6644 J. Agric. Food Chem., Vol. 53, No. 17, 2005
Llovera et al.
Figure 1. Synthesis of esters and the relation of 21 acids used.
tubes, 24 mm × 150 mm (Radleys Discovery Technologies, Essex,
United Kingdom). One gram (1.05 mmol) of PS-carbodiimide resin
was added to a dry reaction tube. Each acid (0.85 mmol) was dissolved
in 7 mL of dry dichloromethane (with 1% dry DMF if required). Figure
1 shows all different acids essayed. The solution was then added to
the dry resin, and the mixture was stirred for 1 h at room temperature.
Then, 75.3 mg (0.66 mmol) of 4HM2F in 3 mL of dry CH2Cl2 was
added and the reaction mixture was stirred at room temperature
overnight. The progress of the reaction was monitored by thin-layer
chromatography analysis on silica gel 60 F254 (chloroform-methanol
5%). The reaction mixture was filtered, and the resin was washed twice
with the reaction solvent. To evaluate the presence of ester and its purity,
the filtrate was subsequently analyzed by GC-MS as described below.
A scheme of the reaction is given in Figure 1.
Calibration Standards. Patulin stock solution was prepared ac-
cording to the AOAC Official Method 2000.02 (2). Calibration
standards were prepared from free patulin extract obtained according
to Brause et al. (16). The extract solution was evaporated under vacuum
conditions at 40 °C, and the final residue was redissolved in 200 µL of
standard solutions containing 0.2 µg/mL of ISs and either 0.15, 0.25,
0.38, 0.50, 0.75, or 1.25 µg/mL of patulin.
Spiked Samples. A working solution of 100 µg/mL of patulin was
used to prepare a set of ethyl acetate solutions containing 0.6, 1.0, 3.0,
and 5.0 µg/mL of patulin and 0.8 µg/mL of the ISs (IS-1, IS-2, IS-3,
and IS-4). Five milliliters of free patulin apple juice was then spiked
with 50 µL of these ethyl acetate solutions. After shaking vigorously
for 1 min to homogenize, samples were extracted and cleaned-up as
described by Brause et al. (16). The final residue was dissolved in 200
µL of ethyl acetate and analyzed by GC-MS as described below. The
essays were carried out in triplicate. A UV/vis UV2 spectrophotometer
from Unicam was used to confirm the concentration of patulin in the
stock solution.
GC-MS Analysis. Analysis was carried out using an Agilent 6890N
gas chromatograph interfaced to a 5973N mass selective detector. Mass
spectrometric data were collected in full-scan and SIM modes. Full
scan data were used for preliminary selection of best target m/z ions
and qualifiers. In other cases, the SIM mode was used to quantify patulin
in apple juice in order to maximize sensitivity and selectivity. SIM
was performed monitoring the ions in one group, and the dwell time
applied for each ion was 50 ms with a rate of 3.03 cycles/s.
One microliter of the extract was injected using the on column mode
and following a ramp pressure technique and track oven temperature
programmed. A fused silica deactivated retention gap of 3m × 0.32
mm (Agilent, Anorsa, Barcelona, Spain) was connected between the
injector and the analytical column using a universal deactivated press
fit connector (Agilent). The carrier gas was helium at a constant flow
rate of 1.5 mL/min. The columns used were a HP-5 MS (cross-linked
5% phenylmethylpolysiloxane) 30 m × 0.25 mm i.d. column, df )
0.25 µm (Agilent, Cromlab, Barcelona, Spain) and a Rtx-200 MS
(crossbond trifluoropropylmethylpolysiloxane) 30 m × 0.25 mm i.d.
column, df ) 0.25 µm (Restek, Teknokroma Barcelona, Spain). GC
temperature parameters varied slightly according to the column used.
The oven temperature was programmed at 140 °C and was ramped at
10 °C/min up to a maximum of 280 °C for 10 min when a DB-5 MS
was used. When a Rtx-200 MS was used, the GC temperature was
programmed at 140 °C, initially ramped at 5 °C/min to 170 °C, further
ramped at 15 °C/min up to 280 °C, and then held until a total run time
of 24 min. The GC-MS transfer line was held at 280 °C, and the
quadrupole analyzer and the ion source heaters were maintained at 150
and 230 °C, respectively.
Ion abundance ratios and retention times were applied as criteria
for identification of patulin in samples. The ions selected for patulin
identification were as follows: m/z 110 (M - C2H4O)+, which was
used as the target ion; and 126 (M - CO)+, 136 (M - H2O)+, and 154
(M+), which were used as qualifier ions. The specificity of the
chromatographic method coupled to a MS detector was readily
demonstrated by establishing the ratios for patulin identification, m/z
126/110 ) 55.0 ( 5%, m/z 136/110 ) 34.0 ( 5%, and m/z 154/110)
27.0 ( 5% (mean ( RSD %, n ) 14). Those ratios were calculated
from the different mass spectra obtained analyzing spiked samples with
100 µg/L of patulin in full scan mode (40-400 amu). To confirm the
presence of patulin in samples analyzed in SIM mode, the specified
ratios were calculated by integrating the individual ion chromatograms.
The ion selected for ester quantification was m/z 96 (C5H4O2)+.
Statistical Analysis. Calibration curves were generated in EXCEL
using least-squares linear regression analysis. We established the utility
of esters as ISs by predicting the concentration of spiked samples
throughout the calibration range for three different levels of patulin.
Predictions were made on the basis of the fitted line and the estimated
standard deviation (Sx) of a predicted value for xi (17). The calculation
of the limit of detection (LOD) was based on the residual standard
deviation of the regression line (Sy/x) and the slope (b), LOD ) (3.3 ×
Sy/x)/b.
RESULTS AND DISCUSSION
Preparation of Putative IS. Table 1 summarizes the results
of the solid phase syntheses carried out as indicated in Figure
1. The efficiency of the process and the chromatographic
behavior in the GC-MS analysis is shown for each synthesized
ester. The first six compounds listed in Table 1 were discarded
due to the low yields of their synthesis (<50%). The yield of
the synthesis for the other esters was 80-85%. One microliter
of a 100 µg/mL ethyl acetate solution of each of the chosen
esters was injected into the GC-MS to determine its relative
retention time (rRt) to patulin. The predominant fragments from
their mass spectra obtained in scan mode (scan range 40-400
amu) were also determined. Compounds with rRt values similar
to that of patulin were then selected for the recovery assays.
All of the selected compounds exhibited an m/z 96 fragment
(C5H4O2)+ (Table 1), which was very similar to m/z 110
(C5H2O3)+, the main fragment selected for patulin. Thus, in the
absence of sources of interference, MS performance should
affect fragments in the same way.
Recovery studies were carried out by spiking 5 mL of apple
juice with 50 µL of a mixture of selected esters, each in a
concentration of 0.8 µg/mL. The samples were then extracted
and cleaned-up as described in the Experimental Section (16).
Table 2 shows the recovery efficiency of the esters (Qis)
throughout the selected extraction method. Qis was calculated
for each ester according to the following equation:
Qis ) AhIS-spk × 100/AhIS-std
where AhIS-spk is the mean chromatographic peak area of m/z 96
corresponding to each ester from the spiked samples and AhIS-std
is the expected mean value for each ester. The esters with
recovery rates greater than 90%, IS1, IS2, IS3, and IS4, were