Rapid Determination of Chloramphenicol Residues in Honey by LDTD-APCI-MS/MS 1
1
1
1
2
Grégory Blachon , Geneviève Paillard , Virginie Dumarchey , Réal Paquin , Pierre Picard , Yves Babin
P-T-108
3
1. Université Laval, Québec, Canada - 2. Phytronix Technologies, Québec, Canada 3. Ministère de l’Agriculture, des Pêcheries et de l’Alimentation du Québec, Québec, Canada
INTRODUCTION
We present a method for the detection of CAP in honey, using a new commercially available technology based on Laser Diode Thermo Desorption (LDTD) 3 of the analyte, followed by atmospheric pressure chemical ionization (APCI) and MS/MS analysis. The sample preparation is a simple liquid-liquid extration of the analyte using ethyl-acetate, and the analysis of each sample by the instrument takes no more than 8 seconds. A linear regression shows a correlation of R = 0.9997 over a range of 0.1 to 50 ppb. The LOD obtained is 0.1 ppb, well below the MRPL. The relative standard deviation (RSD) of 15% for 8 different honeys at the MRPL suggests that no further purification is necessary.
LDTD Technology
RESULTS
Instrumentation (Figure 3)
Desorption Profiles
– LDTD model T-960, Phytronix Technologies – TSQ® QuantumTM Ultra AM, Thermo Fisher Scientific
Precision, Sensitivity and Ruggedness
Honey samples were spiked with CAP standards in aqueous solution before liquid-liquid extraction. An example of desorption profile is shown in figure 6 for a white honey spiked to 0.1, 0.3 and 1.0 ng/g. Preliminary tests have shown that the use of fatty acids, together with a buffer solution (pH 7.4), enhances the signal observed for CAP transitions. This effect will be the subject for further investigations.
Sample preparation
8 honey samples of different floral and geographical origins and a blank honey were fortified with 0.1, 0.3 and 3 ppb of CAP in 3 replicates and subjected to the proposed analytical method. All calibration curves show close agreement with each others (Figure 8). This suggests that 1 single calibration curve could be used to quantify CAP in honey samples of different floral and geographical origins. RSD values (n=24/concentration) show good repeatability over the calibration range for the combined data of the 8 3,5 honeys. y = 0,7438x + 0,0653
Time (s)
Extraction :
CAP 321 → 152
0,5
0
Piston head
Piston
LDTD Parameters • Laser power pattern :
• APCI (-) • Scan time : 0.02 s • Q1 width : 0.50 amu • Q3 width : 0.50 amu • Q2 CID : 1.5 mTorr (Ar) • SRM transition (Table 1)
• Increase laser power to 25 % in 1.0 s • Hold at 25 % for 2.0 s • Decrease laser power to 0 % • Carrier gas flow : 3 L/min (Air) • Corona voltage value : 5 kV
Table 1 : MS parameters for each transition Figure 1 Schematic of the LDTD ionization source.
Transition (m/z)
LazWellTM Plate (Figure 2) CAP
d5-CAP
Tube Lens (V)
Collision Energy (V)
321 → 257
52
13
321 → 152
52
19
326 → 262
52
14
326 → 157
52
20
Figure 5 Chemical structure of chloramphenicol
Linearity To evaluate the linearity of the method, a white honey sample was fortified at different concentrations from 0.1 to 50 ppb in 3 replicates. The analysis of the extracts was repeated on 3 successive days, and the resulting linear relation is shown in Figure 7. We also give the RSD for each concentration (n = 9). ppb CAP
RSD (%)
0.1
27
1
0.1
24
0.3
15
3
11
1,5
2
2,5
3
CAP (ppb)
0.3
16
1
14
3
14
5
10
15
9
50
8
LOD and LOQ The LOD of the proposed method is 3-times lower than the government agencies to verify honey quality. Meanblank : SDblank : Meanslopes : LOD = LOQ =
MRPL of 0.3 ng/g used by several y0 = 0.064 s0 = 0.026 m = 0.762
3s 0 m
LOD = 0.1 ppb
10 s0 m
LOQ = 0.3 ppb
CONCLUSIONS y = 0.73x + 0.07 R = 0.9997
CAP Quantitation Figure 2 LazWellTM sample plate.
RSD (%)
Figure 8 Analysis of 8 different spiked honeys
MS Parameters
– Standard 96-well plate format. – Low volume delivery (from 1 to 10 µL of sample per well). – No carryover. – No sample desalting needed. – No mobile phase needed. – Sample dried at room temperature.
0,5
d5 CAP 326 → 157
LazWell Sample Plate
Corona Discharge Needle
ppb CAP
0
Figure 6 Desorption profiles obtained for a honey spiked at 3 levels
Mass Spectrometer Inlet
y = 0,7018x + 0,0634
1,5
d5 CAP 326 → 262
Figure 4 Preparation of 8 different honeys
Transfer Tube
2
1
– Plug-and-play ionization source can be interfaced to most popular mass spectrometers. – Thermal desorption is induced by a laser diode. – The sample is carried by a carrier gas to a corona discharge region for APCI. – Loader holds up to 10 LazWellTM plates (960 samples).
IR Laser Beam
y = 0,6764x + 0,0967 y = 0,7431x + 0,087 y = 0,8184x + 0,0753 y = 0,751x + 0,078
2,5
• Weigh 3 g of honey Figure 3 LDTD-MS/MS analytical system. • Add d5-CAP (Internal Standard) • Dilute with 2 mL water • Extract analyte with 5 mL Ethyl Acetate • Organic layer filtered with 0.2 µ m Nylon • Dilution 1/4 of the filtered extract with 1:9:2 Stearic acid (5 mg/mL in Ethyl acetate) / Methanol / Phosphate Buffer solution pH = 7.4 • Transfer manually 2.0 µL onto LazWellTM to perform LDTD-MS/MS analysis
LDTD (Figure 1)
Carrier Gas
y = 0,9067x + 0,0217 y = 0,7571x + 0,0519
3
CAP 321 → 257
CAP / d5-CAP
Chloramphenicol (CAP) is a broad-spectrum antibiotic effective against a wide range of microorganisms, but it is well known to have serious toxic effects on humans such as aplastic anemia, and a suspected carcinogenicity.1 CAP is therefore totally prohibited in food in most countries. EU has determined for CAP a minimum required performance limit (MRPL) of 0.3 µg/kg in food of animal origin. Many methods have been recently published for determining CAP residues in food at and below the MRPL, especially for shrimps, milk and honey.2
METHOD
– Very easy sample preparation. – Ultra-fast Chloramphenicol thermal desorption in 3 seconds. – Analysis by the instrument takes less than 8 seconds for each sample. – Linear over a wide range of concentration (R = 0.9997 from 0 to 50 ppb). – Reproducible for different honeys, suggesting that no further purification is necessary. – Suitable method for rapid detection of CAP in honey with a limit of detection of 0.1 ppb.
References
• The following equation was used to obtain the area ratio
1. 2. 3
(Area count )321→257 + (Area count )321→152 CAP = d5 − CAP (Area count )326→262 + (Area count )326 →157 Figure 7 Linearity of the method over a range of 0.1 to 50 ppb for a single honey
J. A. Turton, D. Yallop, C. M. Andrews, R. Fagg, M. York, T. C. Williams Hum. Exp. Toxicol. 18, 566, 1999 L. Santos, F. Ramos Current Pharma. Anal. 2, 53, 2006 J. Wu, C. S. Hughes, P. Picard, S. Letarte, M. Gaudreault, J. F. Levesque, D. A. Nicoll-Griffith, K. P. Bateman Anal. Chem. 79, 4657 – 4665, 2007
