9034 J. Agric. Food Chem., Vol. 58, No. 16, 2010
Li et al.
MATERIALS AND METHODS
standard curves. Composition was calculated as moles relative to moles
of the terminal units. For example, the unfractionated tannin yielded
2.2 mol of terminal catechin, 1.97 mol of terminal epigallocatechin gallate,
3.45 mol of terminal myricetin, and 30.1 mol of extender epigallocatechin
benzylthiol ether, so the mole ratio was 0.29:0.26:0.45:3.96 for terminal
catechin/terminal epigallocatechn gallate/terminal myricetin/extender
epigallocatechin.
The average degree of polymerization (mDP) was calculated from the
molar ratio of all the extender units plus terminal units to terminal units.
Molecular weights were calculated on the basis of the molecular weight of
each subunit.
Chemicals. Analytical or HPLC grade solvents and reagents were used.
Trifluoroacetic acid (TFA), phenylmethane thiol (benzyl mercaptan),
catechin, epicatechin, cyanidin chloride, delphinidin chloride, cesium tri-
fluoroacetate, 2,5-dihydroxybenzoic acid, eriodictyol, quercetin, myricetin,
luteolin, apigenin, kaempferol, and Toyopearl HW50F were from Sigma-
Aldrich (St. Louis, MO). Epigallocatechin gallate, epicatechin gallate, and
epigallocatechin were provided by Lipton Tea Co. (Newark, NJ). Catechin
benzylthioether, epicatechin benzylthioether, epigallocatechin benzylthio-
ether, and gallocatechin benzylthioether were isolated by Moore and
Hagerman (personal communication). AB-8 macroporous resin was pur-
chased from Naikai Chemical Plant, Tianjin, China. Dowex 50 ꢀ 8-400
cation-exchange resin was from Supelco, Bellefonte, PA.
HPLC-ESI-MS analysis was performed on an Agilent 1100 LC inter-
faced to a Esquire LC ion trap MS (Bruker Daltonics, Billerica, MA). The
chromatographic conditions were the same as described above except that
solvent A was 0.5% acetic acid in water and solvent B was acetonitrile. A
source temperature of 500 °C, negative ion mode, was used with a sprayer
needle voltage of 3.5 kV. The capillary temperature was 350 °C. The full
scan mass spectra of the eluate from m/z 100 to 700 were measured using
500 ms for the collection time; three microscans were summed.
Direct injection ESI-MS analysis employed the electrospray ioniza-
tion source (ESI) operated in negative mode. A nominal target mass
was set to 1000 before fine-tuning. The capillary, skimmer 1, and trap
drive voltages were 3500, -51.8, and 72.9 V, respectively. The ion
charge control was on with a target of 30000. The 300 °C nitrogen dry
gas flow rate was 4 L/min, and the nebulization gas pressure was 11 psi.
A syringe pump was used to infuse the sample at 10 μL/min. Each data
point in the spectrum consisted of an average of eight scans over a mass
range of m/z 50-2000.
Anthocyanidin monomer analysis was carried out on the products
of the acid butanol reaction (14) followed by analysis with HPLC
on a Hewlett-Packard 1050 (Hewlett-Packard, Waldbronn, Germany)
equipped with a diode array detector and controlled with Agilent
ChemStation software (A.09.03). The column used was a 30 mmꢀ2.1 mm
i.d., 5 μm, ODS2 C18 with a 4 mm ꢀ 4 mm i.d. guard column of the same
material (Grace Davison, Deerfield, IL). Separation was achieved with
a gradient of 0.13% TFA in H2O (A) and 0.1% TFA in acetonitrile (B)
at 0.5 mL/min in a 45 min program as follows: 0-10 min, 5% B; 10-
15 min, increase to 20% B; 20-25 min, increase to 55% B; 25-35 min,
55% B; 35-40 min, decrease to 5% B; 40-45 min, re-equilibrate at 5% B.
The samples were filtered through a 0.22 μm cellulose acetate spin filter
before injection of 10 μL, and the eluate was monitored at 540 nm. The
peaks were identified by comparison of the retention time and spectra with
those of the commercial cyanidin and delphinidin standards, and peak
areas were calculated with the ChemStation software.
Lyophilized persimmon tannin (0.3 g) was dissolved in about 2 mL of
methanol and fractionated by gravity chromatography (30 cm ꢀ 500 mm i.d.)
on Toyopearl TSK HW-50 (F) using 1000 mL of acetone/methanol (1:4, v/v)
to yield fraction PT14, 1000 mL of acetone/methanol (2:3, v/v) to yield
fraction PT23, and 250 mL of 60% aqueous acetone to yield fraction
PT60. The three fractions were evaporated under reduced pressure at
30 °C, lyophilized, and stored at -20 °C for further analysis.
GPC was carried out on the Agilent series 1050 HPLC equipped with a
300 mm ꢀ 7.5 mm i.d., 5.0 μm, Varian PL Gel MIXED-D size exclusion
column (Varian, Inc., Amherst, MA). An isocratic mobile phase of 1%
water in tetrahydrofuran was delivered at 1 mL/min. The samples were
dissolved at 1 mg/mL in tetrahydrofuran containing 1% water and filtered
through a 0.22 μm nylon spin filter before injection of 10 μL. The eluent
was monitored at 280 nm. Molecular weights were estimated on the basis
of a standard curve generated with narrow polydiversity polystyrene
standards (Mp 0.162-91.8 kDa).
Characterization of Persimmon Proanthocyanidin. Mature and
fully colored fruits of the astringent Shanxi heart-shaped yellow persim-
mon (D. kaki) were harvested in late October from an orchard in Shan’xi
(China). After harvest, fruits were held at 100 °C for about 5 min to
inactivate polyphenol oxidase and were then stored deep frozen at -20 °C.
As we previously reported (6), the yield of proanthocyanidin from
persimmon is optimized by extraction with acidic methanol, similar to
the group II proanthocyanidins from Sorghum grain (13). We therefore
extracted the pulverized fruit with methanol/HCl (1%, v/v) at 90 °C and
purified it on AB-8 macroporous resin followed by polysulfone ultra-
filtration using a membrane with a molecular weight cutoff of 10 kDa.
MALDI-TOF mass spectra were collected on a Bruker Reflex II
MALDI-TOF mass spectrometer (Billerica, MA) equipped with delayed
extraction and a N2 laser (337 nm). Mass spectra were calibrated with
bradykinin (M þ H 1060.6, monoisotopic) as an external standard. All
samples were analyzed in the positive ion reflectron mode to detect either
[M þ Na]þ or [M þ Cs]þ using an acceleration voltage of 25.0 kV and a
reflectron voltage of 26.25 kV. All of the sample/matrix mixtures were
applied (1 μL) onto a stainless steel target and dried at room temperature.
To prepare the samples, 25 μL of a 30 mg/mL solution of the tannin in
methanol was dried under nitrogen. The sample was reconstituted in 25 μL
of 100% ethanol immediately before analysis and was then mixed with the
matrix solution (50 mg/mL 2,5-dihydroxybenzoic acid (DHB) in 100%
ethanol). For detection of [M þ Na]þ, samples (2 μL) were vortexed with
DHB (20 μL), and 300 shots were acquired using a laser attenuation of 68.
For detection of [M þ Cs]þ, samples (2 μL) were vortexed with DHB
(16 μL) followed by the addition of Dowex 50W ꢀ 8-400 cation-exchange
resin (3 μL) in 100% EtOH. The deionized sample/matrix solution (12 μL)
was vortexed with cesium trifluoroacetate (0.8 μL of 10 mM), and 500
shots were acquired using a laser attenuation of 69.
Calculated monoisotopic masses are based on the equation [M þ Na]þ
=
290.08 ꢀ CAT þ 306.07 ꢀ GCAT þ 152.01 ꢀ GALLOYL þ 318.04 ꢀ
MYR - 2.02 ꢀ B - 4.04 ꢀ A þ 22.99. CAT, GCAT, GALLOYL, and
MYR are the numbers of (epi)catechin, (epi)gallocatechin, galloyl esters, and
myricetin subunits, respectively. For the average composition, the average
molecular weights are substituted for the monoisotopic weights. For the Csþ
adducts, Cs (132.91 amu) is substituted for Na (22.29 amu).
Thiolysis was carried out on both the crude persimmon proanthocya-
nidin and the Toyopearl fractions (PT14, PT23, and PT60, see below).
About 5 mg of sample was dissolved in 1 mL of methanol, mixed with an
equal volume of thiolysis reagent (5% (v/v) benzyl mercaptan in methanol
containing 0.2 M HCl), and heated at 60 °C for 2 h (6). The products were
analyzed with an Agilent series 1100 HPLC (Santa Clara, CA) equipped
with a diode array detector in line with a 12 channel CoulArray detector
(ESA, Chelmsford, MA). The column used was a 150 mm ꢀ 4.6 mm i.d.,
5 μm XDB-C8, with a 4 mm ꢀ 4 mm i.d. guard column of the same
material (Agilent, USA). The flow rate was 0.5 mL/min, and the gradient
of 0.13% TFA in H2O (A) and 0.1% TFA in acetonitrile (B) was as
follows: 0-3 min, 15% B; 3-8 min, 15-20% B; 8-10 min, 20-25% B;
10-30 min, 25-35% B; 30-34 min, 35-70% B; 34-44 min, 70% B;
44-47 min, 20% B; 47-52 min, 20-15% B; 52-57 min, re-equilibration
with 15% B. Thiolysis products were filtered through a 0.22 μm cellulose
acetate spin filter before injection of 10 μL, and the eluant was monitored
by DAD from 200 to 800 nm. The electrochemical detector, which
provided added sensitivity, was set to monitor over the voltage range
from 180 to 840 mV with 60 mV increments between channels. Data
collected at 220 nm were integrated (Agilent Chem Station software ver.
A.09.03) and were then converted to moles relative to the terminals using
For 13C NMR, samples were dissolved in 50% H2O/50% acetone at
300 mg/mL (crude persimmon tannin) or at 65 mg/mL (fraction 4). A
Bruker AC-500 MHz NMR spectrometer was employed in the 125 MHz
mode using a 5 mm probe at 15 °C with acetone as the reference. For
the degradation product the spectrum was collected only between 70 and
170 ppm to obtain sufficient signal/noise.
To improve NMR spectroscopic resolution, tannin was chemically
degraded by heating a sample (28 mg/mL) in 6.25% HCl in methanol.
After reaction at 70 °C for 2 h, the reaction mixture was cooled and applied
onto a column of Toyopearl TSK HW-50 (F) resin (50 cm ꢀ 3 cm i.d.)
equilibrated with methanol. The column was eluted with methanol at