed
during breadmaking.
wheat breada
rye breada
a Number of experiments n
) 3, mean coefficient of variation <3%.
RESULTS
Synthesis of the [14C]-Labeled Tracer. Starting from
vanilline and 2-[14C]-labeled malonic acid, 8-[14C]-labeled
(E)-ferulic acid was obtained in a Knoevenagel condensation.
Because free ferulic acid is poorly water soluble and has dif-
ferent electronic properties than an ester, the acid was converted
into 8-[14C]-(E)-ferulic acid-(D-galactopyranose-6-yl)ester as
outlined in Figure 1. After acidic hydrolysis of the ketal, the
product was obtained as a mixture of diastereomers of the
carbohydrate unit. Therefore, it was not possible to characterize
the product by NMR. The product was used as a radio tracer in
the subsequent studies.
Modified Osborne Fractionation of Tracer Breads. After
adding the radio tracer to the flour, wheat and rye breads were
prepared in a 10 g scale. The breads were extracted by a
modified Osborne fractionation. The albumin/globulin-, prola-
min-, SDS-soluble, and SDS-insoluble glutelin fractions were
examined by means of liquid scintillation measurements to
monitor the retention of the radio tracer. The results are shown
in Table 1. As a control experiment, the radio tracer was added
to wheat starch, which was chosen as a control instead of flour,
to exclude the possible formation of artifacts especially during
the extraction of the albumins/globulins. As the unmodified
tracer was rather water soluble, more than 95% of the tracer in
HPLC/MS. To determine the structure of the modified
ferulate compounds, all radioactive fractions of the bread
hydrolyzates were examined by means of HPLC/MS measure-
ments. In addition, MS-MS spectra were recorded from the
eluted peaks. To ensure sensitive measurement of possible
ferulic acid-amino acid cross links, 7-S-cysteinohydroferulic acid
was synthesized (28) and used as a tuning standard for the ESI
mass spectrometer. LC/MS analysis of the fractions t )
18-20 min of the HPLC separation of both the rye and the
wheat bread hydrolyzates (Figure 4) gave a signal with m/z
[M + H]+ ) 374, which corresponded to the structure of a
dehydro-ferulic acid-tyrosine cross-link. To verify the structure

5282 J. Agric. Food Chem., Vol. 53, No. 13, 2005
Piber and Koehler
Figure 7. Speculative structures of dehydro-ferulic acid-tyrosine heterodimers related to those known from ferulate homocoupling.
of this compound, dehydro-ferulic acid-tyrosine was synthesized
by treating polymer-bound ferulate with tyrosine and hydrogen
peroxide in the presence of horseradish peroxidase and subse-
quent alkaline hydrolysis (Figure 5). After SPE purification,
the products were analyzed by HPLC/MS. As depicted in Figure
6, the synthesis led to a product mixture. The major components
were several peaks giving signals at m/z ) 566.1 ([M + H]+).
From the molecular masses and the fragmentation pattern of
MS-MS experiments, it could be assumed that this class of
compounds was different isomers containing two ferulic acid
Figure 8. Differences in retention time of unlabeled and [2H]3 labeled
dehydro-ferulic acid-tyrosine. RP-HPLC/MS−MS chromatogram of (A) an
enzymic wheat bread hydrolyzate containing dehydro-ferulic acid-tyrosine
and (B) synthetic [2H]3-dehydro-ferulic acid-tyrosine.
and one tyrosine unit. Another class of compounds was
represented by four peaks, all having signals at m/z ) 374 of
the [M + H]+ ions and all showing similar fragmentation
patterns (m/z 374 f m/z 310 and m/z 293); one isomer (peak
#6) additionally showed the fragmentation m/z 374 f m/z 315.
From the mass spectrometry data, it could be concluded that
these four compounds were different isomers of dehydro-ferulic
acid-tyrosine (DFT) representing different types of linkage. As
outlined in Figure 7, structures of the heterodimers related to
those known from ferulate homocoupling are possible. However,
as the spectra of all four compounds were very similar, an
assignment of the type of linkage between ferulic acid and
tyrosine was not possible. Furthermore, it was not possible to
obtain NMR spectra of these compounds, because a certain
degree of isomerization could always be detected after HPLC
purification and re-injection. Current investigations are underway
to determine which isomer of dehydro-ferulic acid-tyrosine was
formed.
of a product with the mass of [2H3]-labeled DFT. Therefore,
there was no need to synthesize the labeled galactose ester.
Relative Quantitation of Dehydro-Ferulic Acid-Tyrosine.
DFT was detected in all hydrolyzed samples of wheat and rye
(flour, dough, bread). This was also the case for samples that
had been prepared without any addition of the exogenous
ferulate tracer. To get information on the formation of DFT
within the course of bread preparation, the relative changes of
the concentrations of DFT were determined in the hydrolyzates
of flour, dough, and bread using a deuterium labeled isomer of
DFT (Figure 6, peak # 6) as an internal standard, which also
allowed the estimation of the magnitude of the amount of DFT
by assuming a similar response factor of the exogenous and
natural DFT-isomer. To do so, [2H]3-dehydro-ferulic acid-
tyrosine was synthesized from [2H]3-ferulic acid and L-tyrosine
using hydrogen peroxide and horseradish peroxidize as already
described for the nondeuterated compound. Addition of [2H]3-
DFT to the hydrolyzed samples allowed the estimation of the
content of DFT in the hydrolyzates by HPLC/MS-MS analysis
(Figure 8). Dough and bread were prepared from wheat and
rye flour without the addition of exogenous ferulate compounds.
The relative changes of the content of DFT from flour to dough
and bread from wheat and rye, respectively, are shown in Table
2. The influence of L-ascorbic acid, which is widely used as a
dough improver, on the formation of DFT was also examined
in these experiments.
Comparing the LC/MS and LC/MS-MS data of the synthe-
sized and the isolated compounds, only one of the substances
from the synthesis showed a MS pattern identical to that of the
isolated product. However, these two substances did not have
the same retention time during HPLC. Therefore, additional
experiments for the identification of dehydro-ferulic acid-
tyrosine had to be performed. For this reason, [2H]3-ferulic acid
was synthesized by a Knoevenagel reaction of [2H]3-vanilline
and malonic acid. 10.8 mg of [2H]3-ferulic acid was added to
10.0 g of wheat flour, and bread was baked from this mixture.
After hydrolysis, the HPLC/MS signal for m/z ) 374 decreased
in favor for a new signal at m/z ) 377, which is a strong indi-
cation for the formation of dehydro-ferulic acid-tyrosine from
the added [2H]3-ferulic acid. For this experiment, the free acid
([2H]3-ferulic acid) was added because it led to the formation
It is evident that the concentration of DFT increased by more
than 300% during the formation of wheat dough from roughly

Protein−Carbohydrate Cross-Links in Grain-Based Products
J. Agric. Food Chem., Vol. 53, No. 13, 2005 5283
based on two, three, and four tyrosine and ferulic acid building
blocks. Isolation of the relevant compound by HPLC showed
that its structure was not stable, as isomerization was detected
by rechromatography of the compounds. This phenomenon still
has to be clarified.
Table 2. Relative Quantitation of Dehydro-Ferulic Acid-Tyrosine (DFT)
in Wheat (cv. Soissons) and Rye (cv. Nikita) Flour, Dough, and Bread
(Content of DFT in Wheat Flour
) 100%)
DFT (%, in relation to wheat flour)
wheat (cv. Soissons) rye (cv. Nikita)
no additive no additive
-ascorbic acida -ascorbic acida
100.0 18.8
56.3
106.5
The newly identified cross-link contains ferulic acid and
tyrosine. As low amounts of free tyrosine are present in flour
and as ferulate is present in endosperm-AX as well as in lignin
originating from the outer layers of the kernel, it might be
possible that DFT comes from these sources. However, as the
major portions of ferulate and tyrosine are present in AX
(29-31) and (gluten) proteins, respectively, it is very likely that
DFT represents a new covalent cross-link between AX and
proteins in cereal flour. The increase of the DFT concentration
during wheat dough mixing shows that ferulate may have an
impact on the dough properties by forming cross-links during
bread preparation. As the radioactive tracer experiments showed
that ferulate was also incorporated in the rye flour proteins,
further investigations need to be performed to identify chemical
modifications of ferulate in rye dough and bread. Obviously,
mixing and/or air are required for the cross-link reaction, because
baking of the dough did not affect the DFT concentration. The
fact that ascorbic acid led to an increase in the formation of
DFT may be explained with the possible role of dehydro-
ascorbic acid as a hydrogen acceptor during the oxidative
coupling of ferulate and tyrosine.
L
L
flourb
±
±
±
16.2
16.8
20.3
±
±
±
1.5
4.6
2.0
doughb
breadb
376.1
370.6
544.7
578.7
±
±
91.3
32.0
17.3
23.4
±
±
3.6
5.6
a 100 mg/kg of flour. b Number of experiments n
) 3.
0.2 µg/g in flour to 0.7 µg/g in dough, and then remained almost
constant during baking. Addition of L-ascorbic acid (100
mg/kg flour) led to an increase of over 40% in the formation of
DFT during wheat dough preparation. With rye, a much lower
amount of DFT was detected in flour compared to wheat. The
content of DFT in rye flour was only about a sixth of the content
of DFT in wheat flour with roughly 0.03 µg/g. More important,
no significant increase of DFT in rye was observed during dough
preparation and baking. A possible explanation for this observa-
tion might be the lower tyrosine content of rye flour as compared
to wheat flour (0.29% vs 0.33% (m/m)). Furthermore, the
concentration of ferulate in rye arabinoxylans is lower than that
in wheat arabinoxylans (450 µg/g vs 700 µg/g water soluble
arabinoxylan).
With a concentration of about 80 µg ferulate/g and about 3
mg tyrosine/g in wheat flour, the formation of roughly 0.8 µg
DFT/g during dough preparation would correspond to the
linkage of 0.50% of the ferulate and about 0.013% of the
tyrosine present in wheat flour. Assuming an arabinoxylan
molecule with a molecular mass of 100.000 and a protein
molecule with a molecular mass of 40.000, the identified amount
of formed DFT during dough preparation would lead to about
300 µg newly cross-linked arabinoxylan-protein/g of flour.
Although a final statement on the significance of the newly
identified cross-link on the functional properties of the dough
with the results obtained is not possible yet, the DFT cross-
link seems to be of minor importance for the functionality of
dough. However, with one ferulate-modification identified, it
is not clear how many other reactions ferulate is capable of
during dough and bread preparation. This assumption is
confirmed by the results of Table 1, which shows that additional
cross-links may be formed, as about 24% of the added ferulic
acid radio tracer has reacted with prolamines in wheat bread.
Nevertheless, for the first time, a chemical reaction was
identified that may be discussed for the chemical properties of
arbinoxylans and ferulate during wheat dough preparation.
DISCUSSION
Previous studies (17, 21) revealed that ferulate undergoes
chemical modification within the course of the breadmaking
process. In this study, one possible product of this modification,
the dehydro-ferulic acid-tyrosine cross-link, has been identified
for the first time in wheat and rye bread. However, it was
necessary to use a radioactive ferulic acid-galactose ester as a
tracer to enable selective and sensitive detection of ferulate
modifications during breadmaking. A galactose ester was used
instead of free ferulic acid, because the latter is poorly water
soluble and has different chemical properties than an ester.
Furthermore, the acetone ketal of galactose was commercially
available as one of the starting materials for the synthesis. The
product of the synthesis was water soluble and contained ferulic
acid as an ester, similar to ferulate present in AX.
In the first experiments, Osborne fractions were investigated
for the presence of ferulate modifications, as, especially in the
prolamin and glutelin fractions, increased radioactivity would
indicate protein-bound ferulate. The experiments clearly showed
incorporation of ferulate into the protein fractions. The fact that
prolamin fractions were stronger affected than the glutelin
fractions cannot be explained up to now. Possibly, in a dough,
the monomeric prolamins are more mobile than the polymeric
glutelins. Whole bread was then used for the identification
studies to identify all modifications present in bread.
The identification of the cross-link was based on mass
spectrometric data. It was not possible to record NMR spectra
of the isolated compound because it was present in very low
amounts. To verify the identification of the cross-link, its
structure was confirmed by chemical synthesis. However, this
was difficult to achieve as complex mixtures of products were
formed. As initial experiments in the liquid phase gave even
more complex mixtures, a polymer-based solid-phase synthesis
was carried out which generated a more limited mixture, which,
nevertheless, contained more than 20 individual compounds.
These compounds represented different isomers of substances
ABBREVIATIONS USED
AX, arabinoxylan; DFT, dehydro-ferulic acid-tyrosine; DMF,
N,N-dimethylformamide; DMSO, dimethyl sulfoxide; dpm,
disintegration per minute; ESI, electrospray ionization; HPLC,
high performance liquid chromatography; RP-, reversed-phase;
SD, standard deviation; SDS, sodium dodecyl sulfate; SPE,
solid-phase extraction; THF, tetrahydrofuran; TFA, trifluoro-
acetic acid; TLC, thin-layer chromatography.
ACKNOWLEDGMENT
We would like to thank Mrs. I. Otte and Mrs. S. Heinel for
performing the HPLC/MS measurements, and Prof. W. Schwab
for his assistance during the scintillation measurements.

5284 J. Agric. Food Chem., Vol. 53, No. 13, 2005
Piber and Koehler
LITERATURE CITED
(18) Wang, M.; van Vliet, T.; Hamer, R. J. How gluten properties
are affected by pentosans. J. Cereal Sci. 2004, 39, 395-402.
(19) Wang, M.; Oudgenoeg, G.; van Vliet, T.; Hamer, R. J. Interac-
tion of water unextractable solids with gluten protein: effect on
dough properties and gluten quality. J. Cereal Sci. 2003, 38,
95-104.
(20) Frederix, S. A.; Van Hoeymissen, K. E.; Courtin, C. M.; Delcour,
J. A. Water-extractable and water-unextractable arabinoxylans
affect gluten agglomeration behavior during wheat flour gluten-
starch separation. J. Agric. Food Chem. 2004, 52, 7950-7956.
(21) Wang, M.; van Vliet, T.; Hamer, R. J. Evidence that pentosans
and xylanase affect the re-agglomeration of the gluten network.
J. Cereal Sci. 2004, 39, 341-349.
(22) Takasaki, S.; Kawakishi, S. Effects of tyrosine residue on gluten
formation. New Food Ind. 1998, 40, 37-42.
(23) Geissmann, T.; Neukom, H. Composition of the water soluble
wheat flour pentosans and their oxidative gelation. Lebensm.-
Wiss. Technol. 1973, 6, 59-62.
(24) Sidhu, J. S.; Hoseney, R. C.; Faubion, J.; Nordin, P. Reaction
of 14C-cysteine with wheat flour water solubles under ultraviolet
light. Cereal Chem. 1980, 57, 380-382.
(25) Kieffer, R.; Belitz, H.-D.; Zweier, M.; Ipfelkofer, R.; Fischbeck,
G. A 10-g micro-version of the rapid-mix-test. Z. Lebensm.-
Unters.-Forsch. 1993, 197, 134-136.
(26) Wieser, H.; Antes, S.; Seilmeier, W. Quantitative determination
of gluten protein types in wheat flour by reversed-phase high-
performance liquid chromatography. Cereal Chem. 1998, 75,
644-650.
(27) Schmitz, I.; Zahn, H.; Klostermeyer, H.; Rabbel, K.; Watanabe,
K. On the occurrence of isopeptide bonds in heated milk protein.
Z. Lebensm.-Unters.-Forsch. 1976, 160, 377-381.
(28) Jackson, G. M.; Hoseney, R. C. Fate of ferulic acid in overmixed
wheat flour doughs: partial characterization of a cysteine-ferulic
acid adduct. J. Cereal Sci. 1986, 4, 87-95.
(29) Hatcher, D. W.; Kruger, J. E. Simple phenolic acids in flours
prepared from Canadian wheat: relationship to ash content, color,
and polyphenol oxidase activity. Cereal Chem. 1997, 74, 337-
343.
(30) Abdel-Aal, E.-S. M.; Hucl, P.; Sosulski, F. W.; Graf, R.; Gillott,
C.; Pietrzak, L. Screening spring wheat for midge resistance in
relation to ferulic acid content. J. Agric. Food Chem. 2001, 49,
3559-3566.
(1) de Vries, J. W.; Prosky, L.; Li, B.; Cho, S. A historical
perspective on defining dietary fiber. Cereal Foods World 1999,
44, 367-369.
(2) Theander, O.; Westerlund, E.; Aman, P. Structure and compo-
nents of dietary fiber. Cereal Foods World 1993, 38, 135-141.
(3) Ishii, T. Structure and functions of feruloyated polysaccharides.
Plant Sci. 1997, 127, 111-127.
(4) Delcour, J. A.; Van Win, H.; Grobet, P. J. Distribution and
Structural Variation of Arabinoxylans in Common Wheat Mill
Streams. J. Agric. Food Chem. 1999, 47, 271-275.
(5) Izydorczyk, M. S.; Biliaderis, C. G. Cereal arabinoxylans:
advances in structure and physiochemical properties. Carbohydr.
Polym. 1995, 28, 33-48.
(6) Maes, C.; Delcour, J. A. Structural characterisation of water-
extractable and water-unextractable arabinoxylans in wheat bran.
J. Cereal Sci. 2002, 35, 315-326.
(7) Andersson, R.; Aman, P. Cereal Arabinoxylan: Occurrence,
Structure and Properties. In AdVanced Dietary Fibre Technology;
McCleary, B. V., Prosky, L., Eds.; Blackwell Science: Oxford,
UK, 2001; pp 301-314.
(8) Michniewicz, J.; Biliaderis, C. G.; Bushuk, W. Effect of added
pentosans on some properties of wheat bread. Food Chem. 1992,
43, 251-257.
(9) Denli, E.; Ercan, R. Effect of added pentosans isolated from
wheat and rye grain on some properties of bread. Eur. Food
Res. Technol. 2001, 212, 374-376.
(10) Delcour, J. A.; Vanhamel, S.; Hoseney, R. C. Physicochemical
and functional properties of rye nonstarch polysaccharides. II.
Impact of a fraction containing water-soluble pentosans. Cereal
Chem. 1991, 68, 72-76.
(11) Labat, E.; Rouau, X.; Morel, M.-H. Effect of flour water-
extractable pentosans on molecular associations in gluten during
mixing. Lebensm.-Wiss. Technol. 2002, 35, 185-189.
(12) Courtin, C. M.; Delcour, J. A. Arabinoxylans and Endoxylanases
in Wheat Flour Bread-making. J. Cereal Sci. 2002, 35, 225-
243.
(13) Do¨rfer, J. Untersuchungen zur Aktivita¨t pentosanabbauender
Enzyme in Roggenmehlen mit auffa¨lliger Teigbeschaffenheit.
Getreide Mehl. Brot. 2001, 55, 154-157.
(14) Izydorczyk, M. S.; Biliaderis, C. G.; Bushuk, W. Oxidative
gelation studies of water-soluble pentosans from wheat. J. Cereal
Sci. 1990, 11, 153-169.
(15) Brenna, O.; Cantarelli, C. The possible role of wheat and rye
phenolics on functional properties of pentosans in dough. Bulletin
de Liaison - Groupe Polyphenols 1992, 16, 96-99.
(16) Yeh, Y. F.; Hoseney, R. C.; Lineback, D. R. Changes in wheat
flour pentosans as a result of dough mixing and oxidation. Cereal
Chem. 1980, 57, 144-148.
(31) Beaugrand, J.; Croˆnier, D.; Debeire, P.; Chabbert, B. Arabi-
noxylan and hydroxycinnamate content of wheat bran in re-
lation to endoxylanase susceptility. J. Cereal Sci. 2004, 40, 223-
230.
Received for review February 21, 2005. Revised manuscript received
April 25, 2005. Accepted April 29, 2005.
(17) Jackson, G. M.; Hoseney, R. C. Effect of endogenous phenolic
acids on the mixing properties of wheat flour doughs. J. Cereal
Sci. 1986, 4, 79-85.
JF050395B