This is an Application Brief and does not contain a detailed Experimental section.
We demonstrate an analytical method to detect trace levels of the neonicotinoid pesticide, imidacloprid, in bee pollen using UPLC-MS/MS. The first step is extracting a sample from the complex bee pollen using Sep-Pak C18 cartridges to remove numerous matrix interferences, followed by quantitative analysis in the pg/g range with the highly sensitive ACQUITY UPLC I-Class System paired with a Xevo TQ-S triple quadrupole mass spectrometer.
Analysis of complex environmental samples for trace levels of contaminants presents a challenge in achieving the necessary sensitivity of the instrument employed to detect sub-ng/g levels through often intense matrix effects. Monitoring honey bee populations for lethal and sub-lethal effects due to pesticide exposure requires reliably quantifying levels of pesticides in both bees and bee products.
A local case in which imidacloprid pesticide was used to treat an invasive insect species prompted the monitoring of area honey bee populations. Honey bee samples from this area and their product, bee pollen, were subjected to analysis of imidacloprid residues. Both sample types are highly complex, with bee pollen containing on average 20% protein and 6% total lipid content,1 as well as a plethora of vitamins, fatty acids, enzymes, and pigments. In the following work, a specialized extraction technique utilizing Waters’ Sep-Pak C18 Cartridges to remove numerous matrix interferences and the highly sensitive Xevo TQ-S for reliable detection of imidacloprid in the pg/g range was employed.
300 mg of bee pollen samples were homogenized in a mortar and pestle, then diluted 1:10 in Milli-Q water. 12 mL of 2% triethylamine in acetonitrile was added to the pollen slurry, vortex mixed, and then added to a QuEChERS AOAC tube 1 (6 mg MgSO4 and 1.5 g sodium acetate). Upon centrifugation, the organic layer was removed and diluted to 15% water content and passed through a Sep-Pak C18 1-cc Cartridge under vacuum. The resulting extract was dried down to near dryness under a gentle stream of nitrogen, and reconstituted in four times the original volume in water.
The samples were then analyzed on an ACQUITY UPLC I-Class System, coupled to Xevo TQ-S. The MRM transition 256.2 → 175.2 was used for quantification, and 256.2 → 209.2 for confirmation. Extraction recovery and matrix suppression were assessed using store-bought organic bee pollen nutritional supplement spiked with 1 ng/g imidacloprid standard, and those results are displayed in Table 1. Both matrix-matched and solvent calibration curves were used for quantification of samples and recovery tests, respectively, and had an r2 > 0.99. Method LOD and LOQ were determined by matrix-matched standards that exceeded a peak-to-peak signal-to-noise (S/N) of 3 and 10, respectively. Well conserved ion ratios were used as an additional criterion for identification.
Of the samples from the exposure area tested, imidacloprid residues were detected in three bee pollen samples, though below the LOQ in matrix, with the exception of one sample that contained 29.3 ng/g. MRMs of the aforementioned sample, one below the LOQ, and a matrix-matched 100 pg/g standard, which surpassed the peak-to-peak S/N ratio of 10, and was considered the method LOQ, as shown in Figure 2 a, b and c.
The detection of imidacloprid bee pollen was performed using the highly sensitive and robust Xevo TQ-S, which proved to be a powerful tool in solving this unique environmental monitoring challenge.
The authors wish to extend special thanks to Kenneth Warchol for providing the bee and bee pollen samples.
720004604, February 2013