1894 J. Agric. Food Chem., Vol. 54, No. 5, 2006
Rizzi
plus 0.15 mmol of catechol derivative, and mixtures were shaken briefly
to obtain complete mixing. Also, control reactions were run in which
the catechol was omitted. After 1.0 h, the mixtures were optionally
centrifuged to separate precipitated solids (15 min at 5000 rpm). Using
syringe manipulation, the clear supernatant solutions were forced
through C-18 SepPak cartridges, and cartridges were rinsed with 5.0
mL of distilled water followed by 5.0 mL of air. Strecker aldehydes
were eluted from the water-free cartridges with 4.0 mL of methylene
chloride, and final volumes were adjusted to 5.00 mL prior to analysis.
For GC/MS analyses to determine relative and absolute aldehyde yields,
methylene chloride solutions were spiked with an internal standard (d4-
pyrazine) and analyzed by direct injection. Yield data were calculated
assuming 100% SepPak extraction efficiency and GC response factors
of 1.00. For initial qualitative identification of aldehydes, 2,4-
dinitrophenylhydrazone derivatives were prepared by shaking the 5 mL
of methylene chloride solutions with 3 mL of 2,4-DNP reagent (see
Materials and Methods) for 5 min. The separated methylene chloride
phase was evaporated to dryness under a stream of N2, and residues
were fractionated by preparative TLC to obtain pure samples of 2,4-
DNP derivatives for analysis by TLC, ES/MS, and UV-vis spectros-
copy. Isolated derivatives were identified by comparing their data with
those of standard reference compounds. Strecker aldehydes could not
be detected in the control experiments either as free aldehydes or as
2,4-DNP derivatives.
Quantification of Methional 2,4-Dinitrophenylhydrazone. UV-
vis spectroscopy was used to quantify the yield of methional produced
in the reaction of K3Fe(CN)6/methionine/caffeic acid. The total crude
2,4-DNP product obtained under the General Procedure was dissolved
in 5.00 mL of chloroform, and using quantitative techniques, a 0.500
mL aliquot was fractionated by preparative TLC, Rf 0.24 (solvent A).
The isolated TLC band was eluted with MeOH (volume adjusted to
2.00 mL) and analyzed by UV-vis. Using the observed absorption
(A ) 0.483) at the absorption maximum (358 nm) and the molar
extinction (ꢀ) of pure methional 2,4-DNP (22, 400), the total yield of
methional 2,4-DNP was 0.122 mg (0.29% molar yield) equivalent to
8.9 ppm of free aldehyde in the original reaction mixture (5.0 mL)
based on the limiting reagent, caffeic acid.
catechol (1,2-dihydroxybenzene), oxygen, and a phenolase
isolated from Atropa belladonna L. (4). To explain the fact that
two atoms of oxygen were consumed per molecule of catechol,
a two-step mechanism was postulated involving the initial
formation of an o-quinone followed by conjugate addition of
the amino acid and reoxidation to yield a highly colored
4-amino-1,2-benzoquinone intermediate. It was further suggested
that this intermediate reacted with glycine and alanine to produce
observed low yields of R-ketoacids, but neither decarboxylation
nor aldehyde formation were observed indicative of SD.
However, in more recent studies, Strecker aldehydes have been
detected in reactions of amino acids and various catechol
derivatives with polyphenolases isolated from a variety of
sources including the microbe Alternaria tenuis, as well as
extracts of tea leaves, cocoa beans, and coffee beans (5). In
some instances, amino acids were proven to be aldehyde
precursors by radiochemical tracer studies (6); however, the
direct involvement of o-quinoid intermediates in food-related
Strecker reactions has yet to be chemically established. Although
many reports have associated the Strecker degradation with
relatively high temperatures such as those used in roasting or
cooking, the reaction will also take place near room temperature
(ca. 25 °C) with enzymes (5) and in some nonenzymic model
systems (7). Preliminary experiments in our lab demonstrated
the formation of highly colored intermediates and Strecker
aldehydes in model unbuffered reactions of R-amino acids with
1,4-benzoquinone in aqueous solution at 22 °C (8).
The objective of this study was to further characterize the
nature of the low-temperature reactions of amino acids and
catechol-derived oxidation products, viz., o-quinones. In par-
ticular, our plan was to investigate the oxidative chemistry of
R-amino acids with catechol derivatives including natural
flavonoids in a simplified nonenzymic model system in which
potassium ferricyanide was used instead of polyphenoloxidases.
Preparation of Caffeic Acid Quinone and Reaction with 1,2-
Diaminobenzene (OPD). A solution containing 52.0 mg (0.29 mmol)
of caffeic acid in 50 mL of MeOH was treated with 63.0 mg (0.27
mmol) of potassium periodate and stirred for 30 min at 22 °C. Following
the addition of OPD (82.0 mg, 0.76 mmol), the mixture was stirred for
1 h. MeOH was evaporated under a stream of dry N2, and the black
residue was fractionated by prep TLC (solvent C) to afford 46.2 mg
(78% molar yield) of 2,3-diimino-2,3,5,10-tetrahydrophenazine 5 as a
deep reddish-black solid. Dilute solutions of the compound in MeOH
were clear bright yellow. TLC: Rf 0.36 (solvent C); PMR: δ 7.057 (s,
2H, H-1 and H-4), 7.64-7.66 (m, 2H, H-7 and H-8), and 7.97-7.99
(m, 2H, H-6 and H-9) ppm. 13C NMR: δ 103.0 (C-1 and C-4), 128.0
(C-6 and C-9), 128.5 (C-7 and C-8), 141.1 (C-5a and C-9a), 143.7
(C-4a and C-10a), and 146.4 (C-2 and C-3) ppm. UV: nm, (log ꢀ):
426 (3.48), 259 (3.92), 217 (3.69). ES/MS (+) ion mode: m/z 211
amu (MH+) indicative of MW 210 amu.
Ferricyanide Oxidation of 4-Methylcatechol in the Presence of
L-pro(OMe). A solution containing 22.5 mg (0.18 mmol) of 4-meth-
ylcatechol and 50.0 mg (0.30 mmol) of l-pro(OMe) hydrochloride in
4.0 mL of 0.1 M pH 7.17 phosphate buffer was treated at room
temperature (ca. 22 °C) with 1.0 mL of similar buffer containing 197
mg (0.60 mmol) of K3Fe(CN)6. The mixture, which instantly became
deep blood red and later changed to purple, was extracted after 5 min
with 5 mL of ethyl acetate. TLC analysis of the clear, deep purple
extract indicated a single colored product at Rf 0.52 (solvent B). UV:
504.0 nm. ES/MS (+) ion mode: m/z 250 amu (MH+) indicative of
MW 249 amu.
MATERIALS AND METHODS
Reagents. All chemical reagents and solvents were high quality
commercially available materials. C-18 SepPak cartridges were obtained
from Waters Inc. Analytical and preparative TLC was done on Analtech
silica gel GF plates (0.25 mm layer thickness) using 4:1 hexane-ethyl
acetate [A], 1:1 methanol-ethyl acetate [B], or 90:16:8 CHCl3-MeOH-
HOAc [C] v/v as eluting solvents. Phosphate buffer was prepared by
combining 0.10 M aqueous solutions of Na2HPO4 and NaH2PO4. 2,4-
Dinitrophenylhydrazine reagent was a saturated solution of 2,4-
dinitrophenylhydrazine in 1:10 diluted hydrochloric acid. 2,4-Dinitro-
phenylhydrazone standards were prepared by adding pure aldehydes
to excess 2,4-DNP reagent followed by purification by preparative TLC.
Instrumental Analyses. For GC/MS, an Agilent 6890 GC/5973
quadrupole MS combination was used. GC was performed on a 30M
× 0.32 mm i.d. Restek Stabilwax column (film thickness 0.25 µM)
temperature programmed from 40 to 200 °C at 7 °C/min followed by
200-240 °C at 25 °C/min. MS identification was by direct comparison
with standard reference spectra. Electrospray ionization MS (ES/MS)
employed a Micromass Platform LCZ instrument using direct injection
infusion, with source and desolvation temperatures of 130 and 300 °C,
respectively, or a Thermo LCQ ion trap mass spectrometer. With ES/
MS, both positive (+) and negative (-) modes were used to optimize
the detection of analyte ions. NMR spectra were obtained with a 500
MHz Varian Unity Inova spectrometer in CD3OD using solvent signals
for calibration, and multiplicity in analyte signals is indicated by s )
singlet and m ) multiplet. UV-vis data were obtained in MeOH
solution using a computer interfaced Hitachi U-3010 spectrophotometer.
General Strecker Degradation Procedure. Reactions were per-
formed in common screw-capped vials at ambient room temperature
(ca. 22 °C) for periods of 1.0 h. At zero time, 1.00 mL of 0.1 M
phosphate buffer (pH 7.17) containing 0.50 mmol of K3Fe(CN)6 was
added to 4.00 mL of like buffer containing 0.30 mmol of amino acid
RESULTS AND DISCUSSION
It is well-established that Strecker aldehydes are formed under
mild conditions in reaction mixtures containing amino acids,
catechols, molecular oxygen, and a polyphenoloxidase (PPO)
catalyst. Since catechols are known to be oxidized to o-quinones