Journal of Agricultural and Food Chemistry
Article
displaced fractions was done on silica plates (silica gel 60 F254) with
n-butanol/glacial acid/water (4:2:1) as an eluent. Detection of reductic
acid was done under UV light, and detection of partly co-eluating,
unreacted D-galacturonic acid was done by dipping the plate into a
methanolic 10% sulfuric acid solution and heating it at 120 °C for 10
min. The effluents from the resin containing only reductic acid were
pooled and concentrated to dryness under reduced pressure at 35 °C.
The light brown syrup was washed with cold acetone to obtain crude
crystals, which were recrystallized from a mixture of ethyl acetate and
ethyl alcohol or sublimed.
min) was detected. Quantitation was performed by external calibration
with pure reference compounds.
α-Dicarbonyl Compounds as Quinoxalines. The derivatized
samples were analyzed by HPLC-DAD using a flow rate of 0.5 mL/
min, an oven temperature of 40 °C, and a solvent gradient starting
with a mixture (95/5; v/v) of water and acetonitrile and increasing the
acetonitrile content after 5 min to 10% within 3 min. After another 5
min the content was increased to 20% within 7 min and after 5 min
increased to 50% within 10 min. Quantitation was performed by
comparing the peak areas with those of a standard solution at 317 nm
containing known amounts of each pure reference or pure authentic
quinoxaline isolated and elucidated from reaction mixtures in our
working group.
Aldehydes as 2,4-Dinitrophenylhydrazones. Identification of
reaction products was achieved by LC-MS measurements in the
total ion mode. HPLC separation (Thermo Fischer Scientific Accela
pump, CTC PAL autosampler) was carried out by injecting the
derivatized samples using a solvent gradient starting with a mixture
(40/60; v/v) of water and acetonitrile and increasing the acetonitrile
content after 7 min to 100% within 13 min. Ionization (TSQ Vantage
System with Ion Max Source, H-ESI II probe) in the ESI negative
mode followed, using a spray voltage of 3 kV, a vaporizer temperature
of 450 °C, a sheath gas pressure of 60 psi, and a capillary temperature
of 270 °C. Spectra were obtained by setting the scan at 50 to 1000 m/
z. Thermo Excalibur 2.1.0.1139 software was used for the identification
of the substances.
Gas Chromatography−Mass Spectrometry (GC-MS). GC was
performed on a Shimadzu GC-2010 by using a capillary DP-5 column
(30 m × 0.25 mm, 0.25 mm, Supelco SLB-5MS, Bellefonte, PA, USA).
The samples were applied by split injection (1:10) at an injection
temperature of 250 °C and an oven temperature of 100 °C. After 4
min, the temperature of the oven was raised at a rate of 8 °C/min to
200 °C and held for 10 min, then raised again at a rate of 40 to 275 °C
and held for 2 min. The flow of the carrier gas, helium, was 2 mL/min.
MS analysis was performed with a GCMS-QP2010plus (Shimadzu) in
tandem with the GC.
Model Reactions. D-Galacturonic Acid. To investigate the
degradation behavior and the formation of key intermediates and
degradation products, D-galacturonic acid monohydrate (0.5 M) was
dissolved in water, and the pH adjusted with sodium hydroxide and
hydrochloric acid solutions to 3.0, 5.0, or 8.0. Aliquots of the solution
were filled into glass ampules and incubated at 100 °C for 2 h in a
heating block. The reaction was stopped at various time points and
worked up as follows. Browning was measured as the absorption at
420 nm with a spectrophotometer (UV-1650 PC, Shimadzu, Duisburg,
Germany), and the CIELab color with a reflectance attachment with a
Spectralon integrating sphere (Specord 40, Analytik Jena, Jena,
Germany). pH values were measured with a pH electrode (SenTix
Mic, WTW, Weilheim, Germany; pH meter CG 820, Schott, Mainz,
Germany) after cooling the sample to room temperature. For organic
acids and aci-reductones, samples were diluted (1:10) with 1% m-
phosphoric acid and 10 mM dithiothreitol and measured by HPLC-
DAD. For furans and furanones, samples were diluted (1:100; v/v)
with 10 mM phosphate buffer (pH 6.0) and subjected to HPLC-DAD.
For quinoxalines, o-phenylenediamine hydrochloride (OPD; 20 mM)
was added (1:1; v/v) and the samples were reincubated for at least 3 h
at room temperature and subsequently analyzed by HPLC-DAD. For
aldehydes, 2,4-dinitrophenylhydrazine (DNPH) and toluene were
added to the sample (1:2:1; v/v/v) and reincubated for 12 h at room
temperature and in the dark. The samples were vigorously shaken, and
the organic layer was taken and dried under a stream of nitrogen. The
residue was dissolved in acetonitrile and subjected to LC-MS. For
unknown carboxylic acids, dicarboxylic acids, and sugar derivatives, an
aliquot of the solution was dried under a stream of nitrogen and
derivatized with 200 mL of N,O-bis(trimethylsilyl)acetamide with 5%
trimethylchlorosilane in anhydrous pyridine (1:1; v/v) and subjected
to GC-MS.
Reductic Acid and Furan-2-carbaldehyde. To investigate the
browning behavior of degradation products of D-galacturonic acid, 0.05
M reductic acid or furan-2-carbaldehyde with and without L-alanine
(0.005 M) was dissolved in water and the pH adjusted to 5.0 or 8.0.
The solutions were further treated and analyzed like the model
solutions of D-galacturonic acid. All experiments were conducted in
three independent replications, and the arithmetical mean of all
quantified values was calculated.
Characterization and Quantitation. Organic Acids and aci-
Reductones. The diluted samples were analyzed by HPLC-DAD using
a phosphate buffer (20 mM, pH 2.8) as an isocratic eluent and
monitoring the effluent at wavelengths of 214 nm for organic acids
(formic acid (3.9 min), acetic acid (6.3 min), succinic acid (11.4 min),
2-ketoglutaric acid (6.9 min), malic acid (4.2 min)) and 254 nm for
reductic acid (7.5 min) detection. Identification of the organic acids
was accomplished by comparing their retention time with authentic
known acids and by comparing with the results of GC-MS analysis
after silylation of the sample. Standard curves were prepared for each
compound, and the peak areas of the compounds in the samples were
measured and the concentrations calculated from the standard curves.
Furans and Furanones. The diluted samples were analyzed by
HPLC-DAD using a solvent gradient starting with a mixture (99/1; v/
v) of phosphate buffer (5 mM, pH 6.0) and methanol and increasing
the methanol content after 10 min to 20% within 5 min and after
another 5 min to 99% within 5 min. By monitoring the effluent at a
wavelength of 280 nm, furan-2-carbaldehyde (6.7 min), 5-formyl-2-
furanoic acid (16.5 min), and norfuraneol (3.6 min) were detected,
and by monitoring at a wavelength of 254 nm 2-furanoic acid (8.5
High-Performance Liquid Chromatography (HPLC)-DAD.
The HPLC apparatus consisted of one pump (LC-9A, Shimadzu), a
gradient mixer (FCV-9 AL, Shimadzu), a column oven (VDS Optilab,
Berlin, Germany), and an autosampler (Gina50, Dionex, Germering,
Germany). The effluent was monitored by a diode array detector
(DAD, Gynkotek UVD 340S, Gynkotek, Germering, Germany)
operating in a wavelength range between 200 and 500 nm. Separations
of organic acids and aci-reductones were performed on a stainless steel
column (150 × 4.6 mm), packed with ODS-AQ material (3 mm,
YMC, Kyoto, Japan), with a flow rate of 0.6 mL/min at a column
temperature of 25 °C. Separations of furans and furanones were
performed on a stainless steel column (125 × 4.6 mm, Macherey-
̃
Nagel, DA1/4 ren, Germany), packed with an RP-18 material
(Nucleosil, 3 mm), with a flow rate of 0.5 mL/min at a column
temperature of 40 °C. For the separation of α-dicarbonyl compounds
as quinoxalines the same column, but a column temperature of 35 °C,
was used.
Gel Permeation Chromatography−Refractive Index Detec-
tion (GPC-RI). For the characterization of the molecular weight
distribution of the thermally treated samples at 100 °C for 2 h, the
samples were diluted with water and analyzed by GPC-RI. The
molecular distribution of the reacted D-galacturonic acid was compared
to standard solutions of D-glucose, a pullulan standard of 40 kDa, and a
reaction mixture consisting of D-glucose and L-alanine.
RESULTS AND DISCUSSION
■
Thermal treatment of unbuffered solutions of D-galacturonic
acid started at pH 3.0, 5.0, or 8.0 at 100 °C led to 10 times
higher browning after 2 h compared to L-arabinose and D-
glucose or D-galactose.
The browning was measured as the absorption at 420 nm. It
was highest at pH 5.0 and a little lower at pH 3.0 and 8.0 at the
beginning of the reaction. The pH was not controlled during
3495
dx.doi.org/10.1021/jf303855s | J. Agric. Food Chem. 2013, 61, 3494−3500