302 J. Agric. Food Chem., Vol. 56, No. 2, 2008
Luchtefeld et al.
were prepared as described (12). Data were fit to a straight line by
linear regression analysis using Microsoft Excel.
Negative-ion, atmospheric pressure chemical ionization (APCI)
LC-MS analysis was performed using a Finnigan MAT LCQ Duo
ion trap (West Palm Beach, FL), equipped with a Spectra System P4000
pump, UV6000 LP UV diode array detector, and AP3000 auto sampler.
A C18 Ace analytical column from MacMod Analytical, Inc. (Chadds
Ford, PA), 10 cm × 2.1 mm i.d. and 100 Å pore size, was used for
separation. To determine the molecular weight of impurities, such as
triapocynin, a full-scan mass spectrum over a mass range of 100–750
amu was obtained and an isocratic mobile phase of 35:65:0.38 methanol/
water/ammonium acetate (v/v/w) at pH 7 was used. For the determi-
nation of the concentration of unreacted apocynin, a gradient elution
was used. Solvent A was 480:20:0.38 methanol/water/ammonium
acetate (v/v/w) at pH 7, and solvent B was 20:480:0.38 H O/CH OH/
Figure 1. Conversion of apocynin to diapocynin. This is an oxidation–re-
duction reaction, requiring the in situ generation of sulfate radicals, which
remove one hydrogen from each molecule of apocynin, producing
diapocynin.
(
12, 13). Diapocynin was separated from apocynin using
isocratic, reverse-phase high-performance liquid chromatography
HPLC) (14). The mobile phase was strong enough to cause
2
3
ammonium acetate (v/v/w) at pH 7. The mobile phase started with 100%
A for the first min, followed by a linear increase to 100% B from 1 to
(
1
6 min. This was followed by 100% B from 16 to 31 min and then a
diapocynin to elute before apocynin (14). Still, diapocynin is
not readily available from commercial sources, and no methods
have been reported for its analysis.
linear decrease to 100% A from 31 to 40 min. The injection volume
was 20 µL, and the eluent flow rate was 0.25 mL/min. The analysis
used selected ion monitoring (SIM) of ions with m/z 163.5–166.5 and
In preparation for possible cell culture or animal studies, the
objectives of this study were to develop methods based on
reverse-phase HPLC with ultraviolet (UV) detection and liquid
chromatography-mass spectrometry (LC-MS) to determine the
purity of diapocynin, to establish ways to optimize the synthesis
and purification procedures, and to measure the log P and pKa
of apocynin and diapocynin. These methods were used to
analyze diapocynin in different stages of its synthesis and
purification, to measure its solubility in octanol and water, and
to determine the suitability of dimethylsulfoxide (DMSO) as a
vehicle or solvent for preparing doses of diapocynin.
-
3
27.5–330.5, corresponding to the [M - H] ions produced by apocynin
and diapocynin, respectively. Apocynin standards (0.05–8 µg/mL or
.3–48 µM) were prepared in 1 mM NaOH. Data were fit to a straight
0
line by linear regression analysis using Microsoft Excel.
The pK of apocynin was determined by titrating it with an aqueous
a
solution with NaOH and observing the pH and midpoint. An Orion
520A pH meter was used to measure the pH. The pK
was determined by measuring the apparent log P at different pH values
2.0, 7.4, and 7.9). The diapocynin was at least 99% pure, on the basis
a
of diapocynin
(
of the HPLC and LC-MS analyses. That is, diapocynin accounted for
at least 99% of the total UV peak area by HPLC, and there was less
than 1% unreacted apocynin, as determined by LC-MS analysis. The
log P was determined by the shake flask method (15). Diapocynin in
the aqueous phase was determined using the HPLC method with UV
detection. The concentration of diapocynin in the octanol phase was
determined by extracting the octanol with 0.1 N NaOH, followed by
dilution with deionized water and analysis by HPLC. Five different
measurements of log P were made at each pH; therefore, results could
be reported as an average ( standard deviation.
The concentrations of diapocynin in aqueous suspensions in 10 mM
phosphate buffer at pH 7.0–8.5 were determined by filtering mixtures
of excess diapocynin and buffer that had been sonicated 10 min and
diluting it into 10 mM NaOH until the diapocynin concentration was
within the range of concentrations of standards used in the calibration
curve (1.8–92 µM).
MATERIALS AND METHODS
HPLC-grade methanol was from Burdick and Jackson (Muskegon,
MI). All other chemicals were from Aldrich-Sigma (St. Louis, MO).
Apocynin (acetovallinone) was recrystallized from water and dried in
a desiccator before use. Diapocynin was synthesized by dissolving 2 g
of recrystallized apocynin in 200 mL of deionized water with stirring
and heating until the solution was boiling gently. To this was added
0
.15 g of ferrous sulfate heptahydrate and 1.6 g of sodium persulfate.
A brown precipitate formed. After 5 min, the solution was cooled and
filtered. The precipitate was dissolved in 3 N NH OH and then
4
reprecipitated by adding 6 N HCl. The precipitate was filtered and
washed 3 times with 100 mL of boiling water. The diapocynin
precipitate was further purified by washing 3 times with 100 mL of
boiling methanol. The water and methanol washings were analyzed
for impurities by HPLC and LC-MS. The purified product was dried
in a desiccator before being used in measurements of solubility, log P,
Solutions of diapocynin in DMSO were prepared by sonicating
mixtures of diapocynin and solvent for 10 min. The DMSO solutions
1
did not need filtering. DMSO solutions were diluted
NaOH, followed by another /50 dilution in water.
/10 in 4 mM
1
The diluted samples were analyzed by HPLC with UV detection.
The concentrations of diapocynin in samples that had been stored
for 30 days in closed vials at room temperature were determined
by analyzing them by HPLC, using freshly prepared diapocynin
standards.
a
pK , and stability in DMSO.
H nuclear magnetic resonance (NMR) and H decoupled C NMR
spectra of diapocynin in DMSO-d were obtained using a Bruker ARX
1
1
13
6
1
5
1
00 MHz NMR. A 30° pulse width was used for the H NMR, with a
s pulse delay. A 30° pulse width was used for the C NMR spectra,
1
3
with a 2 s pulse delay. The hydrogen and carbon chemical shifts were
referenced to the DMSO peaks, which were set to 2.50 ppm for
hydrogen and 39.50 ppm for carbon, respectively. The attached proton
test (APT) was used to distinguish between two groups of signals,
methyl/methine and methylene/quaternary.
RESULTS AND DISCUSSION
1
The H NMR spectrum of diapocynin (16) contained peaks
with the following chemical shifts, in ppm, and assignments,
The IR spectrum was obtained using a Varian 800 Fourier transform
infrared spectroscopy (FTIR, Palo Alto, CA) with an attenuated total
reflectance (ATR) accessory, from PIKE Technologies (Madison, WI).
Reverse-phase HPLC was performed using a Shimadzu LC-10 HPLC
in parentheses: 2.494 (-CH ), 3.895 (-OCH ), 7.451 and 7.463
3
3
1
3
(
aromatic CH), and 9.465 (-OH). The C NMR spectrum
contained peaks with the following chemical shifts, in ppm, and
assignments by comparing with the APT spectrum, in paren-
theses: 26.25 (-CH3), 55.98 (-OCH3), 109.66 (aromatic C-2,
CH), 124.46 (C-5, C), 125.25 (C-6, CH), 127.86 (C-1, C),
(
Kyoto, Japan), equipped with an autosampler and a LC-10 AV UV–Vis
detector, set at 276 nm. An Alltech Platinum EPS C18 column, with a
µm particle diameter and 100 Å pore size, was used with an isocratic
5
1
47.44 (C-3, C), 149.11 (C-4, C), and 196.15 (CdO). The FTIR
mobile phase consisting of 48:52:0.31 methanol/water/ammonium
acetate (v/v), flowing at 1.0 mL/min. The injection volume was 20
µL. For quantitative analysis, diapocynin standards (0.16–25 µg/mL
or 1.8–76 µM) were prepared in 1 mM NaOH. Diapocynin standards
-1
spectrum contained the following peaks, λmax cm : 3318 (OH),
1666 (CdO), 1588 (aryl CdC), 1286, 1204, 1127, 1083, and
910.