Synthetic Hydroxytyrosol in Rat Plasma
J. Agric. Food Chem., Vol. 46, No. 10, 1998 3999
1
the H NMR spectrum, which was consist about 15% of
the total product. This byproduct had a similar Rf with
DPE in silica gel TLC and could not be easily removed
from DPE. After further isolation and purification, this
byproduct was found to be (3,4-dihydroxyphenylacetic
acid ester of 2-(3,4-dihydroxyphenyl)ethanol from FAB-
MS spectra (m/z 305 [M + H]+, 327 [M + Na]+, 397 [M
1
+ glycerol + H]+), and was finally certified by H NMR
spectrometry (300 MHz, CD3COCD3), as δ 2.61 (2H, t,
J 7′,8′ ) 7.2 Hz, 7′-H), 3.30 (2H, s, 7-H), 4.03 (2H, t, 8′-
H), 6.40 (1H, dd, J 2,6 ) 2.0 Hz, J 5,6 ) 8.0 Hz, 6-H), 6.45
(1H, dd, J 2′,6′ ) 2.0 Hz, J 5′,6′ ) 8.0 Hz, 6′-H), 6.70 (4H,
m, 2-H, 5-H, 2′-H, 5′-H).
This unexpected result was caused by weak basic
reaction condition, high reaction temperature, and long
reaction time. A simple and efficient synthesis method
was developed by using a quick methylation reagent,
TMSCHN2 (Hashimoto et al., 1981), and the reaction
products were reduced by NaBH4 in an ice-water
solution for 1.5 h with strong stirring. After purification
by flash chromatography on silica gel, the yield of DPE
may exceed 90%, and no byproduct interfered. The
F igu r e 1. Hydroxytyrosol synthesis and byproduct formation
through dihydroxyphenylacetic acid.
by flash chromatography on silica gel with CHCl3:MeOH
(7:1) as eluting solvent to yield DPE as a colorless oil. The
yield was about 90%.
1
purity of DPE was about 98% estimated from H NMR.
An im a l Exp er im en t. Thirty-three healthy male Wistar
rats, weighing 170-190 g, were randomly selected to be in 11
groups, fasted overnight before experiment, and orally admin-
istered of DPE (10 mg/mL in 0.5% tragacanth solution) at a
single dose of 1 mL, except for three rats as blank control. At
each timepoint of 2, 5, 10, 20, 30, 60, 180, 360, 1440, and 2880
min after oral administration, each group (three rats) was
sacrificed and blood samples were collected and heparinized.
Rat plasma obtained was centrifuged and stored at -20 °C
until analysis. During the experiment, the rats were not fed
with any other diets other than those of 1440 and 2880 min,
i.e. six rats.
Extr a ction of DP E fr om Ra t P la sm a . To a tube contain-
ing 1 mL of plasma were added 10 µg of I.S. (as DMF solution)
and 1 mL of 0.3 mol/L HCl-acetonitrile (1:1, v/v). The solution
was vortex-mixed for 15 s and incubated for 5 min at room
temperature. Followed by extraction with 3 mL of ethyl
acetate, the content in tube was centrifuged at 1000g for 6
min. This extraction was repeated twice, and organic layers
were combined and dried over Na2SO4. After filtration and
evaporation, the residue was pertrimethylsilylated for 0.5 h
by adding 75 µL of DMF and 75 µL of BSTFA. One microliter
of the reaction solution was submitted to GC-MS analysis.
Va lid a tion of GC-MS Qu a n tita tive An a lysis. DPE was
dissolved in DMF solution at various concentrations (each
containing 5 µg I.S.) and 75 µL of DMF and 75 µL of BSTFA
were added for trimethylsilyl derivation. The reaction was
carried out at room temperature for 0.5 h, and 1 µL of the
reaction solution was injected for GC-MS analysis. Selected
ion monitoring (SIM) was used for quantitative analysis.
Recovery of DPE from rat plasma was carried out as follows:
known amount (0.01, 0.1, and 1 µg) of DPE was added to a
pool rat plasma, respectively, after treated with HCl-aceto-
nitrile, and extracted with ethyl acetate; it was quantified as
described as above.
Qu a n tita tive An a lysis of DP E by GC-MS. Al-
though HPLC and GLC have been used extensively to
determine DPE in olive oil (Akasbi et al., 1993; J aner
del Valle et al., 1980; Solinas et al., 1982; Tsimidou et
al., 1992), these methods cannot be applied to plasma
samples because components in plasma remain very
complex even after some necessary pretreatment. In
our study, GC-MS was proved far more sensitive and
selective since DPE had a high abundance of molecular
ion (m/z 370.25; 30%). To reduce the endogenous
interferences present in ethyl acetate extracts of plasma,
SIM detection method was applied to raise the quanti-
tative sensitivity by detecting the molecular ion at m/z
370.25 of pertrimethylsilylated DPE. Nonadecanoidic
acid was used as an internal standard (I.S.) since both
of them after TMS derivation had the same M+ at m/z
370. The retention times for DPE and I.S. were 8.05
and 11.95 min, respectively.
The standard calibration curve was obtained by
plotting the peak area ratio of DPE to I.S. with the
known concentration in the stock standard solution,
after the same derivation using BSTFA. The calibration
regression line, expressed as correlation coefficients, was
in very good linearity (r2 ) 0.9998) over concentration
range from 0.01 to 1.3 and 1-5 µg. The detection limit
was estimated as the peak height of DPE at least three
times higher than the baseline noise range, as 0.01 µg.
Recover y of DP E fr om Ra t P la sm a . In developing
a quantitative analysis method, the main problem was
how to isolate DPE from plasma with good recovery and
minimal interference, especially from proteins. Due to
weak acidity of DPE, hydrophilic solvent containing
hydrochloric acid was attempted to precipitate protein
and recover DPE quantitatively. We tested four HCl
concentrations (2, 1, 0.5, and 0.3 mol/L) in extraction
and observed that the lower the pH, the faster the
degradation rate of DPE, particularly when DPE level
was lower than 0.1 µg. Thus 0.3 mol/L HCl was chosen
for extraction. On the other hand, MeOH and CH3CN
were tested and found that CH3CN yielded rather good
recovery, but MeOH caused a severe loss of DPE (40-
60%) during following concentration process.
RESULTS AND DISCUSSION
Im p r ovem en t of DP E Syn th esis by In h ibitin g
Byp r od u ct F or m a tion . Synthesis of DPE from 3,4-
dihydroxyphenylacetic acid, a commercially available
compound, was not so successful, according to the
method of Bianco et al. (1988). DPE, after purified, was
identified by 1H NMR (300 MHz, CD3COCD3), compared
with that of Montedoro et al. (1993), as δ 2.52 (2H, t,
2-H), 3.54 (2H, t, J 1, 2 ) 7.0 Hz, 1-H), 6.41 (1H, dd, J 4,
8 ) 2.0 Hz, J 7, 8 ) 7.0 Hz, 8-H), 6.57 (1H, d, 4-H), 6.59
(1H, d, 7-H) (Figure 1). But the yield of DPE was only
about 50%, with the formation of a byproduct shown in
The concentration of DPE in DMF having 5 µg I.S.
was calculated from standard calibration curve. The