4122 J. Agric. Food Chem., Vol. 54, No. 12, 2006
Dong et al.
Atchison, KS). The 200 mg ground sample was placed in a 20 mL
polyethylene vial with a white poly lined plastic cap. A 3 mL volume
of 10% KOH/MeOH (w/v) was added, and then the vial was shaken
on a Precision reciprocal shaking water bath (Winchester, VA) at 80
°C for 60 min. The cap was loosened to release internal pressure around
5 min after heating and then retightened. A 1 mL volume of water was
added, and ergosterol was extracted with one 4 mL volume of hexane.
A 2 mL amount of hexane extract (upper layer) was transferred to a 4
mL screw top vial and dried under nitrogen.
(ii) Single Kernel. A single kernel of wheat or barley was weighed
and placed in a 20 mL polyethylene vial with a white poly lined plastic
cap. A 2 mL amount of 5% KOH (w/v) in MeOH/H2O (95/5, v/v) was
added, and the vial was shaken on a Precision reciprocal shaking water
bath at 80 °C for 6 h followed by shaking on an Eberbach reciprocal
shaker (Ann Arbor, MI) at room temperature for 18 h. The cap was
loosened to release internal pressure around 5 min after heating and
then retightened. A 0.5 mL volume of water was added, and one 3 mL
volume of hexane was used to extract ergosterol. A 1.5 mL amount of
hexane extract was transferred to a 2 mL screw top vial and dried under
nitrogen.
Figure 1. Mass spectrum of the TMS derivative of ergosterol.
infected wheat sample with a certain concentration of ergosterol was
run with every 20 samples as a control to check the reproducibility of
sample analysis.
Derivatization. (i) Ground Sample. A 50 µL amount of TMS
(trimethylsilyl) reagent (TMSI/TMCS ) 100/1, v/v) was added to a 4
mL vial containing dried extract. The vial was rotated to ensure that
the TMS reagent was in contact with all of the extract in the vial. The
vial was then shaken on an Eberbach shaker for 10 min. A 500 µL
volume of isooctane containing 400 ng/mL of Mirex was added and
shaken gently, after which 500 µL of water was added. The vial was
shaken on a vortex mixer (Scientific Products, McGaw Park, IL) until
a milky isooctane layer became transparent. The upper layer (isooctane
layer) was transferred to a GC vial.
(ii) Single Kernel. A 25 µL amount of TMS reagent (TMSI/TMCS
) 100/1, v/v) was added to a 2 mL vial containing dried extract. The
vial was rotated to ensure that the TMS reagent was in contact with all
of the extract in the vial. The vial was then shaken on an Eberbach
shaker for 10 min. A 150 µL amount of isooctane containing 400 ng/
mL of Mirex was added and shaken gently, after which 150 µL of
water was added. The vial was shaken on a vortex mixer, and the clear
upper isooctane layer was transferred to a GC vial with a 200 µL glass
insert.
Standard Curve Preparation. Hulls of mycotoxin-free barley seeds
(cultivar Robust grown in Arizona) were removed by a small-scale
pearling machine to produce ergosterol-free barley. A 10 g amount of
ground ergosterol-free barley was saponified with 10% KOH/MeOH
(w/v) at 80 °C and extracted with hexane as described above. A 2 mL
volume of hexane extract was transferred to a 4 mL screw top vial and
dried under nitrogen. These samples were later used as a barley matrix
to prepare the ergosterol standard curve. A standard curve consisting
of 0.025, 0.05, 0.1, 0.25, 0.5, 1.0, 2.5, 5.0, and 10.0 ng/µL of ergosterol
was prepared by adding the appropriate amounts of 10 ng/µL ergosterol
standard solution to the vial containing the barley matrix and deriva-
tizing by the TMS reagent as described above.
Gas Chromatographic-Mass Spectrometric Analysis (GC-MS).
All samples were analyzed on a Shimadzu GCMS-QP2010 (Shimadzu
Corp., Kyoto, Japan) equipped with the AOC-20 auto sampler/auto
injector. All data were processed using GCMSsolution software (version
2.10 Su2F). Perfluorotributylamine (PFTBA) was used to tune the mass
spectrometer. A J&W DB-5MS capillary column (0.25 µm film
thickness, 0.25 mm i.d., and 29.5 m length after cutting 50 cm of the
leading edge for column maintenance) was used to separate compounds.
A high-pressure injection method (300.0 kPa, 1.00 min) was used in
the splitless injection system. Linear velocity of flow control mode
was used with the following oven temperature program: 80 °C for 1
min and then 50 °C/min to 300 °C holding 6 min. Injection, ion source,
and interface temperatures were kept at 290, 220, and 300 °C,
respectively. Injection volume was 1 µL. Ergosterol was detected using
selected ion monitoring (SIM) mode with electron ionization energy
of 70 eV. The fragment ions of m/z 363, 337, and 468 were used for
ergosterol quantitation. Concentrations of ergosterol in barley or wheat
samples were calculated using a standard calibration curve generated
with each set of samples. Mirex in each sample was used to monitor
the stability or the precision of the instrument. A naturally Fusarium-
RESULTS AND DISCUSSION
The mass spectrum of ergosterol-TMS derivative is shown
in Figure 1. Fragment ions of m/z 337, 363, and 468 are three
characteristic ions of ergosterol-TMS derivative. The peak at
m/z 468 is from a molecular ion (M+), and m/z 363, [M - 105]+,
is due to loss of the trimethylsilanol group and one methyl group.
It was suggested that the m/z 337, [M - 131]+, is produced by
loss of the trimethylsilanol group and the C1-C3 fragment (6).
However, the mass spectrum of deuterated ergosterol-TMS
derivative ([4-2H2]ergosterol-TMS) indicated that the m/z 337
is produced by loss of the trimethylsilanol group and C2-C4
fragment (8). The m/z 363 is the most abundant fragment, so it
was use as the target ion for ergosterol quantitation. The m/z
337 and 468 fragments were used as reference ions, and the
intensity ratios of m/z 363/337 and 363/483 together with the
retention time were used to ensure the correct identification of
ergosterol.
Figure 2 shows the standard curves for high and low
concentration ranges of ergosterol. In general, a 9-point standard
curve (0.025-10 ng/µL) was used to calculate ergosterol
concentrations. A 6-point standard curve (0.025-1.0 ng/µL) was
used to quantify concentrations of ergosterol lower than 1.0 ng/
µL to obtain better accuracy. Both calibration curves are quite
linear with R2 values of 0.999 374 and 0.999 587, respectively.
An ion chromatogram constructed from the sum of SIM signals
is shown in Figure 3 for the TMS derivative of 5 ng/µL of
ergosterol prepared in barley matrix. The peak with a retention
time of 9.335 min came from ergosterol-TMS derivative. Mirex
used to monitor the stability of the instrument was observed at
6.638 min. The concentration of Mirex in each sample was the
same, so the intensity of Mirex reflected the stability of the
instrument or the precision of the instrument. Ergosterol-TMS
standard solutions generated by reacting ergosterol with TMSI/
TMCS (100/1, v/v) silylating reagent at room temperature for
10 min and then extracted to isooctane were very stable as
compared with freshly prepared standard solutions. The standard
solutions can be used for at least 2 months by storing in a -20
°C freezer. The detection sensitivity for ergosterol derivatized
by TMSI/TMCS is similar to that by N,O-bis(trimethylsilyl)-
trifluoroacetamide/pyridine (BSTFA/py) (6, 8, 20). The TMSI/
TMCS derivatization method is, however, simple and more
efficient than BSTFA since the latter requires longer derivati-
zation time (30 min) and/or higher temperature (60 °C). No
column deterioration was observed in our study or in the study
conducted by Nielsen et al. (8), a phenomenon experienced by